The SLC-ome of membrane transport: From molecular discovery to physiology and clinical applications

Abstract

Membrane transporters are essential for human health, mediating the movement of nutrients, electrolytes, metabolites and other molecules across cellular and organellar membranes. Genes encoding these proteins account for approximately 5.2% of the human protein coding genome. Nearly half of these belong to the solute carriers (SLC) supergroup, the largest class of membrane transport proteins, collectively termed the “SLC-ome.” The current SLC-ome comprises 464 SLCs organized into 76 SLC families, of which 24% (111 SLCs) remain orphan transporters with unknown or incompletely characterized function. An additional 52 SLC-like proteins bring the total to 516 membrane transport proteins. SLCs function as molecular gatekeepers, and their dysfunction contributes to a wide spectrum of human diseases, including cancer, diabetes, and immunological, cardiovascular and neurodegenerative disorders. Pathological consequences of SLC defects include hypertension, hyperglycemia, hypercholesterolemia, nutritional deficiencies, metal ion imbalance, oxidative stress, and dysfunction of mitochondria, lysosomes, endoplasmic reticulum and Golgi apparatus. In addition, genetic defects in SLCs are the cause of many rare diseases. Several SLCs require additional subunits to form functional heteromeric complexes, while others exhibit additional or alternative roles, such as acting as transceptors. In this review, we provide updated physiological, structural, mechanistic, and pharmacological insights for each of the 516 human SLC and SLC-like proteins. We also summarize their classification, structural architecture, transport mechanisms and pharmaceutical relevance, and present the most recent SLC gene nomenclature assignments approved by the HUGO Gene Nomenclature Committee (HGNC).

Introduction

Membrane transporters and ion channels serve as gatekeepers of cells and organelles, regulating the uptake and efflux of vital compounds, such as sugars, amino acids, vitamins, trace minerals and electrolytes, as well as waste products and xenobiotics. They play a key role in major physiological processes in the human body, and their dysfunction contributes to the pathogenesis of a wide range of diseases, including cancer, diabetes, and immunological, cardiovascular and neurological diseases. Approximately 5.2 % of the human protein-coding genome is devoted to membrane transport processes, and nearly half of these genes encode members of the solute carrier (SLC) supergroup —the “SLC-ome”— comprising 464 SLCs grouped into 76 SLC families. As shown in Fig. 1, the SLC-ome includes all membrane transporters that are neither ATP-driven pumps, nor ABC (ATP-binding cassette) transporters, nor ion channels, nor aquaporins. This is the largest class of transport proteins, far greater than those of ion channels, ABC transporters and pumps.

Fig. 1. The human SLC-ome.

The figure shows that of the 19,270 human protein-coding genes, 5.2% (1,005 genes) are related to membrane transport. Of these, 46% (464 genes) encode SLC solute carriers (see Table 3). As shown on the right, SLC solute carriers include classical transport proteins as well as all transport proteins that are not water channels, ion channels, pumps or ABC transporters. The SLC nomenclature system aims to provide a systematic categorization of mammalian transport proteins. The percentages of non-SLC transporters and channels in the lower left pie chart are based on data from the HGNC website (https://www.genenames.org/).

The SLC-ome classification traces back to seminal work in the early 1990s, when expression cloning using Xenopus laevis oocytes enabled the identification of several founding members of new transporter families. During this period, the HUGO Gene Nomenclature Committee (HGNC) contacted Matthias Hediger to propose a standardized nomenclature system for these rapidly emerging transporters. This collaboration led to the establishment of the SLC gene nomenclatuer, which has since provides a unified framework for classifying membrane transporters.

In 2004, a special mini-review series was launched to provide an overview of the various SLC transporter families (44). Since then, interest in the SLC genes has grown markedly as their roles in health and disease have become increasingly apparent. The series was republished in an updated form in Molecular Aspects of Medicine nine years later (45). Subsequently, numerous additional SLC-like protein sequences have come to light through database analysis (7), underscoring the need for a new, up-to-date overview of the solute carrier families.

The following fundamental elements and core features of the SLC-ome are addressed in this review:

Section 1. The various types of membrane transport proteins within the SLC-ome

Section 2. The vital compounds, drugs, and waste products handled by SLC membrane transporters

Section 3. The clinical and pharmacological relevance of SLCs

Section 4. The origins of the discovery of the SLC-ome transporter families

Section 5. The establishment and detailed description of the SLC nomenclature system

Section 6. Orphan transporters and SLC-like proteins

Section 7. The structural architecture of the SLC-ome transporters

Section 8. The grouping of SLC families by shared structural folds

Section 9. A comprehensive update of the 464 SLC members that make up the SLC-ome, summarizing the molecular, physiological, clinical, and pharmacological features of each member

Section 10. The description of SLC-like proteins

1. The various types of membrane transport proteins within the SLC-ome

Membrane transporters are known to be either passive or active (46). Passive transporters, also known as facilitative transporters, allow solutes to diffuse across membranes down their concentration gradient. Active transporters create solute gradients across membranes and are classified as primary and secondary active transporters, depending on the mode of coupling to cellular energy. Passive transporters and secondary active transporters are referred to collectively as “secondary transporters”, transferring specific solutes down or against a concentration gradient, and thus include uniporters, cotransporters (also known as symporters), and antiporters (also known as exchangers) (47).

Primary active transporters use a chemical energy source, usually ATP, to move molecules across the membrane against their concentration gradient. Primary active ATP-dependent transporters include members of the ABC transporter family and ion pumps (ATPases). Examples of human ABC transporters include the multidrug resistance protein MDR1, also known as P-glycoprotein (ABCB1), the multidrug resistance-associated proteins MRP4 (ABCC4) and MRP6 (ABCC6), and the breast cancer resistance protein BCRP (ABCG2). They bind and hydrolyze ATP to transport a variety of substances, including xenobiotics, out of cells or into cellular organelles (48, 49). Ion pumps on the other hand hydrolyze ATP to actively pump ions such as Na⁺, H⁺, Ca²⁺, and Cu²⁺ out of cells or into cellular organelles (50–54). Ion pumps also create and maintain electrochemical ion gradients across membranes.

Secondary active transporters include cotransporters (symporters) and exchangers (antiporters). Cotransporters use an electrochemical gradient generated by a primary active transporter, i.e., the Na⁺/K⁺-ATPase (ATP1A1), to move molecules against their concentration gradient. This mode of transport is also called “uphill transport”. Examples include the intestinal glucose transporter SGLT1 (SLC5A1), which couples the cotransport of 2 Na⁺ ions to one glucose molecule, or the presynaptic γ-aminobutyric acid (GABA) transporter GAT1 (SLC6A1), which cotransports 2 Na⁺ ions and 1Cl^- ion per GABA (GABA is a zwitterion and electroneutral at physiological pH). Similarly, in the case of exchangers, the free energy stored in a concentration gradient across a membrane can be used to move another molecule against its gradient in the opposite direction across the membrane. An example is the Na⁺/H⁺ exchanger NHE3 (SLC9A3).

Tertiary active transporters are a variant of mammalian secondary active transporters that use as their driving force an ion gradient established by another secondary active transporter. For example, the H⁺ gradient generated by the Na⁺/H⁺ exchanger NHE3 can drive the activity of the intestinal brush border H⁺-coupled oligopeptide transporter PepT1 (SLC15A1) (55, 56).

Membrane transporters can be further subdivided into electrogenic and electroneutral transporters. Electrogenic transporters translocate charged molecules in such a way that a net charge moves across the membrane during each transport cycle, generating an electrical current and affecting the membrane potential. Examples include pumps that move ions across the membrane and ion-coupled transporters such as SGLT1 (SLC5A1), GAT1 (SLC6A1), PepT1 (SLC15A1), and DMT1 (SLC11A2). In contrast, electroneutral transporters move either uncharged species or a mixture of positively and negatively charged species across the membrane, the latter in such a way that no net charge is translocated. Examples include the glucose transporter GLUT1 (SLC2A1), which allows passive transport of uncharged glucose molecules down their concentration gradient, and the anion exchanger AE1 (SLC4A1), which exchanges Cl^- and bicarbonate (HCO₃^-) ions.

Similar to facilitative transporters, channels allow solutes to move down their electrochemical gradients. However, facilitative transporters typically have a fixed stoichiometry of ion(s)/solute(s) movement per translocation cycle, whereas the flow of ions or solutes through channels is controlled by the open probability of the channel via gating mechanisms and the single channel conductance (i.e., the number of charges per second that pass through the channel at a given voltage). However, as previously reported (57) there is a blurred boundary between the channels and the transporters.

In contrast to transporter- or channel-mediated transport, simple diffusion refers to the free movement of molecules across membranes along their concentration gradient without requiring a transporter or a channel.

The SLC-ome mainly consists of secondary transporters, i.e., passive transporters and secondary active transporters, but not channels. However, despite the exclusion of channels, several currently annotated SLC proteins exhibit channel-like transport mechanisms, which is also supported by structural evidence, such as the SLC14 (urea transporters) (37, 58, 59), SLC31 (copper transporters) (33), SLC41 (MgtE-like magnesium transporters) (39, 60, 61), SLC42 (ammonium transporters) (59, 62–64) and SLC58 (MagT-like magnesium transporters) (65, 66) families.

Conversely, certain channel families, such as the aquaporin (AQP) water channel and the CLC (CLCN) chloride channel families, have not only members with channel properties, but also some with transporter-like properties (67, 68). This further highlights the interface between channel and transport function, and often it is not known whether there are conformational changes associated with the translocation process, a characteristic of transporter function.

Regarding the unique structural architecture of SLCs, since they are mostly secondary transporters, many of them use the “alternating access” transport mechanism, as described later in Section 7.

2. The roles of SLCs range from gating of vital compounds, drugs, and waste products to functioning as sensors

Over the past decades, many different types of SLC families have been identified by different research groups, each responsible for transporting a great variety of solutes essential for daily physiological function, as well as drugs, metabolites and waste products. In addition, several SLCs participate in cellular sensing processes, leading to the term “transceptor”, denoting SLCs that possess both transporter and a receptor-like sensing function (69).

Table 1a lists substrates known to be transported by members of the indicated SLC families, while Table 1b summarizes SLCs reported to be associated with sensing mechanisms. Further details on substrate specificities and sensing functions are provided in the individual SLC family descriptions of Section 9. Information on the chimeric membrane transporter GPR155 (also known as LYCHOS), which combines a transporter domain with a G protein-coupled receptor domain, can be found in the SLC65 family description in Section 9 and the dedicated GPR155 description in Section 10.

Table 1. Roles of SLCs: from transporting essential compounds to functioning as sensors.

a) List of substrates transported by members of the SLC-ome:
Sugars	SLC2, SLC5, SLC45, SLC50, SLC60
Amino acids	SLC3, SLC6, SLC7, SLC16, SLC17, SLC25, SLC32, SLC36, SLC38, SLC43, SLC56, SLC66
Peptides	SLC15, SLC16, SLC21/SLCO, SLC72
Neurotransmitters	SLC6, SLC17, SLC18, SLC22, SLC29, SLC32, SLC36
Nucleosides	SLC28, SLC29
Vitamins	SLC2, SLC5, SLC19, SLC23, SLC46, SLC52, SLC69
Queuine (a micronutrient)	SLC35
Nucleosides	SLC28, SLC29
ATP	SLC17, SLC25, SLC35, SLC62
Phosphate	SLC20, SLC34, SLC53, SLC37
Sodium, potassium, chloride	SLC9, SLC12, SLC74
Nucleotide-conjugated sugars	SLC35
Sugar phosphates	SLC37
Bicarbonate	SLC4, SLC26
Sulfate	SLC26, SLC13
Calcium	SLC8, SLC24, SLC64
Iron, zinc, copper, molybdate	SLC11, SLC30, SLC31, SLC39, SLC40, SLC61
Magnesium	SLC41, SLC57, SLC58, SLC70
Heme	SLC48, SLC21/SLCO, TMEM14C
Ammonium/ammonia	SLC42, SLC73, SLC71
Urea	SLC14
Carnitine	SLC22, SLC25
Polyamines	SLC45
Carboxylates	SLC13, SLC16
Creatine	SLC6, SLC16
Creatinine	SLC22
Organic anions/cations & Xenobiotics	SLC21/SLCO, SLC22, SLC47, SLC51, SLC67
Chemotherapeutic drugs	SLC20, SLC21/SLCO, SLC29, SLC35, SLC46, SLC47
Bile acids	SLC10, SLC21/SLCO
Fatty acids	SLC5, SLC27, SLC59
Acetyl-CoA	SLC33
Cholesterol	SLC65
Sphingosine-phosphate	SLC63
Choline	SLC5, SLC44, SLC49
Lipid-linked oligosaccharides	SLC76
Mitochondrial substrates	SLC25, SLC54, SLC55
b) List of SLCs associated with sensing mechanisms:
Lysosomal nutrient sensing via mTOR (mammalian target of rapamycin) signaling	SLC36A1, SLC38A2, SLC38A9 (amino acid sensing); SLC65A1 (NPC1), GPR155 (cholesterol sensing)
Renal sensing mechanism for changes in acid-base status	SLC4A9
Potential glucose sensor	SLC5A4
Bone phosphate sensing	SLC20-SLC53A1 interplay
Remote sensing and signaling	SLC22A6-SLC22A8 interplay
Potential role in the regulation of food intake and energy production	SLC60A1
Arginine sensor in lysosomes	SLC66A1

3. Clinical and pharmacological relevance of the SLC-ome

a). SLCs as drug targets

Malfunction of SLCs is implicated in a wide spectrum of both common and rare human diseases, reflecting their essential role in transporting vital substances. According to the Online Mendelian Inheritance in Man (OMIM) database (70), ~ 190 SLC genes are associated with inherited disorders (71), many of which are discussed in this review. Furthermore, genome-wide association studies (GWAS) have linked SLC gene polymorphisms to complex diseases (72), further underscoring their importance as therapeutic targets.

Several SLC transporters are targeted by approved drugs. Examples include:

• NKCC2 (SLC12A1): inhibited by loop diuretics to treat hypertension (73).

• NCC (SLC12A3): inhibited by thiazides for diuretic and antihypertensive therapy.

• NHE3 (SLC9A3): inhibited by tenapanor to treat constipation-predominant irritable bowel syndrome (74).

• SGLT2 (SLC5A2): inhibited by gliflozins to lower renal glucose reabsorption in type 2 diabetes (T2D) (75).

• URAT1 (SLC22A12): inhibited by ruzinurad for gout treatment (76).

• SERT (SLC6A4): inhibited by fluoxetine, a selective serotonin reuptake inhibitor (SSRI), for depression (77).

• GAT1 (SLC6A1): inhibited by tiagabine for epilepsy (78).

• VMAT1/2 (SLC18A1/2): inhibited by reserpine for hypertension.

• VMAT2 (SLC18A2): inhibited by tetrabenazine for movement disorders such as in Huntington disease (79).

• ENT1 (SLC29A1): inhibited by dipyridamole to dilate blood vessels in peripheral arterial and coronary artery disease (80).

• NPC1L1 (SLC65A2): inhibited by ezetimibe to lower blood cholesterol levels (81).

Beyond currently marketed drugs, many SLCs are being evaluated as targets in preclinical or clinical development. For example, inhibition of the glycine transporter GlyT1 (SLC6A9) by iclepertin is under investigation for the treatment of cognitive impairment in schizophrenia (82).

Since many SLC polymorphisms underlie rare diseases, most often caused by SLC dysfunction, therapeutic strategies would in many cases require the development of small-molecule activators, a notoriously challenging strategy. An alternative strategy involves using small-molecule correctors that promote the proper localization of disease-causing SLC variants to their target membrane (83). This approach parallels the one used successfully in cystic fibrosis, where correctors bind to the misfolded mutant CFTR proteins, enhancing their trafficking to the cell surface so they can function as a chloride channel (84). This same concept has been applied to the creatine transporter SLC6A8, mutations in which cause creatine transporter deficiency (CTD), accounting for ~2% of all X-linked intellectual disability cases (83). There are currently no treatments for CTD, but a newly developed SLC6A8 corrector restored brain creatine levels in a mouse model carrying an SLC6A8 variant, providing proof of concept for this therapeutic approach (83).

The SLC-ome offers substantial opportunities for therapeutic innovation, including strategies for personalized medicine and drug delivery (see below) (41, 43, 85–91). Notably, about 24% of annotated SLCs are orphan transporters with unknown transport function and/or physiological roles, representing a rich and largely untapped source of potential drug targets.

The growing number of experimentally solved SLC structure, particularly those in complex with therapeutic compounds or other modulators, continues to expand our understanding of SLC druggability and mechanisms of action (92). For example, the 3D structure of the renal Na⁺-glucose cotransporter SGLT2 (SLC5A2) bound to the antidiabetic drug empagliflozin provides valuable insights for the rational design of next-generation SLC-targeted therapies (93).

b). SLCs as drug delivery systems

While SLCs can serve as direct drug targets, as highlighted above, they are also being exploited to facilitate drug delivery across biological barriers, such as the intestinal barrier or the blood-brain barrier (BBB), or into specific tissues and cell types, including cancer cells. Two different SLC-mediated drug delivery strategies have been developed (43):

The prodrug strategy

In this approach, an SLC substrate is covalently linked to the drug molecule via a cleavable bond (94). The resulting prodrug is transported into the target cell by the corresponding SLC. Once in the cytoplasm, the prodrug can be cleaved by enzymes to release the active drug.

The nanoparticle strategy

More recently, a complementary drug delivery approach has been introduced in which drug molecules are encapsulated in nanoparticles (reviewed in (43)). These nanoparticles are chemically modified so that known substrates of a target SLC are attached to their surface, enabling recognition by the desired SLC transporter. Due to their size, nanoparticle binding to the SLC typically triggers endocytosis, allowing them to cross barriers or deliver their cargo to specific tissues. The approach can improve drug delivery efficiency while reducing off-target effects. It has attracted particular interest in oncology as a strategy to target cancer cells with higher precision, thereby minimizing the adverse side effects of chemotherapy (42, 95–97).

c). Role of SLCs in drug absorption, distribution, metabolism, excretion and toxicity (ADMET)

Numerous SLCs play a critical role in drug ADMET, and their functional state can profoundly influence drug pharmacokinetics (98–100). As reviewed (41), this functional state is shaped by both intrinsic factors such as genetic polymorphisms, ethnicity, age, and sex, and extrinsic factors such as diet and concomitant medications. For instance, functional genomics studies have shown that several drug transporters harbor common loss-of-function polymorphisms that contribute to interindividual variability in drug response (72, 101, 102).

d). Role of SLCs in drug approval

This topic has attracted considerable attention in the context of regulatory guidelines designed to ensure the safety and efficacy of drugs in development. Regulatory agencies, including the U.S. Food and Drug Administration (FDA), the European Medicines Agency (EMA), and Japan’s Pharmaceuticals and Medical Devices Agency (PMDA), provide guidance on evaluating the interaction of new chemical entities with transporters, whether as substrates, inhibitors, or inducers (41, 103–108). A major focus is the assessment of transporter-mediated drug-drug interactions (DDIs) (109). Such DDIs can occur when one drug inhibits or induces a transporter, thereby altering the plasma concentration of another drug, which may impact its efficacy or safety and/or lead to serious adverse effects. For example, the histamine H₂ receptor antagonist cimetidine inhibits the OCT2 (SLC22A2)- and MATE1 (SLC47A1)-mediated renal transport of the antidiabetic drug metformin. OCT2 is expressed in the basolateral membrane and MATE1 in the apical membrane of renal proximal tubules cells and their inhibition reduces metformin excretion (110) (see the description of SLC22A2). Understanding the potential for such DDIs is therefore essential during drug development. Furthermore, numerous factors including disease states, co-administered drugs, and nutritional status can alter the functional expression of SLCs in key sites such as the intestine, liver, kidney, brain, lungs, placenta, and tumor tissue (104).

e). Role of SLCs in precision medicine

The functional state of SLCs is a key factor in precision medicine, guiding the selection of appropriate drugs and the optimization of dosing regimens to maximize efficacy while minimizing toxicity. In oncology, insights into the function of SLCs are crucial for designing targeted chemotherapy strategies and nanoparticle-based approaches that enhance drug delivery to tumors (42, 95–97, 102).

4. Origin of the discovery of SLC families and the central role of expression cloning in identifying founding members

Compared to other human gene families, the molecular discovery of SLCs lagged behind for two main reasons:

• First, the hydrophobic nature of these integral membrane proteins made purification difficult. This often prevented the generation of sufficient amounts for antibody production or for amino acid sequencing (via Edman degradation (111, 112)) to create antibody or nucleic acid probes for screening cDNA libraries. An exception was the anion exchanger AE1 (SLC4A1), which could be purified more easily due to its high expression in red blood cells (113). However, most transporters are expressed at low levels in native tissue, and attempts to obtain enough pure protein for antibody generation or amino acid sequencing generally failed. A classic example is the intestinal Na⁺/glucose transporter, where biochemical approaches were unsuccessful (114), necessitating new strategies for molecular identification.

• Second, the structural and amino acid sequence diversity of transporters across families made it nearly impossible to identify new transporters by homology-based cloning before the human genome sequence became available (115–117).

The molecular cloning of SLCs began in the early 1980s, with different strategies. Among the first successes were the mitochondrial ADP/ATP carrier ANT1 (SLC25A4, 1982) (118), the Cl^-/HCO₃^- exchanger AE1 (SLC4A1, 1985) (119), and the glucose transporter GLUT1 (SLC2A1, 1985) (120). Despite these achievements, progress soon stalled due to the persistent challenge of purifying hydrophobic transport proteins.

A breakthrough came with the development of expression cloning in Xenopus laevis oocytes (121, 122). In this approach, cRNAs prepared from cDNA library clones are microinjected into the oocyte cytoplasm, where the encoded transporters are expressed in the plasma membrane. Clones can then be directly selected based on their ability to induce the desired transport function (122, 123). This method circumvented the need for antibodies or sequence-based probes. The strategy leveraged several unique features of Xenopus oocytes:

• Their large size (~1 mm in diameter), which facilitates cRNA microinjection.

• Their suitability for assays such as radioisotope uptake or two-electrode voltage clamp.

• Their low background expression of endogenous transporters.

• Their strong protein-synthetic capacity, enabling efficient expression of membrane proteins from foreign, microinjected cRNAs.

The power of this approach was first demonstrated by the cloning of the Na⁺/glucose cotransporter SGLT1 (SLC5A1) from rabbit small intestine (121, 124), marking the first molecular identification of a Na⁺-coupled transporter. Since then, expression cloning in Xenopus oocytes has led to the discovery over 20 transporters, including the first members of numerous entirely new SLC families as well as critical new representatives of existing families that could not have been identified by homology-based strategies (Table 2a).

Table 2. List of the SLC solute carrier families that were identified from different species by expression cloning.

Important new members of SLC families not identified by homology searches are also included in the list.

a) Expression cloning using Xenopus oocytes (22 cases)
1987	SLC5 – Rabbit intestinal Na+/glucose cotransporter SGLT1 (SLC5A1) (121).
1991	SLC10 – Rat liver Na+/bile acid cotransporter NTCP (SLC10A1) (141).
1991	SLC17 – Rabbit renal phosphate transporter NaPi-1 (SLC17A1) (142).
1992	SLC1 – Rabbit intestinal epithelial and neuronal high-affinity Na+/glutamate transporter EAAC1/EAAT3 (SLC1A1) (134).
1992	SLC3 – Rat and rabbit kidney heavy chain D2/NAA-Tr/rBAT (SLC3A1) (135, 143, 144).
1992	SLC5A3 – Canine renal Na+-coupled myo-inositol transporter SMIT (SLC5A3) (145).
1993	SLC34 – Human and rat renal Na+/PO42− cotransporters NaPi-IIa (SLC34A1) (146).
1993	SLC14 – Rat renal urea transporter UT2 (SLC14A2) (138).
1993	SLC12 – Winter flounder urinary bladder thiazide-sensitive Na+/Cl− cotransporter (SLC12A3) (137).
1993	SLC13 – Rat kidney cortex Na+-sulfate cotransporter Slc13a1 (147).
1994	SLC15 – Rat intestinal H+-coupled oligopeptide transporter PepT1 (SLC15A1) (139).
1994	SLC28 – Rat intestinal pyrimidine nucleoside transporter CNT1 (SLC28A1) (148).
1994	SLC21/SLCO – Rat sodium-independent bile salt and organic anion transporter Oatp1a1, also known as Oatp1 (Slc21a1/Slco1a1) (149). Note that “SLCO1A1” does not exist in humans.
1994	SLC22 – Rat kidney organic cation transporter OCT1 (SLC22A1) (150).
1994	SLC26 – Rat liver Na+-independent SO42- transporter Sat-1 (SLC26A1) (151).
1996	SLC5A5 – Rat thyroid Na+/I- cotransporter NIS (SLC5A5) (152).
1997	SLC4A4 – Salamander renal electrogenic sodium bicarbonate cotransporter NBC1 (SLC4A4) (153).
1997	SLC11 – Rat intestinal H+-coupled divalent metal transporter DCT1 (DMT1/SLC11A2) (136).
1998	SLC7 – Rat (C6 glioma cells) Na+-independent neutral amino acid transporter LAT1 (SLC7A1) (154).
1999	SLC23 – Rat intestinal Na+-coupled vitamin C transporter SVCT1 (SLC23A1) (140).
2001	SLC51 – OSTα (SLC51A) and OSTβ (SLC51B) from the liver of little skate Raja erinacea (155).
2003	SLC43 – Human LAT3 amino acid transporter (SLC43A1) (156).
b) Expression cloning in cultured cells or yeast as an alternative approach to Xenopus oocytes (8 cases).
1991	SLC6A2 – Human cocaine- and antidepressant-sensitive noradrenaline transporter NET (SLC6A2) (157).
1992	SLC18 – Rat vesicular reserpine-sensitive monoamine transporter VMAT1 (SLC18A1) and VMAT2 (SLC18A2) (158, 159).
1992	SLC16 – Hamster H+-coupled pyruvate and lactate transporter MCT1 (SLC16A1) (160).
1993	SLC36 – Plant H+-driven amino acid transporters (161, 162).
1994	SLC27 – Mouse fatty acid transport protein FATP1 (SLC27A1) (163).
1994	SLC10A2 – Hamster Na+-dependent ileal bile acid transporter IBAT (SLC10A2) (164).
1996	SLC39 – Arabidopsis thaliana iron-regulated transporter IRT1 (165).
1997	SLC33 – Human ACATN acetyl-CoA endoplasmic reticulum/Golgi apparatus transporter (SLC33A1) (166).

Expression cloning was later adapted to cultured cells (125, 126) and yeast (127), enabling the discovery of eight additional founding members of new SLC families or key subbranches of existing ones (Table 2b). These approaches required robust selection systems to identify clones with the desired transport activity.

In parallel, additional strategies contributed to the identification of SLCs transporters, including functional complementation (128), positional cloning (129), differential screening (130), analysis of disease-causing mutations (131), exon trapping (132), and, in some cases, serendipity (133).

5. Establishment of the SLC nomenclature system

The SLC classification system has been established by Matthias Hediger in collaboration with the HUGO Gene Nomenclature Committee (HGNC; https://www.genenames.org) (44, 45). Phyllis McAlpine, chair of the HGNC from 1992 to 1996, contacted Hediger in the early 1990s to establish a standardized nomenclature for the genes encoding the rapidly emerging class of membrane transport proteins. At that time, Hediger and colleagues had characterized the molecular properties of several novel transporters through expression cloning, leading to the identification of the founding members of multiple SLC families, including SLC1 (SLC1A1) (134), SLC3 (SLC3A1) (135), SLC5 (SLC5A1) (121), SLC11 (SLC11A2) (136), SLC12 (SLC12A3) (137), SLC14 (SLC14A2) (138), SLC15 (SLC15A1) (139), and SLC23 (SLC23A1) (140) (see Table 2).

The resulting SLC classification system is based on the following standardized principles:

• 1Root symbol: designates each family with the root symbol SLC (for Solute Carrier) followed by a number (e.g., SLC1, Solute Carrier Family 1) (44, 45). Today, the SLC-ome comprises 76 SLC distinct gene families (Table 3) (see also https://www.genenames.org/data/genegroup/#!/group/752 or https://www.bioparadigms.org/slc/intro.htm).
• 2SLC gene symbol structure and maintenance: As recommended by Phyllis McAlpine, the SLC family designation consists of the family number, followed by the letter A (originally introduces as a neutral spacer), and then the number of the individual transporter gene (e.g., SLC1A1, SLC1A2, etc.). Phyllis McAlpine made the initial family assignments based on a list of novel membrane transporters provided by Matthias Hediger. Subsequently, the nomenclature has been maintained in collaboration with Elspeth Bruford, the current lead of the HGNC, with Hediger serving as specialist advisor for the SLC gene series.
• 3Assignment rules (original and current): Originally, a transporter was assigned to an SLC family if the encoded protein shared roughly 20% or more amino acid sequence identity with other family members (44, 45). Following the initial assignments, families have been added in chronological order. The initial assignments were based on the list of novel transporter genes provided to McAlpine and were not strictly chronological, as the order of the list did not reflect the order of discovery. Subsequent assignment decisions have been made jointly by Hediger, HGNC members, and the scientists who identified the transporter genes. Assignments were made regardless of whether the identified transport proteins contained multiple transmembrane helices (TMHs) or only a single TMH.

  Over time, the 20% amino acid identity rule became difficult to apply consistently. Sequence identity values can vary depending on the alignment software and substitution matrices used, the set of sequences included in the alignment, and the choice of the reference sequence in pairwise comparisons. These issues are particularly pronounced when sequence lengths differ significantly, for example due to additional functional domains on the polypeptide chain.

  As a result, more recent SLC classification efforts have shifted towards a phylogeny-based approach. Consequently, families and subfamilies are now primarily defined based on well-supported clades, as well as structural architecture and functional similarities.

• 4Pseudogenes: Some SLC superfamily genes are unitary pseudogenes in humans but retain coding orthologs in other species. In such cases, the locus is assigned the next available family symbol with a “P” suffix (e.g., SLC6A10P) to indicate that it is a pseudogene. The coding orthologs in other species retain the same symbol but without the “P”.
• 5Aliases: Some genes later recognized as SLCs already had widely used non-SLC symbols. In such cases, the original symbol was usually retained as the approved name, with the SLC symbol added as an alias, e.g., OCA2 (SLC13B1).
• 6 Variations and exceptions of the SLC classification system:

Table 3. List of approved SLC families.

The total number of members in each family is shown on the right. The first three columns on the right show the status in 2004 (44), 2013 (45) and 2025, and the numbers also include orphan transporters and pseudogenes. Of the 475 SLC genes in 2025, 11 are pseudogenes (SLC6A10P, SLC6A21P, SLC7A15P, SLC19A4P, SLC22A20P, SLC23A4P, SLC26A10P, SLC35E2A, SLC66A1LP, SLC68A2P, SLC71A3P), resulting in 464 genes encoding SLCs. The right column shows the number of orphan transporters in 2025, which phylogenetically belong to specific SLC families, but whose transport function and/or physiological roles are still not clear. 24% of all SLCs are orphan transporters. Each family name is followed in parentheses by the corresponding family from the Transporter Classification Database (TCDB) (167), the dominant Pfam (Protein families) model name (168) and the structural fold (47, 169, 170).

The HGNC Solute Carrier Family Series	Total 2004	Total 2013	Total 2025	Orphan transporters 2025
SLC1:  High-affinity glutamate and neutral amino acid transporter family (2.A.23/SDF/DAACS)	7	7	7	0
SLC2:  Facilitative GLUT transporter family (2.A.1.1/Sugar_tr/MFS)	14	14	14	1
SLC3:  Heavy subunits of the heteromeric amino acid transporters (8.A.9/SLC3A2_N/single TMH)	2	2	2	0
SLC4:  Bicarbonate transporter family (2.A.31/HCO3_cotransp/NAT)	10	10	10	0
SLC5:  Sodium glucose cotransporter family (2.A.21/SSF/APC)	8	12	12	0
SLC6:  Sodium- and chloride-dependent neurotransmitter transporter family (2.A.22/SNF/APC)	16	21	21	1
SLC7:  Cationic amino acid transporter/glycoprotein-associated family (2.A.3/AA_permease_2/APC	14	14	14	1
SLC8:  Na+/Ca2+ exchanger family (2.A.19.3/Na_Ca_ex/NCX)	3	3	4	0
SLC9:  Na+/H+ exchanger family (2.A.36 and 2.A.37/Na_H_Exchanger/NhaA)	8	13	14	4
SLC10: Sodium bile salt cotransport family (2.A.28/SBF/NhaA)	6	7	7	4
SLC11: Proton-coupled metal ion transporter family (2.A.55.2/Nramp/APC)	2	2	2	0
SLC12: Electroneutral cation-coupled Cl- cotransporter family (2.A.30/AA_permease and SLC12/APC)	9	9	9	0
SLC13: Na+-sulfate/carboxylate cotransporter family (2.A.47.1/Na_sulph_symp/AbgT)	5	5	6	1
SLC14: Urea transporter family (1.A.28.1/UT/UT)	2	2	2	0
SLC15: Proton oligopeptide cotransporter family (2.A.17/PTR2/MFS)	4	5	5	1
SLC16: Monocarboxylate transporter family (2.A.1.13/MFS_1/MFS)	14	14	14	5
SLC17: Organic anion and vesicular glutamate transporter family (2.A.1.14/MFS_1/MFS)	8	9	9	0
SLC18: Vesicular amine transporter family (2.A.1.2/MFS_1/MFS)	3	4	4	0
SLC19: Folate/thiamine transporter family (2.A.48/Folate_carrier/MFS)	3	3	4	0
SLC20: Type III Na+-phosphate cotransporter family (2.A.20.2/PHO4/PiT)	2	2	2	0
SLC21/SLCO: Organic anion transporter family (2.A.60/OATP/MFS)	11	12	12	3
SLC22: The organic cation/anion/zwitterion transporter family (2.A.1.19/Sugar_tr/MFS)	18	23	27	9
SLC23: Na+-dependent vitamin C transporter family (2.A.40.6/Xan_ur_permease/NAT)	4	4	4	0
SLC24: Na+/(Ca2+-K+) exchanger family (2.A.19.4/Na_Ca_ex/NCX)	5	6	5	0
SLC25: Mitochondrial carrier family (2.A.29/Mito_carr/MCF)	27	53	53	14
SLC26: Multifunctional anion exchanger family (2.A.53.2/Sulfate_transp/NAT)	10	11	11	0
SLC27: Fatty acid transporter family (4.C.1.1/AMP-binding/single TMH)	6	6	6	0
SLC28: Na+-coupled nucleoside transport family (2.A.41.2/Gate/CNT)	3	3	3	0
SLC29: Facilitative nucleoside transporter family (2.A.57.1/Nucleoside_tran/MFS)	4	4	5	0
SLC30: Zinc efflux family (2.A.4/Cation_efflux/CDF)	9	10	11	1
SLC31: Copper transporter family (1.A.56/Ctr/Ctr)	2	2	2	1
SLC32: Vesicular inhibitory amino acid transporter family (2.A.18.5/Aa_trans/APC)	1	1	1	0
SLC33: Acetyl-CoA transporter family (2.A.1.25/Acatn/MFS)	1	1	2	1
SLC34: Type II Na+-phosphate cotransporter family (2.A.58/Na_Pi_cotrans/unknown)	3	3	3	0
SLC35: Nucleoside-sugar transporter family (2.A.7/Nuc_sug_transp, UAA, TPT, SLC35F/NST)	17	30	32	18
SLC36: Proton-coupled amino acid transporter family (2.A.18.8/Aa_trans/APC)	4	4	4	1
SLC37: Sugar-phosphate/phosphate exchanger family (2.A.1.4/MFS_1/MFS)	4	4	4	3
SLC38: The System A & N, sodium-coupled neutral amino acid transporter family (2.A.18.6 and 2.A.18.9/Aa_trans/APC)	6	11	12	4
SLC39: Metal ion transporter family (2.A.5/Zip/ZIP)	14	14	14	0
SLC40: The basolateral iron transporter family (2.A.100/FPN1/MFS)	1	1	1	0
SLC41: MgtE-like magnesium transporter family (1.A.26/MgtE/MgtE)	3	3	3	0
SLC42: Rh type glycoprotein family of ammonium transporters (1.A.11/Ammonium_transp/AmtB)	3	3	5	2
SLC43: Na+-independent, system-L-like amino acid transporter family (2.A.1.44/MFS_1/MFS)	2	3	3	0
SLC44: Putative choline transporter CTL1 family (2.A.92/Choline_transpo/SLC44)		5	5	2
SLC45: H+/sugar cotransporter family (2.A.2.4/MFS_1, MFS_2/MFS)		4	4	0
SLC46: Folate transporter family (2.A.1.50/MFS_1/MFS)		3	3	0
SLC47: Multidrug and Toxin Extrusion (MATE) family (2.A.66.1/MatE/MATE)		2	2	0
SLC48: Heme transporter family (2.A.110.1/HRG/unknown)		1	1	0
SLC49: FLVCR-related transporter family (2.A.1.28/MFS_1/MFS)		4	4	2
SLC50: Sugar efflux transporters (2.A.123.1/MtN3_slv/SWEET)		1	1	0
SLC51: Transporters of steroid-derived molecules (2.A.82/Solute_trans_a/structure unknown)		2	5	3
SLC52: Riboflavin transporter family (2.A.125/DUF1011/MFS)		3	3	0
SLC53: XPR1 phosphate exporter (2.A.94/EXS/XPR)			1	0
SLC54: Mitochondrial pyruvate carrier family (2.A.105/MPC/SemiSWEET)			3	0
SLC55: SLC55 LETM mitochondrial cation/proton exchanger family (2.A.97/LETM1/single-TMH)			3	3
SLC56: Sideroflexins (2.A.54/Mtc/unknown)			5	4
SLC57: NIPA-like magnesium transporter family (2.A.7.25/Mg_trans_NIPA/NST)			6	2
SLC58: MagT-like magnesium transporter family (1.A.76.1/OST3_OST6/MagT)			2	0
SLC59: Sodium-dependent lysophosphatidylcholine symporter family (2.A.2.3/MFS_2/MFS)			3	0
SLC60: Glucose transporters (2.A.1.7/MFS_1/MFS)			2	0
SLC61: Molybdate transporter family (MFSD5) (2.A.1.40/MFS_5/MFS)			1	1
SLC62: ATP exporter (2.A.66.9/ANKH/MATE)			1	0
SLC63: Spinster sphingosine-phosphate transporters (2.A.1.49/MFS_1/MFS)			3	1
SLC64: Golgi Ca2+/H+ and/or Mn2+/H+ antiporter (2.A.106.2/UPF0016/SLC64)			1	1
SLC65: NPC-type cholesterol transporters (2.A.6.6/Patched/RND)			7	3
SLC66: PQ-loop amino acid transporter family (2.A.43/PQ-loop/SWEET)			6	3
SLC67: Organic cation transporter-like family (2.A.1.2.53/PF07690/MFS)			2	2
SLC68: Cation symporter family (2.A.2/ MFS_2/MFS)			2	1
SLC69: Vitamin A receptor/transporter family 2.A.90/RBP_receptor/STRA6)			1	0
SLC70: Cyclin M Mg2+ exporter family (1.A.112/ CNNM/CNNM)			4	3
SLC71: Putative ammonia transporter family (2.A.1.2/MFS_1/MFS)			3	2
SLC72: Lysosomal solute carrier family (MFSD1) (2.A.1.53/MFS_1/MFS)			1	0
SLC73: Orphan MFSD6 transporter family (2.A.1.65/MFS_1/MFS)			2	2
SLC74: Lysosomal chloride channel family (MFSD8) (2.A.1.2/MFS_1/MFS)			1	0
SLC75: Tetracycline transporter-like family (MFSD10) (2.A.1.2/ MFS_1/MFS)			1	1
SLC76: Glycolipid translocator family (RFT1) (2.A.66.3/Rft-1/MATE)			1	0
Total	298	395	475	111

• In the SLC8, SLC9, SLC13, SLC18, SLC22, SLC29, SLC59 and SLC65 families, the letter “A” has been extended to “B”, or “C”, etc. to denote subfamilies of the corresponding SLC families.

• SLC51 consists of two distinct members, SLC51A and SLC51B, which are not related by sequence similarity. They encode the actual organic solute carrier α-subunit (SLC51A) and the ancillary β-subunit (SLC51B). To denote new, phylogenetically distant members related to SLC51A, the root symbol SLC51C was introduced (e.g., SLC51C1).

• The SLC21 organic anion transporter family was renamed SLCO, an update implemented to accommodate a special, species-independent classification framework, as required by certain members of the research community (see “Note” in the description of the SLCO family in Section 9).

• 7Gene symbols in other vertebrates: Gene symbols in vertebrates such as mouse, rat, Xenopus laevis, chicken and zebrafish, as well as those named by the Vertebrate Gene Nomenclature Committee, typically adopt the human gene nomenclature for orthologous genes (171).
• 8Protein symbols: The SLC gene nomenclature has been widely used in the literature, but for many SLC proteins, alternative symbols derived from commonly used functional names have remained more popular. In general, the HGNC strives for gene symbols to be consistent with published protein names, but has accepted that this is not always possible for all SLCs. There is no official nomenclature for proteins, and hence, there are often several commonly used symbols for each protein. Conversely, sometimes the same protein symbol has been applied to more than one SLC (e.g., PAT1 for both SLC36A1 and SLC26A6), highlighting the practical utility of the standardized SLC gene nomenclature system.

An alternative classification system for solute carriers and other transport-related proteins was initiated in the 1990s by Milton Saier, called the Transporter Classification Database (TCDB) (172). The TCDB aims to classify proteins based on molecular phylogeny, while assigning members a five-segment number similar to the Enzyme Classification system. Top-level classes are defined by the type of transport and the energy source of transport, while lower-level classes are defined by groups of phylogenetically related proteins that transport similar substrates. As a database of representative sequences, unlike the SLC nomenclature system, the TCDB does not specifically attempt to collect and classify all transporter proteins from a given species. Moreover, it does not inherently take into account orthology between different proteins within a family or subfamily, since it assigns different TCDB numbers to orthologous pairs of proteins. On the other hand, the TCDB provides a broad overview of the vast landscape of transport-related proteins from prokaryotic organisms to humans. Complementary nomenclature systems are important for correctly annotating gene function and for helping researchers navigate among gene families with different functions. In general, nomenclature systems should make it easy to find families of related genes/proteins that are thought to have similar functions (e.g., amino acid transport). To reconcile the SLC and TCDB systems, TCDB name assignment information for SLC families and subfamilies is provided in this review and look-up tables with such information have also been published (7).

A notable difference between the SLC and TCDB classification systems is that while the TCDB assignments may be updated as new information is discovered (172), the SLC gene symbol assignments, which are approved by the HGNC (173) as the official symbols for SLC genes, are expected to remain stable over time, though any information encapsulated in the gene name can still be updated without changing the gene symbol, ensuring consistency across the scientific literature.

Classification of SLCs can be challenging in cases where proteins are distantly related and there is little or conflicting information about their phylogenetic relationship. For example, several proteins of the Major Facilitator Superfamily Domain (MFSD) transporter series (174) were initially classified into the SLC18 family based on the hidden Markov model (HMM) fingerprinting method (7). However, this method relied on the prior classification of proteins into families by TCDB curators. In contrast, multiple sequence alignments and phylogenetic trees of these proteins generated independently of the TCDB failed to support this classification, as SLC18 proteins clustered distinctly from the rest of the proteins. Consequently, these proteins were reclassified as SLC aliases in the new families SLC72 (MFSD1), SLC73 (MFSD6), SLC74 (MFSD8), while MFSD10 has been renamed as SLC75A1.

6. Orphan transporters and SLC-like proteins

Approximately 28% of the annotated SLCs were previously reported to be orphan transporters (175), meaning that their transport function and/or physiological role had not yet been sufficiently elucidated. Our most recent data set (see Table 3) reveals that this proportion has decreased to ~24%.

Among the putative SLC-like proteins reported in our earlier work, 55% turned out to be orphans (7).

Even when a substrate is identified in vitro, it is often unclear whether that represents the physiological substrate, and additional studies are needed to define the true biological function of the protein.

Genetic model systems such as Drosophila melanogaster (176) or Caenorhabditis elegans (177), as well as preclinical animal models including mouse or rat, have proven instrumental for elucidating gene function. However, identifying suitable orthologs in these systems can be challenging due to genetic diversity, as exemplified by the SLCO/SLC21 family (178). In our previous work, we addressed the identification of orthologous SLC-like proteins in seven model organisms, resulting in extensive phylogenetic trees that have been reconciled with a species tree to facilitate ortholog identification (see Supplementary File 1 in (7)). We are optimistic that using model organisms to identify the biologically relevant substrates of these transporters will facilitate the deorphanization of orphan transporters.

In recent years, it has become apparent that several “SLC-like proteins” are not yet represented in the official SLC nomenclature system. To identify these, we turned to sequence databases and annotation systems that are phylogenetically broader and not limited to human proteins, and developed criteria to define “SLC-like” proteins. As a result of this extensive search, a surprising 133 additional human proteins were found that are SLC-like and could potentially be functional SLC transporters, but were not yet part of the official SLC nomenclature (179). Of these, 77 have since been added to the SLC-ome and assigned SLC or SLC alias names. Our current knowledge of the remaining 52 SLC-like proteins is summarized in Section 10.

7. Structural architecture of SLC-ome transporters

It was recognized early on that the membrane-spanning region of secondary transporters typically contains an internal pseudosymmetry, whereby two halves of the membrane-spanning polypeptide region can be superimposed onto each other, and they are typically arranged in an opposite orientation in the membrane bilayer (180). This so-called inverted repeat architecture has been found in a wide variety of secondary active transporters and has become a hallmark of these proteins.

As mentioned in the introductory part, transport-related conformational changes in secondary transporters have linked the inverted repeat architecture to the alternating access mechanism in which substrates are translocated across the membrane as part of the protein cycle between an inward-facing and an outward-facing conformation (181, 182). A characteristic of the alternating access mechanism is the formation of barriers (or gates) on either side of the substrate-binding site, and the inverted repeat regions are ideally suited for this purpose due to their opposite orientation in the membrane. If the location of the substrate remains fixed during the transport process, usually in the center of the membrane-spanning region, it is referred to as a “moving barrier” mechanism. This can be further subdivided into more detailed terms for the mechanism based on whether all structural elements of the transporter move relative to each other (“rocker-switch” model) or the movement is relative to a structural element fixed within the membrane bilayer (“rocking-bundle” model). The substrate-binding site can also be mobile and change location within the membrane-spanning region, which is usually referred to as an “elevator” mechanism, where part of the protein structure containing the substrate-binding site, usually referred to as the “transport” domain, moves relative to a fixed “scaffold” region. In this case, the barrier providing alternating access to the substrate-binding site is typically formed by a “fixed barrier” at the interface of the transport and scaffold domains (183, 184). A detailed description of the structural folds found in SLCs is presented in Section 8 of this review.

Within the membrane-spanning region, short unwound regions lacking a helical secondary structure are often found, which typically make up the substrate-binding site. These breaks can be present in otherwise helical membrane-reentrant or hairpin loops or in fully transmembrane segments, and the exposed backbone atoms typically play a critical role in substrate recognition and binding, especially for charged substrates (184). It has also been suggested that helix dipole moments that thus become oriented directly towards the substrate-binding site contribute to substrate recognition by creating an electrostatically favorable environment for either positively or negatively charged substrates (185).

In terms of transporter proteins, inverted membrane orientation of the two pseudosymmetric halves are not a strict requirement for secondary transport activity, as illustrated by the structural fold of the SWEET transporters (186) that contain the two so-called SemiSWEET protodomains (see “Structure-based classification of SLCs” for further details) in a parallel membrane orientation connected by a membrane-spanning linker helix (187). Vice versa, a structure showing inverted repeat symmetry does not necessarily have alternating-access secondary transport activity, such as in aquaporins (AQPs) or the related GlpF glycerol channel (188), CLC chloride channels (185) and Fluc prokaryotic fluoride channels (189).

The question of predicting the structural fold of human SLCs based on sensitive sequence similarity analyses to proteins with known structure has been previously tackled. As part of our search for additional human SLC-like proteins, 27 different fold families for human SLC proteins have been found (7). Based on this analysis, classical SLC families SLC34, SLC48 and SLC51, and of the more recently included SLC families, SLC56 (sideroflexins) and SLC64 (Golgi Ca²⁺/H⁺ exchangers), are still structural orphans, meaning that no experimentally resolved structure is available for these proteins. Subsequently, the AlphaFold project has given new impetus to structure recognition and created a structural model for all currently known human protein sequences (190). Structures of solute carriers generated by AlphaFold (191) as well as structural predictions from other, similar methods, such as trRosetta (transform-restrained Rosetta) (192) or RaptorX (193) have been reviewed (31). For most structurally orphan SLC families, AlphaFold provides a reasonable predicted structure, one that shows the hallmarks of solute carrier proteins, including a helical bundle that could potentially insert into a membrane bilayer, and in the case of SLC34 proteins, a pair of hairpin loops oriented in an inverted manner, reminiscent of the architecture of several other transporter families (Fig. 2). Interestingly, some of the predicted structures of putative transporter proteins identified in our previous work (7), such as TMEM41-64 proteins, also show a similar inverted repeat hairpin loop architecture (Fig. 2F).

Fig. 2. AlphaFold structure predictions of currently classified SLCs and SLC-like proteins with no experimentally determined structure.

Representative human proteins shown with UniProt accessions in parentheses are (A) SLC34A1 (Q06495), (B) SLC48A1 (Q6P1K1), (C) SLC51A (Q86UW1), (D) SLC55A1/LETM1 (Q6P1Q0) (E) SLC56A1/SFXN1 (Q9H9B4), (F) SLC64A1/TMEM165 (Q9HC07), (G) STARD3 (Q14849), (H) ARV1 (Q9H2C2), (I) TMEM41A (Q96HV5), (J) SIDT1 (Q9NXL6), (K) TMEM245 (Q9H330), (L) LMBR1 (Q8WVP7), (M) LAPTM4A (Q15012), (N) TMEM205 (Q6UW60). If visual analysis found inverted repeat structural elements, those have been colored blue and yellow. Otherwise, predicted transmembrane segments based on UniProt annotations have been colored blue. Larger unstructured loops have been clipped for clarity.

Some SLC families also have unusual transmembrane architectures, such as single-pass transmembrane proteins or proteins that form channel-like solvent-accessible pores. Examples of families harboring a single TMH are SLC3 (see Fig. 9), SLC27/FATP and SLC55/LETM (Fig. 2). SLC3 is a well-known ancillary protein of the SLC7 transmembrane transporters (194), while the transport mechanisms of SLC27 and SLC55 proteins are still not well understood and it is possible that they exert their activity through interactions with other transmembrane proteins. In the case of the SLC27 family, which encodes six FATP fatty acid transport proteins, there is a rather atypical transport mechanism in which the lipids taken up by lipid permeation are trapped by intracellular thioesterification with coenzyme A, for which the FATPs have built-in acyl-CoA synthetase activity on the intracellular side. This prevents exit from the cell, as described in more detail in the description of the SLC27 family in Section 9.

Fig. 2 also shows the AlphaFold structure predictions of the SLC-like proteins STARD3, ARV1, TMEM41A, SIDT1, TMEM245, LMBR1, LAPTM4A, and TMEM205. The properties of these SLC-like proteins are discussed in Section 10.

8. Structure-based classification of SLCs: Most common fold families of the SLC-ome

In this section, we summarize the most common fold families encountered in the currently annotated SLC-ome, including novel families of SLCs (7), and discuss their structural architecture. A circular dendrogram of the SLC-ome, the so-called SLC Atlas, shows the hierarchical clustering of human SLCs and the common fold families to which they belong (Fig. 3). To generate the image in Fig. 3, we first used our previously described HMM (hidden Markov-model) fingerprint-based approach to define clusters of sequences at 0.99 cosine similarity, a vector-based measure of similarity (7). The fingerprints also included similarity values (bit scores) to “pdb70” clusters as used for our previous structural homolog search (7). This technique was useful for separating the heterogeneous set of SLC sequences into clusters that show sequence similarity to a similar set of proteins with known structure, and are thus likely homologous. However, unlike our previous approach, we replaced the dendrograms for each cluster of proteins with proper phylogenetic trees to make them display the predicted evolutionary relationships among family members more accurately. For each cluster, we generated protein sequence alignments using MUSCLE 5.2 (195), followed by building phylogenetic trees using FastTree2 (196, 197). The sequence clusters at this point also included sequences from the TCDB families corresponding to the protein sequences found in each cluster. The phylogenetic trees were midpoint rooted, converted to ultrametric trees using treePL (198, 199), and the subtree featuring all human sequences in the cluster was extracted for each cluster. The final diagram was plotted using custom-made scripts.

Fig. 3. Circular dendrogram showing the hierarchical clustering of human SLC families.

All currently classified SLCs (colored), except SLC3 and SLC27, as well as SLC-like proteins (black), are shown with their assigned gene symbols. SLC3 proteins were omitted since they are ancillary proteins of transporters, whereas SLC27 proteins lack a complex transmembrane domain and instead share higher similarity with acyl-CoA synthetase enzymes (see the SLC27 family description in the text). The dendrogram was generated using a hierarchical clustering of hidden Markov-model (HMM) fingerprints of individual protein sequences as described (7) combined with phylogenetic analysis as described in the text. HMM fingerprints also included similarity to “pdb70” clusters (11), based on which experimentally determined structural folds were assigned to currently classified SLC protein sequences. Structural fold groups are shown as colored outer rings with the names of the structural folds indicated. For more details about individual structural folds, see Fig. 4.

Below, we summarize each individual known structural fold present in the various transport protein families of the SLC-ome. The structural architecture of currently classified SLC proteins is presented in Fig. 4.

Fig. 4. Structural architecture of currently classified SLC proteins.

Experimentally determined structures of representative proteins are shown for each fold family with the fold name indicated. For each protein structure, a single monomer is shown for illustrating the architecture of the polypeptide chain, with repeating structural elements colored blue, yellow, and salmon. For the “Ctr”, “STRA6”, and “CNNM” folds, the homooligomeric state forming the anticipated substrate-binding site is shown, with a single subunit colored blue. For “MgtE”, the two subunits forming the prokaryotic 4U9L structure are fused in homologous human SLC41 proteins. Bound substrates are colored red, stick and sphere representation is used for organic and inorganic substrates, respectively. Coupling ions resolved in the structures are shown in sphere representation in magenta (cations) and green (anions). Black and orange stick representation shows bound lipids and observed sugar moieties, respectively. If the structural fold has representatives only in a single SLC family, that family name is indicated. Protein Data Bank accession codes are indicated in parentheses. The shown structural folds and references to the example structures are the following: MFS – Major Facilitator Superfamily (2), APC/LeuT – Amino acid-Polyamine-organoCation transporters (13), NAT/UraA – Nucleobase/Vitamin C Transporter (14), CDF – Cation Diffusion Facilitators (15), NCX – Na⁺/Ca²⁺ exchangers (16), NHE/NhaA – Na⁺/H⁺ antiporters (17), DAACS/GltPh – Dicarboxylate/Amino Acid:Cation (Na⁺ or H⁺) Symporters (19), DASS/NaDC/AbgT – Divalent Anion:Sodium Symporters (20), MCF/SLC25 – Mitochondrial Carrier Family (21), CNT – Concentrative Nucleoside Transporters (22), MATE – Multidrug And Toxic compound Extrusion transporters (24), NST – Nucleoside-Sugar Transporters (27), PiT – Type III Na⁺/Phosphate cotransporters (28), SWEET – Sugar Will Eventually be Transported family (29), ZIP – Zrt/Irt-like Proteins (30), SLC44 – Choline transporters of the SLC44 family (31), RND – Resistance-Nodulation-Division transporters (32), Ctr – SLC31 Copper transporters (33), MagT – SLC58 Magnesium transporters (34), XPR1 – SLC53 Phosphate carriers (35), STRA6 – SLC69 Vitamin A receptors/transporters (36), UT – SLC14 Urea transporters (37), AmtB – SLC42 Ammonium transporters (38), MgtE – SLC41 Magnesium transporters (39), CNNM – SLC70 Cyclin M Mg²⁺ exporter family (40).

MFS fold – Major Facilitator Superfamily (MFS)

The Major Facilitator Superfamily (MFS) is one of the earliest identified and best characterized transporter superfamilies (200, 201). The first reported crystal structures were those of the H⁺-coupled lactose symporter LacY (202) and the glycerol-3-phosphate-phosphate antiporter GlpT from Escherichia coli (203), and these were also the first structures describing the MFS fold. The currently available MFS structures from both prokaryotes and eukaryotes represent different stages of the transport cycle and illustrate most of the structural elements involved in the transport mechanism, including extracellular and intracellular gates, substrate binding sites, and coupling mechanisms.

MFS fold families:

• SLC2 Facilitative GLUT transporter family

• SLC15 Proton oligopeptide cotransporter family

• SLC16 Monocarboxylate transporter family

• SLC17 Organic anion and vesicular glutamate transporter family

• SLC18 Vesicular amine transporter family

• SLC19 Folate/thiamine transporter family

• SLC21/SLCO Organic anion transporter family

• SLC22 Organic cation/anion/zwitterion transporter family

• SLC29 Facilitative nucleoside transporter family

• SLC33 Acetyl-CoA transporter family

• SLC37 Sugar-phosphate/phosphate exchanger family

• SLC40 Basolateral iron transporter family

• SLC43 Na⁺-independent, system-L-like amino acid transporter family

• SLC45 Putative choline transporter CTL1 family

• SLC46 Folate transporter family

• SLC49 FLVCR-related transporter family

• SLC52 Riboflavin transporter family

• SLC59 Sodium-dependent lysophosphatidylcholine symporter family

• SLC60 Glucose transporters

• SLC61 MFSD5 Molybdate transporter family

• SLC63 Spinster sphingosine-phosphate transporters

• SLC67 Organic cation transporter-like family

• SLC68 Cation symporter family

• SLC71 Putative ammonia transporter family

• SLC72 Lysosomal solute carrier family (MFSD1)

• SLC73 Orphan MFSD6 transporter family

• SLC74 Lysosomal chloride channel family (MFSD8)

• SLC75 Tetracycline transporter-like family (MFSD10)

Members of the SLC2 facilitative GLUT transporter family, representative of MFS architecture transporters, typically have 12 transmembrane helices (TMHs) with intracellular N- and C-termini and 4 inverted TMH trimer repeats, with TMH 1-3 having some sequence similarity to an inverted TMH 4-6 and TMH 7-9 having some sequence similarity to an inverted TMH 10-12 (180, 204). In addition, there is a structural pseudo-symmetry in which TMH1-6 (the N-terminal half) is mirrored by TMH7-12 (the C-terminal half), with the two half-proteins separated by a large cytoplasmic loop between TM6 and TM7 (201, 205, 206). This inverted 6+6 TMH topology provides the basis for the alternating-access mechanisms of the MFS transporters (207). The transport mechanism of the superfamily has been reviewed for the most characterized subfamilies of sugar and drug transporters (208, 209). Interestingly, the substrate range of the superfamily is remarkable, ranging from single ions to trace elements and nutrients to oligopeptides.

LeuT fold – Amino acid-Polyamine-oganoCation superfamily (APC)

After the Major Facilitator Superfamily the Amino acid-Polyamine-organoCation (APC) superfamily is the second largest superfamily of secondary active transporters (210). The APC superfamily includes 8 human families that possess the LeuT fold (211).

LeuT fold families:

• SLC5 Sodium glucose cotransporter family

• SLC6 Sodium- and chloride-dependent neurotransmitter transporter family

• SLC7 Cationic amino acid transporter/glycoprotein-associated family

• SLC11 Proton-coupled metal ion transporter family

• SLC12 Electroneutral cation-coupled Cl^- cotransporter family

• SLC32 Vesicular inhibitory amino acid transporter family

• SLC36 Proton-coupled amino acid transporter family

• SLC38 System A & N, sodium-coupled neutral amino acid transporter family

With the resolution of the 3D structure of the LeuT leucine transporter from Aquifex aeolicus, a structural framework for interpreting structure/function studies of a range of transporters has opened (212). Based on sequence analysis, several other protein families have been shown to be similar to the make-up of the APC superfamily (210). The common transporter core consists of 5+5 TMHs, with the 5-TMH unit repeated in an opposite membrane-spanning orientation. Following the elucidation of the LeuT structure, several other experimentally determined structures of APC transporters have emerged, and the alternating access mechanism has been described in detail. A comprehensive review of these structures and mechanistic transport details has been published (213). Many of the mammalian APC transporters translocate amino acids and monoamine neurotransmitters. In fact, APC transporters have the largest collection of amino acid transporters within the SLC6, SLC7, SLC32, SLC36, and SLC38 families. In addition, transporters of sugars (SLC5 family (214, 215)), divalent metal ions (SLC11 family (216–218)), and cation-coupled chloride cotransporters (SLC12 family (219, 220)) are important representatives of the APC superfamily.

NAT fold – Nucleobase/Ascorbate Transporter (NAT) family, Anion Exchanger (AE) family and Nucleobase Cation Symporter 2 (NCS2) family

In human 3 SLC families harbor the NAT fold (211).

Human NAT fold families:

• SLC4 Bicarbonate transporter family

• SLC23 Na⁺-dependent vitamin C transporter family

• SLC26 Multifunctional anion exchanger family

These transporters share the 7-transmembrane-inverted repeat architecture (211, 221–223) (Fig. 4). Structurally, the helix bundles are divided into a core and gate domain, with the substrate-binding site located in a cleft between the two, however, substrate-binding residues are exclusively located in the core subdomain. While it is thought that the core and gate domains move relative to each other to implement an alternating-access mechanism, the exact details of these movements are unclear (222). The flagship structure of this fold family was that of the prokaryotic uracil/H⁺ symporter UraA (224), but in the subsequent years, the 3D structures of several human members of these families have been resolved (14, 225–228). These include the human AE1 (SLC4A1) Cl^-/HCO₃^- exchanger (14) resolved by X-ray crystallography, and cryo-electron microscopy (cryo-EM)-based structures of the human NBCe1 (SLC4A4) Na⁺-CO₃^2- cotransporter (225) and the rat NDCBE (SLC4A8) Na⁺-CO₃^2-/Cl^- exchanger.

In addition, several structures of the mouse proteins pendrin (SLC26A5) (229) and vitamin C transporter SVCT1 (SLC23A1) (230) are available.

CDF – Cation Diffusion Facilitator (CDF) family, a member of the CDF superfamily

Human CDF fold family:

• SLC30 Zinc efflux transporter family

Originally, the CDF family was identified as a transport protein family specific for heavy metal ions (231). The X-ray structure of the YiiP transporter that catalyzes Zn²⁺/H⁺ exchange across the inner membrane of E. coli in complex with Zn²⁺ has been reported at 3.8 Å resolution (232). This success was followed by the high-resolution structures of the human SLC30 zinc transporters ZnT8 (SLC30A8) (15) (Fig. 4) and ZnT7 (SLC30A7) (233). YiiP is a homodimer held together in parallel orientation by four Zn²⁺ ions at the interface of the cytoplasmic domains. Most members of the CDF family possess six putative TMHs with N- and C-termini on the cytoplasmic side of the membrane.

The CDF proteins are a conserved family of divalent transition metal ion transporters that act as exporters of divalent metal ions such as Zn²⁺, Fe²⁺, Mn²⁺ or Cd²⁺ from the cytoplasm to either the extracellular environment or intracellular compartments. They contain a conserved transmembrane core structure formed by 6 TMHs, and also often harbor a regulatory cytoplasmic C-terminal domain (234, 235). In the transmembrane region, a single divalent metal ion binding site is found, which is thought to correspond to the active site of transport. The C-terminal domain is thought to be involved in transport, as observed in the available structures, and contains further Zn²⁺-binding sites that likely have regulatory and structural roles (15, 232, 236, 237). Based on the comparison of inward-facing and outward-facing conformations, the transport cycle was proposed to involve a scissoring motion of the transmembrane domain of the two monomers relative to each other, as well as more localized tilting and bending of a four-TMH bundle that controls access to the metal-binding site (236).

NCX fold – Ca²⁺:Cation Antiporter (CaCA) family, a member of the CDF superfamily

The NCX plasma membrane Na⁺/Ca²⁺ exchangers represent a large family of proteins that mediate the entry and exit of Ca²⁺ into and out of cells and thus modulate Ca²⁺ signaling and homeostasis in biological systems ranging from bacteria to humans.

Human NCX fold families:

• SLC8 Na⁺/Ca²⁺ exchanger family

• SLC24 Na⁺/(Ca²⁺-K⁺) exchanger family

Currently, only prokaryotic and archaeal structures are available for proteins containing the NCX fold (238–241). The structural architecture of NCX proteins consists of a repeated 5-TMH unit comprising a total of 10 TMHs, which are connected by a long linker sequence in eukaryotic NCX proteins, harboring several regulatory domains (242). Molecular dynamics simulations have been used to study the details of ion recognition and conformational changes during the transport cycle, and the results have proposed a transport mechanism that does not fit into the currently known rocker-switch, rocking bundle or elevator types (16, 243).

NhaA fold – The Monovalent Cation:Proton Antiporter CPA1 and CPA2 families

CPA1 is a large family of transporters found in all kingdoms of life (244). In plants, for example, CPA1 transporters function as Na⁺/H⁺ or K⁺/H⁺ antiporters, contributing to salt tolerance and K⁺ homeostasis in cellular organelles and to cellular Na⁺ extrusion (245). The CPA1 family includes the ubiquitous and pharmacologically important NHEs of the SLC9 family.

CPA2 is a moderately large family represented in bacteria, archaea and eukaryotes. While CPA1 includes the electroneutral antiporters, e.g., human NHEs, CPA2 includes the electrogenic antiporters, e.g., E. coli NhaA (246).

Human NhaA-fold SLC families:

• SLC9 Na⁺/H⁺ exchanger family

• SLC10 Na⁺/bile salt cotransporter family

• GPR155 (G-protein coupled receptor 155). This family is included here even though it is not currently assigned to the SLC nomenclature, as it has been identified as an SLC-like transporter (see Section 10, SLC-Like Proteins).

The crystal structure of NhaA, the major Na⁺/H⁺ antiporter of E. coli, revealed a unique NhaA structural fold shared by prokaryotic and eukaryotic membrane proteins (247–249). Of the 12 NhaA TMHs, TMHs 3-5 and 10-12 are topologically inverted repeats. In addition, TMHs 4 and 11 are unwound and cross each other in the mid-membrane to form an X-shape characteristic of the NhaA fold (250). Analysis of the outward- and inward-facing structures, as well as hydrogen-deuterium exchange mass spectrometry, have suggested an elevator-type mechanism, in which the core domain moves relative to the dimerization domain to induce transport, with the latter remaining relatively immobile during the process (251–254).

For SLC10 proteins, the structure of human NTCP (SLC10A1) has been experimentally resolved (255), and it also has a “crossover” region between TMH3 and TMH8, a hallmark of NhaA/SLC9 Na⁺/H⁺ antiporters (247) and prokaryotic bile acid transporter homologs (256, 257). Bile acid transporters of the SLC10 family also likely translocate their substrates using an elevator mechanism (257).

SLC9D1 (TMCO3) is a novel member of the monovalent cation:proton antiporter 2 (CPA2) family (TC 2.A.37; Pfam: Na_H_Exchanger) (7), which functions as a putative K⁺/H⁺ antiporter at the Golgi apparatus (258). The cryo-EM structure of the E. coli SLC9D1 homolog K⁺/H⁺ exchanger KefC (glutathione-gated K⁺ efflux transporter) was subsequently determined (259). KefC forms a homodimer similar to the inward-facing conformation of the Na⁺/H⁺ antiporter NapA. The KefC monomer consists of 13 TMHs with an extracellular N-terminus and an intracellular C-terminus. The KefC structure was more similar to the structure of the NapA bacterial Na⁺/H⁺ exchanger of the CPA2 family (244) than to NhaA from E. coli (260).

GPR155, also known as LYCHOS (LYsosomal CHOlesterol Signaling), is involved in the sensing of lysosomal cholesterol by binding to cholesterol in the N-terminal permease-like region (261, 262) (see Fig. 45 in the SLC38A9 description, the description of the SLC65A1 cholesterol transporter, and the description of GPR155 in Section 10). At high levels of lysosomal cholesterol, it recruits the mammalian target of rapamycin complex 1 (mTORC1) to lysosomes to promote anabolic metabolism.

The first 10 TMHs of GPR155/LYCHOS show a 5+5 TMH arrangement where the second half of the transporter domain (TMHs 6-10) shows similarity to the N-terminal half of the Na⁺/bile transporters of ASBT (SLC10A2) (7). The remaining seven transmembrane helices (TMHs 11-17) of GPR155 share motifs with adhesion-like G-protein-coupled receptors (GPCRs), and the C-terminus includes a DEP (Dishevelled, Egl-10 and Pleckstrin) domain, which is common in signaling proteins like RGS (Regulator of G-protein Signaling) proteins (262, 263).

DAACS fold – The Dicarboxylate/Amino Acid:Cation (Na⁺ or H⁺) Symporter (DAACS) family (also known as DAACS superfamily)

This family includes transporters that mediate the symport of dicarboxylates and amino acids across cell membranes, typically coupled with the cotransport of sodium Na⁺ or H⁺.

Human DAACS fold family:

• SLC1 family of glutamate and neutral amino acid transporters

Members of the SLC1 family adopt the DAACS structural fold – also known as the GltPh fold – named after the archaeal SLC1 homolog GltPh from Pyrococcus horikoshii, whose crystal structure provided the first detailed insight into this fold (264). The crystal structure of GltPh reveals that the transporter assembles as a trimer in which each monomer is an independent functional unit capable of substrate permeation (265–267).

Membrane transporters of the SLC1 family contain a scaffold (or trimerization) domain that embraces a transport domain showing an inverted repeat architecture with two reentrant helical hairpin loops (HP1 and HP2) in each repeat (264) (Fig. 4). The transport mechanism has been identified as elevator-type through various experimentally determined structures of homologous transporters along different points in the transport process (268). The characteristic of an elevator-type mechanism is the rigid-body sliding of the transport domain carrying the substrate(s) over the static scaffold domain (269, 270). Structure-function studies and molecular dynamics analysis revealed a critical role of the serine residue S364 of the human GLT1/EAAT2 (SLC1A2) reentrant hairpin loop HP1, and more specifically its hydroxyl group, in the coupling of sodium and substrate fluxes (271, 272).

The SLC1 structure has been shown to harbor an anion (i.e., chloride) conductance (see the description of SLC1A7/EAAT5 in Section 9 for details). This conductance is enabled by a pore that is proposed to open only during substrate translocation and is distinct from the substrate translocation pathway (268, 273).

The structures of a thermostabilized human EAAT1/GLAST1 (SLC1A3) variant and human ASCT2 (SLC1A5) have also become available (19, 270, 274, 275).

DASS/AbgT folds – The Divalent Anion:Na⁺ Symporter (DASS) family (also known as the NaDC or AbgT family; member of the Ion Transporter (IT) superfamily)

The DASS family contains both Na⁺-driven anion cotransporters and anion/anion exchangers. The family belongs to a broader ion transporter (IT) superfamily, which comprises 24 families of transporters, including those of the prokaryotic p-aminobenzoyl-glutamate (AbgT) antimicrobial resistance transporter (276–278). The human proteins in the DASS family play major physiological roles and are drug targets.

Human DASS fold family:

• SLC13 Na⁺-sulfate/carboxylate cotransporter family

The hallmark the DASS structure is the structure from the homologous VcINDY protein (TC 2.A.47.5.2), an Na⁺-dependent dicarboxylate transporter that imports TCA cycle intermediates across the inner membrane of Vibrio cholerae. The protein forms a homodimer and each protomer consists of 11 TMHs, which form a scaffolding domain (TMHs 1-4 and 7-9) and a transport domain (TMHs 5, 6, 10, 11 and two hairpins), the latter containing the substrate-binding site (279, 280). The structure shows an inverted symmetry between TMHs 2-6 and 7-11 (277, 279, 280). The identification of the outward-open state confirms an elevator-type transport mechanism (277, 280, 281).

Well-characterized structural representatives of the prokaryotic AbgT transporter family (TCDB 2.A.68) include YdaH from Alcanivorax borkumensis (282) and MtrF from Neisseria gonorrhoeae (283), both of which share a similar fold and oligomeric architecture. These antibiotic efflux transporters mediate bacterial resistance to antimetabolite drugs and thus represent important drug discovery targets for the development of novel antibiotics to combat bacterial infections (276). Both transporters use the H⁺-motive force and function as H⁺ antiporters, although YdaH can bind Na⁺ as well (276, 282–284). Like VcINDY, both MtrF and YdaH exist as homodimers, with the scaffold domains forming the dimer interface. The structures of AbgT proteins were shown to be very similar to that of VcINDY (284). This supports the notion that the VcINDY fold is representative of the entire IT superfamily.

MCF fold – Mitochondrial Carrier Family

Members of this family are found exclusively in eukaryotic organelles, predominantly in the inner membrane of mitochondria, and are encoded in the nucleus. Mitochondrial carriers differ in their mode of transport, with most mediating substrate exchange, some being H⁺ symporters, and a few being unidirectional transporters (285).

MCF fold family:

• SLC25 Mitochondrial carrier family

The general structural design of mitochondrial carriers is that they contain 3 homologous pseudo-symmetric repeats of 2-TMH segments, each having about 100 amino acid residues, with N- and C-termini facing the intermembrane space, giving rise to a total of 6 TMHs (286). In addition, a highly conserved signature motif, Px[DE]xx[KR], is found in the odd-numbered α-helices in these transporters resulting in pronounced kinks due to the proline residues, giving them an L-shape, which helps to block access to the central cavity from the mitochondrial matrix in the cytoplasmic-open state (286). Available structures suggest an alternating-access mechanism in which two distinct hydrogen-bond and salt-bridge networks on the cytosolic and matrix sides of the inner mitochondrial membrane alternately break and reform, driving the carrier between outward-open and inward-open conformations (287, 288).

The monomeric or dimeric state of mitochondrial carriers has been a matter debate as it has been believed that they exist as homodimers and transport substrates with a sequential kinetic mechanism, forming a ternary complex where both exchanged substrates are bound simultaneously. However, this has been clarified showing they are monomers, except for dimeric aspartate/glutamate carrier AGC12 (SLC25A13), and operate with a ping-pong kinetic mechanism in which the substrate import and export steps occur consecutively, consistent with a common transport mechanism in which a single central substrate-binding site is alternately accessible (289). As already noted, AGC12 (SLC25A13) is an exception, as it is unique, having an additional N-terminal domain involved in the dimerization of the protomers (290).

Cardiolipin molecules form an integral part of the structure of SLC25 carriers and even though they are not required for function, they have been proposed to help protect dynamic regions of the protein (291). Three cardiolipin molecules are tightly bound to mitochondrial carriers and are important for their stability and function (292).

Interestingly, the SLC25 members MTCH1 (SLC25A49) and MTCH2 (SLC25A50) lack a single TMH of the 6-TMH bundle of the MCF architecture. These proteins have been shown to function as insertases (293) (see also the SLC25 family description in Section 9), and have been suggested to have lipid scramblase activity (294). AlphaFold2-based structural models of human MTCH1 and MTCH2 suggest that the missing TMH opens a lateral gap in the protein structure, which is thought to form the exit pathway of the inserted peptide, as well as the site of lipid scramblase activity (294).

CNT fold – The Concentrative Nucleoside Transporter (CNT) Family (or superfamily)

Members of the CNT family have been identified in bacteria (e.g., E. coli NupC permease; TC 2.A.41.1.1), in which case they are energized by H⁺ symport, and in mammals (SLC28 family members), in which case they are energized by Na⁺ symport (295, 296).

CNT fold family:

• SLC28 Na⁺-coupled nucleoside transporter family

The core transporter structure is composed of 8 TMHs, two reentrant hairpin loops (HP1 and HP2), and three interfacial helices (IH1-3) (297). These can be divided into a scaffold domain, consisting of TMHs 1-3, IH1 (interfacial helix 1), and TMH6, and a transport domain, which is in turn composed of two pseudo-symmetric structural groups that are oriented oppositely in the membrane (297). Additionally, the human CNT proteins encoded by SLC28A1-3 contain 3 more TMHs at their N-termini, as is visible in the subsequently resolved hCNT3 (SLC28A3) structure (298). The available bacterial structures have shown that CNT transporters operate according to an elevator mechanism, where the transport domain is displaced relative to the scaffold domain during the transport cycle (297, 299).

MATE fold – The Multidrug/Oligosaccharidyl-lipid/Polysaccharide (MOP) Flippase Superfamily

The Multidrug and Toxic compound Extrusion (MATE) family comprises a large group of secondary active transporters found in both prokaryotes and eukaryotes and belongs to the Multidrug/Oligosaccharidyl-lipid/Polysaccharide (MOP) flippase superfamily. In bacteria, MATE transporters contribute to multidrug resistance by harnessing electrochemical gradients of H⁺ or Na⁺ to export diverse cationic compounds, including toxic dyes such as ethidium bromide, tetraphenylphosphonium (TPP⁺), berberine, acriflavine, and norfloxacin. This efflux activity reduces intracellular accumulation and confers resistance. In humans (see the SLC47 family in Section 9), MATE transporters are primarily expressed in the kidney and liver, where they mediate the export of a broad range of mostly cationic xenobiotics and endogenous compounds. The MATE family is also closely related to the Drug/Metabolite Transporter (DMT) superfamily (see below) (278).

Human MATE fold families:

• SLC47 Multidrug and Toxin Extrusion (MATE) family

• SLC62 ATP exporter family

• SLC76 Glycolipid translocator (RFT1) family

The SLC47 multidrug and toxin extrusion (MATE) family was originally thought to belong to the MFS superfamily, but structural evidence later showed that they have a distinct structural architecture (300–303). The MATE fold consists of 12 TMHs organized in two 6-TMH bundles (TMH1-6 and TMH7-12) that are related by rotational symmetry along an axis perpendicular to the membrane (302). MATE transporters have been captured in both outward-open and inward-open conformations, leading to the speculation that the alternating-access transport is implemented by a rocker-switch mechanism similar to MFS transporters (302). However, a number of controversies still exist, in particular about the structure of TMH1 and its role in the transport mechanism, how exactly bound ions facilitate conformational changes during the antiport cycle, and what role lipids play in stabilizing individual protein conformations or lowering the energy barrier for transport-related conformational transitions (302).

The ATP exporter SLC62A1 also carries the MATE fold and has 12 TMHs (7, 304, 305).

Likewise, RFT1 (SLC76A1), also known as Man₅GlcNAc₂-PP-dolichol translocation protein, harbors the MATE fold (7). It is an ER protein with 14 TMHs (306).

NST fold - The Drug/Metabolite Transporter (DMT) Superfamily

The DMT superfamily is a large group of membrane transporters from eukaryotes, bacteria and archaea, comprising over 32 families whose members are involved in the export of a wide range of substrates, including drugs and metabolites (307).

Human NST fold families:

• SLC35 Nucleoside-sugar transporter family

• SLC57 NIPA-like magnesium transporter family

The nucleotide sugar transporters of the SLC35 family are part of the Drug/Metabolite Transporter (DMT) superfamily (308). DMT superfamily members share a similar structural architecture that is composed of 10 TMHs, which can mediate the transmembrane transport of a variety of substrates, such as amino acids, sugar-phosphates, or nucleotide-sugars (307, 309–311). DMT proteins also show an inverted repeat architecture consisting of two 5-TMH bundles that are inserted in the membrane in the opposite orientation. A model of the cytoplasmic-facing state based on the experimentally determined luminal conformation has hinted at a possible alternating-access mechanism of transport (310). Based on these structures, nucleotide and sugar recognition within the binding site seems to happen in two different pockets, hinting at structural determinants of substrate specificity (311).

Interestingly, NIPA-type Mg²⁺ transporters, which have been highlighted in the recent search for SLC-like proteins, show remote sequence similarity to DMT members and thus likely share the same structural fold (7). Additionally, this search also revealed two additional human orphan transporters, TMEM144 and TMEM234, that show remote similarity to SLC35 members (7, 191) (see SLC-like transporters in Section 10).

PiT fold – The Inorganic Phosphate Transporter (PiT) family

Members of the PiT family are inorganic phosphate transporters derived from bacteria, archaea, and eukaryotes. They use either Na⁺ or H⁺ gradients to transport P_i (312).

Human PiT fold family:

• SLC20 Type III Na⁺-phosphate cotransporter family

The structure of a homolog of human SLC20 transporters from the hyperthermophilic bacterium Thermotoga maritima has been solved (28). The structure shows a distinct architecture with a transporter core containing 5+5 TMHs in an inverted repeat arrangement, each containing a membrane-reentrant hairpin loop, similar to SLC1 or SLC13 transporters (28). The structure also contains two additional TM helices, which have been termed the scaffold domain, responsible for forming a dimerization interface (28). This structural arrangement could support an elevator-type transport mechanism based on analogy to SLC1 or SLC13 proteins. However, the exact details of transport-related conformational changes in the protein remain to be elucidated (28).

SWEET and SemiSWEET folds – The Sweet; PQ-loop; Saliva; MtN3 (Sweet) family (a member of the Transporter-Opsin-G protein-coupled receptor (TOG) Superfamily)

Transporters of the SWEET (“Sugars Will Eventually Be Exported Transporter”) family, also known as the PQ-loop, Saliva, or MtN3 family (TC# 2.A.123), are sugar exporters belonging to the Transporter-Opsin-G protein-coupled receptor (TOG) superfamily.

Human SWEET fold families:

• SLC50 Sugar efflux transporter family

• SLC66 PQ-loop amino acid transporter family

The first gene of the SWEET family was identified as MtN3 in Medicago truncatula, which is involved in Rhizobium-induced nodule development (313). The SWEET fold is also known as the MtN3 structural fold (191, 314). The SWEETs are abundant in plants, but also present in the animal kingdom, as well as in humans (315–317).

SWEET proteins possess a characteristic MtN3/Saliva domain, also called the PQ-loop repeat, which is constituted by three TMHs. Eukaryotic SWEETs consist of a tandem repeat of the 3 basic TMH unit separated by a single TMH linker, forming a 3+1+3 TMH structure (316, 318). The SWEET transport cycle is thought to be initiated by the alternate opening of either the TMH1-2 and TMH5-6 hairpins on the cytoplasmic side or the TMH2-3 and TMH6-7 hairpins on the periplasmic/extracellular side (318).

PQ-loop amino acid transporters (SLC66 family) and KDEL receptors are also related to SWEET transporters and share a similar fold (319–321).

Human SemiSWEET fold family:

• SLC54 MPC mitochondrial pyruvate carrier family

In addition to SWEET transporters, “half-transporters” harboring only the 3-TMH bundle structural element also exist, known as SemiSWEETs. Bacterial SemiSWEET transporters function as dimers (316, 322, 323). In human, the mitochondrial pyruvate carrier (SLC54 family) is a representative of SemiSWEET transporters, thus being distantly related to SWEET transporters (288). SemiSWEETs are among the smallest known transporters.

ZIP fold – The Zinc (Zn²⁺)-Iron (Fe²⁺) Permease (ZIP) Family (or superfamily)

The ZIP (Zrt/Irt-like Protein) / SLC39 family members are divalent metal ion transporters mostly mediating the uptake of Zn²⁺, Fe²⁺ and Mn²⁺ (324).

Human ZIP-fold family:

• SLC39 Metal ion transporter family

The structure of a prokaryotic homolog has confirmed previous predictions that the transporter core consists of 8 TMHs (30, 324). The first 4 TMHs are symmetrically related to the last 4 TMHs by a pseudo-twofold axis running parallel to the membrane. Interestingly, the structure seems to contain a binuclear substrate-binding site that is able to house two divalent metal ions (30). The structure of the apo state of the same transporter hinted at an alternating-access mechanism where two bundles (TMHs 1, 4, 5, 6, and TMHs 2, 3, 7, 8) move as rigid bodies relative to each other upon substrate binding (325). Overall, the two bundles are reminiscent of a scaffold and transport domains of elevator-type transporters, and a computationally generated model of the outward-open state has subsequently confirmed that this transport mechanism is plausible for ZIP proteins (325, 326). Extensive structural modeling efforts for the human SLC39 family have suggested structural elements with putative mechanistic roles, such as intracellular and extracellular gates of the binding site, based on comparison to the prokaryotic structure (327).

SLC44 fold – The Choline Transporter-like (CTL)

The CTL family includes several eukaryotic choline transporter-like proteins, while no prokaryotic homologues have been found (328).

Human SLC44 fold family:

• SLC44 Choline-like transporter family

Choline transporters of the SLC44 family have long been considered as structural orphans, until the AlphaFold initiative has provided a structural model of human SLC44A1 using ab initio prediction. The structure corresponded to a yet unknown novel fold that has been subsequently confirmed experimentally (31). In the cryo-EM density map, while densities interpreted as putative cholesterol-binding pockets have been found, the structure has not hinted at a possible transport mechanism (31).

RND fold – The Resistance-Nodulation-Cell Division (RND) superfamily

Human RND fold family:

• SLC65 NPC (Niemann-Pick type C) cholesterol transporter family

RND superfamily transporters are best known as active efflux pumps in Gram-negative bacteria for a wide range of compounds, including toxins and antibiotics. They are also present in organisms ranging from archaea to eukaryotes, where they have diverse functional roles. An important eukaryotic branch of sterol transporters includes the Niemann-Pick type C1 protein NPC1/SLC65A1. Related proteins are the developmental hedgehog (Hh) signaling proteins PTCH1 and PTCH2 (see SLC65 family description in Section 9). RND transporters typically contain 12 TMHs and two large external loops between TMHs 1 and 2 and between 7 and 8 (329) (Fig. 4).

Ctr fold – Copper transporters

Human Ctr fold family:

• SLC31 Copper transporter family

The structure of a vertebrate Ctr homolog from Atlantic salmon (Salmo salar) has been resolved to a near-atomic resolution using the insertion of the BRIL crystallization chaperone between TMH1 and TMH2 (33). The membrane-spanning region of the structure showed a good overlap with previously reported low-resolution 2D electron crystallographic reconstruction of human Ctr1/SLC31A1, also confirming a trimeric assembly (33, 330, 331). The three protomers form a central pore structure where two bound Cu⁺ ions were identified at two distinct locations within the membrane-spanning region. These substrate ions are coordinated by Cu-S bonds with Met residues in a trigonal planar geometry, likely also providing selectivity against Cu²⁺ and other ions. Since Cu⁺ binding seemed to induce an instability in the C-terminal region of the protein, the authors suggest that this region might act as a flexible intracellular gate (33). It is also notable that the conduction rate of Ctr transporters is extremely slow, and therefore the exact mechanism of transport remains unresolved (33). See further details in the description of the SLC31 family in Section 9.

MagT fold – The Magnesium Transporter 1 (MagT1) Family

While MagT1 (SLC58A1) and TUSC3 (SLC58A2) have been proposed to be Mg²⁺ transporters (66), they are also part of the OST-B glycosylation complex (34, 332). Specifically, MagT1 and TUSC3 are the human orthologues of the OST3 and OST6 oligosaccharyltransferases in S. cerevisiae (332, 333).

Human MagT fold family:

• SLC58 MagT-like magnesium transporter family

Recent studies have elucidated the structure of the oligosaccharyltransferase complexes OST-A and OST-B, and thus parts of the structures of human MagT1 and another of its paralogs, DC2/OSTC (34). The structures of MagT1 and DC2/OSTC show a high degree of similarity, with 3 of the 4 TMHs organized in a linear manner (34). In contrast to the initial proposal that they function as Mg²⁺ transporters, the reported structures do not reveal any transporter-like or channel-like transport mechanism. Thus, it is possible that these proteins have a regulatory role or an indirect effect on Mg²⁺ homeostasis.

XPR1 fold – Phosphate transporter family

Human XPR1 fold family:

• SLC53 XPR1 phosphate exporter family

Several structures of the human XPR1 (SLC53A1) protein have appeared that have been resolved with cryogenic EM as either the apo protein or with various positions of the bound P_i substrate (35, 334–336)]. The transmembrane region showed a 10-TMH architecture that was not similar to any previously known membrane protein structures. The transmembrane structure can be divided in various ways into an N-terminal or scaffold domain with 3-5 TMHs, and an EXS (homologous regions found in yeast ERD1, SYG1, and human XPR1), also called channel or core domain with 5-7 TMHs (35, 334–336). These two regions are not structurally similar to each other, and thus there is no internal repeat symmetry within the transmembrane region of the protein structure. Interestingly, the C-terminal TMHs have been suggested to be similar to a prokaryotic chloride-pumping rhodopsin (ClR) (336). The scaffold or N-terminal domain is responsible for the dimerization interface observed with human XPR1, where lipids have also been indicated to play a role (35, 336).

Within the transmembrane region, a channel-like pore was observed between the N-terminal and C-terminal domains that can potentially conduct P_i ions. Comparison of the apo and P_i-bound states as well as MD simulations suggested that there is no significant conformational change between P_i-bound and unbound states, indicating a channel-like transport mechanism (35). Electrophysiology experiments have also confirmed an unsaturable channel-like conductance (335). Interestingly, inositol phosphates and pyrophosphates, especially InsP₈ (1,5-bis-diphosphoinositol 2,3,4,6-tetrakisphosphate), are required for channel activity of XPR1, and cause gating-related rearrangements in the transmembrane region when bound to the intracellular SPX domain (335, 337) (see also the SLC53 family description in Section 9).

STRA6 fold – Retinol transporter family

Human STRA6 fold family:

• SLC69 Vitamin A receptor/transporter family

A single structure describing almost the entire zebrafish stra6 (SLC69A1) protein in complex with calcium-calmodulin (CaM) has been resolved using cryo-EM (36). STRA6 forms an intricately associated symmetric dimer consisting of 9 TMHs in each protomer, as well as an intriguing horizontal intramembrane helix pair that run parallel to the membrane (36). There is no internal repeat symmetry observed within the protomers. An arch-like structure protruding into the extracellular milieu as an extension of transmembrane helices forms a putative RBP-binding site (36) (see also the SLC69 family description in Section 9). The STRA6 structure is bound by two CaM molecules, which was unanticipated given that there is currently no evidence to link retinol transport with cellular Ca²⁺ homeostasis (36).

AmtB fold – Ammonia/Urea transporter family

Human AmtB fold families:

• SLC14 Urea transporter family

• SLC42 Rh type glycoprotein family of ammonium transporters

The AmtB structural fold family in humans comprises the SLC14 (urea transporter) and SLC42 (ammonium transporter) solute carrier families. Transporters of this fold family form trimeric assemblies, where within each protomer, a pseudo-twofold symmetry axis can be observed. This confers an inverted repeat architecture, with each repeat containing 5 TMHs (62, 64, 338, 339). Despite the overall similar architecture of their transmembrane regions, these two families also have functionally important distinct structural features, and therefore, will be discussed separately, and are also shown individually in Fig. 4 as “AmtB/SLC42” and “UT/SLC14” structural folds.

Structures of human UT-A (SLC14A2) and human and bovine UT-B (SLC14A1) as well as the prokaryotic homolog dvUT all show a membrane-spanning pore with specific conserved selectivity sites (338–341). An important distinction from ammonium transporters of the same fold family is that urea transporter structures feature a distinct break in TMHs 1 and their symmetric partner 6, with the N-terminal halves of these helices tilted at a 45° angle relative to the membrane plane, exposing their carbonyl oxygens to the channel pore. These atoms form part of the selectivity filter sites and together with the negative half of the oriented helix dipole, help stabilize bound urea substrate molecules (338, 339). MD simulations of the urea-bound structure confirmed a possible channel-like transport mechanism with little conformational changes in the protein upon urea translocation (339). One of the sections of the selectivity filter, called “S_m”, located in the center of the membrane-spanning region, contains highly conserved threonine residues. These have been suggested to confer urea selectivity and repel charged species such as protons, ammonium or guanidinium by helping in the formation of a hydrophobic constriction that does not permit a hydration sphere (339). See also the description of the SLC14 family in Section 9.

Interestingly, phosphatidylinositol (PI) lipids have been found bound to the human urea transporter structures from the side of the membrane interface (340), which might play a role in stabilizing the trimeric structure or the individual protomers.

First structures of the prokaryotic homolog AmtB have highlighted features of channel-like conductance in the protein, such as a pore with a hydrophobic filter region, a selectivity filter, and bound substrates (62). While similar structural elements are also present in human RhCG (RHCG/SLC42A3), there are notable differences in its conduction mechanism (64). Prokaryotic homologs typically transport the uncharged ammonia (NH₃) species, while RhCG has to take up ammonia from the urinary duct, where the protonated ammonium (NH₄⁺) species predominates (64). It has been suggested based on molecular dynamics and quantum mechanical calculations, that RhCG likely recruits NH₄⁺ and deprotonates it using a His sidechain, after which the resulting NH₃ is transported across the channel (342). For details on the physiological role of SLC42, see the description of the SLC42 family in Section 9.

MgtE fold – MgtE-like magnesium transporters

Human MgtE family:

• SLC41 MgtE-like magnesium transporter family

A prokaryotic homolog of SLC41 proteins, MgtE from Thermus thermophilus was the first representative of this structural family with a resolved structure (343). The overall architecture of the transmembrane region of the protein revealed a symmetric dimeric assembly with 5 TMHs in each protomer (343). The human SLC41 proteins, due to their sizes, likely contain both protomers on the same polypeptide chain. Structures of the T. thermophilus MgtE also show a regulatory N and CBS domains on the cytoplasmic side of the membrane and the N-terminus of the polypeptide chain, which seem to be absent from human SLC41 members. Interestingly, despite extensive structure determination efforts with and without these regulatory domains, as well as various substrates and the apo form, an open channel conformation has not yet been identified (39, 61, 343, 344). Based on available structures, either a cytoplasmic-open or a periplasmic-open conformation can be observed, which is reminiscent of an alternative access mechanism. Curiously, however, the T. thermophilus MgtE channel was reported to have a high Mg²⁺ conductivity (96 pS) (39), which leaves the question of its conducting mechanism still unresolved. More details about the ion selectivity and physiological roles of SLC41 proteins can be under their family description in Section 9.

CNNM fold – CorB/CorC-like magnesium transporters

Human CNNM family:

• SLC70 Cyclin M Mg²⁺ exporter family

The structures of the prokaryotic homologs CorB and CorC of CNNM members have been reported by two independent groups (345, 346). Both proteins contain soluble regulatory domains on the cytoplasmic side (CBS and CorC/HlyC domains), of which the cyclin and cystathionine β-synthase (CBS) domain is also shared by their eukaryotic homologs, which also contain a CNBH (cyclic-nucleotide binding homology) domain on the cytoplasmic C-terminus and an extended N-terminal region (347). The transmembrane region of these proteins shows a symmetric dimeric arrangement with 3 TMHs in each protomer (345, 346). Both structures show an inward-open conformation with one Mg²⁺ ion bound to each protomer (345, 346). Interestingly, in contrast to SLC41 members, the Mg²⁺ ions seem to be directly coordinated by protein residues and are lacking a hydration shell (see description of the SLC41 family in Section 9). While these structures have not shed light on the transport mechanism itself, they have shown that removal of the bound Mg²⁺ causes significant rearrangements in the cytoplasmic ATP-binding CBS domain (346), and that ATP binding is important for Mg²⁺ transport activity (345). A conserved π-helical turn in TMH3 is observable in both structures (346), which might act as a hinge region for possible transport-related conformational rearrangements.

9. Description of the individual SLC families from SLC1 to SLC76

This section privides detailed descriptions of all 76 human SLC solute carrier families and their members, with added emphasis on less well-characterized transporters and orphan transporters. The following conventions, resources, and terminology are applied throughout this section to ensure consistency and comparability:

• TCDB assignment numbers, Pfam family names and structural folds: For each family, the corresponding TCDB assignment numbers, Pfam family names and structural folds are listed in parentheses in the title.

• Family discovery: Each SLC family description begins with a Discovery paragraph, outlining how the family was identified and how its founding member was discovered.

• Phylogenetic analysis: For SLC families with more than three members, unrooted phylogenetic trees are provided. Trees were generated as described previously (7), visualized using the Interactive Tree of Life (iTOL) server (348), and manually annotated. The branch length scale in the phylogenetic figures indicates the average number of substitutions per amino acid position and can be interpreted as a measure of phylogenetic distance.

• Gene symbols and species notation: Approved SLC symbols for genes encoding membrane transport proteins are shown in italics, capitalized for humans (e.g., SLC26A4), and case-sensitive for experimental animal models as follows:

  • -Rodents (mouse, rat): only the first letter is capitalized, remainder lowercase (e.g., Slc26a4)
  • -Fish and frog: all lowercase
  • -Dog: all uppercase

Alias SLC symbols are listed in standard (non-italic) script.

• Pseudogenes: This review focuses on human SLCs and human SLC unprocessed pseudogenes (11 in total). Of these, 8 are unitary pseudogenes with coding orthologs in another species, while 3 were initially classed as protein-coding. Other types of human SLC pseudogenes are not included.

• Protein names: Protein names of SLCs are listed as they appear in the web-based SLC tables and/or as they are commonly used in the literature. Where SLC names themselves are used to designate transporter proteins, particularly in families where protein nomenclature is less well established, they are presented in non-italic characters.

• Expression data: Some SLC family descriptions include information on tissue distribution and cellular localization, derived from the Human Protein Atlas (HPA; https://www.proteinatlas.org/). These data were generated as previously described (349). While the HPA is a valuable resource, its data should be interpreted with caution, as the data has not undergone formal peer reviewed. All citations to the Human Protein Atlas in this review are abbreviated as HPA.

• Terminology:

  • -Homologous genes separated by speciation are referred to as orthologs.
  • -Homologous genes separated by duplication events are referred to as paralogs.
  • -Protein isoforms (or variants) denote products of alternative splicing or other post-transcriptional modifications of a single gene.

• Additional resources: Further information on individual SLC family members can be found at the following online resources:

  • -genenames.org (https://www.genenames.org/data/genegroup/#!/group/752)
  • -bioparadigms.org (https://www.bioparadigms.org/slc/)
  • -Resolute Knowledgebase (https://re-solute.eu/knowledgebase)
  • -IUPHAR Guide to Pharmacology (https://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=863)
  • -TCDB (https://tcdb.org/hgnc_explore.php?stem=SLC)
  • -SOLVO knowledgebase (https://www.solvobiotech.com/knowledge-center/transporters-a-z)

What follows are detailed descriptions of the 76 individual SLC families, presented in numerical order from SLC1 to SLC76:

SLC1 High-affinity glutamate and neutral amino acid transporter family (2.A.23/SDF/DAACS)

Discovery: The rabbit intestinal epithelial and neuronal high-affinity Na⁺/glutamate transporter EAAC1/EAAT3 (SLC1A1) (134) is one of the founding members of the SLC1 family. At the same time two additional Na⁺-coupled glutamate transporters were cloned in parallel to the expression cloning of EAAC1/EAAT3 (SLC1A1) using Xenopus oocytes: 1) The rat brain glial Na⁺-coupled glutamate transporter GLT1/EAAT2 (SLC1A2) was identified at the molecular level using an antibody against the purified glial L-glutamate transporter (350); 2) The rat brain Bergmann glia Na⁺-coupled glutamate transporter GLAST1/EAAT1 (SLC1A3) was identified by serendipity due to a 66 kDa hydrophobic glycoprotein that was copurified during the isolation of a rat brain enzyme (133).

Gene family members (7):
SLC1A1 (EAAC1/EAAT3)	SLC1A5 (ASCT2)
SLC1A2 (GLT1/EAAT2)	SLC1A6 (EAAT4)
SLC1A3 (GLAST/EAAT1)	SLC1A7 (EAAT5)
SLC1A4 (ASCT1)

Molecular aspects, physiological roles and links to disease

The SLC1 family has five glutamate transporters (EAATs) and two alanine, serine and cysteine transporters (ASCTs). SLC1 family members belong to the Dicarboxylate/Amino Acid Cation (Na⁺ or H⁺) Symporter (DAACS) family (TC 2.A.23) and family members adopt the DAACS structural fold, also known as the GltPh fold (264). They have a structural architecture consisting of 8 TMHs and 2 reentrant helical hairpin loops (HP1 and HP2) (351) and use the elevator-type transport mechanism (264, 269, 270). For more details on structure and transport mechanisms of SLC1 family members, see Section 8.

Glutamate transporters of the SLC1 family play important roles in the termination of excitatory neurotransmission and in supplying glutamate to cells throughout the body for metabolic purposes (reviewed in (352)). The two neutral amino acid transporters ASCT1 (SLC1A4) and ASCT2 (SLC1A5) facilitate the electroneutral exchange of amino acids in neurons and cells of peripheral tissues. Consistent with this functional grouping and the phylogenetic tree (Fig. 5), the description of SLC1 family members follows the division into the following subfamilies:

Fig. 5. Phylogenetic tree of the SLC1 family.

1. Glutamate transporter subfamily: EAAC1/EAAT3 (SLC1A1), GLT1/EAAT2 (SLC1A2), GLAST/EAAT1 (SLC1A3), EAAT4 (SLC1A6), and EAAT5 (SLC1A7)

2. Alanine, serine, cysteine transporter subfamily: ASCT1 (SLC1A4), ASCT2 (SLC1A5)

1. Glutamate transporter subfamily

This subfamily includes the high-affinity ion-coupled glutamate transporters EAAC1/EAAT3 (SLC1A1), GLT1/EAAT2 (SLC1A2), GLAST/EAAT1 (SLC1A3), EAAT4 (SLC1A6), and EAAT5 (SLC1A7).

EAAC1/EAAT3 (SLC1A1), EAAT4 (SLC1A6) and EAAT5 (SLC1A7) are neuronal glutamate transporters and GLT1/EAAT2 (SLC1A2) and GLAST/EAAT1 (SLC1A3) are glial glutamate transporters. Both the neuronal and the glial glutamate transporters serve to regulate excitatory neurotransmission, to maintain low extracellular glutamate concentrations to protect against excitotoxicity and/or to provide glutamate for metabolism, including the glutamate-glutamine cycle (Fig. 6). It has furthermore been shown that the Na⁺/K⁺-ATPase co-assembles with GLT1/EAAT2 and GLAST/EAAT1 to maintain the ion gradients that drive transport (353).

Fig. 6. Expression of excitatory amino acid transporters in neurons and astrocytes.

Glutamate is released into the synaptic cleft and acts on AMPA and NMDA glutamate receptors, which are permeable to Na⁺ and Ca²⁺, respectively, resulting in membrane depolarization and action potential generation. High affinity glutamate transporters regulate glutamate concentration and thus neurotransmission by removing the excitatory neurotransmitter from the synaptic cleft. Glutamate transporters are also essential for maintaining extracellular glutamate concentration below neurotoxic levels. The five high-affinity glutamate transporters (EAAT1 to 5) have distinct regional and cellular localizations. GLT1/EAAT2 (SLC1A2) is mainly expressed in glia throughout the central nervous system and is responsible for 90% of the glutamate uptake. EAAC1/EAAT3 (SLC1A1) is a postsynaptically localized neuronal glutamate transporter that is widely distributed throughout the central nervous system. The figure shows the glutamate-glutamine cycle, which helps maintain an adequate supply of the neurotransmitter glutamate in the central nervous system.

All SLC1 glutamate transporters are coupled to the cotransport of three Na⁺, one H⁺, and the countertransport of one K⁺ (352, 354). K⁺ is required to return the empty transporter to the outward-facing open position to prepare it for a new transport cycle. Based on this stoichiometry, these transporters are effective in protecting the CNS from glutamate-induced neurotoxicity. Moreover, all SLC1 glutamate transporters have a Cl⁻ current that is uncoupled from glutamate transport (355–358). The postsynaptic membrane is hyperpolarized by this anionic current, creating a glutamate transporter-dependent feedback mechanism that has previously been studied in the retina (359, 360).

SLC1A1: EAAC1/EAAT3 (SLC1A1) is a high-affinity glutamate transporter found in neurons (Fig. 6) and epithelia (Fig. 17) (134). In the central nervous system, it is predominantly expressed in neurons of different brain areas, especially in the hippocampus, cerebral cortex, olfactory bulb, striatum, superior colliculus and thalamus (361–363). In the kidney, EAAC1 is present in the apical membrane of the proximal convoluted tubules (364). In the intestine, it is expressed in the brush border membrane of enterocytes in the duodenum (134) and absorbed glutamate can be converted to alanine by alanine transaminase in the enterocytes (365). Slc1a1 knockout mice did not develop significant neurological symptoms or neurodegeneration over a period of more than 12 months except that homozygous mutants showed significantly reduced spontaneous locomotor activity (366). On the other hand, Slc1a1 knockout mice developed dicarboxylic aminoaciduria, confirming the role of EAAC1/EAAT3 (SLC1A1) in the reabsorption of glutamate from the renal proximal tubules.

Among the five SLC1 family members, EAAC1/EAAT3 (SLC1A1) is the only isoform that can efficiently transport L-cysteine (L-Cys), a substrate for glutathione synthesis. EAAC1/EAAT3-mediated cysteine transport occurs with lower affinity for glutamate and aspartate, but its efficiency (V_max/K_m) for L-cysteine transport is much higher than that of the GLT1/EAAT2 (SLC1A2) and GLAST/EAAT1 (SLC1A3) paralogs (367). Whereas in astrocytes the alanine, serine and cysteine transporters ASCT1 (SLC1A4) and ASCT2 (SLC1A5) and the system x-cystine/glutamate antiporter xCT (SLC7A11) (see SLC7 summary) contribute primarily to L-Cys uptake, EAAC1/EAAT3 mediates the uptake of L-Cys into neurons. Since L-Cys is the rate-limiting substrate for the neuronal antioxidant glutathione synthesis (368) EAAC1/EAAT3 helps protect neurons from oxidative stress, which is particularly relevant in ischemic stroke, epilepsy, Parkinson disease, Huntington disease, Alzheimer disease and anti-aging protection (369, 370).

The structural basis of EAAC1/EAAT3 (SLC1A1) substrate recognition, including cysteine recognition, was investigated using cryo-EM and biophysical approaches (371). The structures visualize human EAAC1/EAAT3 recognizing four substrates: L-Asp, D-Asp, L-Glu, and L-Cys. Together with binding assays, the studies indicate that EAAC1/EAAT3 transports L-Cys in its thiolate form (i.e., the deprotonated, negatively charged form of the thiol group in L-cysteine). The transporter coordinates acidic amino acids and L-Cys thiolate by fine-tuning the position of specific residues, in particular arginine at position 447 (R447), which interacts with the acidic side chain of the substrate. Interestingly, the neutral amino acid transporters ASCT1 (SLC1A4) and ASCT2 (SLC1A5), which are known to transport cysteine in its thiol form (with an -SH group), have R447 replaced by threonine and cysteine, respectively (371). Furthermore, a mutant of EAAC1/EAAT3 in which R447 was replaced by cysteine did not bind or transport acidic amino acids, while it still transported L-Cys and neutral amino acids via the electroneutral exchange mode, similar to ASCT2 (see ASCT2/SLC1A5 summary) (371). The EAAC1/EAAT3 structures furthermore indicate that the substrate binds before the last sodium seals the gate and that different substrates affect the distribution of the transporter between different conformational states, thereby affecting translocation rates (371). Specifically, different substrates affect how the transporter distributes between a fully outward-facing conformation and intermediate occluded states on the way to the inward-facing conformation. This suggests that translocation rates are substrate-dependent.

Patients with dicarboxylic aminoaciduria, a rare autosomal recessive disorder of urinary glutamate and aspartate transport that can be associated with impaired intellectual development, have been described with loss-of-function mutations in SLC1A1 (372). These mutations of conserved residues impaired or abolished glutamate and cysteine transport by EAAC1/EAAT3 (SLC1A1) and resulted in nearly absent surface expression, consistent with EAAC1 being an epithelial and neuronal glutamate transporter. The results further confirm that EAAC1 is responsible for glutamate and aspartate reabsorption in the human kidney and suggest its potential involvement in the pathogenesis of certain neurological disorders. For example, several studies have linked genetic variants of SLC1A1 and/or dysfunction of EAAC1/EAAT3 to obsessive-compulsive disorder (OCD) (369). While it appears that disruptions in the expression or function of EAAC1/EAAT3 increase the risk of OCD-like behaviors, EAAC1/EAAT3 dysfunction is likely part of a complex system leading to OCD.

SLC1A2: GLT1/EAAT2 (SLC1A2) is a glial type high-affinity glutamate transporter (350). It is predominantly expressed in astrocytes of various brain regions, particularly in the cerebral cortex and hippocampus. GLT1 is a key player in the clearance of released glutamate at excitatory synapses, accounting for the majority of extracellular glutamate removal (373, 374) (Fig. 6). The Slc1a2-knockout mice showed dramatic lethal spontaneous epileptic seizures with behavioral patterns similar to those of N-methyl-D-aspartate (NMDA)-induced seizures (375). Despite depletion of the Slc1a2 gene, the mice did not show any significant changes in the expression of the other major glial glutamate transporter GLAST/EAAT1 (SLC1A3) (376, 377). The mice also exhibited selective neuronal degeneration in the hippocampal CA1 region, consistent with the roles GLT1 plays in neuroprotection. Electrophysiological analysis of the CA1 pyramidal neurons revealed that GLT1 contributes to the removal of glutamate from the synaptic cleft (375). Pharmacologically, GLT1 is distinct from the other high-affinity glutamate transporter paralogs. Evidence suggests altered regulation and different distribution of splice variants of GLAST/EAAT1 (SLC1A3) and GLT1/EAAT2 (SLC1A2) (374, 378, 379).

Several neurodegenerative diseases, including Alzheimer disease, Huntington disease, amyotrophic lateral sclerosis, Parkinson disease, bipolar disorder and epilepsy, have been linked to GLT1/EAAT2 (SLC1A2) dysfunction and mutants of SLC1A2, leading to excessive accumulation of extracellular glutamate, highlighting the importance of identifying compounds that can specifically act as EAAT2 neuromodulators (380–385).

GLT1/EAAT2 is selectively inhibited by dihydrokainate and kainate, which are non-transportable inhibitors of the transporter (386). The search for small molecules as inhibitors or modulators of glutamate transporters was stimulated by the cloning of the SLC1 glutamate transporter. In addition to L-trans-2,4-PDC (387), a series of synthetic cyclic glutamate analogs were generated and tested for inhibitory activity against EAAC1, GLT1 and GLAST, with the ancestor of this class of molecules recognizable as dihydrokainate (reviewed in (352, 388)). WAY-855 was subsequently developed with a conformationally restricted glutamate analog (389), which exhibits potent inhibition of glutamate transport with low selectivity for GLT1. In cryo-EM structures of human GLT1/EAAT2 with and without bound selective inhibitor WAY213613, a trimer was revealed with each protomer consisting of transport and scaffold domains, and WAY213613 occupying both the glutamate binding site and another transport-related cavity of the transporter to interfere with its alternate access transport mechanism (390).

A screening of FDA-approved drugs and dietary supplements led to the discovery that many β-lactam antibiotics are transcriptional activators of EAAT2, resulting in increased EAAT2 protein levels. Increased EAAT2 expression resulting from β-lactam antibiotic administration (e.g., ceftriaxone) is neuroprotective and occurs through nuclear factor NF-κB-mediated EAAT2 promoter activation (reviewed in (391)). Ceftriaxone has neuroprotective effects in both in vitro and in vivo models based on its ability to inhibit neuronal cell death by preventing glutamate excitotoxicity. Activation of the peroxisome proliferator-activated receptor gamma (PPAR-γ) by the diabetes drug rosiglitazone also leads to increased expression of EAATs through promoter activation (392). In addition, several translational activators of EAAT2 have since been described (393) along with treatments that increase surface expression of EAAT2 (394, 395) or prevent its downregulation (396).

Regarding EAAT2 impairment associated with Alzheimer disease, Huntington’s disease, amyotrophic lateral sclerosis, Parkinson disease, etc., the leucine-rich repeat kinase LRRK2 has been shown to be required for proper physiological function and expression of the glial glutamate transporter EAAT2 (SLC1A2) in the plasma membrane (385). Furthermore, mutations in LRRK2 have been shown to contribute to both monogenic and sporadic forms of Parkinson disease, of which the common substitution G2019S has been shown to be associated with a significant deficit in EAAT2 expression. The study also highlights the neurotransmitter transporter specificity of LRRK2-mediated modulation of EAAT2 only, revealing a likely important role for the kinase as a checkpoint protecting neurons from excitotoxicity (385). Thus, in brain disorders associated with impaired glutamate clearance, targeting LRRK2 for EAAT2 regulation may provide alternative novel therapeutic opportunities.

In addition, EAAT2 (SLC1A2) expression has been shown to be modified by single nucleotide polymorphisms (SNPs) that correlate with the pathogenesis of schizophrenia (397). Altered expression and localization of EAAT2 (SLC1A2) have been identified in schizophrenia and other major mental health conditions (398, 399), although the mechanisms by which such alterations contribute to the pathogenesis of schizophrenia have remained elusive. As highlighted in a follow-up study (397), insight into the possible role of altered EAAT2 expression in the pathogenesis of schizophrenia stems from the glutamate hypothesis, according to which NMDA receptor antagonists induce schizophrenia-like symptoms (400), as well as other observations of impaired glutamatergic transmission in the context of EAAT2 dysregulation (401). It was also shown that EAAT2 (SLC1A2) mRNA and protein expression is typically decreased at the regional level in schizophrenia, in the thalamus, hippocampus, and temporal cortex (402–404). In addition, in the thalamus, analysis of cell populations enriched for excitatory relay neurons or astrocytes revealed increased expression of EAAT2b, one of the alternatively spliced transcript variants of SLC1A2 identified in the neuronal population (405). A similar result was seen in the postmortem anterior cingulate cortex of schizophrenia patients, with significantly increased levels of EAAT2b in enriched populations of pyramidal neurons (406). In contrast to EAAT2 (SLC1A2), which is constitutively targeted to the plasma membrane (provided that LRRK2 is intact), the presence of a PDZ-binding domain in the C-terminus of the EAAT2b splice variant has been reported to facilitate the regulation of EAAT2b trafficking and stabilization at the cell surface by Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) and the structural protein DLG1 (Discs large homolog 1), a PDZ protein prominent in astrocytes and colocalized and coimmunoprecipitated with EAAT2b (407).

To investigate whether neuronal expression and function of EAAT2 (SLC1A2) are increased in the frontal cortex in subjects diagnosed with schizophrenia, expression of EAAT2 splice variants was examined in enriched populations of neurons and astrocytes from the dorsolateral prefrontal cortex (DLPFC) of schizophrenia patients (397). While the study revealed no significant changes in EAAT2 protein expression or glutamate uptake in the DLPFC of schizophrenia subjects compared to controls, transcript expression of EAAT2 (SLC1A2) and signaling molecules associated with EAAT2b trafficking via CaMKII and DLG1 were significantly altered, highlighting perturbation of astrocyte and neuronal EAAT2 expression and supporting the hypothesis for the role of dysregulation of the glutamate system in the pathophysiology of schizophrenia.

In terms of expression in peripheral tissues, based on studies in mice, GLT1/EAAT2 (SLC1A2) was detected in the plasma membranes of perivenous hepatocytes, where a C-terminal EAAT2 splice variant called the “a-variant” was most highly expressed (408). L-glutamate formed through the liver mitochondrial glutaminase GLS2 in periportal hepatocytes may be released across the sinusoidal membrane via a glutamate transporter operating in efflux mode, presumably OAT2, which enables the bidirectional transport of glutamate (see the description of SLC22A7) (409). From the blood circulation, it may be delivered into downstream perivenous hepatocytes via the glutamate transporter SLC1A2/GLT1, followed by conversion back into glutamine via glutamine synthetase and release via a glutamate exit transporter, supporting the interorgan L-Gln flux (365). The different locations of GLS2 (periportal hepatocytes) and GS (perivenous hepatocytes) allows the liver to incorporate ammonia either into urea or L-Gln.

SLC1A3: GLAST/EAAT1 (SLC1A3) is a glial-type high-affinity glutamate transporter that is particularly abundant in the cerebellum (133). It is expressed in astrocytes and cerebellar Bergmann glia (363).

The Slc1a3 knockout mice developed normally and were able to perform simple coordinated tasks. However, they exhibited motor discoordination in more difficult tasks, consistent with abnormalities in the cerebellum (410). Electrophysiological studies showed that cerebellar Purkinje cells in knockout mice remained innervated by climbing fibers into adulthood. Knockout mice also showed increased susceptibility to cerebellar injury (410). In the retina, GLAST/EAAT1 (SLC1A3) expression is in Müller cells. GLAST is required for normal signaling between photoreceptors and bipolar cells, as the electroretinogram beta-wave and oscillatory potentials are reduced in GLAST/EAAT1 (SLC1A3)-deficient mice. Furthermore, retinal damage after ischemic stroke is exacerbated in Slc1a3 knockout mice (411).

In the peripheral auditory system, GLAST/EAAT1 has been shown to play important roles in keeping the perilymph glutamate concentration at non-toxic levels during acoustic overstimulation (412). Thus, Slc1a3 knockout mice have been studied in the context of noise trauma. Exposure to intense, loud noise causes swelling of dendrites below the inner hair cells and physical damage to the outer hair cells of the ear, and it has been proposed that glutamate excitotoxicity may be partly responsible for hearing loss (412). This was supported by the observation that Slc1a3 knockout mice showed increased accumulation of glutamate in perilymph within the inner ear after acoustic overstimulation, resulting in exacerbation of hearing loss (412).

The effect of Slc1a3 gene disruption on amygdala-initiated seizures was also examined. In Slc1a3 knockout mice, seizure duration was found to be significantly prolonged compared to wild-type mice. Together with other studies, it was concluded that GLAST/EAAT1 is one of the determinants of seizure susceptibility (413).

Based on studies in rats, GLAST/EAAT1 (Slc1a3) is also expressed in tanycytes, the nutrient-sensing cells that line the third ventricle within the hypothalamus (414). This suggests a role for tanycyte-mediated glutamate transport in neuroendocrine activity.

In an effort to identify glutamate transporter subtype-selective inhibitors, UCPH-101 was identified as the first specific inhibitor of GLAST/EAAT1 (415). UCPH-101 is proposed to target a predominantly hydrophobic crevice in the trimerization domain of the GLAST monomer. X-ray crystal structures of thermostable EAAT1/GLAST variants in complex with the substrate L-aspartate or the allosteric inhibitor UCPH-101 provided new insights for the future design of allosteric compounds with improved selectivity for transporters of the SLC1 family (19).

SLC1A6: EAAT4 (SLC1A6) is a neuronal high-affinity glutamate transporter expressed predominantly in cerebellar Purkinje cells on postsynaptic dendritic spines (416, 417). This transporter exhibits a remarkable thermodynamically uncoupled chloride conductance associated with substrate transport. EAAT4 (SLC1A6) promoter activity was not present in non-neuronal cells (418). The EAAT4 glutamate transporter helps regulate excitatory neurotransmission and prevents glutamate-mediated excitotoxicity in the cerebellum.

SLC1A7: EAAT5 (SLC1A7) is a high-affinity glutamate transporter expressed primarily in the retina (419). EAAT5 is associated with rod photoreceptors and some bipolar cells based on immunocytochemical studies of the rat retina (420).

Substrate binding to glutamate transporters of the SLC1 family is known to generate thermodynamically uncoupled Cl^- fluxes to varying degrees (421). For both EAAT4 and EAAT5, the Cl^- flux is large enough to influence the excitability of neurons (422, 423). These glutamate transporters conduct chloride ions via a channel-like process that is gated by the joint binding of Na⁺ and glutamate but is thermodynamically uncoupled from the glutamate flux (423, 424).

The Cl^- conductance is particularly large in the retina-specific EAAT5 (SLC1A7), where glutamate-evoked currents are mainly carried by chloride ions (419). Therefore, it has been proposed that EAAT5 functions as a slow anion channel rather than a classical high-capacity glutamate transporter (425). These properties are similar to those of the glutamate-evoked chloride conductance previously described in retinal neurons, suggesting that the EAAT5-associated chloride conductance may be involved in visual processing (419) and help maintain cell excitability (423) and osmotic balance (426). Subsequent studies revealed that EAAT5 acts as a major inhibitory presynaptic receptor at the axon terminals of mammalian rod bipolar cells (423). Glutamate transport can lead to inhibition of L-type Ca²⁺ channels in these cells through EAAT5 (SLC1A7)-mediated Cl^- efflux (427). Thus, it has been proposed that the excitatory L-glutamate released from rod terminals provides a negative feedback signal to inhibit the release of further L-glutamate.

EAAT5 protein expression was shown to be strongly reduced at photoreceptor synapses in retinas of experimental autoimmune encephalomyelitis, a mouse model of multiple sclerosis, compared to healthy retinas (428). The data illustrate expression changes in the EAAT5 glutamate transporter in the early preclinical phase of the disease and suggest an involvement of EAAT5 in the previously observed early synaptic changes at photoreceptor synapses. Elucidation of the precise underlying mechanisms still requires further investigation.

2. Alanine, serine, cysteine transporter subfamily

This subfamily includes the neutral amino acid transporters ASCT1 (SLC1A4), and ASCT2 (SLC1A5).

The ASC neutral amino acid transporters exhibit the properties of the classical Na⁺-dependent amino acid transport system ASC (429–432). ASC transporters are believed to function exclusively as Na⁺-dependent amino acid exchangers, while glutamate transporters can mediate both uptake and exchange. Specifically, ASC transporters mediate high affinity Na⁺-dependent exchange of small neutral amino acids such as Ala, Ser, Cys and Thr. The two ASC transporters ASCT1 (SLC1A4) and ASCT2 (SLC1A5) exhibit distinct substrate selectivity (see below). Their structure is predicted to be similar to that of the glutamate transporters (429, 432). ASCT1 and ASCT2 also exhibit thermodynamically uncoupled chloride channel activity associated with substrate transport (433, 434). Whereas EAATs counter-transport K⁺ (see above) ASCTs do not, and their function is independent of the intracellular concentration of K⁺ (434). At low pH (5.5) both ASCT1 and ASCT2 are able to exchange acidic amino acids such as L-cysteate and glutamate (432, 435).

SLC1A4: ASCT1 (SLC1A4) is a trimeric Na⁺-dependent neutral amino acid transporter that has the properties of the previously characterized system ASC (429, 431). It accepts L-alanine, L-serine, L-cysteine, and L-threonine in a stereospecific manner. The currents associated with ASCT1-mediated transport result from activation of a thermodynamically uncoupled chloride conductance with permeation properties similar to those described for the glutamate transporter subfamily (434). In contrast to glutamate transporters, which mediate net flux and complete a transport cycle by the counter-transport of K⁺, ASCT1 mediates only the obligatory exchange of amino acids and is insensitive to K⁺ (434).

According to the HPA, ASCT1 (SLC1A4) is relatively widely expressed at lower levels, whereas in the brain it is prominently expressed in glial cells together with the enzyme 3-phosphoglycerate dehydrogenase, a key enzyme for L-serine biosynthesis, and extensive colocalization has been demonstrated at the cellular level (436). It has therefore been proposed that large amounts of L-serine are synthesized and stored in these glial cells and released through ASCT1 in exchange for other extracellular substrates. Thus, ASCT1 (SLC1A4) is essential for shuttling L-serine from astrocytes into neurons. The transporter also serves as a critical regulator of brain D-serine, a physiological coagonist of NMDA receptors that plays an important role in neurodevelopment (437).

A recent study showed that individuals with biallelic variants in SLC1A4 have spastic quadriplegia, thin corpus callosum, and progressive microcephaly (SPATCCM) syndrome, but individuals with heterozygous variants are not affected (438, 439). One mutation has been shown to cause a dominant-negative N-glycosylation defect of ASCT1, which in turn reduces the plasma membrane localization of ASCT1 and the transport rate of ASCT1 for L-serine, leading to global developmental delay, spasticity, epilepsy, and microcephaly (439).

A novel series of hydroxyproline analogs has been identified as promising agents for pharmacological modulation of ASCT1 (SLC1A4) and ASCT2 (SLC1A5) amino acid exchangers (440). As highlighted in the study, targeting the transport mechanisms that control the reuptake of D-serine represents a novel therapeutic approach for neuropsychiatric disorders such as schizophrenia, which are characterized by hyper- or hypo-functioning NMDA receptors. In addition, inhibition of SLC1A4 is expected to be beneficial in conditions associated with traumatic brain injury and Alzheimer disease, where studies in animal models have shown that D-serine efflux from inflammatory astrocytes can exacerbate NMDA-mediated excitotoxicity (440–443).

SLC1A5: ASCT2 (SLC1A5) is the second paralog of the ASC transport system. ASCT2 shows somewhat different properties in substrate selectivity (430, 432): In addition to the typical system ASC substrates L-alanine, L-serine, L-cysteine, and L-threonine, ASCT2 also accepts glutamine and asparagine as high-affinity and methionine, leucine, and glycine as low-affinity substrates, whereas ASCT1 does not accept these substrates (429–432). Furthermore, according to the HPA, unlike ASCT1, ASCT2 is widely expressed in kidney, intestine, lung, pancreas, liver, adipose tissue, muscle tissue, skin, bone marrow, male and female tissues, while brain expression is rather low.

ASCT2 transports glutamate, albeit with low affinity. Glutamate transport by ASCT2 is intensified at low pH (432). Like ASCT1, ASCT2 mediates the Na⁺-driven obligatory exchange of substrate amino acids (433). In kidney and intestine, ASCT2 was found in brush border membranes of proximal tubule cells and enterocytes, respectively (444).

ASCT2 (SLC1A5) is a key regulator of glutamine metabolism and as such has been shown to play an important role in the aggressive luminal breast cancer subtype (445). It is expected to be a potential therapeutic target for cancer treatment.

In addition, ASCT2 has been shown to serve as a receptor for retroviruses (446). This finding is analogous to that of the cationic amino acid transporter CAT1 (SLC7A1), which was originally identified as a viral receptor (see the description of the SLC7 family).

The cryo-EM structures of human ASCT2 (SLC1A5) were successfully determined in the outward-facing conformation in complex with the substrate L-glutamine (274). The structures provide insights into the conformation of the critical ECL2a loop, which allows for structural flexibility at TMH4b. They also provide new insights into substrate recognition, which involves conformational changes in the HP2 reentrant loop. A putative cholesterol binding site was also observed near the domain interface in the outward-facing state. These advances provide a major step forward in our understanding of substrate recognition and transport mechanisms in the SLC1 family.

Common and distinct functional characteristics of glutamate and ASC transporters:

Despite the different substrate selectivity of glutamate and ASC transporters, these transporters share common substrate recognition properties that reflect their structural similarity. For example, glutamate transporters, in particular EAAC1/EAAT3 (SLC1A1), transport the neutral amino acid cysteine (447), and conversely, the neutral amino acid transporter ASCT2 (SLC1A5) transports glutamate, albeit with low affinity (432). Glutamate transport by ASCT2 is enhanced at low pH (432). ASCT1 (SLC1A4) is inhibited by acidic amino acids such as glutamate, aspartate, cysteate and cysteine sulfinate by lowering the pH (435). ASC transporters, unlike glutamate transporters, do not appear to be coupled to H⁺ transport (433, 434). ASC transporters are not coupled to the counter-transport of K⁺ (433, 434), in contrast to glutamate transporters which require K⁺ for their translocation step. Consistent with this, Y403 and E404 in rat asct2 (Slc1a5), which are responsible for the K⁺ coupling in glutamate transporters, are not conserved in ASC transporters. These results indicate that both types of transporters have structurally similar substrate binding sites with only a few distinct amino acid sequence differences (448).

Orphan transporter family members: N/A

SLC2 Facilitative GLUT transporter family (2.A.1.1/Sugar_tr/MFS)

Discovery: The human erythrocyte glucose transporter GLUT1 (SLC2A1) was identified as the founding member of the SLC2 family by cloning from an expression library using antibodies directed against the human erythrocyte glucose transporter (120).

Gene family members (14):
SLC2A1 (GLUT1)	SLC2A5 (GLUT5)	SLC2A9 (GLUT9)	SLC2A13 (HMIT)
SLC2A2 (GLUT2)	SLC2A6 (GLUT6)	SLC2A10 (GLUT10)	SLC2A14 (GLUT14)
SLC2A3 (GLUT3)	SLC2A7 (GLUT7)	SLC2A11 (GLUT11)
SLC2A4 (GLUT4)	SLC2A8 (GLUT8)	SLC2A12 (GLUT12)

Molecular aspects, physiological roles and links to disease

The SLC2 family consists of 14 members that have been extensively studied (449, 450). It belongs to the Sugar Porter (SP) family (TC 2.A.1), which is part of the MFS superfamily of membrane proteins that harbor the canonical MFS fold. Consistent with this fold, SLC2 transporters typically consist of 6+6 TMHs arranged symmetrically in two bundles, forming two “halves” of the transporter. This structural arrangement is the basis for the alternating-access mechanisms of MFS transporters (207), i.e., the substrate-binding site of a transporter is exposed to only one side of the membrane at a time (see Section 8 for more details).

Crystallization and high-resolution structure determination of GLUT1 and GLUT3 have been reported, which is a major achievement considering that overexpression, purification and crystallization of eukaryotic membrane proteins is a challenging task in structural biology (451). GLUT1 was solved in the inward-open conformation at 3.2 Å resolution (452) and human GLUT3 in complex with D-glucose at 1.5 Å resolution in an outward-occluded conformation (2). The molecular basis for the alternating access transport cycle of this prototypical solute carrier family was revealed by structural elucidation of these transporters in three different functional states. This progress also provides the molecular basis for structure-based ligand design.

In addition, the availability of cryo-EM structures for GLUT1 (453) and GLUT4 (SLC2A4) (454) has fostered the discovery of inhibitors and substrates that are specific for each GLUT, some of which are under investigation for therapeutic use. Thus, the druggable properties of each GLUT protein are beginning to be elucidated (206, 455).

Based on the phylogenetic tree (Fig. 7), the SLC2 family can be divided into the following subfamilies (456):

Fig. 7. Phylogenetic tree of the SLC2 family.

Class I - classical glucose transporters: SLC2A1 (GLUT1), SLC2A2 (GLUT2), SLC2A3 (GLUT3), SLC2A4 (GLUT4), SLC2A14 (GLUT14)

Class II - lack affinity for 2-deoxy-D-glucose; unresponsive to cytochalasin B: SLC2A5 (GLUT5), SLC2A7 (GLUT7), SLC2A9 (GLUT9), SLC2A11 (GLUT11)

Class III - transporters with intracellular targeting signals and a large glycosylated loop exposed to the extracellular space or organellar lumen: SLC2A6 (GLUT6), SLC2A8 (GLUT8), SLC2A10 (GLUT10), SLC2A12 (GLUT12), SLC2A13 (HMIT)

Class I - classical glucose transporters

These include the family members SLC2A1 (GLUT1), SLC2A2 (GLUT2), SLC2A3 (GLUT3), SLC2A4 (GLUT4) and SLC2A14 (GLUT14)

SLC2A1: GLUT1 (SLC2A1) transports D-glucose, D-galactose, D-glucosamine, and the glucose analogs 2-deoxy-D-glucose and 3-O-methyl-D-glucose. It also transports the oxidized form of vitamin C, dehydroascorbic acid (DHA) (457). It has been shown that DHA and 3-O-methyl-D-glucose bind to mutually exclusive sites on the exo- and endofacial surfaces of GLUT1 and are transported by the same GLUT1 protein (458). GLUT1 consists of 12 hydrophobic transmembrane α-helices with both N- and C-termini on the cytoplasmic side and a glycosylated extracellular loop between TMHs 1 and 2.

GLUT1 is expressed in almost all tissues with varying levels of expression that usually correlate with the rate of cellular glucose metabolism (450). It provides glucose for energy production in erythrocytes and brain. Glucose transporters are present at very high concentrations in the erythrocyte membrane (up to 10% of total integral membrane protein), and thus the rate of transport is extremely high (459).

GLUT1 is crucial for the development of the BBB (460). GLUT1 deficiency syndrome (GLUT1DS) is a rare metabolic encephalopathy associated with abnormal brain metabolism. Impaired glucose transport at the level of the endothelial cells of the BBB (see Fig. 33) results in decreased glucose supply to the brain, which in turn leads to brain dysfunction. Genetically, GLUT1DS is due to a pathogenic variant in the SLC2A1 gene, which can be inherited sporadically or in an autosomal dominant manner, but rare cases of autosomal recessive transmission have been reported (461). Epilepsy predominates in childhood, and the disease in adulthood is characterized by paroxysmal exercise-induced dyskinesia and fatigue, followed by epilepsy and migraine. Specific recommendations have been reported that provide clinicians with a systematic guide for the rapid identification, diagnosis, and timely treatment of GLUT1DS (462).

Animal models of Glut1 (Slc2a1) deficiency have been generated to further characterize the role of Glut1 during BBB development and the pathogenic mechanism of GLUT1DS. However, homozygous Slc2a1 deficiency is lethal in mouse embryos and Slc2a1 haploinsufficiency results in compensatory mechanisms (463, 464). Subsequently, a new mouse model was generated by introducing a missense mutation, S324P, in Glut1 (Slc2a1), named Glut1^Rgsc200 (Slc2a1^Rgsc200) (465). The Glut1^Rgsc200 mutant mice have been shown to be a suitable model to study the pathogenic mechanism of GLUT1DS.

As mentioned above, GLUT1 in the endothelial cells of the BBB plays a critical role in cerebral glucose supply, as the brain requires glucose as an energy source under normal, healthy conditions. Delivery of D-glucose across the BBB is mainly mediated by the high-affinity transporter GLUT1, which is prominently expressed in the luminal and abluminal membranes of endothelial cells. The driving force for the facilitated diffusion of D-glucose across the BBB by GLUT transporters is provided by the concentration gradient between D-glucose in blood and brain interstitium (466). Between meals, the D-glucose concentration in blood is 4-6 mM, whereas the D-glucose concentration in brain interstitium is only 1-2 mM. The glucose concentration gradient between blood and brain interstitium is thought to be established and maintained by the uptake of D-glucose into astrocytes and neurons and the metabolic degradation of D-glucose in these cells. The K_m values determined for the uptake of D-glucose by GLUT1, measured in the absence of initial intracellular substrate (trans-zero uptake), range from 0.7 to 3.2 mM (466).

GLUT1 is also widely expressed in cells that form other blood-tissue barriers in addition to the BBB. Examples include 1) the blood-cerebrospinal fluid (CSF) barrier, where GLUT1 is localized to the basolateral membrane of epithelial cells in the choroid plexus (466, 467) (see Fig. 11); 2) the blood-retinal barrier, where GLUT1 facilitates glucose transport across the outer blood-retinal barrier formed by the retinal pigment epithelium (RPE), the inner blood-retinal barrier formed by the endothelium (468); and 3) the blood-testis barrier, where GLUT1 has been reported to be present together with GLUT3 (SLC2A3) and GLUT8 (SLC2A8) in Sertoli cells (469, 470).

At lower levels, GLUT1 is expressed in cardiac muscle and adipose tissue, where it provides glucose in the basal state.

GLUT1 is furthermore present in the placenta in endothelial cells and syncytiotrophoblasts. The expression of GLUT1 in the placenta is important because glucose is the predominant energy substrate for fetal oxidative processes and growth, and due to the lack of endogenous fetal glucose production until the first trimester, the maternal circulation is the only source of glucose for the placenta and fetus. Thus, glucose is taken up by the placenta and transported to the fetus via syncytiotrophoblasts mainly by GLUT1, whose expression increases during gestation (471). GLUT1 is localized to both plasma membranes, but with a threefold higher expression in the microvillus membrane (apical, facing the maternal blood side) compared to the basal membrane (facing the fetal capillaries) (472). The human placenta also expresses the fructose transporter GLUT8 (SLC2A8), the urate transporter GLUT9 (SLC2A9) and the glucose and dehydroascorbate transporter GLUT10 (SLC2A10), and first trimester syncytiotrophoblasts furthermore express the insulin-sensitive glucose transporters GLUT4 (SLC2A4) and GLUT12 (SLC2A12) (473).

Expression of GLUT1 in the basolateral membrane has been shown to increase in maternal diabetes, resulting in increased basal membrane glucose transport activity and pregnancy complications (473, 474), which likely contribute to the increased risk of developing gestational diabetes mellitus in response to maternal obesity (475).

Placental hypoxia is furthermore known to be involved in pregnancy pathologies, including fetal growth restriction and preeclampsia. The expression of GLUT1 (SLC2A1) as well as GLUT3 (SLC2A3) has been shown to be upregulated under hypoxic conditions via the hypoxia-inducible transcription factor HIF-1α (476). Conversely, GLUT1 has been shown to be downregulated in preeclampsia, which affects the intrauterine environment. This in turn affects fetal development and probably also fetal programming (477).

A major driver of cancer cell proliferation is increased uptake of glucose. Glucose is the primary nutrient used by eukaryotic cells to proliferate. In contrast, cancer cells are highly dependent on the glycolytic pathway to meet their energy requirements and prefer glucose fermentation over mitochondrial oxidation, even under aerobic conditions; this is called the Warburg effect (478). Thus, GLUT1 is an important target in cancer treatment because cancer cells upregulate GLUT1 (479).

The development of small-molecule inhibitors targeting GLUT1 has shown promise for anticancer therapy. Compounds such as WZB117, which blocks glucose uptake and induces growth arrest in cancer cells (480), and BAY-876, a highly selective inhibitor that impairs glucose metabolism with potent anticancer activity (481, 482), exemplify this approach. While these inhibitors were initially developed without high-resolution structural information, the more recent availability of cryo-EM structures for GLUT1 (453) and GLUT4 (SLC2A4) (454) has significantly advanced rational drug design efforts. These 3D structures are now being used to refine substrate specificity and develop next-generation GLUT inhibitors with improved selectivity and therapeutic potential.

As noted above, GLUT1 also transports the oxidized form of vitamin C, dehydroascorbic acid (DHA) (457) (Fig. 8). Vitamin C or ascorbic acid is a potent antioxidant in plasma that scavenges oxygen free radicals. Nucleated cells that are in high demand for vitamin C, such as chromaffin cells of the adrenal medulla during norepinephrine synthesis, concentrate vitamin C to high levels through uptake via the SVCT2 (SLC23A2) Na⁺-coupled vitamin C transporter. In contrast, erythrocytes, the most abundant cells in the blood, serve to maintain vitamin C levels in the blood so that the concentration of vitamin C in erythrocytes is similar to that in plasma. Erythrocytes have a unique ability to regenerate vitamin C from its oxidized form, DHA (483, 484). DHA is rapidly taken up by erythrocytes via the abundantly expressed glucose transporter GLUT1 (SLC2A1) (Fig. 8). Intracellularly, DHA is immediately reduced to vitamin C by glutathione, glutaredoxin or thioredoxin reductase (484).

Fig. 8. The transport and metabolism of vitamin C and dehydroascorbic acid (DHA) in the human body.

The vitamin C transporter SVCT1 (SLC23A1) absorbs L-ascorbate (vitamin C) across the intestinal brush border membrane. How it leaves the epithelial cells is still unknown. There are two possible routes to the brain. 1) across the choroid plexus (blood-cerebrospinal fluid barrier) via the vitamin C transporter SVCT2 (SLC23A2) located in the basolateral membrane and GLUT12 (SLC2A12) located in the apical membrane of choroid plexus epithelial cells and then from the cerebrospinal fluid across the ependymal cell layer via unknown transporters; and 2) across the BBB as DHA via GLUT1/SLC2A1 (not shown; see Fig. 33). Since DHA is present in blood at relatively low concentrations compared to vitamin C, the SVCT2 route via the choroid plexus appears to be the predominant pathway. In neurons, vitamin C is oxidized to DHA, which is released by GLUT3 (SLC2A3). Astrocytes import DHA via GLUT1 and reduce it to vitamin C. The exit pathway of vitamin C from astrocytes is currently unknown. Neurons import vitamin C via SVCT2, closing the vitamin C recycling pathway. DHA is rapidly taken up by erythrocytes via the abundantly expressed glucose transporter GLUT1 (SLC2A1). Intracellularly, DHA is reduced to vitamin C by glutathione, glutaredoxin or thioredoxin reductase, with the end result of replenishing plasma vitamin C levels.

Next, intracellular vitamin C is used to maintain plasma vitamin C levels. The proposed mechanism for this, based on a chemical knockout model, is as follows (484) (Fig. 8): 1) Intracellular vitamin C is oxidized to the ascorbate radical; 2) an electron leaves the cell while being carried by the transmembrane protein cytochrome b561; 3) outside the cell, the electron reduces an external ascorbate radical from the plasma that was formed as the first step of plasma vitamin C oxidation; and 4) reduction of the external ascorbate radical to vitamin C is the final step to replenish plasma vitamin C levels (484). Vitamin C generated in erythrocytes via GLUT1-mediated DHA uptake is also required to maintain the structural integrity of erythrocytes. The question arises as to whether the excess plasma glucose concentration that occurs in diabetic patients competes with DHA transport and compromises erythrocyte integrity, and whether vitamin C supplementation would resolve this problem. In fact, vitamin C supplementation favorably improved metabolic profiles in patients with T2D, including improvements in glycemic indices and insulin sensitivity, although the vitamin C status did not specifically affect osmotic fragility of erythrocytes (485).

SLC2A2: GLUT2 (SLC2A2) was first characterized by cDNA cloning of the Slc2a2 gene from rat and human liver cDNA libraries (486, 487). GLUT2 is a low affinity glucose and fructose transporter expressed in the basolateral membranes of intestinal and renal absorptive epithelial cells (see Fig. 14) and in the sinusoidal membrane of hepatocytes. The expression profile of GLUT2 according to the HPA is as follows: strongest expression in hepatocytes, robust expression also in duodenum and kidney, moderate expression in Leydig cells, low expression in bile duct cholangiocytes, absence of expression in pancreas and other tissues.

GLUT2 has the unique property among glucose transporters of having a low apparent affinity for glucose (K_m ~ 17 mM). It can also transport galactose (K_m ~ 92 mM), mannose (K_m ~ 125 mM) and fructose (K_m ~ 76 mM) with low affinity. Interestingly, it has a very high affinity for glucosamine (K_m ~ 0.8 mM) (488). Cellular GLUT2 expression is usually very high and therefore the rate of glucose uptake is not a limiting factor in glucose utilization.

GLUT2 in the human liver is considered to be a bidirectional transporter. It takes up glucose into hepatocytes for storage as glycogen during the fed state and releases glucose generated by either gluconeogenesis or glycogenolysis during the fasting state (450). However, although GLUT2 is essential for glucose uptake, the release of glucose from hepatocytes does not require the presence of GLUT2, suggesting that an alternative unknown system releases glucose in the case of hepatic glucose production (489).

GLUT2 is the major basolateral glucose efflux mechanism in the intestine. The uptake of glucose depends on the presence of the Na⁺/glucose cotransporter SGLT1 (SLC5A1) located in the apical membrane (121). It can transport glucose as well as galactose (450). Fructose uptake is mediated by GLUT5, which is also located in the apical membrane (450).

In the kidney, GLUT2 is located on the basolateral membrane of the epithelial cells that are involved in the reabsorption of glucose. Expression of GLUT2 is essential to glucose reabsorption. In contrast to the situation in intestinal cells, suppression of GLUT2 expression by knocking out the Slc2a2 gene induces a massive glucosuria (490), indicating that this transporter is absolutely necessary for the process of glucose reabsorption in the kidney.

GLUT2 is highly expressed in pancreatic β-cells of rat but not human (450). Animal studies revealed that GLUT2 in pancreatic β-cells acts as a glucose sensor that detects small increases in glucose levels leading to increased insulin secretion. However, in human islets it is not clear which glucose transporters take over the blood glucose sensing function to control insulin secretion. GLUT1 and GLUT3 were proposed as candidates to render this task (491), although according to the HPA it appears that they are not much expressed in the endocrine pancreas.

The three-dimensional structure of GLUT2 is currently unknown. However, homology-based models of GLUT2 in different conformations based on the structures of other human SLC2 members (2, 454, 492), as well as the E. coli homolog XylE (493), have been generated (494).

These structural models have been used to analyze genetic variants of SLC2A2 that cause Fanconi-Bickel syndrome (495). These are deleterious variants in SLC2A2 that cause a rare glycogen storage disease characterized by hepatorenal glycogen accumulation leading to severe renal tubular dysfunction and impaired glucose and galactose metabolism.

Interestingly, a follow-up study reported that GLUT2 (SLC2A2) may also function as a bidirectional urate transporter, and an association between genetic variations in SLC2A2 and gout/serum urate was reported (496). This finding adds to the physiological roles of other SLC2 family members, especially GLUT9 (SLC2A9) and possibly also GLUT12 (SLC2A12) as urate transporters (see descriptions of SLC2A9 and SLC2A12). However, further studies are still needed to clarify the precise physiological role of GLUT2 (SLC2A2) as a urate transporter.

SLC2A3: GLUT3 (SLC2A3) was first cloned from a human fetal skeletal muscle cell line and shares 64% sequence identity with SLC2A1 (487). GLUT3 is a high-affinity glucose transporter with a K_m for D-glucose of ~1.4 mM, similar to that of GLUT1 (SLC2A1), while that of GLUT2 (SLC2A2) is ~11 mM (466, 497, 498). It also transports dehydroascorbate (see below). With respect to D-galactose, it is transported with low affinity (K_m ~ 8.5 mM) (497), while GLUT3 is not able to transport fructose.

GLUT3 (SLC2A3) is closely related to the paralog GLUT14 (SLC2A14), which is largely specific to the testis (spermatocytes, spermatids), with 95% amino acid sequence identity (see SLC2A14 description below). GLUT14 is a transporter of D-glucose and dehydroascorbate. The two paralogs, GLUT3 (SLC2A3) and GLUT14 (SLC2A14), are thought to have arisen from a gene duplication event.

The major glucose transport systems used by neurons and astrocytes in the brain are GLUT3 (SLC2A3) and GLUT1 (SLC2A1), respectively (466). Neurons have high energy demands and rely heavily on glucose as their primary energy source (499). The properties of the glucose transporter GLUT3 make it well-suited to mediate neuronal glucose uptake in the brain’s low-glucose environment. This is because:

• 1)GLUT3 has a higher apparent affinity and turnover number for glucose than other “Class I” classical glucose transporters, ensuring efficient neuronal uptake;
• 2)D-glucose is the primary physiological substrate (500); and
• 3)glucose concentrations in the brain interstitial fluid are much lower (1-2 mM) than in (~ 5 mM). Thus, GLUT3, as a high-affinity glucose transporter, is optimally positioned to support the basal “housekeeping” glucose demands of neurons.

Once in the cytosolic space, glucose is phosphorylated by glucokinase to form glucose 6-phosphate (501).

In other tissues, GLUT3 has been detected at the mRNA and/or protein level in placenta, kidney, and at lower levels in adipose tissue and small intestine (502, 503). The HPA suggests that expression at the mRNA and/or protein level roughly decreases in the order bone marrow (hematopoietic cells), lymphoid tissues (spleen, lymph nodes), testis (pachytene spermatocytes, round early spermatocytes, elongated or late spermatids strong), epididymis, lung (macrophages, endothelial cells), brain areas (strongest in cerebral cortex), placenta (trophoblastic cells), breast (glandular cells); at lower levels in intestine (glandular cells in duodenum and colon), kidney (cells in glomeruli), liver (cholangiocytes), muscle and adipose tissue.

GLUT3 is also expressed in the retina, specifically in Müller glial cells according to HPA, where it is likely to play an important role, together with GLUT1 (SLC2A1), in retinal glucose metabolism and metabolic reprogramming in retinal diseases in this most energy-demanding of tissues (504).

While the main physiological substrate was reported to be D-glucose (500), it was later discovered that it also transports dehydroascorbate (DHA). The K_m values of DHA for GLUT1 and GLUT3 were 1.1 and 1.7 mM, respectively, similar to those previously reported for glucose (457). Thus, the apparent affinities of both glucose transporters for DHA are comparable and suitable for the transport of either substrate. The concentration of DHA in blood is low (~0.5 μM) (505), but DHA can be accumulated via GLUT3 against a concentration gradient by a mechanism involving facilitated uptake followed by retention of reduced ascorbic acid (vitamin C) (506).

Anerobic glycolysis can occur in the brain as part of the astrocyte-neuron lactate shuttle (507–509). During high activity in excitatory neurons, astrocytes can supply neurons with lactate as an energy source when neuronal glucose uptake via GLUT3 is insufficient to meet energy demands. Glutamate released from neuronal terminals is taken up by astrocytes via the excitatory amino acid transporter GLT1/EAAT2 (SLC1A2) for conversion to glutamine by ATP-dependent glutamine synthetase (510). Glutamine is then delivered to neurons to replenish the neurotransmitter pool of glutamate. Glutamate uptake by astrocytes also stimulates glucose uptake from blood vessels via GLUT1. This occurs as follows: Glutamate taken up by astrocytes leads to intramitochondrial acidification, a process reported to involve the uncoupling protein UCP4 (SLC25A27), which facilitates H⁺ translocation across the inner mitochondrial membrane (see the SLC25 family description) (511). As a consequence, intramitochondrial acidification reduces the rate of mitochondrial respiration. Reduced ATP production is effectively compensated for by increased glycolysis. This process stimulates glucose uptake by astrocytes via GLUT1 (SLC2A1) to initiate glycolysis. The resulting pyruvate is then converted to L-lactate by lactate dehydrogenase. L-lactate exits astrocytes, presumably via MCT4 (SLC16A3), and enters neurons where it is oxidized in mitochondria to fuel the citric acid cycle and promote oxidative phosphorylation. Thus, glutamate-induced mitochondrial acidification in astrocytes promotes glucose uptake via GLUT1 from the blood for glycolysis in astrocytes, the end product of which is lactate, which is used as an energy source by neurons to enhance neuronal survival during.

On the other hand, neurons are highly sensitive to oxidative stress, especially during high activity (512), and protection by vitamin C is very important. According to current knowledge, vitamin C is taken up by neurons via SVCT2 (SLC23A2) to scavenge free radicals during neurotransmission, and the resulting DHA exits neurons via GLUT3 and then enters astrocytes via GLUT1, where it is recycled to vitamin C. From there, it exits astrocytes via an unknown mechanism and reenters neurons via SVCT2. It was previously shown that astrocytes transport DHA via the glucose transporter GLUT1 (SLC2A1) (513). Thus, GLUT3 appears to have two transport roles in neurons in opposite direction: 1) uptake of glucose as an important source of energy; and 2) export of DHA, the oxidized form of vitamin C, formed after scavenging free radical during high neuronal activity.

A polymorphism in SLC2A3, rs12842, has been found to be associated with an increased risk of attention-deficit/hyperactivity disorder (ADHD). Epidemiologic and genetic studies have reported a link between antecedent ADHD and Alzheimer disease, as both share a dysregulation of brain glucose (514).

During sperm development, GLUT3 is highly expressed in spermatocytes and spermatids. In principle, it could deliver glucose to these germ cells. The glucose concentration in human seminal fluid is ~400 μM and actual results indicate that lactate is the major energy source during spermatogenesis and for maintaining sperm motility and velocity. Thus, lactate is considered to be an essential Sertoli cell-derived energy metabolite that nourishes spermatogenic cells (515). Sertoli cells take up glucose from the blood and produce lactate through glycolysis. In addition, spermatocytes and spermatids are highly sensitive to oxidative stress, suggesting that dehydroascorbate would be a more reasonable substrate to be transported than glucose (see the SLC2A14 description for more details).

As mentioned in the description of SLC2A1, the glucose transporter GLUT1 is localized in both the microvillus membrane (apical) and the basal membrane (facing the fetal capillaries) of syncytiotrophoblasts. In contrast, GLUT3 is mainly localized in the microvillus membrane (516). It has been shown that GLUT3 plays an important role in glucose transport across syncytiotrophoblasts in early pregnancy, while its expression decreases significantly in the second and third trimesters (516, 517). Since GLUT3 has a higher affinity for glucose and a greater transport capacity than GLUT1, this transporter is particularly important to ensure sufficient glucose transport capacity in early pregnancy, supporting the concept that both glucose transporters are required for optimal fetal development (517). It has also been suggested that the asymmetric distribution of GLUT1 and GLUT3 across the placental barrier may help prevent glucose loss during periods of maternal hypoglycemia (518).

GLUT3 is expressed in human white blood cells, human lymphocytes, monocytes/macrophages, neutrophils and platelets. Interestingly, in these cells, GLUT3 is largely confined to intracellular storage vesicles. From there, it translocates to the plasma membrane in response to stimulation by multiple pathways (519). A recent study shows that activated T cells massively upregulate GLUT3 expression with acetyl-CoA generation as a limiting step in the epigenetic regulation of inflammation-associated genes involved in inflammatory diseases (520). The results suggest that inhibition of GLUT3-dependent acetyl-CoA generation is likely to be a promising strategy to alleviate T helper 17 cell-mediated inflammatory diseases.

GLUT3 has been identified as a ferroptosis-related gene involved in rheumatoid arthritis (521). In this study, synovial fibroblast-like cells of rheumatoid arthritis (RA-FLS) were treated with RSL3, a selective ferroptosis inducer, and it was found that RSL3 is able to downregulate GLUT3 expression and induce ferroptosis in RA-FLS. It was also found that RSL3 can induce ferroptosis in RA-FLS via downregulation of GLUT3. Ferroptosis is a type of programmed cell death dependent on iron and characterized by the accumulation of lipid peroxides. It has been found that cells overexpressing GLUT3 feature increased glycolysis and glycolysis reprogramming through the PI3K/AKT/mTOR/HIF1α signaling pathway while generating glutathione peroxidase (GPX4), a phospholipid hydroperoxidase that protects cells against membrane lipid peroxidation, thereby inhibiting RA-FLS ferroptosis and promoting RA-FLS cell proliferation (521).

The cation channel TRPM7 was shown to be a novel “hot spot” of glycolytic reprogramming through transcriptional upregulation of GLUT3 via its calcium channel functionality (522). Ca²⁺ influx through TRPM7 activates calcineurin. Calcineurin in turn dephosphorylates the cAMP response element binding protein (CREB)-regulated transcription coactivator 2 (CRTC2), followed by the nuclear translocation of CRTC2 and activation of CREB. CREB then transduces the Ca²⁺ signal to activate SLC2A3 transcription (522). Deletion of endothelial TRPM7 also impaired retinal vessel growth. It has been proposed that inhibition of TRPM7-dependent glycolysis, which has been shown to be activated via GLUT3, could be exploited for cancer therapy and pathological angiogenesis.

SLC2A4: GLUT4 (SLC2A4) is the insulin-responsive glucose transporter that is highly expressed in skeletal muscle, cardiomyocytes, and adipocytes (486, 523–525). Under basal, low insulin conditions, GLUT4 is primarily located in intracellular membrane compartments, but when circulating insulin levels rise after a carbohydrate meal, it induces GLUT4 relocation to the plasma membrane to enhance glucose uptake and metabolism in these tissues, thereby preventing chronic elevations in blood glucose (450). Insulin resistance is characterized by a failure of insulin to stimulate translocation of GLUT4 to the cell surface in muscle and adipose tissue. Combined with defective insulin secretion from pancreatic beta cells and hepatic insulin resistance with impaired insulin suppression of glucose production in hepatocytes, this leads to type 2 diabetes mellitus (526).

The cryo-EM structures of human GLUT4 bound to its small molecule inhibitor cytochalasin B have been reported, laying the foundation for further mechanistic investigation of the modulation of GLUT4 trafficking (454). The cellular trafficking of GLUT4 has been shown to be regulated by posttranslational modifications. Cys223, which is at the intracellular end of TMH6, can be palmitoylated by DHHC-7 (“Asp-His-His-Cys”-7) palmitoyl acyltransferases, an event that controls insulin-dependent translocation of GLUT4 to the plasma membrane (527). Additional structural determination of the palmitoylated GLUT4 will be necessary to reveal the molecular basis for the regulation of the trafficking.

Insulin-induced GLUT4 translocation has been extensively studied (528). As part of the insulin-induced translocation of GLUT4 from the cytosol to the cell membrane, the binding of insulin to its receptors initiates a signal transduction cascade that results in the activation of Akt, which acts on GLUT4-containing vesicles in the cytosol to facilitate their fusion with the cell membrane.

GLUT4 is the major glucose transporter in adipose and muscle cells and is recycled between the plasma membrane and intracellular storage vesicles (529). Its steady-state distribution is regulated by insulin- and/or contraction-dependent signaling cascades. The two closely related Rab GTPase-activating proteins TBC1D1 and TBC1D4 play a critical role in regulating GLUT4 translocation in response to insulin and contraction in skeletal muscle (530). Rab8 and Rab10 have been identified as major GTPases involved in GLUT4 translocation in muscle and fat cells, respectively (529). In muscle cells, GLUT12 has been described to undergo regulated trafficking in response to stimulation by insulin, but whether GLUT12 expression could compensate for GLUT4 is unknown (see the SLC2A12 description).

Interestingly, both Rab GTPase-activating proteins, TBC1D1 and TBC1D4, also control the uptake of long-chain fatty acids (LCFAs) into skeletal muscle via the fatty acid transporter FATP4 (SLC27A4) (530). In response to muscle contraction or insulin stimulation, the LCFA translocase CD36 (also known as the scavenger receptor SR-B2) and the glucose transporter GLUT4 translocate to the plasma membrane to increase cellular uptake of fatty acids and glucose, respectively (531, 532). CD36 is a membrane protein with two transmembrane segments and a large extracellular loop with a hydrophobic sequence to which lipid ligands can bind, with both its N- and C-termini facing the cytosol (531). Notably, CD36 and GLUT4 may be mobilized from distinct intracellular pools within the endosomal compartment.

Furthermore, TBC1D4 controls LCFAs entry via CD36 and/or FATP4 (SLC27A4) transporters in human adipocytes (533) (see the SLC27 family description).

SLC2A14: GLUT14 (SLC2A14) is a transporter of dehydroascorbate and D-glucose that is mostly specific to the testis. SLC2A14 maps to chromosome 12p13.3, about 10 Mb upstream of GLUT3, with which it shares remarkable (95%) sequence identity and is likely to have arisen from a duplication of the GLUT3 (SLC2A3) gene (534).

GLUT14 has two alternatively spliced forms, a short form of GLUT14 (GLUT14-S) yielding a protein of 497 amino acids and a long form (GLUT14-L) yielding a protein of 520 amino acids differing from GLUT14-S only at the N-terminus (534, 535). GLUT14-L contains 12 putative TMHs along with sugar transporter signature motifs previously shown to be essential for sugar transport activity. Interestingly, the ortholog of GLUT14 is not found in mice. In terms of the transport function of the two GLUT14 (SLC2A14) isoforms, it was shown that both mediate the cellular uptake of dehydroascorbic acid (DHA) (536). D-glucose is also transported, whereas vitamin C and fructose were not transported by GLUT14.

In contrast to the tissue-wide expression pattern of GLUT3, both isoforms of GLUT14 are quite specifically expressed in the testis at high levels, approximately four times higher than GLUT3 (534). According to the HPA, GLUT14 is highly and specifically expressed in the testis and, at the tissue and cellular level, immunostaining is strongest in pachytene (late) spermatocytes. Male germ cells at various stages of differentiation from pachytene spermatocytes to mature caudal epididymal spermatozoa have an intrinsic propensity to generate ROS (537). Spermatozoa are particularly susceptible to oxidative stress because their plasma membranes contain high levels of polyunsaturated fatty acids, which can lead to impaired sperm function often observed in infertility patients (537, 538).

Vitamin C is widely recognized as a potent antioxidant for its ability to cause cellular damage through its ROS scavenging activity. When vitamin C performs its antioxidant function, it produces dehydroascorbic acid, which must be removed from cells via GLUT transporters similarly as in neurons (539) or regenerated via glutathione dehydrogenase (540). Therefore, in spermatocytes, GLUT14 can either extrude dehydroascorbic acid or take it up for intracellular reduction to vitamin C to provide antioxidant defense and prevent mitochondrial dysfunction, since mitochondria can regenerate vitamin C from its oxidized forms (541).

The question remains whether spermatocytes first take up vitamin C via the Na⁺-coupled vitamin C transporter SVCT2 (SLC23A2) or whether they take up DHA for intracellular reduction to vitamin C. Vitamin C is present in seminal fluid at high concentrations compared to blood plasma (400 μM vs. 60 μM), ready to effectively protect human sperm from oxidative stress and DNA damage (542). According to the HPA, SVCT2 (SLC23A2) is expressed at moderate levels in early and late spermatids. So, whether vitamin C is first taken up by SVCT2 in early/late spermatids, followed by oxidation to DHA, which then exits through GLUT14, possibly for regeneration in Sertoli cells, or whether GLUT14 in late (pachytene) spermatocytes takes up DHA for intracellular conversion to vitamin C, remains to be clarified.

Outside the testis, low levels of expression of GLUT14 were seen in Paneth cells (the secretory cells located in the crypts of Lieberkühn), exocrine gland cells (pancreas), granulosa cells of the ovary, inhibitory and excitatory neuronal cells, microglial cells, granulocytes, Langerhans cells, and erythroid cells, the latter showing the highest expression within the group.

Studies of a well-phenotyped cohort revealed that SLC2A14 SNPs are associated with inflammatory bowel disease (IBD), strengthening the hypothesis that genetically determined local dysregulation of vitamin C as an antioxidant contributes to the development of IBD (536). Previously, the SLC23A1 gene, which encodes the intestinal Na⁺-dependent apical vitamin C transporter SVCT1, was also found to be associated with IBD, particularly with Crohn disease (543). Since genetic variations in the two intestinal vitamin C transporter genes SLC23A1 and SLC2A14 are both associated with IBD, it has been hypothesized that a localized vitamin C deficiency caused by decreased membrane transport may be a causative or contributing factor in the etiology of intestinal inflammation.

Paneth cells might play a role in these pathologies as they are known to protect the intestinal mucosal barrier from development of IBD (544). Paneth cells accomplish this through autophagy to prevent inflammation in the intestine and to avoid the stress caused by reactive oxygen species. However, whether GLUT14 (SLC2A14) and possibly also SVCT1 (SLC23A1) are present at sufficient levels in Paneth cells would still need to be verified.

Class II - lack affinity for 2-deoxy-D-glucose; unresponsive to cytochalasin B

These include SLC2A5 (GLUT5), SLC2A7 (GLUT7), SLC2A9 (GLUT9) and SLC2A11 (GLUT11)

SLC2A5: GLUT5 (SLC2A5) is a highly specific fructose transporter that mediates the uptake of dietary fructose across the apical membrane of enterocytes in the small intestine (545–547) (see Fig. 14). As further highlighted by the HPA and consistent with previous studies (547), GLUT5 is not only highly expressed in the duodenum, but is also abundant in bone marrow, kidney, skeletal muscle, testis, tongue and brain. In testis, GLUT5 is thought to be responsible for fructose uptake by spermatozoa, providing fructose as an energy source (546).

While GLUT5 transports exclusively fructose, GLUT1 is a glucose-specific transporter and GLUT2 transports both glucose and fructose (498).

SLC2A5 expression in the small intestine is regulated at the transcriptional level by the presence of fructose in the intestine (548) and by diurnal rhythm independent of fructose availability (549).

Circulating levels of fructose are generally much lower than glucose, despite high consumption of fructose in the form of sugar-sweetened beverages, including energy, fruit, and soft drinks. This is because most dietary fructose is rapidly metabolized in the liver after intestinal absorption. Once in the portal circulation, ingested fructose is targeted to liver hepatocytes via GLUT2 (SLC2A2) and GLUT8 (SLC2A8) (550) (see also the SLC2A8 description). Ingested fructose is almost completely removed from the portal blood after first-pass, with only a small fraction entering the systemic circulation (551). According to a review of the subject (552), ingested fructose increases the availability of intrahepatic carbohydrate metabolites, leading to de novo lipogenesis and deleterious effects on hepatic insulin sensitivity that are worse than those induced by other carbohydrates such as glucose and galactose.

In addition, fructose is the only sugar known to increase serum urate, and increased dietary intake has been implicated in rising serum urate levels in humans and the risk of gout, as well as the increasing global prevalence of metabolic syndrome (MetS) (553, 554). Specifically, fructose intake effectively increases urate production in the liver through increased purine catabolism and de novo synthesis (555).

Fructose filtered in the kidney by the glomeruli is reabsorbed in the renal proximal tubules by GLUT5 (SLC2A5) or GLUT2 (SLC2A2) and converted to glucose or lactate (556).

GLUT5 expression has been shown to be increased in the intestine and skeletal muscle of type 2 diabetics and in certain cancers that are highly dependent on fructose uptake and metabolism (547, 557). In insulin resistance, this upregulation provides an alternative route for energy supply. The liver nuclear receptor LXRα (encoded by NR1H3) was subsequently shown to be responsible for the upregulation of GLUT5 (SLC2A5) mRNA and protein levels (557). As a ligand-activated transcription factor, LXRα may provide novel pharmacological strategies for selective modulation of GLUT5 activity in the treatment of metabolic diseases and cancer.

In another study, the determinants of fructose-mediated cell proliferation were elucidated. In short, overexpression of GLUT5 in non-fructolytic cells enabled growth in fructose-containing media. Specifically, GLUT5 allowed fructose to flow through glycolysis using hexokinase instead of fructokinase (or ketohexokinase), the enzyme that normally catalyzes the phosphorylation of fructose to produce fructose-1-phosphate in the liver (558). It was therefore concluded that GLUT5 is a potent driver of fructose-dependent cell proliferation. In colorectal cancer, for example, there is increased GLUT5 expression compared to healthy controls and a significant positive correlation between GLUT5 expression levels and cancer grade, and GLUT5 blockade has been shown to inhibit the proliferation of colorectal cancer cells (559).

Subsequently, it was shown that loss-of-function variants of the SLC2A5 gene do not contribute to the clinical presentation of acquired fructose malabsorption (560). It seems reasonable to speculate that other members of the SLC2 family compensate for the loss of GLUT5 (SLC2A5) function, such as the putative fructose transporter GLUT7 (SLC2A7), which is also expressed in the brush border membrane of the small intestine (see the SLC2A7 description below).

GLUT5 plays a critical role in metabolic disorders such as obesity induced by a high-fructose diet. Thus, GLUT5 could be a promising diagnostic and therapeutic target in both metabolic diseases and cancer (561). Further studies of GLUT5 will contribute to the development of targeted therapeutic tools, for which a cryo-EM structure would be helpful. However, due to various difficulties, the molecular structure and mechanism of GLUT5 have not been fully elucidated (561).

SLC2A7 (“Semi-deorphanized”): GLUT7 (SLC2A7) is primarily expressed in the small intestine and colon based on studies in rat (562), while the human ortholog is expressed exclusively in the duodenum according to the HPA. The rat protein was found to be expressed in the apical membrane of the small intestine and colon and was shown to have a high affinity (K_m <0.5 mM) for glucose and fructose (562). Surprisingly, later reevaluations questioned whether GLUT7 is really a transporter of glucose or fructose, in part also due to the low turnover rates of the transporter, and it was suggested that the physiological substrate of GLUT7 awaits further investigation (450, 563). However, a follow-up study clarified that GLUT7 expressed in oocytes is capable of transporting both glucose and fructose, as originally proposed (564). Thus, in principle, GLUT7 would be expected to contribute to at least fructose uptake across the apical membrane of enterocytes. This study also showed that the GLUT7-mediated uptake of both glucose and fructose was inhibited by the flavonoid apigenin, but not by the flavonoids quercetin or EGCG (epigallocatechin gallate). By confirming that fructose is indeed transported by GLUT7, this finding is also consistent with the hypothesis that a conserved isoleucine-containing motif present in GLUT2 (SLC2A2), GLUT5 (SLC2A5), and GLUT7 (SLC2A7) may be essential for fructose transport (565).

Meanwhile, a cryo-EM structure of human GLUT7 has been reported in which this transporter was stabilized in the outward-facing conformation in the absence of substrate (492). A structural model of GLUT7 was generated suggesting that, similar to other members of the GLUT family, GLUT7 undergoes a global rocker-switch-like reorientation of the transmembrane bundles to facilitate substrate translocation across the membrane (492). The cryo-EM structure in the open state provides new insights into the molecular architecture of GLUT7 with details of the substrate-binding cavity. However, the question of whether the physiological substrate of GLUT7 is glucose, fructose or some other compound is still unresolved. Therefore, GLUT7 can at best be classified as a “semi-deorphanized” transporter.

SLC2A9: GLUT9 (SLC2A9) is a urate transporter that is most highly expressed in the kidney, with prominent expression also in the liver (Fig. 33) (566). The SLC2A9 gene has been shown to encode two alternative RNAs that encode different N-terminal cytoplasmic tails (450, 566), the original GLUT9, also referred to as GLUT9a, and a splice variant that encodes a protein with a shorter N-terminus, GLUT9b. Importantly, the N-terminal tails direct differential targeting of GLUT9 to opposite poles of epithelial cells (567–569). GLUT9a is targeted to the basolateral membrane and GLUT9b to the apical membrane of polarized cells (568). GLUT9a is present in many tissues, including liver, kidney, intestine, leukocytes and chondrocytes, while GLUT9b is expressed only in liver, kidney and placenta (570). The HPA also shows prominent expression of GLUT9 (SLC2A9) variants in salivary gland, adrenal gland, and male/female tissues (cytotrophoblasts/spermatids).

Although initially thought to be a glucose or fructose transporter (500, 571), it has now been established that GLUT9 (SLC2A9) is a urate transporter (572, 573). Specifically, it was found that urate transport in Xenopus oocytes expressing SLC2A9 is about 50 times faster than that for glucose, and that glucose stimulates SLC2A9-mediated urate transport (574).

In the human kidney, GLUT9 is expressed in the proximal tubule (568), whereas in the mouse its ortholog is present in the basolateral and apical membranes of the distal convoluted tubules. The difference in expression may account for species differences in urate handling (575).

Urate transport mediated by GLUT9a and GLUT9b has the same K_m value of 0.6 mM (450). The transport is electrogenic and depends on the membrane potential (573). Urate transport can be inhibited by the uricosuric agents benzbromarone and losartan and there is partial inhibition by phloretin but not by cytochalasin B (450).

Urate is biosynthesized from purine by xanthine oxidase mainly in the liver and excreted in urine and feces. Although several transporters responsible for renal and intestinal handling of urate have been reported, information on hepatic transporters is limited. In mice, specific deletion of GLUT9 in the liver alone led to hyperuricemia, suggesting a role for GLUT9 in transporting urate into hepatocytes and making it available for degradation by uricase (575). However, this finding seems to be irrelevant for human physiology, since humans do not express uricase, so the uptake of urate into hepatocytes seems unnecessary. A recent study shed new light on this topic by showing that GLUT9 (SLC2A9) strongly and MRP4 (ABCC4) moderately contribute to the sinusoidal efflux of urate into the blood (576).

Urate as a major transport substrate of GLUT9 was further revealed by GWAS in a search for gene loci associated with urate levels. In these studies, SLC2A9 was identified as the most important locus (574, 577–581). This locus has also been found to be significantly associated with gout (579).

Elevated urate levels and gout prevalence have been linked to genetic variation not only in SLC2A9, but also in BCRP (ABCG2) and SLC17A3, the latter encoding the NPT4 anion transporter (578). All three genes are involved in renal urate transport. In the small intestine, basolateral GLUT9 (SLC2A9) mediates urate uptake into enterocytes, which is coupled to apical efflux via ABCG2, a mechanism well established in mice (582). Data from the HPA indicate that both basolateral GLUT9 and apical ABCG2 are abundantly expressed in the human intestine, making it likely that the same urate efflux pathway operates in the human small intestine (583).

By contrast, loss-of-function variants in the renal urate reuptake transporter URAT1 (SLC22A12) are substantially protective against gout (584).

As indicated above, GLUT9 has also been shown to contribute to glucose and fructose transport, although urate is recognized as the major substrate. The ability of renal GLUT9 to exchange glucose for urate may explain the known correlation between glycosuria in diabetes and the reduction of plasma urate levels (585). Increased urinary glucose levels have been proposed to accelerate GLUT9-mediated urate efflux across the apical membrane of the renal proximal convoluted tubule (574). A direct confirmation of this hypothesis is given by the recent observation that treatment of diabetic patients with the antidiabetic drug empagliflozin inhibiting SGLT2 (SLC5A2) is effective in controlling the patients’ hyperuricemia (586).

Another study uncovered an interesting link between renal GLUT9 secretion, iodine intake, hyperuricemia, and preeclampsia. First, it was shown that the N-terminal domain of human GLUT9a has a unique effect on transport function and increases interaction with small negatively charged ions such as chloride (587). Experiments showed that uric acid transport mediated by GLUT9a, but not by GLUT9b, is chloride dependent. Interestingly, replacing chloride with iodide resulted in a loss of currents for GLUT9a but not for GLUT9b. Iodide inhibits GLUT9a with an IC₅₀ of approximately 35 μM. The results of this study have implications for hyperuricemia-associated diseases, such as preeclampsia during pregnancy. Several studies have shown an association between iodine deficiency and preeclampsia (588, 589). Thus, it is hypothesized that the function of the renal GLUT9 (SLC2A9) uric acid transporter is regulated by iodide. Based on this finding, it can be concluded that iodine intake is beneficial in individuals with human hyperuricemic diseases such as preeclampsia. In fact, this has already been suggested in another study (590).

In terms of GLUT9 expression in spermatids, uric acid helps to protect sperm function and improve sperm viability, motility, and fertilization ability (591). It does this by helping to eliminate endogenous free radicals and by enhancing certain specific enzymes that are considered critical for sperm function. For example, urate serves as a cosubstrate for the enzyme cyclooxygenase, which produces prostaglandins and thromboxane (592). The enzyme is required for proper sperm function and the fertilization process (591).

Cryo-EM structures of GLUT9 (SLC2A9) with and without urate have been reported. The studies reveal the effect of SLC2A9 genetic variants that predispose to disease (593). Additional cryo-EM structures of GLUT9 in apo- and urate-bound states have been reported (594). They reveal a unique urate recognition mode of GLUT9 that is distinct from the substrate recognition mode of conventional GLUTs. The findings provide insight into the substrate specificity conferred to this member of the GLUT family and provide novel insights into the development of uricosuric drugs targeting GLUT9.

SLC2A11: GLUT11 (SLC2A11), like GLUT7, can transport both glucose and fructose with relatively low K_m values when expressed in Xenopus oocytes (500) and has no rodent ortholog. It is expressed as 3 sequence variants, GLUT11-A, GLUT11-B and GLUT11-C, which differ at the N-terminus and probably result from different promoter usage. The 3 GLUT11 variants are differentially expressed (595) and their functional, cellular and subcellular properties were subsequently determined (596). GLUT11-A is present in heart, skeletal muscle and kidney, GLUT11-B in placenta, adipose tissue and kidney and GLUT11-C in adipose tissue, heart, skeletal muscle and pancreas (596). However, the physiological roles of these GLUT11 (SLC2A11) variants are still unknown and require further investigation.

Class III – transporters with intracellular targeting signals and a large glycosylated loop exposed to the extracellular space or organellar lumen

These transporters include SLC2A6 (GLUT6), SLC2A8 (GLUT8), SLC2A10 (GLUT10), SLC2A12 (GLUT12) and SLC2A13 (HMIT)

SLC2A6 - Orphan transporter: GLUT6 (SLC2A6) is ubiquitously expressed, particularly in the spleen, brain, and leukocytes, as well as in muscle and adipose tissue (450, 597). It has been shown to be localized intracellularly in the lysosome of mouse macrophages, to be regulated by inflammatory stimuli and to be involved in glycolysis in macrophages (598).

Data show that this transporter is not a major regulator of systemic metabolic physiology, as Slc2a6 knockout in mice had minimal effects on whole body metabolic physiology (599).

In the pancreas, the transporter has been shown to play a role in islet insulin secretion, although the transporter was not directly involved in glucose-stimulated insulin secretion in human islets, but further investigation is still needed to elucidate the precise role of GLUT6 in islet function (600).

Using transcriptomics to reveal the genetic mechanisms regulating myoblast differentiation in a classical myogenic differentiation model, Slc2a6 was identified in genetically diabetic db/db mice. The transporter was shown to be involved in the regulation of skeletal muscle myogenesis and to influence muscle development by targeting the glycolysis-related gene LDHB (lactate dehydrogenase B) (601). The study revealed that the Slc2a6-LDHB pathway may be an upstream regulator of myogenic differentiation, but the underlying details and how Slc2a6 is involved still need further investigation.

SLC2A8: GLUT8 (SLC2A8) is highly expressed in testis and at lower levels in cerebellum, brainstem, hippocampus, hypothalamus, adrenal gland, liver, spleen, brown and white adipose tissue and lung (602). GLUT8 (SLC2A8) has a dileucine internalization and intracellular retention motif that confines it to the endosomal/lysosomal compartments (602, 603), and when mutated it can be expressed at the cell surface in Xenopus oocytes or mammalian cells and functions as a glucose transporter (K_m ~2 mM) (602). Fructose and galactose could also significantly inhibit glucose uptake, suggesting that fructose and galactose might also serve as substrates for GLUT8. Fructose transport by GLUT8 was also measured in another study using ¹⁴C-fructose radioisotope uptake studies (604).

New information on the physiological role of GLUT8 comes from a recent study showing that GLUT8 mediates fructose uptake in the liver in response to a high-fructose diet (550). The study further found that fructose taken up by hepatocytes leads to triacylglycerol accumulation and lipogenesis based on experiments in mice and HepG2 cells. The question then arises as to how GLUT8 moves from its endosomal/lysosomal localization to the plasma membrane to facilitate fructose uptake. The study reveals that this is enabled by reduced expression of the membrane protein TM4SF5 in hepatocytes during a high-fructose diet. TM4SF5 is a membrane glycoprotein with four transmembrane segments (605) that has been implicated in non-alcoholic fatty liver disease (NAFLD). TM4SF5 binds to several partners, including epidermal growth factor receptor and mTOR. Interestingly, it also binds to several SLC solute carriers such as the glucose/fructose transporter GLUT8 (SLC2A8), the heterodimeric cystine-glutamate exchanger 4F2h (SLC3A2)/xCT (SLC7A11), and the liver fatty acid transporters FATP2 (SLC27A2) and FATP5 (SLC27A5). TM4SF5 forms massive protein-protein complexes in plasma and endosomal membranes called TM4SF5-enriched microdomains.

As mentioned above, dietary fructose has been shown to induce loss of TM4SF5 binding to GLUT8 by reducing TM4SF5 expression (550). This in turn triggers the translocation of GLUT8 from its intracellular lysosomal location to the plasma membrane, allowing fructose to be taken up and metabolized by hepatocytes, leading to lipogenesis (550). Thus, excessive fructose consumption leads to excessive lipogenesis, which can ultimately lead to hepatic steatosis. Given the clinical significance of this finding, TM4SF5 and GLUT8 have been proposed as promising therapeutic targets for the treatment of hepatic steatosis.

Regarding the high expression of GLUT8 in the testis, expression is restricted to Leydig cells (606) and germ cells, with highest expression in late spermatids according to the HPA. Fructose is an important source of energy for ATP production in sperm and is essential for sperm viability and motility (607). Leydig cells could in principle synthesize fructose from glucose as a source of energy for sperm production (608) but it is unclear how the fructose would be delivered from Leydig cells across the blood-testis barrier to the spermatids.

SLC2A10: GLUT10 (SLC2A10) is a transporter of glucose and dehydroascorbate (DHA). Northern blot analysis of human tissues revealed high expression in liver and pancreas and at lower levels in placenta, kidney, heart, and lung (609). The HPA indicates that GLUT10 (SLC2A10) is most highly expressed in liver, gastrointestinal tract, female and male tissues, choroid plexus, endocrine tissues, heart and smooth muscle (including smooth muscle cell-enriched organs such as the aorta (610)).

GLUT10 was found to be mainly expressed in the endoplasmic reticulum (611). While other GLUT/SLC2 transporters can transport DHA into cells, such as GLUT1 (SLC2A1) into erythrocytes (483), GLUT10 can import cytosolic DHA into the ER, where it can be converted by reduction to vitamin C by glutathione and other thiols (611, 612).

GLUT10 (SLC2A10) has been reported to localize to the mitochondria of smooth muscle cells and insulin-stimulated adipocytes, thus delivering DHA directly to the mitochondria for reduction back to vitamin C with the goal of protecting the mitochondria (613). In support of this, missense mutations in SLC2A10 cause hereditary arterial tortuosity syndrome (ATS), a rare congenital connective tissue disorder (614), because they lead to accumulation of reactive oxygen species. It was concluded that GLUT10 maintains the integrity of large arteries by maintaining redox homeostasis and mitochondrial function (615).

In a follow-up study, the clinical implications of SLC2A10 mutations were presented in a cohort of 21 patients receiving medical care in the Qatari Healthcare System who received a genetic diagnosis after detection of SLC2A10 mutations (616). While the outcome of ATS in these patients is apparently mild, the presence of comorbidities has had a significant impact on the quality of life of these individuals. Craniofacial features and connective tissue manifestations have been commonly observed in individuals with ATS. The aforementioned role of GLUT10 in recycling DHA to vitamin C makes this transporter important for vitamin C-dependent hydroxylation of prolyl and lysyl residues, as part of the proper maturation of collagen. This would explain why GLUT10 dysfunction contributes to the phenotypic expression of ATS. In addition, vitamin C has also been reported to inhibit collagen degradation through the TGF-β signaling pathway (617, 618). Thus, loss of GLUT10 function may also potentially increase collagen degradation and disruption of extracellular matrix proteins that are critical for the structural integrity of blood vessel walls and other connective tissues (616).

GLUT10-mediated DHA transport has furthermore been shown to influence adipogenesis via vitamin C-dependent DNA demethylation. Vitamin C acts as a cofactor for TET enzymes (619), a family of ten-eleven translocation (TET) methylcytosine dioxygenases that catalyze DNA demethylation in order to regulate gene expression. TET-regulated gene expression has been reported to be important for the proper development of white adipose tissue to protect against high-fat diet-induced metabolic dysregulation (620). Specifically, studies in mice demonstrated that Glut10 (Slc2a10) regulates adipogenesis via ascorbic acid-dependent DNA demethylation, highlighting the importance of GLUT10 as a high-fat diet-associated susceptibility locus for T2D (620).

While high expression levels of GLUT1 (SLC2A1) and/or GLUT3 (SLC2A3) are known to be associated with poor survival in many cancer types, leading to increased aggressiveness and invasiveness of tumors (621), a recent study uncovered GLUT10 as a novel immune regulator involved in lung cancer immune cell infiltration (622). Through transcriptome experiments, database analysis and studies using human samples, GLUT10 was shown to be a novel immune signaling molecule involved in tumor immunity, particularly in immune cell infiltration of lung adenocarcinoma. Knockdown of SLC2A10 broadly activated immune and inflammatory signaling. GLUT10 is aberrantly expressed in several tumors, and the expression levels of GLUT10 (SLC2A10) were closely correlated with cancer prognosis. Low SLC2A10 expression was associated with poorer prognosis and increased malignancy of lung cancer. GLUT10 expression was shown to be closely associated with the infiltration of various types of immune cells, particularly macrophages. GLUT10 may modulate immune cell infiltration of lung adenocarcinoma through the cyclooxygenase-2 (COX-2, PTGS2) pathway and prostaglandin E₂ (PGE₂), known to be important inflammatory players associated with cancer cell survival, invasion, growth and immune escape (622, 623).

SLC2A12: GLUT12 (SLC2A12) was identified in MCF-7 breast cancer cells by homology to the insulin-regulated glucose transporter GLUT4 (SLC2A4) (624). In normal adult human tissues, GLUT-12 was found to be prominently expressed in skeletal muscle and fat (624). According to the HPA, GLUT12 (SLC2A12) is highly expressed in the choroid plexus, and at lower levels in stomach, small intestine, prostate, muscle, and kidney. It is also expressed in syncytiotrophoblasts where it can be activated by insulin (473). Further studies revealed that GLUT12 localizes to the Golgi apparatus (625) with its subcellular targeting directed by a dileucine motif within its N-terminal cytoplasmic domain (626).

Analysis of glucose transport in GLUT12 in cRNA-injected Xenopus oocytes showed that the uptake of radiolabeled 2-deoxy-D-glucose was highly increased compared to water-injected control oocytes (627, 628). When 2-deoxy-D-glucose uptake was measured in the presence of different sugars, the substrate selectivity order was D-glucose > 2-deoxy-D-glucose > D-galactose > D-fructose > L-glucose. Furthermore, the transport of 2-deoxy-D-glucose into the oocyte was inhibited by cytochalasin B, which can directly bind to GLUT4 (628). Based on studies in MCDK overexpressing GLUT12, glucose transport by GLUT12 is H⁺-coupled (629). Further studies using the Xenopus oocyte expression system showed that glucose transport by GLUT12 increased in the presence of Na⁺ (627). Confirming that the GLUT12 transport mechanism is coupled to ion movement, electrophysiological analysis revealed that GLUT12-mediated glucose transport is electrogenic (627).

A subsequent study has shown that SLC2A12 is associated with serum urate levels, susceptibility to gout and hyperuricemia (630). GLUT12 has subsequently been found to function as a physiological urate transporter (631), and its dysfunction has been demonstrated to increase the blood urate concentration (631). But because this transporter is relatively weakly expressed in the kidney and liver, it is unclear how it would affect urate levels in the body. Since GLUT12 is expressed in muscle, one might speculate that it is involved in the transport of uric acid produced in muscle into the blood. In any case, whether urate is really a major physiological substrate of GLUT12 should be considered with caution because urate transport mediated by GLUT12 transiently expressed in HEK293 cells did not show saturation under the experimentally used maximal urate concentration (500 μM), suggesting a much lower urate affinity of the transporter compared to that of the urate transporter GLUT9 (SLC2A9).

Previous studies found that insulin stimulates the translocation of GLUT12 from its intracellular membrane compartments to the plasma membrane in human skeletal muscle cells, similar to GLUT4 (SLC2A4) (632). This finding would be more consistent with the role of GLUT12 as a sugar transporter. Changes in GLUT4 (SLC2A4) and GLUT12 (SLC2A12) expression have been described during exercise (633) and in the development of insulin resistance (634). If there is a loss of function of GLUT4, as observed in human adipose tissue in insulin resistance (449), the question is whether this loss could be compensated by an increase in GLUT12 expression and whether the response to insulin would be the same (627).

Regarding fetal development, studies have shown that during the gestational period in rats from day 15 to 21, GLUT12 is expressed in insulin-sensitive tissues such as heart, skeletal muscle and brown adipose tissue, suggesting that it plays a role during fetal development (635).

Interestingly, a follow-up study identified GLUT12 (SLC2A12) as an exporter that regulates vitamin C supply from the blood to the brain (636). Using cell-based transport analyses of vitamin C efflux and Slc2a12 knockout mice, GLUT12 was identified as a physiologically important vitamin C efflux protein expressed in the choroid plexus, where it is highly expressed (636) (see Fig. 8). Previously, SVCT2 (SLC23A2) was found to mediate the transport of vitamin C from the blood to the cytoplasm on the basal (blood) side of the plasma membrane in choroid plexus epithelial cells (140, 637, 638). GLUT12 (SLC2A12) would then be expected to facilitate the apical exit of vitamin C into the cerebrospinal fluid (see Fig. 11).

GLUT12 has also been shown to be highly upregulated in breast ductal carcinoma (639), while SVCT2 was found to be downregulated (640). Since vitamin C promotes apoptosis in breast cancer, decreased vitamin C uptake via SVCT2 and increased release via GLUT12 would be consistent with the goal of cancer cells to promote survival and growth.

SLC2A13: HMIT (SLC2A13) is an H⁺-coupled myo-inositol transporter (641). High expression is found in the cerebral cortex, hippocampus, hypothalamus, brainstem, and other brain regions (450), as well as in endocrine tissues (especially parathyroid gland), gastrointestinal tract (stomach, small intestine, colon), kidney, placenta, and heart muscle. To achieve maximum plasma membrane expression using the Xenopus oocyte expression system, it was necessary to mutate two internalization motifs and one ER retention signal (450). Transport activity was specific for myo-inositol, and it was strongly activated by acidifying the extracellular medium. The K_m for myo-inositol was 100 μM. Transport was inhibited by phloretin, phlorizin, and cytochalasin B and no glucose transport activity was detected.

In the brain, HMIT has been detected in both neurons and glial cells (450). In neurons, it was identified in intracellular vesicles. These vesicles can be stimulated to translocate and fuse with the plasma membrane to allow cellular uptake of myo-inositol. HMIT is translocated at synapses and growth cones, induced by neuronal activation, increased Ca²⁺ influx or protein kinase C activation (642).

One of the primary functions of HMIT is to supply myo-inositol, which is necessary for the creation of phosphatidylinositol, a critical molecule for intracellular signaling (642).

HMIT was furthermore identified as a novel γ-secretase associated protein (643). γ-Secretase is a multi-subunit protease complex and functions as an integral membrane protein that cleaves single-pass transmembrane proteins at residues within the transmembrane domain (644). It also catalyzes the formation of amyloid β-peptide (Aβ), which has a well-known pathogenic role in Alzheimer disease. When searching for genes and proteins that reduce the production of Aβ upon treatment with siRNA, the researchers identified HMIT and phospholipid-transporting ATPase IIA (ATP9A), along with a few others, as potential candidates (643). This groundbreaking discovery suggests that HMIT may be a valuable target for Aβ-reducing therapy to treat Alzheimer disease.

The precise role of HMIT (SLC2A13) in other tissues is yet to be determined.

Orphan transporter family member (1)

SLC2A6 (GLUT6)

SLC3 Heavy subunits of the heteromeric amino acid transporters (8.A.9/SLC3A2_N/single TMH)

Discovery: Rat and rabbit kidney heavy chain D2/NAA-Tr/rBAT (SLC3A1) (135, 143, 144) are the founding members of the SLC3 family. Expression cloning by three different groups simultaneously identified the protein.

Gene family members (2)

SLC3A1 (rBAT), SLC3A2 (4F2hc)

Molecular aspects, physiological roles and links to disease

The SLC3 family belongs to the rBAT Transport Accessory Protein Family (TC 8.A.9), where rBAT stands for “related to b^0,+ amino acid transport.” The two sole human members of the SLC3 family are single transmembrane domain type II glycoproteins that serve as ancillary proteins to allow functional expression of certain amino acid transporters of the SLC7 family (194, 645) (Fig. 9). Specifically, SLC3 family members serve as part of heterodimeric amino acid transporter (HAT) complexes. These are composed of a light subunit (SLC7 family member) and a heavy subunit (SLC3 family member) linked by a disulfide bridge (Fig. 9).

Fig. 9. Heterodimeric cystine, dibasic amino acid and neutral amino acid exchanger.

The exchanger consists of two subunits, rBAT and b^0,+, encoded by SLC3A1 and SLC7A9, respectively. a) The rBAT–b^0,+ heterodimer is localized at the apical membrane of the renal proximal tubules where it is responsible for the reabsorption of cystine together with the other dibasic amino acids. Genetic variants causing cystinuria have been identified in both the SLC3A1 and SLC7A9 genes. b) rBAT mediates trafficking of the amino acid exchanger b^0,+ to the plasma membrane. The two subunits rBAT and b^0,+ are linked by a disulfide bridge (S-S). c) LAPTM4b has been shown to recruit the LAT1-4F2hc (SLC7A5-SLC3A2) complex from the plasma membrane to the lysosome. This increases the uptake of leucine into lysosomes, thereby raising their intracellular levels. This, in turn, results in stronger activation of V-ATPase and, via Ragulator, the positioning of mTORC1 on the lysosome (see the SLC38A9 description and Fig. 45 for further details). It is believed that cancer cells hijack this process to boost anabolic signaling and support tumor growth (18). Figure created de novo; elements of the conceptual framework are based on multiple sources, including (18).

SLC3 homologs are found throughout the metazoans of the animal kingdom. Both rBAT (SLC3A1) and 4F2hc (SLC3A2) have a short N-terminal domain, a TMH, and a bulky extracellular N-glycosylated domain. The latter shows similarity to a family of α-glucosidases from both prokaryotic and eukaryotic organisms (135), but no α-glucosidase activity could be demonstrated for any of the SLC3 family members (646). Thus, despite cryo-EM structural information, the physiological roles of these glucosidase homologies remain unclear.

SLC3A1: D2/NAA-Tr/rBAT (SLC3A1) is a type II membrane glycoprotein expressed mainly in the apical membrane of the small intestine and kidney (135, 144, 647). At a lower level, expression is also present in pancreas and liver (648). D2/NAA-Tr/rBAT (SLC3A1) has a single transmembrane domain and a large extracellular C-terminal domain that is N-glycosylated and shows similarity to the extracellular region of the 4F2 heavy chain cell surface antigen and, as already noted, to a family of α-glucosidases (135) (Fig. 9). It forms a heterodimeric complex with the non-N-glycosylated cystine, dibasic amino acid and neutral amino acid transporter b^0,+ (SLC7A9) (194, 365). It is essential for trafficking of the SLC7A9 amino acid transporter to the plasma membrane.

Cystinuria is an inherited disease characterized by impaired renal reabsorption of cystine and other dibasic amino acids in the proximal renal tubule and in the epithelial cells of the gastrointestinal tract (Fig. 9). The resulting renal hyperexcretion of cystine causes it to precipitate in the distal tubule, forming cystine stones (649). The heterodimeric transporter rBAT-b^0,+, composed of the cystine, dibasic, and neutral amino acid transport subunits encoded by SLC3A1 and SLC7A9, is localized to the apical membrane of the renal proximal tubules, where it reabsorbs cystine along with the dibasic amino acids arginine, ornithine, and lysine. Genetic variants that cause cystinuria have been identified in both the SLC3A1 and SLC7A9 genes (194, 365). While mutations in SLC3A1 are usually inherited in an autosomal recessive manner, SLC7A9 mutations show a wide variability of inheritance, ranging from autosomal recessive to dominant.

SLC3A2: 4F2hc (SLC3A2), also named CD98, is also a type II membrane glycoprotein. Like SLC3A1, SLC3A2 is essential for trafficking the corresponding amino acid transporters to the plasma membrane (650). 4F2hc (SLC3A2) is responsible for trafficking by heteromerization with the amino acid transporters LAT1 (SLC7A5), y⁺LAT2 (SLC7A6), y⁺LAT1 (SLC7A7), LAT2 (SLC7A8), asc-1 (SLC7A10) and xCT (SLC7A11) (194). Like rBAT (SLC3A1), 4F2hc (SLC3A2) has extracellular domains with homology to α-glucosidases (650).

4F2hc (SLC3A2) is ubiquitously expressed in normal tissues and has been found to be expressed in several tumor cell lines (651). 4F2hc was first identified as a surface antigen typical of activated lymphocytes, responsible for cell proliferation and growth (652–654). Subsequently, 4F2hc was proposed to be involved in cell adhesion and fusion pathways, particularly through integrin signaling, via both its extracellular and intracellular parts, modulating mechanical and metabolic signaling (655–658). Later, it was found that the metabolic changes that natural killer cells undergo during inflammation to support their high energy requirements for effector function and proliferation include interleukin-18-mediated upregulation of the heterodimeric amino acid transporter 4F2hc (SLC3A2) -LAT1 (SLC7A5), thereby facilitating amino acid-induced recruitment of mTORC1 to the lysosome (659).

Studies in cancer cells have shown that the lysosomal protein LAPTM4B (see Section 10, SLC-Like Proteins) recruits LAT1/4F2hc from the plasma membrane to the lysosome (Fig. 9C). This results in leucine being taken up by the lysosomes, which triggers recruitment of mTORC1 to the lysosome via the V-ATPase/Ragulator pathway (Fig. 9C) (18). This process switches the cell from catabolism to anabolism and links leucine uptake to growth regulation (660, 661).

4F2hc was also found to be involved in herpes simplex virus 1 (HSV-1) de-envelopment, specifically showing that HSV-1 recruits 4F2hc and β1 integrin to the nuclear membrane for viral de-envelopment (662). 4F2hc has also been shown to facilitate parasite entry into cells. For example, Plasmodium vivax was shown to directly bind host 4F2hc/CD98hc (SLC3A2) to enter immature red blood cells (663).

4F2hc (SLC3A2) has 4 asparagine residues, N365, N381, N424, N506, which represent N-glycosylation sites on its extracellular surface. The role of N-glycosylation on 4F2hc trafficking and stability was investigated using single, double, triple and quadruple mutants (664). Only the quadruple mutant severely impaired both stability and trafficking of 4F2hc to the plasma membrane. The reduced presence of 4F2hc at the plasma membrane correlated with a reduced presence of LAT1 (SLC7A5) and its trafficking activity (664). This finding potentially opens new perspectives for human therapy where inhibition of 4F2hc (SLC3A2) trafficking would act synergistically with LAT1 (SLC7A5) inhibitors in clinical trials for cancer therapy (665) (see the SLC7 family description).

Orphan transporter family members: N/A

SLC4 Bicarbonate transporter family (2.A.31/HCO3_cotransp/NAT)

Discovery: The origin of the discovery of the first SLC4 family member, anion exchanger AE1 (SLC4A1), was “Band 3”, which appeared as a prominent band on SDS-polyacrylamide gel electrophoresis of unpurified erythrocyte membrane proteins (for review, see (666)). This finding provided a first unique clue to the characterization of this protein, as it appeared to be abundant enough for an in-depth biochemical approach. Band 3 turned out to be a predominant glycoprotein expressed on the erythrocyte membrane, where it mediates chloride/bicarbonate anion exchange across the plasma membrane, a process necessary for efficient respiration. These observations led to the molecular identification of the erythrocyte Cl^-/HCO₃^- exchanger AE1 (SLC4A1) as the first identified member of the SLC4 family (119). A cDNA library from the spleen of mice subjected to anemic stress was screened with a polyclonal antibody against purified mouse erythrocyte Band 3.

This success made AE1 one of the best studied membrane transport proteins. AE1 plays a key role in the carriage of carbon dioxide (CO₂) from the systemic capillaries to the pulmonary capillaries (667, 668). Its discovery paved the way to identifying two other members of this protein family, AE2 (SLC4A2) and AE3 (SLC4A3). The identification of additional members of the SLC4 family, in particular the four Na⁺-coupled bicarbonate transporters, required a separate cloning strategy due to their distant phylogeny. Expression cloning of the salamander renal electrogenic sodium bicarbonate cotransporter NBC1 cDNA, corresponding to human NBCe1 (SLC4A4), was successfully used to identify the first member of the Na⁺-coupled bicarbonate transporter subfamily (153).

Gene family members (10):
SLC4A1 (AE1)	SLC4A4 (NBCe1)	SLC4A8 (NDCBE)	SLC4A11 (BTR1)
SLC4A2 (AE2)	SLC4A5 (NBCe2)	SLC4A9 (AE4)
SLC4A3 (AE3)	SLC4A7 (NBCn1)	SLC4A10 (NBCn2)

Molecular aspects, physiological roles and links to disease

The SLC4 family has 10 members, eight of which are HCO₃^- transporters [AE1 (SLC4A1), AE2 (SLC4A2), AE3 (SLC4A3), NBCe1 (SLC4A4), NBCe2 (SLC4A5), NBCn1 (SLC4A7), NBCBE (SLC4A8) and NBCn2 (SLC4A10), one is an acid-base sensor [AE4 (SLC4A9)] and one is an NH₃-stimulated H⁺ (OH^-) transporter [BTR1 (SLC4A11)]. SLC4A6 does not exist (669).

The SLC4 family belongs to the anion exchanger family (TC 2.A.31), a subfamily of the amino acid-polyamine-organocation (APC) superfamily. Members of the AE family are found in animals, plants and yeast. SLC4 transporters typically form dimers, and each SLC4 monomer consists of an N-terminal cytoplasmic core domain, which is involved in interactions with cytoskeletal proteins and plays a cell structural role, followed by a large transmembrane domain involved in transport function. Structural insights revealed a 7+7 TMH inverted repeat fold (223) (see Section 8).

The SLC4 family can be grouped into the following subfamilies based on the phylogenetic tree and functional properties (Fig. 10) (670):

Fig. 10. Phylogenetic tree of the SLC4 family.

• 1) The Cl⁻/HCO₃⁻ exchangers AE1-3 (SLC4A1-3)
• 2) The electrogenic Na⁺/HCO₃⁻ cotransporters NBCe1 (SLC4A4) and NBCe2 (SLC4A5)
• 3) The electroneutral Na⁺-coupled HCO₃^- transporters NBCn1 (SLC4A7), NDCBE (SLC4A8), and NBCn2 (SLC4A10)
• 4) AE4 (SLC4A9)
• 5) BTR1 (SLC4A11)

1) The Cl⁻/ HCO₃⁻ exchangers (SLC4A1-3)

SLC4A1: AE1 (SLC4A1), also known as Band 3, is an anion exchanger mainly expressed in erythrocytes and renal cells. SLC4A1 facilitates the exchange of Cl⁻ and HCO₃⁻ and also anchors the cytoskeleton to the plasma membrane, thus maintaining the stability of the erythrocyte membrane (671). In erythrocytes it plays a crucial role in the respiratory system and in the regulation of the acid-base balance of the extracellular fluid (672). In fact, the two key molecular elements in erythrocytes are HbA (hemoglobin) and AE1 (SLC4A1). HbA is present in high concentrations and serves as the main H⁺ buffer of erythrocytes. In the capillaries of the lungs, erythrocytes take up oxygen and release carbon dioxide (step 1). In other tissues of the body, the reverse reaction occurs; erythrocytes take up carbon dioxide and release oxygen (step 2). These steps are detailed as follows.

Step 1: In the lungs, O₂ enters the erythrocytes via lipid permeation. For this purpose, the permeability of the erythrocyte plasma membrane is kept high due to high membrane cholesterol content, allowing complete O₂ loading during the passage of erythrocytes through the lung capillaries (673). Inside the erythrocyte, CO₂ bound as carbamate to the N-terminus of HbA and protons bound to histidine residues of HbA are released into the cytosol, allowing HbA to bind O₂. Next, bicarbonate (HCO₃⁻) present in plasma is taken up by erythrocytes via AE1 (SLC4A1) and combined with protons from HbA by carbonic anhydrases I and II (CA1, CA2) to form water and CO₂. CO₂ then passively exits the erythrocyte via aquaporin 1 (AQP1) or the Rh-associated glycoprotein RhAG (SLC42A1), followed by removal of CO₂ by exhalation. Plasma HCO₃⁻ can also be directly dehydrated to CO₂ by extracellular carbonic anhydrase IV (CA4) located on endothelial cells lining capillaries in the lung.

Step 2: In the tissue space, CO₂ produced, e.g., in muscles after exercise, enters the erythrocytes via AQP1 or RhAG (SLC42A1). CO₂ and H₂O form HCO₃⁻ and H⁺ via catalysis by carbonic anhydrase II. H⁺ is then bound by HbA and CO₂ interacts with the N-terminal α-amino groups of HbA to form carbamates.

This is accompanied by the release of oxygen and its diffusion into the tissues. The bicarbonate formed then exits the erythrocyte via AE1 (SLC4A1) in exchange for chloride. The association of protons with the amino acids in hemoglobin causes a conformational change in the protein, ultimately reducing the affinity of the binding sites for oxygen molecules. This reduced affinity, which is the principle of the Bohr effect, facilitates the delivery of O₂ to tissues.

AE1 (SLC4A1) is also expressed in the basolateral membrane of α-intercalated cells in the collecting duct of the kidney and in the male reproductive tract. Subsequent studies showed that the kidney AE1 is the result of alternative splicing and was therefore named kAE1 (674). It is derived from the same gene but lacks amino acids 1-65. This loss of these residues is thought to create a more stable and open structure in the kAE1 isoform compared to the erythrocyte isoform (eAE1) (675). Renal α-intercalated cells play an important role in acid excretion to regulate blood pH (676). First, inside the cells, carbonic anhydrase II (CA2) provides protons and bicarbonate by hydrating CO₂. The resulting protons are then secreted across the apical membrane into the lumen by the apical V-ATPase proton pump and the H⁺/K⁺-ATPase, with the released protons mostly buffered by ammonia or phosphate. The resulting bicarbonate is released into the circulation via basolateral kAE1.

It has been reported that AE1 (SLC4A1) is also expressed in the epididymis (677) and it has been further shown to be involved in the process of sperm capacitation and in the rearrangement of the sperm membranes, playing a role in the acrosome reaction (678).

The structure and substrate binding of AE1 (SLC4A1) and a novel inhibitor were also reported (679). Cryo-EM structures in the apo-, bicarbonate-, and inhibitor-bound states, combined with uptake and computational studies, provided new insights into substrate recognition and transport mechanisms and revealed sterol binding sites.

The turnover rate of a membrane transporter is the maximum rate at which the substrate can be translocated across the plasma membrane under substrate-saturating conditions. Erythrocyte AE1 (eAE1) has a turnover rate of ~50,000 s^-1, making it one of the fastest secondary transporters known (680). A follow-up study of the turnover rates of two other SLC4 family members, NBCe1-A (see SLC4A4 summary) and the kidney-specific basolateral anion exchanger kAE1, both of which play an important role in renal bicarbonate absorption, indicates that they also have very high turnover rates (681). These unusually high turnover rates of SLC4 proteins are proposed to result from the small vertical shift of the ion coordination site as part of the elevator-type transport mechanism, resulting in rapid protein and membrane reorganization during outward- and inward-facing transitions (681).

SLC4A2: AE2 (SLC4A2) is a widely distributed Cl⁻/HCO₃⁻ exchanger. In epithelial cells, it is expressed at the basolateral membranes (667). It is particularly highly expressed in gastric parietal cells, choroid plexus epithelial cells, small intestine and renal collecting duct cells (682). The Slc4a2^−/− mice die at or before weaning and show severe growth retardation, failure of tooth eruption, osteopetrosis and gastric achlorhydria with gastric mucosal dysplasia (683).

In gastric parietal cells, AE2 contributes to H⁺ secretion by exporting bicarbonate into the blood via Cl⁻/HCO₃⁻ exchange, which balances the H⁺ pumped into the lumen of the gastric gland, and by providing chloride for delivery into the gastric lumen, presumably via cystic fibrosis transmembrane conductance regulator (CFTR) (670, 684).

In the choroid plexus, AE2 (SLC4A2), similar to NBCn2 (SLC4A10) (see below), is highly expressed and has been localized to the basolateral membrane (685, 686) (Fig. 11). One of its main functions in the choroid plexus is to regulate intracellular pH through HCO₃⁻ efflux, which contributes to the acidification of the cytoplasm. AE2 may also provide an important pathway for Cl⁻ influx across the basolateral membrane of the choroid plexus.

Fig. 11. Membrane transporters in the choroid plexus.

The figure shows membrane transport proteins mentioned in the text that have been localized to the apical or basolateral membranes of choroid plexus epithelial cells. These cells form the barrier between blood and cerebrospinal fluid (CSF). The barrier is located along the floor of the lateral ventricles and on the roof of the third and fourth ventricles. Above the epithelial layer are the ependymal cells (not shown), which line the ventricles and the neural canal and are responsible for transporting electrolytes and solutes between the CSF and the brain parenchyma.

In the kidney, HCO₃⁻ not reabsorbed by the proximal tubules is reabsorbed by the epithelial cells of the thick ascending limb, with AE2 (SLC4A2) mediating the extrusion of HCO₃⁻ during transepithelial HCO₃⁻ reabsorption. In the thick ascending limb, NaCl reabsorption is facilitated by NKCC2 (SLC12A1) and NHE3 (SLC9A3) at the apical membrane (see Fig. 12), a process that is inhibited by the basolateral SLC4 bicarbonate transporters such as AE2 (SLC4A2), NBCn1 (SLC4A4) and NBCn2 (SLC4A5) because they increase intracellular Na⁺ and Cl⁻ concentrations. Subsequently, it was shown that under high dietary sodium conditions, these bicarbonate transporters are functionally upregulated by interaction with IRBIT (inositol 1,4,5-trisphosphate receptor-binding protein released with inositol 1,4,5-trisphosphate) and long-IRBIT, thereby inhibiting Na⁺ reabsorption. The upregulation effectively blocks the Na⁺ retention-induced increase in arterial blood pressure (687). This finding may explain why genetic variations in bicarbonate transporter genes of the SLC4 family are associated with hypertension (687).

Fig. 12. Kidney acid-base control via SLC4 bicarbonate transporters and SLC42 Rhesus glycoprotein ammonium transporters.

The kidney contributes to acid-base homeostasis through the recovery of filtered bicarbonate in the proximal tubule. In the distal tubule, intercalated cells generate new bicarbonate, which is consumed by the titration of non-volatile acid. Proximal tubular acidosis results from dysfunction of the proximal tubule, where the majority of bicarbonate is reabsorbed. Intercalated cells are also involved in the excretion of ammonia/ammonium.

In bone, AE2 (SLC4A2) is localized to the contra-lacunar plasma membrane of osteoclasts where lacunar acidification involves HCO₃⁻ extrusion and Cl⁻ loading via the AE2 anion exchanger to sustain bone resorption. Consistent with this, targeted Slc4a2^−/− mice exhibited severe osteopetrosis.

Additional roles of AE2 in various organs such as intestine, liver, pancreas, esophagus have been reviewed elsewhere (688).

The cryo-EM structures of human AE2 have been reported in five major operating states, revealing an interlock mechanism for interactions between the cytoplasmic N-terminal domain and the transmembrane domain, and the self-inhibitory effect of the C-terminal loop (689). This work provides the structural basis for the pH-balancing activities of AE2.

SLC4A3: The anion exchanger AE3 (SLC4A3) is mainly expressed in excitable tissues such as brain (690, 691), retina (692) and heart muscle (693, 694). According to the HPA, it is also expressed in the ovary and the pituitary gland. It extrudes intracellular HCO₃⁻ in exchange for extracellular Cl⁻. The SLC4A3 gene encodes two variants of AE3, a full-length brain AE3 and a cardiac AE3. An SLC4A3 genetic variant has been shown to be associated with epilepsy, likely causing neuronal changes in cell volume and abnormal intracellular pH, resulting in hyperexcitability and generation of seizures (695). In support of this association, the grossly normal Slc4a3^−/− mice show increased susceptibility to pharmacologically induced seizures (686, 696).

Strikingly, the mice did not show a cardiac phenotype despite the high expression of Ae3 (Slc4a3) in the myocardium. A distinct Ae3 isoform is also expressed in Müller cells and horizontal neurons of the retina (692). One line of Slc4a3^−/− mice showed late-onset retinal degeneration and progressive blindness (697). AE3 (SLC4A3) deficiency in the retina might alter the potential of the chloride balance to reduce γ-aminobutyric acid (GABA)- or glycine-mediated inhibitory input (686).

2) The electrogenic Na⁺/HCO₃⁻ cotransporters NBCe1 (SLC4A4) and NBCe2 (SLC4A5)

SLC4A4: NBCe1 (SLC4A4) is an electrogenic Na⁺/HCO₃⁻ cotransporter that is highly expressed in kidney, pancreas and brain (670). Under normal conditions, NBCe1 (SLC4A4) mediates cellular acid extrusion, which is achieved in human tissues by different splice variants: NBCe1-A, which is expressed in the basolateral membrane of the renal proximal tubules and mediates the basolateral exit of bicarbonate into the blood; NBCe1-B, which is present in several organs, including in particular pancreatic ductal cells, where it accumulates bicarbonate in the lumen of the exocrine ducts; and NBCe1-C, which is specifically expressed in the brain, where it may regulate intracellular pH in astrocytes (670). Two additional variants have been identified in mouse reproductive tissues, NBCe1-D and NBCe1-E (698). According to this study, three variants of NBCe1 have been reported to be present in the kidney: NBCe1-A, NBCe1-B and NBCe1-D.

In the renal proximal tubules, approximately 80% of filtered Na⁺/HCO₃⁻ is reabsorbed. NBCe1-A is the major basolateral efflux transporter involved in this process and is predominantly expressed in the S1 and S2 segments of the proximal tubule (Fig. 12). The driving force for Na⁺/HCO₃⁻ reabsorption is provided by protons secreted by the apical Na⁺/H⁺ exchanger (NHE3/SLC9A3). As a first step in the reabsorption process, Na⁺/HCO₃⁻ is converted to CO₂ and H₂O, catalyzed by luminal carbonic anhydrase IV (CA4). The resulting CO₂ then diffuses into the tubular cells. This may occur either by lipid permeation or by gas-permeable membrane proteins such as AQP1 or even by NBCe1. Looking at this step in a very general context, although the background membrane gas permeability in cell membranes of certain cell types may be sufficiently high to meet physiological demands, the central pores of gas-permeable membrane proteins such as AQPs and Rh proteins are believed to be the major pathways for dissolved gases to cross biological membranes (for a review, see (699)). This is particularly important for gastric gland apical membranes, RBC membranes, renal proximal tubule apical membranes, and oocyte membranes, which have been shown to have low CO₂ or O₂ permeabilities (699). Since AQP1 was found to be widely expressed in the apical and basolateral membranes of renal proximal tubule cells (700), it would be ideally suited to deliver CO₂ into the renal proximal tubule cells. Once inside the cells, CO₂ combines with H₂O to form Na⁺/HCO₃^- + H⁺ catalyzed by cytoplasmic CA2. H⁺ then returns to the tubular lumen via the NHE, while Na⁺/HCO₃^- exits the cell via NBCe1-A at the basolateral membrane. NBCe1-A has a Na⁺:HCO₃⁻ coupling ratio of 1:2 in the proximal renal tubules (701).

NBCe1 (NBCe1-A / NBCe1-B) has the ability to bind two ions of either HCO₃^- or CO₃^2- (701). Specifically, it has been proposed that NBCe1 moves 1Na⁺ + 1HCO₃^- + 1CO₃^2- when operating in the efflux mode, whereas it moves 1Na⁺ + 2HCO₃^- when operating in the influx mode. It has also been shown that NBCe1 has unique ordered substrate binding kinetics, with binding of HCO₃^- preceding binding of Na⁺ (701). This kinetic order (2HCO₃^-→1Na⁺) is different from that of other Na⁺-coupled transporters such as SGLT1/SLC5A1 (2Na⁺→1glucose) (214, 702), the high affinity glutamate transporters of the SLC1 family (2Na⁺→1Glu→1Na⁺) (352) and the NaPi-IIb/SLC34A2 phosphate transporter (2Na⁺→1P_i→1Na⁺) (703), in which Na⁺ binding precedes the binding of the driven solutes.

Under normal physiological conditions (including normal membrane potential and electrochemical ion gradients), this stoichiometry for the bicarbonate export mode of NBCe1-A supports basolateral efflux of Na⁺ and HCO₃⁻ into the interstitial space and blood (Fig. 12).

IRBIT has been reported to substantially stimulate the transport activity of the NBCe1-B/C splice variants to approximately the same level as that of NBCe1-A (670, 704–707). The unique N-terminal variable region (Nt-VR1) of the NBCe1-B/C splice variants contains the structural determinants for IRBIT binding (Shirakabe et al., 2006). In support of the importance of NBCe1 (SLC4A4) in renal bicarbonate reabsorption, numerous SLC4A4 mutations have been reported to cause severe proximal renal tubular acidosis with excessive urinary excretion of Na⁺/HCO₃⁻ (708).

In pancreatic ductal cells, NBCe1-B is expressed at the basolateral membrane where it facilitates the accumulation of cytosolic bicarbonate, a first step towards the secretion of Na⁺/HCO₃^- from the blood into the lumen of the exocrine ducts (670). In pancreatic ductal cells, in contrast to renal proximal tubule cells, NBCe1-B acts in an influx mode instead of an efflux mode. A possible explanation may be that the depolarized membrane potential of pancreatic cells caused by CFTR activation and subsequent Cl^- secretion favors influx.

SLC4A4 has been proposed as a therapeutic target for the treatment of pancreatic ductal adenocarcinoma because an emerging key driver of cancer progression is tumor acidity, which favors the selection of malignant cancer cells and influences the tumor microenvironment (709).

NBCe1 (SLC4A4) also plays an important role in the CNS, where pH and HCO₃⁻ are important because the function of many ion channels is pH sensitive, along with the release of neurotransmitters, affecting the excitability of neurons. In rat brain, NBCe1-B and NBCe1-C were found to be the main variants expressed, where NBCe1-B is mainly distributed in neurons and NBCe1-C is expressed in glial cells (710).

SLC4A5: NBCe2 (SLC4A5), also known as NBC4, is an electrogenic Na⁺/HCO₃^- cotransporter (711). It is widely expressed in mammalian tissues (712), especially in the choroid plexus (713–717) (see Fig. 11), kidney outer medullary collecting duct (718), pelvic uroepithelium, bile duct and liver (719). According to the HPA, NBCe2 (SLC4A5) is most highly expressed in the choroid plexus and the retina; and expression is also shown in the thyroid gland, parathyroid gland, testis, mammary gland, and fallopian tube. Isoforms with different tissue distribution have been detected, resulting from alternative promoters and splicing (670).

Studies in transgenic mice have shown that NBCe2 (SLC4A5) plays a critical role in the choroid plexus and the outer blood-retinal barrier. Slc4a5^-/- mice show significant remodeling of choroid plexus epithelial cells, including abnormal mitochondrial distribution, cytoskeletal protein expression, and ion transporter polarity (715). Specifically, loss of NBCe2 (Slc4a5) resulted in severe ventricular hypovolemia and decreased intracranial pressure. In addition, these mice developed severe retinopathy with retinal detachment, which was attributed to the loss of photoreceptors and retinal ganglion cells.

SLC4A5 mutations have not been linked to any monogenic human disease, but polymorphisms have been associated with an increased risk of elevated blood pressure and an increased sensitivity of blood pressure to high sodium intake (720–723). Subsequently, a mouse mutant with retinal abnormalities (the tvrm77 phenotype) was discovered in an ocular N-ethyl-N-nitrosourea mutagenesis screen, and genetic mapping identified a cryptic splice site mutation in Slc4a5 (708, 724). These mutant mice, called tvrm77 mice, serve as an excellent model for exudative retinal detachment, retinal neovascularization and retinal dysplasia. In addition, as previously reported, Slc4a5 mutant strains exhibited abnormalities in the epithelia of the bile duct, liver, brain, and kidney, and the tvrm77 mice may be useful in defining the precise role of SLC4A5 in these other organs.

In the thyroid, NBCe2 (SLC4A5)-mediated bicarbonate transport may help maintain the alkaline intrafollicular pH that is thought to be critical for iodide coupling to thyroglobulin and internalizing iodinated thyroglobulin, possibly working together with NBCe1-B (SLC4A4) (725).

3) The electroneutral Na⁺-coupled HCO₃^- transporters NBCn1 (SLC4A7), NDCBE (SLC4A8), and NBCn2 (SLC4A10)

SLC4A7: NBCn1 (SLC4A7), also known as NBC3, is the first reported electroneutral Na⁺/HCO₃^- cotransporter (726, 727). Based on rat Northern blots, Slc4a7 mRNA is present in spleen and testis and at lower levels in heart, brain, lung, liver and kidney. In the rat kidney, Slc4a7 was found to be expressed in the basolateral membrane of the thick ascending limb of Henle (728) (Fig. 12) and in the inner medullary collecting ducts (729). Its protein expression is upregulated during metabolic acidosis, which is thought to facilitate transepithelial ammonium reabsorption and increased basolateral HCO₃^- uptake in thick ascending limbs (730). According to the HPA, NBCn1 (SLC4A7) is most highly expressed in retina, duodenum, kidney and tongue, and at lower levels in mammary gland, testis and skin, gall bladder, urinary bladder, endocrine tissues, adipose tissue, lung, brain areas, heart, skeletal muscle and smooth muscle.

SLC4A7 has two alternative promoters and up to six major cassette exons, and 18 full-length NBCn1 variants have been identified to date (731, 732). The resulting sequence variations in these isoforms have no significant functional effect, but some of them contain binding domains for IRBIT to increase functional expression (733).

Several Slc4a7^−/− mouse models have been generated that provide insight into the functional role of Slc4a7. Disruption of the gene in these different animal models resulted in visual and hearing defects and loss of duodenal bicarbonate secretion, among other symptoms. Specifically, there was progressive degeneration of photoreceptor cells in the retina (734), degeneration of hair cells in the organ of Corti and morphologic changes in the cochlear duct (735), and failure to secrete HCO₃⁻ in the duodenum (736). The latter finding suggests that NBCn1 (SLC4A7) is critical for the import of HCO₃⁻ from the blood into the duodenal epithelial cells, thereby ensuring the integrity of the duodenal mucosa. In addition, there was reduced locomotor activity in mice by affecting their exploratory behavior or emotionality (737).

SLC4A8: NDCBE (SLC4A8) is a Na⁺-driven Cl⁻/CO₃²⁻ exchanger cloned from human brain (738). It is abundantly expressed in brain, pituitary, testis, retina and at lower levels in kidney (670). NDCBE (SLC4A8) exchanges one Na⁺ ion and one CO₃²⁻ ion for one Cl⁻ ion.

The Na⁺-driven Cl⁻/CO₃²⁻ exchanger plays an important role in regulating pH in specific cells throughout the brain, kidney and the rest of the body (739). Knockout mice show abnormal regulation of Na⁺ reabsorption in the kidney and decreased neuronal excitability (740, 741).

The SLC4A8 gene contains two alternative promoters and up to four cassette exons, and five full-length NDCBE variants, NDCBE-A to -E, have been identified to date (705, 742).

The cryo-EM structure of NDCBE (SLC4A8) was subsequently reported, and the work uncovers the molecular determinants involved in NDCBE and SLC4 transport (223).

SLC4A10: NBCn2 (SLC4A10), also known as NCBE, is a Na⁺-coupled HCO₃⁻ transporter that drives cellular uptake of HCO₃⁻ and thus mediates acid extrusion (667, 743). It has been reported to be highly expressed in brain and at low levels in pituitary, testis, kidney and ileum (743). According to the HPA, it is most highly expressed in the choroid plexus, excitatory and inhibitory neurons, and retina, and at lower levels in several organs as indicated above.

Mice deficient in Slc4a10 have reduced brain ventricle size, show reduced neuronal excitability (744) and exhibit early hearing loss (745). NBCn2 (SLC4A10) was found to mediate acid extrusion in principal neurons and interneurons, in epithelial cells of the choroid plexus (744) (Fig. 11) and in inner ear fibrocytes. In mice, NBCn2 has been shown to play an important role in regulating the intracellular pH excitability of neurons. In choroid plexus epithelial cells, NBCn2 is expressed in the basolateral membrane, probably allowing transepithelial electrolyte transport and controlling CSF production (744). A failure in CSF production is likely to cause ventricular collapse in Slc4a10 knockout mice. In the inner ear, impaired pH regulation of fibrocytes may be the cause of hearing loss (745).

In humans, heterozygous genomic deletions involving all or part of SLC4A10 have been associated with autism spectrum disorder, with additional features such as impaired motor and language skills or epilepsy (746, 747). A recent study shows that autosomal recessive SLC4A10 loss-of-function mutations cause a neurological disorder associated with impaired GABAergic transmission (748). The mutants cause intellectual disability with striking radiological abnormalities of the lateral ventricles, closely resembling the phenotype seen in Slc4a10 knockout mice. Because NBCn2 (SLC4A10) is localized to inhibitory presynapses, its disruption impairs the release of GABA. Changes in the GABAergic system are proposed to contribute to the pathomechanistic basis of this developmental disorder.

The SLC4A10 gene contains three alternative promoters and up to seven cassette exons (732, 749). To date, SLC4A10 is known to express 15 “full-length” NBCn2 variants (NBCn2-A to -N, plus rb3NCBE) as well as a specific variant, rb7NCBE, containing only the N-terminal domain (705).

4) AE4 (SLC4A9)

SLC4A9: AE4 (SLC4A9) is exclusively expressed in the kidney (667). Although its function remained elusive for a long time, subsequent studies in mice showed that AE4 is localized to the basolateral membrane of β-intercalated cells in the renal collecting duct, where it serves as part of the renal sensing mechanism (transceptor function) for changes in acid-base status (see Fig. 12) (667, 750–753). In particular, based on studies in Slc4a9 knockout mice, which exhibit severe dysregulation of acid-base balance, it has been shown that AE4 is essential for the upregulation of the Na⁺-independent Cl^-/HCO₃^- exchanger pendrin (Slc26a4) and thus for the prevention of life-threatening hypochloremic metabolic alkalosis (754). The study identifies AE4 as an essential part of the renal sensing mechanism for changes in acid-base status.

5) BTR1 (SLC4A11)

SLC4A11: BTR1 (SLC4A11) is an NH₃-stimulated H⁺ (OH⁻) transporter (755–757). Unlike other members of the SLC4 family, BTR1 does not transport bicarbonate. BTR1 (SLC4A11) is widely expressed in human tissues with particularly high expression in kidney, salivary glands, testis, thyroid glands, and trachea (757) as well as in the corneal endothelium (718).

Dysfunction of BTR1 leads to congenital hereditary endothelial dystrophy (758, 759) and Fuchs endothelial corneal dystrophy (760).

Phosphatidylinositol 4,5-bisphosphate (PIP₂) is well known for its role as an intracellular regulator of the activity of transport proteins, including the SLC4 family members NBCe1 (SLC4A4) (761, 762) and NBCn1 (SLC4A7) (763). The cryo-EM structure of human BTR1 (SLC4A11) has been successfully resolved, showing the transporter in the outward-facing state in complex with its activating ligand PIP₂ and in the inward-facing state with the pathogenic R125H mutation (764). PIP₂ binds to the interface between the transmembrane domain and the N-terminal cytoplasmic domain of BTR1. The results provide new insights into the mechanisms by which the transport activity and conformational changes of BTR1 are regulated by PIP₂ binding.

Orphan transporter family members: N/A

SLC5 Sodium glucose cotransporter family (2.A.21/SSF/APC)

Discovery: The first member of the SLC5 family was discovered by expression cloning with Xenopus oocytes of the rabbit intestinal Na⁺/glucose cotransporter SGLT1 (SLC5A1) (121). This advance also marked the first identification and characterization of a Na⁺-coupled transporter at the molecular level, which is important because the thermodynamic coupling between the downhill transport of an ion such as Na⁺ and the uphill transport of a substrate such as glucose is a core feature of secondary transporters (765). Distinct mechanisms of Na⁺ binding in SGLT1 and H⁺ binding in LacY permease (TC 2.A.1.5.1) have been discussed in light of their different structural folds, namely SGLT1 as a member of the APC transporter superfamily and LacY permease as a member of the MFS transporter superfamily (766) (see “Structure-based classification of SLCs”).

Gene family members (12):
SLC5A1 (SGLT1)	SLC5A4 (SGLT3)	SLC5A7 (CHT)	SLC5A10 (SGLT5)
SLC5A2 (SGLT2)	SLC5A5 (NIS)	SLC5A8 (SMCT1)	SLC5A11 (SMIT2)
SLC5A3 (SMIT1)	SLC5A6 (SMVT)	SLC5A9 (SGLT4)	SLC5A12 (SMCT2)

Molecular aspects, physiological roles and links to disease

The SLC5 family consists of 12 members, of which 5 are hexose transporters (SLC5A1, SLC5A2, SLC5A4, SLC5A9 and SLC5A10) and the others are myo-inositol (SLC5A3 and SLC5A11), iodide (SLC5A5), monocarboxylate (SLC5A8 and SLC5A12), choline (SLC5A7) and vitamin (biotin and pantothenic acid; SLC5A6) transporters, as reviewed in (214). All these membrane proteins share the common Na⁺-coupling transport mechanism.

The SLC5 family belongs to the Solute/Sodium Symporter (SSS) family (TC 2.A.21), which in turn is part of the APC superfamily (214, 767, 768). SLC5 family members incorporate the alternating access mode of membrane transport as they contain the LeuT structural fold (212), a paradigm structure found in membrane transporters of the APC superfamily (46) (see Section 8). SSS family proteins typically contain 10-14 transmembrane helices (TMHs). Members of the SSS family have been widely identified in prokaryotes and eukaryotes, and the functionally characterized members have been shown to catalyze solute uptake by Na⁺ cotransport, as exemplified by the E. coli Na⁺-proline symporter PutP (46, 769), the of E. coli Na⁺-pantothenate symporter (PanF) (770), and the rabbit intestinal Na⁺-glucose cotransporter (46, 121, 214). The tertiary structures of two representative members, the sodium/galactose symporter from Vibrio parahaemolyticus (vSGLT) (771) and the sialic acid transporter SiaT from Proteus mirabilis (772) confirmed that they share the APC structural fold with a transporter core formed by 5+5 TMHs in an inverted repeat arrangement.

Based on the phylogenetic tree of the family, there are distinct branches for the human SLC5 proteins in accordance with their substrate selectivity, with the sugar and myo-inositol transporters representing one branch, the iodide, monocarboxylate and vitamin transporters another branch, and the third branch formed by the choline transporter, which is Na⁺ and Cl⁻ coupled. The following describes the members of the SLC5 family, which are divided into the following four groups according to substrate type and phylogenetic relationship (see Fig. 13):

Fig. 13. Phylogenetic tree of the SLC5 family.

1. Hexose transporters: SGLT1 (SLC5A1), SGLT2 (SLC5A2), SGLT3 (SLC5A4), SGLT4 (SLC5A9), SGLT5 (SLC5A10)

2. myo-Inositol transporters: SMIT1 (SLC5A3), SMIT2 (SLC5A11)

3. Iodide, multivitamin and monocarboxylate transporters: NIS (SLC5A5), SMVT (SLC5A6), SMCT1 (SLC5A8), SMCT2 (SLC5A12)

4. Choline transporter: CHT (SLC5A7)

1) Hexose transporters: SGLT1 (SLC5A1), SGLT2 (SLC5A2), SGLT3 (SLC5A4), SGLT4 (SLC5A9), SGLT5 (SLC5A10)

SLC5A1: SGLT1 (SLC5A1) is the major transporter responsible for the uptake of glucose and galactose across the intestinal brush border membrane (121, 122, 214, 773). It is coupled to the cotransport of two Na⁺ ions (Fig. 14). Michaelis-Menten (K_m) values for α-methyl-D-glucopyranoside (αMeGlc), a non-metabolized analog of glucose, were found to be 110 μM for rabbit, 400 μM for rat, and 800 μM for human SGLT1 (121, 122, 774, 775). Since SGLT1 (SLC5A1) is the only mechanism for absorption of these sugars in the human small intestine, its genetic defects lead to glucose-galactose malabsorption, a rare hereditary disease with massive and life-threatening diarrhea due to osmotic imbalance (776–778). At somewhat lower levels, SGLT1 is also expressed in the proximal convoluted tubules of the kidney, where it reabsorbs filtered glucose that has not been absorbed in the early proximal tubules by SGLT2 (SLC5A2) (see below). SGLT1 (SLC5A1) is also found in the heart muscle, epididymis, and gallbladder, according to the HPA.

Fig. 14. Na⁺-glucose cotransporters in intestine (a) and kidney (b).

SGLT1 (SLC5A1) is expressed in the brush border membrane of enterocytes as well as in the apical membranes of kidney proximal tubule S3 segments. SGLT2 (SLC5A2) is specifically expressed in the apical membranes of kidney proximal tubule S1 segments. GLUT2 (SLC2A2) is expressed in the basolateral membranes of intestinal and renal proximal tubule cells. GLUT5 (SLC2A5) facilitates fructose uptake in the intestine in the apical and basolateral membranes. Basolateral GLUT2 in enterocytes aids in the transport of fructose as well. SGLT2 inhibitors (gliflozins) lower blood glucose levels and have shown great efficacy in the treatment of T2D. Gliflozins improve glycemic control and reduce body weight and systolic and diastolic blood pressure. Inhibition of Na⁺-coupled glucose reabsorption via SGLT2 increases Na⁺ exposure at the macula densa. As a result, macula densa cells reduce renin release to suppress the renin-angiotensin-aldosterone system (RAAS) and vasoconstrict afferent arterioles to reduce hyperfiltration commonly associated with diabetes. Tight junctions shown in (a) seal the paracellular pathway but can also form selective semipermeable paracellular pathways for small cations, anions and water (12).

Another important function of SGLT1 is to facilitate intestinal water absorption during digestion. Due to the presence of an unstirred layer near the surface of the brush border membrane, high concentrations of monosaccharides, amino acids and oligopeptides are likely to accumulate at the villous surface after a meal as a result of carbohydrate and protein digestion. There they are available for absorption by Na⁺-coupled glucose transport [i.e., via SGLT1 (SLC5A1)), Na⁺-coupled amino acid transport [i.e., via B⁰AT1 (SLC6A19); see the SLC6 family description] and H⁺-coupled oligopeptide transport [i.e., via PepT1 (SLC15A1); see the SLC15 family description and Fig. 17)]. The resulting ion-coupled solute transport creates an osmotic gradient that drives fluid absorption via the paracellular route through epithelial tight junctions (779–781) (Fig. 17). Cl^- also follows the paracellular route and enters the interstitium driven by the electrical gradient. In addition, it has been shown that SGLT1 itself transports water across the brush border membrane in the intestine (782). Thus, ion-coupled solute transport via SGLT1 and other ion-coupled transporters drives fluid absorption through both transcellular and paracellular routes. Furthermore, crypt cells cooperate during digestion to recycle Na⁺ and water from the blood back into the intestinal lumen (783). Cl^- first enters crypt cells at the basolateral membrane via the Na⁺/K⁺/2Cl^- cotransporter NKCC1 (SLC12A2) and leaves the cell apically via the CFTR chloride channel. Na⁺ ions then move paracellularly into the intestinal lumen, driven by the negative electrical potential of the lumen. As a result, Na⁺ and Cl^- enter the intestinal lumen, allowing water to follow osmotically to maintain the fluidity of the chyme.

Disturbances in fluid cycling in the intestine have pathological consequences. Patients with glucose-galactose malabsorption suffer from life-threatening diarrhea unless glucose and galactose are removed from the diet (776). Unabsorbed luminal glucose reverses osmotic flow and causes diarrhea. In cholera, the diarrheal bacterium Vibrio cholerae promotes cAMP-mediated fluid loss. Cholera toxin binds to GM1 gangliosides on the cell surface of enterocytes (784), followed by endocytosis and induction of cAMP production, leading to hypersecretion of Cl^- from crypt cells into the intestinal lumen, resulting in diarrhea.

Specifically, cAMP leads to increased apical membrane insertion of the CFTR apical Cl^- channel in crypt cells, resulting in loss of Cl^- ions and HCO₃^- through CFTR into the lumen. In addition, cAMP causes Na⁺ loss by inhibiting the apical Na⁺/H⁺ exchanger NHE3 (SLC9A3) in villus cells (785–787). This leads to dehydration with high water loss and electrolyte depletion due to loss of NaCl and withdrawal of H₂O from the epithelium, which passively follows the NaCl into the intestinal lumen. In oral rehydration therapy, luminal glucose is provided to stimulate Na⁺ and water absorption in villus cells. In cholera patients, oral rehydration therapy prevents death caused by severe diarrhea (786, 788). Improved oral rehydration therapy includes not only glucose but also amino acids to accelerate Na⁺-coupled transport and the rehydration process.

In the human heart, SGLT1 expression has been reported to be elevated in hypertrophic, ischemic, and diabetic cardiomyopathy (789). Moreover, selective inhibition of SGLT1 was shown to have a protective effect against myocardial infarction-induced ischemic cardiomyopathy in a preclinical model (790). Recent SGLT1 knockdown have shown a decrease in cardiac fibroblast activation in diabetic cardiac fibrosis (791) and that SGLT1 is involved in cardiac fibrosis via the p38 and ERK1/2 signaling pathways. The findings highlight that SGLT1 is a potential therapeutic target for the prevention of diabetic cardiac fibrosis (791).

SLC5A2: Following the discovery of SGLT1, the high-affinity, low-capacity Na⁺-glucose cotransporter (SLC5A1), its paralog SGLT2 (SLC5A2) was identified as a low-affinity, high-capacity Na⁺-coupled glucose transporter (K_m for αMeGlc: 1.6 mM) (122, 792). SGLT2 is predominantly expressed in the kidney and is responsible for the bulk reabsorption of filtered glucose in the early segments (S1) of the renal proximal tubule. In contrast, any remaining glucose in the downstream segments (late S2 and S3) is reabsorbed by SGLT1 (122, 792) (Fig. 14). Thereby SGLT2 greatly contributes toward maintaining normal blood glucose levels required as a source of energy and to warrant normal neurologic function (792).

Loss-of-function genetic mutations in the human gene encoding SGLT2 (SLC5A2) or MAP17 (PDZK1IP1), the necessary activator of SGLT2 (214, 793, 794), lead to familial renal glycosuria (795–798). Patients with this disease have a decreased renal tubular absorption of glucose from the urine in the absence of hyperglycemia and in the absence of any other signs of tubular dysfunction.

The discovery of SGLT2 as the major renal reabsorption mechanism of glucose (792) led to the development of novel drugs such as canagliflozin, empagliflozin and dapagliflozin that inhibit SGLT2 to effectively lower blood glucose levels in patients with T2D.

Phlorizin was first isolated from apple tree bark in 1835 (reviewed in (799)). It was later found to be a potent but rather non-selective inhibitor of both SGLT1 and SGLT2, but at that time phlorizin did not seem to have any obvious medicinal value. In 1886, its hypoglycemic and renal glucosuric effects were described (799). However, glucosuria was considered an indication of a form of diabetes, and although many groups investigated the effects of phlorizin, no medicinal use was attributed to it (800). In the late 1950s, phlorizin was shown to inhibit intestinal Na⁺-dependent glucose uptake (801). In the mid-1980s further studies showed that phlorizin could reduce the hyperglycemia of partially pancreatectomized diabetic rats (802). This led to a reconsideration of the potential medical value of phlorizin, but its low solubility and low potency hampered development. The cloning of SGLT1 (121) and SGLT2 (792) opened the door to screening for new and improved phlorizin derivatives.

Since phlorizin is a non-selective inhibitor with poor oral bioavailability, phlorizin derivatives were synthesized, the first being Tanabe’s T-1095 (803), which is a prodrug that is absorbed in the intestine and rapidly converted in the liver to the active metabolite T-1095A. However, T-1095 did not enter clinical development. Subsequently, other O-glycoside derivatives of phlorizin were developed, i.e., Kissei’s remogliflozin (804) and sergliflozin (805). Inhibition of renal SGLT1 and SGLT2 with this compound increased urinary glucose excretion in diabetic animals. Remogliflozin is currently used to treat non-alcoholic steatohepatitis (NASH) and T2D. Sergliflozin is a prodrug of sergliflozin-A. It is an investigational anti-diabetic compound developed by GlaxoSmithKline that was not further developed after Phase II. In addition, fused aromatic O-glycosides have been developed (806, 807).

The next important structures to be developed are aromatic and heteroaromatic C-glycosides in which the glucose moiety is directly linked to the aglycone by a carbon-carbon bond. This was based on a method developed by Abbott Laboratories for the preparation of C-glycoside derivatives of phlorizin (808). Dapagliflozin is a representative compound of aromatic C-glycosides to increase the chemical stability of the glycosidic bond, developed at Bristol Myers Squibb in 2000 (809). These C-glycosides are more metabolically stable than O-glycosides due to their resistance to gastrointestinal β-glucosidases and are rapidly absorbed in the gastrointestinal tract without modification of the prodrug form. Further molecular modification produced the selective SGLT2 inhibitor dapagliflozin (approved in Europe since 2012) and other C-glycoside derivatives with varying degrees of selectivity for SGLT2 inhibition (e.g., canagliflozin, approved in the USA in 2013, followed by empagliflozin and ertugliflozin) (810).

Beneficial cardio-renal effects of these agents, particularly in reducing blood pressure and the risk and progression of heart failure and chronic kidney disease independent of glycemic control, have also been reported and have generated indications beyond the management of diabetes (811, 812). Among these, the SGLT2 inhibitor canagliflozin has been shown to improve hemodynamics and have beneficial effects on cardiac function (813). To specifically address the cardiorenal protective effects of SGLT2 inhibitors, studies were conducted using a highly differentiated opossum kidney culture model that recapitulates key morphologic and functional features of the renal proximal tubule. The studies showed that canagliflozin, but not empagliflozin, reduced the function of the Na⁺/H⁺ exchanger NHE3 (SLC9A3)-dependent fluid transport and endocytosis, independent of SGLT2 inhibitors. It has been reported that this is due to inhibition of NHE3 and mitochondrial complex I. The investigators concluded that canagliflozin suppresses Na⁺-dependent fluid transport, possibly due to reduced ATP generation in canagliflozin-treated cells, as well as albumin uptake in proximal tubule cells (814). Therefore, consistent with previous observations, it has been hypothesized that SGLT2 inhibitors have protective effects on cardiorenal functions beyond glycemic control and that canagliflozin contributes to renal and extrarenal protection in diabetic and even non-diabetic patients. However, the molecular mechanisms underlying these protective effects remain to be elucidated. It also remains to be determined whether the same is true for the human kidney.

Prior to the medical application of SGLT2 inhibitors, metformin, originally discovered in 1918, was the preferred first-line oral glucose-lowering agent for the treatment of T2D. The ability of metformin to counteract insulin resistance and treat adult-onset hyperglycemia gained credibility and it was introduced in the US in 1995. Long-term cardiovascular benefits of metformin were identified by the United Kingdom Prospective Diabetes Study in 1998, providing a new rationale for the adoption of metformin as initial therapy (815). SGLT2 inhibitors can be readily combined with other diabetes medications such as metformin to achieve optimal HbA1c levels, weight loss, and blood pressure control (816–818).

The discovery of SGLT2 inhibitors has led to the development of clinical game changers in the treatment of T2D. In another major breakthrough, the high-resolution structure of the 14-TMH SGLT2 Na⁺-glucose cotransporter bound to the single TMH activator MAP17 and to the anti-diabetic drug empagliflozin has been solved (93). The structure reveals the interaction between MAP17 and transmembrane helix 13 of SGLT2, with empagliflozin occupying the sugar-binding site and the outer vestibule to inhibit SGLT2 by occluding the transport cycle. Subsequently, a detailed analysis of the SGLT2-MAP17 structure and the transport and inhibition mechanisms was reported (819). These studies shed new light on the rational design of modulators and drugs for SLC5 solute carriers.

Despite the narrow tissue distribution of SGLT2 in the early proximal tubules of the kidney, SGLT2 has been shown to be a contributor to high glucose demand in pancreatic cancer, prostate cancer and astrocytoma. In addition, the function of SGLT2 in these cancers has been demonstrated by imaging with an the SGLT-specific positron emission tomography (PET) imaging probe, α-methyl-4-deoxy-4-¹⁸F-fluoro-D-glucopyaranoside (Me-4FDG). Me-4FDG is a very sensitive probe for visualization of high-grade astrocytoma by PET (820, 821). PET imaging of SGLTs has been shown to be effective in assessing the pharmacodynamics of SGLT inhibitors, investigating metabolism in diabetic patients, and staging various cancers (821). It also remains to be determined whether SLGT2 inhibitors such as empagliflozin are beneficial in the treatment of these cancers, for example whether they reduce astrocytoma growth in patients (822).

SLC5A4: SGLT3 (SLC5A4) does not transport glucose, but its interaction with the sugar depolarizes the plasma membrane in a saturable, Na⁺-dependent, phlorizin-sensitive manner. This makes it a potential glucose sensor (transceptor) (823, 824). It is highly expressed in the duodenum and at lower levels in other tissues such as muscle. Specifically, it has been reported to be expressed in cholinergic neurons in the enteric nervous system and at neuromuscular junctions and it has been proposed to regulate intestinal motility in response to glucose (823, 825). A follow-up study demonstrated that human SGLT3 functions as a sugar sensor in vivo and strikingly, a single amino acid mutation on TMH4 (E457Q) converts this sugar sensor into a sugar transporter with properties similar to its close paralog, SGLT1 (SLC5A1). It is worth noting that residue 457 is required for glucose translocation in SGLT1 (826, 827) and when mutated causes glucose-galactose malabsorption (776). A subsequent study suggests that SGLT3 expression in the small intestine of mice and humans is localized almost exclusively to the intestinal epithelium and probably not to cholinergic neurons (828). In this study, it was proposed that activation of the SGLT3 by luminal glucose reduces the driving force for SGLT1 (SLC5A1)-mediated glucose uptake due to the glucose-induced Na⁺ uptake by SGLT3 (SLC5A4) in more proximal parts of the small intestine, resulting in greater distal delivery of glucose to act on endocrine cells, thereby increasing secretion of the incretin GLP-1 (glucagon-like peptide-1), which acts to increase insulin secretion from pancreatic beta cells. Downregulation of SGLT3 (SLC5A4) in obesity would then lead to increased SGLT1 (SLC5A1)-mediated glucose uptake in the proximal part of the small intestine, resulting in reduced distal release of GLP-1 and decreased pancreatic insulin secretion. Further studies are still required to validate this concept, as well as the precise cellular localization of SGLT1 (SLC5A1) in the intestine and its role in obesity.

SLC5A9: SGLT4 (SLC5A9) is a Na⁺-dependent mannose transporter that is highly expressed in the duodenum and only at low levels in other organs such as liver, kidney, pancreas and lung (829). The substrate preference of this transporter was subsequently reevaluated and it was concluded that under physiological conditions it mainly transports mannose but not 1,5-anhydroglucitol (see below, SLC5A10) (830), and according to a previous study, it also transports fructose (829). Since SGLT4 (SLC5A9) is highly expressed in the intestine and only at lower levels in the renal proximal tubules under normal conditions, its main physiological function is more likely the intestinal absorption of certain sugars such as mannose and possibly fructose. To date, no SGLT4 deficiency has been reported and knockout mouse models are lacking, hindering progress in understanding the exact physiological roles of this transporter.

Of note, a recent genome-wide characterization of 54 urinary metabolites revealed that SGLT4 (SLC5A9) function is also associated with xylose transport (831). Given the prominent expression of SGLT4 in the small intestine, it is tempting to suggest that this transporter is an essential component of the D-xylose absorption test used to assess the digestive and absorptive capacity of the small intestine, especially in cases of suspected malabsorption problems (832). In this test, a patient typically drinks a xylose-containing solution, and then blood and urine samples are collected at specified intervals to measure the xylose levels in these samples. If the xylose levels in the blood and urine are low, this indicates a potential problem with the absorption process.

SLC5A10: SGLT5 (SLC5A10) is highly and almost exclusively expressed in the apical membrane of the proximal renal convoluted tubules (833, 834). It mediates the Na⁺-coupled renal reabsorption of mannose and fructose, two of the major dietary sugars, along with glucose and galactose, which are absorbed in the intestine and subsequently filtered and reabsorbed in the kidney. Consistent with this, urinary fructose excretion was shown to be a distinctive feature of an SGLT5 knockout mouse model (835). Thus, SGLT5 prevents the appearance of fructose in the urine, which would otherwise stimulate bacterial growth in the urinary tract and increase the risk of urinary tract infections. It is likely that SGLT5 (SLC5A10) also limits the urinary loss of mannose, which plays an important role in protein glycosylation.

It also mediates renal reabsorption of 1,5-anhydroglucitol, an abundant polyol in the blood. Frequent heterozygous mutations in SLC5A10 lower blood levels of 1,5-anhydroglucitol (836), confirming that SGLT5 (SLC5A10) reabsorbs monosaccharides. In healthy individuals, 1,5-anhydroglucitol levels are kept relatively constant by intestinal absorption and renal reabsorption. Interestingly, during hyperglycemia, glucose that could not be completely reabsorbed by SGLT1 and SGLT2 likely leads to inhibition of 1,5-anhydroglucitol reabsorption via SGLT5, and thus to the observed decrease in blood 1,5-anhydroglucitol levels. The 1,5-anhydroglucitol test is currently FDA-approved for use in diabetes patients to measure blood levels of 1,5-anhydroglucitol to determine the history of hyperglycemic episodes (837), in addition to HbA1c and fructosamine testing.

1,5-anhydroglucitol is also a major player in rare forms of neutropenia type 4, a disorder of the hematopoietic system associated with mutations in the glucose-6-phosphate transporter SLC37A4 (G6PT) or phosphatase (G6PC3) genes, resulting in accumulation of 1,5-anhydroglucitol-6-phosphate in neutrophils. Mutations in SLC5A10 lower blood levels of 1,5-anhydroglucitol and thus favor neutropenia (830).

2) myo-Inositol transporters: SMIT1 (SLC5A3) and SMIT2 (SLC5A11)

SLC5A3: SMIT1 (SLC5A3), a renal Na⁺-coupled myo-inositol transporter, belongs to a separate subbranch of the SLC5 family, and expression cloning was therefore used to identify the founding member of this subbranch, the canine SMIT (SLC5A3) (145). According to the HPA, SMIT1 (SLC5A3) is also relatively highly expressed in other tissues, especially in the thyroid, choroid plexus and retina. In the epithelial cells of the choroid plexus, it may be located in the basolateral membrane (838) (see Fig. 11). SMIT1 (SLC5A3) regulates the intracellular concentration of the osmolyte myo-inositol, allowing cells to survive in a hypertonic environment, and its expression is upregulated by extracellular hypertonicity via transcriptional mechanisms (839). Gene knockout in mice (Slc5a3^−/−) results in severe myo-inositol deficiency issues during embryogenesis and early fetal development (840). The phenotype of these mice includes abnormal respiratory rhythmogenesis leading to death shortly after birth and a dramatic delay in osteoblastic differentiation and bone formation. However, how exactly the Slc5a3 knockout leads to these phenotypes has not yet been fully elucidated (841).

In addition to its role as an osmolyte transporter, SMIT1 (SLC5A3) can also deliver myo-inositol into cells as a substrate for phosphorylation to generate the phospholipid PIP₂, a key modulator of many ion channels. The SMIT1 protein was subsequently shown to be expressed in vascular smooth muscle cells (VSMCs) where it modulates arterial contractility through an association with the potassium channel Kv7.4 (KCNQ4)-Kv7.5 (KCNQ5) heteromers (842).

SMIT1 (SLC5A3) also plays a key role in cancer cell survival and proliferation by shaping cellular osmoregulation and regulation of metabolic demand (843). Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive cancer with a poor prognosis, largely due to the rapid development of chemoresistance in patients (844). The chemotherapeutic drug gemcitabine, which is often used as the primary chemotherapeutic agent for PDAC, must be transported into cells via nucleoside transporters, most commonly via SLC29A1, SLC28A1, and SLC28A3 (845). Gemcitabine disrupts DNA synthesis and induces apoptosis of cancer cells (846). Mitochondria are crucial for cell energy production and apoptosis regulation and thus play an important role in the induction of drug resistance in cancer. SMIT1 (SLC5A3) was discovered to be a key modulator promoting chemoresistance in PDAC and its expression levels were significantly upregulated in gemcitabine-resistant PDAC cells (847), thereby enhancing their cell survival by stabilizing the mitochondrial functions and inhibiting apoptosis. Moreover, SMIT (SLC5A3) inhibition was shown to enhance the efficacy of gemcitabine and thus significantly reduce tumor growth (847). Specifically, SMIT (SLC5A3) inhibition disrupted mitochondrial dynamics, leading to increased reactive oxygen species production, mitochondrial fission, and impaired oxidative phosphorylation. SLC5A3 inhibition was also shown to activate the PINK1 [phosphatase and tensin homologue (PTEN)-induced kinase 1]/Parkin-mediated mitophagy pathway. Parkin is an E3 ubiquitin ligase that tags damaged mitochondria with ubiquitin, marking them for degradation by autophagy. While PINK1/Parkin-mediated mitophagy is a critical process for mitochondrial quality control by acting as a cellular mechanism to remove damaged or dysfunctional mitochondria, excessive removal of damaged and healthy mitochondria depletes mitochondrial reserves and sensitizes cells to apoptosis (847). These results suggest that targeting SMIT (SLC5A3)-mediated mitochondrial regulation is a promising therapeutic strategy to overcome gemcitabine resistance in PDAC.

SLC5A11: SMIT2 (SLC5A11) is a Na⁺-coupled inositol transporter (848) expressed in brain, intestine and kidney (849, 850). In the brain, strong expression was found in the hypothalamus and substantia nigra, and immunohistochemistry showed cytoplasmic staining in neurons (849). In the small intestine, SMIT2 (SLC5A11) was found intracytoplasmic in villus epithelial cells and myenteric ganglion cells (849). In rabbit kidney, SMIT2 was localized in the cortex where it was proposed to serve as an apical myo-inositol transporter (850). SLC5A11 variants were found to be associated with inverse salt sensitivity of blood pressure (851). There is also evidence that SLC5A11 in humans interacts with immune-related genes and may function as an autoimmune modifier (852). Overall, despite these interesting observations, more work is needed to clarify the cellular and subcellular localizations and physiological roles of SMIT2 (SLC5A11).

3) Iodide, multivitamin and monocarboxylate transporters: NIS (SLC5A5), SMVT (SLC5A6), SMCT1 (SLC5A8), SMCT2 (SLC5A12)

SLC5A5: NIS (SLC5A5) is an iodide transporter required for thyroid hormone biosynthesis. Iodide anion (I⁻) absorbed in the gastrointestinal tract is actively transported by NIS (SLC5A5) from the blood across the basolateral membrane of thyroid follicular cells of the thyroid gland. NIS (SLC5A5) transports I⁻ electrogenically, with a stoichiometry of 2 Na⁺ and 1 I⁻ (853). It transports a wide variety of other monovalent anions such as BF₄⁻, TcO₄⁻, and SCN⁻ (854, 855). Since SLC5A5 is part of a separate subbranch of the SLC5 family, expression cloning was used to identify the first member of this subbranch, the rat thyroid Na⁺/I⁻ cotransporter NIS (Slc5a5) (152). Slc5a5 knockout mice demonstrated that NIS is the only protein that actively accumulates I⁻ in the thyroid (856).

NIS is highly expressed in thyroid gland, stomach and salivary gland (857). Its expression is upregulated at the transcriptional level by TSH (858) and downregulated at the post-transcriptional level by I^- itself (859). NIS expression in the intestinal brush border membranes has been demonstrated for rat and mouse enterocytes (860, 861) as well as human duodenal enterocytes (862) and it has been proposed that NIS contributes to Na⁺-coupled intestinal absorption of iodide. However, several other transporters are also known to be involved in iodide transport (863). These include the anion exchangers pendrin (SLC26A4) and SLC26A7 (see the SLC26 family description), the chloride channel anoctamin 1/TMEM16A (ANO1) (864, 865) and the sodium-dependent multivitamin transporter SMVT (SLC5A6) (866). Anoctamin 1, pendrin and SLC26A7 have been implicated as mediators of apical iodide efflux. To what extent SMVT (SLC5A6) also contributes to intestinal iodide absorption remains to be determined.

In the kidney, most of the absorbed iodide is excreted by glomerular filtration without reabsorption by the renal tubules (867). Regarding the role of NIS (SLC5A5) expression in the stomach and salivary gland, the current concept is that while some of the absorbed iodide undergoes organification in the thyroid and much of the iodide is excreted by the kidney, a fraction of the iodide is also secreted via NIS expressed in the basolateral membrane of the salivary glands and stomach, and then enters the gastrointestinal lumen from where it is reabsorbed via NIS or other transporters such as SMVT in the small intestine (862, 868). This results in an entero-thyroidal circulation of iodide with NIS-mediated secretion in the stomach and salivary glands.

Interestingly, in light of the above, NIS has been described at the subcellular level in polarized cells in basolateral or apical localizations. For example, in the thyroid it is located at the basolateral membrane of follicular cells, whereas in the intestine it has been shown to be apically localized. This suggests that NIS membrane targeting in polarized cells is cell type specific.

According to the HPA, NIS (SLC5A5) is also highly expressed in the choroid plexus. In fact, iodine transport systems have been previously identified in the epithelium of the choroid plexus (869, 870). Whether NIS (SLC5A5) is expressed at the basolateral (blood-side) membrane of the choroid plexus epithelium, allowing accumulation of iodide in the cerebrospinal fluid (869), or at the apical (CSF-side) membrane, allowing clearance of excess iodide from the central nervous system (870), remains to be determined. Fig. 11 tentatively shows NIS at the apical membrane.

Thyroid hormone synthesis is known to be affected by genetic defects or environmental factors leading to congenital hypothyroidism, the most common endocrine disorder in newborns and one of the most common preventable causes of impaired intellectual development (871, 872). Iodide transport defect is a rare autosomal recessive disorder caused by the inability of thyroid follicular cells to accumulate iodide, resulting in congenital dyshormonogenic hypothyroidism (872, 873). An iodide transport defect is suspected when radioiodine accumulation in a eutopic thyroid gland as well as in the salivary glands is reduced or absent. Biallelic loss-of-function variants in the SLC5A5 gene encoding NIS lead to defective iodide accumulation and thus to congenital dyshormonogenic hypothyroidism (857, 874). More than forty pathogenic SLC5A5 gene variants have been identified in patients with congenital dyshormonogenic hypothyroidism. In-depth molecular characterization of NIS variants has provided information on the transport mechanism and revealed specific amino acid residues critical for substrate binding, substrate specificity and stoichiometry, as well as membrane protein folding and targeting to the plasma membrane (857, 873–875).

Radioiodine therapy with NIS (SLC5A5) as the underlying molecular vehicle is routinely used to treat hyperthyroidism and certain types of thyroid cancer. Recent breakthroughs in NIS research are opening the door to new applications using NIS as a powerful “theranostic tool” for diagnostic imaging and therapeutic radionuclide delivery. The transport of specific radiotracers allows efficient non-invasive monitoring of the biodistribution of functional NIS expression by whole-body scintigraphy, single-photon emission computed tomography or PET. Application of therapeutically active radionuclides delivered via NIS induces cytoreductive effects. This strategy enables cytoreductive gene therapies based on targeted NIS expression in thyroid and nonthyroid cancer cells (857, 876, 877).

SLC5A6: SMVT (SLC5A6) is a widely distributed multivitamin Na⁺ cotransporter (878). It mediates Na⁺-dependent uptake of pantothenic acid, biotin, the vitamin-like substance α-lipoic acid and iodide and plays an important role in their absorption across the digestive tract (see Fig. 30). In addition, it is predicted to be able to transport these B-group vitamins across the choroid plexus into the cerebrospinal fluid (Fig. 11). The protein shares high sequence identity and similarity with NIS (SLC5A5), so it is perhaps not surprising that it also behaves as a Na⁺/iodide cotransporter (866). However, the precise role of SMVT in the transport and homeostasis of iodide is still unclear (879).

The intestinal Slc5a6 knockout mice exhibited growth failure, decreased bone density and length, lethargy, hunchback posture, and intestinal inflammation (880, 881), but no iodine deficiency or thyroid dysfunction phenotype has been reported. Supplementation with biotin and pantothenic acid was able to rescue the phenotype (882). Furthermore, several patients with biallelic SLC5A6 variants also showed failure to thrive and triple vitamin replacement therapy had beneficial effects in the patients (879, 883).

SLC5A8: SMCT1 (SLC5A8) is an Na⁺-coupled electrogenic transporter of monocarboxylates such as lactate, pyruvate, nicotinate (niacin/vitamin B3), and short-chain fatty acids (SCFA; e.g., acetate, propionate, and butyrate) (884–888). The Na⁺:substrate stoichiometry depends on the monocarboxylate transported (from 4:1 for propionate to 2:1 for lactate) (889). SMCT1 (SLC5A8) has been shown to transport iodide as well by a passive mechanism (884). SMCT1 (SLC5A8) expression in mice is found in the colon and kidneys, and to a lesser extent in the brain and retina. In the mouse colon, SMCT1 (SLC5A8) is postulated to contribute to the apical uptake of short-chain fatty acids such as acetate, propionate, and butyrate generated by bacterial fermentation of dietary fiber (890) (see Fig. 33). This function would coincide with MCT1 (SLC16A1), which mediates basolateral exit of SCFA from the colonic epithelium (891) (see the SLC16 family description).

It should be noted, however, that the expression of SLC5A8 in human tissues may differ from rodents, as suggested by the HPA, which, although not peer-reviewed, suggests negligible expression in colon but high expression in cervix, thyroid, and somewhat lower levels in kidney and adrenal. Further studies will be needed to clarify the exact expression of SLC5A8 and also its close homolog SLC5A12 (see below) in the human gastrointestinal tract.

Regarding its expression in the cervix, SMCT1 (SLC5A8) was successfully shown to act as a suppressor in the progression of cervical cancer by regulating the Wnt signaling pathway, a finding that suggests a putative strategy for the treatment of cervical cancer (892). SMCT1 (SLC5A8) has also been shown to function as a tumor suppressor gene in other cancers, including colorectal cancer (889). One of the short chain fatty acids that serves as a substrate for SMCT1 (SLC5A8) is butyrate, which is known to induce apoptosis in a variety of tumors, illustrating that SMCT1 (SLC5A8) can act as a tumor suppressor.

Since SMCT1 (SLC5A8) has been established as an important tumor suppressor, the molecular and cellular effects of SLC5A8 missense variants on its tumor suppressive function were investigated using various in vitro assays (893). The study confirms that decreased SMCT1 (SLC5A8) expression caused by SLC5A8 missense variants significantly impairs the tumor suppressive function of the transporter. Further research is needed to determine whether SLC5A8 variants influence colorectal cancer susceptibility.

Regarding expression in the thyroid, as noted above, SMCT1 (SLC5A8) probably transports iodide by a passive mechanism and it has been proposed that this transporter serves as a putative human iodide transporter located at the apical membrane of thyrocytes where it has been localized by immunohistochemistry at the apical pole of thyroid cells facing the colloid lumen (884). The results suggest that SMCT1 (SLC5A8) mediates iodide transport from the thyroid cell through the apical membrane into the colloid lumen in parallel with pendrin (SLC26A4) and SLC26A7 (see the SLC26 family description).

In the kidney, based on studies in mice, Slc5a8 functions together with Slc5a12 (see below) in the apical membrane of renal proximal tubule cells to reabsorb lactate, and the studies suggested that it may also be involved in urate reabsorption (894).

A recent investigation of protein-protein interactions revealed that the adaptor protein PDZK1 is a binding partner of both SMCT1 and SMCT2, and additionally identified a molecular complex of SMCT1-PDZK1 and the urate transporter URAT1 (SLC22A12) (895).

SLC5A12: SMCT2 (SLC5A12) is a low-affinity Na⁺-coupled electroneutral transporter of monocarboxylates (215, 888, 896). There is 57% amino acid sequence identity between low-affinity SMCT2 (SLC5A12) and high-affinity SMCT1 (SLC5A8). It is highly expressed in kidney and at somewhat lower levels in small intestine (duodenum) and epididymis. In mouse kidney, SMCT2 has been shown to be expressed in the apical membrane along the entire length of the proximal tubule (S1/S2/S3 segments), while the expression of SMCT1 is mostly restricted to the S3 segment (896). Thus, the low-affinity transporter SMCT2 initiates lactate absorption in the early parts of the proximal tubule, followed by the involvement of the high-affinity transporter SMCT1 in the later parts of the proximal tubule. The basolateral exit of monocarboxylate is thought to be mediated by MCT1 (SLC16A1) located on the basolateral membrane of early (primarily S1) proximal tubule segments (see the SLC16 family description).

4) Choline transporter: SLC5A7 (CHT)

SLC5A7: CHT (SLC5A7) is a Na⁺-coupled choline cotransporter expressed in tissues containing cholinergic neurons. There, its transport activity is the rate-limiting step for acetylcholine synthesis. CHT (SLC5A7) is essential for choline reuptake via the Na⁺ gradient in the presynaptic membrane of the neuromuscular junction (897).

In contrast to the other members of the SLC5 family, the transport process mediated by CHT is Cl⁻-dependent and regulated by extracellular pH (898). Ablation of CHT in mice (Slc5a7^-/-) is lethal at birth due to deficits in cholinergic synaptic activity (899).

SLC5A7 choline transporter mutations have been reported in patients with severe congenital myasthenic syndromes (900). Congenital myasthenic syndromes are a heterogeneous group of disorders characterized by impaired neuromuscular signal transmission. A total of 35 genes expressed at the neuromuscular junction and harboring pathogenic variants have been reported to cause the syndromes, including SLC18A3 encoding the vesicular acetylcholine transporter VAChT, SLC25A1 encoding the mitochondrial citrate transporter, and SLC5A7 encoding the choline transporter discussed here (897). Clinical manifestations of the disease include muscle weakness, hypotonia, severe fatigue, and paroxysmal apnea (901, 902). Cases of congenital myasthenic syndromes resulting from SLC5A7 gene mutations are rare, with just over 20 cases reported worldwide (902).

In addition to expression in cholinergic neurons, CHT (SLC5A7) is also significantly expressed in the colon. Immunohistochemical analysis of rat colon showed that CHT (SLC5A7) is expressed in the cholinergic system of the colon. Intense immunostaining was observed in both muscular and mucosal layers (903). Furthermore, upregulation of CHT (SLC5A7) was shown to alleviate stress-induced hyperalgesia that occurs during irritable bowel syndrome (903). Another study showed that CHT (SLC5A7) is downregulated in colorectal cancer (CRC) and functions as a tumor suppressor. Specifically, DNA promoter methylation caused inactivation of CHT (SLC5A7) in CRC, and targeted demethylation of CHT (SLC5A7) might be a therapeutic strategy for CRC and other cancers (904).

Orphan transporter family members: N/A

SLC6 Sodium- and chloride-dependent neurotransmitter transporter family (2.A.22/SNF/APC)

Discovery: The molecular characterization of the first member of the SLC6 family was the rat brain GABA transporter GAT1 (SLC6A1), cloned from purified protein (905). The human cocaine- and antidepressant-sensitive noradrenaline transporter NET (SLC6A2) was subsequently identified from the neuroblastoma cell line SK-N-SH using expression cloning (157).

Gene family members (19 + 2 pseudogenes):
SLC6A1 (GAT1)	SLC6A3 (DAT/ DAT1)	SLC6A5 (GlyT2)	SLC6A7 (PROT)
SLC6A2 (NAT1/NET1/NET)	SLC6A4 (SERT)	SLC6A6 (TauT)	SLC6A8 (CRTR/CT1)
SLC6A9 (GlyT1)	SLC6A13 (GAT2)	SLC6A17 (NTT4/XT1)	SLC6A21P (pseudogene)
SLC6A10P (pseudogene)	SLC6A14 (ATB0+)	SLC6A18 (B0AT3/XT2)
SLC6A11 (GAT-B/GAT-3)	SLC6A15 (B0AT2/NTT7-3)	SLC6A19 (B0AT1/HND)
SLC6A12 (BGT1)	SLC6A16 (NTT5)	SLC6A20 (SIT, IMINO, XT3)

Molecular aspects, physiological roles and links to disease

Members of the SLC6 family are Na⁺-coupled secondary active transporters, some of which also cotransport Cl^-, hence the name Na⁺/Cl^-dependent transporters. The SLC6 family is part of the Neurotransmitter Sodium Symporters (NSS) family (TC 2.A.22) and the Pfam domain name is SNF (Sodium:Neurotransmitter symporter Family). The NSS family is a member of the APC superfamily, and SLC6 family members share the LeuT fold, one of the two paradigm structural folds of the APC superfamily. The LeuT fold was first identified in the crystal structure of the bacterial Na⁺-coupled amino acid importer LeuT (212). It consists of a 5+5 transmembrane spanning domain inverted structural repeat and embodies the basic building blocks that facilitate the alternating access mechanisms of APC transporters (46, 906) (see Section 8). Typically, SLC6 proteins have 12 TMHs, 10 of which form the core of the transporter, and the N- and C-termini are located on the intracellular side (907).

Based on the phylogenetic tree of the SLC6 family (78) and the substrates the family members transport, the SLC6 transporters have been divided into four subgroups (Fig. 15):

Fig. 15. Phylogenetic tree of the SLC6 family.

• 1)The neurotransmitter transporter subgroup: GAT1 (SLC6A1), NET (SLC6A2), DAT (SLC6A3) and SERT (SLC6A4)
• 2)The neurotransmitter amino acid transporter subgroup: GlyT2 (SLC6A5), PROT (SLC6A7), GlyT1 (SLC6A9) and ATB^0,+ (SLC6A14)
• 3) The nutrient amino acid transporters subgroup
  • a)Ancillary protein-associated subgroup: B⁰AT3 (SLC6A18), B⁰AT1 (SLC6A19) and SIT1 (SLC6A20) (The amino acid transporters of this subgroup either require ancillary proteins to which they bind, in order to functionally express in the plasma membrane or they associate with such ancillary proteins).
  • b)Ancillary protein-independent subgroup: B⁰AT2 (SLC6A15), NTT5 (SLC6A16), NTT4/XT1 (SLC6A17)
• 4)The osmolyte transporter subgroup: TauT (SLC6A6), CT1 (SLC6A8), GAT3 (SLC6A11), BGT (SLC6A12) and GAT2 (SLC6A13)

In addition, a K⁺-coupled member of the SLC6 family of the tobacco hornworm Manduca sexta is discussed (subgroup 5)

1) The neurotransmitter transporter subgroup:

This subgroup includes the GABA transporter GAT1 (SLC6A1) and the monoamine transporters NET (SLC6A2), DAT (SLC6A3) and SERT (SLC6A4). Members of this subgroup are known targets for inhibitory drugs used to treat depression, epilepsy and movement disorders such as Parkinson disease. For example, a well-established strategy for treating neurological disorders such as epilepsy is to increase GABA levels in the synaptic cleft by inhibiting GABA reuptake transporters. Another example is the serotonin transporter SERT (SLC6A4), which removes synaptic serotonin and is a known target of antidepressants. The monoamine transporters are targeted not only by medications but also by drugs of abuse. For example, amphetamine and cocaine bind DAT (SLC6A3) with nanomolar affinity, increasing extracellular dopamine levels. NET (SLC6A2) and SERT (SLC6A4) are also affected by both drugs.

With respect to the specific localization of monoamine transporters within synaptic terminals, electron microscopy studies have shown that DAT, NET and SERT are localized on the presynaptic plasma membrane, but not within the synaptic active zones. Instead, they are found in the perisynaptic region, i.e., just outside the synaptic cleft (908). This implies that released transmitters diffuse into the surrounding zone from the synaptic cleft, where they are then taken up back into the terminal.

SLC6A1: GAT1 (SLC6A1) is a GABA transporter that cotransports Na⁺, Cl⁻ and GABA at a ratio of 2:1:1 (909). It is the major GABA transporter in the brain and thus plays an important role in the regulation of GABAergic signaling (910). GAT1 is highly expressed in GABAergic neurons in the neocortex, hippocampus, cerebellum, basal ganglia, brainstem, spinal cord, olfactory bulb and retina (78) and at lower levels in liver and testis according to the HPA. The transporter is strongly inhibited by antiepileptic drugs such as tiagabine and nipecotic acid, which probably exert their effects by increasing the extracellular level of GABA (78).

The pathophysiology due to severely impaired GAT1 function can lead to a wide range of neurodevelopmental phenotypes, including autism, epilepsy and neurodevelopment delay, as shown in recent studies of reduced surface expression, ER retention, and degradation of SLC6A1 genetic variants (911–913). Similarly, another study found that pathogenic variants in SLC6A1 were associated with a clinical phenotype of developmental delay, behavioral problems, and seizures (914). Genetic variance within SLC6A1 has also been associated with pathological anxiety, supporting the concept that GAT1 is a promising target for the treatment of anxiety disorders with panic symptoms (915).

The cryo-EM structure of human GAT1 in complex with its clinically used inhibitor tiagabine shows that this compound locks GAT1 in the inward-open conformation and thus suppresses neurotransmitter uptake, consistent with a two-step mechanism of inhibition (916).

Betaine is an osmolyte that exerts protective effects in the central nervous system and shows therapeutic potential in alleviating certain neurological disorders, but the underlying cellular and molecular mechanisms responsible for its neuroprotective effects remain elusive (917, 918). Betaine has been shown to modulate GAT1 at low concentrations, providing a possible mechanism for the beneficial effect of betaine in protecting neurons from excitotoxicity (918). This modulation is mediated by a temporal inhibition of the transporter, with prolonged occupancy of the transporter by betaine preventing the rapid transition of the transporter to the inward conformation. Specifically, betaine has been shown to play a dual role in GAT1: at mM concentrations it acts as a slow substrate, and at μM concentrations it slows the GAT1 transport cycle, thereby inhibiting GABA uptake (918). This makes betaine a promising neuromodulator of inhibitory pathways, improving GABA homeostasis via GAT1 and conferring neuroprotection against excitotoxicity.

SLC6A2: NET (SLC6A2) is a Na⁺- and Cl⁻-dependent noradrenaline transporter (919). It also transports dopamine, and it was shown that each transport cycle involves the cotransport of one dopamine molecule, one Na⁺ ion, and one Cl⁻ ion (920). Synthetic substrates are amphetamine, methamphetamine, and the neurotoxin MPP⁺ (1-methyl-4-phenylpyridinium). NET (SLC6A2) mRNA has been detected in the brainstem, especially in the locus coeruleus, the large noradrenergic nucleus of the brain, and other brain areas, as well as in the adrenal medulla, vas deferens, and placental syncytiotrophoblast (921). The norepinephrine transporter NET plays a critical role in brain norepinephrine homeostasis and is a target for antidepressants and drugs of abuse.

In Slc6a2 knockout mice, the synaptic life span of the norepinephrine is prolonged, and the mice have a lower body weight and a reduced locomotor response to novelty (922).

Variations in the gene encoding the human norepinephrine transporter (NET, SLC6A2) have been shown to be associated with ADHD in a longitudinal study (923).

SLC6A3: DAT (SLC6A3) mediates the cotransport of two Na⁺ ions and one Cl⁻ ion with each dopamine substrate (907, 924). DAT is almost exclusively expressed in the brain, particularly in the midbrain, where it plays a central role in dopamine transmission by mediating the clearance of extracellular dopamine (925). DAT colocalizes with markers for tyrosine hydroxylase and dopamine D2 receptors (926). Tyrosine hydroxylase and DAT regulate dopamine neurotransmission at the biosynthesis and reuptake steps, respectively (927) and D2 autoreceptors provide feedback inhibition of dopamine release (928, 929). By controlling the reuptake of extracellular neurotransmitter molecules into presynaptic neurons, DAT directs the spatial and temporal dynamics of dopamine neurotransmission (930). The dopamine neurotransmitter system is involved in movement, mood, reward, and cognition, and DAT-mediated reuptake of released dopamine is the main mechanism for the termination of dopaminergic neurotransmission. Many disorders such as depression, bipolar disorder, Parkinson disease, and attention deficit hyperactivity disorder are associated with abnormal dopamine levels and DAT is critical to their etiology. Medications used to treat these disorders, along with many addictive drugs, target this transporter, enhancing dopaminergic signaling by suppressing transmitter reuptake. Synthetic substrates are amphetamine, methamphetamine, and MPP⁺.

SLC6A4: SERT (SLC6A4), also known as 5-hydroxytryptamine transporter or 5-HTT, is crucial for regulating synaptic serotonin levels by transporting serotonin back into the presynaptic neuron, a process called reuptake. It is expressed in the midbrain and brainstem and in certain peripheral tissues (931). In peripheral tissues, it is present in intestine, lung and testis as suggested by the HPA. In brain, SERT (SLC6A4) is highly expressed on extrasynaptic axonal membranes, while it is cytoplasmic in cell bodies and dendrites (932). SERT transports 1 Na⁺, 1 CI⁻ together with 1 serotonin molecule with the counter-transport of 1 K⁺ (933). It also transports the neurotoxin 5,7-dihydroxytryptamine. Synthetic substrates are amphetamine, methamphetamine, and MPP⁺.

SERT (SLC6A4) removes synaptic serotonin and is a known target of antidepressants (Fig. 16). The selective serotonin reuptake inhibitor (SSRI) fluoxetine (Prozac), used to treat depression, obsessive-compulsive disorder and bulimia, works by increasing serotonin levels in the synaptic cleft. In an attempt to improve the efficacy of Prozac, docking of over 200 million small molecules against the inwardly open state of SERT, followed by inhibition assays and further structure-based optimization led to the selection of two potent (low nanomolar) inhibitors that exhibited anxiolytic and antidepressant-like activity in mouse assays, with potencies up to 200-fold better than fluoxetine (Prozac) (934). The results of these studies provide the blueprints for the future rational design of neuromodulators.

Fig. 16. Representative serotonin (5-HT) neuron (left) and GABA neuron (right) with synapses.

The neurotransmitters serotonin (5-HT) and GABA are cleared by SERT (SLC6A4) and GABA (SLC6A1) transporters, respectively, as shown in the figure. The antidepressant SSRI fluoxetine and paroxetine are highly specific inhibitors of SERT (SLC6A4).

SERT is involved in many physiological functions, including mood, aggression, appetite, sleep, cognition, and motor activity. Anxiety, depression, suicide, schizophrenia, autism, substance abuse, and gastrointestinal disorders have been linked to changes in SERT activity, binding site density, and gene polymorphisms (932).

SERT (SLC6A4) has also attracted interest in pain studies, in addition to its use as a target for SSRIs (935). In humans, functional nucleotide variations in the SLC6A4 gene are associated with certain pathological pain conditions and differences in response to pharmacological therapy (935). These findings reflect the importance of SERT in the complex physiology and management of pain, as well as the scientific and clinical challenges that must be addressed to optimize SERT (SLC6A4)-related analgesic therapies (935).

Most studies of genetic variants of SLC6A4 have focused on common regulatory variants (e.g. 5-HTTLPR, rs25531) and their association with psychiatric disorders (936–938). However, rare disruptive coding variants in SLC6A4 have also been reported and linked to severe psychiatric phenotypes including depressive disorder, obsessive-compulsive disorder, and substance use disorder (939) (940) (preprint).

2) The neurotransmitter amino acid transporter subgroup:

These include GlyT2 (SLC6A5), PROT (SLC6A7), GlyT1 (SLC6A9) and ATB^0,+ (SLC6A14).

SLC6A5: GlyT2 (SLC6A5) is a Na⁺- and Cl^--coupled glycine transporter whose expression is predominantly restricted to neurons at glycinergic nerve terminals, but it is also found in GABAergic Golgi cells of the cerebellum (941, 942). Expression is highest in the brainstem, cerebellum and spinal cord (943, 944).

GlyT2 plays an important role in regulating glycine levels at glycinergic inhibitory synapses, where glycine activates glycine receptors, causing an influx of chloride ions that hyperpolarizes the postsynaptic cell (942, 945). Glycinergic neurons are major contributors to the regulation of neuronal excitability, primarily in caudal areas of the nervous system. These neurons control the flow of sensory information between the periphery and the CNS and various motor activities such as locomotion, breathing or vocalization (946). Glycinergic neurons also express VIAAT (SLC32A1), a vesicular transporter of glycine and GABA. Thus, after presynaptic release of glycine, GlyT2 transports glycine from the synaptic cleft back into the presynaptic neuron, where it can be recycled into synaptic vesicles to maintain glycinergic neurotransmission.

GlyT2 shares 48% amino acid sequence identity with GlyT1 and exhibits significantly different properties from those of GlyT1 (947, 948). For example, the stoichiometry of the GlyT2-mediated transporter is 3 Na⁺ to 1 Cl^- to 1 glycine, allowing GlyT2 to maintain nanomolar extracellular glycine levels, whereas that of GlyT1 is 2 Na⁺ to 1 Cl^- to 1 glycine, suggesting that glycine could be imported or exported depending on the physiological conditions governing the electrochemical gradients driving the coupling ions (949).

In addition, glycinergic neurons control pain transmission in the dorsal spinal cord, and their function is reduced in chronic pain states and thus moderate inhibition of GlyT2 may potentiate glycinergic inhibition by slowing the removal of glycine from the synaptic cleft to inhibit pain (945). Overall, it has become evident that GlyT2 represents an attractive target for pharmacological intervention against devastating conditions (945, 946). Interestingly, inhibitors have been shown to bind to a site known as the lipid allosteric site, where there is an interplay between these inhibitors and cholesterol binding to GlyT2. In particular, it has been shown that both glycine transport and sensitivity to lipid inhibitors are affected by interactions with cholesterol molecules recruited from the membrane to a specific site on GlyT2 (950).

Mutations in SLC6A5 are responsible for congenital hyperekplexia, a rare, potentially treatable neurogenetic disorder characterized by generalized stiffness and an excessive startle reflex to unexpected stimuli in newborns (951, 952).

SLC6A7: PROT (SLC6A7) is a predominantly brain-specific transporter of L-proline that is expressed by certain subpopulations of glutamatergic neurons (953). According to the HPA, PROT is highly expressed in the cerebellum and cerebral cortex, among other brain areas, and at low levels in the small intestine and enteroendocrine cells. The L-proline transporter PROT is closely linked to glutamatergic neurotransmission, where L-proline modulates the function of the NMDA receptor. NMDA receptor-mediated excitotoxicity is a major cause of neuronal death after stroke, triggered by the uncontrolled release of glutamate during the ischemic process. The distribution of NMDA receptors in glutamatergic neurotransmission has been shown to correlate with the proline transporter PROT (954). Inhibition of PROT increases extrasynaptic levels of proline and favors NMDA receptor function in glutamatergic synapses, leading to neuroprotection (955). Newly developed inhibitors of PROT such as LQFM215 have been shown to promote neuroprotection and neuro-repair in an acute ischemic stroke model (955).

SLC6A9: GlyT1 (SLC6A9) is a Na⁺- and Cl^--coupled glycine transporter expressed at the highest levels in the spinal cord, brainstem, diencephalon, and retina, and to a lesser extent in the olfactory bulb and brain hemispheres (944). According to the HPA, it is also widely expressed in peripheral tissues, particularly the skin, esophagus, adrenal gland, and vagina. In the CNS, GlyT1 (SLC6A9) is expressed in astrocytes and is involved in glutamatergic neurotransmission, as GlyT1 inhibition can modulate glutamatergic neurotransmission through NMDA receptors (956).

One functional role of GlyT1 (SLC6A9) is to take up waste glycine from synaptic sites after glycine release into astrocytes through cotransport with 2Na⁺ and 1 Cl^- per 1 glycine (949, 957). However, when intracellular glycine concentration is increased and/or after acute increase in intracellular Na⁺ concentration or acute membrane depolarization, GlyT1 can transport in the reverse mode and release glycine (Huang et al. 2004) (949, 958, 959). The role of Cl^- in the transport cycle was subsequently investigated, revealing unexpected insight into Cl^--dependent conformational changes in the GlyT1 transporter (960).

In addition, it has been shown that dopamine-induced activation of G protein-coupled dopamine receptor D5 on astrocytes induces GlyT1 function in the reverse mode via activation of phospholipase C (PLC), causing astrocytes to release glycine. Glycine is a co-agonist of NMDA receptors, which in turn play an essential role in synaptic plasticity, neural development, and glutamate-induced neurotoxicity (961, 962). Thus, cortical dopamine release can alter neuronal excitability through glycine release from astrocytes.

GlyT1 (SLC6A9) plays an important role in Parkinson disease in terms of involuntary movements (dyskinesia) and psychosis. Evidence for this comes from the demonstration that inhibition of GlyT1, which removes glycine from the synaptic cleft to enhance the action of glycine at NMDA receptors, alleviates the aforementioned Parkinson disease symptoms (963–965). First, the effect of bitopertin, a selective GlyT1 inhibitor (965), on the severity of parkinsonism was tested (963). Bitopertin has been shown to increase glycine levels in the striatum and cerebrospinal fluid (965), which could potentiate NMDA receptor activity and regulate dopaminergic and glutamatergic neurotransmission (966). GlyT1 inhibition may reduce both parkinsonism and L-DOPA-induced dyskinesia, potentially providing a novel approach to the treatment and prevention of Parkinson disease (963). As the compound has undergone several clinical trials for psychiatric indications (967), it may be suitable for repurposing for Parkinson disease if efficacy is demonstrated in preclinical models (963).

Bitopertin has also been investigated as a potential treatment for schizophrenia symptoms because in patients with schizophrenia, prolonged hypofunction of NMDA receptors can lead to impaired synaptic plasticity and impaired cognitive function. In fact, the majority of patients with schizophrenia suffer from cognitive impairment (968, 969). As there is increasing evidence that hypofunction of glycine and glutamate co-activated NMDA receptors is key to the pathophysiology of schizophrenia, enhancing neurotransmission through NMDA receptors holds promise for the treatment of these disorders (970, 971). GlyT1 has also been shown to co-localize with NMDA receptors in glutamatergic neurons (972–974). The development of GlyT1 inhibitors to increase glycine availability in the synaptic cleft and restore impaired NMDA activity is considered an important approach to treat schizophrenia-associated cognitive impairment (971).

The novel GlyT1-specific inhibitor iclepertin has been developed and has shown promising results in non-clinical studies as well as in Phase I and II clinical trials, demonstrating safety, tolerability and pro-cognitive effects in patients with schizophrenia (82). In addition, sarcosine, an amino acid found naturally in humans, has been shown to be a potent and prototypical endogenous inhibitor of GlyT1. If successful, iclepertin could become the first pharmacotherapy for the treatment of cognitive impairment associated with schizophrenia.

Cryo-EM structures of human GlyT1 in the apo state (GlyT1Apo), in complex with iclepertin and sarcosine, allowed the investigation of the binding pocket and the inhibitory mechanism of drugs targeting GlyT1 (971). Three cholesterol binding sites were identified in GlyT1, two of which are conformation dependent. Transport kinetics studies show that a delicate binding equilibrium for cholesterol is critical for the conformational transition of GlyT1. The study represents a major breakthrough in our understanding of the physiological and pharmacological aspects of GlyT1 (971).

Mice lacking Glyt1 (Slc6a9) expression have been shown to be nonviable (975). Newborn Glyt1-deficient mice are anatomically normal with no airway or lung malformations, but show severe motor and respiratory deficits and die during the first postnatal day (975). In brain stem slices from Glyt1-deficient mice, in vitro respiratory activity was markedly reduced, and insufficient synaptic clearance was shown to lead to elevated extracellular glycine concentrations, thereby inducing glycine receptor hyperactivity. Thus, the study demonstrated that GlyT1 is essential for lowering glycine concentrations at inhibitory glycine receptors and that it plays a critical role in regulating centrally generated rhythmic motor functions that are essential for autonomous neonatal life, such as breathing (975).

Likewise, rare loss-of-function mutations within the human SLC6A9 gene have been associated with GlyT1 encephalopathy (976). The disease causes severe postnatal respiratory failure, muscle hypotonia, and arthrogryposis, a condition with multiple joint contractures or stiffness. As highlighted above, GlyT1, which is expressed by major glial cell populations and a subset of glutamatergic neurons, facilitates rapid glycine clearance from the synaptic cleft and additionally regulates NMDA receptor function by controlling the extracellular glycine concentration at a subset of excitatory synapses (944, 975). Newborn infants showed severe respiratory failure requiring continuous ventilation, encephalopathy, hypotonia progressing to limp hypertonia in response to loud sounds and tactile stimulation, global developmental delay and dysmorphic features, in addition to muscle abnormalities (977). In all patients, mildly elevated CSF glycine concentration was observed without changes in serum glycine, which may be a good diagnostic marker for GlyT1 encephalopathy in the future (978).

GlyT1 (SLC6A9) has also been reported to function to accumulate glycine as an osmolyte in embryos (979). Early preimplantation mouse embryos are very sensitive to increases in external osmolarity. While somatic cells have characteristic organic osmolyte transporters, early embryos instead have their own unique organic osmolyte transporters, the main one being the classical “System Gly”, which accepts glycine and sarcosine (N-methylglycine) and is equivalent to GlyT1 (SLC6A9) (980).

SLC6A10P – Pseudogene: SLC6A10P was previously thought to be protein coding and known as SLC6A10 but is now considered to be an unprocessed transcribed pseudogene.

SLC6A14: ATB^0,+ (SLC6A14) is a Na⁺ and Cl^--dependent transporter of neutral and cationic amino acids. It represents the B^0,+ amino acid transporter system which is specific for neutral (index “0”) and basic (index “+”) amino acids (981). SLC6A14 has the highest affinity for the nonpolar amino acids isoleucine, leucine, methionine, valine and serine but glutamine, asparagine and arginine are among its substrates as well (981, 982). ATB^0,+ (SLC6A14) is expressed in lung, trachea, salivary gland, at somewhat lower levels also in stomach, mammary gland, epididymis, prostate and hippocampus (981). ATB^0,+ (SLC6A14) shows high affinity for the neutral amino acids Ile, Leu, Met, Val and Ser (981) and also transports several D-amino acids (983). Because of its expression in the colon at the luminal membrane, ATB^0,+ (SLC6A14) has been proposed contribute to the uptake of D-serine derived from bacteria (983). The transport mediated by ATB^0,+ (SLC6A14) is coupled to the cotransport of 2 Na⁺ ions and 1 Cl^- ion (981).

ATB^0,+ (SLC6A14) is also known as CT2 (carnitine transporter 2) because it has been shown to also transport L-carnitine (984), the compound that is critical in facilitating the transfer of long-chain fatty acids into mitochondria for β-oxidation. Both the carnitine transporter CT1/OCTN2 (SLC22A5) and ATB^0,+ (SLC6A14) show significant expression in colon tumor cells compared to normal colon tissue, which shows little or no expression (985).

Cancer cells need a constant supply of nutrients and ATB^0,+ (SLC6A14) is highly expressed in several types of cancer, and in colon cancer it has been shown to function as a tumor promoter (986, 987). Notably, cancer-associated expression of this transporter is observed in tissues that normally do not express this transporter, suggesting a metabolic rewiring of cancer cells. The ATB^0,+ (SLC6A14) transport substrates carnitine, leucine, glutamine, and arginine are all important in supporting cancer growth through metabolic rewiring.

To upregulate ATB^0,+ (SLC6A14) in cancer cells, the protein must first exit the ER and traffic to the plasma membrane through a process that is triggered by active heat shock proteins HSP70 (HSPA14) and HSP90-β (HSP90AB1), which rescue the transporter from proteolytic degradation (988). ER exit then requires recognition of ATB^0,+ (SLC6A14) by the vesicle trafficking protein SEC24C subunit of the COPII coat protein complex, which promotes vesicular trafficking to the plasma membrane. The SEC24C protein is phosphorylated by protein kinase B, also known as Akt, which is hyperactivated in many cancers, and AKT has been shown to regulate ER export of the amino acid transporter (989). Inhibition of SLC6A14 trafficking prior to Akt action by inhibiting HSPs is considered one of the promising anticancer strategies, and the HSP90 inhibitor ganetespib has undergone phase II clinical trials in combination with chemotherapeutics (990, 991). In addition, given the central role of Akt in switching cancer cell metabolism by enabling adaptation to nutrient availability, combination therapies with both Akt and SLC6A14 inhibitors are expected to be promising therapeutic strategies (989).

Serine is also an important amino acid in tumorigenesis. In a search for serine transporters involved in tumorigenesis using an arrayed RNAi screen, SLC6A14 and SLC25A15 emerged. SLC6A14 was identified as a major cytoplasmic transporter of serine, which can then enter the mitochondria via the SLC25A15 serine transporters (992). While cells can perform de novo serine synthesis, most transformed cells rely on serine uptake to meet their increased biosynthetic needs. Dual targeting of SLC6A14 and SLC25A15 has been shown to reduce serine uptake and growth of colorectal cancer cells in vitro and in vivo (992).

ATB^0,+ (SLC6A14) has also been shown to play a critical role in controlling macrophage inflammation (982). In macrophages undergoing inflammation, acetylation of residue K636 of ATB^0,+ (SLC6A14) was shown to reduce ubiquitination, leading to increased levels of ATB^0,+ (SLC6A14) expression in the plasma membrane, and the resulting increase in asparagine uptake was found to enhance interleukin-1β secretion by macrophages. The enhanced interleukin-1β secretion was found to involve interaction of asparagine with liver kinase B1 (LKB1), a serine/threonine kinase, although exactly how asparagine regulates the LKB1 activity and function remains unknown (982). In any case, interaction of asparagine with LKB1 in the cytosol has been reported to inhibit the nuclear accumulation of LKB1 (982). The latter normally inhibits the mitogen-activated protein kinase (MAPK) pathway-dependent NOD-like receptor family pyrin domain containing 3 (NLRP3) inflammasome activation and thus interleukin-1β secretion. In contrast, in the presence of asparagine taken up by ATB^0,+ (SLC6A14), nuclear LKB segregation is inhibited, allowing inflammasome activation and interleukin-1β secretion. These findings identify ATB^0,+ (SLC6A14) as a potential target for the modulation of inflammatory diseases.

3) The nutrient amino acid transporters subgroup:

a) Ancillary protein-associated subclass

This subgroup includes B⁰AT3 (SLC6A18), B⁰AT1 (SLC6A19) and SIT1 (SLC6A20).

Amino acid transporters of this subgroup either require ancillary proteins for their expression in the plasma membrane (B⁰AT3/SLC6A18 and B⁰AT1/SLC6A19) or optionally associate with such ancillary proteins (SIT/SLC6A20). In the intestine or lung, the auxiliary protein is the plasma membrane-anchored angiotensin-converting enzyme 2 (ACE2), and in the kidney it is collectrin, a collecting duct-specific short isoform of ACE2. ACE2 is also the receptor of the SARS-CoV-2 coronavirus that caused the 2019 coronavirus pandemic. ACE2 is a type I transmembrane protein consisting of a single TMH and a luminal N-terminal domain containing a carboxypeptidase. The latter removes the terminal amino acid from peptides generated during the digestive process after a protein-rich meal, and it preferentially releases large neutral amino acids, including tryptophan or phenylalanine, which are substrates of B⁰AT1 (993). In the proximal tubules of the kidney, collectrin, a smaller isoform of ACE2 that lacks the extracellular peptidase domain, is responsible for amino acid transporter-trafficking to the plasma membrane (994).

SLC6A18: B⁰AT3 (SLC6A18), also known as XT2, has long been an orphan transporter. Slc6a18 knockout mice have been generated and Slc6a18 null mice exhibited high levels of glycine in the urine (995), supporting the hypothesis that this orphan transporter functions as an amino acid transporter. Its long-sought functional expression was finally achieved by co-expression of B⁰AT3/XT2 (SLC6A18) with the ancillary protein ACE2, the extended version of collectrin, in Xenopus oocytes (996). B⁰AT3 (SLC6A18) was shown to function as a Na⁺ and Cl^- dependent neutral amino acid transporter (996).

B⁰AT3/XT2 localized to the apical membrane of the mouse renal proximal tubule, mainly in the S3 segments, where it was found to be highly expressed (997). Expression was also detected in mouse enterocytes lining the brush border membrane of intestinal villi. The HPA suggests that B⁰AT3 (SLC6A18) is expressed nearly exclusively in the proximal tubules of the kidney.

In the kidney, B⁰AT3/XT2 (SLC6A18) has been shown to be required for renal reabsorption of residual tubular amino acids not reabsorbed in the early proximal tubule segments by B⁰AT1 (SLC6A19) or of neutral amino acids released by amino acid exchangers, e.g., during cystine reabsorption with release of neutral amino acids via b^0,+ (SLC7A9) (996, 997),

SLC6A19: B⁰AT1 (SLC6A19), also known as XT3, is an apical transporter in epithelial cells that is highly and specifically expressed in the brush border membrane of the small intestine (Fig. 17) and early renal proximal tubules (S1 segment) starting at the glomerulus (997). B⁰AT1 (SLC6A19) transports all neutral amino acids and large aliphatic amino acids such as methionine, leucine, valine, etc. are the preferred substrates (998, 999). A stoichiometry of 1 Na⁺/amino acid was determined and the transport was chloride-independent. Thus, it mediates the absorption of neutral amino acids generated after a protein-rich meal by aminopeptidases and carboxypeptidases, including ACE2. In the early proximal tubules of the kidney, it reabsorbs filtered neutral amino acids.

Fig. 17. Intestinal absorption of amino acids and oligopeptides.

AA, Amino acid, NAA, neutral amino acids. LAT4 (SLC43A2) was reported to be mainly present in crypt cells (6) (see the SLC43 family description).

For trafficking to the apical membranes, it requires the ancillary proteins ACE2 in the intestine and collectrin in the kidney. Hartnup disease, an autosomal recessive metabolic disorder, is caused by mutations in SLC6A19 and intestinal absorption and renal reabsorption of neutral amino acids are impaired (1000–1002). Patients with Hartnup’s disease present with neutral aminoaciduria, indicanuria (caused by a defect in tryptophan absorption), photosensitive pellagra-like rash (due to niacin deficiency), cerebellar ataxia, anxiety, depression, and mild intellectual disability (1003). The lack of absorption of essential amino acids such as tryptophan, which can be converted to serotonin, melatonin, and niacin, is responsible for the symptoms. In most Hartnup patients, a high-protein diet has been found to overcome the lack of transport of neutral amino acids (1004). The reason is that the brush border intestinal di- and tripeptide transporter PepT1 (SLC15A1) can transport oligopeptides containing just about any amino acid (139) (see the SLC15 family description). Under normal healthy conditions, the PepT1-mediated contribution to amino acid absorption becomes important after high dietary protein intake when amino acid transporters such as B⁰AT1 (SLC6A19) are saturated, and PepT1 can provide the additional absorptive capacity needed (1005).

As mentioned above, in addition to being an enzyme, ACE2 is also a functional cell surface receptor through which SARS-CoV-2 enters host cells (1006). A recent cryo-EM structure has clarified the interaction of B⁰AT1 (SLC6A19) with ACE2, showing that ACE2 is an integral part of the transporter in the membrane (1007). The data show that the ACE2-B⁰AT1 complex is assembled as a dimer of heterodimers, with the collectrin-like domain of ACE2 mediating homodimerization. The receptor binding domain (RBD) of the surface spike glycoprotein of SARS-CoV-2 is recognized by the extracellular peptidase domain of ACE2 mainly through polar residues. These findings provide important insights into coronavirus recognition and the role of B⁰AT1 (SLC6A19) in the gastrointestinal route of infectivity.

Interestingly, many critically ill COVID-19 patients developed gastrointestinal disturbances, including vomiting and diarrhea, and it has been hypothesized that the SARS-CoV-2 spike protein, when bound to intestinal ACE2 in complex with B⁰AT1 (SLC16A19), negatively regulates neutral amino acid uptake (1004). This explains not only the gastrointestinal but also the systemic disturbances in COVID-19 and suggests that amino acid supplementation would be helpful.

To gain insight into the role of an allosteric site of B⁰AT1 (SLC16A19) in the transport and inhibition mechanism of the SLC6-AC2 heterodimeric amino acid transporters, high-resolution cryo-EM structures were determined with and without inhibitors derived from high-throughput screening (1008). The structures studied showed that these inhibitors bind to an allosteric binding site in the vestibule of the transporter, thereby preventing the movement of TM1 and TM6 required for the transporter to undergo a conformational change from an outwardly open to an occluded state.

To understand the mechanism behind the structural and conformational dynamics of B⁰AT1 (SLC16A19) in the presence of lipids, a state-of-the-art all-atom molecular dynamics simulation of B⁰AT1 in lipid mimetic bilayers was employed to shed light on how lipid modulates B⁰AT1 structure (1009). This all-atom molecular dynamics study provides a platform for future investigations of the structure-function mechanism of B⁰AT1 in realistic lipid mimetic bilayers. It also provides a framework for the development of new therapeutic agents targeting this transporter.

SLC6A20: SIT1 (SLC6A20), also known as IMINO system, is expressed primarily in epithelial cells of the small intestine (Fig. 17), but expression has also been detected in proximal tubule S3 segments of the kidney and the choroid plexus and lung, among other tissues, based on studies in the rat (1010, 1011). In the choroid plexus, it may be expressed on the apical (CSF-facing) side (see Fig. 11) (1012). SIT1 (SLC6A20) mediates the Na⁺-coupled uptake of mainly proline and does not efficiently recognize other α-amino acids such as alanine or lysine (1010, 1011). The coupling stoichiometry was shown as 2 Na⁺, 1 Cl^- per proline molecule. Interestingly, functional expression in Xenopus oocytes did not necessarily require co-expression with the ancillary protein ACE2, even though ACE2 forms a heteromeric complex with SIT, yet trafficking of SIT1 was enhanced by co-expression with ACE2 (1010, 1013). Proline is a prerequisite for collagen synthesis and thus the expression in the lung may be important because it has a high collagen content. SIT1 (SLC6A20) has also been shown to transport the osmolyte betaine (1011) where it may complement the betaine transporter BGT (SLC6A12) in the sinusoidal membrane of the liver (see the SLC6A12 description). Iminoglycinuria is an autosomal recessive abnormality of renal transport of the imino acids proline and hydroxyproline. Interestingly, studies of patients have identified mutations in the gene encoding the proton amino acid transporter PaT2 (SLC36A2) as the major cause of Iminoglycinuria. However, mutants of SLC6A20 may contribute to Iminoglycinuria in combination with mutants in other amino acid transporter genes such as SLC36A2, possibly also SLC6A18, SLC6A19 or SLC6A20 (1014).

Genetic variations of SLC6A20 and also SLC6A19 have also been found to be associated with nephrolithiasis, possibly due to glycinuria (1015, 1016). However, the mechanism underlying this association is unclear. One possibility might be that the resulting free urinary glycine increases urinary oxalate concentrations and promotes oxalate stone formation, possibly by affecting the expression of renal SLC solute carriers for oxalate or citrate.

Proline may be required for collagen synthesis in certain brain compartments (1017). SIT1 (SLC6A20) is prominently expressed in the choroid plexus and may regulate extracellular proline concentrations in the central nervous system (see Fig. 11). SIT (SLC6A20) was also shown to regulate glycine homeostasis in the brain (1018). Glycine is a major inhibitory neurotransmitter with co-agonist activity for NMDA receptors, and while the GlyT1 (SLC6A9) and GlyT2 (SLC6A5) transporters regulate brain glycine levels, SLC6A20 has now been shown to also regulate brain glycine levels and NMDA receptor function. Thus, SIT (SLC6A20) is considered an attractive target for the treatment of brain disorders with suppressed NMDA receptor function, such as schizophrenia (1018).

SIT1 (SLC6A20) expressed in the lung is considered a potential new target for the treatment of COVID-19 (1019–1021), as it is associated with the ancillary protein ACE2, which is also the receptor of SARS-CoV-2 (see the description of SLC6A19 and Fig. 33, bottom part). The structures of ACE2-SIT1 recognized by the omicron variants of SARS-CoV-2 have now been resolved (1022). Furthermore, Cryo-EM structure of ACE2-SIT1 in complex with the highly potent inhibitor tiagabine was reported (1013). The study advances our understanding of the ACE2-SIT1 complex. This, in turn, is relevant to a better understanding of the binding of SARS-CoV-2 to its receptor ACE2 in human lung alveolar cells, where SIT1 and ACE2 are functionally expressed.

SLC6A21P - Pseudogene: This is a unitary pseudogene in human. Coding versions of this gene are present in rodents but these remain uncharacterized.

b) Ancillary protein-independent subclass

This subgroup includes B⁰AT2 (SLC6A15), NTT5 (SLC6A16), NTT4/XT1 (SLC6A17).

SLC6A15: B⁰AT2 (SLC6A15), also known as SBAT1, is abundantly expressed in the hypothalamus and other parts of the brain (1023, 1024). Its expression is mainly neuronal, including localization in many GABAergic neurons and spinal cord motor neurons, but it was also found in astrocytes near ventricles (1024). Interestingly, expression in the brain also appeared to coincide with areas known to regulate mood, behavior and food intake (1024). The HPA additionally suggests prominent expression in the retina, pituitary, and placenta. B⁰AT2 (SLC6A15) is a Na⁺- and Cl^--dependent electrogenic amino acid transporter with a Na⁺/amino acid cotransport ratio of 1:1 (1023, 1025). It transports large neutral amino acids, with proline and methionine and the branched-chain amino acids (BCAAs) leucine, isoleucine, and valine as preferred substrates (1023, 1025). Both proline and BCAAs are involved in the synthesis of the neurotransmitter glutamate, while methionine can be converted to homocysteic acid, a potent NMDA receptor agonist (1026). In brain areas, B⁰AT2 (SLC6A15) is expressed in neurons that release various neurotransmitters, so a possible role of B⁰AT2 is to also transport neurotransmitter precursors and neuromodulators (1024, 1027).

Functional coding variants of SLC6A15 have been identified as a risk factor for the development of major depressive disorder due to defective function of this transporter in neurons (1028, 1029). Studies in mice subjected to either conventional Slc6a15 knockout or virus-mediated hippocampal Slc6a15 overexpression revealed a role for this transporter in modulating emotional behavior, possibly through its effects on glutamatergic neurotransmission (1030). These findings suggest that targeting this transporter may be an innovative approach in the treatment of certain disorders affecting glutamatergic transmission, highlighting the need for the development of selective inhibitors.

Loratadine was first reported as a non-amino acid inhibitor of B⁰AT2 (1031). Subsequently, computational approaches led to the understanding of the molecular basis of B⁰AT2 inhibition by loratadine and to the discovery of a new inhibitor, tiagabine, an anticonvulsant drug prescribed off-label for the treatment of anxiety with antidepressant properties (1032). While tiagabine was considered a novel B⁰AT2 (SLC6A15) inhibitor, adding a new dimension to the pharmacological complexity of the drug, subsequent high-throughput screening yielded additional novel inhibitors with high potency, selectivity and physicochemical properties (1033) (preprint). These inhibitors are expected to allow targeting the transporter in relevant biological systems and to initiate studies to clarify the precise role of B⁰AT2 (SLC6A15) in diseases.

SLC6A16 – orphan transporter: The NTT5 (SLC6A16) transporter is an orphan transporter. Its structure appears to be that of a Na⁺- and Cl⁻-dependent transporter of the SLC6 family, but its substrates are unknown. The SLC6A16 mRNA is highly expressed in peripheral tissues, especially in testis, pancreas and prostate (1034). Transient transfection with epitope-tagged transporter constructs demonstrated NTT5 is predominantly intracellular, suggestive of a vesicular location (1035).

SLC6A17: NTT4/XT1 (SLC6A17; also called B⁰AT3, although this name is used for the protein encoded by SLC6A18) is a Na⁺-coupled vesicular neutral amino acid transporter reported to be expressed exclusively in the nervous system and specifically on synaptic vesicles in glutamatergic and some GABAergic neurons (1036, 1037). The HPA is consistent with high expression in excitatory and inhibitory neurons, but also suggests prominent expression in horizontal cells of the retina, in melanocytes and in the endocrine system, particularly the pituitary gland.

Using a combination of molecular manipulations to increase the expression of the NTT4/XT1 (SLC6A17) protein at the plasma membrane, the substrate profile of this vesicular transporter was determined and included leucine, proline, methionine and glutamine, similar to that of the closely related plasma membrane transporter B⁰AT2/SBAT1 (SLC6A15) (1037). Consistent with this, in a previous study, vesicle content and uptake were measured by gas chromatography in wild-type PC12 cells expressing endogenous rat Slc6a17 and compared with cells in which Slc6a17 expression was reduced by siRNA, and the data showed transport of proline, glycine, leucine and alanine (1038). It has been hypothesized that NTT4/XT1 (SLC6A17) may contribute to glutamate synthesis by supplying amino acids, that it may provide amine groups for the synthesis of glutamate from α-ketoglutarate (1037).

Amino acid transport was found to be coupled to Na⁺ cotransport but independent of Cl^- and was inhibited by low pH, analogous to SLC6A15 (1037).

In the mouse brain, NTT4/XT1 (SLC6A17) has been localized in both excitatory and inhibitory neurons, with high expression in synapses. NTT4/XT1 expression overlaps with VIAAT (SLC32A1) and VGLUT2 (SLC17A6) (1039). NTT4/XT1 (SLC6A17) has been localized in vesicles of presynaptic terminals (1036).

Mutations in SLC6A17 have been shown to cause intellectual disability (1040, 1041). The expression of functional NTT4/XT1 was found to be critical for synaptogenesis and neuritogenesis during prenatal development. The studies highlighted the importance of NTT4/XT1 in supplying glutamine to synaptic vesicles (1040).

An in silico structural model of the vesicular neutral amino acid transporter NTT4 (SLC6A17) was created, shedding light on the molecular determinants of substrate selectivity (1042). In addition, molecular dynamics simulations were performed to clarify the effects of the pathogenic mutations at the molecular level (1042).

4) The osmolyte transporter subgroup

This subgroup includes the taurine transporter TauT (SLC6A6), the creatine transporter CT1 (SLC6A8), the GABA transporter GAT3 (SLC6A11), the betaine transporter BGT (SLC6A12) and the GABA transporter GAT2 (SLC6A13).

SLC6A6: TauT (SLC6A6) is a Na⁺ and Cl^- dependent transporter of taurine (1043). Northern blot analysis showed that it is highly expressed in the kidney and evenly distributed in different parts of the brain. In situ hybridization in different brain areas localized TauT mRNA in the corpus callosum, striatum and anterior commissure (1043). According to the HPA, TauT is most highly expressed in the retina (rod and cone photoreceptors) and bone marrow (monocytes), and is also expressed at lower levels in proximal enterocytes, among other cell types.

Taurine is an organic osmolyte involved in cell volume regulation. It is one of the most abundant amino acids in the brain, retina, muscle tissue, and organs throughout the body, and as a sulfur-containing β-amino acid it also has important effects on antioxidant defense networks (1044).

Diet is an important source of taurine in humans, although it can also be synthesized in the brain and liver, especially during development (1045). After intestinal absorption, the kidney reabsorbs taurine in the proximal tubules with the involvement of the TauT transporter.

In the liver, TauT is likely involved in the formation of bile salts (i.e., before secreting bile acids, hepatocytes conjugate them with the amino acids taurine or glycine).

In the retina, taurine is critical for photoreceptor development and acts as a cytoprotectant against stress-induced neuronal damage and other pathological conditions (1046).

Slc6a6^−/− mice have severe retinal degeneration, significantly decreased taurine concentrations in plasma, kidney, liver and eyes (1047), and suffer from renal taurine loss and impaired ability to lower urine osmolality and increase urinary water excretion (1048).

Pathogenic variants of SLC6A6 have been shown to cause autosomal recessive retinal degeneration and cardiomyopathy (1049, 1050). In one affected family, the blood taurine levels of the two affected individuals were extremely low. After 24 months of daily oral taurine supplementation, the cardiomyopathy disappeared and the retinal degeneration was arrested in the younger affected sibling, but the older affected sibling had a completely destroyed retina before treatment.

The involvement of TauT (SLC6A6) in taurine transport at the blood-testis barrier has also been reported (1051).

TauT (SLC6A6) overexpression has been shown to correlate with poor prognosis and aggressive tumor behavior in certain cancers, such as gastric cancer (1052). Although current inhibitors have not been shown to be effective in the treatment of cancer, inhibition of TauT may offer therapeutic potential in the future. Understanding the structure of TauT may facilitate the development of targeted therapeutics and ultimately advance this potential cancer therapeutic. Single-particle cryo-EM reconstruction was used to determine the structure of TauT in complex with taurine, β-alanine, and in a substrate-free state (1053). The study elucidates the mechanisms of substrate binding of TauT as well as the conformational changes that are associated with it.

SLC6A8: CT1 (SLC6A8) is a Na⁺- and Cl⁻-dependent transporter required for cellular uptake of creatine, a key high-energy phosphate storage molecule important for maintaining ATP homeostasis (78, 1054). For example, in fast-twitch skeletal muscle, large pools of phosphorylcreatine are available for immediate regeneration of ATP hydrolyzed during short periods of intense work (1055). The substrate-specific creatine transporter CT1 (SLC6A8) is located in the plasma membrane of various energy-demanding cells and organs such as skeletal muscle, cardiac muscle and brain. At lower levels it is also present in the gastrointestinal tract, kidney, bladder, bone marrow, retina, etc., as suggested by the HPA.

Creatine transporter deficiency (CTD) is an X-linked disorder caused by mutations in the SLC6A8 gene. It disrupts creatine transport, leading to intellectual disability, behavioral abnormalities, speech delay, autism, epilepsy, and poorly developed muscle mass (1056–1058).

Studies in Slc6a8^−/y knockout mice revealed severely reduced skeletal muscle phosphocreatine/creatine levels, impaired motor function with severe muscle wasting, and increased glucose metabolism due to activation of AMPK (AMP-activated protein kinase), a central regulator of energy homeostasis (1057). Subsequent studies of Slc6a8^−/y mice revealed that the muscles of these animals exhibit an atrophic phenotype accompanied by fiber ultrastructural changes, decreased performance, and increased expression of key E3 ubiquitin ligases associated with the progression of atrophy (1058). In addition, mitochondria were shown to exhibit significant morphological abnormalities, reduced membrane potential, and impaired mitochondrial Ca²⁺ uptake, coupled with changes in the expression of proteins involved in mitochondrial Ca²⁺ homeostasis.

Creatine and phosphocreatine are increasingly recognized as playing a critical role in energy homeostasis in the retina. It has been suggested that at the inner blood-retinal barrier, creatine synthesized in Müller cells is transported to the photoreceptor cells via CT1 (SLC6A8) to ensure energy supply (1059). It should be noted, however, that MCT12 (SLC16A12) also functions as a creatine transporter, which is expressed in the kidney and in the retina (see the SLC16 family description). It is possible that in fact MCT12 (SLC16A12) is primarily responsible for this function, as it is this transporter whose genetic defects cause juvenile cataracts (1060).

Increased creatine transporter expression has also been associated with cancer progression. High levels of CT1 (SLC6A8) have been found in triple negative breast cancer, hepatocellular carcinoma or non-small cell lung cancer (1061). In triple-negative breast cancer, creatine transporter overexpression under hypoxic conditions and creatine accumulation led to cell growth and survival (1062). Blocking CT1 has been shown to delay cancer progression, and an inhibitor of this transporter (RGX-202) is being tested in the first phase of clinical trials for the treatment of advanced colorectal cancer (1063).

SLC6A11: GAT3 (SLC6A11) is a GABA transporter that is highly and almost exclusively expressed in brain (1064). In addition, the HPA suggests expression in the retina. GAT3 also recognizes β-alanine as a substrate (1065). It is predominantly expressed in glia cells, and localizes to GABAergic synapses, highlighting its role in regulating GABA signaling (1064) (see Fig. 16). A recent study showed that GAT3 (SLC6A11) is regulated by the circadian clock. Specifically, circadian oscillation of extracellular GABA was found to be regulated in the suprachiasmatic nucleus by GAT3 (SLC6A11), with uptake peaking during circadian day, leading to the daytime trough and nighttime peak (1066).

SLC6A12: BGT (SLC6A12) is a Na⁺-coupled betaine and GABA transporter that is only partially dependent on Cl⁻ (1067, 1068). It is mainly expressed in liver, kidney and brain (1069). BGT (SLC6A12) plays a role in osmoregulation together with TauT (SLC6A6). In epithelial cells it is expressed on the basolateral membranes (1070).

In the kidney, BGT (SLC6A12) delivers betaine into renal epithelial cells as a known osmolyte to protect them from hypertonic stress in the inner medulla and maintain normal cell volume. In the kidney, BGT (SLC6A12) expression is regulated by the tonicity-responsive enhancer-binding protein (TonEBP) transcription factor (1071).

BGT (SLC6A12) is highly expressed in the liver, where it influences liver physiology and disease. It is present in the plasma membranes of parenchymal and sinusoidal endothelial cells and recent studies in Slc6a12^−/− mice have shown that BGT deficiency prevents acute liver failure (1072). In addition, BGT (SLC6A12) was found to be significantly downregulated in patients with liver failure and in mice with experimental acute liver failure. Similarly, BGT (SLC6A12) deficiency or treatment with a BGT inhibitor, NNC 05-2090, decreased apoptosis and stimulated expression of the anti-apoptotic gene c-Met (1072). The results suggest that BGT (SLC6A12) is a promising drug target for the treatment of acute liver failure.

The brain localization of BGT (SLC6A12) is less well defined. It has been reported that a polymorphism in SLC6A12 increases the risk of temporal lobe epilepsy (1073). However, the importance of BGT (SLC6A12) in the regulation of brain GABA levels is rather unlikely due to the low expression of BGT, its slow turnover rate and low affinity for GABA compared to GAT1 (1068).

SLC6A13: GAT2 (SLC6A13) is highly expressed in kidney. It is also expressed in liver and brain areas (910, 1074). However, although it is expressed in brain areas, its expression pattern doesn’t overlap with markers of GABAergic signaling (78). Studies of Slc6a13^−/− mice revealed that GAT2 is unimportant for GABA inactivation and that it acts as a taurine transporter as well (1075). In the liver, it was proposed to maintain low GABA levels in blood plasma, which is important if GABA acts as a peripheral signaling molecule (1075). Likewise, in the kidneys, GAT2 (SLC6A13) immunolabeling was observed in the cortex on basolateral, blood-facing membranes (1075). GAT2 (SLC6A13) is also believed to be an important taurine transporter. In the liver, GAT2 (SLC6A13) was found to be expressed at the sinusoidal membrane of the periportal region where taurine may be used for conjugation with bile acids in parallel to TauT (SLC6A6). Considering the fairly high expression levels in proximal tubules, it is plausible that GAT2 (SLC6A13) plays a specific role in the renal handling of taurine (1075). At the BBB it was proposed that GAT2 (SLC6A13) serves as an efflux transporter for brain taurine (1075).

5) SLC6-family K⁺-coupled transporter from tobacco hornworm Manduca sexta

Although not directly relevant to human SLC physiology, the following provides interesting information on an alternative ion-coupling mechanism of a distant SLC6 family member: To investigate the mechanism of cation selectivity of SLC6 family members, a K⁺-coupled neutral amino acid transporter KAAT1 (SLC6 family) was identified in the midgut of Manduca sexta larvae by expression cloning with Xenopus oocytes (1076), leading to the first identification of a K⁺-coupled amino acid transporter. KAAT1 exploits the high luminal K⁺ concentration and pH of the larval lumen of the anterior midgut to drive amino acid uptake in a K⁺-coupled manner (1077). KAAT1 has an unusual cation selectivity, being activated by K⁺ and Li⁺ in addition to Na⁺ (1078).

Orphan transporter family member (1)

SLC6A16 (NTT5)

SLC7 Cationic amino acid transporter/glycoprotein-associated family (2.A.3/AA_permease_2/APC)

Discovery: Amino acids play crucial roles in cellular metabolism, growth, and proliferation (365, 1079). The first family member of the SLC7 family was identified as mCAT-1 (SLC7A1), for mouse cationic amino acid transporter (1080), originally discovered as the receptor for murine ecotropic leukemia viruses with unknown function (1081). The identification of the second branch of this amino acid transporter family was achieved in 1998 by expression cloning, screening for aldosterone-dependent transporters. The resulting rat Na⁺-independent neutral amino acid transporter (SLC7A5) was designated LAT1 (L-type amino acid transporter 1). At the same time, LAT1 was isolated from C6 glioma cells by expression cloning (154). For functional expression in Xenopus oocytes, LAT1 required the coexpression of SLC3A2 heavy chain 4F2 cell surface antigen (see SLC3 family description).

Gene family members (13 + 1 pseudogene):
SLC7A1 (CAT-1)	SLC7A3 (CAT-3)	SLC7A5 (LAT1)	SLC7A7 (y+LAT1)
SLC7A2 (CAT-2)	SLC7A4 (CAT-4)	SLC7A6 (y+LAT2)	SLC7A8 (LAT2)
SLC7A9 (b0,+)	SLC7A11 (xCT)	SLC7A14
SLC7A10 (asc-1)	SLC7A13 (AGT-1)	SLC7A15P (pseudogene)

Molecular aspects, physiological roles and links to disease

SLC7 has 13 members, all of which are amino acid transporters that play important functional roles in nutrient uptake, reuptake or recycling of neurotransmitters, acid-base balance, nitrogen balance, energy balance, redox homeostasis, and in providing amino acids as essential metabolites for fatty acid biosynthesis, membrane lipids, and nucleotide synthesis (194, 365, 1082).

SLC7 family members belong to the Amino Acid Transporter (AAT) family (TC 2.A.3.1), which is part of the APC transporter superfamily. They share the LeuT fold, which consists of a 5 + 5 transmembrane spanning domain inverted structural repeat (see Section 8).

In cancer cells and diabetes, dysregulation of these amino acid transporters drives metabolic reprogramming that alters intracellular amino acid levels.

Some members of the SLC7 family require an SLC3 ancillary glycoprotein for functional expression in the plasma membrane - see below under “Glycoprotein-associated transporters”. These transporters form a heterodimeric amino acid transporter (HAT) complex consisting of an SLC3 single TMH protein, called the “heavy chain”, and an SLC7 multi-TMH membrane transport protein, called the “light chain” (see also the SLC3 family description and Fig. 9).

The SLC7 family can be divided into the following subgroups (see Fig. 18):

Fig. 18. Phylogenetic tree of the SLC7 family.

• 1)Non-Glycoprotein associated transporters: SLC7A1 (CAT-1), SLC7A2 (CAT-2) SLC7A3 (CAT-3), SLC7A4 (CAT-4), SLC7A14, SLC7A13 (AGT-1) and Slc7a15 (rodent only)
• 2)Glycoprotein associated transporters: SLC7A5 (LAT1), SLC7A6 (y⁺LAT2), SLC7A7 (y⁺LAT1), SLC7A8 (LAT2), SLC7A9 (b^0,+), SLC7A10 (asc-1), SLC7A11 (xCT) and Slc7a12 (rodent only)

1) Non-Glycoprotein associated transporters

a) Cationic amino acid transporters:

These include SLC7A1 (CAT-1), SLC7A2 (CAT-2), SLC7A3 (CAT-3), SLC7A4 (CAT-4) and SLC7A14.

The transport of cationic amino acids into cells is critical for arginine-, lysine-, and ornithine-dependent metabolic pathways. In particular, arginine is the precursor for the synthesis of nitric oxide (NO), creatine and urea. Ornithine is the starting point for polyamine synthesis.

SLC7A1: The discovery of the ecotropic retrovirus receptor as a cationic amino acid transporter CAT-1 (SLC7A1) was a landmark in the field of nutrient transport (1083). CAT-1 was the first amino acid transporter to be cloned and others were subsequently identified and characterized. These cationic amino acid transporters support important metabolic functions such as protein synthesis, nitric oxide synthesis, polyamine biosynthesis and interorgan amino acid flux and also have important oncogenic roles (1084).

CAT-1 (SLC7A1) is expressed almost ubiquitously, the only exception being the liver (1085). It is a high affinity cationic amino acid transporter. The transport system is consistent with the previously described system y⁺ (1086). A characteristic of this system is “trans-stimulation”, i.e., concentration-dependent stimulation of transport by substrates at the opposite side of the membrane. CAT-1 has the most pronounced trans-stimulation among the CATs (1083, 1087). CAT-1 (SLC7A1) mediates the basic supply of the cationic amino acids in most cells, including nitric oxide-producing cells.

Knockout studies in mice have shown that deletion of Slc7a1 is lethal (1088). The homozygous knockout mice die on the first day after birth, are 25% smaller than their wild-type littermates, and suffer from severe anemia. Indeed, L-arginine import via the cationic amino acid transporter CAT-1 was later shown to be essential for both differentiation and proliferation of erythrocytes (1089). The relatively normal development of most tissues in homozygous Slc7a1 knockout mice up to birth is probably due to the expression of CAT-3 (SLC7A3) during embryogenesis and fetal development (1090). Studies have shown that CAT-1 is involved in tumor progression in several cancers. CAT-1-mediated arginine was shown to contribute to nitric oxide production for the survival of breast cancer cell lines (1091). Increased arginine uptake due to CAT-1 overexpression was shown to be involved in the reprogramming of energy metabolism in ovarian cancer to induce tumor progression, and the study also revealed that it inhibits the expression of chemokine CCL4, resulting in reduced lymphocyte immune infiltration into tumors as part of immune escape (1092).

Arginine, whose endogenous synthesis can be initiated by citrulline, is transported across cell membranes by the cationic amino acid transporters SLC7A1, SLC7A2 and SLC7A3 (1093). Dietary supplementation may be necessary in certain pathological situations requiring high arginine levels. Of particular relevance, in human T cells, SLC7A1 is responsible for arginine uptake and its knockdown affects T cell proliferation as reviewed (1093). Briefly, arginine starvation induces T cell cycle arrest via GCN2 activation (1094), decreases T cell antigen receptor zeta chain expression (1095) and reduces T cell proliferation and cytokine production (1096). GCN2 (general control nonderepressible 2) is a serine/threonine protein kinase that senses amino acid deprivation by binding to uncharged tRNA to modulate amino acid metabolism in response to nutrient deprivation (1097). Arginine supplementation increases CD8+ T cell-mediated anti-tumor activity by promoting the generation of central memory-like T cells with high survivability in a mouse model (1098). Arginine is therefore important for the T cells to proliferate, activate and have proper effector function (1099).

SLC7A2: CAT-2 (SLC7A2) is a low affinity cationic amino acid transporter. There are two splice variants, CAT-2A and CAT-2B. The low-affinity CAT-2A is most abundant in liver, and is also expressed in skeletal muscle and pancreas (1083). CAT-2B is the inducible CAT isoform and it is often induced together with the inducible isoform of nitric oxide synthase (iNOS), which is generally coexpressed with CAT-1 (SLC7A1). CAT-2B has been shown to be present only in activated mouse macrophages and lymphocytes (1100).

Deficiency of CAT-2 was shown to protect mice from hyperoxia-induced lung injury (1101). The pathology of acute lung injury involves apoptosis of pulmonary endothelial cells induced by nitric oxide generated by iNOS. Mice deficient in CAT-2 (Slc7a2^-/-) and control mice were subjected to hyperoxia (>95% oxygen) or control oxygen levels (21% oxygen) for 60 h. In wild-type mice exposed to hyperoxia, the exhaled nitric oxide was twofold greater than in wild-type mice exposed to normoxia, whereas in slc7a2^-/- mice there was no significant difference between exhaled nitric oxide in animals exposed to hyperoxia or normoxia. Consistent with the hypothesis that CAT-2 deficiency protects mice from acute lung injury, the study concluded that iNOS-derived nitric oxide production is dependent on the uptake of L-arginine by CAT-2 (1101). CAT-2 may be a target for acute lung injury therapy, where blocking it could prevent or ease the signs of lung injury.

Elevated CAT-2 (SLC7A2) expression has been shown to be associated with an abnormal neuroinflammatory response and nitrosative stress in Huntington disease (1102). SLC7A2, in its function as an arginine transporter, is an important regulator of innate and adaptive immunity in macrophages (1103, 1104). SLC7A2 is also a top upregulated gene upon huntingtin deletion (1105). Neuroinflammation is associated with Huntington disease, and SLC7A2 has been shown to play a role in the neuroinflammatory stress response in Huntington disease cells and to be upregulated at the transcriptional level in Huntington disease cell models and patients (1102). This leads to an overactive response to neuroinflammatory responses in Huntington disease cells, with abnormally high iNOS induction and nitric oxide production, resulting in increased protein nitrosylation. Depletion of extracellular arginine or knockdown of SLC7A2 blocked iNOS induction and nitric oxide production, suggesting a novel pathway linking arginine uptake to nitrosative stress via SLC7A2 upregulation in Huntington disease pathogenesis. Arginine supplements may pose a greater risk to Huntington disease patients due to upregulation of SLC7A2 expression.

SLC7A3: CAT-3 (SLC7A3) is a cationic amino acid transporter. In mice, CAT-3 has been found in many developing tissues and in neurons (1083). In human, it is found in thymus, uterus, testis, hippocampus (neurons), stomach and smooth muscle. Unlike the other CATs, the SLC7A3 gene is located on the X chromosome. It has been found to be highly resistant to variation in humans, as evidenced by the low frequency of deleterious variants. Nevertheless, several missense variants have been identified in patients with autism spectrum disorder, which are characterized by a moderate loss of transport function, leading to autism and epilepsy, probably in association with other genetic factors (1106). It has been proposed that since NO is an important cell-cell signaling molecule in the central nervous system, reduced availability of arginine could reduce NO synthesis and signaling in neurons, leading to cognitive impairment. Alternatively, since arginine regulates the mammalian mTOR pathway, which controls neuronal survival, differentiation, development, and synaptic plasticity, this pathway may be affected by the genetic variants (1106).

SLC7A4 – orphan transporter: CAT-4 (SLC7A4) is expressed in brain, testis and placenta (1107). SLC7A4 and SLC7A14 (see below) are both more distantly related members of the cationic amino acid transporter group. The transport function of CAT-4 (SLC7A4) and its physiological role in mammalian cells remain elusive (1083, 1107).

SLC7A14: SLC7A14 has been reported to function as a lysosomal transporter for cationic amino acids (1108). High levels of expression have been found in the cerebellum and spinal cord (1108). According to the HPA, SLC7A14 is expressed almost exclusively in neurons, oligodendrocytes, horizontal cells and bipolar cells. Mutations in the SLC7A14 gene have been shown to cause autosomal recessive retinitis pigmentosa, resulting in loss of photoreceptor function (1109, 1110). Subsequently, they have also been shown to cause dysfunction of inner ear mechanosensory hair cells, leading to auditory neuropathy, a condition in which hearing is impaired to the point that the affected individual has difficulty understanding spoken words in noisy environments. SLC7A14 has been shown to be highly expressed in the lysosomes of mammalian cochlear inner hair cells (1111). Thus, autosomal recessive mutations of SLC7A14 have been shown to cause both loss of photoreceptor function and loss of inner ear hair cell function. The loss-of-function mutations of SLC7A14 altered protein trafficking and increased basal autophagy, leading to progressive cell degeneration (1111). This study links autophagy-lysosomal dysfunction to syndromic vision and hearing loss in mice and humans.

b) Transporters of other substrates:

These include SLC7A13 (AGT-1) and Slc7a15 (rodent only).

SLC7A13: AGT-1 (SLC7A13) is a Na⁺-independent L-aspartate/L-glutamate exchanger that also transports cystine. It is expressed on the apical membrane of the S3 segments of the renal proximal tubule (1112). Its genetic defects have been shown to be responsible for a specific type of cystinuria (1112). It is likely that the Na⁺-coupled glutamate transporter EAAC1/EAAT3 (SLC1A1), which is highly expressed in the S3 segments of the renal proximal tubule, reabsorbs glutamate and aspartate released into the renal tubular lumen via AGT-1 (SLC7A13) counter-transport, thereby preventing urinary loss of these acidic amino acids.

SLC7A15P – Pseudogene: This gene is coding in rodents, dog and chicken but not in primates. The human gene (SLC7A15P) at the equivalent location to mouse Slc7a15 has been pseudogenized and classed as a unitary pseudogene. ArpAT (Slc7a15) is expressed in the brain where it transports the amino acid L-DOPA (L-3,4-dihydroxyphenylalanine), the precursor for the synthesis of the neurotransmitters dopamine, norepinephrine and epinephrine (1113).

2) Glycoprotein-associated transporters, These include SLC7A5 (LAT1), SLC7A6 (y⁺LAT2), SLC7A7 (y⁺LAT1), SLC7A8 (LAT2), SLC7A10 (asc-1), SLC7A11 (xCT) and Slc7a12 (rodent only)

a) Glycoprotein-associated amino acid transporter – association with rBAT (SLC3A1):

This is represented by a single member, SLC7A9 (b^0,+).

SLC7A9: b^0,+ (SLC7A9) transports cystine, dibasic and neutral amino acids Na⁺-independently. It functions as an exchanger and preferentially exchanges extracellular cationic amino acids and cystine for intracellular neutral amino acids (194). It is expressed on the apical membrane of renal proximal tubule S1 and S3 segments and in intestinal epithelial cells (1114) (Fig. 17). As highlighted in the SLC3 family description, b^0,+ (SLC7A9) forms a heteromeric complex with the ancillary heavy chain D2/NAA-Tr/rBAT (SLC3A1). This is a necessary step for the functional expression of b^0,+ (SLC7A9) in the plasma membrane (see Fig. 9). In addition, genetic variants of the heterodimeric cystine transporter complex SLC3A1/SLC7A9 cause cystinuria, with SLC7A9 mutations resulting in a wide clinical variability (1115–1118) (Fig. 9).

The cryo-EM structure of the human heteromeric amino acid transporter b^0,+ (SLC7A9)-rBAT (SLC3A1), which exists as a dimer of the heterodimer, has been determined alone and in complex with arginine at resolutions of 2.7 and 2.3 Å, respectively. A ligand amino acid molecule is bound to the substrate-binding pocket, near which an occluded pocket important for substrate transport has been identified (1119). Several key residues in the pocket, as well as residues whose mutations correlate with cystinuria, have been biochemically investigated. The results provide new insights into the mechanism of action of the b^0,+-rBAT amino acid transporter complex.

b) Glycoprotein-associated amino acid transporters – association with 4F2hc (SLC3A2):

This subfamily includes SLC7A5 (LAT1), SLC7A6 (y⁺LAT2), SLC7A7 (y⁺LAT1), SLC7A8 (LAT2), SLC7A10 (asc-1) and SLC7A11 (xCT).

SLC7A5: LAT1 (SLC7A5) is widely expressed and transports large neutral amino acids such as branched-chain and aromatic amino acids, several of which are essential amino acids. Transport occurs pH-independent and Na⁺-independently, and LAT1 is functioning as an amino acid exchanger (154, 194, 1120–1122).

In analogy to the heteromeric complex of b^0,+ (SLC7A9) and D2/NAA-Tr/rBAT (SLC3A1) (Fig. 9), LAT1 (SLC7A5) forms a heterodimeric complex with the ancillary heavy chain 4F2hc (SLC3A2), also named CD98, which is required for its functional expression in the plasma membrane, and likewise LAT1 is covalently linked through a conserved disulfide bond to LAT1, thereby forming the heterodimeric LAT1-4F2hc complex (see description of the SLC3 family) (154, 194, 1121, 1122). The cryo-EM structure of the human LAT1-4F2hc heterodimer was determined at 3.3 Å resolution, providing insight into how LAT1 forms the disulfide-linked heterodimer with 4F2hc (1123). LAT1 embodies the canonical APC/LeuT fold and has an extended cavity that could accommodate bulky amino acids and drugs. 4F2hc binds to LAT1 through extracellular, transmembrane and putative cholesterol-mediated interactions. The results elucidate the principles of glycoprotein and solute carrier assembly and help design compounds and antibodies targeting the LAT1-4F2hc amino acid transporter complex.

LAT1 is a major leucine influx transporter in the plasma membrane in many cell types, including neurons, trophoblasts, and cancer cells (1124). In trophoblasts, for instance, LAT1 overexpression has been shown to significantly increase the uptake of essential amino acids, such as leucine, and to activate mTORC1, thereby promoting cell growth via nutrient-sensing pathways (661).

In addition to transporting branched-chain amino acids such as leucine and aromatic amino acids, LAT1 can also transport thyroid hormones. Furthermore, it can carry pharmaceutical drugs that mimic these structures. Because it is abundantly expressed in the BBB (see Fig. 33) and in tumors, LAT1 is being exploited as a target for drug delivery systems that rely on amino acid-mimicking prodrugs (1125). The anti-Parkinson disease drug L-DOPA, the anti-cancer drug melphalan and the anticonvulsant gabapentin are known to utilize LAT1/4F2hc for transport across the BBB and into target tissues (1126) (1127, 1128).

LAT1 (SLC7A5) has been shown to be overexpressed in various types of cancers, where it provides amino acids critical for tumor growth (1129–1133). The findings make this transporter a promising target for cancer therapy.

The mTOR signaling pathway is frequently activated in human cancers, which hijacks the process, promoting tumor growth by altering cancer cell metabolism (18). As described in the SLC3 family description and illustrated in Fig. 9C, studies on cancer cells have demonstrated that the lysosomal protein LAPTM4B (see Section 10, SLC-Like Proteins) has the capacity to directly recruit LAT1 from the plasma membrane to the lysosome, thereby shifting the cell from a catabolic to anabolic state (18, 660). Note that the LAT1/LAPTM4B interaction enables leucine to be imported into lysosomes, thereby contributing to the activation of mTORC1 via the V-ATPase from the luminal side (see Fig. 9C). In contrast, SLC38A9 regulates arginine-dependent release of lysosomal leucine, which is crucial for cytosolic sensing coupled to the activation of mTORC1 (see the description of SLC38A9).

In addition, the recruitment of mTORC2 to lysosomes in cancer cells is thought to play a role in cell migration and survival (1134). LAT1 has been shown to promote cell migration induced by the PKC activator phorbol ester phorbol 12-myristate 13-acetate (PMA) through the activation of mTORC2 at the lysosome (1134). The study revealed that in PMA-treated cells LAT1 translocates via endocytosis to lysosomes. The N-terminal ubiquitination of LAT1 has been implicated in PMA-triggered endocytosis leading to lysosomal localization. There, it interacts directly with Rictor (rapamycin-insensitive companion of mammalian target of rapamycin), which is a component of mTORC2. This results in the recruitment of mTORC2 to the lysosomes. Targeting LAT1 could block metastasis and disease progression in various types of cancer.

Therefore, LAT1 inhibition is a promising strategy for blocking cancer progression. Inhibiting LAT1 starves cancer cells, leading to reduced growth, arrested cell cycles, and even death.

Nanvuranlat (JPH203/KYT-0353) was developed as a first-in-class LAT1-specific inhibitor (1135). Nanvuranlat has shown antitumor activity in preclinical studies and efficacy in biliary tract cancer in clinical trials (1130, 1136, 1137). Inhibition of LAT1 by Nanvuranlat in cancer cells has been shown to suppress the G0/G1-S transition of the cell cycle by downregulating cyclin D1 via p38 MAPK activation (1136). Thus, this transporter has emerged as a promising target for cancer therapy. Nanvuranlat and its N-acetyl metabolite were extensively pharmacologically characterized, including structural insights into their LAT1 interactions (665). Both compounds were found to be highly selective towards LAT1 over LAT2 (SLC7A8) and other amino acid transporters. Nanvuranlat was shown to be a competitive, non-transportable LAT1 inhibitor (Ki = 38.7 nM). In contrast, its N-acetyl metabolite retains selectivity but with reduced affinity (Ki = 1.68 μM). Nanvuranlat has the potential for prolonged therapeutic efficacy, as it exhibits a sustained inhibitory effect on LAT1 even after its removal. These mechanistic and metabolic insights provide important information for understanding clinical efficacy of Nanvuranlat and advancing LAT1-targeted cancer therapies.

Further efforts to develop LAT1-selective compounds have included both substrates and inhibitors, with the aim of using them for clinical applications. For example, LAT1-specific 3-fluoro-α-methyl-L-tyrosine radiotracers were developed as LAT1-specific PET (positron emission tomography) and SPECT (single photon emission computed tomography). Tumor uptake of 3-[¹⁸F]fluoro-L-α-methyl-tyrosine was highly correlated with LAT1 (SLC7A5) expression. These probes have proven useful in clinical trials for predicting prognosis and assessing response to therapy (1138–1140). The addition of an α-methyl group to aromatic amino acids made these compounds highly selective for LAT1 (1141). A bicyclic phenylalanine analog, (R)-2-amino-1,2,3,4-tetrahydro-2-naphthoic acid, was also reported to exhibit high LAT1 selectivity comparable to that of α-methylphenylalanine (1142). However, the affinity of all these compounds for LAT1 is significantly lower than that of Nanvuranlat, despite the high selectivity they show (665).

Other LAT1 selective inhibitors that were based on the structure of the thyroid hormone T3, such as SKN103, have also been reported (1143). However, compared to Nanvuranlat, these compounds also showed lower affinity (665).

Yet another series of inhibitors, JX-119, JX-078, and JX-075, were shown to be high-affinity inhibitors with IC₅₀ values between 100 and 250 nM for the inhibition of leucine uptake (1133). Based on the reported IC₅₀ values, they appeared to have slightly lower LAT1 affinity than Nanvuranlat. The mechanism of transport inhibition has been investigated as well. Cryo-EM structures of the corresponding LAT1-4F2hc complexes with these three inhibitors revealed that the protein adopts an outward-facing occluded conformation (1133). In the structure, the inhibitors were found to bind in the classical substrate binding pocket with their tails wedged between the substrate binding site and TMH10 of LAT1. The underlying structural insight provides a basis for designing future drugs to target LAT1 very specifically (1133).

In addition, compounds based on dithiazole and dithiazine scaffolds have been identified as potent LAT1 inhibitors, although their selectivity profiles require further investigation (1144).

Collectively, the exceptional selectivity of Nanvuranlat for LAT1 over LAT2, coupled with its high affinity, has established it as a lead compound in the development of LAT1-targeted therapies (665). Moreover, in clinical trials, Nanvuranlat demonstrated acceptable safety and antitumor activity in patients with advanced solid tumors (1145). In addition, a recent Phase II clinical trial in patients with biliary tract cancer met its primary endpoint of improving progression-free survival (PFS), further highlighting the potential of Nanvuranlat as a therapeutic (1146). To clarify how LAT1 discriminates substrates and inhibitors including the clinically relevant drugs, the structural basis of anticancer drug recognition by LAT1 has been determined using the lipid nanodisc system and cryo-EM (1147). Nanvuranlat was shown to trap LAT1 in an outward-facing state with a U-shaped conformer, with its aminophenylbenzoxazole moiety pushing against TMH3 and bending TMH10. Physiological substrates such as L-phenylalanine have been shown to lack such effects, whereas melphalan used to treat various cancers presents a steric hindrance, explaining its inhibitory activity. The substrate-like behavior of the “classical” System L inhibitor BCH (2-amino-bicyclo-(2,2,1)-heptane-2-carboxylate) was confirmed and this compound was shown to induce an occluded state critical for transport. The results will guide future drug design and provide a structural basis for substrate recognition and inhibition of LAT1 (1147).

As mentioned above, LAT1 (SLC7A5) is also abundantly expressed in the blood-brain barrier (BBB), where it is believed to play an important role. Alterations in the function of LAT1 may be responsible for differences in cerebral phenylalanine content between individuals, as well as for the absence of intellectual disability in some patients with untreated phenylketonuria. The common variant rs113883650 of the SLC7A5 gene has been evaluated in relation to brain phenylalanine levels, suggesting that the variant may influence the amount of phenylalanine in the brain across the BBB (1148).

SLC7A6: y⁺LAT2 (SLC7A6) is widely expressed and transports cationic amino acids Na⁺-independently and large neutral L-amino acids Na⁺-dependently (1149). Analogous to LAT1, y⁺LAT2 forms a heteromeric complex with 4F2hc (SLC3A2) (1150). It functions as an exchanger and preferentially exchanges intracellular cationic amino acid for extracellular neutral amino acid and Na⁺ (1149, 1151). y⁺LAT2 (SLC7A6) and y⁺LAT1 (SLC7A7) (see below) are closely related transporters with similar transport properties and kinetics of arginine transport indicates comparable affinity for the substrate, but they have different tissue distributions (1151). According to the HPA, y⁺LAT2 (SLC7A6) is most strongly expressed in skeletal myocytes, microglial cells, neurons, type I alveolar epithelial cells and syncytiotrophoblasts but at negligible levels in kidney tubule cells, while y⁺LAT1 (SLC7A7) is most strongly expressed in kidney proximal tubule cells, proximal enterocytes and immune cells (see below).

SLC7A7: y⁺LAT1 (SLC7A7) is abundantly expressed in kidney, small intestine, peripheral blood leukocytes and at lower levels in lung and spleen (1152, 1153) (see Fig. 17). y⁺LAT1 (SLC7A7) forms a heterodimer with 4F2hc (SLC3A2) to form the active transporter responsible for y⁺L amino acid transport (1152). y⁺LAT1 transports cationic amino acids Na⁺-independently and large neutral L-amino acids Na⁺-dependently. It acts as an exchanger and preferentially exchanges intracellular cationic amino acids for extracellular neutral amino acids and Na⁺. In polarized kidney proximal tubule and small intestine epithelial cells, it mediates basolateral arginine, lysine or ornithine efflux in exchange for neutral amino acids and Na⁺ (1151, 1152).

SLC7A7 genetic variants are associated with lysinuric protein intolerance (LPI). LPI is a rare metabolic disorder that is characterized by poor intestinal absorption and renal reabsorption and increased urinary excretion of the cationic amino acids L-ornithine, L-arginine, and L-lysine, leading to urea cycle defects with protein intolerance (1153, 1154). Consistent with this, a case report of a rare kidney transplant in a patient with LPI showed a significant improvement in protein tolerance. This was most likely due to the fact that after kidney transplantation, the renal wasting of ornithine, lysine and arginine is abolished because the y⁺LAT1 transporter is functioning normally in the transplanted kidney (1155).

LPI is also linked to immune dysfunctions. This may be due to altered arginine metabolism in mononuclear cells, which can lead to an overproduction of the pro-inflammatory pathway of nitric oxide synthase (1156). Downregulation of SLC7A7 has been shown to induce an inflammatory phenotype in human macrophages and airway epithelial cells (1156). The results support a pathogenic model of LPI lung complications in which immune and airway epithelial cells are involved in a positive feedback loop responsible for inflammatory responses. Accordingly, pro-inflammatory cytokines from y⁺LAT1-deficient cells are expected to hyperstimulate production of the chemokine molecule RANTES by airway epithelial cells that recruit circulating monocytes into the airways (1156). This highlights the importance of macrophage-epithelial communication in lung homeostasis and host-pathogen interactions.

To clarify the relationship between specific mutations in y⁺LAT1 and LPI, the cryo-EM structures of the human y⁺LAT1-4F2hc complex in the apo state in an inward-open conformation and in the native substrate-bound state in an outward-open conformation were determined (1157). The study suggests that D243 in y⁺LAT1 plays a critical role in coordination with Na⁺ ions and substrate selectivity. Molecular dynamics simulations provide new insights into the substrate binding mechanisms and work cycles of heteromecic amino acid transporters, revealing the different transport mechanisms of cationic and neutral amino acids (1157).

SLC7A8: LAT2 (SLC7A8) transports neutral L-amino acids and T3/T4 thyroid hormones in a Na⁺-independent manner, acting as an exchanger (194). The SLC7A8/SLC3A2 heterodimer is primarily expressed in the renal proximal tubule, small intestine, BBB and placenta, where it is responsible for the flux of amino acids across cell barriers (1127, 1158–1160) (see Fig. 17).

Mutations in the SLC7A8 gene have been implicated in age-related hearing loss in humans and mice (1161). The underlying reason is that loss of function of LAT2 leads to damage in the inner ear, resulting in impairment of the delicate hair or nerve cells in the cochlea. Slc7a8^-/- knockout mice exhibited deafness with age-related hearing loss characteristics, defective hearing at high frequencies with early onset in homozygotes and progressive worsening with age in heterozygotes (1161).

The SLC7A8 gene has been shown to be associated with early disease progression in osteosarcoma, a malignant bone tumor that mainly affects children and adolescents (1162). Functional studies have shown that LAT2 facilitates the uptake of the chemotherapy drug doxorubicin. Lack of LAT2 expression is reported to be a prognostic factor for poor prognosis and reduced overall survival in patients without metastases (1162). The findings are expected to provide new opportunities to personalize the treatment of osteosarcoma patients.

The cryo-EM structure of the 4F2hc-LAT2 protein complex has been reported showing the 12 TMH LAT2 transporter with the substrate binding site (1163).

SLC7A10: The asc-1 (SLC7A10) transporter has been reported to be expressed in astrocytes and neurons (1164), where it is responsible for the release of D-serine and glycine and the uptake of D- and L-serine (1164–1169). According to the HPA, asc-1 (SLC7A10) is also expressed in adipose tissue and the mammary gland in addition to the brain. The asc-1 transporter functions Na⁺-independently and preferentially as an exchanger (194).

Given the expression of asc1 (SLC7A10) in adipose tissue, its role in obesity and insulin resistance has been investigated (1170). The transporter plays an important role in metabolic regulation in adipose tissue depots and adipocyte subtypes. In addition, asc-1 (SLC7A10) has a critical role in adipocyte differentiation where it is part of a regulatory network in subcutaneous fat involved in metabolic syndrome (MetS). Its expression has been shown to be inversely correlated with obesity and insulin resistance (1171). Thus, this transporter may serve as a target for therapeutics to reduce the risk of insulin resistance and T2D.

In glutamatergic synapses, binding of the co-agonists D-serine and glycine to the NMDA receptor at its allosteric site is required to transduce glutamatergic signals (1172). In Slc7a10 knockout mice, reduced glycine but not D-serine levels were observed in the nervous system (1167). These mice show severe symptoms such as tremors, ataxia and seizures, leading to early postnatal death.

The cryo-EM structure of the human Asc-1 transporter (SLC7A10) in complex with 4F2hc (SLC3A2) has been reported in the apo-, D-serine-, and L-alanine-bound states at 3.6 Å, 3.5 Å, and 3.4 Å, respectively (1173). The study reveals the alternating access mechanism underlying conformational changes in the complex and provides insight into substrate recognition and the transport cycle.

SLC7A11: xCT (SLC7A11) is a Na⁺-independent cystine-glutamate exchanger, expressed in macrophages, brain, retinal pigment cells, liver, kidney, epididymis, fallopian tube and thyroid gland (194). It functions as an exchanger and preferentially exchanges extracellular cystine for intracellular glutamic acid (194).

Regarding the role of SLC7 family members in cancer development, the cystine/glutamate exchanger xCT-4F2hc encoded by SLC7A11 and SLC3A2 has been suggested as a possible therapeutic target for cancer treatment, as it is overexpressed in a variety of cancers. The cystine/glutamate exchanger xCT provides intracellular cystine/cysteine for the production of glutathione, a major cellular antioxidant (1174), and inhibition of xCT causes accumulation of reactive oxygen species, which suppresses tumor growth. Recent studies have shown that xCT (SLC7A11) regulates fibrosis in liver fibrosis, cardiomyopathy, and numerous other pathophysiological processes (1175). In addition, the structure of the xCT-4F2hc complex bound to the xCT inhibitor erastin revealed novel molecular mechanisms underlying erastin-induced ferroptosis (1175).

c) Glycoprotein-associated amino acid transporter – association with unknown SLC3 heavy chain:

Slc7a12 (rodent only): Asc-2 (Slc7a12) is a neutral amino acid transporter expressed in skeletal muscle and kidney collecting duct of mice (1176). It is a heterodimeric amino acid transporter that is not associated with 4F2hc or rBAT, but is thought to be associated with another unknown heavy chain of the SLC3 family. In mouse kidney, Asc-2 is expressed in the collecting duct where it may transport alanine into cells to synthesize osmolytes.

Orphan transporter family member (1)

SLC7A4 (CAT-4)

SLC8 Na⁺/Ca²⁺ exchanger family (2.A.19.3/Na_Ca_ex/NCX)

Discovery: The cardiac sarcolemmal Na⁺/Ca²⁺ exchanger NCX1 (SLC8A1) was cloned as a founding member of the SLC8 family by screening an expression library with an antibody (1177).

Gene family members (4):
SLC8A1 (NCX1)	SLC8A3 (NCX3)
SLC8A2 (NCX2)	SLC8B1 (NCLX)

Molecular aspects, physiological roles and links to disease

Proteins of the SLC8 family belong to the Ca²⁺/Cation Antiporter (CaCA) family (TC 2.A.19.3). They are ubiquitous in animals, plants, yeast, archaea and divergent bacteria and have widely divergent sequences (295). SLC8 transporters share the structure of the NCX fold and thus have 10 TMHs, representing two repeated 5-TMH units. Between TMH5 and TMH6 they have a long cytosolic loop containing the Ca²⁺-binding regulatory domains CBD1 and CBD2 (242) (see also Section 8 and the description of the SLC24 family).

NCX1 (SLC8A1), NCX2 (SLC8A2), and NCX3 (SLC8A3) and their splice variants mediate Ca²⁺ fluxes across cell membranes in a tissue-specific manner and thus contribute significantly to the regulation of Ca²⁺-dependent events in many cell types. The different SLC8 paralogs and splice variants contribute to diverse functions such as excitation-contraction coupling, long-term potentiation and learning in the brain, blood pressure regulation, immune response, insulin secretion, and mitochondrial bioenergetics. And as reviewed, altered NCX SLC8 protein expression and regulation contributes to impaired Ca²⁺ homeostasis in heart failure, arrhythmia, cerebral ischemia, hypertension, diabetes, renal Ca²⁺ reabsorption, muscle dystrophy, etc. (1178). The phylogenetic tree (Fig. 19) shows that SLC8B is on a different branch relative to SLC8A. NCLX (SLC8B1) of branch B is a mitochondrial carrier (see below).

Fig. 19. Phylogenetic tree of the SLC8 family.

SLC8A1: NCX1 (SLC8A1) is widely expressed, most strongly in cardiac muscle. It uses the electrochemical Na⁺ gradient to mediate the electrogenic counter-transport of 3 Na⁺ ions for 1 Ca²⁺ ion across the sarcolemmal membrane (1179). In cardiomyocytes, upon depolarization of the plasma membrane, Ca²⁺ influx occurs through the opening of voltage-gated L-type Ca²⁺ channels, which then triggers Ca²⁺ release from the terminal cisternae of the junctional sarcoplasmic reticulum, followed by contraction of the myocyte. The entry of Ca²⁺ with each contraction requires an equal amount of Ca²⁺ extrusion within a single heartbeat, which is achieved by NCX1 (1180). The transporter may also be directly involved in the regulation of excitation-contraction coupling.Global deletion of Slc8a1 in mice results in NCX1-null embryos that cannot survive due to a lack of spontaneous heartbeat (1181), although mice with a cardiac-specific knockout of NCX1 can live to adulthood, suggesting that NCX1 is important for the development of the heartbeat function (1182). Selective inhibition or activation of NCX1 (SLC8A1) variants in specific diseases may be of great clinical relevance. Of particular interest would be the ability to target the NCX1 isoform expressed in the heart. This is where Na⁺ and Ca²⁺ cycling is critical in every heartbeat. An NCX inhibitor, ORM-10103 (see below), was shown to have improved selectivity over existing small molecule NCX inhibitors and to abolish induced arrhythmias without affecting the L-type calcium channel (1183).

NCX1 (SLC8A1) operates in either forward or reverse mode as part of its contribution to maintaining intracellular Ca²⁺. In forward mode, NCX1 transports Ca²⁺ out of the cell, and in reverse mode, it takes up extracellular Ca²⁺. Depolarization of the plasma membrane as it may occur in excitotoxicity and/or alterations of the Na⁺ and Ca²⁺ electrochemical gradients can reverse the direction of the NCX1 exchanger (1184). Over the past decades, a variety of NCX1 blockers targeting reverse or forward NCX1 have been developed and used to study the pathophysiological roles of NCX1 in hypoxic/ischemic brain, heart, and kidney tissue (1185–1187). The aforementioned ORM-10103 inhibitor, which has been investigated for its potential use in the treatment of certain arrhythmias, inhibits both the forward and reverse modes of NCX1 (1183). The relatively specific amiloride derivative 5-(N-4-chlorobenzyl)-2’,4’-dimethylbenzamil (CB-DMB) blocker specifically inhibits the forward mode of NCX1 (1185, 1188).

Pharmacological inhibition of NCX1 (SLC8A1) can overcome apoptosis evasion and reduce cancer cell proliferation. Blocking the forward mode of the Na⁺/Ca²⁺ exchanger with CB-DMB has been shown to suppress the growth of glioblastoma cells through Ca²⁺-mediated cell death (1187). NCX1 (SLC8A1) upregulation was found to be associated with poor prognosis in patients with gastric cancer (1189) and NCX1 (SLC8A1) inhibition by CB-DMB significantly reduced cancer proliferation and induced Ca²⁺-dependent cell death in gastric cancer cells, both alone and synergistically with cisplatin treatment (1189).

Modulation of NCX1 (SLC8A1) expression in monocytes has been shown to be associated with multiple sclerosis progression (1190). NCX1 regulates Na⁺ and Ca²⁺ homeostasis in monocytes and its disruption by SLC8A1-AS1 long non-coding RNA (lncRNA; see Abbreviations and Glossary) has been reported to be associated with NCX1 dysregulation in multiple sclerosis. This leads to dysfunction of the immune regulatory network, as NCX1 levels in monocytes have been found to correlate with the proportion of circulating regulatory T cells (1190). SLC8A1-AS1 lncRNA is an antisense SLC8A1 RNA (HGNC ID: 44102) and the presence of lncRNAs is known to have regulatory effects on gene expression at multiple levels (1191). The results of this study reveal a stage-specific dysregulation of the NCX1 exchanger in monocytes during the progression of multiple sclerosis due to a disturbed ion balance in monocytes. This may influence the immune regulatory network during the course of disease pathogenesis (1190). These findings may be relevant to the identification of novel biomarkers and/or therapeutic targets in multiple sclerosis.

SLC8A2: NCX2 (SLC8A2) is a Na⁺/Ca²⁺ exchanger that is expressed almost exclusively in the brain, especially in the cerebellum. It is involved in Ca²⁺ clearance following neuronal depolarization (1192). Knocking out NCX2 in mice resulted in improved performance in several hippocampus-dependent learning and memory tasks. This result supports the idea that NCX2 plays a predominant role in Ca²⁺ clearance at pre- and postsynaptic sites in the hippocampus (1193, 1194).

SLC8A3: NCX3 (SLC8A3) is a Na⁺/Ca²⁺ exchanger that has been identified in the plasma membrane and in intracellular compartments. It is highly expressed in skeletal muscle, retina and brain. In humans, three splice variants have been identified that encode unique NCX proteins (1195). NCX3 knockout mice have been generated and their pathology appears to be largely related to defects in muscle fibers and at the neuromuscular junction (1196).

SLC8B1: NCLX (SLC8B1) is a mitochondrial Na⁺/Ca²⁺ exchanger involved in the regulation of energy metabolism. It is found in mitochondria of pancreatic β-cells, B-lymphocytes, and in cells of the ovary, adrenal gland, kidney, intestine, lung, heart, among others (1178). Mitochondria have emerged as major Ca²⁺ signaling organelles, buffering Ca²⁺ by uptake through the mitochondrial Ca²⁺ uniporter and efflux through the Na⁺/Ca²⁺ exchanger NCLX (1197). In airway smooth muscle mitochondria, NCLX has a complex role in airway smooth muscle remodeling during asthma and its inhibition has been proposed as a potential strategy for the treatment of asthma (1197). In B lymphocytes, NCLX serves as a regulator of the adaptive immune response. Specifically, while B cell receptor engagement causes rapid cytosolic Ca²⁺ increases through the ubiquitous store-operated calcium entry pathway, NCLX maintains optimal SOCE activity by expelling mitochondrial Ca²⁺ (1198). Furthermore, in the heart, enhanced NCLX-dependent mitochondrial Ca²⁺ efflux has been shown to reduce pathological remodeling in heart failure (1199).

Orphan transporter family members: N/A

SLC9 Na⁺/H⁺ exchanger family (2.A.36 and 2.A.37/Na_H_Exchanger/NhaA)

Discovery: The primary structure of a mammalian Na⁺/H⁺ exchanger NHE1 (SLC9A1) was determined following a unique genetic complementation approach of a mouse fibroblast mutant by gene transfer (1200).

Gene family members (14):
SLC9A1 (NHE1)	SLC9A5 (NHE5)	SLC9A9 (NHE9)	SLC9C2 (NHE11)
SLC9A2 (NHE2)	SLC9A6 (NHE6)	SLC9B1 (NHA1)	SLC9D1 (TMCO3)
SLC9A3 (NHE3)	SLC9A7 (NHE7)	SLC9B2 (NHA2)
SLC9A4 (NHE4)	SLC9A8 (NHE8)	SLC9C1 (Sperm-NHE)

Molecular aspects, physiological roles and links to disease

There are 13 human electroneutral cation/proton antiporters belonging to the SLC9 gene family, called Na⁺/H⁺ exchangers (NHEs) (1201–1203) and one putative K⁺/H⁺ exchanger (258).

The SLC9 family members belong to the monovalent cation:proton antiporter (CPA) superfamily. The superfamily has been structurally divided into CPA1 (TC 2.A.36) and CPA2 (TC 2.A.37) (244, 1204). While CPA1 members are predominantly Na⁺/H⁺ exchangers, CPA2 members are K⁺ efflux antiporters or cation/H⁺ exchangers (1205). All but one of the SLC9 members belong to CPA1, with the exception of SLC9D1, which belongs to CPA2. SLC9D1 was added to the SLC9 family for the putative K⁺/H⁺ exchanger previously approved as TMCO3.

SLC9 family members of the CPA1 family share the NhaA structural fold. According to the crystal structure of the E. coli NhaA Na⁺/H⁺ antiporters, they have 12 TMHs and contain inverted repeats, and studies suggest an elevator-type mechanism (see the Section 8).

The SLC9 family can be grouped into five subfamilies (Fig. 20):

Fig. 20. Phylogenetic tree of the SLC9 family.

• 1)Subfamily A, part 1 - plasma membrane exchangers: SLC9A1-5 (NHE1 to 5).
• 2)Subfamily A, part 2 - endomembrane (organellar) alkali cation exchangers (eNHEs): SLC9A6-9 (NHE6 to 9).
• 3)Subfamily B, NhaA-like transporters: SLC9B1 and SLC9B2 (NHA1 and NHA2).
• 4)Subfamily C: SLC9C1 encoding sperm NHE (NHE10), and SLC9C2/NHE11.
• 5)Subfamily D: SLC9D1 (TMCO3) encoding a putative Golgi K⁺/H⁺ exchanger.

1) Subfamily A, part 1 - plasma membrane exchangers

SLC9A1: NHE1 (SLC9A1) is an electroneutral secondary active Na⁺/H⁺ exchanger (1:1 stoichiometry) expressed on the plasma membrane of most mammalian cells, where it plays a critical role in the regulation of intracellular pH and volume homeostasis (1202). It is ubiquitously expressed with minimal basal activity in most tissues, but can be activated by intracellular H⁺. It is the principal mechanism for H⁺ extrusion in many cell types (1202, 1206). An important characteristic is the localization of NHE1 together with other ion channels/transporters in distinct microdomains of the plasma membrane, allowing it to regulate local pH_i, depending on the cell type and state, in order to influence specific cellular processes, such as Ca²⁺ homeostasis, membrane excitability, neurotransmission, etc. (1206, 1207). In epithelial tissues such as kidney and the gastrointestinal tract it is localized to the basolateral membrane. NHE1 also shows prominent expression in stomach (1208).

The NHE1 knockout (Slc9a1^−/−) mouse was shown to die at a young age due to severe epilepsy. However, it does not show any obvious gastrointestinal abnormalities (1209). Although NHE1 (SLC9A1) is highly expressed in all gastric epithelial cell types, surprisingly, the Slc9a1^-/- parietal cells had a normal appearance.

Amiloride is the first NHE1 inhibitor identified (1210, 1211). More selective and potent inhibitors have since been developed, including lipophilic amiloride derivatives (e.g., ethyl isopropoyl amiloride, EIPA) and benzoylguanidines such as cariporide (1210).

NHE1 is abundantly expressed in brain tissues and is closely related to the development of epilepsy, where it is upregulated. Acidosis after an epileptic seizure has been shown to activate NHE1, resulting in intracellular Na⁺ overload and subsequent intracellular Ca²⁺ overload via reversal of the Na⁺/Ca²⁺ exchanger (1212). Intracellular Ca²⁺ overload then activates calpain, a Ca²⁺-dependent neutral protease, and triggers several pathways involved in apoptosis. A subsequent study showed that inhibition of NHE1 expression prevents neuronal cell death in epilepsy (1212).

Activation of NHE1 is regulated by the extracellular environment and protein cofactors, including calcineurin B-homologous protein 1 (CHP1), an obligate binding partner that promotes the biosynthetic maturation of NHE1 (SLC9A1), its expression on the cell surface, and its sensitivity to pH (1213). Neurological disorders are associated with dysfunction of either protein (1214, 1215). The structure and transport mechanism of the human NHE1-CHP1 complex was determined in both the inward-facing and inhibitor (cariporide)-bound outward-facing conformations (1213). NHE1 (SLC9A1), has been shown to assemble as a symmetric homodimer, with each subunit undergoing a conformational change during cation exchange. The cryo-EM map reveals the binding site for the NHE1 inhibitor cariporide. It also illustrates how inhibitors block transport activity (1213).

Inhibition of myocardial NHE1 has been considered as a potentially attractive therapeutic approach for the treatment of myocardial salvage after ischemic insult. However, this has been found to be challenging due to the side effects arising from the ubiquitous expression of NHE1 (1216).

NHE1 promotes tumorigenesis at several levels. Acid extrusion, for example, is known to protect breast cancer cells in the acidic tumor environment by maintaining a permissive pH_i in the slightly alkaline range (pH 7.0 to 7.2) (1217). Therefore, the NHE1 inhibitor cariporide was tested for its efficacy in reducing acid extrusion (1218). However, cariporide only acutely reduced acid extrusion and minimally affected steady-state pH_i. It was found that compensation by an alternative acid extrusion mechanism, namely cariporide-induced increase in NBCn1 (SLC4A7) expression, shifted pH_i regulation to a CO₂/HCO₃^--dependent mechanism. It was concluded that cariporide had no net effect on epidermal growth factor receptor 2 (HER2/ERBB2)-driven breast cancer development (1218).

NHE1 also plays an important role in cell migration, a key step in the metastatic spread of cancer cells from a primary tumor to distant organs in the body (1219). Cell motility is controlled by cell-matrix interactions, the actomyosin cytoskeleton, and the regulation of cell volume. According to the “Osmotic Engine Model”, a cell migrating in confinement establishes a spatial gradient of ion transporters, ion channels and aquaporins in the cell membrane such that localized swelling at the leading edge and shrinkage at the trailing edge facilitate net cell movement. As part of this process, NHE1 polarizes at the leading edge of the cell, where it promotes migration. NHE1 promotes isosmotic cell swelling, consistent with its role in isosmotic volume regulation (1219). In contrast, SWELL1 (LRRC8A; a subunit of the volume-regulated anion channel VRAC)(1220, 1221) and aquaporin 4 (AQP4) are preferentially enriched at the rear of confined migrating cells and mediate local cell shrinkage via regulatory volume decrease (1219). The coordinated action of isosmotic swelling and shrinkage at the cell poles mediated by NHE1 and SWELL1, respectively, supports efficient confined migration. Kaplan-Meier survival analysis showed that patients expressing high levels of both NHE1 and SWELL1 had shorter distant metastasis-free survival (1219). Thus, these findings suggest that inhibition of NHE1 and SWELL1 might be a potential therapy to suppress breast cancer dissemination and metastasis.

NHE1 also promotes macropinocytosis in oncogene-driven (i.e., KRAS-driven) cancer cells by alkalinizing the submembranous pH (1222–1224). Oncogenic mutations drive increased proliferation through various adaptations, and macropinocytosis is one of them. It is an endocytic pathway that allows the internalization of extracellular fluid along with nutrients such as amino acids. It thus provides a nutrient scavenging pathway for Ras-driven cancer cells to support cancer cell metabolism and tumor growth (1223, 1224).

SLC9A2: NHE2 (SLC9A2) is an epithelial Na⁺/H⁺ exchanger expressed in the intestine, stomach, gallbladder BBB and kidney (1202). In the intestine, it is present in the apical membrane of epithelial cells, especially in the colonic crypt to villus axis, and regulates the intracellular pH of epithelial cells (1208). In parietal cells of the stomach, it is not directly involved in the process of acid secretion, but rather to keep the viability of the parietal cells, and it is probably also involved in the differentiation of parietal cells. In the gallbladder it is expressed in the apical membrane and facilitates Na⁺ uptake and transepithelial Na⁺ absorption. In the endothelial cells of the BBB, together with NHE1 (SLC9A1), it plays an important role in maintaining the volume and ionic composition of the cerebrospinal fluid (1225). In the kidney, it is expressed in the apical membrane of the thick ascending limb of Henle and in the distal convoluted tubules.

SLC9A3: NHE3 (SLC9A3) is present in Na⁺-absorbing tissues such as the small intestine, colon, gallbladder, renal proximal tubule, and thick and thin limbs of the loop of Henle (1202). In addition to its functional role in Na⁺ absorption by intestinal and renal epithelial cells, it removes protons from the cytoplasm to maintain an outward H⁺ gradient across the brush border membrane of enterocytes. As previously proposed (1226), this gradient is required for the transport activity of H⁺-coupled solute transporters such as PepT1 (SLC15A1), DMT1 (SLC11A2) and PAT1 (SLC36A1) in conjunction with the unstirred layer present at the extracellular brush border membrane surface. Consistent with the functional role of NHE3 (SLC9A3) in Na⁺ absorption in kidney and intestine, Slc9a3 knockout mice suffer from low blood pressure and metabolic acidosis (1227).

NHE3 (SLC9A3) is subject to acute, mid- and long-term regulation, including circadian regulation, making it a highly regulated transport protein (1202, 1208, 1228). The C-terminus of NHE3 (SLC9A3) is necessary for its short-term regulation and acts as a scaffold binding several proteins such as CHP1, ezrin, CaMKII, CaM, PLC, IRBIT, megalin, as well as the lipids PIP₂ and phosphatidyl inositol 3,4,5-trisphosphate (PIP₃) (1228).

NHE3 (SLC9A3) also binds to NHERF1 (Na⁺/H+ exchanger regulatory factor or NHERF1) (SLC9A3R1), which functions as a membrane-cytoskeletal adaptor and is required for cAMP-mediated inhibition of NHE3 in the intestine (1229, 1230). Enterotoxins produced by E. coli increase cAMP levels, which stimulates Cl^- secretion in crypt cells and inhibits NHE3 NaCl absorption in villus cells of the intestinal epithelium. cAMP-dependent inhibition of NHE3 mediated NaCl absorption contributes to diarrhea. Inhibition by cAMP has been shown to require NHERF1 (1229). The NHERF family has four members, NHERF1 (NHERF1), NHERF2 (NHERF2), NHERF3 (PDZK1), and NHERF4 (NHERF4), each with unique molecular organization and tissue distribution. As reviewed later (1231), the N-terminal domain of NHERF1 has two PDZ domains that target specific proteins, and the C-terminal region has an ezrin-binding domain. While most NHERF1-associated proteins bind to the first PDZ domain, NHE3 interacts with the second PDZ domain.

NHE3 (SLC9A3) has been shown to be involved in the pathogenesis of IBD in a subset of patients, as rare mutations in SLC9A3 have been shown to be a risk factor for very early onset IBD (1232). Consistent with this, Slc9a3 knockout mice develop inflammatory bowel disease-like dysbiosis and spontaneous colitis (1233). NHE3 has been shown to have roles in epithelial cell functions beyond canonical ion transport in the context of injury, in particular a role in colonic epithelial cell proliferation and migration during wound healing (1234).

Tenapanor is a small-molecule inhibitor of NHE3 that increases luminal sodium and water content and is approved for the treatment of irritable bowel syndrome (IBS). Tenapanor has been reported to reduce IBS-related pain by strengthening the intestinal barrier, thereby decreasing permeability to macromolecules and antigens, and reducing TRPV1-mediated pain signaling in colonic sensory neurons (74).

SLC9A4: NHE4 (SLC9A4) is highly expressed in the basolateral membranes in parietal cells of the stomach and at lower levels in kidney macula densa, thick ascending limb and distal convoluted tubule among others. The characterization of Slc9a4^−/− mice confirmed that NHE4 is required for acid secretion, because the analysis of stomach contents revealed that these mice were hypochlorhydric and displayed reduced numbers of structurally abnormal parietal cells. Silencing NHE4 (SLC9A4) results in abnormal differentiation of gastric mucosal epithelial cells (1208, 1235). Therefore, NHE4 plays an essential role in the maturation and differentiation of parietal cells.

NHE4 is also expressed in the basolateral membrane of the cells of the thick ascending limb of Henle, together with NHE1 (SLC9A1), where they contribute to intracellular pH recovery from intracellular acid load. Interestingly, Slc9a4^−/− mice show a defect in ammonia uptake in the thick ascending limb. The proposed explanation is that under normal conditions, ammonium ions (NH₄⁺) entering the luminal membrane of thick ascending limb cells generate NH₃ and H⁺ inside the cell. NH₃ then exits the basolateral membrane while H⁺ is exported through NHE4 or alternatively HCO₃⁻ is imported through NBCn1 (SLC4A7) (1236). In Slc9a4^−/− mice, thick ascending limb cells are unable to secrete excess acid load, resulting in renal ammonium secretion (1208).

SLC9A5: NHE5 (SLC9A5) is predominantly expressed in the brain, especially in the cerebellum. For normal neuronal membrane excitability, precise control of the neuronal pH milieu is essential. However, little is known about the specific role of NHE5 (SLC9A5) in the CNS. To this end, the kinetic and pharmacological properties of NHE5 were determined in a NHE-deficient CHO cell model (1237). Extracellular monovalent cations, such as H⁺ and Li⁺, but not K⁺, acted as effective competitive inhibitors of the influx of radioisotope cation ²²Na⁺ via NHE5, and the pharmacological and biochemical properties distinguish it from other members of the SLC9 family. Clearly, further work is necessary for the elucidation of the physiological role of NHE5 (SLC9A5) in neuronal cell function.

2) Subfamily A, part 2 - endomembrane (organellar) alkali cation exchangers (eNHEs)

NHE6 (SLC9A6), NHE7 (SLC9A7), NHE8 (SLC9A8) and NHE9 (SLC9A9) are members of SLC9 subfamily A, part 2, “endomembrane alkali cation (Na⁺ or K⁺)/H⁺ exchangers (eNHEs)”, which play a complex role in the regulation of the luminal pH of secretory vesicles and endosomes (1238). of these, the most closely related members NHE6 (SLC9A6), NHE7 (SLC9A7) and NHE9 (SLC9A9) form a separate clade as shown in the phylogenetic tree (Fig. 20).

NHE8 (SLC9A8) also belongs to subfamily A, part 2 (Fig. 20), although it is more distantly related to the other members and is expressed at the plasma membrane as well as in intracellular compartments such as the Golgi apparatus, ER and endosomes (1208). Its localization varies depending on the cell type and its specific function, and in the small intestine it is mainly located at the cell membrane.

The directionality of membrane transport mediated by eNHEs has been intensively discussed. For example, NHE7 was originally proposed to function as a proton leak at the Golgi, but subsequent data indicate that NHE7 functions as a proton loader. Below is a brief overview of the peculiarities of the proton leak and proton loader hypotheses.

Proton leak hypothesis: According to the proton leak hypothesis, after the vacuolar V-ATPase continuously loads the TGN lumen with protons in an ATP-dependent manner (1239), NHE7 and other eNHEs function as proton leaks, allowing the exit of luminal H⁺ in exchange for cytosolic Na⁺ or K⁺. This would allow these organelles to function as repositories for H⁺ that can extrude H⁺ when needed. However, given the similar concentrations of monovalent cations in the cytosol and Golgi (for Na⁺ it is ~16 mM in the cytosol and ~27 mM in the Golgi (1240); and for K⁺ it is 140 to 150 mM in the cytosol and ~107 mM in the Golgi (1241)), there would be little or no driving force for H⁺ export from the Golgi by countertransport coupling to these ions. Consistent with this, it has been shown that increasing the cytosolic Na⁺ concentration to mimic physiological extracellular Na⁺ concentrations, which are known to drive NHE1-mediated proton extrusion from the plasma membrane, has minimal effect on Golgi pH (1241, 1242). Nevertheless, H⁺ leak activity at the Golgi along the H⁺ ion gradient generated by the V-ATPase is still possible.

Proton loader hypothesis: According to the proton loader hypothesis, NHE7 and other eNHEs mediate luminal H⁺ loading in the Golgi by exchanging H⁺ for luminal Na⁺ and thus act in an additive manner together with the V-ATPase to induce luminal acidification (1243–1245). Thus, both an eNHE exchanger and the V-ATPase are involved in the maintenance of steady-state luminal pH.

Whether the individual eNHEs, NHE6 (SLC9A6), NHE7 (SLC9A7), NHE8 (SLC9A8) and NHE9 (SLC9A9), function as proton leaks or proton loaders may also depend on the primary localization in the endocytic pathway and their specific properties. Different eNHEs have been shown to be localized in different intracellular compartments (1246). The individual localization, proposed function as proton leak or loader of the eNHEs is provided in the summaries below.

SLC9A6: NHE6 (SLC9A6) is a ubiquitously expressed electroneutral cation (Na⁺ or K⁺)/H⁺ exchanger (1247). According to the HPA, NHE6 (SLC9A6) shows highest expression in neurons in the cerebral cortex. It has also been shown to be abundantly expressed in the CA1 area of the mouse hippocampus, and immunohistochemistry of dendritic spines revealed expression in early and recycling endosomes (1247). It is generally believed that in neurons and other cell types, NHE6 assembles as a homodimer and localizes to early and recycling endosomes where it modulates luminal acidification driven by the electrogenic vacuolar V-ATPase (1238, 1247–1249).

Deleterious mutations in NHE6 (SLC9A6) and also in NHE9 (SLC9A9) (see below) have been reported to be associated with neurodevelopmental disorders. Mutations in SLC9A6 cause Christianson syndrome, an X-linked disorder associated with intellectual disability, microcephaly, seizures, ataxia, and absence of speech (1250). SLC9A6 loss-of-function mutations are common in patients with X-linked intellectual disability (1250–1253). They have been shown to lead to overacidification of endosomes, suggesting that SLC9A6 acts as a proton leak (1249). Based on subsequence studies in mice, structural and functional experimental models of learning were found to be impaired in Slc9a6 knockout hippocampal neurons (1254). Impaired hippocampal plasticity associated with loss of recycling endosomal NHE6 (SLC9A6) was shown to be ameliorated by the neurotrophic tyrosine receptor kinase B (TrkB) agonist 7,8-dihydroxyflavone (1254). TrkB activation may thus serve as a potential clinical intervention to ameliorate cognitive deficits in Christianson syndrome and other neurodegenerative disorders (1254).

Mutations of NHE6 (SLC9A6) have also been associated with Parkinsonism and NHE6 has been shown to be downregulated in the substantia nigra in patients with Parkinson disease (1255). Both endosomal NHE6 and NHE9 were found to be dysregulated in Parkinson disease (1255).

SLC9A7: NHE7 (SLC9A7) is ubiquitously expressed according to the HPA, with the highest expression in the brain in neurons and glial cells as well as in dendritic cells of the immune system. NHE7 is located within the trans-Golgi complex and accumulates predominantly in TGN and post-Golgi vesicles (1256), where it is thought to regulate luminal pH as well as glycosylation of exported cargoes (1257).

As mentioned above, the directionality of NHE7 (SLC9A7) was originally a matter of debate as it was proposed to function as a proton leak pathway (1239) while later additional experimental evidence supported the concept that it acts as a proton loader, working together with the V-ATPase to establish steady-state luminal acidification (1243–1245). Importantly, NHE7 was shown to transport Li⁺ and Na⁺, but not K⁺, and to be activated by cytosolic H⁺, leading to the conclusion that NHE7 acts as a proton loading transporter rather than a proton leak (1245). In addition, NHE7 knockdown in pancreatic ductal adenocarcinoma cells was shown to lead to alkalinization of the Golgi (1243). It has been proposed that NHE7-mediated H⁺ transport into the trans-Golgi network (TGN) is involved in the assembly and secretion of post-Golgi vesicles whose cargo is destined for the extracellular space (1243). In this way, acidification of the TGN may not only represent a repository for cytoplasmic acid, but also the means to expel H⁺ extracellularly, which in the case of tumor growth would help to maintain the slightly decreased optimal extracellular pH of about 6.8.

In further support of the proton loader hypothesis is the finding on the role of NHE7 (SLC9A7) in enhancing the maturation of macropinosomes in hepatocellular carcinoma to potentiate the uptake of small extracellular vesicles (1258). Small extracellular vesicles, which include exosomes, play a critical role in cancer development and progression by facilitating cell-to-cell communication (1259). Upon activation of macropinocytosis triggered by oncogene activation, macropinosomes facilitate the internalization of small extracellular vesicles. The study showed that NHE7 alkalizes intracellular pH and acidifies endosomal pH, leading to the maturation of macropinosomes (1258). Inhibition of NHE7 in mouse tumors delayed tumor development and suppressed lung metastasis. Clinically, NHE7 expression was upregulated and associated with poor prognosis in hepatocellular carcinoma.

In another study, it has been proposed that a unique genetic variant of SLC9A7 (L515F) converts the NHE7 exchanger into a proton leak leading to luminal alkalization (1257). As mentioned above, genetic evidence highlights the critical role of eNHEs in neuronal development and plasticity, as their dysfunction is associated with neurological disorders (1238, 1257). A unique genetic variant (L515F) in NHE7 (SLC9A7) has been identified that causes non-syndromic X-linked intellectual disability, a rare disorder characterized by macrocephaly, minimal speech, muscle weakness, and hypotonia (1257). In line with the proton leak hypothesis, the authors proposed a model in which the mutation converts the exchanger into a hyperactive proton leak responsible for luminal alkalization (1257). Studies in transfected cells showed that expression of the SLC9A7 mutant resulted in decreased acidification of the TGN and associated compartments. This, in turn, is expected to result in impaired post-translational glycosylation of proteins required for normal development and thus impaired processing of cargo by these organelles along the secretory and endocytic pathways (1238, 1257). However, further investigation is required to determine the precise molecular and cellular mechanisms underlying this disorder and how the finding of this NHE7 (SLC9A7) mutation can be interpreted in light of the now widely accepted proton loader hypothesis describing the proper function of NHE7.

SLC9A8: NHE8 (SLC9A8) is ubiquitously expressed according to the HPA and findings from different groups (1246, 1260). Highest expression was found in kidney, muscle, testis, stomach, liver and bone marrow. NHE8 (SLC9A8) was originally cloned from a mouse kidney library and localized to the renal proximal tubules (1260). Subsequent immunolocalization identified NHE8 (SLC9A8) on the apical membrane of rat kidney proximal tubule segments S1 to S3 and also in coated pits or subapical tubules, suggesting involvement in endocytosis (1261). Based on expression studies in yeast, NHE8 was described as an organellar exchanger that transports Na⁺ and K⁺ ions in exchange for H⁺, and based on expression studies in COS-7 cells, it was found to be present in the Golgi apparatus, specifically in the mid- to trans-Golgi (1246).

In the stomach, NHE8 (SLC9A8) has been detected in the apical membrane of surface mucus cells (1262). In addition, it has been reported that Slc9a8^-/- mice have a reduced gastric mucosal surface pH and that in parallel, the expression of the chloride anion exchanger DRA (SLC26A3) was significantly reduced in the gastric surface mucosal cells, leading to a defect in bicarbonate secretion (1262). Based on these findings, it has been proposed that NHE8 (SLC9A8) plays a protective role in the stomach by contributing to the protection of the gastric mucosa and thus to the protection against gastric ulcer and gastric cancer (1262).

NHE8 (SLC9A8) was also found to be highly expressed in human and mouse ocular tissues, and loss-of-function of NHE8 in mice resulted in decreased tear volume, increased corneal staining, and increased tumor necrosis factor alpha (TNF-α) expression in the ocular surface (1263). The results suggest that NHE8 plays an important role in ocular surface protection by participating in tear production. In the colon, NHE8 deficiency has been shown to lead to altered mucus layer formation, which results in undesired increased adhesion of unwanted bacteria (1208).

In the rat intestine, immunohistochemistry indicated that NHE8 is expressed in the brush border membrane (1264). NHE8 has also been shown to function as an acid extrusion mechanism in the apical membrane of enterocytes. In addition, it has been proposed that it helps regulate the pH and fluidity of the crypt lumen and thus has important functions in mucus release (1265). Consistent with its role in mucosal protection, NHE8 (SLC9A8) predominates in secretory cells while NHE2 (SLC9A2) predominates in absorptive cells (1208).

NHE8 has also been reported to play a key role in controlling protein trafficking and late endosomal morphology in mammalian cells, but depletion of NHE8, despite its function as a Na⁺ (K⁺)/H⁺ antiporter, did not affect the overall pH within dense multivesicular bodies (1266).

NHE8 (SLC9A8) deficiency at the Golgi apparatus in mouse germ cells has been shown to result in defective acrosome formation and male infertility (1267). The luminal pH of acrosomes is ~5.3 and thus more acidic than that of the Golgi (pH ~5.9-6.3). NHE8 may play a critical role in the fusion of Golgi-derived vesicles during formation of the acrosomal cap, possibly acting as a proton leak to fine-tune vesicle acidification. NHE8 (SLC9A8) has therefore been proposed as a candidate drug target for male contraception (1267).

SLC9A9: NHE9 (SLC9A9) is an endosomal cation (Na⁺ or K⁺)/H⁺ exchanger that, analogous to NHE6 (SLC9A6), counteracts acidity generated by the vacuolar V-ATPase, thereby limiting luminal acidification. According to the HPA, it is ubiquitously expressed, with the highest levels found in the spinal cord, brain areas and lymphoid tissues, but also at significant levels in kidney, liver, lung, endocrine tissues, adipose tissue, male and female tissues and at the cellular level in microglia, oligodendrocytes, astrocytes and inhibitory neurons.

NHE9 (SLC9A9) has been shown to regulate the luminal pH of late and recycling endosomes (1268, 1269).

In a study focusing on the role of NHE9 (SLC9A9) in the internalization of exosomes via the endocytotic pathway, where endosomal pH is critical for exosome internalization, NHE9 was found to be located on the endosomal membrane. It has been proposed that NHE9 modulates endosomal pH by transporting H⁺ out of endosomes in exchange for sodium or potassium ions, thereby functioning according to the proton leak hypothesis (1270).

Genetic variants in NHE9 (SLC9A9) are associated with attention deficit hyperactivity disorder (1269, 1271–1273) and autism spectrum disorder with epilepsy (1274).

In addition, NHE9 (SLC9A9) expression is upregulated in glioblastoma, one of the most aggressive forms of brain cancer (1268). The study demonstrates that NHE9-mediated proton leak from endosomes drives oncogenic signaling in glioblastoma, particularly through epidermal growth factor receptor (EGFR) signaling, a key driver of tumor progression. While receptor tyrosine kinase inhibitors have shown limited clinically efficacy and EGFR persists at the plasma membrane to promote tumor growth and invasiveness, the study highlights the importance of endolysosomal pH in receptor sorting and turnover. Consequently, NHE9 which mediates inside-out control of oncogenic signaling, has been proposed as a promising drug target for anticancer therapy (1268).

Of the known endosomal pH regulators, only NHE9 (SLC9A9) has been genetically linked to severe COVID-19 risk (1275). Endocytosis is a prominent mechanism for SARS-CoV-2 entry into host cells such as lung alveolar cells. It has been shown that limiting the acidification of early endosomes by increasing the expression of NHE9 results in reduced infectivity of SARS-CoV-2 spike-bearing virus in host cells via NHE9 acting as a proton leak pathway specifically on endosomes (1276). In addition, it was shown that the early endosome membrane lipid phosphatidylinositol 3-phosphate (PI3P) acts as a link between luminal pH changes and early endosome trafficking, such that in cells with high NHE9 expression, PI3P persists longer on early endosomes and, through NHE9-mediated alkalinization, inhibits early endosome perinuclear movement, ultimately leading to early endosomes falling off microtubules and impairing viral cargo delivery to late endosomes (1276). NHE9 thus offers a unique opportunity as a viable therapeutic target to prevent SARS-CoV-2 entry into host cells.

The cryo-EM structure of Equus caballus NHE9 (SLC9A9) is reported at 3.2 Å resolution (253). The 13-TMH architecture and ion-binding site of NHE9 were found to be remarkably similar to those of the distantly related bacterial Na⁺/H⁺ antiporters. The study reports on the conserved architecture of the NHE ion-binding sites, the elevator-like structural transitions, the functional implications of autism disease mutations, and the role of phosphoinositide lipids in promoting homodimerization (253). The structure of NHE9 is expected to serve as a suitable template for homology modeling of the structures of other NHE paralogs.

Na⁺/H⁺ exchangers function exclusively as homodimers, with the architecture of the dimerization interface varying among orthologs and paralogs. This is especially true for the organellar isoforms such as NHE6, NHE7 and NHE9, which contain β-hairpin domains and interact with lipids like phosphatidylinositol-4,5-bisphosphate (PIP₂) at their dimerization interfaces (254). Cryo-EM studies have resolved the structures of E. coli NhaA and Equus caballus NHE9 in complex with cardiolipin and phosphatidylinositol-3,5-bisphosphate (Ptdlns(3,5)P₂), showing lipid binding at the dimer interface (254). Notably, the endosome-specific lipid PtdIns(3,5)P₂ stabilizes the NHE9 homodimer and enhances its transport activity, confining NHE9 function to endosomes, where the lipid is present, and preventing activity at the plasma membrane, where it is absent (254). These findings highlight how specific lipids regulate Na⁺/H⁺ exchanger activity by promoting dimer stabilizing in response to membrane identity and trafficking cues.

3) Subfamily B, NhaA-like transporters

SLC9B1 and SLC9B2 – Orphan transporters (SLC9B2 characteristics partially solved): While NHE1 (SLC9A1) through NHE9 (SLC9A9) are well known for their roles in human physiology and disease, much less is known about the two members of the SLC9B subfamily, NHA1 (SLC9B1) and NHA2 (SLC9B2), which share a higher sequence similarity to bacterial NhaA proteins compared to the NHEs than the SLC9A paralogs (1202, 1277, 1278).

NHA1 (SLC9B1) is mainly expressed in the testes and is important for sperm motility and fertility in males, but is not linked to human disease (1202).

NHA2 (SLC9B2) is expressed ubiquitously and was subsequently shown to be involved in the regulation of blood pressure and electrolyte balance in the kidney, in insulin secretion and in systemic glucose homeostasis. NHA2 has also been implicated in the pathogenesis of polycystic kidney disease (1277, 1279). Loss of SLC9B2 (NHA2) has been shown to result in decreased insulin secretion (1280). Therefore, to increase insulin secretion, an NHA2 activator would have to be developed. But loss of NHA2 also lowers blood pressure, so NHA2 inhibitors would be beneficial to lower blood pressure (1277).

In the testis, NHA2 is important for sperm motility and male fertility, as Slc9b2 knockout in mice significantly reduced sperm motility and led to a lower pregnancy rate (1281).

To date, no transport assay has been described for mammalian NHA1 (SLC9B1), and thus its ion preference and transport kinetics remain unknown. In contrast, the transport properties of NHA2 (SLC9B2) have been partially characterized. Human NHA2 was shown to rescue the Na⁺/H⁺ exchanger-deficient phenotype in yeast, restoring salt resistance at acidic extracellular pH (1282). Studies using proteoliposome-reconstituted NHA2 suggest it functions as an electroneutral Na⁺/H⁺ exchanger (1283), and it has been shown to mediate Li⁺/ H⁺ exchange in the kidney (1284). Additionally, NHA2 shares significant sequence similarity with the Na⁺, Li⁺/H⁺ antiporter NhaA of E. coli (Ec-NhaA), which plays a key role in pH and Na⁺ homeostasis in enterobacteria (1285, 1286).

4) Subfamily C

SLC9C1: Sperm-NHE (SLC9C1) is a sperm-specific Na⁺/H⁺ exchanger involved regulating intracellular pH of spermatozoa (1202, 1287). It is essential for sperm motility and male fertility, and mutations in SLC9C1 have been linked to human asthenozoospermia (1288).

Activation of SLC9C1 alkalinizes the sperm cytoplasm, stimulates soluble adenylyl cyclase (sAC) to raise cAMP, and thereby promotes CatSper Ca²⁺ channel opening, motility, and fertilization competence (3551, 3552.

Cryo-EM studies revealed that, unlike typical NHE family members, SLC9C1 combines three distinct functional modules within a single polypeptide: a canonical NHE transport module, consisting of 13 TMHs that mediate Na⁺/H⁺ exchange; an S1–S4 voltage-sensing domain (VSD) located C-terminal to the transport module; and a distal cyclic-nucleotide binding domain (CNBD) (3553, 3554). This unprecedented combination of an SLC transport module, a VSD, and a CNBD within a single molecule exemplifies a previously unrecognized mode of electromechanically regulated transport and offers a new perspective on how transporters can be gated and integrated into signaling networks.

Sperm-NHE has been detected along the entire length of the flagellum (3555). Its specific localization and function provide a compelling basis for potential therapeutic strategies, either to enhance sperm motility in asthenozoospermic men or to develop non-hormonal contraceptives (3551).

SLC9C2 - Orphan transporter: NHE11 (SLC9C2) is a sperm-specific transporter that is important for male fertility. It localizes to the sperm head and likely resides in the plasma membrane overlaying the acrosome in mature sperm (1289). The precise physiological role of NHE11 (SLC9C2) has yet to be elucidated but its predicted unique localization and functional domains suggests that it modulates intracellular pH of the sperm head in response to changes in membrane potential and cyclic nucleotide concentrations that occur as part of the of sperm capacitation events. It might be an attractive target for male contraceptive drugs due to its exclusive sperm-specific expression (1289).

5) Subfamily D

SLC9D1 -Orphan transporter: SLC9D1 (TMCO3) encodes a putative K⁺/H⁺ exchanger localized to the Golgi apparatus. It was added to the SLC9 family based on its membership in the Monovalent Cation:Proton Antiporter (CPA) superfamily, specifically the CPA2 family (TC 2.A.37; Pfam: Na_H_Exchanger) (7). TMCO3 is predicted to contain an N-terminal secretory signal peptide and to functions as a K⁺/H⁺ exchanger at the Golgi apparatus, where it has been shown to play a key role in longitudinal growth in both mice and humans (258).

Mutations in the SLC9D1 gene have been shown to be associated with the cornea guttata and anterior polar cataract, a rare autosomal dominant inherited eye disease (1290). In addition, SLC9D1 has been reported to be a prognostic marker that is highly expressed in hepatocellular carcinoma and correlates with poor prognosis (1291).

The cryo-EM structure of the SLC9D1 homolog in E. coli K⁺/H⁺ exchanger KefC (glutathione-gated K⁺ efflux transporter) was subsequently determined (259). KefC forms a homodimer similar to the inward-facing conformation of the Na⁺/H⁺ antiporter NapA. The KefC monomer consists of 13 TMHs with an extracellular N-terminus and an intracellular C-terminus. The KefC structure was more similar to the structure of the NapA bacterial Na⁺/H⁺ exchanger of the CPA2 family (244) than to NhaA from E. coli (260), which represents the CPA1 family.

The precise functional properties and physiological roles of SLC9D1 (TMCO3) are still subject to further investigation.

Orphan transporter family members (4)

SLC9B1 (NHA1), SLC9B2 (NHA2), SLC9C2 (NHE11), SLC9D1 (TMCO3)

HGNC update

SLC9D1 is the new symbol for the gene previously approved as TMCO3.

SLC10 Sodium bile salt cotransport family (2.A.28/SBF/NhaA)

Discovery: The founding member of the SLC10 family is the rat liver Na⁺/bile acid cotransporter NTCP (SLC10A1), which was identified by expression cloning with Xenopus oocytes (141). Subsequently, the hamster Na⁺-dependent ileal bile acid transporter IBAT (SLC10A2) was identified by expression cloning in cultured cells (164).

Gene family members (7):
SLC10A1 (NTCP)	SLC10A3 (P3)	SLC10A5 (P5)	SLC10A7 (P7/C4orf13)
SLC10A2 (ASBT)	SLC10A4 (P4)	SLC10A6 (SOAT)

Molecular aspects, physiological roles and links to disease

The SLC10 family consists of seven members, and three of them, NTCP (SLC10A1), ASBT (SLC10A2) and SOAT (SLC10A6), mediate Na⁺-dependent substrate transport (1292). Together with the orphan transporter SLC10A4, they form separate branches on the phylogenetic tree (see Fig. 21).

Fig. 21. Phylogenetic tree of the SLC10 family.

Members of the SLC10 family belong to the Bile Acid:Na⁺ Symporter (BASS) family (TC 2.A.28) and, similar to the SLC9 family, share the NhaA structural fold. SLC10 transporters mediate the translocation of a wide range of molecules across membranes, including bile acids in humans and small metabolites in plants. The substrate specificity of several human SLC10 family members remains undetermined (1293).

The structure of the human sodium taurocholate co-transporting polypeptide (NTCP; SLC10A1) has been resolved (255), revealing a 9-TMH topology (see Section 8 for further details). Additionally, the crystal structure of a BASS transporter from Neisseria meningitidis in complex in complex with pantoate has been determined, providing further insight into Na⁺ coupling, substrate binding, and translocation mechanisms (1294). SLC10 bile acid transporters are proposed to operate via an elevator-type transport mechanism (257, 1294).

SLC10A1, SLC10A2: Enterohepatic bile acid transport involves the apical ASBT (SLC10A2)-mediated uptake of bile acids in the ileum, the basolateral exit into the portal vein via OSTα-OSTβ (encoded by SLC51A and SLC51B; see SLC51 family), the hepatic sinusoidal uptake via NTCP (SLC10A1), and exit via the canalicular bile salt export pump BSEP (ABCB11) pump, with bile acid delivery back into the intestine (Fig. 33). This process constitutes the enterohepatic bile acid circulation, whose aim is to maintain low plasma levels of bile acids (1293, 1295).

NTCP also transports steroids and xenobiotics, including 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins) (1296).

Regulation of bile acid transporters by nuclear receptors is critical for maintaining bile acid homeostasis, and defects in these transporters or their regulation can lead to hepatobiliary pathologies such as cholestasis, gallstone formation and liver cancer (1297).

The regulation of human apical ASBT (SLC10A2) was revealed by determining site-specific phosphorylation occupancy using targeted mass spectrometry (1298). The protein was found to be phosphorylated at multiple sites such as T330, S334, and S335, where S335 was the predominant phosphorylation site. The critical involvement of PKC in the regulation of ASBT (SLC10A2) activity through phosphorylation at S335 was demonstrated, and there was a proportional relationship between the phosphorylation level of S335 and ASBT bile acid uptake activity.

SLC10A3 - Orphan transporter: P3 (SLC10A3) is widely expressed in human tissues but its function and substrate specificity are unknown (1293).

SLC10A4 - Orphan transporter: P4 (SLC10A4) is highly expressed in the pituitary gland and midbrain as suggested by the HPA and other studies (1293). In rat midbrain, Slc10a4 is specifically expressed in dopaminergic neurons of the substantia nigra (1299). Moreover, microarray analyses of the developing human ventral mesencephalon revealed increased Slc10a4 expression during development (1300), suggesting a potential role in both developing and mature dopaminergic neurons of the substantia nigra. Notably, P4 (SLC10A4) exhibits vesicular localization (1301), implicating a possible function in neurotransmitter storage or exocytosis (1293).

SLC10A5 - Orphan transporter: P5 (SLC10A5) is highly expressed at the mRNA level, but not detected at the protein level, in liver and to a lesser extent in intestine, kidney and brain according to the HPA. Human SLC10A5 deficiency was shown to cause hypercholanemia and SLC10A5 was shown to be involved in bile acid transport in hepatocytes (1302). SLC10A5 was localized to the sinusoidal membrane, and its deficiency inhibited bile acid uptake, indicating that SLC10A5 functions as an influx transporter. The study highlights the importance of SLC10A5 in bile acid metabolism and in the pathogenesis of hypercholanemia.

SLC10A6: SOAT (SLC10A6) is a Na⁺-dependent organic anion transporter that is highly expressed in skin and esophagus and primarily transports sulfated steroids (1293, 1303). As reviewed (1304), SOAT (SLC10A6) transports 3ʹ- and 17ʹ-monosulfated steroid hormones such as estrone sulfate and dehydroepiandrosterone sulfate to specific target cells as precursors for the synthesis of estrogens and androgens. However, further studies are still needed to clarify the precise physiological function of this transporter and its potential as a drug target, e.g., for endocrine-based therapy of steroid-responsive diseases such as hormone-dependent breast cancer (1304).

SLC10A7 - Orphan transporter: P7 (SLC10A7) is ubiquitously expressed. Its transport function is unknown. A recent study revealed an important role of P7 in bone mineralization and transport of glycoproteins to the extracellular matrix, and P7 was localized to the Golgi (1305). Furthermore, SLC10A7 mutations have been identified as responsible for a new congenital disorder of glycosylation (CDG) (1306). Subsequently, it was shown that SLC10A7 regulates O-N-acetylgalactosamine (O-GalNAc) glycosylation and Ca²⁺ homeostasis in the secretory pathway (1307). Loss of SLC10A7 was shown to lead to severe Golgi O-GalNAc glycosylation defects in SLC10A7-CDG patient fibroblasts. O-GalNAc glycosylation is predominant for TGN46, a membrane protein located in the trans-Golgi network (TGN) that cycles between the TGN and the plasma membrane and is involved in the sorting and transport of secretory proteins (1308). TGN46 also serves as a marker protein for the TGN.

Orphan transporter family members (4)

SLC10A3 (P3), SLC10A4 (P4), SLC10A5 (P5), SLC10A7 (P7)

SLC11 Proton-coupled metal ion transporter family (2.A.55.2/Nramp/APC)

Discovery: The rat intestinal H⁺-coupled divalent metal transporter DMT1 (Slc11a2) (formerly known as DCT1 or NRAMP2) was identified by expression cloning with Xenopus oocytes (136). It represents the major iron importer in enterocytes and erythroblasts. Around the same time a positional cloning strategy led to the identification of the same gene as being defective in an inbred mouse strain (mk) with microcytic anemia (129). Although its paralog NRAMP1 (Slc11a1) was already discovered in 1993 by positional cloning of a mouse gene affecting the capacity with which macrophages can fight bacterial pathogens (1309), this information did not lead to the identification of SLC11A2 as a divalent metal ion transporter. The precise functional properties of NRAMP1 still remained unclear (see the SLC11A1 description below).

Gene family members (2):
SLC11A1 (NRAMP1)	SLC11A2 (DMT1/NRAMP2)

Molecular aspects, physiological roles and links to disease

NRAMP1 (SLC11A1) and DMT1 (SLC11A2) belong to the Metal Ion (Mn²⁺-ion) Transporter (Nramp) family (TC 2.A.55), which is part of the APC superfamily. NRAMPs have 11 or 12 TMHs and contain the LeuT fold (see “Structural insights and pharmaceutical perspectives” below and Section 8). The name “natural resistance-associated” derives from the role of NRAMP1 in resistance to intracellular bacterial pathogens (1310).

SLC11A1: NRAMP1 (SLC11A1) plays a critical role in macrophages and neutrophils in the defense against microbial infections (1311). According to the HPA, NRAMP1 is highly expressed in monocytes, macrophages, Hofbauer cells and Kupffer cells. The NRAMP1 protein has endocytic targeting signals at the N- and C-termini (1312, 1313). The subcellular expression of NRAMP1 has been shown to be limited to late endosomes, lysosomes, and phagosomes (1314). It colocalizes with lysosomal-associated membrane protein 1 (LAMP1), a marker of late endosomes/lysosomes of resting macrophages (1315). During the phagosome maturation process, both NRAMP1 (SLC11A1) and LAMP1 proteins are recruited to the phagosomal membrane with similar kinetics (218, 1315). One of the primary functions of NRAMP1 (SLC11A1) is to contribute to antimicrobial function of macrophages by expelling essential divalent metal ions such as Mn²⁺ and Fe²⁺ from the phagolysosome via H⁺-coupled metal ion cotransport to the cytosol (216, 1316, 1317). Ferroportin (SLC40A1) then allows Fe²⁺ to exit the macrophage. This limits the ability of the mycobacteria to produce and activate enzymes such as superoxide dismutase, which prevents the propagation of the ingested microorganisms (1318). By contrast, an increased concentration of metal ions in the phagosome caused by a defective NRAMP1 transporter can promote mycobacterial growth and render the invaded organism susceptible to the pathogen (1310, 1319). NRAMP1 is therefore required for natural resistance to intracellular bacterial infection.

The structural basis for metal ion transport by human NRAMP1 (SLC11A1) has been reported as part of a cryo-EM structural analysis of the human SLC11 proteins NRAMP1 and DMT1 (216). Based on transport studies in proteoliposomes, the apparent K_m values of NRAMP1 and DMT1 for Mn²⁺ transport were found to be 5 μM and 36 μM, respectively, and those for Fe²⁺ transport were found to be 1.4 μM and 2.5 μM, respectively (216). Both proteins were found to catalyze selective divalent metal ion transport coupled to the cotransport of H⁺ by an alternating access mechanism, but also mediated uncoupled H⁺ flux. This work provides the basis for the structure-based design of specific modulators that interfere with the function of NRAMP1 and DMT1 (216).

In addition to its expression in macrophages and polymorphonuclear leukocytes, NRAMP1 has also been reported to be expressed in late endosomes and lysosomes of CD11c⁺ myeloid dendritic cells, where it exerts pleiotropic effects on cytokine transcription, MHC class II molecule expression, and processing of protein antigens for presentation to T cells, highlighting its importance in regulating susceptibility to both infectious and autoimmune diseases (1320).

NRAMP1 plays a critical role in the pathogenesis of tuberculosis, a disease caused by Mycobacterium tuberculosis that primarily affects the lungs. In patients with tuberculosis, both serum iron levels and NRAMP1 expression are reduced compared to healthy individuals, leading to the hypothesis that M. tuberculosis lowers iron availability by downregulating NRAMP1 expression (1321). This is consistent with the primary function of NRAMP1 of exporting divalent metals such as Fe²⁺ and Mn²⁺ from phagosomes, depriving intracellular pathogens of essential nutrients. The reduced expression of functional NRAMP1 in macrophages is thought to be induced by M. tuberculosis to facilitate its active replication within alveolar macrophages (1321). Additionally, the bacterium increases the expression of the transferrin receptor to enhance acquisition of iron (1321). However, it remains unclear whether NRAMP1 downregulation is the cause of reduced serum iron levels in tuberculosis, or whether this is mainly due to increased transferrin receptor expression or other contributing factors.

Low NRAMP1 expression may predispose individuals to tuberculosis by impairing phagolysosomal function in pulmonary macrophages. Supporting this idea, several studies have identified associations between SLC11A1 polymorphisms (e.g., D543N) and increased susceptibility to tuberculosis (1321–1324). These findings further reinforce the role of NRAMP1 in host defense and the pathogenesis of tuberculosis.

Several studies have also implicated NRAMP1 in human susceptibility to autoimmune diseases (1313), including rheumatoid arthritis (1325) and type 1 diabetes (T1D) (1326). In dendritic cell, NRAMP1 influences cytokine production, MHC class II expression and T cell antigen processing, functions consistent with its role in both resistance to infectious diseases and susceptibility to autoimmunity. Interestingly, the D543N polymorphism has been associated with a protective effect against rheumatoid arthritis (1327), while the same variant has been reported to be a risk factor for tuberculosis (see above).

SLC11A2: DMT1 (SLC11A2), previously known as NRAMP2, is a H⁺-coupled divalent metal ion transporter. It has a vital role in iron homeostasis by mediating iron uptake in the intestine and recovering iron from recycling endosomes in tissues after transferrin-receptor mediated endocytosis. In the brush border membrane of duodenal enterocytes, it facilitates dietary absorption of non-heme ferrous iron (Fe²⁺) (Fig. 22). As indicated in Fig. 22, it also transports other divalent metal ions (136). In other cells in the body, including erythroid precursors, following transferrin receptor-dependent endocytosis it mediates iron uptake from acidified endosomes to the cytoplasm. The cryo-EM structural basis of divalent metal ion transport by DMT1 has been reported as highlighted above (see description of NRAMP1/SLC11A1).

Fig. 22. Intestinal absorption of dietary non-heme iron, zinc, and manganese, and excretion of zinc and manganese.

Hephaestin, multicopper oxidase (accepts 1 electron of Fe²⁺ to form Fe³⁺); TF, transferrin; ZNT5B, splice variant of ZnT5 (SLC30A5).

Four different DMT1 (SLC11A2) isoforms exist due to alternative transcription initiation sites and alternative RNA splicing at the 3ʹ end. The 1A and 1B isoforms differ in the N-terminal region. Alternative splicing in the 3ʹ end is responsible for two additional isoforms with different C-terminal and 3ʹ UTR regions, one of which contains an iron responsive element (IRE) (1328). These DMT1 isoforms are tissue specific and show functional differences in protein trafficking that account for their distinct roles in iron homeostasis. While DMT1-IRE mRNA is predominantly expressed in the duodenum, DMT1-nonIRE mRNA has a broader distribution, although it is particularly important in erythroid precursors as part of the transferrin receptor-mediated iron uptake pathway (218). The interaction of the iron regulatory protein IRP1 with the 3’-DMT1-IRE protects the transcript from endonucleolytic cleavage (1329). Thus, in enterocytes, cellular accumulation of iron leads to dissociation of IRP1 from the DMT1-IRE, resulting in its mRNA degradation, thereby preventing excessive intestinal iron absorption.

The iron-responsive element of DMT1-IRE was also shown to control Notch-mediated cell fates (1330). The study demonstrated that the DMT1 (SLC11A2) splice variants with and without the IRE serve as switches that control Notch-mediated cell fate decisions such as cell renewal and differentiation in healthy mammalian cells and tissues as well as in cancer cells.

The key role of SLC11A2 in iron metabolism is highlighted in the microcytic anemia (mk) mice and Belgrade rats, which have severe defects in intestinal iron absorption and erythroid iron utilization (129). A unique mutation in Slc11a2, G185R, has occurred spontaneously in mk (microcytic anemia) mice and Belgrade rats and is responsible for this phenotype. Hypochromic microcytic anemia due to a mutation in SLC11A2, G75R (analogous to rodent DMT1 G185R), has been shown to be the cause of a very rare disorder, hypochromic microcytic anemia with iron overload (1331). Injection of iron dextran into the DMT1 G185R Belgrade rat ameliorated its anemic state but led to hepatic iron deposition via DMT1-independent iron acquisition at high serum iron levels (1332), likely via the divalent metal ion transporter ZIP14 (SLC39A14). The iron accumulation observed in G75R patients was also likely caused by ZIP14 due to excessive transfusion and oral iron supplementation, the gold standard treatments for general anemic patients (1331).

DMT1 (SLC11A2) was shown to be prominently expressed in rat choroid plexus (136), with intracellular immunoreactivity in choroid plexus epithelial cells of the lateral ventricle (1333) (see Fig. 11), whereas no expression was detected in brain capillary endothelial cells. Also NRAMP1 (SLC11A1) appears to be expressed in choroid plexus according to the HPA. DMT1 may be involved in the transport of iron transferrin across the choroid plexus into the cerebrospinal fluid and further into the brain interstitium. However, as previously highlighted (1333), iron transport into the CSF is thought to be quantitatively too small to account for the amount found in the brain (1334), and transport of solutes such as iron from the ventricular CSF into the brain only occurs where there is close proximity to the corresponding ventricular system (1335).

As reviewed elsewhere (1336), there is also increasing evidence for the involvement of DMT1 in both physiological ageing and neurodegeneration and neuroinflammation, as seen in Parkinson disease, ischemia and Alzheimer disease.

Structural insights and pharmaceutical perspectives

The elucidation of the ScaDMT structure of Staphylococcus capitis (1337) was the first breakthrough in the structural characterization of the SLC11/NRAMP family, followed by studies of the Deinococcus radiodurans (Dra) Nramp homolog (1338, 1339) and the bacterial Eremococcus coleocola (EcoDMT) homolog (1340). These studies revealed a LeuT structural fold, characterized by a modular organization of pairs of structurally related elements oriented in opposite directions in the membrane (1341). For this purpose, the first transmembrane segment of each five-helix repeat is unwound in the center, and residues in this region constitute the substrate binding site. Molecular dynamics simulations and site-directed mutagenesis revealed a novel H⁺ coupling mechanism that differs from that of other H⁺ transporters (1342).

The mechanistic basis of the inhibition of DMT1 was subsequently elucidated for bis-isothiourea substituted compounds (1343). The characterization of DMT1 inhibitors have provided first detailed insights into the pharmacology of a human iron transport protein (1344). A series of alternative DMT1 inhibitors, XEN601 and XEN602, have also been tested by Xenon Pharmaceuticals for the treatment of inherited forms of iron overload (1345, 1346). However, while the inhibitors have proven to be valuable research tools, the company is no longer attempting to develop them for the treatment of iron overload disorders due to severe toxic side effects. New therapeutic alternatives are therefore needed.

Orphan transporter family members: N/A

SLC12 Electroneutral cation-coupled Cl⁻ cotransporter family (2.A.30/AA_permease and SLC12/APC)

Discovery: The winter flounder urinary bladder thiazide-sensitive Na⁺/Cl⁻ cotransporter (137) is the first identified member of the SLC12 family. Shortly thereafter, the Na⁺-K⁺-Cl⁻ cotransporter (NKCC) was identified from a shark rectal gland library by screening with monoclonal antibodies against the native shark cotransporter (1347).

Gene family members (9):
SLC12A1 (NKCC2)	SLC12A4 (KCC1)	SLC12A7 (KCC4)
SLC12A2 (NKCC1)	SLC12A5 (KCC2)	SLC12A8 (CCC9)
SLC12A3 (NCC)	SLC12A6 (KCC3)	SLC12A9 (CIP)

Molecular aspects, physiological roles and links to disease

The SLC12 family belongs to the cation-chloride cotransporter (CCC; TC 2.A.30; Pfam “AA_permease”) family which in turn is a part of the APC superfamily. CCC members are found in animals, plants, fungi, archaea and bacteria, and they catalyze NaKCl, NaCl or KCl cotransport across plasma membranes. In human physiology, these transporters are important for cell volume regulation, salt and water reabsorption in the kidney, exocrine fluid secretion, hearing, olfaction, visual processing and GABA-mediated modulation in neurons, among others (1348). The SLC12 family members typically have a membrane topology that includes 12 TMHs flanked by a short cytoplasmic N-terminal region and a long cytoplasmic carboxyl-terminal region, and an extracellular loop containing N-linked glycosylation sites located between TMHs 7 and 8 for NKCC1/2 and NCC and between segments 5 and 6 for KCCs (1348–1351). The cryo-EM structures of the SLC12 transporter confirm some key features established by traditional biochemical and molecular methods, including the formation of a dimer as the functional unit (1352). The information obtained from more than 40 structures of Na-K-Cl transporters has been thoroughly reviewed (220). This includes the binding sites for cations, anions, and inhibitors, the external and internal gates, the subtle movements of the TMDs, and the larger movements of the monomers and C-terminal tails between different conformations.

In accordance with the phylogenetic tree (Fig. 23), the description of the family members is divided into 1) the Na⁺-K⁺-Cl⁻ cotransporter / NaCl cotransporter group, 2) the KCl cotransporter group and 3) other SLC12 family members.

Fig. 23. Phylogenetic tree of the SLC12 family.

Na⁺-K⁺-Cl⁻ cotransporter / NaCl cotransporter group

SLC12A1: The Na⁺-K⁺-Cl⁻ cotransporter NKCC2 (SLC12A1; also known as BSC1) is expressed in the apical membrane of the thick ascending limb of Henle where it mediates NaCl reabsorption serving as the main renal salt transport pathway (1348). In doing so, it plays a key role in blood pressure regulation and the reabsorption of the divalent cations Ca²⁺ and Mg²⁺, as well as in maintaining the countercurrent mechanism responsible for urine concentration (1348, 1353, 1354). It is inhibited by loop diuretics such as furosemide, and its inactivating mutations result in Bartter type I syndrome characterized by hypotension, hypokalemia, hypercalciuria and metabolic alkalosis (1355–1358) (see also ClC-K channel description in Section 10).

SLC12A2: The bumetanide-sensitive NKCC1/BSC2 (SLC12A2) is ubiquitously expressed (1348). In polarized secretory epithelia it is expressed at the basolateral membrane where it facilitates Cl⁻ loading into cells. Cl^- is then secreted across the apical membrane via CFTR with water and Na⁺ following paracellularly as part of intestinal fluid recycling during digestion to maintain chyme fluidity (1348, 1359) (see the SLC5A1 description).

Similarly, in secretory cells of the lung airway epithelium, Cl⁻ is imported through basolateral NKCC1 to maintain the intracellular Cl⁻ levels necessary for Cl⁻ secretion via apical CFTR channels, thereby driving water secretion (1360) (see Fig. 33 at the bottom).

SLC12A3: The thiazide-sensitive NaCl cotransporter NCC (SLC12A3; also known as TSC) is responsible for the reabsorption of ~5% of the filtered load of NaCl in the kidney and loss of function mutations of the SLC12A3 gene cause Gitelman syndrome, an autosomal recessive salt wasting disorder (1348, 1354, 1361).

2) KCl cotransporter group

SLC12A4: KCC1 (SLC12A4) encodes the first reported Na⁺-independent KCl cotransporter (1362). KCC1 shares only 33-36% amino acid identity with the three Na⁺-dependent cation chloride cotransporters, which share ~75% amino acid identity. This places this cotransporter in a subfamily distinct from NCC, NKCC1 and NKCC2. In terms of tissue distribution, KCC1 (SLC12A4) is widely expressed, with particularly high expression in the choroid plexus (Fig. 11). KCC1 is a housekeeping transporter that also plays a crucial and specific role in pathological situations, with supplementary functions in hematopoietic and cancer cells (1363).

SLC12A5: KCC2 (SLC12A5) is expressed in several brain areas and the retina (1348, 1364). It is a neuron-specific KCl cotransporter responsible for establishing the Cl^- ion gradient in neurons by maintaining low intracellular Cl^- concentrations (see Fig. 16). KCC2 is the main Cl^- extruder of neurons, and it ensures the proper inhibitory function of the neurotransmitters GABA and glycine (1365). To dampen neuronal excitability, GABA hyperpolarizes membrane potentials by opening Cl^--permeable GABAA receptor channels, thereby facilitating Cl^- influx down its electrochemical gradient, which is entirely dependent on KCC2. A pathogenic SLC12A5 missense variant has been identified as the cause of infantile epilepsy with migrating focal seizures due to impaired KCC2 chloride extrusion (1366). KCC2 is considered a promising drug target for epilepsy treatment (1364).

SLC12A6: KCC3 (SLC12A6) is widely expressed, including in the retina and bone marrow (1367–1369). It plays a critical role in the nervous system in regulating cell volume through electroneutral efflux of K⁺ and Cl^- and diffusion of water molecules (1370). In addition, KCC3 and also KCC4 (see below) are required for inner ear structure and function. KCC3 is expressed in many cells involved in the inner ear K⁺ recycling pathway (1371).

SLC12A7: KCC4 (SLC12A7) is widely expressed, also in heart muscle and on the basolateral membrane of renal tubule α-intercalated cells (1368). Among the KCCs, KCC4 is the most strongly activated by cell swelling (1368). Slc12a7 knockout mice are deaf, but otherwise seem to be neurologically intact (1372). It has been shown that the inner ear of Slc12a7 knockout mice loses almost all outer hair cells between postnatal days P14 and P21 (1372). Outer and inner ear hair cells probably degenerate in the absence of KCC4 due to the disrupted ionic microenvironment. Mutations in human genes that disrupt physiological K⁺ (and Cl^-) flux in the inner ear have been described to cause hereditary deafness (1373). Slc12a7 knockout mice also exhibit renal tubular acidosis (1372). In Slc12a7 knockout mice, α-intercalated cells have significantly higher total intracellular Cl^- levels, suggesting an impairment of basolateral Cl^- efflux (1372). Since α-intercalated cells exchange intracellular Cl^- for basolateral HCO₃^- (see Fig. 12), this may explain the acidosis and alkaline urine of KCC4 knockout mice.

3) Other SLC12 family members

SLC12A8: CCC9 (SLC12A8) was originally proposed to be involved in the transport of polyamines and amino acids across the cell surface (1374). However, it was later discovered that CCC9 (SLC12A8), which is highly expressed in the intestine, is the first identified nicotinamide mononucleotide (NMN) transporter (1375). NMN, a biosynthetic precursor of NAD⁺, promotes cellular NAD⁺ production. This in turn counteracts age-related pathologies associated with a decrease in tissue NAD⁺ levels (1375). SLC12A8 deficiency significantly decreases NAD⁺ levels in the jejunum and ileum, which has been shown to be associated with reduced NMN uptake. CCC9 (SLC12A8) is also expressed in a specific neuronal subpopulation in the lateral hypothalamus where it has been reported to play an important role in the regulation of energy expenditure, whole body metabolism and skeletal muscle function (1376).

SLC12A9: CIP (SLC12A9) was originally cloned and characterized as a putative heterologous CCC-interacting protein (CIP) that may modify the activity of SLC12 family members through heterodimer formation (1377). It was later shown to function as a co-transporter of NH₄⁺ and Cl^- ions out of lysosomes, thus acting as a lysosomal detoxifier (1378).

In an effort to identify new genes regulating lysosomal volume, a genome-wide activation screen was performed to detect suppression of enlarged lysosomes in cells of the FIG4^-/- HAP1 cell line (1379). Pathogenic variants of FIG4 (“Factor-Induced Gene 4” encoding phosphoinositide 5-phosphatase) cause enlarged lysosomes and neurological and developmental disorders (1380). The screen using FIG4^-/- cells led to the identification of SLC12A9 as another protein capable of correcting lysosomal swelling. Biallelic loss-of-function variants of SLC12A9 were subsequently found to cause lysosomal dysfunction and a syndromic neurodevelopmental disorder. The data link SLC12A9 to a role in lysosomal function, likely related to osmoregulation, and the recent demonstration of the cotransport of NH₄⁺ and Cl^- ions out of lysosomes may be in line with this concept. Thus, the study adds to the list of neurogenetic conditions caused by defective lysosomal membrane proteins (1379).

Structural insights and pharmaceutical perspectives

The X-ray structure of the C-terminal domain of an SLC12 homolog from the archaeon Methanosarcina acetivorans revealed a novel fold of a regulatory domain distantly related to universal stress proteins (1381). The cryo-EM structures of human KCC1 (SLC12A4) in KCl or NaCl showed that KCC1 exists as a dimer, with both extracellular and transmembrane domains involved in dimerization. The studies reveal one potassium and two chloride binding sites in KCC1. The KCC1 structures allow modelling of a potential ion transport mechanism in KCCs and provide the rationale for drug design (1382). Single-particle electron cryo-microscopy studies revealed an outward-facing conformation of NKCC1/SLC12A2, showing bumetanide wedged into a pocket in the extracellular ion translocation pathway (1351). The cryo-EM structures of human NCC/SLC12A3 alone and in complex with a thiazide diuretic, together with functional studies, reveal major conformational states, an intriguing regulatory mechanism, and illustrate how thiazide diuretics specifically interact with NCC to inhibit its transport function (1383). As highlighted above, the available structural information of the Na-K-Cl transporters has been thoroughly reviewed (220). These findings also provide a blueprint for future drug design and help interpret disease-related mutations.

Orphan transporter family members: N/A

SLC13 Na⁺-sulfate/carboxylate cotransporter family (2.A.47.1/Na_sulph_symp/AbgT)

Discovery: The first members of the SLC13 family were identified by Xenopus oocyte expression cloning. The founding member was the rat kidney cortex Slc13a1 Na⁺-sulfate cotransporter (147). The transporter was subsequently cloned from rat intestine (1384). This was followed by the cloning of the Slc13a2 rabbit renal Na⁺-dicarboxylate cotransporter (1385), the Xenopus Slc13a2 (1386) and the winter flounder Slc13a3 (1387).

Gene family members (6):
SLC13A1 (NaS1)	SLC13A3 (NaC3/NaDC3/SDCT2)	SLC13A5 (NaC2)
SLC13A2 (NaC1/NaDC1/SDCT1)	SLC13A4 (NaS2)	OCA2 (SLC13B1)

Molecular aspects, physiological roles and links to disease

The SLC13 family consists of five electrogenic Na⁺-coupled anion cotransporters and a putative electrodiffusion chloride channel. The family can be further divided into the Na⁺-sulfate cotransporters NaS1 (SLC13A1) and NaS2 (SLC13A4), the Na⁺-di- and tricarboxylate cotransporters NaC1 (SLC13A2; also known as NaDC1), NaC3 (SLC13A3; also known as SDCT2 and NaDC3), NaC2 (SLC13A5) (1388, 1389), and the putative chloride channel SLC13B1/OCA2. NaS1 is responsible for sulfate homeostasis, while NaC transporters regulate oxidative metabolism, and OCA2 increases organelle pH. NaS1 (SLC13A1) plays an important role in growth and fertility, while NaC1 (SLC13A2) plays a crucial role in the regulation of urinary citrate concentration which in turn is linked to calcium nephrolithiasis, while OCA2 (SLC13B1) helps regulate melanin production.

The mammalian SLC13 proteins are members of the larger Divalent Anion Sodium Symporter (DASS) family (TC 2.A.47) (1390), and they have the structural architecture of the DASS/AbgT fold (Fig. 4). Extensive structural studies of the bacterial homologs of the SLC13 family have revealed an 11-TMH topology and a dimeric state, together with an elevator-type transport mechanism (279, 281, 1391, 1392). Based on AlphaFold2 modeling, DASS/AbgT transporters, including the distantly related OCA2 (SLC13B1), can exist in inward- and outward-facing conformations and an elevator-type transport mechanism has been proposed (1393). In addition, the first cryo-EM structure of the human Na⁺-dependent citrate transporter NaC2 (SLC13A5) has been reported (1394), and subsequently additional cryo-EM-derived structures of human NaS1 (SLC13A1) and NaC1/NaDC1 (SLC13A2) revealed multiple conformations along the transport cycle (280). Collectively, these studies uncovered the intricacies of substrate and inhibitor binding and conformational changes during the transport cycle within human SLC13 members and provided critical insight into the optimization of anti-obesity drugs.

In accordance with the phylogenetic tree (Fig. 24), the description of the family members is divided into 1) the Na⁺-sulfate cotransporters group, 2) the Na⁺-di- and tricarboxylate cotransporter group, and 3) the SLC13B1 (OCA2) branch.

Fig. 24. Phylogenetic tree of the SLC13 family.

Na⁺-sulfate cotransporter group

SLC13A1: NaS1 (SLC13A1) is an apical plasma membrane Na⁺-coupled sulfate cotransporter that primarily mediates sulfate reabsorption across renal proximal tubule cells (147, 1389). At lower levels, it is also expressed in the intestine, where it facilitates sulfate absorption. Plasma sulfate levels are maintained by reabsorption of filtered sulfate in the renal proximal tubule, which requires the apical transporter NaS1 (SLC13A1) together with the basolateral membrane sulfate anion exchanger SLC26A1 (see the SLC26 family description for details). A homozygous loss-of-function variant in the SLC13A1 gene was identified in a patient with unexplained skeletal dysplasia (1395). It was concluded that loss of the SLC13A1 gene results in profound hypersulfaturia and hyposulfatemia, primarily associated with abnormal skeletal development, which may predispose to degenerative bone and joint disease (1395, 1396).

SLC13A4: NaS2 (SLC13A4) facilitates the cellular uptake of sulfate and plays a critical role in cellular sulfate metabolism, especially during development. Once inside cells, sulfate is converted to 3ʹ-phosphoadenosine 5ʹ-phosphosulfate (PAPS), which serves as a sulfate donor for sulfonation by sulfotransferases (1397). Enzymatic sulfation is an important metabolic pathway for many compounds, including steroid hormones, bile acids, neurotransmitters, and small peptides. Sulfation is a widely used biotransformation that serves to regulate biological activity (1398). Xenobiotics are also subject to sulfation, a process that reduces their biological activity. Modification of proteoglycan glycosaminoglycan chains by sulfation is important for modulating the physiological roles of proteoglycans, particularly in developing tissues. In the developing brain, sulfation reactions alter the functions of extracellular matrix components, which in turn regulate local growth factor interactions critical for neurogenesis, axon guidance, and synaptogenesis (1399, 1400).

Observations in humans and mouse models suggest that dysregulated sulfate levels are linked to neurodevelopmental disorders such as autism, and the sulfate transporter NaS2 (Slc13a4) is critical for brain development, as Slc13a4 deficiency in adult mice has been shown to cause behavioral problems, including impaired social interaction and long-term memory (1401). Conditional gene deletion has shown that these phenotypes have a developmental origin and that full biallelic expression of Slc13a4 is required during postnatal development. Specifically, it has been shown that administration of N-acetylcysteine between developmental stages P14 and P30 prevents the onset of phenotypes in adult Slc13a4^+/- mice.

NaS2 (Slc13a4) has been shown to be prominently expressed at the basolateral membrane in choroid plexus epithelial cells (see Fig. 11), indicating that NaS2 (SLC13A4) regulates the uptake of sulfate from the blood into the choroid plexus epithelium (1401). Based on the experimental findings, it was concluded that NaS2 (SLC13A4) plays an important role in sulfate delivery to the brain, which is critically important during postnatal development. Consistent with these findings, the HPA suggests that NaS2 (SLC13A4) is most highly expressed in the choroid plexus and in syncytiotrophoblasts.

In addition, the sulfate transporter NaS2 (SLC13A4) plays an anti-apoptotic role by regulating intracellular sulfate levels, which modulate caspase-3 sulfation and influence cell survival (1402). Thiosulfate, a sulfur donor, has been investigated for its protective effects in ischemic injuries (e.g., in the brain), likely by replenishing sulfur-containing metabolites and reducing oxidative stress. While NaS2 (SLC13A4) is involved in maintaining sulfate homeostasis and may contribute to anti-apoptotic signaling via caspase-3 sulfation, it is unclear whether its transport activity limits apoptosis during ischemic brain injury.

Interestingly, SLC13A4 mRNA levels were found to be markedly reduced in head and neck squamous cell carcinoma compared to paracancerous tissue, and decreased SLC13A4 expression was associated with poor overall survival. NaS2 (SLC13A4) has been proposed as a biomarker for head and neck tumors (1403).

Na⁺-di- and tricarboxylate cotransporter group

SLC13A2: NaC1 (SLC13A2), also known as NaDC1 or SDCT1, absorbs dietary dicarboxylates across the intestinal brush border membrane, including citrate and other Krebs cycle intermediates. It is also expressed in the kidney, where it reabsorbs most of the filtered citrate across the apical membrane of renal proximal tubule cells (1404).

Low urinary concentrations of the Ca²⁺ chelator citrate can promote Ca²⁺ stone formation because urinary citrate can inhibit the crystallization and precipitation of Ca²⁺ stones by chelating Ca²⁺ ions. Therefore, sufficient urinary citrate concentration is key to preventing stone formation. Since NaC1/NaDC1 (SLC13A2) reabsorbs citrate in the proximal tubuli it is also a determinant of renal stone formation (1405).

In addition, the oxalate transporter SLC26A6 is a determinant of renal stone formation due to its ability to interact with and inhibit the activity of the citrate transporter NaC1/NaDC1 (SLC13A2) (1405, 1406). Specifically, in the absence of SLC26A6, the risk of stone formation is increased in the following two ways: 1) Increased activity of NAC1/NADC1 due to lack of interaction with SLC26A6 is expected to result in increased reabsorption of filtered citrate, decreased excretion of urinary citrate, and increased risk of calcium nephrolithiasis, and consistent with this, Slc26a6 knockout mice exhibited increased renal reabsorption of citrate, decreased excretion of urinary citrate, and increased risk of stone formation (1407); and 2) since SLC26A6 is thought to play an important role in the intestine in reabsorbing oxalate that has been passively absorbed through tight junctions, a defect in intestinal oxalate secretion is expected to lead to increased net oxalate absorption, hyperoxalemia, and hyperoxaluria. Indeed, Slc26a6-null mice had hyperoxaluria (1407). Of note, while in the kidney, a defect in renal oxalate secretion in Slc26a6-null mice would be expected to reduce urinary oxalate, the defect in intestinal oxalate secretion, which leads to increased oxalate absorption, appears to override the effect of reduced oxalate secretion in the kidney, resulting in overall hyperoxaluria.

The STAS domain of SLC26A6 and the cytosolic H4c domain of NaC1/NaDC1 were shown to mediate the physical and functional interactions of these transporters, revealing a molecular pathway that senses and tightly regulates oxalate and citrate levels and controls Ca²⁺-oxalate stone formation (1407, 1408). For detailed information on the structure and function of SLC26A6 and the role of its STAS domain, see the description of SLC26A6.

Furthermore, a link between nephrolithiasis and hypertension has been proposed based on the above-mentioned molecular interaction of NaC1/NaDC1 and SLC26A6 to modulate succinate/citrate and oxalate transport in epithelial cells, whereby the succinate/citrate homeostatic pathway is regulated by IRBIT and affects both blood pressure and the risk of calcium oxalate stone formation (1405, 1409). The following molecular steps are proposed: 1) apical uptake of succinate/citrate is inhibited by SLC26A6 as mentioned above; 2) both apical and basolateral succinate transport are orchestrated by the succinate signaling pathway, whereby luminal succinate stimulates the succinate receptor SUCNR1, which triggers the release of IRBIT via activation of the IP₃ receptor. IRBIT then translocates to the membrane to bind succinate transporters at the apical membrane (NaC1/NaDC1) and basolateral membrane (NaC3/SDCT2/SLC13A3) to coordinate appropriate succinate uptake across the epithelium. According to this model, impaired SLC26A6 function has three independent deleterious consequences: 1) as noted above, it leads to hyperoxalemia and hyperoxaluria due to impaired SLC26A6-mediated oxalate secretion in the intestine; 2) also as noted above, it leads to hypocitraturia; and 3) it leads to hypersuccinatemia and hyposuccinaturia caused by increased NaC1/NaDC1-mediated succinate uptake due to decreased inhibition of NaC1/NaDC1 by SLC26A6 in the kidney, and the increased luminal succinate in the juxtaglomerular apparatus, which increases renin secretion and thereby causes hypertension. These findings further demonstrate that proper succinate homeostasis protects against kidney stone disease and hypertension, and that succinate and citrate transport and signaling pathways are potential therapeutic targets for the treatment of both urolithiasis and hypertension (1407, 1409, 1410).

Despite the beneficial effects of succinate, this compound can also cause inflammation when produced in excessive amounts, for example by certain gut bacteria. This is especially true in the context of IBD as a result of alterations in succinate-metabolizing gut bacteria (1411). It has been suggested that there is a transepithelial pathway for succinate that delivers succinate from the gut microflora to macrophages involving SLC13A2 (in complex with the SLC26A6 transporter to regulate succinate homeostasis) and the succinate receptor SUCNR1 on macrophages. In macrophages, excess cytoplasmic succinate then acts as an inflammatory signal, stabilizing HIF-1α, which induces upregulation of glycolysis-related enzymes and expression of proinflammatory cytokines, thereby switching these cells to a proinflammatory state (1412). SUCNR1 has been reported to further enhance the pro-inflammatory state in macrophages (1413). More studies are still needed to clarify the specifics of this transepithelial pathway.

SLC13A3: NaC3 (SLC13A3), also known as SDCT2 or NaDC3, is a plasma membrane Na⁺-dicarboxylate cotransporter that shares moderate (close to 50%) amino acid identity with NaC1 (SLC13A2) previously identified in kidney and intestine (1414). NaC3 (SLC13A3) is abundantly expressed on the basolateral membrane of renal proximal tubule S3 segments, consistent with sites of high-affinity dicarboxylate transport (1414, 1415). In the basolateral membrane of renal proximal tubule cells, it takes up Krebs cycle intermediates for metabolic purposes and for secretion of organic anions by dicarboxylate/organic anion exchange via the organic anion exchangers OAT1 (SLC22A6) and OAT3 (SLC22A8) (Fig. 33). NaC3 mediates Na⁺-dependent transport of di- and tricarboxylates with a substrate preference for succinate over citrate but excluding monocarboxylates. In contrast to NaC1 (SLC13A2), NaC3 (SLC13A3) exhibits a unique pH dependence for succinate transport (optimal pH 7.5-8.5) and a high affinity for dimethylsuccinate, two features characteristic of basolateral transport (1414).

NaC3 (SLC13A3) also transports other important metabolites into the cell, including glutathione (1416), mercaptosuccinate, and N-acetylaspartate (1417). N-acetylaspartate is primarily synthesized in neurons by the enzyme N-acetyltransferase-8-like and is broken down in oligodendrocytes by aspartoacylase into acetate and aspartate. N-acetylaspartate links the metabolism of axons with oligodendrocytes to support myelination, allowing neuron-derived N-acetylaspartate to signal in the oligodendrocyte nucleus to support or maintain myelination (1418). The HPA suggests that NaC3 (SLC13A3) is abundantly expressed in oligodendrocytes, the predominant cell type of brain white matter, which is mainly composed of myelinated axons, which seems consistent with the following pathologies.

Loss-of-function mutations in SLC13A3 are the cause of a rare group of disorders called “acute reversible leukoencephalopathy with elevated urinary α-ketoglutarate” (ARLIAK). These are autosomal recessive disorders of brain white matter characterized by acute reversible neurological deterioration. They are characterized by developmental abnormalities or white matter degeneration, leading to acute episodes of deterioration and increased urinary accumulation of dicarboxylic acids, particularly α-ketoglutarate (1419–1421). However, whether the clinical manifestations of ARLIAK are due to defective axon myelination and/or other pathological effects resulting from NaC3 dysfunction remains to be determined.

SLC13A5: NaC2 (SLC13A5) is a Na⁺-coupled citrate transporter expressed in the plasma membrane of specific cell types in liver (1422, 1423) and brain (1424, 1425). According to the HPA, NaC2 is highly and almost exclusively expressed in hepatocytes and salivary epithelium, with only low levels detected in astrocytes in the brain. NaC2 is an electrogenic transporter with a Na⁺ to citrate stoichiometry of 4:1 (1426).

In the liver, NaC2 facilitates the uptake of circulating citrate for metabolic energy production and for the synthesis of fatty acids and cholesterol (1422). In the brain, it is thought to contribute to the trafficking of tricarboxylic acid cycle intermediates and related metabolites between glia and neurons (1424).

Loss of function mutations in SLC13A5 cause autosomal recessive epileptic encephalopathy with seizure onset in the first days of life, accompanied by neurological impairment, developmental delay (1427–1429) and dental hypoplasia (1430). This rare SLC13A5 deficiency disorder has drawn growing interest for therapeutic development (1431–1433), including gene replacement strategies, though major hurdles remain due limitations of current disease models and incomplete understanding of the disease mechanisms (1431, 1433).

A subsequent large-scale functional study of SLC13A5 variants systematically assessed the effects of hundreds of missense mutations on protein stability, trafficking, and citrate transport activity (1434). This work provides an unprecedented functional map of pathogenic and benign variants, clarifies the structural determinants of transport, and offers a powerful resource for variant interpretation in clinical genetics and for future gene-therapy design.

Beyond the neurological implications, the high expression of NaC2 in hepatocytes and its role in importing citrate make it an attractive metabolic drug target. Inhibition of hepatic citrate uptake has been proposed as a strategy to reduce lipogenesis and treat obesity (1394). Cryo-EM structures of human NaC2 in complexes with citrate or a small molecule inhibitor revealed the structural basis for citrate binding and for inhibition. The studies show how the inhibitor occupies the citrate-binding site and blocks the transport (1394). These structures also explain how disease-causing mutations disrupt transport activity.

SLC13B1 (OCA2) branch

OCA2 (SLC13B1) – Orphan transporter: OCA2 (SLC13B1), also known as pink-eyed dilution and oculocutaneous albinism II, is encoded by the OCA2 gene, and is a melanosomal transmembrane protein reported to be essential for skin and eye pigmentation. OCA2 has been given the alias SLC13B1 because OCA2 and SLC13 proteins share the “AbgT” structural fold and are the only proteins in humans that are predicted to have the “DASS/AbgT” structural fold. The TCDB classifies these proteins in family #2.A.45 (the Arsenite-Antimonite (ArsB) efflux family), in different subfamilies (OCA2: #2.A.45.2; SLC13: #2.A.45.1). A phylogenetic tree was constructed from the entire #2.A.45 family of TCDB and the SLC13 and OCA2 families, supporting their placement in the same family (7).

OCA2 has been proposed to act as an electrodiffusion chloride channel protein that increases organelle pH to regulate melanin synthesis by modulating melanosome pH (1435–1437). Mutations of the OCA2 gene cause albinism (1438, 1439). The HPA shows high expression in skin as well as in choroid plexus at the mRNA level. The role of OCA2 in the choroid plexus remains unknown and needs to be further investigated.

Oca2 (Slc13b1) shares the DASS/AbgT topology and possesses a GOLD (Golgi dynamics)-like domain (1393). GOLD domains are known as protein modules involved in Golgi function and secretion (1440). OCA2 also possesses binding residues corresponding to key citrate binding sites in the Na⁺-carboxylate transporters NaCT1-3, and citrate docking to OCA2 at the putative binding site has been successful (1393), further highlighting the structural similarity between OCA2 and the other SLC13 transporters. Based on AlphaFold2 modeling, OCA2, like other DASS transporters, can exist in inward- and outward-facing conformations supporting an elevator-type transport mechanism.

Orphan transporter family member (1)

OCA2 (SLC13B1)

HGNC update

OCA2 has been given the alias SLC13B1.

SLC14 Urea transporter family (1.A.28.1/UT/UT)

Discovery: The molecular basis of urea transport in kidney and erythrocytes remained a puzzle until the rat renal urea transporter UT2 (SLC14A2) was discovered (138). Subsequently, the erythrocyte urea transporter from human bone marrow cells was cloned (SLC14A1, also referred to as HUT11/UT-B1) (1441).

Gene family members (2):
SLC14A1 (splice variants UT-B1 to UT-B2)	SLC14A2 (splice variants UT-A1 to UT-A6)

Molecular aspects, physiological roles and links to disease

The SLC14 family belongs to the Urea Transporter (UT) family (TC 1.A.28). It carries the UT fold, which contains two homologous halves with opposite orientations in the membrane, and each half contains five TMHs and a tilted reentrant helix (37) (see Fig. 4). Its members are facilitative transporters that allow urea to move down its concentration gradient. Several splice variants of the SLC14 urea transporters have been identified, some with cell-specific expression patterns. In general, SLC14 urea transporters have 10 TMHs with a glycosylated extracellular loop between TMH5 and TMH6 and intracellular C- and N-termini. An exception is the SLC14A2 UT-A1 splice variant, which consists of two functional urea transporters linked to form 20 TMHs.

The crystal structure of a bacterial homolog (dvUT from Desulfovibrio vulgaris) of the kidney urea transporter revealed that it is a homotrimer, with each subunit containing a continuous membrane-spanning pore formed by the two homologous halves of the protein (58, 59). A similar architecture was found for the bovine urea transporter UT-B (37). All structural studies of human UT-A (SLC14A2) and human and bovine UT-B (SLC14A1) as well as the prokaryotic homolog dvUT show a membrane-spanning pore with specific conserved selectivity sites and revealed a channel-like transport mechanism (338–341). Specific structural insights of UT-A and UT-B in the context of interactions with small molecule inhibitors and their selectivity have also been reported (340). The studies reveal phospholipids associated with the urea transporters. The structures of the urea transporters improve our understanding of their function and aid in the development of new diuretics that target the urea transporters.

The urea transporters expressed in the kidney serve as a major component of the urinary concentrating mechanism and nitrogen excretion (9) (Fig. 25). As previously discussed (9, 25), there is also experimental evidence for active urea secretion in the proximal tubules of the kidney (Fig. 25), although the underlying transport mechanism has not been elucidated at the molecular level.

Fig. 25. Schematic model of the role of renal urea transporters in the urine concentrating mechanism.

Urea is reabsorbed by UT-A1 and UT-A3 in the inner medullary collecting duct (CD) and accumulates in the medullary interstitium, while UT-B plays a role in countercurrent exchange between the descending vasa recta (DVR) and ascending vasa recta (AVR) to maintain medullary hypertonicity (9). Consistent with knockout data (10), UT-A2 contributes to urea accumulation in the medullary interstitium by reabsorption of urea in the descending thin limb (tDL), a process that requires active urea secretion in the proximal tubules (see orange arrows with question mark). However, the molecular nature of active urea secretion remains elusive, despite a hypothesized mechanism (25). Independent of urea, NaCl reabsorption and accumulation in the renal medulla via NKCC2 expressed in the thick ascending limb (TAL) is considered to play an important role in urine concentrating mechanisms by initiating hypertonicity in the inner medullary interstitium.

SLC14A1: UT-B1 (SLC14A1) is highly expressed in erythrocytes and human erythrocytes lacking UT-B1 (Kidd blood group Jk antigen null) have a decreased permeability to urea (1442). As a result, these erythrocytes are unable to lose urea rapidly enough and take some of the urea out of the renal medulla and into the bloodstream, thereby reducing the efficiency of countercurrent exchange (see below) and the ability to concentrate urine (1443, 1444). In the kidney, it is expressed in the descending vasa recta and papillary epithelia (9).

A splice variant, UT-B2, is expressed in brain and bladder (1445). The distribution of rat Slc14a1 mRNA (referred to as UT3 mRNA) was studied in brain in detail by in situ hybridization (1446) and was identified in astrocytes throughout the central nervous system as well as in Bergmann glia in the cerebellum, among other brain areas. A follow-up study confirmed the expression in rat brain also at the protein level in specific brain areas. While the significance of urea formation in the normal brain is not fully understood, the fate of urea after synthesis may simply be its excretion into the bloodstream either across the BBB or into the cerebrospinal fluid. Expression in astrocytes is likely to be a mechanism for adjusting the osmotic balance when urea is formed in neurons during ornithine production.

SLC14A2: UT-A (SLC14A2) has 6 splice variants called UT-A1 to UT-A6. Interestingly, UT-A1 shows a duplication of two functional urea transporter structures (UT-A2 and UT-A3) connected by a large hydrophobic intracellular loop, resulting in a transporter with 20 TMHs. (9). Urea uptake by UT-A1 and UT-A3 in the inner medullary collecting duct and by UT-B1 in the descending vasa recta is mainly responsible for medullary urea accumulation in the urinary concentration process (Fig. 25). Vasopressin, an antidiuretic hormone, regulates the UT-A isoforms via phosphorylation and trafficking of the glycosylated transporters to the plasma membrane. These urea transporters are an integral part of countercurrent multiplication, the process in the kidney that creates an osmotic gradient allowing water to be reabsorbed from the tubular lumen while producing concentrated urine. The details of these processes are reviewed elsewhere (9).

Pharmaceutical aspects

UTs are considered a potential diuretic target since urea transporter inhibitors could provide novel diuretics that do not disrupt electrolyte balance. The identification and discovery of small molecule urea transporter inhibitors as a new type of diuretic is underway (1447).

Orphan transporter family members

N/A

SLC15 Proton oligopeptide cotransporter family (2.A.17/PTR2/MFS)

Discovery: Researchers originally believed that ingested protein had to be broken down into individual amino acids by peptidases before absorption could occur in the intestinal lumen. However, studies beginning in the 1970s showed that much of the absorption of protein digestion products in the human small intestine occurs via di- and tripeptides (1226, 1448). This eventually led to the identification of mammalian SLC15 oligopeptide transporters, which are distinct from amino acid transporters. The founding member of this family is the rat intestinal H⁺-coupled oligopeptide transporter PepT1 (SLC15A1), which was identified by expression cloning using Xenopus oocytes (139). Subsequently, the cloning of the rabbit intestinal H⁺-coupled oligopeptide transporter PepT1 (SLC15A1) (1449) and the rabbit renal H⁺-coupled oligopeptide transporter PepT2 (SLC15A2) (1450) were reported.

Gene family members (5):
SLC15A1 (PepT1)	SLC15A3 (PHT2)	SLC15A5
SLC15A2 (PepT2)	SLC15A4 (PHT1)

Molecular aspects, physiological roles and links to disease

The SLC15 family belongs to the Proton-dependent Oligopeptide Transporter (POT/PTR) family (TC 2.A.17) which is part of the MFS superfamily, members of which contain a canonical 12-TMH fold split into two six-helix bundles and follow the alternating access transport mechanism. The cryo-EM structure of PepT2 (SLC15A2) has revealed the structural basis for H⁺-coupled peptide and prodrug transport (1451).

PepT1 (SLC15A1) mainly mediates intestinal absorption of luminal di/tripeptides from total dietary protein digestion. PepT2 (SLC15A2) facilitates renal tubular reuptake of di/tripeptides, PHT2 (SLC15A3) and PHT1 (SLC15A4) interact with both di/tripeptides and histidine in specific immune cells (139, 1452, 1453). SLC15A5 has no known physiological function. Members of the SLC15 family interact with a large number of peptidomimetic and peptide-like drugs, and PepT1 (SLC15A1) is being exploited as a vehicle for transporter-mediated drug delivery (43).

The phylogenetic tree (Fig. 26) shows two distinct branches, and therefore the description of the family members is divided into 1) the H⁺-coupled oligopeptide transporter group and 2) the lysosomal transporter group.

Fig. 26. Phylogenetic tree of the SLC15 family.

1) H⁺-coupled oligopeptide transporter group

This group includes the epithelial oligopeptide transporters PepT1 (SLC15A1) and PepT2 (SLC15A2) and the orphan transporter SLC15A5.

SLC15A1: PepT1 (SLC15A1) is a H⁺-coupled oligopeptide transporter that is abundantly expressed in the brush border membrane of enterocytes in duodenum, jejunum and ileum, with little or no expression in normal colon (139, 1452–1454) (see Fig. 17). PepT1 is the principal route of transport of di-and tripeptides across the intestinal brush border membrane. PepT1 is also expressed in the early proximal tubules of the kidney, together with PepT2 (SLC15A2) in a sequential manner (see the SLC15A2 description below) (1455, 1456). PepT1 is a low affinity, high capacity peptide transporter, in contrast to renal PepT2 (SLC15A2), which is a high affinity, low capacity peptide transporter (1452, 1453). PepT1 transports almost any di- and tripeptide, beta-lactam antibiotics of the cephalosporin and penicillin classes, certain angiotensin-converting enzyme (ACE) inhibitors, antitumor agents such bestatin, and prodrugs such as valacyclovir used in the treatment of viral infections (139).

The proton motive force drives PepT1-mediated transport as follows: Neutral and cationic dipeptides are cotransported with 1 H⁺ while anionic dipeptides are cotransported with 2 H⁺ (1457–1459). A specialized mucus barrier on the apical surface of enterocytes (1460) helps maintain an acidic microclimate at the brush border surface at approximately pH 6.0 (1461), which facilitates H⁺-coupled oligopeptide uptake (Fig. 17). This in turn leads to increased H⁺ efflux back into the lumen by the apical Na⁺/H⁺ exchanger NHE3 (SLC9A3), followed by Na⁺ export by the basolateral Na⁺, K⁺ ATPase. Thus, di- and tripeptides are actively taken up by enterocytes via PepT1, where they undergo rapid intracellular hydrolysis by intracellular dipeptidases. Individual amino acids are then released via basolateral amino acid transporters (Fig. 17).

Slc15a1 knockout mice were generated to study the contribution of PepT1 to amino acid absorption. The study showed that PepT1 becomes particularly important after high dietary protein intake when amino acid transporters are saturated because PepT1 can provide the additional absorptive capacity needed (1005). Oligopeptide transport has been shown to compensate for the loss of intestinal amino acid transporter function, such as in Hartnup disease caused by mutations in SLC6A19 (see the SLC6 family description). The Slc15a1 knockout mice were healthy and no alterations in body weight, development and fertility were observed, even though dipeptide absorption in the intestine was significantly reduced (1462).

PepT1 plays an important role in the development and progression of colonic diseases. While PepT1 expression is normally restricted to the small intestine, its expression is increased in colon biopsies from patients with IBD and colorectal cancer (1463, 1464). Studies in mice showed that overexpression of human PepT1 in the colon exacerbates experimental colitis, whereas Slc15a1^-/- mice are protected against dextran sulfate sodium (DSS)-induced colitis and colitis-associated cancer (1463, 1465). Further studies in mice highlighted a role for PepT1 in altering the composition of the microbiota and consequently the susceptibility to colitis and cancer (1466). Specifically, the absence of PepT1 in Slc15a1^-/- mice had drastic consequences on the composition and localization of the intestinal microbiota, leading to a thicker mucus layer and an increased number of goblet cells per crypt. This provides a more favorable condition to promote reduced susceptibility to colitis (1466).

PepT1 can transport a wide range of bacterial di- and tripeptides, which act as microbial agents contributing to the pathogenesis of intestinal inflammation by triggering downstream proinflammatory effects (1466). In the healthy colon, where PepT1 expression is suppressed, epithelial cells are protected despite constant exposure to bacterial flora that release such peptides into the intestinal lumen. However, in the presence of PepT1, several proinflammatory factors have been shown to be taken up by the transporter. For example, fMLP, a tripeptide produced by E. coli and commonly found in the intestinal lumen, is a transport substrate of PepT1 (1467). Briefly, PepT1 mediates the entry of fMLP into human polarized Caco2-BBE cells, a clone of the Caco2 cell line that exhibits a brush-like morphology (1468). These cells express PepT1, resulting in fMLP uptake accompanied by cytosolic acidification due to H⁺ co-transport function. The fMLP uptake caused directed movement of neutrophils across epithelial monolayers, whereas inhibition of PepT1-mediated fMLP transport decreased neutrophil transmigration (1467). Thus, it was proposed that PepT1-mediated fMLP uptake in colonic epithelial cells influences neutrophil-epithelial interactions. The results highlight the importance of hPepT1 in mediating intestinal inflammation. PepT1 has also been shown to transport the proinflammatory muramyl dipeptide (MDP) (1469) and the proinflammatory tripeptide L-Ala-γ-D-Glu-meso-diaminopimelic acid (tri-DAP) (1470). MDP is a component of peptidoglycan, a component of the cell wall of both gram-negative and -positive bacteria, whereas tri-DAP is a peptidoglycan degradation product of gram-negative bacteria.

The downstream effects of these peptides in intestinal epithelial cells, such as in the colon, have been investigated in several studies. For example, treatment of Caco2-BBE cells with fMLP stimulated activation of the nuclear factor NF-κB and activator protein 1 (AP-1), which regulate important biological and pathological processes (1471). A model of inflammatory activation has been presented (1466), according to which the accumulation of bacterial di- and tripeptides stimulates the NF-κB pathway, leading to the activation of proinflammatory cytokines. In addition, IBD can cause disruption of barrier function, resulting in the transport of bacterial di-tripeptides via the paracellular pathway. Once in the lamina propria, they are taken up by macrophages, possibly via SLC15A3, where they can signal upregulation of major histocompatibility class I molecules (1466). This links proinflammatory bacterial peptides to macrophages as a central component of the innate immune system responsible for defense against a variety of pathogens.

PepT1 (SLC15A1) has also been shown to be essential for the growth of pancreatic ductal adenocarcinoma (PDAC) cells, and inhibition of PepT1 reduced cancer cell proliferation (1472). PepT1 is upregulated in PDAC cell lines and patient-derived xenografts (PDXs), whereas it is expressed at very low levels in normal pancreas. PepT2 (SLC15A2) is also overexpressed in PDAC cell lines and PDXs and at low levels in normal pancreas, but it is not functional due to its intracellular localization. It was concluded that PepT1 is critical for cancer cell survival and that tumor-derived lactic acid generated by the Warburg effect in the tumor microenvironment supports the transport function of PepT1 in maintaining amino acid nutrition in cancer cells by inducing matrix metalloproteinases (MMPs) and dipeptidyl peptidase 4 (DPP-4) to generate peptide substrates for PepT1 and by generating a H⁺ gradient across the plasma membrane to energize PepT1. The studies highlight a link between PepT1 function and extracellular protein degradation in the tumor microenvironment as a key determinant of pancreatic cancer growth and establish PepT1 as a potential therapeutic target for PDAC (1472). Specifically, it has been proposed that tumor-derived lactic acid elicits transcriptional activation of MMPs and DPP-4, which then degrade collagen within the extracellular matrix into di- and tripeptides, and that the resulting small peptides and amino acids induce PepT1 expression, which, owing to the acidic pH in the tumor microenvironment, becomes functional and brings in dipeptide substrates inside the cancer cells that get hydrolyzed into amino acids to promote DNA and protein synthesis and support tumor growth.

SLC15A2: PepT2 (SLC15A2) is a H^±coupled oligopeptide transporter that is highly expressed in kidney and has a significantly higher substrate affinity compared to PepT1 (SLC15A1) (1452, 1453). PepT2 transports neutral substrates with a 2 to 1 proton to substrate stoichiometry and charged substrates with variable coupling ratios (1473). PepT2 is responsible for reabsorption of filtered di- and tripeptides and peptidomimetics across the renal proximal tubuli. And as mentioned in the description of SLC15A1, while PepT1 is expressed in the early proximal tubule S1 segments of the kidney, PepT2 is expressed primarily in the S3 segments (1455, 1456).

Using nonisotopic in situ hybridization, the expression of SLC15A2 mRNA has been studied in the CNS. It is expressed in brain by astrocytes, subependymal cells, ependymal cells and epithelial cells of the choroid plexus. Furthermore, PepT2 is expressed in retina by Müller cells and in dorsal root ganglia by satellite cells. SLC15A2 mRNA expression in astrocytes was found to be moderate and relatively homogenous throughout the brain except for an area in ventral forebrain where SLC15A2 mRNA levels were below average. The data suggest that removal of neuropeptide fragments from brain extracellular fluid occurs via PepT2 expressed in astrocytes, ependymal cells and choroid plexus epithelial cells. PepT2 was found to be expressed on apical membranes of the choroid plexus epithelial cells where it is responsible for the efflux of peptidomimetics from cerebrospinal fluid into choroidal tissue (1452, 1474) (see Fig. 11).

SLC15A5 – Orphan transporter: SLC15A5 has been proposed as an additional member of the SLC15 family that is mainly expressed in pituitary, liver, thymus and spleen tissues (1475, 1476). Interestingly, the HPA suggests that it is not expressed at significant levels in any tissue, although at the single cell level it shows significant expression in early spermatids and at lower levels in astrocytes, cone photoreceptors and excitatory neurons. Thus, further studies are needed to clarify the expression and functional role of SLC15A5 in human tissues.

2) The lysosomal transporter group

The SLC15 family includes the lysosomal transporters PHT1 (SLC15A4) and PHT2 (SLC15A3), which have low sequence similarity to SLC15A1 and SLC15A2. Their substrate selectivity covers histidine, oligopeptides and peptidoglycan fragments from the cell wall of gram-positive bacteria. There have been many more studies done with SLC15A4 and therefore SLC15A4 will be reviewed before SLC15A3.

SLC15A4: SLC15A4 is an endo-lysosomal transporter expressed in immune cells, preferentially in plasmacytoid dendritic cells, especially after Toll-like receptor (TLR) stimulation, where it plays a critical role in autoimmune and other inflammatory diseases (1477–1479). SLC15A4 expression and cellular localization have also been reported in inflamed colonic epithelia of IBD patients, suggesting induction of SLC15A4 expression in response to epithelial inflammation in the colonic epithelium (1480).

Genome-wide analyses as well as data from mouse models demonstrate that the function of SLC15A4 is closely related to the pathogenesis of T2D (1481), systemic lupus erythematosus (SLE) (1482–1486) and IBD (1478) (1483–1486).

SLC15A4 is an important component of the TLR-mediated inflammatory response system (1487). The mechanism of action has been hypothesized to involve H⁺-coupled transport of histidine and bacterial peptidoglycans from endosomes or lysosomes to the cytosol. Regulation of lysosomal pH and histidine concentration by finely tuned H⁺-coupled transport provides a milieu for optimal functionality of endolysosomal components, including those of TLR signaling pathways, cathepsin stability, and V-ATPase activity. These components can be modulated by histidine levels due to the buffering capacity of histidine and overall lysosomal acidity (1488). Thus, the absence of SLC15A4, as is the case in the feeble mouse model and Slc15a4^-/- mice (1489), leads to a failure in the homeostasis of the lysosomal environment and likely explains the disruption of the TLR signaling pathway in SLC15A4-deficient cells. Importantly, Slc15a4 feeble and Slc15a4^-/- mice show pronounced reductions in lupus erythematosus (1490) (1482, 1491) and IBD (1478, 1492) manifestations.

Moreover, SLC15A4 has been shown to interact with TASL, the “TLR adaptor interacting with SLC15A4 on the lysosome”, which modulates TLR7, TLR8, and TLR9 signaling and mediates induction of type I interferon (IFN-I) genes by recruiting the interferon regulatory factor IRF5 (1493). Consequently, loss of TASL or mutations that impair complex formation have been reported to mimic the phenotype of SLC15A4 deficiency, resulting in impaired type I IFN but not NF-κB or MAPK signaling. This suggests that recruitment of TASL is a critical role of SLC15A4 in the TLR7-9 pathway.

In addition, SLC15A4 (and also SLC15A3, see below) is involved in the transport of peptidoglycans from lysosomes to the cytosol, where they are ligands of NOD (nucleotide-binding oligomerization domain)-like receptors (1488). Specifically, SLC15A4 facilitates the transport and thereby the endo-lysosomal exit of the bacterial-derived peptidoglycans muramyl dipeptide (MDP), L-Ala-γ-D-Glu-meso-diaminopimelic acid (Tri-DAP), and γ-D-glutamyl-meso-diaminopimelic acid (iE-DAP) all of which are NOD-like receptor ligands (1494, 1495). The two major NOD-like receptors NOD1 and NOD2 directly bind to ligands through their variable tandem C-terminal leucine-rich repeat domains, which allow them to detect these bacterial peptidoglycans. They are localized in the cytosol bound to the membranes of early endosomes and interact with the actin cytoskeleton, acting as innate immunity “sensors” towards pathogen-derived components, triggering downstream production of proinflammatory cytokines via activation of NF-κB in response to bacterial stimulants. Interestingly, members of the SLC46 family were also shown to be involved in muropeptide uptake and NOD signaling, for example, in mammalian keratinocytes (1495, 1496) (see the description of SLC46A2).

SLC15A4 also plays a key role in the integration of metabolic and inflammatory signals (1497). SLC15A4 localizes to the endolysosomal compartment with LAMP1 (Lysosome-associated membrane protein 1) and LAMTOR (Late endosomal/lysosomal adaptor and MAPK and mTOR activator) and associates with the Ragulator and mTORC1 supercomplex consisting of mTORC1 and Rag proteins (1487) (see the description of SLC38A9). Critical for mTORC1 activity is endo-lysosomal acidity, the regulation of which is one of the important roles of SLC15A4. In addition, SLC15A4 mediates the stabilization of mTORC1 at the endolysosomal membrane as it associates with the ragulator components (1479). Therefore, loss of SLC15A4 function destabilizes Ragulator-mTORC1 complex formation, resulting in decreased mTORC1 activity.

Thus, SLC15A4, which binds TASL, regulates the TLR signaling pathway and acts as an important metabolic regulator. In addition, TASL has been reported to affect the behavior and function of tumors such as melanoma and breast cancer. This suggests that SLC15A4-TASL complex is also involved in cross-talk between tumor immune and metabolic pathways (1498, 1499).

Slc15a4 was originally cloned from rat brain and Slc15a4 mRNA was detected by in situ hybridization throughout the brain (1500), with particularly strong signals found in the hippocampus, choroid plexus, cerebellum and pontine nucleus. The human homologue was cloned from an intestinal cDNA library and Caco-2 cells (1501). Heterologous expression was achieved at the plasma membrane in Xenopus oocytes and studies revealed H⁺-dependent transport of ¹⁴C-histidine with a K_m of 17 μM (1500). Competitive inhibition of ¹⁴C-histidine uptake was observed for various di- and tri-peptides as well as carnosine as substrates. Analogous results were obtained with the human SLC15A4 paralogs using transiently transfected COS-7 cells, which were found to be able to express this endosomal/lysosomal transporter also in the plasma membrane (1502), and kinetic analysis of histidine transport revealed a K_m value in the low millimolar range (K_m = 0.16 mM) (1503). Thus, various expression systems have been used in an attempt to generate plasma membrane expression for functional analysis, but these studies have led to significant inconsistencies in published work with respect to substrate selectivity, transport kinetics, and pH dependence, as SLC15A4 transport functions may be affected by the lysosomal environment. In addition, experimental validation and detailed kinetic analysis for SLC15A4 substrates is still incomplete and has been hampered by the native expression of SLC15A4 in the endosomal-lysosomal environment.

Given the critical role of SLC15A4 in systemic inflammatory and autoimmune diseases, significant efforts are underway to identify molecules capable of modulating SLC15A4. Until recently, the pharmacological development of SLC15A4 inhibitors was limited by the lack of appropriate functional assays and scarce structural information. However, new developments are beginning to emerge. A chemoproteomic strategy has led to the development of the SLC15A4-related inhibitor AJ2-30 with anti-inflammatory activity (1504). An SLC15A4-mediated NOD2 activation reporter assay was used to monitor AJ2-30-mediated inhibition of transport function. Interestingly, AJ2-30 showed no activity against the closely related SLC15A3 transporter, which also transports NOD2 ligands. AJ2-30 has not yet been directly tested in a classical transport assay.

To further advance the pharmacological development of inhibitors and to address the inconsistencies reported on the functional properties of SLC15A4, new assays based on solid-supported membrane electrophysiology (SSME) and microscale thermophoresis using isolated lysosomal membranes were developed, allowing the protein to be studied in its native environment (1505). Using this approach, first recordings of electrophysiological properties and direct evidence for H⁺ cotransport by SLC15A4 were obtained. In addition, assessments of SLC15A4 substrate selectivity and transport kinetics for L-histidine and the identified SLC15A4 substrates L-arginine, L-lysine, His-Leu and Leu-Leu were performed (1505).

Understanding how SLC15A4 recruits TASL at the molecular level is key to developing therapeutics against SLE and other autoimmune diseases. In this regard, the cryo-EM structure of SLC15A4 stabilized in an outward-open conformation, as well as a model of the SLC15A4/TASL complex in which the first 16 N-terminal TASL residues fold into a helical structure that binds to the central cavity of SLC15A4 in the inward-open conformation, have been reported. These structures provide important insights into the molecular basis of SLC15A4/TASL-mediated type I interferon production (1506). Additional analysis of cryo-EM structures of SLC15A3 and SLC15A4 in their apo (outward-facing), substrate-bound, and TASL-bound (inward-facing) states revealed detailed structural mechanisms of substrate and TASL recognition (1507).

As mentioned above, SLC15A4 is also highly expressed in the choroid plexus. There it may be responsible for the removal of neuropeptide degradation products from the cerebrospinal fluid, similar to PepT2 (SLC15A2) (1474, 1494) (see Fig. 11).

SLC15A3: SLC15A3 plays an important role in TLR-mediated inflammatory responses as well (1508). Similar to SLC15A4 (see above), SLC15A3 is involved in the recognition of microbial pathogens by TLRs and NOD-like receptors and mediates the transport of bacterial peptidoglycans across the endolysosomal membrane (1509). Analogous to SLC15A4, SLC15A3 facilitates the transport of certain bacterial peptidoglycans, such as muramyl dipeptide (MDP), which are NOD2 ligands (1509).

SLC15A3 has been shown to be highly expressed in macrophages and monocytes (1510, 1511). In addition, the HPA suggests that SLC15A3 is most highly expressed in Müller and Schwan glial cells, in addition to monocytes and macrophages (including Hofbauer and Kupffer cells), and at lower levels also in dendritic and Langerhans cells. Müller glial cells are critical modulators of the retinal immune response by expressing receptors for cytokines and by releasing cytokines to regulate inflammation (1512). Analysis of diabetic mouse retinas revealed a significant increase in Slc15a3 expression associated with the pathogenesis of diabetic retinopathy, including microglial activation (1513). Schwann cells have also been shown to be immunocompetent (1514). Furthermore, in a mouse genome-wide transcriptional profiling study examining altered gene expression in the Schwann cell signaling network during sciatic nerve regeneration, there was significant upregulation of Slc15a3 (1515).

Interestingly, the newly identified partner of SLC15A4, TASL, which is required for endosomal TLR signaling, does not interact with SLC15A3 (1516).

SLC15A3 has been shown to be transcriptionally activated by the transcription factors p65 and HIF1α through direct binding to the SLC15A3 promoter (1517). Activation by p65 and HIF1α has been reported to contribute to poor outcomes in ischemic stroke associated with systemic inflammation. The reason is that it promotes microglial cells to polarize toward the proinflammatory M1 phenotype, thereby contributing to poor outcomes in ischemic stroke associated with systemic inflammation.

The rather exclusive expression of SLC15A3 in immunocompetent cells makes it a promising therapeutic target for the treatment of inflammatory diseases (1497, 1510). SLC15A3 has been shown to be associated with inflammatory diseases such as Crohn disease (1518), SLE (1519), and STING-associated vasculopathy of infancy (1520), where it regulates the inflammatory signaling of NOD2, MAVS (mitochondrial antiviral signaling protein), and STING (stimulator of interferon genes), respectively. Targeting SLC15A3 may offer advantages as it may have only moderate side effects.

Clinical relevance and pharmaceutical aspects

The role of PepT1 (SLC15A1) in drug delivery, including the prodrug strategy, has been reviewed elsewhere (43).

SLC15A3 and SLC15A4 are promising therapeutic targets for the treatment of inflammatory diseases (1497). The development of specific SLC15A4 inhibitors is a promising therapeutic strategy for the treatment of autoimmune diseases such as SLE and IBD. SLC15A4 has also been shown to control endolysosomal TLR responses by recruiting the innate immune adaptor TASL (1493), supporting SLC15A4-TASL targeting as a potential therapeutic strategy for SLE and related diseases (1506, 1507, 1521). As mentioned above, a chemoproteomic approach to developing a therapeutic strategy for patients with autoimmune diseases such as SLE and IBD led to the development of SLC15A4 inhibitors with anti-inflammatory activity (1504).

Orphan transporter family member (1)

SLC15A5

SLC16 Monocarboxylate transporter family (2.A.1.13/MFS_1/MFS)

Discovery: The H⁺-coupled pyruvate and lactate transporter MCT1 (SLC16A1) is the founding member of this family. It was identified as a mevalonate transporter by expression cloning from a mutant CHO (Chinese hamster ovary) cell line that exhibited enhanced mevalonate uptake, and the cDNA was isolated from these cells selected for growth in low concentrations of mevalonate when synthesis is blocked (160). It was subsequently shown that the wild-type version of the identified transporter corresponds to an H⁺-coupled pyruvate and lactate transporter, which was named MCT1 (SLC16A1) (1522).

Gene family members (14):
SLC16A1 (MCT1)	SLC16A2 (MCT8)	SLC16A3 (MCT4)	SLC16A4 (MCT5)
SLC16A5 (MCT6)	SLC16A8 (MCT3)	SLC16A11 (MCT11)	SLC16A14 (MCT14)
SLC16A6 (MCT7)	SLC16A9 (MCT9)	SLC16A12 (MCT12)
SLC16A7 (MCT2)	SLC16A10 (TAT1/MCT10)	SLC16A13 (MCT13)

Molecular aspects, physiological roles and links to disease

Of the 14 members of the SLC16 family, MCT1 (SLC16A1), MCT2 (SLC16A7), MCT3 (SLC16A8), and MCT4 (SLC16A3) are H⁺-coupled plasma membrane transporters of glycolysis products such as lactate and pyruvate, and ketone bodies (acetoacetate, β-hydroxybutyrate) across cell membranes. They provide electroneutral co-transport of monocarboxylates along with protons in a stoichiometric ratio of 1:1. Other SLC16 family members such as MCT7 (SLC16A6), MCT8 (SLC16A2), TAT1 (SLC16A10), MCT12 (SLC16A12) and MCT13 (SLC16A13) mediate facilitative transport of zwitterionic or amphipathic metabolites such as taurine, thyroid hormone, aromatic amino acids, creatine and oligopeptides. They function as either influx or efflux transporters depending on the substrate concentration gradient. Several orphan transporters still remain in this family (MCT5 (SLC16A4), MCT6 (SLC16A5), MCT9 (SLC16A9) and MCT14 (SLC16A14)), which have demonstrated important pathophysiological roles.

The SLC16 family belongs to the Monocarboxylate Transporter (MCT) family (TC 2.A.1.13), which is part of the MFS superfamily. All family members are predicted to have 12 TMHs with intracellular C- and N-termini and a large intracellular loop between TMH6 and TMH7 (1523). The SLC16 transporters are involved in a wide range of metabolic pathways (1524), including energy metabolism of brain, skeletal muscle, heart and tumor cells, T lymphocyte activation, intestinal metabolism, thyroid hormone metabolism and spermatogenesis.

Members of the MCT family are not themselves glycosylated, which has implications for their expression in the plasma membrane. MCT1-4 and likely also other SLC16 family members therefore associate with glycosylated ancillary proteins to facilitate proper membrane expression (1524). The non-glycosylated MCTs form heterodimers with the highly N-glycosylated ancillary glycoproteins basigin (BSG/CD147) or embigin (EMB/GP70) (1523, 1525, 1526). These are immunoglobulin proteins of the immunoglobulin (Ig) superfamily and type 1 transmembrane proteins with a single TMH, a short intracellular C-terminal domain, and a large extracellular highly glycosylated N-terminal domain. As described in Section 10 (SLC-Like Proteins), the XKR8 lipid scramblase requires BSG as an ancillary chaperone as well. In fact, there might be regulatory overlaps between these proteins in specific cell types, such as cancer and immune cells, where BSG is highly expressed. The BSG chaperone differs from the trafficking chaperones of the SLC7-family of non-glycosylated amino acid transporters, which require the SLC3 family type 2 membrane glycoproteins for plasma membrane expression (type 2 membrane proteins have a short intracellular N-terminus and a large highly N-glycosylated extracellular C-terminal domain; see the SLC3 family description).

The previously reported phylogenetic grouping (1524) has been updated for the detailed description of the individual members of the SLC16 family (Fig. 27), which is as follows

Fig. 27. Phylogenetic tree of the SLC16 family.

1) Proton-linked monocarboxylate transporters

SLC16A1 (MCT1), SLC16A7 (MCT2), SLC16A8 (MCT3), SLC16A3 (MCT4) and orphan transporter SLC16A5 (MCT6)

2) Facilitative transporters of zwitterionic or amphipathic compounds and related orphan transporters

• a)Thyroid hormone SLC16A2 (MCT8) and aromatic amino acid transporter SLC16A10 (TAT1)
• b)Taurine transporter: SLC16A6 (MCT7)
• c)Creatine transporter: SLC16A12 (MCT12)
• d)Oligopeptide transporter: SLC16A13 (MCT13) and orphan transporter SLC16A11 (MCT11)
• e)Orphan transporter SLC16A4 (MCT5)
• f)Orphan transporters SLC16A9 (MCT9) and SLC16A14 (MCT14)

1) Proton-coupled monocarboxylate transporters

SLC16A1 (MCT1), SLC16A7 (MCT2), SLC16A8 (MCT3), SLC16A3 (MCT4) and orphan transporter SLC16A5 (MCT6). The predominant role of MCTs 1-4 is the transport of L-lactate, pyruvate and the ketone bodies β-hydroxybutyrate and acetoacetate into and out of cells, with L-lactate being the most important substrate (1524). Since it is lactic acid and not lactate that is produced and used in metabolism, the ability of MCTs to transport lactate with a proton is ideally suited to its metabolic role (1524). (1527).

SLC16A1: MCT1 (SLC16A1) is a widely expressed H⁺-linked monocarboxylate transporter expressed in the plasma membrane of cells. It transports a variety of monocarboxylates including lactate, pyruvate, acetate, and the ketone bodies acetoacetate and β-hydroxybutyrate. It functions via alternating outward-open and inward-open conformational states. According to the cryo-EM structure of MCT1 (1528), there are three critical amino acid residues, K38, D309 and R313, required for H⁺ recognition, all of which are conserved among the H⁺-coupled MCTs (MCT1, MCT2, MCT3 and MCT4). The protonation and deprotonation of D309 is particularly important for the conformational transition during the transport process.

As mentioned in the final paragraph of the SLC16 family description on the clinical relevance and pharmaceutical aspects, MCT1 (SLC16A1) together with MCT4 (SLC16A3) are crucial for tumor metabolism and have recently become significant targets for anti-cancer drug development (1529). Specifically, lactate released by hypoxic cancer cells via MCT1 (SLC16A1) can be taken up by oxidative cancer cells via MCT4 (SLC16A3), which perform aerobic glycolysis to fuel oxidative phosphorylation (1530).

Slc16a1 knockout mice are embryonically lethal but Slc16a1^+/- mice developed normally (1531). However, when fed high fat diet, they displayed resistance to development of diet-induced obesity, as well as less insulin resistance and no hepatic steatosis as compared to wt littermates. The enhancement in expression of several genes involved in lipid metabolism in the liver of wt mice under high fat diet was prevented in the liver of Slc16a1^+/- mice. These findings highlight the importance of MCT1 in the regulation of energy balance, especially during exposure to an obesogenic diet.

Lactate and ketone bodies are important respiratory substrates for some tissues including heart, skeletal muscle and brain. Thus, lactate released from skeletal muscle during exercise and ketone bodies derived from fatty acid metabolism in the liver are transported from the blood across the BBB into the brain by MCT1 and then utilized by the brain as alternative energy metabolites (1532). In addition, in heart and skeletal muscle, the uptake of ketone bodies may be facilitated by MCT1 (1524). This is important because trained muscle is able to utilize fatty acids and ketone bodies more efficiently, reducing muscle glycogen depletion and lactate production and delaying the onset of fatigue. MCT1 delivers lactate as part of the lactate shuttle to promote mitochondrial biogenesis and increase TCA flux (1533, 1534). Using a mouse model with specific deletion of Slc16a1 in skeletal muscle, it was shown that the MCT1-mediated lactate shuttle also has an active role in promoting mitochondrial biogenesis and TCA flux, in addition to the known function of lactate as an energy fuel to feed the TCA cycle (1534).

In the brain, there are predominantly three MCT paralogs, each of which has a distinct cellular distribution (1535). MCT1 is expressed by endothelial cells of the BBB, astrocytes and oligodendrocytes. At the BBB, MCT1 plays an important role in the delivery of lactate and ketone bodies into the brain, particularly when blood glucose availability is reduced (1536) (see Fig. 33). MCT4 expression is restricted to astrocytes, whereas MCT2 is found almost exclusively in a subset of neurons, including their postsynaptic sites, where it may affect synaptic transmission (1537). This expression pattern is consistent with the concept that lactate is released from astrocytes via MCT1/MCT4 and taken up by neurons via MCT2 (1535). MCT1 and MCT2 are also present on the apical (CSF-facing) side of choroid plexus epithelial cells, whereas MCT4 is present on the basal side of the choroid plexus epithelium (1535) (Fig. 11). Maintaining adequate levels of lactate in the CSF is critical and increases with age and the pathogenesis of Parkinson disease, while it is decreased in patients with dementia (1535). It has been suggested that MCT1 facilitates the transport of lactate from the choroid plexus epithelial cells into the CSF, whereas lactate is transported from the CSF into the choroid plexus epithelial cells via MCT2 (SLC16A7) and released across the basolateral membrane via MCT4 (1535).

In contrast to glycolytic cells that produce lactic acid, other cells use lactic acid as a substrate for lipogenesis and gluconeogenesis. Tissues that carry out such processes, such as liver, kidney tubules, and adipose tissue, express either or both MCT1 and MCT2, depending on the species (1524). In hepatocytes, MCT1 is expressed at significant levels in the plasma membrane on the sinusoidal side, based on studies in mice (1538). Thus, in hepatocytes, MCT1 may either transport L-lactate into hepatocytes for gluconeogenesis, especially after exercise, or mediate the release of lactate and ketone bodies from the liver, which are then transported from the blood to the brain, skeletal muscle, or heart muscle (1532).

One of the metabolic signals important for food anticipation under caloric restriction is β-hydroxybutyrate. Based on liver-specific knockout studies in mice, Mct1 has been implicated in food anticipation activity (FAA). Lack of MCT1 in the liver, but not in neuronal or glial cells, was shown to reduce FAA in mice and this was associated with a reduction in blood β-hydroxybutyrate levels (1539). This finding underscores the importance of food anticipation of liver-derived β-hydroxybutyrate under caloric restriction with subsequent delivery to the blood via MCT1.

In the digestive tract, MCT1 is highly expressed in the colon, where it localizes to the basolateral membrane together with BSG/CD147 (891, 1538), and at lower levels, MCT4 (SLC16A3) also appears to be expressed there, according to the HPA. Here, the MCTs participate in the absorption of short-chain fatty acids (SCFAs) such as acetate, propionate, and butyrate, which are produced by the bacterial fermentation of dietary fiber (1540). MCT1 and MCT4 mediate the electroneutral exit of SCFA in the basolateral membrane with a H⁺ to SCFA stoichiometry of 1:1. Transepithelial SCFA transport is coordinated with the high-affinity, low-capacity Na⁺-coupled monocarboxylate transporter SMCT1 (SLC5A8) located in the brush border membrane in colonic cells (at least in rodents; see the description of SLC5A8). Once transported across the intestinal epithelium into the bloodstream, SCFA can play critical roles in host metabolism with beneficial effects on preventing the development of metabolic disorders and neurodegenerative diseases (1541, 1542). In addition, SCFA can serve as an effective fuel for the failing heart, outperforming ketone oxidation (1543).

In addition, MCT1, and to some extent also MCT4 (SLC16A3), is significantly expressed in the basolateral membrane of the proximal small intestine, particularly in the duodenum, where low-affinity, high-capacity SMCT2 (SLC5A12) is present in the brush border membrane (1538, 1544). Since bacterial fermentation in the small intestine is minimal, SMCT2 and MCT1/MCT4 are likely responsible for the uptake of acetate, lactate and other monocarboxylates from food sources, including milk and fermented products.

The kidney has a high rate of reabsorption of monocarboxylates such as lactate from the glomerular filtrate. The situation is similar to that of the Na⁺-glucose cotransporters, where low-affinity, high-capacity SGLT2 (SLC5A2) in the early part (S1 segments) of the proximal convoluted tubules mediates most of the glucose reabsorption, and high affinity, low-capacity SGLT1 (SLC5A1) in the distal part (S3 segment) of the proximal tubules absorbs the remaining glucose, in both cases with GLUT2 (SLC2A2) in the basolateral membrane, thereby reducing urinary glucose to low levels (see Fig. 14). Similarly, low-affinity, high-capacity SMCT2 (SLC5A12) in the proximal tubule (in this case in all three segments S1, S2 and S3) and high-affinity, low-capacity SMCT1 (SLC5A8) in more distal part of the proximal tubules (S3 segments) facilitate the uptake of monocarboxylates across the apical membrane (896). MCT1 then acts as an efflux pathway for the reabsorbed lactate into the blood. It may also take up lactate or pyruvate from the circulation for gluconeogenesis and β-oxidation (1538). Based on immunohistochemical studies in mice, basolateral MCT1 was found to be expressed in the most proximal part of the proximal convoluted tubules (S1) in the mouse kidney (1538), where lactate reabsorption is likely coordinated with SMCT2 in the apical membrane. It is currently unclear whether an additional basolateral monocarboxylate transporter is present in the S3 segments, possibly the orphan transporter MCT5 (SLC16A5), which according to the HPA is highly expressed in renal tubules, or MCT4 (SLC16A3), which is also known to be highly expressed in renal tubules.

In the retina, lactate is one of the major nutrients for rods and cones of photoreceptor cells (1538). MCT1 and MCT3 are localized in the apical and basolateral membranes, respectively, of the retinal pigment epithelium (RPE) and facilitate the transfer of H⁺-lactate across the pigmented epithelium between the outer retina and the blood vessels. Excess lactate may also be extruded across the pigment epithelium via these transporters under certain conditions (1538).

MCT1 plays a critical role in spermatogenesis and knockout studies of MCT1 have shown a complete absence of spermatozoa (1545). In the female genital tract, MCTs are expressed throughout, and the granulosa is a predominant expression site of MCT1 according to findings in the mouse ovary (1538, 1546). Oocytes can consume lactate by converting it to pyruvate. Therefore, similar to testicular Sertoli cells and spermatogenic cells, the lactate shuttle is established between cumulus cells and oocytes.

In its role in cellular metabolism and energy supply, MCT1 is also expressed in activated B cells where it mediates monocarboxylate transport functionally coupled to antibody production (1547). MCT1 deficiency reduced cell proliferation of activated CD8⁺ T cells and derailed their cellular metabolism. MCT1 expression levels were found to be significantly elevated in B cells from patients with SLE, and transporter deficiency ameliorated the symptoms in a murine lupus model. Inhibition of MCT1 effectively blocked the exceptionally rapid phase of T cell division that is key to an effective immune response during T cell activation. Thus, MCT1 was found to be an attractive target for immunosuppressive therapy (1547).

As mentioned above, ketone metabolism is an important source of energy for many tissues during fasting. Consistent with this, mutations in the SLC16A1 gene are the cause of monocarboxylate transporter 1 deficiency, a rare disorder that causes recurrent ketoacidosis triggered by fasting or infection (1548) with vomiting, dehydration, Kussmaul breathing, and decreased consciousness. The mean lactate transport activity of heterozygous carriers of inactivating SLC16A1 mutations, both symptomatic and asymptomatic, was significantly reduced and all patients presented with episodes of ketoacidosis in the first years of life. Patients with homozygous mutations had a more severe phenotype with earlier onset of disease, more severe ketoacidosis, developmental delay and an increased prevalence of epilepsy.

The frequency of ketoacidosis episodes decreased over time with complete resolution by the age of seven years. In addition, a 28-month-old female patient with recurrent ketoacidosis and hypoglycemia due to a homozygous pathogenic variant in SLC16A1 presented with seizures. This was the first report of neuroimaging findings in MCT1. Mutations in SLC16A1 cDNA have also been reported to cause defects in lactate transport in erythrocytes (1549), and the SLC16A1 polymorphism rs1049434 was found to be associated with the incidence of muscle injuries in elite soccer players (1550). Furthermore, gain-of-function SLC16A1 promoter mutations have been reported and shown to be associated with hyperinsulinism by inducing SLC16A1 expression in pancreatic β-cells as a result of failed silencing of SLC16A1 (1551).

SLC16A7: MCT2 (SLC16A7), which has a higher affinity for pyruvate and lactate than MCT1 (SLC16A1), is expressed in tissues such as neurons that have a high demand for lactic acid for use as a respiratory fuel (1524, 1552). MCT2 also transports ketone bodies such as β-hydroxybutyrate and acetoacetate with relatively high affinity (1553).

MCT2 is highly expressed in numerous cancers including skin, pancreatic, lung, breast, esophageal and prostate cancer (1553, 1554). Chemical analogs of α-ketoglutarate that can enter cells via MCT2 have been designed to explore the MCT2 pharmacophore and to develop compounds that can interfere with intracellular targets associated with cell proliferation and tumor survival (1553).

In parallel, the SLC16A7 gene transcription has been shown to respond with high sensitivity to hypoxia, intracellular pH, and lactate (1555). Such responses amalgamate key metabolic stimuli to further malignancy, as in glioblastoma, and silencing of such factors would be expected to lead to metabolic strategies to reprogram cancer metabolism, for example by forcing intracellular pH to return to normal levels.

In neurons, MCT2 (SLC16A7) is a key component of astrocyte-neuron cross-talk and a link between metabolism, cortical structure, and state-dependent brain function (1556). The current concept is that neuronal activity is fueled by an activity-dependent lactate transfer from astrocytes, known as the astrocyte-neuron lactate shuttle (ANLS). In this ANLS, energy demands triggered by neuronal activity increase astrocytic glucose uptake and glycolytic metabolism. As a result, lactate is produced and transferred to the neuron for ATP production. Thus, ANLS requires glucose uptake by astrocytes via GLUT1 (SLC2A1) and astrocytic monocarboxylate transporters MCT1 (SLC16A1) and MCT4 (SLC16A3) to deliver lactate to the neuron, which in turn takes it up via neuronal MCT2 (SLC16A7).

MCT2 (SLC16A7) enables the insulin-independent uptake of ketone bodies, such as β-hydroxybutyrate, into neurons. This process is crucial for energy metabolism during midlife, when insulin resistance can hinder glucose utilization (1557). Brain insulin resistance is a critical feature of aging, obesity, type 2 diabetes (T2D), and Alzheimer disease (1558, 1559). It disrupts cognition, memory, and metabolic regulation. Studies have reported that MCT2-mediated ketone body uptake is particularly crucial during midlife (1557). This period has been identified as a “critical window” beginning around age 44, when insulin resistance in the brain starts to impair glucose metabolism (1557). MCT2 helps maintain neuronal function and stability during this vulnerable phase by facilitating alternative energy uptake. Thus, increasing ketone body production provides an alternative energy source for the brain and could stabilize neural networks and mitigate age-related decline if implemented during the midlife “critical window.” These findings underscore the potential of targeting metabolic pathways, such as enhancing MCT2 function and adopting ketogenic dietary strategies, to promote brain health and counteract age-related changes (1557).

SLC16A8: MCT3 (SLC16A8) is highly expressed in the basolateral membrane of the RPE where it contributes to the regulation of pH and lactate concentrations in the outer retina, together with MCT1 (see above) (1560).

SLC16A3: MCT4 (SLC16A3) is most highly expressed in skeletal muscle, and to a lesser extent in astrocytes, white blood cells, chondrocytes, and cancer cells. It plays an important role in the efflux of lactate and other monocarboxylates from cells (1524).

Lactate produced by muscle during exercise is an important end product of glycolysis that enters the lactate shuttle. According to the shuttle concept, lactate produced by glycolytic muscle fibers is subsequently used by oxidative muscle fibers (1561). MCT4, expressed in glycolytic muscle fibers facilitates lactate efflux. The lactate shuttle hypothesis was described several decades ago (1561) and lactate was thought to be the cause of fatigue in exercising muscles. To gain further insight in this, mice with global deletion of Mct4 (Slc16a3^-/-) or muscle-specific deletion of the accessory protein BSG/CD147 (Bsg^-/-) were generated (1562). Mice with both knockouts showed normal muscle morphology and contractility. However, Slc16a3^-/- mice exhibited an exercise intolerant phenotype. In vivo measurements of compound muscle action potentials showed a decrease in the evoked response in the Slc16a3^-/- mice, and the studies indicated that increased lactate was not the cause of muscle fatigue during exercise. Rather, it is thought that the observed exercise intolerance is related to a functional and structural impairment of the neuronal component of the motor unit, with α motor neurons and neuromuscular junctions being most affected (1562). To further elucidate the precise contributions of MCT to exercise-induced muscle fatigue, skeletal muscle-specific knockout models will be required.

Subsequently, pharmacological MCT inhibitors were used in mice to investigate the role of MCTs in exercise duration (1563). MCT1 inhibition by administration of α-cyano-4-hydroxycinnamate, a potent and non-competitive inhibitor of monocarboxylate transporters, significantly shortened treadmill running time and increased blood lactate concentration immediately after exercise. Muscle lactate concentration was also increased and muscle glycogen content was decreased. MCT4 inhibition by bindarit administration reduced treadmill time to an even greater extent. Bindarit administration also increased muscle lactate, but did not alter blood lactate and glucose concentrations or muscle glycogen content immediately after exercise. The data highlight the critical role of MCT1 and MCT4 in exercise endurance, though, as the authors of this study note, the data should be considered with some caution because the inhibitors used are not completely specific for these MCTs (1563).

As already mentioned, MCT4 is thought to play an important role to shuttle L-lactate between astrocytes and neurons in the brain (508, 509). In astrocytes, glucose can be converted directly to L-lactate by glycolysis or stored in the form of glycogen, and as neuronal activity intensifies, astrocytic glycogen is mobilized to supply neurons with lactate when neuronal glucose is insufficient to meet energy demands.

MCT4 (SLC16A3) is thought to participate in the uptake of monocarboxylates in the gastrointestinal tract where it is located in the basolateral membrane (see the MCT1/SLC16A1 description).

In addition, MCT4 is highly expressed in metastatic tumors, where high expression significantly correlates with aberrant cell proliferation, invasion and distant metastasis, with poor prognosis in colorectal cancer, hepatocellular carcinoma, gastric cancer, prostate cancer, bladder cancer, etc. (1564). At these sites, extracellular lactate contributes to malignancy and immune response evasion due to MCT4 efflux function, while its deficiency results in intracellular accumulation of lactate with induced reactive oxygen species-dependent cellular apoptosis. Thus, there is increasing evidence that selective inhibition of MCT4 will provide promising clinical benefits (1564, 1565) (see below, “Clinical Relevance and Pharmaceutical Aspects”).

SLC16A5 – Orphan transporter: MCT6 (SLC16A5) is included here because it is phylogenetically most closely related to the classical H⁺-coupled monocarboxylate transporters. However, its transport function is still unclear. It is expressed in several tissues and has been suggested to transport drugs such as probenecid, which increases uric acid excretion in gout patients, nateglinide, which lowers blood glucose levels in T2D by stimulating pancreatic insulin secretion, and the diuretic bumetanide (1566, 1567). However, subsequent untargeted metabolomics studies identified the potential role of MCT6 in lipid and amino acid metabolism (1568). Additional studies have indicated that MCT6/SLC16A5 functions as a chloride-sensitive organic anion transporter under acidic conditions, associated with the auxiliary protein BSG/CD147 (1569). However, further work is needed to fully elucidate the function and physiological role of this transporter.

2) Facilitative transporters of zwitterionic and amphipathic compounds

a) Thyroid hormone SLC16A2 (MCT8) and aromatic amino acid transporter SLC16A10 (TAT1)

SLC16A2: MCT8 (SLC16A2) is a transporter of thyroid hormones (e.g., T3 and T4). It is highly expressed in the liver as well as in the microvessels of the brain, where it helps T3 cross the BBB (1570). MCT8 (SLC16A2) is also expressed throughout the proximal tubule on the basolateral side of epithelial cells, based on studies in mice, where it may contribute to thyroid hormone reabsorption (1571). SLC16A2 mutations lead to Allan-Herndon-Dudley syndrome, a brain development disorder that causes severe intellectual disability and movement problems (1572). Mutations in SLC16A2 are associated with elevated serum T3 levels and severe psychomotor retardation, consistent with a central role for MCT8 in brain development.

SLC16A10: TAT1 (SLC16A10) is a Na⁺-and H⁺-independent amino acid transporter that mediates the transport of aromatic amino acid across the basolateral membrane of epithelial cells. It may also enable hepatocytes to function as a sink that controls the extracellular aromatic amino acid concentration (1573). TAT1 also mediates the transport of aromatic amino acids across the plasma membrane of non-epithelial cells such as skeletal myocytes.

TAT1-transported amino acids such as tryptophan or tyrosine are known precursors of serotonin, catecholamines, and thyroid hormone. Thus the absence of TAT1 may affect neurotransmitter and thyroid hormone availability, leading to neurological disorders (1573).

Slc16a10^-/- mice grew and reproduced normally, showed no gross phenotype and no obvious neurological defects. The study of these mice, however, showed that TAT1 is essential for the control of extracellular aromatic amino acid homeostasis in mice (1573).

b) Taurine transporter SLC16A6 (MCT7)

SLC16A6: MCT7 (SLC16A6) is a facilitative taurine transporter expressed in the liver, brain, endocrine pancreas, and several cancer types, notably melanoma. It mediates taurine efflux under normal osmotic conditions, helping cells rapidly restore osmotic balance (1574). Taurine is a non-proteinogenic amino sulfonic acid essential for the function of the heart, skeletal muscle, retina, and nervous system. After rapid absorption from the small intestine, primarily via the Na⁺-coupled taurine transporter TAUT (SLC6A6) and the H⁺-coupled amino acid transporter PAT1 (SLC36A1), taurine is distributed through the bloodstream to peripheral tissues. Cellular taurine export relies largely on MCT7, including efflux from polarized intestinal epithelia (1574), a mechanism distinct from the hypo-osmotic activation of volume-regulated anion channels (VRACs) (1574).

The functional surface expression of MCT7 is regulated by auxiliary proteins such as basigin (BSG/CD147) and embigin (EMB/GP70) (1574). Unlike its H⁺-coupled MCT paralogs, MCT7 exhibits pH-independent taurine transport, a property likely explained by the absence of the conserved acidic residue (equivalent to D309 in MCT1) that is required for H⁺ recognition. Cryo-EM structures of MCT1 reveal three key residues, K38, D309, and R313, essential for H⁺ coupling (1528), and MCT7 lacks the position corresponding to D309 (1574).

A recent SLC superfamily interactome study further identified a phosphorylation-dependent degron motif in MCT7 that promotes ubiquitin-mediated degradation, demonstrating that post-translational modification can regulate the stability of this transporter (1575). Such control could, in principle, adjust taurine export under different metabolic or osmotic conditions, although this functional link was not directly tested.

c) Creatine transporter SLC16A12 (MCT12)

SLC16A12: MCT12 (SLC16A12) acts as a creatine transporter in the retina, and a heterozygous mutation in SLC16A12 causes a syndrome characterized by juvenile cataracts (1060, 1576). In the kidney, MCT12 is expressed in both proximal tubules and the thick ascending limbs of the loop of Henle, and localization studies indicate that MCT12 resides on the basolateral membrane. In contrast, another creatine transporter, CRT1 (SLC6A8), is expressed on the apical membrane (1576) (see the SLC6 family description). Like MCT7, MCT12 is characterized as a facilitative diffusion-type transporter, functioning as an influx or efflux transporter depending on the substrate concentration gradient (1577).

d) Oligopeptide transporter: SLC16A13 (MCT13) and orphan transporter SLC16A11 (MCT11)

SLC16A13: MCT13 (SLC16A13) is expressed in the liver and duodenum, and SLC16A13 is a susceptibility gene for T2D (1578). MCT13 functions as an efflux transporter of oligopeptides and peptidomimetics such as the β-lactam antibiotic cephradine (1527). In polarized intestinal epithelia, MCT13 was mainly localized to the basolateral membrane and functioned as a substrate releaser. The expression of MCT13 in the plasma membrane is facilitated by interaction with the auxiliary protein BSG/CD147 (1527).

Intestinal absorption of di- and tripeptides and peptide-like drugs such as β-lactam antibiotics (including cephradine) is mediated by the intestinal H⁺-coupled oligopeptide transporter PepT1 (SLC15A1) from the lumen into enterocytes (see the SLC15 family description). While oligopeptides are cleaved into single amino acids inside enterocytes, MCT13 may be well suited for basolateral release of peptide-like drugs such as β-lactam antibiotics into the blood. However, further studies are still required with a more complete understanding of the substrate specificity of MCT13 to see for which peptidomimetics this basolateral release can be mediated (1527). In particular, which oligopeptides and/or β-lactam antibiotics are transported by MCT13, and whether this is limited to zwitterionic compounds or whether net-charged compounds are also transported requires further experimental analysis.

SLC16A11 - Orphan transporter: The precise function of MCT11 (SLC16A11), which is closely related to SLC16A13, is still unknown. Studies have shown that mutations of SLC16A11 contribute to the pathogenesis of T2D (1579, 1580). T2D affects Latinos at twice the rate of populations of European ancestry, and a risk haplotype spanning SLC16A11 has been identified that explains ~20% of the increased T2D prevalence in Mexico. The T2D risk haplotype is associated with five coding variants, most of which are missense mutations that result in reduced expression of SLC16A11 at the plasma membrane due to decreased affinity for the auxiliary protein BSG, and/or increased affinity for proteosomal proteins. In addition, pyruvate transport was suggested to be reduced by 50% in cells with the risk haplotype (1581). In primary human hepatocytes, knockdown of SLC16A11 using siRNAs resulted in higher intracellular concentrations of acylcarnitines, diacylglycerols and triacylglycerols (1580, 1581). The accumulation of acylcarnitines indicates decreased fatty acid beta-oxidation in mitochondria. Further studies are needed to elucidate the intracellular consequences of reduced SLC16A11 activity in the pathogenesis of T2D (1580).

e) Orphan transporter SLC16A4 (MCT5)

SLC16A4 – Orphan transporter: MCT5 (SLC16A4) still does not have any known transport substrates and is therefore considered an orphan transporter (1529). According to the HPA, it is most highly expressed in the tubules of the kidney, among other tissues.

f) Orphan transporters SLC16A9 (MCT9) and SLC16A14 (MCT14)

SLC16A9 - Orphan transporter: MCT9 (SLC16A9) is an orphan transporter that is highly expressed in kidney and exocrine tissues. Recent evidence suggests that it mediates urate uptake in mammalian cells and that urate uptake is enhanced by heat shock (1582). Moreover, whole-exome sequencing revealed a rare missense variant in SLC16A9 in a pedigree with early-onset gout (1583). In addition, MCT9 has been proposed to function as a transporter of carnitine (987, 1584). To fully elucidate the function of MCT9, further work is still required.

SLC16A14 - Orphan transporter: MCT14 (SLC16A14) was found to be abundantly expressed in mouse CNS and kidney, and it was hypothesized that it may function as a neuronal aromatic amino acid transporter (1585). Further validation of this finding is awaited.

Clinical relevance and pharmaceutical aspects with a focus on cancer treatment

As already highlighted in the SLC16A1 description, MCT1 (SLC16A1) together with MCT4 (SLC16A3) are crucial for tumor metabolism and are important targets for the development of anti-cancer drugs (1529). While normal cells derive their energy from oxidative phosphorylation, cancer cells derive it from oxidative glycolysis, the so-called “Warburg effect”. This results in the production of large amounts of lactate that must be rapidly effluxed across the cell membrane via H⁺-coupled transport by MCT4 (SLC16A3). Consistent with this, MCT4 (SLC16A3) inhibition induces accumulation of intracellular lactic acid and subsequent cell death (1586, 1587). Expression levels of MCT4 (SLC16A3) and ancillary glycoprotein BSG/CD147 correlate with worse prognosis in many types of cancer (1588). The development of drugs targeting specific MCT paralogs such as MCT4 is expected to provide novel approaches for cancer chemotherapy (1524).

In addition, lactate released by hypoxic cancer cells via MCT4 (SLC16A3) can be taken up by oxidative cancer cells via MCT1 (SLC16A1) (1530). These cells perform aerobic glycolysis to fuel oxidative phosphorylation. Therefore, therapeutic strategies targeting MCT1 are considered very promising for cancer treatment as well. Several MCT inhibitors have been investigated, including AR-C122982 (also known as SR13800) (1589), AZD3965 (a variant of AR-C155858) (1590), BAY-8002 (1591), and 7ACC2 (a potent inhibitor of both monocarboxylate transporters and the mitochondrial pyruvate carrier) (1528, 1592).

Structural studies of the MCT1-BSG complex revealed the mode of substrate recognition and H⁺-coupled transport and uncovered the action of anticancer drug candidates (1528). Based on cryo-EM structures of human MCT1 bound to lactate or inhibitors in the presence of the ancillary glycoprotein BSG, MCT1 exhibits similar outward-open conformations when complexed with lactate or the inhibitors BAY-8002 and AZD3965. Inward-open structures were captured in the presence of the inhibitor 7ACC2 or when the H⁺-coupling D309 was neutralized by mutating to asparagine. Building on the understanding of the transport mechanism, silybin, a prevalent flavonolignan in Silybum marianum extracts and a compound with potential antitumor properties, was identified as a selective MCT1 inhibitor (1593). Overall, these findings provide a molecular basis for the substrate recognition and transport mechanisms of MCTs, as well as for structure-guided drug discovery targeting MCTs (1593).

Orphan transporter family members (5):
SLC16A4 (MCT5)	SLC16A9 (MCT9)	SLC16A14 (MCT14)
SLC16A5 (MCT6)	SLC16A11 (MCT11)

SLC17 Organic anion and vesicular glutamate transporter family (2.A.1.14/MFS_1/MFS)

Discovery: The founding member of the SLC17 family is the rabbit renal phosphate transporter NaPi-1 (SLC17A1) (142). This transporter was later found to be an organic anion transporter that is also involved in apical uric acid export in human renal proximal tubuli, because a gain-of-function mutation of this transporter reduces the risk of renal underexcretion gout (1594).

Gene family members (9):
SLC17A1 (NPT1)	SLC17A4	SLC17A7 (VGLUT1)
SLC17A2 (NPT3)	SLC17A5 (Sialin)	SLC17A8 (VGLUT3)
SLC17A3 (NPT4)	SLC17A6 (VGLUT2)	SLC17A9 (VNUT)

Molecular aspects, physiological roles and links to disease

The SLC17 family comprises nine members, all of which are dedicated to anion transport, including the polyspecific anion exporters SLC17A1-4 (NPT1, NPT3, NPT4, SLC17A4), the anionic vesicular glutamate transporters SLC17A6-8 (VGLUT2, VGLUT1, VGLUT3), and a vesicular nucleotide transporter SLC17A9 (VNUT) (Fig. 28). These transporters belong to the Anion:Cation Symporter (ACS) family (TC 2.A.1.14), which is part of the MFS superfamily, whose members typically have 12 TMHs.

Fig. 28. Phylogenetic tree of the SLC17 family.

SLC17A1, SLC17A2, SLC17A3 and SLC17A4: NPT1 (SLC17A1), NPT3 (SLC17A2), NPT4 (SLC17A3) and SLC17A4 are polyspecific anion exporters located at the plasma membrane. They facilitate voltage-driven transport of several organic anions such as penicillin, probenecid, uric acid, p-aminohippuric acid and acetylsalicylic acid (1595, 1596). Therefore, these transporters could potentially contribute to the excretion of metabolites and anionic drugs.

Expression of the SLC17A1–4 proteins is relatively restricted. SLC17A1 is expressed in the apical membrane of the proximal tubule in the kidney (1597) and to a lesser extent the sinusoidal membrane of hepatocytes in the liver (1598). SLC17A3 is also expressed in the kidney and localized to the apical side of renal tubules (1599). SLC17A2 mRNA is expressed in muscle, liver, and kidney, and SLC17A4 is expressed in kidney, liver and intestine (1600).

Although these transporters are considered to be polyspecific anion exporters, their transport properties have only recently been characterized in greater detail. It is interesting to note that increased urate levels and gout prevalence have been associated with genetic variation in NPT1 (SLC17A1) and NPT4 (SLC17A3), in addition to GLUT9 (SLC2A9) and BCRP (ABCG2) (578). The kidney- and liver-specific expression of NPT1 and NPT4 is consistent with a role in urate transport and the development of gout.

Subsequent metabolomics, genomics and biochemical studies showed that there are many more anionic physiological substrates for the SLC17A1-4 group (1601). One of these of particular interest is N-lactoyl-phenylalanine (Lac-Phe), a metabolite produced by intense exercise (1602).

Lac-Phe mediates the effects of metformin on food intake and body weight (1603). The antidiabetic drug metformin has been shown to be a potent pharmacological inducer of the anorexigenic metabolite Lac-Phe. Metformin stimulates Lac-Phe biosynthesis by inhibiting mitochondrial complex I (NADH dehydrogenase), which induces a metabolic shift (1604). This results in an increase in glycolytic flux followed by an intracellular lactate mass action. Lac-Phe is then produced in cells that express the Lac-Phe biosynthetic enzyme carnosine dipeptidase II (CNDP2), such as proximal enterocytes. Genetic ablation of Lac-Phe biosynthesis in male mice rendered the animals resistant to the effects of metformin on food intake and body weight, and further studies identified Lac-Phe as a critical mediator of the weight-lowering effects of metformin (1603), acting as an appetite suppressant and protecting against obesity via G protein-coupled receptors (1605). The anionic membrane transporters SLC17A1/3 facilitate the renal excretion of this important signaling metabolite.

Hepatocyte nuclear factor 1 alpha (HNF1α, HNF1A) has been shown to upregulate the expression of renal SLC17A1 and SLC17A3 (1606). In particular, based on studies in mice, the Slc17a1 and Slc17a3 genes were expressed at reduced levels in the kidneys of Hnf1a^-/- mice (1606). It was concluded that HNF1α is required for optimal SLC17A1 and SLC17A3 transcription in the kidney, highlighting a critical role for HNF1α in the homeostasis of weight-lowering Lac-Phe and thus the action of metformin.

The SLC17A4 transporter has long been considered an orphan transporter. A genome-wide association study later found that genetic variation in the SLC17A4 locus is associated with free (unbound) T4 thyroid hormone concentrations (1607). Subsequently, functional studies confirmed that SLC17A4 facilitates the uptake of T3 and T4 thyroid hormones (1608). SLC17A4 is predominantly expressed in hepatocytes, enterocytes and colonocytes, suggesting a role for this transporter in the metabolic clearance and entero-hepatic cycle of thyroid hormones. Whether SLC17A4 is also able to transport glucuronidated thyroxine as part of thyroid hormone excretion remains to be determined (1608).

Genetic variation in the SLC17A4 locus has also been associated with the progression of elevated serum urate levels and gout (1609). However, there was only marginal induction of urate uptake in SLC17A4-expressing COS-1 cells, and excess urate up to 1 mM did not inhibit SLC17A4-mediated thyroid hormone uptake, suggesting that SLC17A4-mediated thyroid hormone transport would not be modulated by circulating urate (1608).

SLC17A5: Sialin (SLC17A5) is widely expressed and mediates the H⁺-coupled lysosomal exit of various substrates across the lysosomal membrane. It was functionally characterized in 1989 (1610) and cloned in 1999 (1611). Among these substrates is sialic acid, a sugar whose function has been implicated in cellular communication and structural/modulatory processes. Lysosomal degradation is the primary pathway for the catabolism of sialoglycoconjugates. Once delivered to the lysosome, sialic acid residues are sequentially removed by hydrolysis of terminal glycosidic linkages by acidic sialidases (neuraminidases). Sialin (SLC17A5) then exports the free sialic acid from the lysosome. Sialin (SLC17A5) is required for normal CNS myelination (1612).

Free sialic acid storage disorders (FSASDs) are a spectrum of autosomal recessive lysosomal storage disorders caused by the absence or malfunction of sialin (SLC17A5) (1613). The type of SLC17A5 variant determines the severity of the FSASD phenotype. Variants that lead to non-functional sialin protein result in severe infantile sialic acid storage disease, whereas variants that allow residual sialin function result in a milder phenotype, such as in Salla disease, which is inherited in an autosomal recessive fashion. Patients with Salla disease, the majority of whom have been identified in Finland, have moderate to severe psychomotor retardation, spasticity, ataxia, developmental delay, and neurocognitive impairment. They secrete large amounts of free (unbound) sialic acid and store 10 to 30 times the normal amount in various tissues and cultured fibroblasts (1611, 1614, 1615).

The feasibility of base editing as a novel therapeutic approach for the FSASD variant SLC17A5 c.115C>T has been demonstrated (1615). This progress highlights the utility of the base editing approach for the treatment of monogenic diseases caused by dysfunction of SLC solute carriers.

Sialin (SLC17A5) is the only member of the SLC17 family known to transport sialic acid but, as mentioned above, it also transports other substrates. These include acidic amino acids such as glutamate and aspartate as well as N-acetylaspartylglutamate (NAAG) (1616). NAAG acts as a neuromodulator of glutamatergic synapses by triggering activation of the presynaptic metabotropic glutamate receptor (1617). Transgenic studies in mice have shown that sialin is a major vesicular transporter for NAAG and thus plays a critical role in neuronal activity (1616).

To address the structural basis of how sialin (SLC17A5) transports substrates, cryo-EM structural analysis and molecular dynamics simulations were performed on human sialin in apo cytosol-open, apo lumen-open, NAAG-bound, and inhibitor-bound states (1618). Sialin adopts the classic MFS fold consisting of 12 TMHs with two pseudo-symmetric domains, the N domain (TMHs 1-6) and the C domain (TMHs 7-12). The study identified 1) a positively charged cytosol-open vestibule that accommodates either NAAG or the sialin inhibitor Fmoc-Leu-OH, while its luminal cavity likely binds sialic acid; and 2) key residues required for sialic acid or NAAG binding. Based on these findings, a working model of SLC17 family members in general was generated (1618). In short, sialin senses the proton in the lumen through its putative proton sensor, which is a cluster of residues in the N-terminal domain. The sialic acid then binds to the cavity open to the lumen. As the proton is translocated into the cytosol, sialin changes to the cytosol-open state and sialic acid is released into the cytosol.

SLC17A6, SLC17A7, SLC17A8: The three vesicular glutamate transporters VGLUT1 (SLC17A7), VGLUT2 (SLC17A6) and VGLUT3 (SLC17A8) play critical roles in excitatory neurotransmission by transporting glutamate down the electrochemical gradient into neuronal synaptic vesicles. The three VGLUTs share a high degree of sequence identity and have similar transport properties, but differ greatly in their distribution in the brain, suggesting different physiological roles (1619, 1620). VGLUT1 (SLC17A7) is expressed primarily in the cortex, VGLUT2 (SLC17A6) in the diencephalon and brainstem (1621, 1622), and VGLUT3 (SLC17A8) often by neurons associated with a different neurotransmitters (1623–1625). Consistent with their different distributions, genetic inactivation of each VGLUT paralog produces different neurological effects (1620, 1624, 1626–1628).

VGLUT2 (SLC17A6) is also expressed in the pancreas, as suggested by the HPA. Indeed, glutamate has been proposed to play a role in the release of insulin and glucagon from pancreatic cells via exocytosis, and VGLUT2 mRNA has been found to be expressed in both α- and β-cells (1629). However, how glutamate stimulates exocytosis, for example in insulin granules, how glutamate uptake by VGLUT2 into insulin granules is regulated, whether metabotropic glutamate receptors are involved and what role all this plays in diabetes and obesity is not yet clear (1630, 1631).

VGLUT3 (SLC17A8) also plays an important role in hearing, as highlighted by the finding that its dysfunction is responsible for DFNA25, an autosomal-dominant form of progressive, high-frequency nonsyndromic deafness (1632). The SLC17A8 missense mutation A211V was found to segregate with DFNA25 deafness. The A211 residue is conserved in VGLUT3 across species and in all human VGLUT subtypes (VGLUT1-3), suggesting an important functional role. In the cochlea, VGLUT3 accumulates glutamate in the synaptic vesicles of the sensory inner hair cells before releasing it onto receptors of auditory-nerve terminals and the A211V mutation leads to a lack of auditory-nerve responses to acoustic stimuli.

A subpopulation of specialized astrocytes has been identified that also express VGLUT1 or VGLUT2 (1633, 1634). These astrocytes were reported to have a discrete molecular signature similar to that of glutamatergic synapses, with a defined anatomical distribution and functional competence for VGLUT-dependent glutamate release. The results demonstrated that astrocyte glutamate exocytosis exists in the adult brain (1633) and that different groups of specialized astrocytes have distinct roles in brain function. For example, glutamatergic astrocytes contribute to the function of cortico-hippocampal and nigrostriatal circuits during normal behavior and pathological processes. The actions identified demonstrate the functional relevance of these specialized astrocytes, for example, in enhancing hippocampal memory, counteracting hyperexcitation during seizures, and overactivating the subthalamic nucleus in Parkinson disease. The study highlights the potential of these targets for CNS protective therapies.

A H⁺ electrochemical gradient across the synaptic vesicle membrane, generated by the vacuolar-type H⁺-ATPase (V-ATPase), provides the driving force for vesicular uptake of most neurotransmitters (e.g., monoamines and GABA) (1635, 1636). In contrast, vesicular glutamate uptake via VGLUTs uniquely relies on the membrane potential (Δψ) as the primary driving force, consistent with a facilitated diffusion mechanism (1637, 1638).

VGLUTs are allosterically regulated by H⁺ and Cl^- and exhibit an associated Cl^- conductance. These properties appear to coordinate VGLUT activity with the large ionic shifts that accompany the rapid recycling of synaptic vesicles driven by neural activity (1639).

The cryo-EM structure of rat VGLUT2 has been reported, revealing a structure-based mechanism for substrate recognition and allosteric activation by low pH and Cl^- (1640). Consistent with the structure-function arrangement of MFS transporters, it uses the alternating access mechanism and most of the 12 TMHs are distorted or kinked by proline and/or glycine. Because it transports negatively charged substrates, the central cavity of VGLUT2 has a positive charge. The results indicate that a potential permeation pathway for Cl^- intersects the glutamate binding site. These results also demonstrate how the activity of VGLUTs can be coordinated with large shifts in H⁺ and Cl^- concentrations during the synaptic vesicle cycle to ensure normal synaptic transmission.

Because analysis of the individual physiological roles of VGLUTs has been hampered by the lack of specificity of pharmacological tools, new molecular probes for VGLUTs were developed by raising several mouse monoclonal antibodies (1641). One antibody recognized an epitope spanning the three extracellular loops in both VGLUT1 and 2, but not in VGLUT3. The antibody binds from the luminal side and acts allosterically to inhibit VGLUT function, presumably by limiting conformational changes.

SLC17A9: VNUT (SLC17A9) is quite widely distributed and functions as an ATP transporter in lysosomes and other secretory vesicles (1642–1644). It is involved in various pathological processes (1645). Like VGLUTs, VNUT is driven by the electrochemical gradient (1646).

According to the HPA, VNUT (SLC17A9) is most highly expressed in hepatocytes, cholangiocytes, gastric mucus secreting cells, intestinal goblet cells and plasma cells, all cells with secretory function. VNUT (SLC17A9) is responsible for vesicular storage of ATP in ATP-secreting cells and for vesicular release of ATP to initiate purinergic signaling (1647).

Mice lacking VNUT show loss of vesicular storage and secretion of ATP in all secreting cells tested, and it has been shown that VNUT is an essential membrane component for the initiation of purinergic signaling (1647, 1648).

In the liver, VNUT mediates the release of ATP into the blood via lysosomal exocytosis. There, ATP activates purinergic signaling. VNUT-induced purinergic signaling has been shown to exacerbate glucose metabolism by reducing insulin sensitivity in the liver (1647). VNUT has been shown to play an important role in hepatic lipid metabolism and in the development of non-alcoholic steatohepatitis (NASH) because VNUT is a key player in postprandial triglyceride release leading to the progression of steatohepatitis. It has been concluded that under conditions of energy excess, purinergic signaling induced by VNUT in the liver leads to excess energy storage in the form of fat in the liver, thereby exacerbating metabolic diseases such as NASH (1647). Therefore, VNUT is proposed as a novel therapeutic target for intervention in metabolic diseases such as NASH.

Hepatocytes and cholangiocytes also constitutively release substantial amounts of ATP into the bile. This may involve VNUT-mediated lysosomal exocytosis of ATP into the bile. ATP is rapidly degraded by membrane-bound ecto-ATPases and ecto-5ʹ-nucleotidases to adenosine and P_i (1649), from where P_i can be reabsorbed via NaPi-IIb (SLC34A2) from primary hepatic bile (1650).

VNUT (SLC17A9) was further shown to load ATP into secretory vesicles in adrenal chromaffin cells, T cells, and zymogen granules of pancreatic cells (1651). SLC17A9 deficiency reduced lysosomal ATP accumulation, which impaired lysosomal function and led to T cell death (1651).

Moreover, VNUT has been reported to concentrate purine nucleotides such as ATP in synaptic vesicles of neurons and lysosomes of glial cells, allowing their release by exocytosis (1646), and it has been proposed that ATP release plays an important role in purinergic signaling, which also plays an important role in nociceptive pain transmission (1645).

Finally, VNUT plays a critical role in the induction of lysosomal cell death (1652): Lysosomes serve as the cellular recycling center and are filled with numerous hydrolases capable of degrading most cellular macromolecules, and permeabilization of lysosomal membranes and subsequent leakage of lysosomal contents into the cytosol leads to lysosomal cell death. Loss of SLC17A9 function was shown to result in cell death mediated mainly by lysosomal cathepsin proteases. Cell death induced by SLC17A9 deficiency could be rescued by transcription factor EB (TFEB), a helix-loop-helix leucine zipper transcription factor that acts as a master regulator of lysosomal biogenesis, exocytosis and autophagy (1652, 1653). SLC17A9, through its role in lysosomal ATP accumulation, also regulates cell viability by controlling the activity of cathepsin D, a major lysosomal hydrolase. Consistent with this, reducing lysosomal ATP by suppressing SLC17A9 function leads to impaired cathepsin D function and lysosomal cell death (1643).

Interestingly, TFEB also regulates the lysosomal nutrient-sensing complex, which includes the amino acid sensor SLC38A9 (1654) (see the SLC38A9 description). Briefly, to promote autophagy during nutrient starvation, TFEB functions to coordinate cellular endocytosis to drive assembly of the lysosomal nutrient-sensing complex machinery that tethers and reactivates mTORC1, the mammalian target of rapamycin complex 1 (1655). The cryo-EM structure of the lysosomal mTORC1–TFEB–Rag–Ragulator megacomplex was subsequently reported (1656).

Clinical relevance and pharmaceutical aspects

VGLUTs have potential as targets for the treatment of Parkinson disease (1657). As highlighted above, progress in the structural understanding of substrate and inhibitor binding greatly enhances the potential of these transporters for the development of novel CNS protective therapies. In addition, inhibition of glutamate release by VGLUTs expressed in specific subpopulations of astrocytes, as mentioned above, offers a potentially promising therapeutic strategy in hyper-glutamatergic brain disorders, as it does not directly block glutamatergic neurotransmission (1633, 1634).

The development of specific inhibitors for VNUT has been proposed to be an excellent therapeutic strategy to block purinergic chemical transmission and to control pain perception (1645).

Orphan transporter family members: N/A

SLC18 Vesicular amine transporter family (2.A.1.2/MFS_1/MFS)

Discovery: The rat vesicular reserpine-sensitive monoamine transporters VMAT1 (SLC18A1) and VMAT2 (SLC18A2) were identified by molecular cloning virtually at the same time in two different laboratories using different approaches (158, 159):

Approach 1 (158): This group’s approach was based on studies of the mechanism of MPP⁺ toxicity. Systemic administration of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) damages dopaminergic neurons in the substantia nigra, mimicking the pathophysiology of Parkinson disease. As was shown later (1658, 1659), MPTP is first converted to the active metabolite MPP⁺ by monoamine oxidase B in astrocytes. MPP⁺ is then released from astrocytes and transported into dopaminergic neurons via the dopamine transporter DAT (SLC6A3). MPP⁺ inhibits ATP production and stimulates the formation of superoxide radicals, the latter leading to the production of peroxynitrite, which damages specific proteins by oxidation and nitration, such as tyrosine hydroxylase, the rate-limiting enzyme in dopamine synthesis, and the presynaptic protein α-synuclein. Peroxynitrite also nicks DNA, which in turn activates poly(ADP-ribose) polymerase (PARP), which consumes ATP, and the ultimate result is energy failure due to MPP⁺-induced mitochondrial respiration blockade and cell death. Because Chinese hamster ovary (CHO) fibroblasts exhibit much greater sensitivity to MPP⁺ than rat PC-12 cells derived from a rat pheochromocytoma, the investigators used CHO cells as a sensitive host to identify the basis of PC-12 cell resistance to the toxin. After expressing cDNA clones from a PC-12 cell expression library in CHO cells, stable transformants were selected in the presence of 1 mM MPP⁺. A cDNA clone was then identified that conferred strong resistance to the toxin. The clone encoded rat VMAT1, which was responsible for the resistance in PC-12 cells. The resistance could be reversed by the adrenergic blocking agent reserpine, an antihypertensive and antipsychotic drug, further confirming the correctness of the identified clone. A transporter different from VMAT1 was subsequently identified by the same group (158) in rat brain (VMAT2), which showed 62% amino acid identity with VMAT1.

Approach 2 (159): The other group independently cloned rat VMAT2 using expression cloning in cells from the CV-1 cell line (derived from the kidney of the African green monkey). The cells were transfected with cDNA prepared from mRNA of rat basophilic leukemia (RBL) cells, which are known to take up serotonin and then store it in secretory granules containing a VMAT. cDNA clones were then identified that induced increased serotonin uptake in CV-1 cells. The clones responsible for this increased uptake encoded a vesicular transporter that functions to accumulate the neurotransmitter in acidic organelles. The sequence of the identified transporter was found to be identical to rat VMAT2.

Gene family members (4):
SLC18A1 (VMAT1)	SLC18A2 (VMAT2)	SLC18A3 (VAChT)	SLC18B1 (VPAT)

Molecular aspects, physiological roles and links to disease

The SLC18 family includes the closely related vesicular monoamine transporters VMAT1 (SLC18A1) and VMAT2 (SLC18A2), the vesicular acetylcholine transporter VAChT (SLC18A3), and the vesicular polyamine transporter VPAT (SLC18B1) (Fig. 29). They mediate H⁺-coupled antiport of positively charged amine neurotransmitters and hormones from the cytoplasm to secretory vesicles using the H⁺ gradient established by the V-Type ATPase that acidifies secretory vesicles (1660), which is the driving force for accumulation of positively charged neurotransmitters in the lumen of the secretory vesicles.

Fig. 29. Phylogenetic tree of the SLC18 family.

The SLC18 family belongs to the Drug:H⁺ Antiporter-1 (12 Spanner) (DHA1) family (TC 2.A.1.2), which is part of the MFS superfamily, and thus its family members possess the MFS fold and contain 12 TMHs arranged in two pseudosymmetrical halves. Each of these halves contains 6 TMHs with a primary binding site for neurotransmitters, polyamines and inhibitors located near the membrane center. It is thought that conformational changes driven by the electrochemical H⁺ gradient shift the binding site from one side of the membrane to the other, allowing neurotransmitter transport from the cytoplasm to the lumen of the synaptic vesicle as part of the alternating access mechanism.

SLC18A1, SLC18A2: VMAT1 (SLC18A1) and VMAT2 (SLC18A2) form the biogenic monoamine transporter group (Fig. 29). Biogenic monoamines include catecholamines (dopamine, epinephrine, and norepinephrine), histamine, and serotonin. They function as neurotransmitters in the central and peripheral nervous systems that are responsible for motor control, cognition, memory, mood, sleep, etc. (1660, 1661). To precisely control their quantitative release into extracellular spaces, such as synaptic clefts, and to prevent their premature degradation or potential damage to neurons, these monoamines must be stored in secretory vesicles. VMATs are responsible for the uptake of monoamines from the cytoplasm into storage vesicles (1660–1662). They also protect against neurotoxicants by sequestering them in the lumen of the vesicles. For example, they protect against the Parkinson disease-associated neurotoxicant MPTP by transporting its active metabolite, MPP⁺, into the vesicular lumen.

The two VMATs, VMAT1 (SLC18A1) and VMAT2 (SLC18A2), share a relatively high sequence similarity but have distinct tissue distributions: VMAT1 (SLC18A1) is preferentially expressed in large dense core vesicles of several neuroendocrine cells, including adrenal medullary chromaffin cells and enterochromaffin cells, whereas VMAT2 (SLC18A2) is distributed between both central and peripheral sympathetic monoaminergic neurons. These VMAT H⁺ antiporters concentrate single positively charged amines to high levels, up to 500 mM, within the secretory vesicles, exceeding the concentration found in the cytosol by 10,000-fold (1661). The vacuolar ATPase is responsible for generating the acidic luminal environment necessary to drive amine uptake.

The VMATs and their specific inhibitors have a long history of clinical importance in the study of neuronal and endocrine responses. As mentioned above, the adrenergic blocker reserpine is a competitive inhibitor of both VMAT1 and VMAT2. VMAT2 is also inhibited by the non-competitive inhibitor tetrabenazine (TBZ), which is used in the treatment of Huntington disease. Amphetamines used in the treatment of ADHD can efficiently enter storage vesicles and displace stored catecholamines, triggering their synaptic release and inducing intense psychostimulation.

The structure of VMAT1 and how drugs interact with this transporter has been revealed by a cryo-EM study. Specifically, the study reveals the structures of human VMAT1 in cytoplasmic-open and luminal-open conformations, in unbound form, and in complex with four monoamine neurotransmitters, as well as reserpine, amphetamine, and MPP⁺ (1663). The structures and functional studies reveal the mechanisms of monoamine storage and pharmacological interactions.

In addition, the structure of VMAT2 and the mechanism of inhibition by TBZ have been reported based on cryo-EM structural analysis of VMAT2 complexed with TBZ (1664). The study shows that TBZ interacts with residues in a central binding site, locking VMAT2 in an occluded conformation and providing a mechanistic basis for non-competitive inhibition.

In addition, the cryo-EM structures of human VMAT2 complexed with serotonin and clinical drugs were presented by two different groups (1665, 1666). Reserpine and the antihypertensive drug ketanserin were shown to occupy the substrate binding pocket and lock VMAT2 in the cytoplasmic and luminal states, respectively. TBZ was shown to bind to a VMAT2-specific pocket and to trap VMAT2 in an occluded state. The studies also reveal the structural basis of the VMAT2 trafficking cycle.

These structural insights establish the molecular architectures of VMAT and the mechanistic of substrate recognition, transport, and drug inhibition as a foundation for the rational design of small molecule therapeutics.

VMAT1 and 2 encode a large luminal loop containing several N-linked glycosylation sites (1664). The N-linked glycosylation of this loop together with a cytoplasmic C-terminal trafficking signal region of VMAT2 are required for targeting to the correct vesicles, the latter consisting of an extended dileucine motif, KEEKMAIL, and an additional sorting signal, the acidic cluster DDEESESD, located at the very distal C-terminal part of VMAT2 (1667). Specifically, it is proposed that luminal glycosylation of VMAT2 in the ER and Golgi compartments serve to sort the transporter into a compartment where interactions with the C-terminal tail of VMAT2 target the transporter to the vesicule.

VMAT2, but not VMAT1, has been reported to be constitutively phosphorylated (1668). Phosphorylation has been reported to occur at the cytoplasmic N- and C-terminal regions via protein kinases: At the N-terminus, where there are two putative PKC phosphorylation sites at serine 15 and serine 18, by PKC (1669), and at the C-terminus, where there are two serine residues within the acidic cluster (see above), by casein kinase II (1668, 1670). Although the precise role of phosphorylation remains to be established, N-terminal phosphorylation is necessary for the maintenance of VMAT2-mediated monoamine uptake (1671) and C-terminal phosphorylation plays an important role in targeting VMAT2 to the correct vesicles, as in the absence of phosphorylation by casein kinase, VMAT2 is transported into tiny synaptic vesicles rather than massive dense core granules.

A functional gene variant in SLC18A1, rs1390938 (T136I), was found to significantly increase the transport of monoamines into synaptic vesicles and it was hypothesized that the alteration in the magnitude of monoamine release contributes to the severity of alcohol withdrawal in patients of European descent (1672). Alcohol withdrawal after alcohol addiction depends on several neurochemical adaptations in the brain. One of these is linked to SLC18A1 rs1390938. The study showed that this variant not only leads to increased transport of monoamine neurotransmitters into vesicles, but is also associated with reduced withdrawal severity.

The allele frequencies of gene variants associated with psychiatric disorders have been evaluated in terms of human-unique psychological traits that may be favorable in specific human populations. To this end, the allele frequencies of SLC18A1 variants at position T136 were found to have interesting geographic variations (1673).

A homozygous SLC18A2 variant, P387L, has been reported as the cause of a rare disorder called infantile parkinsonism-dystonia 2 (1674). The mutation is predicted to affect the packaging of dopamine and serotonin into vesicles for subsequent synaptic transmission. It is an infantile-onset movement disorder that mimics cerebral palsy, and affected individuals show clinical features of dopamine deficiency, such as severe parkinsonism, dystonia, and oculogyric crisis. The mutated amino acid is a highly conserved proline residue located on the luminal side between TMHs 9 and 10 of VMAT2. The phenotypic and genotypic spectrum of SLC18A2-related disorders was then further investigated (1675, 1676). Treatment with dopamine receptor agonists such as pramipexole could improve some of the symptoms in affected individuals.

SLC18A3: VAChT (SLC18A3) is a vesicular transporter for the neurotransmitter acetylcholine (ACh) that has multiple functions in different parts of the body, typically acting as an excitatory mediator (1660, 1661). Acetylcholine is released in both the central and peripheral nervous systems. In the central nervous system, it is involved in memory, motivation, arousal, and attention, originating in the brain from the basal forebrain and mesopontine tegmentum. In the peripheral nervous system, the neurotransmitter is released at the neuromuscular junction. In the autonomic nervous system, a branch of the peripheral nervous system, it acts both as a neurotransmitter between preganglionic and postganglionic neurons as well as a final release product from parasympathetic postganglionic neurons. More precisely, it acts as a major neurotransmitter of the parasympathetic nervous system, which contracts smooth muscles, dilates blood vessels, increases bodily secretions, and slows heart rate, sweating, and increases gastrointestinal motility.

ACh is synthesized in the cytosol and its active transport into synaptic vesicles is mediated by the vesicular acetylcholine transporter VAChT (SLC18A3), from which it is secreted in response to Ca²⁺ signals. VAChT is a proton antiporter that pumps ACh into synaptic vesicles through the coupled efflux of protons, similar to the other vesicular neurotransmitter transporters.

Genetic variants in SLC18A3 have been reported to cause congenital myasthenic syndrome, a rare genetic disorder of neuromuscular junction function that results in muscle weakness and fatigue (1677). Specifically, the homozygous mutation D398H was identified in one of the patients as the cause of the disease.

Vesamicol is a well-characterized inhibitor of VAChT with nanomolar affinity (1678). To investigate the inhibitory mechanism of vesamicol as a basis for subsequent drug development efforts, the cryo-EM structure of human VAChT in complex with vesamicol or its native substrate Ach was determined (1679). The structures incorporate the MFS fold and adopt a lumen-facing conformation in which the two lobes of TMHs 1-6 and 7-12 open to the luminal side, similar to the reported lumen-facing ketanserin-bound VMAT2 structures (see the SLC18A2 description above). On the cytosolic side, two layers of hydrophobic interactions seal the intracellular gate between TMHs 4-5 and TMHs 10-11. The architecture of the intracellular gate is similar to that of VMAT2. The structure-function data of this study provide new insights into the mechanism by which vesamicol exerts its activity, revealing distinct binding patterns of vesamicol and ACh to VAChT. The work pinpoints the critical residues of VAChT and clarifies how VAChT recognizes its substrates.

The structural data also provide a good explanation for the aforementioned D398H mutation, which causes congenital myasthenic syndrome. It is likely that this mutation disrupts the substrate recognition of VAChT (1679).

In terms of trafficking, endocytosis of VAChT is thought to depend on a dileucine motif similar to VMAT2 or a distinct, downstream tyrosine-based motif (1680). VAChT also contains an upstream acidic residue and a phosphorylation site that mimic the acidic residues in the extended dileucine motif of VMAT2.

SLC18B1: VPAT (SLC18B1) is vesicular polyamine transporter that is highly expressed in lung, brain and placenta (1681). It is present in secretory granules of mast cells, in synaptic vesicles in neurons and in synaptic-like microvesicles in astrocytes (1682). Mast cells, which are abundant in human lung tissue, are granule-filled secretory cells that play an important role in host defense by secreting various intragranular contents, including histamine and serotonin, in response to Fc receptor stimulation (1683). The granules also contain spermine and spermidine, which modulate mast cell function. VPAT controls the storage and release of spermine and spermidine in mast cells, but via a different subpopulation of granules than those that release histamine and serotonin. In addition, VPAT serves as the gatekeeper for spermine and spermidine in the neuronal and astrocyte vesicles where it is important for both short-term and long-term memory via regulation of polyamine levels in the brain. Removal of VPAT leads to a reduction in polyamine levels in the brain, resulting in reduced GABA signaling due to a long-term reduction in glutamatergic signaling (1684).

Clinical relevance and pharmaceutical aspects

The most commonly used VMAT inhibitors are reserpine, a sympatholytic and antihypertensive agent, and TBZ used to treat movement disorders such as Huntington disease. While reserpine irreversibly binds to both VMATs, TBZ reversibly binds only to VMAT2 (79). Vesamicol, an experimental drug, is a non-competitive and reversible blocker of VAChT (1660).

Orphan transporter family members: N/A

SLC19 Folate/thiamine transporter family (2.A.48/Folate_carrier/MFS)

Discovery: A mouse folate transporter mRFC1 (Slc19a1) was first identified by functional complementation, in which methotrexate-resistant human breast cancer cells were transfected with cDNAs from a mouse cDNA library to identify the cDNA that could restore methotrexate sensitivity in these cells (1685).

Gene family members (3 + 1 pseudogene):
SLC19A1 (RFC)	SLC19A2 (THTR2)	SLC19A3 (THTR2)	SLC19A4P (pseudogene)

Molecular aspects, physiological roles and links to disease

This family consists of three members in human, reduced folate carrier RFC (SLC19A1) and thiamine transporter THTR1 (SLC19A2), which deliver their substrates to systemic tissues, and thiamine transporter THTR2 (SLC19A3), which mediates intestinal thiamine absorption (1686).

The SLC19 family belongs to the Reduced Folate Carrier (RFC) family (TC 2.A.48), which is part of the MFS superfamily. Members of the family consist of 12 TMHs. Cryo-EM structures of SLC19A1 have been determined that reveal residues critical for substrate recognition (see the SLC19A1 description below).

Folates carry a negative charge and thiamine a positive charge at physiological pH. Folates are essential for life and folate deficiency contributes to cardiovascular disease, fetal abnormalities, neurological disorders and cancer (1686), but mammals cannot synthesize folates de novo and therefore intestinal uptake is essential. Folate transport is mediated by three distinct systems, endocytosis via folate receptor alpha FRα (FOLR1), transport via reduced folate carrier RFC (SLC19A1), and H⁺-coupled folate transport via PCFT (SLC46A1) (1686, 1687). H⁺-coupled folate transport via PCFT (SLC46A1) is the mechanism by which folates are absorbed across the brush border membrane of the small intestine (see the SLC46 family description).

Thiamin, or vitamin B1, is an essential water-soluble vitamin absorbed in the intestine via THTR2 (SLC19A3). It is converted in the liver by thiamine pyrophosphokinase to thiamine pyrophosphate (TPP), a coenzyme required for carbohydrate and amino acid metabolism (1688).

SLC19A1: RFC (SLC19A1) is a widely expressed bidirectional exchanger. One of its roles is to transport folates and antifolate drugs, such as methotrexate and pemetrexed, which are used to treat cancer, rheumatoid arthritis, and psoriasis (1689).

Subsequent research has demonstrated that RFC (hereafter designated SLC19A1) facilitates the transport of cyclic dinucleotides (CDNs), which act as immune signaling molecules (1690–1692). One such immune signaling molecule, 2ʹ3ʹ–cGAMP, is synthesized by cyclic GMP-AMP synthase and elicits immune responses by binding to the stimulator of interferon genes (STING, STING1). SLC19A1 also transports bacterial and synthetic CDN-type drugs.

Knowledge of the function of SLC19A1 as a transporter of CDNs into cells is expected to facilitate the development of more targeted CDN-based cancer therapies (1691, 1693). This knowledge also has implications in the context of host responsiveness to CDN-producing pathogenic microorganisms (1694).

On the other hand, as already mentioned, SLC19A1 is the primary transporter of folates and chemotherapeutic agents such as methotrexate into the cells (1689, 1695, 1696). Import of folates by SLC19A1 is driven by countertransport of organic anions, including thiamine monophosphate (TMP), thiamine pyrophosphate (TPP), etc. (1686, 1697).

Like the other family members, SLC19A1 consists of 12 TMHs. High-resolution cryo-EM structures of human SLC19A1 and its complex with 5-methyltetrahydrofolate have been reported, and residues critical for substrate recognition and folate transport specificity have been determined (1698). Additionally, the study revealed that SLC19A1-mediated folate transport is coupled with the countertransport of organic phosphate anions such as TPP. According to the alternative access mechanism, SLC19A1 switches between inward- and outward-facing conformations to transport its substrates across the cell membrane.

Using an antibody-facilitated conformation screening strategy, cryo-EM structures of human SLC19A1 in its outward-open state, both with and without bound substrates, revealed the detailed mechanisms of substrate recognition and conformational changes that occur during transport (1692). This approach revealed that SLC19A1 mediates the import of folates and CDNs through its coupling with the export of various organic anions. These anions include ATP, ADP, AMP, 5-aminoimidazole-4-carboxamide ribotide monophosphate (an intermediate necessary for purine synthesis (1699)), and glucose 6-phosphate, a crucial intermediate in carbohydrate metabolism. TPP stood out among these as one of the preferred coupled substrates of SLC19A1, binding to inward-open SLC19A1 in a manner similar to folates (1692).

SLC19A1 is widely expressed in tissues (e.g., brain, placenta, small intestine, colon, and kidney), where it plays a central role in folate uptake (1686). Particularly high expression of SLC19A1 is present in the choroid plexus at the apical membrane facing the CSF, where it likely plays an important role in transporting folates into the CSF (1686). This process presumably occurs together with the folate transporter PCFT (SLC46A1) and the folate receptor FRα (FOLR1) at the basolateral (blood-facing) membrane (Fig. 11) (1700–1703). However, the precise mechanism by which folate and its derivatives are taken up in the CSF is not yet completely clear. The concentration of folate in the CSF is two to three times higher than in the blood, highlighting the importance of folate delivery to the CSF. It has also been proposed that SLC19A1 and PCFT (SLC46A1) may exist as homo-oligomers to facilitate intracellular trafficking and folate transport function (1704), although further validation of this hypothesis is needed.

A homozygous loss-of-function mutation in the SLC19A1 gene leads to embryonic lethality in mice but not in humans (1705, 1706). In human, SLC19A1 plays an important role in hematopoietic cells, which are highly dependent on transport systems to deliver folate from the extracellular environment. Consistent with this, a rare homozygous deletion in the SLC19A1 gene has been shown to cause folate-dependent recurrent megaloblastic anemia (1706). However, the extent to which a lack of SLC19A1 function in humans affects the transport of cyclic dinucleotides and immune responses via STING—the pathway of the innate immune system that helps defend against viral and bacterial infections—has yet to be investigated in affected individuals. Additionally, loss of SLC19A1 function has been shown to lead to methotrexate resistance in cancer cells (1707)..

The role of genetic mutations in folate-related transport genes in influencing the ability to meet maternal dietary requirements of folate for optimal human embryo development, as well as in increasing the risk of neural tube defects (NTDs) in newborns, was evaluated (1708, 1709). The analysis included the genes encoding the plasma membrane folate transporters SLC19A1 and PCFT (SLC46A1), the mitochondrial folate carrier SLC25A32, and the folate receptors FOLR1, FOLR2, and FOLR3. Numerous novel disease-associated variants were revealed in these genes. However, although these genes are closely associated with NTDs, it was concluded that more research is needed to determine the extent to which genetic variants, alongside environmental and nutritional factors, contribute to maternal folate deficiency and the risk of NTDs.

Since SLC19A1 transports TMP and TPP, but not free thiamine, it does not contribute to thiamine absorption in the intestine. THTR2 (SLC19A3) in the brush border membrane together with THTR1 (SLC19A2) in the basolateral membrane (Fig. 30) facilitate the transport of thiamine from the intestine into the portal circulation see below). If THTR1 (SLC19A2) is not functional, as occurs in rare genetic disorders of thiamine transport (see below under “Clinical relevance and pharmaceutical aspects”), SLC19A1 could still serve as a route for providing thiamine to nonpolarized cells, for example through the uptake of TMP, followed by hydrolysis to thiamine via thiamine monophosphatase (1710, 1711).

Fig. 30. Intestinal absorption of water soluble vitamins.

Abbreviations: PCFT, proton coupled folate transporter; THTR, thiamin transporter, SVCT, sodium coupled vitamin C transporter; SMVT, sodium coupled multi-vitamin transporter; SMCT, sodium coupled monocarboxylate; RFVT, riboflavin transporter.

Folate and TPP are cofactors that play a role in anabolic and catabolic enzyme functions, respectively. For instance, folate is essential for synthesizing nucleic acids through its role in single-carbon transfer reactions. TPP, on the other hand, is a cofactor for enzymes that break down glucose to produce energy. Thus, it is tempting to speculate that the coordinated exchange of these two molecules by SLC19A1 is an important mechanism for regulating cellular metabolism (1698).

SLC19A2, SLC19A3: THTR1 (SLC19A2) and THTR2 (SLC19A3) are high affinity transporters involved in the cellular accumulation of thiamine (vitamin B1). Thiamine, found in meat, grains and beans, is an essential nutrient for cellular function. THTR1 (SLC19A2) expression is relatively widespread in human tissues, with the highest expression in skeletal muscle (1712). THTR2 (SLC19A3) is also relatively widely expressed and represents the major thiamine absorptive pathway in the human small intestine, where it localizes to the brush border membrane of enterocytes, facilitating passive thiamine uptake (1713). Basolateral exit occurs via THTR1 (SLC19A2). Although THTR2 was previously thought not to be essential for intestinal absorption, as other thiamine transporters such as OCT1 (SLC22A1) can perform this task (1714, 1715), a follow-up study in mice highlighted its importance (1716) (see the description below under “Clinical Relevance and Pharmaceutical Aspects”). THTR2 (SLC19A3)-mediated thiamine transport also plays a central role in maintaining appropriate levels of free thiamine, TMP, and TPP in various tissues in the brain, heart, fat, liver, and kidney (1716). In whole human blood, the vitamin is present as free thiamine and TMP, in addition to TPP, the concentration of which depends on metabolism and health status (1717, 1718).

In addition to thiamine transport via THTR2 (SLC19A3), TPP can be taken up into tissues via SLC19A1 in exchange for folate, as mentioned above (1698, 1711). In all tissues, the active form of thiamine is TPP, which serves as an important coenzyme in the oxidative decarboxylation of α-keto acids as part of key enzymes involved in glucose, amino acid, and lipid metabolism in mitochondria, to which it is transported via SLC25A19 (see the SLC25 family description and Fig. 36, top left). In the glycolytic pathway, TPP is required to produce energy from glucose because it is a cofactor for pyruvate dehydrogenase to produce acetyl CoA, the first substrate in the citric acid cycle. In its absence, pyruvate is instead converted to lactate. TPP is also a cofactor of α-ketoglutarate dehydrogenase, and in its absence the citric acid cycle is interrupted, resulting in a switch from aerobic to anaerobic metabolism, also with accumulation of lactate, which can be an issue in metformin-associated lactic acidosis (see below). Lastly, TPP is a cofactor of transketolase as part of the pentose phosphate pathway, which produces important molecules such as ribose-5-phosphate used in nucleotide synthesis (see below).

Thiamine has a short half-life, and while there is no long-term storage of thiamine in the body, excess thiamine can be stored to a limited extent in hepatocytes, erythrocytes, and other cell types in the form of TPP (1711, 1719). In the liver, it is absorbed via THTR2 across the sinusoidal membrane (see Fig. 33) and stored as TPP (formed via thiamine pyrophosphokinase), where it acts as a reservoir for other organs, including the brain, allowing the body to remain healthy for up to two weeks. TPP can also enter hepatocytes via SLC19A1 (1711). Similar uptake and storage mechanisms exist in erythrocytes.

In the brain, both THTR2 (SLC19A3) and THTR1 (SLC19A2) are expressed at the BBB. Mouse studies report THTR2 on the abluminal (brain-facing) side and THTR1 on the luminal side of brain microvascular endothelial cells (1716, 1720) (see Fig. 33). However, human data suggest that THTR2 may instead be luminal (blood-facing) (1721), which could reflect a species difference. If THTR2 is luminal in humans, the mechanism by which thiamine exits the abluminal side to enter brain tissue remains to be clarified.

SLC19A4P - Pseudogene: This duplicated gene is adjacent to SLC19A3 in the human genome and is currently considered to be a unitary pseudogene in human. Coding versions of this gene are present in other vertebrates, including dog and cat, but these remain uncharacterized.

Clinical relevance and pharmaceutical aspects

Genetic disorders of thiamine transport are rare but rather severe (1721). Thiamine-responsive megaloblastic anemia (TRMA) syndrome is an autosomal recessive disorder caused by mutations in the gene SLC19A2 (THTR1) (1722–1724). It is characterized by diabetes mellitus resulting from islet dysfunction due to thiamine deficiency, as well as sensorineural deafness resulting from the high thiamine requirements of cochlear or acoustic nerve cells, and neurological manifestations. TRMA patients still have partial thiamine transport capacity because of the redundancy with THTR2 (SLC19A3), and therefore generally have near-normal blood thiamine levels. So what causes the disease is defective THTR1 transport into cells that require increased levels of the vitamin. This results in an impairment of mitochondrial energy production, de novo synthesis of nucleic acids and synthesis of heme precursors (1725). Specifically, in megaloblastic anemia, intracellular thiamine deficiency affects the de novo synthesis of nucleic acids due to the lack of functional thiamine TPP-dependent transketolase and TTP-dependent α-ketoglutarate, which supplies metabolites to the Krebs cycle required to produces the heme precursor succinyl-CoA. Oral thiamine supplementation helps maintain health and correct hyperglycemia in patients (1726, 1727). However, it did not improve the hearing problem (1728).

A genetic disease involving THTR2 (SLC19A3) is biotin-thiamine-responsive basal ganglia disease, a rare disorder that affects the nervous system, especially the basal ganglia in the brain, causing subacute encephalopathy (e.g., confusion, dysphagia, dysarthria, and seizures) (1729). Specific treatment includes initiation of thiamine and biotin supplementation. Administration of high-dose thiamine restores its levels in cells and cerebrospinal fluid, reduces lactic acidosis, and limits cerebral edema and necrosis (1730). To evaluate the impact of THTR2 on tissue thiamine status and metabolism in the brain, a human SLC19A3 transgene was expressed in the intestine of total body Slc19a3^-/- mice. The studies showed that loss of THTR2 in Slc19a3^-/- mice reduces the levels of nucleic acid and amino acid derivatives in the brain, which could be rescued by expression of human THTR2 in the intestine, leading to normal circulating thiamine levels. Therefore, it has been proposed that THTR2 inhibition alters the brain metabolome and reduces the thiamine reservoir for thiamine diphosphate biosynthesis (1716).

Folate transporters play an important role in delivering folate analogs to neoplastic and inflammatory cells to treat cancer or inflammatory and autoimmune diseases. Antifolates such as methotrexate and pemetrexed are structurally similar to folates but work by inhibiting enzymes involved in folate metabolism, thus disrupting DNA synthesis and cell growth (1731). They have provided much of our knowledge about the mechanisms of folate transport (1686).

THTR1 and THTR2 interact with drugs such as the Janus kinase (JAK2) inhibitor fedratinib and the antidiabetic drug metformin (1732). Inhibition of THTR2 has a significant impact on thiamine intestinal absorption and renal reabsorption, resulting in thiamine deficiency.

Metformin-associated lactic acidosis (MALA) is a relatively rare but serious adverse event with high mortality. It has been suggested that metformin competitively inhibits THTR2-mediated thiamin transport into hepatocytes, leading the MALA via the pathways summarized above (1733).

Orphan transporter family members: N/A

HGNC update

The unitary pseudogene SLC19A4P has been named and added to the SLC9 family.

SLC20 Type III Na⁺-phosphate cotransporter family (2.A.20.2/PHO4/PiT)

Discovery: PiT-1 and PiT-2 (SLC20A1 and SLC20A2) are sodium phosphate (NaPi-III) cotransporters that are widely expressed transmembrane proteins. Originally described as retroviral receptors, they were later shown to be NaPi cotransporters (1734, 1735).

Gene family members (2):
SLC20A1 (PiT-1)	SLC20A2 (PiT-2)

Molecular aspects, physiological roles and links to disease

The SLC20 family is a member of the Inorganic Phosphate Transporter (PiT) family (TC 2.A.20.2. Members of this family have a transporter core containing 5+5 TMHs in an inverted repeat arrangement (see Section 8).

SLC20 PiT transporters play an important role in soft tissues, supporting intracellular processes such as biosynthesis of cellular components such as ATP, nucleic acids, phospholipids and contributing to metabolic pathways such as energy transfer, protein activation, and carbon and amino acid metabolic processes. In addition, they are increasingly recognizes as being important in serving as phosphate sensors that mediate P_i-dependent intracellular signaling (1736).

In humans, phosphorus makes up about 1% to 1.4% of lean body mass. About 85% is found in bones, 5% in blood and teeth and the remainder (about 10%) in soft tissues (1737, 1738). Phosphate homeostasis in the human body is maintained by a balance between intestinal absorption, renal excretion, and transport into and out of bone. This balance is mainly controlled by the coordinated actions of 1,25-dihydroxyvitamin D (the active form of vitamin D; also known as calcitriol), parathyroid hormone (PTH), and fibroblast growth factor-23 (FGF23), which regulate phosphate reabsorption in the proximal tubule of the kidney via the sodium-phosphate cotransporters Npt2a (SLC34A1) and Npt2c (SLC34A3) (1738, 1739). In terms of soft tissues, the PiT transporters play important roles in skeletal muscle, liver and bone marrow and probably also in bone metabolism (see below).

SLC20A1, SLC20A2: PiT-1 (SLC20A1) and PiT-2 (SLC20A2) are both widely expressed electrogenic Na⁺-coupled phosphate transporters (1740). They are also referred to as “type III phosphate transporters”. Regulation of intracellular levels of inorganic phosphate (P_i) requires careful control of cellular P_i uptake and efflux for the survival of an organism. The PiT transporters are responsible for cellular uptake, while efflux is controlled by the XPR1 (SLC53A1) P_i exporter (see the SLC53 family description).

As highlighted above, the PiT transporters have important metabolic housekeeping functions, expressed in many different tissue types, but need to be regulated to avoid elevated intracellular phosphate levels, which can lead to vascular calcification associated with cardiovascular morbidity and mortality (1741). Increased SLC20A1 expression has also been observed in aggressive tumors such as claudin-low breast cancer, presumably to restore cellular phosphate and accommodate rapid growth (1742).

According to the HPA, although the PiT transporters are ubiquitously expressed and have complementary functions in certain organs, there are also marked differences in their expression patterns: PiT-1 (SLC20A1) is most highly expressed (in decreasing order) in bone marrow, colon, liver, retina, kidney, various brain regions, placenta, lung, and muscle tissue, and at the single cell level in distal enterocytes, monocytes, dendritic cells, and photoreceptor cells, whereas PiT-2 (SLC20A2) is most highly expressed in choroid plexus, muscle tissue, thyroid gland, salivary gland, pancreas, liver, duodenum, and at the single cell level in neurons, oligodendrocytes, keratinocytes, squamous epithelial cells, serous glandular epithelial cells, and myocytes.

PiT-2 (SLC20A2) is abundantly expressed on the apical membrane (CSF/ventricular side) of the choroid plexus epithelium (1743) (Fig. 11). Its dysfunction therefore prevents P_i backflow from the cerebral ventricles into the blood vessels, resulting in ventricular P_i accumulation (1743). Consistent with this, elevated cerebrospinal fluid P_i levels have been reported in patients with SLC20A2 mutations (1744, 1745) and Slc20a2 homozygous knockout mice (1743, 1746, 1747).

Loss-of-function mutations in SLC20A2 are the major genetic cause of primary familial brain calcifications (PFBC), also known as Fahr disease. It is a rare autosomal dominant or recessive neurodegenerative disorder characterized by bilateral and symmetric microvascular calcifications affecting multiple brain regions, leading to cognitive and neuropsychiatric impairments (1747, 1748). The phenotypes associated with Fahr disease are very similar to those in Slc20a2 knockout mice. Therefore, Slc20a2 knockout mice are suitable for the future evaluation of neuropharmacological intervention strategies for the treatment of patients with Fahr disease.

Additional roles of the phosphate transporters PiT-1 (SLC20A1) and PiT-2 (SLC20A2) in the brain beyond the choroid plexus have been proposed but are still poorly understood. For example, PiT-1 (SLC20A1) and PiT-2 (SLC20A2) have been reported to regulate neuronal plasticity and cognition independently of phosphate transport (1749).

In skeletal muscle, P_i is taken up from the extracellular space by SLC20 PiT transporters, a process stimulated by insulin. Insulin plays an important role in promoting P_i uptake by myocytes, in addition to its well-known stimulation of GLUT4 (SLC2A4) glucose transport. Myocytes require P_i for energy storage in the form of ATP and creatine phosphate. They need ATP for glucose phosphorylation and glycolysis, and creatine phosphate for rapid ATP production during periods of increased energy demand (1750). The two transporters PiT-1 and PiT-2 have been shown to have overlapping functions in muscle, as studies in Slc20a1^-/- or Slc20a2^-/- mice showed that Pit-1 and Pit-2 are essential for normal skeletal muscle fiber function, and combined genetic deletion of both transporters caused severe myopathy (1751).

In the liver, P_i is required for glycolysis and the synthesis of phosphoproteins and lipids. Hepatocytes express PiT-1 (SLC20A1) and PiT-2 (SLC20A2), probably mainly in the sinusoidal membrane (1752) and NaPi-IIb (SLC34A2) in the canalicular membrane (1650). The absence of PiT-1 during development causes hepatic hypoplasia (1753). Interestingly, the absence of PiT-2 in human adults has a beneficial effect, as it improves glucose metabolism (1752): Knockout of PiT-1 in hepatocytes improved glucose tolerance and insulin sensitivity, enhanced insulin signaling, and reduced hepatic lipogenesis. The ubiquitin-specific protease USP7, identified as a PiT-1 binding partner (1752), was shown to interact with the insulin receptor substrate IRS1 under serum starvation conditions, but dissociated from IRS1 after insulin stimulation (1752, 1754). Consequently, deletion of PiT-1 inhibited USP7-IRS1 dissociation upon insulin stimulation, thereby preventing IRS1 ubiquitination and proteasomal degradation. This resulted in a delayed insulin negative feedback loop and sustained insulin signaling.

PiT-1 is furthermore required for normal hematopoiesis and mice lacking PiT-1 develop a profound underproduction anemia due to blockade of erythroid differentiation (1755). They also have severe B cell deficiency. It was concluded that PiT-1 function is required at multiple stages of hematopoiesis and specifically for terminal erythroid differentiation and B cell development. PiT-1 may also function in a similar manner in human hematopoiesis, as supported by the finding that a point mutation in SLC20A1 was identified in a patient with low-grade human myelodysplastic syndromes, a disease that resembles the hematologic phenotype of mice lacking PiT-1, in which cell cycle defects may play a pathophysiologic role (1755). Consistent with a role for PiT-1 in regulating cell proliferation, overexpression of human PiT-1, but not human PiT-2, in density-inhibited cell lines enhanced cell proliferation (1756).

In the skeleton, phosphate is a key component of hydroxyapatite, the main mineral of bone, as well as an integral second messenger in the regulation of bone metabolism. The SLC20 and SLC34 phosphate transporters, which are partially expressed in bone cells, probably have no critical function in P_i transport (1757). However, PiT (SLC20) transporters may be involved in bone P_i sensing.

Two concepts, A) and B), have been proposed with respect to the putative role of PiT-1 in cellular phosphate sensing:

• A): PiT-1 (SLC20A1) expression was found to be low in the phosphate-replete state of cultured cells, but was strikingly induced following phosphate starvation (1758), and the following mechanism has been proposed for regulating intracellular P_i levels (1759): 1) In the presence of sufficient P_i, cells continuously internalize and degrade PiT-1 via the ESCRT (endosomal sorting complexes required for transport) machinery (1758). 2) P_i scarcity causes recycling of PiT-1 from early endosomes to the plasma membrane, thereby increasing the capacity for P_i influx via PiT-1 (1758); and 3) phosphate influx via SLC20 phosphate transporters is balanced by efflux through XPR1 (SLC53A1). To achieve this balance, inositol pyrophosphate (PP-InsP) signaling molecules, such as 1,5-bis-diphosphoinositol 2,3,4,6-tetrakisphosphate (InsP₈), interact with the N-terminal SYG1-Pho81-XPR1 (SPX) domain of XPR1 to increase P_i efflux during P_i abundance, whereas during P_i deficiency, InsP₈ synthesis is reduced, resulting in decreased P_i efflux (337, 1760, 1761) (see the SLC53 family description). Thus, cytosolic levels of InsP₈ are regulated by fluctuations in cytosolic P_i levels (1762–1764). In response to cytosolic P_i deprivation, InsP₈ levels decrease, thereby reducing the functionality of the XPR1 phosphate exporter.
• B): As a refinement of concept A, a subsequent study showed that organelle-localized XPR1 (SLC53A1) regulates cellular P_i uptake by PiT-1 in mammalian cells through direct interaction with the phosphate transporter PiT-1, which prevents PiT-1 degradation (1765). XPR1 binding to PiT-1 is proposed to involve a transmembrane helix and a cytosol-facing loop in XPR1 and to occur in a subset of intracellular LAMP1 (lysosomal-associated membrane protein 1)-positive vesicles (XLPVs). Since XLPVs are part of the endosomal-lysosomal system, PiT-1 may be recycled between early endosomes and the plasma membrane. According to the proposed model, XPR1 is recycled between XLPVs and the plasma membrane, and the XLPV pool of PiT-1 is stabilized by its interaction with InsP₈-bound XPR1. In response to cytosolic P_i deprivation, intracellular InsP₈ levels decrease, which in turn deprives XPR1 of its functionalizing ligand within minutes, resulting in a rapid reduction in the rate of cellular P_i efflux with no change in P_i uptake. XPR1 then directs PiT-1 to the degradation pathway, reducing the availability of PiT-1 for transport to the plasma membrane, which in turn reduces the rate of P_i influx to terminate the P_i correction (1765). According to this model, there is a temporal separation of cellular control over P_i release and P_i uptake to ensure an effective system for maintaining P_i homeostasis.

Clinical relevance and pharmaceutical aspects

A high-affinity SLC20 inhibitor, EOS789, has been described, although SLC34 transporters are also inhibited by EOS789, but at slightly higher concentrations. The properties of EOS789 suggest that it has therapeutic potential for hyperphosphatemia associated with chronic kidney disease. PiT-1 (SLC20A1) may serve as a potential target for therapies aimed at reducing vascular calcification (1766) as well as for treating certain forms of breast cancer (1742).

Orphan transporter family members: N/A

SLCO Organic anion transporter family (2.A.60/OATP/MFS)

Note on the nomenclature deviation (SLCO for SLC21): The SLC21/SLCO organic anion transporting polypeptides (OATPs) form a relatively large family of Na⁺-independent transport systems that mediate the transmembrane transport of a wide range of amphipathic endogenous and exogenous organic compounds. The family can be divided into 6 subfamilies when considering a phylogenetic tree that includes OATPs from a variety of organisms (178, 1767). Originally, the OATP genes were grouped into the SLC21 gene classification system, and assignments were made roughly in the chronological order of discovery. However, this grouping did not reflect the classification into the 6 different subfamilies. Therefore, a new classification system was proposed (1767) and subsequently approved by the HGNC. In this system, the 6 subfamilies are named OATP1 to 6, and further subbranches are indicated by A, B, C, etc. (Fig. 31). The individual genes were numbered according to the chronological order in which they were discovered and in order to keep the root symbol “SLC”, the letter “O” from “OATP” was added to this root symbol. For example, the gene encoding OATP1C1 became SLCO1C1. To facilitate the transition from the old to the new system, comparison tables were created (1767).

Fig. 31. Phylogenetic tree of the SLC21/SLCO family.

Discovery: The rat sodium-independent bile salt and organic anion transporter Oatp1a1 (also known as Oatp1, Slc21a1, and Slco1a1) was the first member of the OATP/SLCO gene family to be identified, which was achieved by expression cloning using Xenopus oocytes (149). It has been reported that rat Oatp1a1 is functionally and pharmacologically similar to human OATP1B1 (SLCO1B1) (1768). However, it is important to note that rat has four additional paralogs (Slco1a3-6), while human has only one paralog, SLCO1A2, meaning that SLCO1A1 does not exist in human (see Supplementary File 1 in (7)). Therefore, from a phylogenetic viewpoint, rat Oatp1a1 is most similar to human OATP1A2 (SLCO1A2).

Gene family members (12):
SLCO1A2 (OATP1A2)	SLCO1B7 (OATP1B7)	SLCO2B1 (OATP2B1)	SLCO4C1 (OATP4C1)
SLCO1B1 (OATP1B1)	SLCO1C1 (OATP1C1)	SLCO3A1 (OATP3A1)	SLCO5A1 (OATP5A1)
SLCO1B3 (OATP1B3)	SLCO2A1 (OATP2A1)	SLCO4A1 (OATP4A1)	SLCO6A1 (OATP6A1)

Molecular aspects, physiological roles and links to disease

The SLC21/SLCO family has 12 members (Fig. 31) and belongs to the Organo Anion Transporter (OAT) family (TC 2.A.60), which is part of the MFS superfamily. The functionally well-characterized members of the SLC21/SLCO family are predicted to have 12 transmembrane domains and are Na⁺-and ATP-independent transporters that mediate the transport of a wide range of amphipathic endogenous and exogenous organic compounds (178). All of them contain the superfamily signature D-X-RW-(I,V)-GAWW-XG-(F,L)-L at the boundary between extracellular loop 3 and TMH6, which has been suggested to play a role in protein folding or in stabilizing the protein within the membrane (1769). The region includes the conserved residues W258, W259 and F262, and when mutated to alanine, the transport function was also significantly reduced (1770).

Cryo-EM structures of human OATP1B1 (SLCO1B1) have provided detailed insights into the architecture of the protein, as well as its mechanisms of substrate recognition and transport (1771) (see the SLCO1B1 description below).

The human SLCO family contains 11 organic anion transporting polypeptides (OATPs). As shown in the phylogenetic tree (Fig. 31) these are divided into six subfamilies, SLCO1 through SLCO6, with each subfamily belonging to a distinct phylogenetic branch. These subfamilies are subsequently divided into smaller subcategories based on sequence identity and designated with a capital letter followed by a number. What follows is a description of the individual members of the different branches.

Subfamily 1A

The only human member of this branch is OATP1A2 (SLCO1A2). Slco1a1 (also known as Oatp1a1 and Slc21a1), which was originally identified by expression cloning, does not exist in humans (149).

SLCO1A2: OATP1A2 (SLCO1A2) is a unidirectional uptake carrier and is predominantly expressed in the brain, including the BBB, and also in the renal proximal tubules, liver and small intestine (1772) (Fig. 33). According to the HPA, it is most highly expressed in brain areas in oligodendrocytes and endothelial cells, with significantly lower or negligible expression in other organs, except in the liver where there is moderate expression and where it has been reported to be expressed in cholangiocytes (1773). Expression in cholangiocytes suggests that it may contribute to the reabsorption of xenobiotics excreted into the bile (1773). Western blot analysis of human brain tissue showed expression of OATP1A2 in the frontal cortex of the brain, with immunofluorescence localizing the transporter to human brain microvessels and brain capillary endothelial cells (1773, 1774).

OATP1A2 transports a wide range of organic anionic, neutral, cationic and amphipathic endogenous and xenobiotic molecules such as bile acids, steroid conjugates, thyroid hormones T3 and T4, prostaglandin E₂ (PGE₂), linear and cyclic peptides, mycotoxins, fexofenadine (antihistamine), ouabain and statins. In addition, as part of the visual cycle, it mediates the apical uptake of all-trans-retinol from the subretinal space into the human retinal pigment epithelium (RPE), which is essential for vision, which depends on retinoid exchange between the RPE and photoreceptors (1775).

Genetic polymorphisms of SLCO1A2 play an important clinical role in drug disposition and CNS drug entry (1773, 1776).

Subfamily 1B

SLCO1B1: OATP1B1 (SLCO1B1) is predominantly expressed at the sinusoidal (basolateral) membrane where it plays an important role in the uptake of bile acids and bilirubin as well as in the clearance of drugs and endogenous molecules from the blood (Fig. 33). OATP1B1 is responsible for the influx of statins from the blood into the liver for clearance, and SLCO1B1 genetic loss-of-function variants are known to be strong predictors of statin-associated muscle pain and risk of rhabdomyolysis (1777), highlighting the importance of genotyping prior to prescription of statins (1778).

Genetic defects in SLCO1B1 and SLCO1B3 cause autosomal recessive Rotor syndrome, also known as Rotor-type hyperbilirubinemia. The disease is characterized by non-hemolytic jaundice due to chronic elevation of predominantly conjugated bilirubin as a result of insufficient uptake of bilirubin by hepatocytes via SLCO1B1 and SLCO1B3 for clearance from the body.

Cholestasis is a condition in which the flow of bile from the liver through the biliary system stops or slows, leading to excessive accumulation of bile components such as bile acids, cholesterol and bilirubin in the liver and circulation. This, in turn, causes damage to hepatocytes and the entire organism. Interleukin 6 (IL-6) has been shown to reduce OATP1B3 expression levels in cholestatic diseases (1779). To investigate the regulatory mechanism of OATP1B1 under cholestasis, cholestatic mice were treated with C170 of STING (stimulator of interferon genes). Cyclic GMP-AMP synthase (cGAS)-STING signaling plays an essential regulatory role in antibacterial, antiviral and antitumor immunity via the induction of cytokines, especially type I interferons, and there is increasing evidence that inflammatory diseases also lead to aberrant cGAS-STING signaling in macrophages. Therefore, the STING inhibitor C170 was administered to cholestatic mice and it was shown that the STING inhibitor inhibited inflammatory factors such as TNF-α and IL-6 and reduced the accumulation of bile acids and bilirubin in serum. The mechanism of IL-6 downregulation of OATP1B1 via β-catenin was further confirmed in HepG2 and PLC5/PRF/5 cells by downregulating β-catenin/TCF4 signaling (1780).

The transport function of OATP1B1 has also been shown to be regulated by PKC, the action of which leads to accelerated internalization and reduced recycling of the transporter protein (1781). PKC is also known to stabilize β-catenin, which together could lead to additional downregulation of OATP1B1 in cholestasis (1782).

Structural analysis using cryo-EM has provided new insights into the architecture and transport mechanism of human OATP1B1 (SLCO1B1) (1771). The transmembrane domain of OATP1B1 adheres to the classic MFS structure and comprises 12 TMHs. The N- and C-termini are located in the cytoplasm. The transmembrane domain consists of an N-terminal and a C-terminal part that exhibit twofold pseudo-symmetry around an axis perpendicular to the membrane plane. These two parts are connected by an intracellular helix. Cryo-EM structures in both outward- and inward-open states have also been reported in complex with the representative endogenous metabolites bilirubin and estrone-3-sulfate, the antiviral drug simeprevir used to treat chronic hepatitis C virus infection, and the fluorescent indicator 2ʹ,7ʹ-dichlorofluorescein (1771). The structures reveal major and minor substrate binding pockets, as well as the conformational changes that occur during transport. Combined with mutagenesis studies and molecular dynamics simulations, the study clarifies the transport mechanism of OATP1B1. The data support the “rocker-switch” alternate access transport mechanism (1771). According to this model, the transport protein flips or oscillates between two states like a rocker switch, allowing access to the substrate binding site from either side of the membrane (1783).

SLCO1B3: The closely related transporter OATP1B3 (SLCO1B3), also known as OATP8 or SLC21A8, has significant substrate overlap with OATP1B1 and it is also found almost exclusively at the sinusoidal membrane of hepatocytes (1784). OATP1B3 is involved in the uptake of bilirubin, the major breakdown product of heme, into centrilobular hepatocytes. This is followed by conjugation of bilirubin and excretion into the bile via MRP2 (ABCC2) (1784).

As mentioned in the OATP1B1 description, genetic defects in SLCO1B3 are the cause of Rotor syndrome type of hyperbilirubinemia, an autosomal recessive form of primary conjugated hyperbilirubinemia (1785, 1786) accompanied with development of mild jaundice. They also have delayed plasma clearance of bromsulphthalein, an anionic diagnostic dye, and marked urinary excretion of coproporphyrin I, a metabolic intermediate in the biosynthesis of heme (1787, 1788).

OATP1B3 (SLCO1B3) also plays an important role in drug clearance from the body (1789) (Fig. 33). In fact, OATP1B1 and OATP1B3 are among the most highly expressed uptake or efflux membrane transporters expressed in the liver and thus play a major, clinically important role in the hepatic uptake and clearance of many drugs (1790). Both transporters are polymorphic and functionally relevant, and ethnically dependent polymorphisms have been identified and characterized that alter drug pharmacokinetics, drug treatment response, and risk for drug-induced toxicities (1791).

To gain general insight into the activation and inhibition of OATPs by modulators (1792), structure-function studies were performed to investigate how the skin antifungal drug clotrimazole activates SLCO1B3. The studies revealed that amino acid residues G45 and V386 in TMHs 1 and 8 are critical for the activation of OATP1B3-mediated estradiol-17β-glucuronide (E17βG) uptake by clotrimazole (1792). Clotrimazole itself is not transported by OATP1B3, but efficiently crosses the plasma membrane by lipid permeation. Clotrimazole is predicted to bind to an allosteric site consisting of hydrophobic residues located in the cytoplasmic region of the transporter.

SLCO1B7 - Orphan transporter: OATP1B7 (SLCO1B7) was originally thought to be a pseudogene, but subsequently hybrids of alternative splice products have been reported to be expressed, affecting drug disposition (1793, 1794). Specifically, the expression of OATP1B3-1B7 has been reported to result from splicing of mRNAs encoded by SLCO1B3 and SLCO1B7, with SLCO1B3 encoding the first part of a 12 transmembrane domain protein and SLCO1B7 being the origin of the rest of the mRNA. The resulting protein, OATP1B3-1B7 (also known as LST-3TM12), was reported to be expressed in the liver with enrichment in granular intracellular structures of hepatocytes, which were identified as the smooth ER (1793). The HPA suggests a strong expression of SLCO1B7 in the mammary gland and at lower levels in the liver. Expression of this transporter increased cellular levels of dehydroepiandrosterone sulfate (DHEAS) using a heterologous expression system (1793). Non-synonymous, naturally occurring SNPs located within the gene region of SLCO1B7 were reported to affect the function of OATP1B3-1B7 as determined by cellular accumulation of DHEAS in HeLa cells (1795). The basic premise of these transport experiments was that even if OATP1B3-1B7 is localized in intracellular vesicular structures, there would still be a change in cellular levels, assuming passive diffusion of the substrate. It has been suggested that this study provides the basis for pharmacogenetic studies in humans, which will ultimately help to decipher whether OATP1B3-1B7 plays a role in the liver or is just a genetic relic (1795).

Subfamily 1C

SLCO1C1: OATP1C1 (SLCO1C1) has been identified as a high affinity thyroid hormone transporter. Thyroid hormones are essential for cell growth and metabolism. The developing central nervous system is particularly sensitive to thyroid hormone deficiency, which can result in neurological deficits and cognitive impairment. In contrast to most other family members, this transporter has a relatively selective substrate preference and thus plays an important role in the disposition of thyroid hormones in brain and also in testis (1796).

OATP1C1 (SLCO1C1) plays a critical role in the delivery of thyroid hormone T4 across the blood-cerebrospinal fluid barrier and the BBB. OATP1C1 is prominently expressed in astrocytes, microglial cells, pyramidal neurons and interneurons in the motor cortex and on the apical and basolateral surface of choroid plexus epithelial cells in humans (1797, 1798).

In choroid plexus epithelial cells, OATP1C1 (SLCO1C1) is thought to function in the transport of thyroid hormone T4 from the blood into the cerebrospinal fluid (1535, 1797, 1799) (Fig. 11). In rodents, Slco1c1 is thought to work in concert with the thyroid hormone transporter Mct8 (Slc16a2), which is expressed on the apical side of choroid plexus epithelial cells (1535, 1796–1799). However, in humans, MCT8 expression appears to be relatively low in the choroid plexus according to the HPA. Therefore, it is possible that in humans OATP1C1 facilitates thyroid hormone transport at both the apical and basolateral membranes.

Because OATP1C1 is a major contributor to thyroid hormone homeostasis in the brain by facilitating T4 transport across both the blood-cerebrospinal fluid barrier and the BBB, novel in vitro assays were subsequently developed to screen chemicals for their ability to inhibit the thyroid hormone transporter OATP1C1 (SLCO1C1) (1800).

Subfamily 2

SLCO2A1: OATP2A1 (SLCO2A1) is widely distributed in tissues and functions as a prostaglandin uptake transporter. It plays an important role in delivering prostaglandins to their site of action in target tissues through release and distribution of newly synthesized prostaglandins and/or metabolic clearance of prostaglandins from the circulation (1801). OATP2A1 represents the rate-limiting step in prostaglandin (especially PGE₂) inactivation by transporting PGs into cells for degradation (1801).

In the lung, OATP2A1-mediated transport of prostaglandin E₂ (PGE₂) exerts anti-inflammatory and anti-fibrotic effects. The transporter removes PGE₂ produced by epithelial and inflammatory cells from the lumen and, in cooperation with basolateral MRP4 (ABCC4), carries it across alveolar type 1 cells to the interstitial space, where PGE₂ exerts its antifibrotic action (1802)(see Fig. 33, bottom part).

In the hypothalamus, OATP2A1 (SLCO2A1) has been shown to be essential for body temperature regulation during fever by contributing to a high concentration of PGE₂ in the hypothalamic interstitial fluid to enable the febrile response (1803).

Loss-of-function mutations of SLCO2A1 lead to disorders such as primary hypertrophic osteoarthropathy (1804) and chronic enteropathy (1805), while pharmacological inhibition of OATP2A1 may enhance tissue repair and regeneration (1806, 1807).

Cryo-EM structures of OATP2A1 in both apo and PGE₂-bound states have revealed its architecture and substrate recognition mechanism, showing how the fatty acid-like PGE₂ binds within the central cavity as part of an alternating access transport cycle (1808). These structural insights represent a critical step toward understanding OATP2A1 function and advancing its potential as a therapeutic target.

SLCO2B1: OATP2B1 (SLCO2B1) has a relatively broad substrate specificity at an acidic pH (pH 6.8) for various endogenous products and drugs, while at pH 7.4, it transports mainly steroid hormone conjugates. A metabolic-scale gene activation screen indicated that SLCO2B1 can also act as a heme transporter to enhance cellular iron availability (1809). According to the HPA, OATP2B1 (SLCO2B1) is expressed in liver, adrenal gland, heart muscle, testis, stomach, pancreas, small intestine, lung and brain areas.

OATP2B1 (SLCO2B1) is expressed on the basolateral side of hepatocytes (Fig. 33), but it was found on the apical membrane of enterocytes (1772, 1810). However, its precise localization in enterocytes – i.e., apical versus basolateral – remains controversial (Keiser, 2017, #2395; Kobayashi, 2003, #2396), which may be due in part to species differences (1811). The broader substrate specificity at lower pH is due to the lower K_m values at pH 6.5 compared to pH 7.4. Thus, there would be an increased affinity of the transport process in the acidic microclimate at the intestinal brush border membrane surface.

N-glycosylation of OATP2B1 at residues N176 and N538 was shown to be essential for plasma membrane surface expression, but not for transport function per se. Immunofluorescence analysis showed that deglycosylated OATP2B1 is largely retained in the ER, which may trigger an ER-associated degradation pathway involving the ubiquitin-proteasome system, leading to degradation of OATP2B1 (1812).

Most clinical evidence concerning the role of OATP2B1 in drug pharmacokinetics relates to intestinal OATP2B1 (1813). A significant (>80%) decrease in exposure to drugs such as the antihistamine fexofenadine and the beta-blocker celiprolol has been attributed to the inhibition of intestinal OATP2B1 when these drugs are administered with common fruit juices. These juices are potent inhibitors of intestinal OATP2B1 and OATP1A2, decreasing exposure to co-administered substrates by approximately 85% (1814).

The effect of SLCO2B1 genetic mutations on transporter function and drug response has been studied (1815). A rare SLCO2B1 mutation resulted in the C520S OATP2B1 variant. This abolished in vitro uptake activity of OATP2B1. Loss-of-function variants such as this would be expected to alter the pharmacokinetics of drugs mediated by OATP2B1, given that OATP2B1 is a prevalent pathway for their disposition. Interestingly, this study shows that other variants, including those previously associated with altered pharmacokinetics, have normal or only modestly reduced protein abundance and transport function (1815). Nevertheless, the results of this study are expected to help predict the consequences of loss-of-function variants such as the aforementioned rare variant, which are difficult to detect in clinical studies.

Subfamily 3

SLCO3A1: OATP3A1 (SLCO3A1) is a ubiquitously expressed organic anion transporter (1816). It has been detected in numerous fetal and adult tissues (1817). In the central nervous system it has been localized in the choroid plexus and neurons as well as in the testis (1818) and heart (1819), among other organs. Physiological substrates of OATP3A1 include estrone-3-sulfate, prostaglandin E₁ and E₂, vasopressin, and thyroxine (1817, 1818, 1820). In addition to these endogenous substances, OATP3A1 is also able to transport drugs such as the HMG-CoA reductase inhibitor simvastatin (1819), the endothelin A receptor antagonist BQ-123, and the opioid analgesic peptide deltorphin II.

There are three different isoforms of OATP3A1, termed V1, V2 and V3, resulting from alternative splicing (1816). OATP3A1_V1, OATP3A1_V2 and OATP3A1_V3 have similar tissue distribution but distinct subcellular localization. They have similar substrate specificities and were able to transport prostaglandin E₁ and E₂, vasopressin, thyroxine and BQ-123 (1818), but OATP3A1_V3 additionally has dehydroepiandrosterone sulfate as a novel OATP3A1 transport substrate (1816). In terms of subcellular localization, OATP3A1_V1 is basolateral and OATP3A1_V2 and OATP3A1_V3 are apical in polarized MDCKII cells (1816).

Based on their distinct expression patterns but overlapping functions, OATP3A1 isoforms are proposed to contribute to transcellular (neuro)steroid transport in the central nervous system. For example, a strong signal for OATP3A1_V3 was detected in the outer cortex, where it is expressed in neurons, especially in axons, as well as in glial cells (1816). In addition, OATP3A1_V3 was shown to be highly expressed in the apical membrane of choroid plexus epithelial cells, while OATP3A1_V1 was localized to the basolateral membrane (1816) (Fig. 11). OATPs are usually bidirectional transporters that act as electroneutral exchangers of organic anions (178). Therefore, in the choroid plexus, the OATP3A1 isoform could in principle either remove organic anions into the blood for subsequent elimination by the liver or kidney, or deliver organic anions into the cerebrospinal fluid.

In the testis, expression of OATP3A1_V3 was observed in germ cells and at lower levels in Sertoli cells (1816).

SLCO3A1 has also been identified as a novel Crohn disease-associated gene. Specifically, it has been suggested that this transporter mediates inflammatory processes in intestinal epithelial cells through NF-κB transcriptional activation, resulting in a higher incidence of bowel perforation in Crohn disease patients (1821).

In addition, it has been reported that OATP3A1 functions as a bile acid efflux transporter and that its expression is upregulated in cholestatic liver through the FGF19-ERK/NFκB-SP1/p65 signaling pathways. Thus, this transporter serves as a bile acid efflux transporter that is up-regulated as an adaptive response to cholestasis (1822). It remains to be determined whether endogenous hepatic uptake substrates for OATP3A1 are required for exchange with bile acid efflux in the cholestatic liver, since, as mentioned above, OATPs are bidirectional transporters (1822).

Subfamily 4 (branches 4A and 4C)

Note: The name OATP4B1 is not used anymore. OATP4B1 is identical to OATP4C1.

SLCO4A1: OATP4A1 (SLCO4A1) acts as a transporter for a number of endogenous OATP substrates such as steroid hormone conjugates (estrone sulfate) and prostaglandins, including the major endogenous pro-inflammatory prostaglandin PGE₂, as well as thyroid hormones, xenobiotics, and drugs (1810). OATP4A1 is expressed in placenta and brain (1823). In human placenta, it may enable transplacental transfer of thyroid hormone. The HPA suggests highest expression in trophoblast cells of the placenta, but expression is also shown in renal tubules, lung, small intestine, stomach, esophagus, heart, skeletal muscle, skin, brain areas, and retina. SLCO4A1 was shown to be highly expressed in ovarian cancer (1824) but has also been shown to be expressed in other cancers, including colorectal cancer, and to play an important role in colorectal cancer cell proliferation and carcinogenesis (1825). OATP4A1 (SLCO4A1) was reported to promote colorectal cancer progression and to be regulated by miR-1224-5p (1826).

Increased OATP4A1 (SLCO4A1) expression has been shown to be associated with upregulation of specific inflammatory pathways in high-grade serous ovarian cancer, while decreased levels are associated with dysfunction of the mitochondrial electron transport chain pathway. These molecular pathophysiological conditions have been reported to be tumor-specific (1827).

SLCO4C1: OATP4C1 (SLCO4C1) is a kidney-specific organic anion transporter (1828, 1829) expressed in the basolaterale membrane of kidney proximal tubule cells (Fig. 33). OATP4C1 transports cardiac glycosides such as digoxin and ouabain, thyroid hormones, cAMP and methotrexate in a Na⁺-independent manner (1830). OATP4C1 has also been shown to excrete uremic toxins such as guanidino succinate, asymmetric dimethylarginine and trans-aconitate, resulting in lower blood pressure and reduced renal inflammation, and it has been suggested that drugs that upregulate SLCO4C1 may have therapeutic potential for patients with chronic kidney disease (CKD) (1831).

In a search for tumor suppressor genes in head and neck cancer that are inactivated by somatic mutation and promoter methylation, SLCO4C1 was identified (1832). Knockdown of SLCO4C1 in endometrial cancer cell lines revealed that this transporter inhibits cancer proliferation and metastasis by inactivating the PI3K/Akt signaling pathway (1833). SLCO4C1 was also found to function as a tumor suppressor in hepatocellular carcinoma cells, where SLCO4C1 downregulation promoted proliferation, invasion, migration and apoptosis. Anoikis resistance is a hallmark of cancer metastasis, and SLCO4C1 was found to be one of the genes in an anoikis prognostic model that serves as a marker for cancer survival (1834).

Subfamily 5

SLCO5A1 - Orphan transporter: OATP5A1 (SLCO5A1) expression has been detected in brain, heart and skeletal muscle (1835). According to the HPA, there is also expression in the prostate, thymus and tongue, and at the single cell level, the highest expression is in inhibitory neurons. The transport function of OATP5A1 remains unknown despite many interesting findings, as shown below, and extensive expression studies in Xenopus oocytes and HeLa cells (1836, 1837). It has even been suggested that OATP5A1 may have a non-classical function unrelated to membrane transport, e.g., one involving reorganization of cell shape, such as during differentiation and migration (1837).

SLCO5A1 expression has been detected in human bone tumors, prostate cancer (1838) and normal and cancerous breast tissue (1839). OATP5A1 has also been found in drug resistant small cell lung cancer cells (1840), and in primary liver cancer and in liver metastases from colorectal cancer (1841). OATP5A1 has been found at the plasma membrane of lactiferous duct epithelial cells in normal breast tissue and at the plasma membrane and in the cytoplasm of malignant breast tumor specimens (1842). OATP5A1 (SLCO5A1) has also been identified as a core immune-related prognostic marker of the clinicopathological features of uveal melanoma, an aggressive malignancy with poor prognosis (1843).

In addition, OATP5A1 has been reported to upregulate synapse assembly and organization genes that contribute to impulsivity in juvenile myoclonic epilepsy, and SLCO5A1 loss-of-function was proposed to be an impulsivity and seizure mechanism (1836).

Subfamily 6

SLCO6A1 - Orphan transporter: OATP6A1 (SLCO6A1) is highly expressed in the testis, followed by the spleen, brain, fetal brain, and placenta (1844). OATP6A1 has been identified as a gonad-specific cancer/testis antigen expressed in human lung cancer (1845). Its transport function is unknown.

Pharmacogenomics

Polymorphisms in the genes encoding members of the SLCO, SLC22, and SLC47 families have been implicated in modulating drug transport and drug distribution. For example, numerous chemotherapeutic drugs have been shown to be transported by OATP1A2, OATP1B1, OATP1B3 and OATP2B1 (1772). As mentioned above, certain SLCO genes are characterized by particularly high genetic variability, with a minor allele frequency of >5% identified for SLCO1B1, SLCO1B3, and SLCO1A2 (1790, 1846, 1847). Several of these variants affect transporter expression, localization, and/or function and can thus significantly alter drug disposition, shaping interindividual and interethnic differences. For example, OATP1B1 is important for transporting HMG-CoA reductase inhibitors, known as statins, to their site of action in hepatocytes, and variability in SLCO1B1 has important implications for statin pharmacokinetics, statin treatment response, and risk of statin-induced myopathy (1790).

Orphan transporter family members (3)

SLCO1B7 (OATP1B7), SLCO5A1 (OATP5A1), SLCO6A1 (OATP6A1)

SLC22 Organic cation/anion/zwitterion transporter family (2.A.1.19/Sugar_tr/MFS)

Discovery: The founding member of this family is the rat kidney organic cation transporter OCT1 (SLC22A1), which was identified by expression cloning using Xenopus oocytes (150). It is mainly expressed in the sinusoidal membrane of hepatocytes. Also identified by expression cloning was the winter flounder renal organic anion transporter fROAT (1848).

Gene family members (26 + 1 pseudogene):
SLC22A1 (OCT1)	SLC22A9 (OAT7)	SLC22A16 (CT2)	SV2A (SLC22B1)
SLC22A2 (OCT2)	SLC22A10 (OAT5, gene/pseudogene)	SLC22A17 (BOIT)	SV2B (SLC22B2)
SLC22A3 (OCT3)		SLC22A20P (OAT6, pseudogene)	SV2C (SLC22B3)
SLC22A4 (OCTN1)	SLC22A11 (OAT4)		SVOP (SLC22B4)
SLC22A5 (OCTN2)	SLC22A12 (URAT1)	SLC22A23 (Nritp)	SVOPL (SLC22B5)
SLC22A6 (OAT1)	SLC22A13 (OAT10)	SLC22A24 (NET46)
SLC22A7 (OAT2)	SLC22A14 (OCTL2)	SLC22A25 (UST6)
SLC22A8 (OAT3)	SLC22A15 (FLIPT1)	SLC22A31

Molecular aspects, physiological roles and links to disease

The SLC22 family consists of 25 members (plus two pseudogenes). They include organic cation transporters (OCTs), organic zwitterion/cation transporters (OCTNs), and organic anion transporters (OATs). The transporters function via different mechanisms, including facilitated diffusion as uniporters (e.g., OCTs, some OCTNs), anion exchange (e.g., OATs), or Na⁺/zwitterion cotransport (e.g., certain OCTNs)(1849, 1850). The family also includes atypical SLC22 members (SLC22B series).

The SLC22 family belongs to the Organic Cation Transporter (OCT) family (TC 2.A.1.19), which is part of the MFS superfamily that typically has 12 TMHs. The structural basis of the SLC22 organic cation transporters was revealed by the cryo-EM structure of human OCT3 bound with and without the inhibitors corticosterone and decynium-22 (1851) (see the SLC22A3 description for more details).

Several of the SLC22 transporters are extensively involved in small intestinal absorption and hepatic and renal excretion of drugs, and their functional expression therefore influences the pharmacokinetics, clinical efficacy and safety of drugs. Regulatory guidelines from the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) recommend the evaluation of OAT transporter interactions with new chemical entities and circulating metabolites, especially when significant renal elimination of negatively charged drugs is observed. To this end, the International Transporter Consortium (ITC) was established to identify a subset of clinically relevant transporters and outline decision criteria that should be used to predict the clinical significance of changes in transporter activity (1852) (https://www.itc-transporter.org/). Subsequent ITC publications have highlighted the development of tools and approaches to address drug development issues and provided the most up-to-date recommendations on data evaluation and decision points (41, 1732, 1853–1856). The SLC22 family has been divided into 6 groups as illustrated in the phylogenetic tree (Fig. 32).

Fig. 32. Phylogenetic tree of the SLC22 family.

Group 1 – Electrogenic cation transporters (1857)

SLC22A1: OCT1 (SLC22A1) is a hepatic uptake transporter expressed on the sinusoidal membrane of hepatocytes, facilitating the disposition and hepatic clearance of cationic drugs and endogenous cationic compounds (1858, 1859) (Fig. 33). Its importance lies in its role as the primary step in active hepatic elimination of cationic drugs. It works together with MATE1 (SLC47A1) to facilitate the biliary elimination of OCT1 substrates transported into the liver (1860).

Fig. 33. SLC solute carriers in the intestine, kidney, liver, blood-brain barrier, and lung, with implications for drug metabolism and therapeutics.

Also shown are the following ABC transporters mentioned in the text: BSEP (ABCB11) [see descriptions of NTCP (SLC10A1) and ASBT (SLC10A2)]; MRP4 (ABCC4) [see descriptions of GLUT9 (SLC2A9) and OATP2A1 (SLCO2A1)]; MRP6 (ABCC6) [see description of ANKH (SLC62A1)]; BCRP (ABCG2) [see description of GLUT9 (SLC2A9)]; and CFTR (ABCC7) [see description of pendrin (SLC26A4)]. Intestinal absorption of SCFAs (short-chain fatty acids) is believed to occur mainly in the colon via SMCT1 (SLC5A8) and/or SMCT2 (SLC5A12) in the apical membrane and MCT1 (SLC16A1) in the basolateral membrane. Note that the subcellular localization of OATP2B1 in enterocytes is still unclear. Also note that the localization of THTR2 (SLC19A3) at the BBB may be luminal (blood-facing) in humans; see discussion in the descriptions of THTR2 (SLC19A3) and THTR1 (SLC19A2). UA^-, urate (produced in hepatocytes via xanthine oxidase); MC, monocarboxylate; BA^-, bile acid; PEG₂, prostaglandin E2, ACE2, angiotensin-converting enzyme 2, SIRT1, sirtuin 1 (a metabolic sensor).

The antidiabetic drug metformin is a major substrate of OCT1, and genetic polymorphisms of OCT1 have been associated with altered metformin pharmacokinetics, safety, and efficacy, although functional SNPs in other genes are also important (1861). Subsequent studies have shown that the polymorphisms rs622342 and rs72552763 of the SLC22A1 gene are associated with variations in hemoglobin A1c (HbA1c) levels and metformin plasma concentrations in patients with T2D, providing further support for the hypothesis that genetic variants influence metformin response (1862).

Primary human hepatocytes (PHH) are key to drug discovery, but lose function when cultured. In contrast, PHH spheroids can be cultured under physiologically relevant conditions, ensuring the expression and function of drug metabolizing enzymes and drug transport proteins. OCT1 has been used as a model to study both transport kinetics and the long-term regulation of transporter activity by physiologically relevant signaling pathways (1863). Such an improvement in the in vitro testing protocol for therapeutic drugs is considered a significant step forward for the benefit of patients.

SLC22A2: OCT2 (SLC22A2) is highly and almost exclusively expressed on the basolateral side of proximal tubule cells (1850) (Fig. 33). It facilitates renal secretion of mostly cationic drugs and endogenous compounds in concert with MATE1 (SLC47A1) and MATE2-K (a splice variant of SLC47A2) located at the apical membrane (1864) to secrete OCT2 substrates into the urine. The transporter is polyspecific, facilitating diffusional movement in either direction, with the driving force presumed to be the electrical gradient of cationic compounds (1850). Although OCT2 is bi-directional, it usually functions as an uptake transporter. A wide range of drugs have been observed to function as substrates or inhibitors of OCT2. Notably, drug substrates for OCT2 include cisplatin and oxaliplatin for chemotherapeutics, metformin for oral antidiabetic medication, cimetidine and ranitidine for proton-pump inhibitors, lamivudine for antivirals, and dofetilide as an antiarrhythmic drug. Overall, assessing OCT2 function is crucial in determining the active renal secretion of cationic drugs, and both the FDA and EMA recommend evaluating OCT2 for drugs with high renal elimination.

Cisplatin, which is administered intraperitoneally, is used as a first-line chemotherapy treatment for patients diagnosed with various types of cancer. It is an organometallic platinum coordination complex that interferes with DNA replication (1865, 1866). Due to the high chloride concentration of the blood of about 100 mM, free cisplatin exists in its intact form. Once inside a cell, cisplatin becomes cytotoxic by losing a chloride ligand, allowing it to bind to DNA and form DNA adducts, followed by inhibition of DNA synthesis and cell growth. Several membrane transporters have been implicated in the transport of cisplatin into cells, including Na⁺, K⁺-ATPase and the SLC solute carriers OCT2 (SLC22A2) and CTR1 (SLC31A1) (see also the SLC31 family description for details). Cisplatin nephrotoxicity is a major clinical problem leading to acute kidney injury due to cisplatin accumulation in renal tubular cells (1867). (1865). OCT2 plays a critical role in the development of the nephrotoxic effect of cisplatin by increasing drug uptake in kidney cells (1868). In vivo studies have shown that OCT2 in humans and Oct1 and Oct2 in mice are essential for the active secretion of cisplatin across the basolateral membrane of renal proximal tubule cells (1869) (Fig. 33), possibly together with the copper transporter CTR1 (SLC31A1) (1865). Cisplatin is then transported across the apical membrane into the lumen via MATE1 (SLC47A1). Alternatively, it is further conjugated with glutathione (GSH) in the proximal tubule cells and subsequently secreted by the multidrug resistance-related protein MRP2 (ABCC2) (1868).

The antihyperglycemic drug metformin exerts its glucose-lowering effect by inhibiting hepatic gluconeogenesis (1870). It enters enterocytes via ENT4/PMAT (SLC29A4) and OCT3 located in the apical brush border membrane (1871). However, its release through the basolateral membrane has been considered inefficient due to the absence of intestinal efflux transporters leading to accumulation of the drug inside enterocytes (1872), and thus it has been proposed that metformin predominantly takes the paracellular route across the intestine to access the circulating blood. From there, metformin enters hepatocytes via OCT1 (SLC22A1) and OCT3 (SLC22A3) (1873). Metformin, which is not subject to significant hepatic metabolism or biliary excretion, is cleared by renal excretion exclusively via OCT2 located in the basolateral membrane. This is followed by extrusion through the apical membrane via MATE1 (SLC47A1) and MATE2 (SLC47A2). Genetic polymorphisms of SLC22A2 have been found to have positive and negative effects on the glucose-lowering effect of metformin due to alterations in renal clearance. These findings may benefit patients with such polymorphisms through individualized therapy (1873).

SLC22A3: OCT3 (SLC22A3) is an uptake transporter with widespread tissue distribution (1858, 1874). OCT3 is involved in biliary and renal excretion of drugs. Endogenous substrates of OCT3 include epinephrine, norepinephrine, histamine, and agmatine. Drug substrates include the antiarrhythmic agents lidocaine and quinidine, the antidiabetic agent metformin, the adrenergic agonist etilefrine, the chemotherapeutic agent oxaliplatin, and the dopamine neurotoxin MPP⁺, which induces parkinsonism (1849, 1851). OCT3 also plays an important role in human skin and its genetic variants were shown to contribute to the variability of expression and activities of OCT3 in human skin. (1875). Studies of a genetic variant of SLC22A3 that leads to reduced expression of OCT3 indicated that this transporter contributes to variability of sebum levels in skin, regulation of sweating, and cell proliferation.

In the brain, OCT3 is expressed in glial and neuronal cells where it plays a role in the transport of neurotransmitters such as norepinephrine, epinephrine, dopamine, and serotonin (1876). The roles of the organic cation transporters OCT1, OCT2 and OCT3 have been investigated in psychiatric and psychostimulant abuse. Data suggest an important role for OCT3 in the actions of amphetamine-type stimulants, ethanol, and as a site of action for stress (corticosterone)- and ethanol-induced enhancement of neurochemical reward, as well as in the reinforcing effects of cocaine (1877). In the latter case, while cocaine has no direct effect on OCT3, corticosterone has been reported to enhance the ability of cocaine to increase dopamine signaling and reinstate cocaine seeking (1878).

As mentioned in the SLC22 introduction, cryo-EM structures of human OCT3 (SLC22A3) bound with and without the two inhibitors corticosterone and decynium-22 have been reported (1851). The structure revealed a classical MFS fold for OCT3, with the transporter composed of twelve TMHs in an outward-facing conformation. The translocation pathway was shown to be located at the interface of the two 2-fold pseudo-symmetrically related transmembrane domains consisting of TMHs 1-6 and TMHs 7-12. The substrate binding site is located in the center of the transporter between the two domains, halfway across the membrane. The structure shows a prominent, partially resolved density of the extended extracellular ectodomain (ECD), which is also characteristic of certain other MFS transporters. Its role in OCT3 is currently unexplored, but is likely to be involved in molecular gating or protein-protein interactions. In particular, the ECD is likely to play a role in the recruitment of CD63, a binding protein known to interact with OCTs and regulate their trafficking to the plasma membrane or intracellular compartments (1879). Indeed (1880) an intact ECD has been shown to be critical for oligomerization and trafficking to the plasma membrane for OCT1 (1880) and OCT2 (1881). When expressed in human embryonic kidney 293 cells, human OCT3 transported its substrate MPP⁺ and showed inhibition by two key molecules, decynium-22 (a cationic derivative of quinoline) and corticosterone (1851). The cryo-EM structural insights pave the way for a better understanding of OCT3 function and reveal the exact mechanisms that confer polyspecificity to the entire SLC22 family (1882). Understanding the structure-function relationships of SLC22 transporters will have direct therapeutic implications and will help to predict and prevent unwanted interactions with newly developed drugs.

Group 2 – Zwitterion transporters (1857)

SLC22A4: OCTN1 (SLC22A4) has been reported to be ubiquitously expressed, including in liver, intestine, kidney, brain, lung and muscle, and is located on the apical membrane of proximal tubule cells and on the luminal side of respiratory epithelial cells (1883, 1884). However, according to the HPA, it is only ubiquitously expressed at moderate to low levels in tissues such as intestine, kidney, muscle, brain, prostate, but highest expression is shown in bone marrow and at the single cell level in ciliated cells of lung, bronchus and oviduct, proximal enterocytes, neutrophils and monocytes. A systematic review of all known information on the localization of OCTN1 (SLC22A4) concludes that this transporter is not ubiquitously expressed, but is highly expressed in neutrophils, monocytes and developing red blood cells. It has been reported to function as either an organic cation/H⁺ proton exchanger, a cation exchanger, or a Na⁺-dependent or Na⁺-independent zwitterion transporter, protecting cells and tissues from oxidative and/or inflammatory damage (1849, 1885). Carnitine, which is required for the transfer of long-chain fatty acids across the inner mitochondrial membrane for subsequent β-oxidation, can serve as a substrate for OCTN1. However, the antioxidant amino acid ergothioneine (a thiourea derivative of histidine containing a sulfur atom on the imidazole ring) has been reported to be the most potent substrate of this transporter (1886, 1887). In the kidney, OCTN1 participates in actively secreting and reabsorbing small organic cations and zwitterions, including carnitine and ergothioneine.

OCTN1 (SLC22A4) is also known for its involvement in chronic inflammatory diseases such as asthma, Crohn disease and rheumatoid arthritis (1888). OCTN1 may be required for the transport of gut microbiome products, and it has been hypothesized to be involved in polyamine transport, which has subsequently been experimentally demonstrated (1889). OCTN1 may also be involved in the release of polyamines (1890). Thus, the link between polyamines and inflammation is also related to the function of OCTN1 (1891). Polyamines such as putrescine, spermidine and spermine are multifunctional polycations that interact with a variety of molecular targets to influence cellular processes such as cell growth, embryonic development, spermatogenesis and DNA repair (1891–1893). As mentioned above, polyamines are derived not only from cell biosynthesis, but also from diet and gut microbiota, and are absorbed and distributed to tissues by specific transporters.

In line with the importance of polyamines, three SLCs have been identified as polyamine transporters: the vesicular polyamine transporter SLC18B1 (VPAT), SLC22A4 (OCTN1) and the H⁺-coupled polyamine and sugar transporter SLC45A4 (1891, 1894).

OCTN1 is thought to be involved in the apical uptake transport of the antiepileptic drug gabapentin in the intestine and kidney (1895). OCTN2 (SLC22A5) (see below) and to a lesser extent OCTN1 transport important respiratory drugs such as the bronchodilator ipratropium and the muscarinic antagonist tiotropium. Expression of these transporters in the lung may have an impact on the disposition and absorption of these drugs in the pulmonary system (1896). Both OCTN1 and OCTN2 are involved in renal disposition and tubular reabsorption of hepatitis B drug entecavir (1897). OCTN1 also transports the tyrosine kinase inhibitor saracatinib, a therapeutic drug for the treatment of rheumatoid arthritis (1898).

SLC22A5: OCTN2 (SLC22A5), also known as CT1 (carnitine transporter 1), is a high-affinity carnitine transporter expressed in kidney and other tissues (1899). According to the HPA, in addition to kidney, there is abundant expression in intestine (enterocytes), skeletal and cardiac muscle, and brain areas such as the cerebellum. In the mouse brain, this transporter has been shown to be expressed in astrocytes, neurons and oligodendrocytes from the developing and mature forebrain (1900). In bovine BBB forming capillary endothelial cells it has been proposed to be located in both the apical and basolateral membranes together with ATB^0,+ (SLC6A14), which is also proposed to transport carnitine (1901). OCTN2 transports carnitine in a Na⁺-dependent manner and with an activity higher than that of OCTN1 (1902).

Carnitine homeostasis is critical for cellular metabolism, acting as a shuttle of acyl groups for fatty acid oxidation. Carnitine can be converted to acylcarnitine by acyltransferase isoenzymes in various subcellular compartments. In the heart, acylcarnitine may serve as a reservoir of activated acyl groups that can be transferred to CoA to provide an immediate energy source via fatty acid oxidation (1903). In addition to OCTN2, specific transporters that mediate the flux of carnitine and its derivatives across cell membranes are CT2 (SLC22A16), ATB^0,+ (SLC6A14), and probably MCT9 (SLC16A9), together with the mitochondrial carrier CAC (SLC25A20). These proteins form a carnitine network that plays a central role in driving fatty acid β-oxidation in mitochondria to maintain energy homeostasis in the human body. The network is also highly relevant to cancer metabolism (1904).

Carnitine is a critical intracellular cofactor for the transport of activated long-chain fatty acids into the mitochondrial matrix, thereby facilitating β-oxidation and enhancing energy production. The body meets only a quarter of its carnitine needs through endogenous biosynthesis, mainly in the liver, kidney, and to some extent in the brain, while the rest must be absorbed from the diet through the consumption of meat, dairy products, vegetables, etc. via OCTN2 expressed in the apical membrane of enterocytes (1905, 1906). Thus, carnitine homeostasis is maintained by a balance between intestinal absorption, synthesis and renal reabsorption (1904, 1907, 1908).

Because dietary carnitine intake and renal reabsorption compensate for reduced carnitine synthesis, individuals with defects in carnitine biosynthesis do not appear to have carnitine deficiency under normal dietary conditions (El-Hattab and Scaglia, 2015). However, patients with SLC22A5 loss-of-function mutations suffer from primary carnitine deficiency, an autosomal recessive disorder characterized by progressive cardiomyopathy, skeletal myopathy, hypoglycemia, and hyperammonemia (1909).

Genetic epidemiologic studies have also shown an association between mutations in the carnitine transporter genes OCTN2 and also OCTN1 and a propensity to develop Crohn disease as a result of carnitine deficiency in the intestinal epithelium (1910). Therefore, it has been suggested that carnitine supplementation as a means of enhancing fatty acid oxidation may be therapeutically beneficial in patients with IBD.

SLC22A16: CT2 (SLC22A16) is a high-affinity L-carnitine transporter that is abundantly expressed in the male reproductive tract and bone marrow (1911, 1912). In the male reproductive system, free carnitine is highly enriched in epididymal tissue and semen (1913, 1914). The epididymis serves as a site for sperm maturation and storage, where the levels of free carnitine directly influence the processes of sperm maturation and metabolism, ultimately affecting sperm motility and fertilization ability (1913–1915). In addition, L-carnitine acts as a mitochondrial protector by regulating the acetyl-CoA/CoA ratio and preventing ROS-mediated cell apoptosis, among other functions (1916). Deficiencies in carnitine or its transporters are associated with asthenozoospermia, reduced sperm quality, and suboptimal fertility outcomes in couples (1917). CT2 (SLC22A16) has been shown to be localized to the luminal membrane of epididymal cells and the plasma membrane of Sertoli cells in the testis (1911, 1917). High expression in the epididymis may suggest that it serves as an antiporter exporting polyamines in exchange for L-carnitine during sperm capacitation (1918).

CT2 (SLC22A16) is also prominently expressed in hematopoietic tissues, including CD34⁺ cells and leukemia cells (1919). Acute myeloid leukemia (AML) cells have a unique dependence on mitochondrial metabolism and fatty acid oxidation. SLC22A16 knockdown reduced basal oxygen consumption without a concomitant increase in glycolysis (1920). It was concluded that blocking fatty acid oxidation via CT2 may be a potential therapeutic strategy to target these malignant cells.

In addition, CT2 mediates cisplatin uptake and resistance to cisplatin in lung cancer (1921).

Slc22a21 (mouse only): Mouse Octn3 (Slc22a21) is specifically expressed in the testis. Human CT2 (SLC22A16) shows about 33% similarity to mouse Octn3, shares similar functional properties of Na⁺-independence and high affinity for carnitine, and is also expressed in the testis (1922).

SLC22A15: FLIPT1 (SLC22A15) is ubiquitously expressed with the highest expression in bone marrow and brain (1857). It prefers zwitterionic compounds over cations and anions (1923). The zwitterions ergothioneine, carnitine, carnosine, gabapentin and MPP⁺, thiamine and cimetidine were identified as substrates. SLC22A15-mediated transport of several substrates was found to be Na⁺-dependent (1923). It has been proposed that SLC22A15, like OCTN1 (SLC22A4), is primarily involved in the transport of the zwitterion ergothioneine, a sulfur-containing histidine derivative with potent antioxidant activity, leading to ergothioneine deposition in the brain (1857, 1887). Transport of several substrates was Na⁺ dependent. According to RNAseq data, it is highly expressed in oligodendrocytes and at lower levels in excitatory and inhibitory neurons and in astrocytes (1923).

Group 3 – Anion transporters

Branch a:

SLC22A6, SLC22A8: OAT1 (SLC22A6) and OAT3 (SLC22A8) are primarily renal anion-exchanging antiporters expressed in the basolateral membranes of renal proximal tubule cells and are responsible for transporting substrates against their concentration gradients into the proximal tubule cells in exchange for α-ketoglutarate and other dicarboxylates (Fig. 33) (1849, 1924–1926). Exported intracellular dicarboxylates are replenished by the Na⁺-coupled dicarboxylate transporter NaDC (SLC13A3) (1927).

In the kidney, transporters located in the tubular epithelium are involved in the disposition and excretion of prescription drugs and their metabolites, and OAT1 (SLC22A6) and OAT3 (SLC22A8), expressed on the basolateral membrane of the renal proximal tubules, are important secretory transporters (1926). This includes the secretion of therapeutic drugs such as beta-lactam antibiotics, loop diuretics, non-steroidal anti-inflammatory drugs and antiviral nucleoside analogues (1928). Due to their roles in drug accumulation in the kidney as well as overall drug elimination, OAT1 and OAT3 are important determinants of the efficacy, side effects and renal toxicities of a number of drugs.

OAT1 and OAT3 have overlapping specificities for endogenous substrates and drugs (41, 1852) and thus both are often involved in drug-drug interactions (DDIs) (1928). DDIs affect drug safety and efficacy by modulating drug levels. Concomitant medications that result in DDIs have been associated with serious adverse drug reactions, leading to the withdrawal of drugs such as terfenadine and cerivastatin. DDIs are often mediated by drug-metabolizing enzymes, but transporters in the intestine, liver, and kidney are also important targets. Concomitant medications that inhibit these transporters may result in high drug levels and adverse effects.

Important drug substrates for both OAT1 and OAT3 include the antibiotic tetracycline, the antiviral drugs adefovir, cidofovir, tenofovir, the histamine receptor subtype 2 antagonist cimetidine, the loop diuretics bumetanide and furosemide, the nonsteroidal anti-inflammatory drugs ibuprofen, indomethacin, ketoprofen (1849). Inhibitors of OAT1/OAT3 include the classical uricosuric drug probenecid used in the treatment of gout (1929), the antibiotic rifampicin (1930), also natural phenylpropanoids and flavonoids found in common dietary and herbal supplements (1931), and the β-lactam penicillin antibiotic cloxacillin, which is both a substrate and a clinically relevant inhibitor of OAT3 (1932). Inhibition of OATs results in reduced or delayed renal elimination and as noted above, the resulting DDIs at these transporters can alter the pharmacokinetics and toxicity of drugs. DDIs are of particular concern for drugs with a narrow therapeutic index, such as methotrexate (1933), where small dose differences can lead to therapeutic failure or adverse effects.

OAT3 (SLC22A8) is also highly expressed in the choroid plexus, where it is thought to play an important role in the extraction of drugs and toxic compounds from CSF to blood across the apical membrane of choroid plexus epithelial cells (1934, 1935) (Fig. 11).

The concept of “remote sensing and signaling” has been reported for organic anion transporters and arose from the analysis of SLC22 family members, particularly OAT1 and OAT3, which have overlapping substrate specificity and broad tissue expression in major epithelial tissues that separate body fluids (1936). Because these proteins have similar substrates and inhibitors, it was proposed that they work together to facilitate the movement of certain classes of compounds in the human body, facilitating organ crosstalk and thus serving as major hubs in a body-wide signaling network. For example, OAT1 and OAT3 have been shown to regulate circulating levels of toxins, such as mercury conjugates, mycotoxins, and gut microbial-derived metabolites, and it is believed that OAT1 and OAT3 have pronounced roles in the handling of exogenous compounds, including drugs, natural products, and toxins (1936).

To elucidate the molecular basis for the selective uptake and elimination of organic anions in the kidney by OAT1, cryo-EM structures of rat OAT1 bound to α-ketoglutarate, the antiviral tenofovir and the inhibitor probenecid were reported (1937). OAT1 adopts an inward open state and forms a binding site in the center of the membrane within the 12 canonical TMHs of the MFS fold. In vivo cellular assays have explained the molecular basis for α-ketoglutarate-driven drug elimination and the allosteric regulation of organic anion transport in the kidney by Cl^- (1937). Cl^- has been identified as an allosteric regulator of OAT1 and OAT3, but has no effect on OAT2 (1924, 1925, 1938). While probenecid inhibits OAT1 transport, Cl^- binding enhances it. Cryo-EM structural work provides new insight into the role of Cl^-, although the precise mechanisms linking metabolite cycling, drug transport, and intracellular Cl^- still remain unclear (1937).

SLC22A20P - Pseudogene: This gene is currently considered to be a unitary pseudogene in human, with coding orthologs in most other vertebrates. The function of rodent OAT6 (Slc22a20) has been described, but there is no reported functionality of the human homologue (1939).

Branch b:

SLC22A11: OAT4 (SLC22A11) has been reported to function as a broad specificity bidirectional organic anion/dicarboxylate exchanger with endogenous substrates such as steroid sulfate conjugates and xenobiotic substrates such as drugs (1849, 1940, 1941). In the kidney, OAT4 localized to the luminal side of proximal tubules contributes to both the reabsorption and secretion of endogenous substances, as well as drugs and other xenobiotics (1940, 1942). In the placenta, OAT4 is believed to be important for transferring toxic anionic compounds from fetal to maternal circulation (1849)

Organic anion secretion begins at the blood side of the proximal tubule cells via the organic anion transporters OAT1-3 (SLC22A6-8) at the basolateral membrane, which mediate this step by exchanging an extracellular organic anion for an intracellular dicarboxylate such as α-ketoglutarate (1943). The exit of organic anions into the tubular filtrate has long been unclear. It was generally assumed that OAT4 in the apical membrane is mainly involved in the reabsorption of urate and certain anionic drugs and toxicants (1944, 1945). However, subsequent evidence has shown that OAT4 interacts asymmetrically with anionic substrates, and OAT4 has been proposed to be an important element of renal organic anion secretion as well (1946). Studies have linked OAT4 to the secretion of dianionic angiotensin II receptor blockers (e.g., olmesartan and azilsartan) (1947). OAT4 has been proposed to use extracellular Cl^- as a counterpart for dianion efflux (1947). In contrast, OAT4-mediated uptake of the monoanionic angiotensin II receptor blocker losartan was little affected by extracellular Cl^-. This finding suggests that only OAT4-mediated dianion transport is Cl^- sensitive (1947) and that OAT4 facilitates losartan reabsorption rather than secretion due to lack of Cl^- exchange. Instead, it has been proposed that URAT1 (SLC22A12) may contribute significantly to losartan secretion (1947). Thus, in addition to its reabsorptive function of monoanionic compounds, OAT4 appears to play an important role in the renal secretion of dianionic compounds in exchange for Cl^-.

SLC22A12: URAT1 (SLC22A12) is a urate transporter whose function has been extensively studied in the proximal tubules of the kidney (1948, 1949) (Fig. 33). It is expressed almost exclusively in the proximal tubules of the kidney, according to the HPA.

URAT1 (SLC22A12) was identified in studies of Japanese patients with exercise-induced acute kidney injury, and is an anion-exchanging uptake transporter localized at the apical membrane of renal proximal tubular cells (1948). URAT1 is localized to the apical membrane of the S1 segment of the proximal tubules where it plays a key role as 98% of filtered urate is reabsorbed from this segment and thus URAT1 plays an important role in uric acid homeostasis (1948, 1950). While URAT1 is the major pathway for apical urate reabsorption, GLUT9 (SLC2A9) is the principal pathway for basolateral urate exit from the proximal tubule cell in the human kidney (1951).

Urate transport by URAT1 is driven by intracellular lactate via an exchange mechanism. Lactate is accumulated in the proximal tubule cells by the Na⁺-coupled monocarboxylate transporters SMCT1 (SLC5A8) and SMCT2 (SLC5A12) at the apical membrane and the monocarboxylate transporter MCT1 (SLC16A1) at the basolateral membrane (Fig. 33). Since URAT1 exchanges urate for some of the same monocarboxylates that SMCT1/2 and MCT1 transport, its transport function is indeed driven by these additional transporters. Similarly, organic anion transport mediated by OAT10 (SLC22A13) is driven by SMCT1/2 and MCT1 (see below and Fig. 33).

URAT1 is proposed to be part of a multimolecular complex that allows the cooperation of several transporters by interacting with PDZK1 as a binding partner (1952). PDZK1 acts as a scaffold protein to regulate the activity of various transport proteins including URAT1 in the renal proximal tubules (1953).

Circulating uric acid is thought to protect the body from oxidative damage. Conversely, excess serum uric acid leads to gout. It has been found that loss-of-function variants of SLC22A12 cause renal hypouricemia (RHUC), a heterogeneous inherited disorder characterized by impaired tubular uric acid reabsorption with severe complications, such as acute kidney injury. Exercise-induced acute kidney injury is a known complication of RHUC. It is caused by accelerated breakdown of purine nucleotides during strenuous muscular exercise, resulting in increased production of uric acid (1948, 1954). On the other hand, dysfunctional variants of URAT1 (SLC22A12) have substantial protective anti-gout effects (584). Therefore, URAT1 inhibitors are highly potent and promising uricosuric agents for the treatment of hyperuricemia. SHR4640, also named ruzinurad, is a selective URAT1 inhibitor developed for the treatment of hyperuricemia and gout (76).

Branch c:

SLC22A7: OAT2 (SLC22A7) is found predominantly on the sinusoidal membrane of liver cells and to a lesser extent also on the basolateral membrane of the renal proximal tubule epithelium (1938, 1955, 1956). It functions as an organic anion and dicarboxylate exchanger, although its precise transport mechanism has yet to be fully clarified. OAT2 facilitates the transport of various endogenous compounds such as dicarboxylic acids, nucleotides, uric acid, lipid hormones, mediators, and creatinine, as well as xenobiotics such as anticancer drugs, antivirals, and antibiotics (1957). Furthermore, OAT2 (SLC22A7) enables the bidirectional transport of glutamate, indicating that hepatocytes and OAT2 can facilitate the release of hepatic glutamate into the blood circulation. Additionally, OAT2 transports pyrimidine bases, nucleosides, and nucleotides.

Furthermore, OAT2 (SLC22A7) has been demonstrated to be the primary transporter responsible for the uptake of nicotinic acid, also known as niacin or vitamin B3 (1958). After intestinal absorption of nicotinic acid via SMCT1 (SLC5A8) (Fig. 30), OAT2 delivers it to the liver and other tissues, where it serves as a precursor for NAD⁺ synthesis. In the kidney, OAT2 may play a role in renal nicotinic acid secretion (1958).

Branch d:

SLC22A9: OAT7 (SLC22A9) is a liver-specific transporter localized to the sinusoidal membrane of hepatocytes that facilitates the exchange of sulfate conjugates for the short-chain fatty acid butyrate (1959) (Fig. 33). It transports estrone sulfate, the major circulating estrogen metabolite and a precursor of active estrogen, and dehydroepiandrosterone sulfate (1959, 1960). It also transports the statin drug pravastatin (1961).

SLC22A25 - Orphan transporter: UST6 (SLC22A25) is an orphan transporter that is abundantly expressed in the liver and is significantly associated with atrial fibrillation based on an exome-wide association study (1962).

SLC22A10 – Gene/pseudogene: OAT5 (SLC22A10) was provisionally named OAT5 after its identification (1956, 1963, 1964) and its mRNA expression was reported to be abundant and specific in the liver (1956, 1963), including fetal liver (1965). However, transport substrates could not be identified and the cellular and subcellular localization of the OAT5 protein remained unknown (1963). Subsequently it was reported that an allele present in 98% of genomes results in a truncated protein that cannot function as a transporter (1966). A further SNP found in up to 50% in some populations can result in an unstable and likely degraded protein, so the gene is now classified as a polymorphic or segregating pseudogene. However, it cannot be excluded that a combination of SNPs exist in some individuals that in fact encode a functional protein.

SLC22A24: NET46 (SLC22A24) is a renal organic anion transporter. It is localized in the S3 segments of the renal proximal tubules where it functions in the reabsorption of steroid compounds with anionic moieties (sulfates, carboxylic acids, glucuronides) in exchange for dicarboxylates such as glutarate or succinate (1967, 1968). In particular, it facilitates the reabsorption of the conjugated steroids estradiol 17β-D-glucuronide, androstanediol glucuronide, and estrone 3-sulfate, as well as bile acids.

Slc22a19 (rodent only): Rodent Oat5 (Slc22a19) mediates excretion or reabsorption of steroid sulfates in mouse kidney (1969). There is no ortholog in humans.

Slc22a22 (rodent only): OAT-PG (Slc22a22) is exclusively expressed in the proximal tubules of the rat renal cortex and localizes to the basolateral membrane where it mediates PGE₂ clearance to regulate the physiological function of PGE₂ in the renal cortex (1970, 1971).

Slc22a26 to Slc22a30 (rodent only): These are putative organic anion transporters as part of a cluster on mouse chromosome 19q that do not exist in humans. The substrates for some of these transporters have not yet been defined (1939).

Branch e)

SLC22A13: OAT10 (SLC22A13) is a transporter highly expressed in renal tubules and transports organic anions including nicotinate, β-hydroxybutyrate, p-aminohippurate, and orotate (1972) (Fig. 33). Similar to URAT1 (see above), OAT10-mediated uptake of organic anions is driven by intracellular monocarboxylates taken up by MCT1 in the basolateral membrane and SMCT1 (SLC5A8) and SMCT2 (SLC5A12) in the apical membrane, as these transporters share common substrates (1972) (Fig. 33).

Group 4 – Mitochondrial riboflavin transporter

SLC22A14: OCTL2 (SLC22A14) has been reported to function as a mitochondrial riboflavin transporter required for sperm oxidative phosphorylation and male fertility, and Slc22a14 deficiency in mice results in decreased sperm motility and male infertility (1973). According to the HPA, it is also not exclusively expressed in early and late spermatids, but also in brain areas at low levels. SLC22A14-mediated riboflavin delivery into spermatids is likely important for energy generation and motility of the resulting spermatozoa. Riboflavin is the precursor of FMN and FAD, which are coenzymes of many enzymes in the TCA cycle in mitochondria.

Group 5 – Orphan transporters

SLC22A17 - Orphan transporter: BOIT (SLC22A17), also known as neutrophil gelatinase-associated lipocalin receptor (NgalR), stands out as an interesting orphan transporter that has been shown to bind iron-bound lipocalin 2 (LCN2) (1974). LCN2 is a circulatory protein responsible for the transportation of small and hydrophobic molecules (steroids, free fatty acids, prostaglandins and hormones) to target organs after binding to megalin/glycoprotein GP330 and SLC22A17 (1975). LCN2 is known to be involved in iron metabolism by binding to bacterial and mammalian siderophores (small iron-binding molecules) present in the body (1976). SLC22A17 binds to and mediates endocytosis of filtered protein in the kidney and assists in iron uptake, but does not drive transport of the usual substrates of the SLC22 family in its function as a cell surface receptor. Although it has been shown to be involved in endocytosis, SLC22A17 is still considered an orphan transporter because no specific substrates or transport mechanisms have been identified.

SLC22A23 – Orphan transporter: Nritp (SLC22A23) is an orphan transporter expressed in various tissues (1977). SNPs of SLC22A23 were associated with IBD, endometriosis-related infertility and the clearance of antipsychotic drugs, and it may play a role in the development of laryngeal cancer (1977, 1978).

SLC22A31 - Orphan transporter: SLC22A31 is an orphan transporter that has been shown to be associated with COVID-19 severity (1979). The HPA indicates that it is highly expressed in cerebellum and at lower levels in lung, pancreas, thyroid, male and female tissues, among others. It has also been reported to be involved in the progression of thyroid cancer: zinc finger protein ZNF217 was shown to mediate transcriptional activation of grainyhead like transcription factor 3 to regulate SLC22A31 and promote malignant progression in thyroid cancer cells (1980).

Group 6 – Synaptic vesicle transporters

Members of group 6, classified as SLC22B, are atypical SLC22 family members (7, 174). They include the SV2 and SVOP transporters, which are transporter-like components of synaptic vesicles that share significant sequence similarity. Both SV2 and SVOP have been shown to have a high affinity for NAD⁺, suggesting that their action may be influenced by the rate of synaptic glycolysis, which consumes NAD⁺, or by synaptic redox potential (1981). SVOPL is a paralog of the synaptic vesicle protein SVOP (1968).

SV2A (SLC22B1), SV2B (SLC22B2), SV2C (SLC22B3) - Orphan transporters: The three SV2 (SLC22B) paralogs SV2A (SLC22B1), SV2B (SLC22B2) and SV2C (SLC22B3) form a distinct SLC22 subbranch of membrane proteins containing 12 TMHs (1982, 1983). SV2 proteins are mainly localized in the membrane of presynaptic vesicles and are thought to play a crucial role in synaptic function and neurotransmitter release. In particular, they have been implicated as regulators of several pathologies such as epilepsy (1982, 1984, 1985). SV2A has attracted particular interest as it has been shown to be the drug target for the antiepileptic drugs levetiracetam, brivaracetam and seletracetam (1986).

SV2A (SLC22B1) is a synaptic vesicle membrane protein expressed in neurons and endocrine cells where it is involved in neurotransmitter release. It has been found to be expressed in all glutamatergic and GABAergic cortical subtypes (1983). It has been shown to bind levetiracetam, and its selective analog brivaracetam (1987). When expressed in yeast cells deficient for hexose transport, human SV2A exhibited cellular galactose uptake (1987). Whether the sugar transport capacity of SV2A plays a direct role in modulating synaptic function remains to be determined.

SV2B (SLC22B2) expression is restricted to glutamatergic neurons only (1983). SV2B (synaptic vesicle glycoprotein 2B) has been implicated in the pathogenesis of Alzheimer disease by negatively regulating the amyloidogenic processing of AβPP at presynapses (1985, 1988).

SV2C (SLC22B3) has very limited expression in a small subgroup of GABAergic interneurons (1983). It has been shown to modulate dopamine release and its function is disrupted in Parkinson disease (1985, 1989).

SVOP and SVOPL (SLC22B4 and SLC22B5) - Orphan transporters: SVOP and SVOPL are more distantly related to SV2. They are both localized to vesicles (1981, 1990). Studies of Svop in knockout mice showed no essential role in fertility or viability of the mice and there were no obvious phenotypes (1991). According to the HPA, SVOP is most highly expressed in bone marrow, with more moderate expression in intestine, kidney, male and female tissues, and different brain regions. SVOPL is a paralog of SVOP (1968). Little is known at the gene and protein level. SVOP (SLC22B4) has been reported to be a nucleotide-binding protein (1981). The SVOPL gene has been shown to be maternally imprinted with the paternal allele inactivated by DNA methylation (1992). Expressed earlier in development, SVOPL expression levels in the brain have been reported to decrease with age (1991).

Orphan transporter family members (9)

SLC22A17 (BOIT), SLC22A23 (Nritp), SLC22A25 (UST6), SLC22A31, SV2A (SLC22B1), SV2B (SLC22B2), SV2C (SLC22B3), SVOP (SLC22B4), SVOPL (SLC22B5)

HGNC update

SLC22A18 (ORCTL2) has been removed from the SLC22 family and renamed as SLC67A1 following refined phylogenetic analysis.

SLC23 Na⁺-dependent vitamin C transporter family (2.A.40.6/Xan_ur_permease/NAT)

Discovery: The rat intestinal Na⁺-coupled vitamin C transporter SVCT1 (SLC23A1) is the founding member of this family (140). It was identified by expression cloning in Xenopus oocytes.

Gene family members (3 + 1 pseudogene):
SLC23A1 (SVCT1)	SLC23A2 (SVCT2)	SLC23A3 (SVCT3)	SLC23A4P (pseudogene)

Molecular aspects, physiological roles and links to disease

The family includes the Na⁺-coupled vitamin C transporters SVCT1 (SLC23A1), SVCT2 (SLC23A2), and orphan transporter SVCT3 (SLC23A3). The SLC23 family belongs to the Nucleobase/Ascorbate Transporter (NAT) family (TC 2.A.40.6) which is part of the Human NAT fold families, members of which share the 7-TMH inverted repeat architecture (see the description below for information on the cryo-EM structure of SVCT1 and Fig. 4).

As an essential vitamin that cannot be synthesized by the human body, vitamin C serves as a co-factor for a number of enzymes, including those involved in the synthesis of collagen and catecholamines, and promotes wound healing. There are two types of Na⁺-coupled cotransporters within the SLC23 family that supply our body with vitamin C (140, 1993):

SLC23A1: SVCT1 (SLC23A1) is expressed in various organs, such as the intestines, kidneys, liver, lungs, and skin (140, 1993). It helps absorb vitamin C across the intestinal brush border membrane (see Fig. 30) and facilitates its reabsorption in the proximal tubule of the kidney (see below). Thus, it plays an important role in maintaining vitamin C levels throughout the body.

Knockout of Slc23a1 (Svct1) in mice, which unlike humans can synthesize vitamin C in the liver, resulted in a significant decrease in plasma ascorbic acid levels and substantial urinary vitamin C loss due to the absence of reabsorption via Svct1 in the proximal tubules of the kidney (1994). Interestingly, intestinal absorption of vitamin C was minimally impacted by Slc23a1 deletion, and knockout mice fed a diet supplemented with ascorbic acid or 6-bromo-6-deoxy-L-ascorbate (a specific ascorbate transporter substrate) excreted both compounds. This suggests that the intestine is also capable of absorbing these compounds through a different, as yet unknown mechanism. Approximately 45% of pups from Slc23a1 knockout dams died during the perinatal period, However, this outcome could be prevented by ascorbic acid supplementation during pregnancy (1994).

The cryo-EM structure of mouse SVCT1 (Slc23a1) has revealed the structural basis of Na⁺-coupled vitamin C transport (230). Mouse SVCT1 functions as a homodimer with its gate domain forming the dimeric interface. Each subunit consists of 14 transmembrane segments (TM1–TM14) where TM1–TM7 and TM8–TM14 are inverted structural repeats related by a pseudo twofold axis parallel to the membrane. The structure shows that vitamin C binds to the core domain of each subunit. Two potential sodium ions are identified near the binding site, and the coordination of sodium ions by vitamin C explains their role in the ion coupling mechanism. SVCT1 employs a transport mechanism that combines elevator-like shift with local structural changes to translocate its substrate (230).

SLC23A2: SVCT2 (SLC23A2) is abundantly expressed in tissues requiring high levels of vitamin C, including the brain, lungs, liver, skin, spleen, muscles, adrenal glands, eyes, kidneys, and reproductive organs, according to the HPA and other sources (140, 1993). SVCTs display a unique preference for L-ascorbic acid (vitamin C) over dehydroascorbic acid (DHA), the oxidized form of vitamin C, and various analogs and intermediates of vitamin C metabolism (1993, 1995). By contrast, DHA is a substrate of GLUT1 (SLC2A1), GLUT3 (SLC2A3), GLUT4 (SLC2A4) and GLUT10 (SLC2A10) (457, 611, 1996) (see the description of the SLC2 families). DHA is toxic at high concentrations and therefore, when it enters cells via GLUT transporters, it is reduced to vitamin C as part of a recycling process. Neurons, for example, require high levels of vitamin C due to their high rate of oxidative metabolism compared to other cells, leading to the oxidation of vitamin C to DHA. The following steps are involved in the process of recycling DHA back into vitamin C in the CNS (see Fig. 8):

• 1)DHA exits neurons through the GLUT3 (SLC2A3) transporter
• 2)Astrocytes take up DHA through the GLUT1 (SLC2A1) transporter
• 3)Astrocytes convert DHA into vitamin C
• 4)Vitamin C leaves the astrocytes through a mechanism that is still unknown (possibly via a transporter similar to OAT2 (SLC22A7) (1997))
• 5)The released vitamin C is then transported back into neurons via SVCT2 (SLC23A2).

The concentration of vitamin C varies in the extracellular space between neurons and astrocytes, ranging from 200 to 400 μM, while within neurons the concentration is 10 mM, and within astrocytes it is 1 mM (1998). This recycling scheme with the GLUT and SVCT transporters helps to maintain the specific concentration levels.

To test the physiological impact of SVCT2-mediated vitamin C transport, Slc23a2 knockout mice were generated (637). Please note the subsequent nomenclature update, which clarifies that this study focuses on Slc23a2/SVCT2 and its knockout rather than Slc23a1, as indicated in the publication. Slc23a2 knockout mice died within a few minutes of birth with respiratory failure and intraparenchymal brain hemorrhage (637). Ascorbic acid levels in these Slc23a2 knockout newborn mice were undetectable in tissues, such as brain, pituitary, adrenals and pancreas and were markedly reduced in liver, kidney and muscle as well as in blood (637, 1999). Vitamin C was not detected in these fetal mice and prenatal supplementation of pregnant females did not elevate blood ascorbic acid in Slc23a2^-/- fetuses, suggesting that SVCT2 (Slc23a2) is important in placental ascorbic-acid transport. Brain hemorrhage was unlikely to be simply a form of scurvy since Slc23a2^-/- mice showed no hemorrhage in any other tissues. The levels of several markers of oxidative stress were increased in placenta and cortex in response to low ascorbic acid levels (1999). Furthermore, apoptosis in the brain was increased in Slc23a2 knockout mice (1999). SVCT2 is therefore required for appropriate ascorbic acid levels in fetal and placental tissues as well as for the development and functional maintenance of the CNS (637, 1999).

A brain-specific Slc23a2 knockout mouse model has subsequently been established to investigate the role of SVCT2 in the central nervous system (2000). Although the model is still under investigation, the current findings highlight the critical role of SVCT2 in maintaining vitamin C levels in the brain. These levels are essential for neurological health and development (2000).

Various genetic variants in SLC23A1 and SLC23A2 have been identified that may influence plasma levels of vitamin C and disease susceptibility (2001). Certain SLC23A2 variants have been linked to an altered risk of gastric cancer, underscoring the significance of these transporters in disease contexts.

SLC23A3: SVCT3 (SLC23A3) is highly expressed in the kidney and at lower levels in the intestine and liver (1993). The murine Slc23a3 gene, which encodes the orphan transporter SVCT3, was initially cloned from the mouse yolk sac. Subsequent studies have shown that it is expressed in the kidney (2002, 2003). When studying the regulation of Slc23a3 mRNA expression in response to vitamin C deficiency in mouse kidney, it was found that Slc23a1 mRNA levels were increased while Slc23a3 mRNA levels were unchanged (1993), raising the question of whether SVCT3 (SLC23A3) is indeed a vitamin C transporter. A recent study indicated that SVCT3 functions as a renal hypoxanthine transporter, because Na⁺-dependent 10 nM [³H]-hypoxanthine uptake could be demonstrated in SLC23A3 cRNA-injected Xenopus oocytes (2004).

SLC23A4P - Pseudogene: Slc23a4 was identified in rat as the first Na⁺-dependent nucleobase transporter (SNBT1) (2005). There are protein coding orthologs in many species but the equivalent gene in human, SLC23A4P, is a unitary pseudogene.

Orphan transporter family member (1)

SLC23A3 (SVCT3)

SLC24 Na⁺/(Ca²⁺-K⁺) exchanger family (2.A.19.4/Na_Ca_ex/NCX)

Discovery: NCKX1 (SLC24A1) was purified from bovine retinal rod outer segments in 1988 (2006). To clone the Na⁺/Ca²⁺-K⁺ exchanger, a bovine retinal expression library was screened using polyclonal antibodies raised against the purified protein (2007).

Gene family members (5):
SLC24A1 (NCKX1)	SLC24A3 (NCKX3)	SLC24A5 (NCKX5)
SLC24A2 (NCKX2)	SLC24A4 (NCKX4)

Molecular aspects, physiological roles and links to disease

Members of the SLC24 family are K⁺-dependent Na⁺/Ca²⁺ exchangers (NCKX) that use both the inward Na⁺ and outward K⁺ gradients to extrude Ca²⁺ from cells. There are five human SLC24 members, SLC24A1 (NCKX1), SLC24A2 (NCKX2), SLC24A3 (NCKX3), SLC24A4 (NCKX4) and SLC24A5 (NCKX5), that play a role in biological processes ranging from vision in retinal rod and cone photoreceptors to olfaction and skin pigmentation.

The SLC24 family has 5 members (Fig. 34) and belongs to the Ca²⁺:Cation Antiporter (CaCA) family (TC 2.A.19.4), which, together with the SLC8 family, is part of the NCX fold families.

Fig. 34. Phylogenetic tree of the SLC24 family.

The current structural model suggests that human NCKX proteins contain 11 TMHs, labelled 0 to 10, with TMH0 being a cleavable signal peptide (2008). Cleavage of the signal peptide results in an external N-terminal domain containing glycosylation sites. NCKX2 contains two clusters of five TMHs separated by a large hydrophilic loop located in the cytosol and a short C-terminal loop located in the extracellular space (2008). A characteristic feature of the NCKX sequences is a high degree of sequence similarity in the two central domains within the two clusters of five TMHs, referred to as the α1 and α2 repeats, each containing 45-50 residues. These α repeats are found in all NCX (SLC8 family) and NCKX (SLC24 family) sequences. The proposed model is consistent with the NCX crystal structures obtained for the archaebacterial Na⁺/Ca²⁺ exchanger NCX from Methanococcus jannaschii (16). NCKX1 has been shown to form homodimers and to complex with cGMP-gated channels in rod photoreceptors (2009, 2010).

The NCKX/SLC24 K⁺-dependent Na⁺/Ca²⁺ exchanger family has similar but also distinct properties to the NCX/SLC8 Na⁺/Ca²⁺ exchanger family, which shares ~20% amino acid sequence identity with NCKX members (2011, 2012): 1) both families are predominantly involved in Ca²⁺ extrusion, which requires the energy of the Na⁺ gradient generated by Na⁺ pumps; 2) both are also capable of bidirectional exchange, e.g., influx and efflux, depending on the membrane potential and the prevailing cation gradients, which under normal physiological conditions support Ca²⁺ extrusion, while under compromised metabolic integrity of cells, such as ischemic episodes, may cause a collapse of transmembrane cation gradients and a concomitant Ca²⁺ influx via reverse exchange, most likely leading to cell death in cells expressing NCKX (and/or NCX) proteins (2010). The NCKX/SLC24 and NCX/SLC8 families are distinguished by two key features: their coupling ratios. The coupling ratio is 4 Na⁺:1 Ca²⁺ + 1 K⁺ for NCKX exchangers and 3 Na⁺:1 Ca²⁺ for NCX exchangers (2013, 2014). From a thermodynamic point of view, the additional coupling to the cotransport of 4 Na⁺ ions and the countertransport of 1 K⁺ ion provides an additional safeguard in the export of calcium, having a stoichiometry that allows NCKX proteins to maintain a Ca²⁺ efflux function even under conditions of partially disturbed transmembrane electrochemical ion gradients. This stoichiometry is particularly important in cells where ion fluxes are high and ion concentrations fluctuate, such as in excitable cells like neurons or in retinal cone photoreceptors (see below), where spontaneous voltage fluctuations occur, or in cells where calcium efflux must be ensured, such as in enamel-forming ameloblasts, where NCKX members provide Ca²⁺ for extracellular mineralization (2015).

During phototransduction in vertebrate photoreceptors, light leads to hyperpolarization of rod photoreceptors by inducing closure of cation-specific channels in the plasma membrane, and in the dark, cation channels open and mediate an inward current due to the entry of sodium ions into the cell. Calcium ions also enter the cell through cGMP-gated channels and are rapidly extruded from the cell by the Na⁺/Ca²⁺ exchanger NCKX1. Upon illumination, the change in Ca²⁺ shaped by NCKX1 activity is a critical factor in the kinetics and termination of the light response and adaptation to changing light levels (2006, 2010, 2016).

Thus, the NCKX/SLC24 family members contribute to a variety of physiological responses in different cells and tissues. They all share a common role in shaping the spatial and temporal nature of cellular Ca²⁺ signals. Careful regulation of the NCKX exchangers is therefore required to ensure the fidelity of the downstream biological processes they control. For example, rod photoreceptors reveal regulatory processes whereby Na⁺-induced NCKX1-mediated Ca²⁺ extrusion ceases within seconds. The ways in which NCKX protein function can be modulated have been the subject of extensive review elsewhere (2010).

SLC24A1: NCKX1 (SLC24A1) is expressed almost exclusively in the outer segments of rod photoreceptors, where it is the only Ca²⁺ extrusion pathway (2017). A mutation in SLC24A1 is implicated in autosomal recessive congenital stationary night blindness (2018).

SLC24A2: NCKX2 (SLC24A2) is widely expressed in neurons throughout the brain, in addition to cone photoreceptors, and is thought to be a key component of neuronal Ca²⁺ efflux when intracellular Ca²⁺ levels are elevated above baseline levels (2019).

SLC24A3, SLC24A4: NCKX3 (SLC24A3) and NCKX4 (SLC24A4) have broad expression patterns, but are also abundant in the brain (2010). NCKX4 has been shown to affect tooth enamel formation (2015) and neuronal satiety signaling pathways (2020) in addition to its role in visual transduction in cone photoreceptors and odor detection in olfactory sensory neurons (2010, 2016, 2021, 2022). Mutations in SLC24A4 (NCKX4) have been associated with amylogenesis imperfecta leading to abnormal enamel formation, highlighting the importance of NCKX function in ameloblasts (2015). Further deepening of understanding of how SLC24 transporters impact physiology and pathophysiology may reveal potential new therapeutic applications, e.g., for NCKX2 in the context of brain ischemia (2010).

SLC24A5: NCKX5 (SLC24A5) plays an important role in skin pigmentation and melanosome maturation in humans, and a “light skin allele” of SLC24A5 has been shown to be responsible for lighter skin in Europeans but not in East Asians (2023). NCKX5 has been reported to localize to mitochondria from where it enables mitochondrial Ca²⁺ transfer into melanosomes (2024). A functional study of mutations in NCKX exchangers revealed associations of mutations in SLC24A5 (NCKX5) with non-syndromic oculocutaneous albinism due to melanin pigment deficiency (2025).

Orphan transporter family members: N/A

SLC25 Mitochondrial carrier family (2.A.29/Mito_carr/MCF)

Discovery: ADP/ATP carrier from beef heart mitochondria (SLC25A4) (118) is the first identified member of the SLC25 family. The protein has been purified from beef heart since 1974 and was easily accessible (2026). This progress allowed the determination of its primary structure, which is in fact the first primary structure of a solute carrier reported. Examination of the ADP/ATP carrier sequence revealed that the whole structure of around 300 amino acids can be divided into three related domains, each about 100 residues in length, which turned out to be a unique feature of related mitochondrial carriers.

Gene family members (53):
SLC25A1 (CIC)	SLC25A11 (OGC)	SLC25A21 (ODC)	SLC25A31 (ANT4)
SLC25A2 (ORC2/ORNT2)	SLC25A12 (AGC1)	SLC25A22 (GC1)	SLC25A32
SLC25A3 (PHC)	SLC25A13 (AGC2)	SLC25A23 (APC2)	SLC25A33 (PNC1)
SLC25A4 (ANT1)	SLC25A14 (UCP5)	SLC25A24 (APC1)	SLC25A34
SLC25A5 (ANT2)	SLC25A15 (ORC1/ORNT1)	SLC25A25 (APC3)	SLC25A35
SLC25A6 (ANT3)	SLC25A16 (GDC)	SLC25A26 (SAMC)	SLC25A36 (PNC2)
UCP1 (SLC25A7)	SLC25A17	SLC25A27 (UCP4)	SLC25A37 (MFRN1)
UCP2 (SLC25A8)	SLC25A18 (GC2)	SLC25A28 (MFRN2)	SLC25A38 (GLYC)
UCP3 (SLC25A9)	SLC25A19 (TPC)	SLC25A29 (BAC)	SLC25A39
SLC25A10 (DIC)	SLC25A20 (CAC)	SLC25A30 (UCP6)	SLC25A40
SLC25A41(APC4)	SLC25A45	MTCH1 (SLC25A49)	SLC25A53 (MCART6)
SLC25A42	SLC25A46	MTCH2 (SLC25A50)
SLC25A43	SLC25A47	SLC25A51 (MCART1)
SLC25A44	SLC25A48	SLC25A52 (MCART2)

Molecular aspects, physiological roles and links to disease

The SLC25 family is the largest protein family of solute transporters with 53 members (Fig. 35) and is a member of the mitochondrial carrier (MC) family (TC 2.A.29). SLC25 transporters have a threefold pseudo-symmetric structure (see Section 8). SLC25 transporters are involved in vital cellular processes, including oxidative phosphorylation of lipids and sugars, amino acid metabolism, macromolecular synthesis, ion homeostasis, cellular regulation, and differentiation, by transporting nucleotides, amino acids, carboxylic acids, fatty acids, inorganic ions, and vitamins across the inner mitochondrial membrane (285, 286, 288). Recent progress in the structural determination of mitochondrial carriers has provided important insights into the transport mechanism, which generally involves the coordinated movement of six structural elements that result in the alternating opening and closing of the matrix or cytoplasmic side of the carriers (286). Mitochondrial carriers have exceptionally dynamic properties with large conformational changes during the transport cycle, allowing them to transport large substrates without significant proton leak.

Fig. 35. Phylogenetic tree of the SLC25 family.

As shown in Fig. 36, members of the SLC25 family transport a wide variety of substrates across the inner mitochondrial membrane.

Fig. 36. Overview of the SLC25 mitochondrial carriers and their role in metabolism and energy production.

MtDNA, mitochondrial DNA.

SLC25 family members also include Ca²⁺-dependent mitochondrial carriers. While Ca²⁺ entry into mitochondria through the Ca²⁺ uniporter (MCU) complex of the in the inner mitochondrial membrane (see below) is well known to affect mitochondrial energy metabolism by activating mitochondrial dehydrogenases and F-ATP synthases (2027, 2028), there are also SLC25 mitochondrial carriers harboring Ca²⁺-binding domains facing the intermembrane space. These carriers provide an additional pathway for Ca²⁺ signaling to mitochondria. Examples are the aspartate/glutamate carriers aralar/AGC1 (SLC25A12) and citrin/AGC2 (SLC25A13) (290) and the ATP-Mg²⁺/P_i carriers APC1 (SLC25A24), APC2 (SLC25A23) and APC3 (SLC25A25) (2029). These calcium-dependent mitochondrial carriers contain a transporter regulatory domains that comprise N-terminal extensions of about 180 amino acid residues and contain several calcium-binding EF hands.

The MCU itself is not part of the SLC25 family. Although it is called a uniporter, it has not been included in the SLC nomenclature system. The MCU complex consists of the pore-forming subunit MCU (TC 1.A.77), the mitochondrial calcium uptake regulatory subunits MICU1, MICU2 and MICU3 (TC 8.A.44), and the structural protein EMRE (TC 8.A.45). The pore-forming subunit turned out to be a calcium channel rather than a transporter and was therefore not assigned an SLC name. Several structures of MCU from fungi have been determined, demonstrating a tetrameric architecture and a Ca²⁺ selectivity mechanism, and confirming that MCU is a channel rather than a transporter (2030, 2031). Interestingly, MCU has no detectable sequence homology to any other cation channel (2030, 2032–2034). The structure of the MICU1-MICU2 complex has revealed the regulation of the MCU complex (2035). MICU1 is a peripheral regulatory protein with two EF-hand motifs that sense cytosolic [Ca²⁺] and when it rises, it opens MCU, allowing mitochondrial Ca²⁺ uptake (2036) to enhance the energy metabolism, as indicated above.

The descriptions of the SLC25 family members are provided clockwise through the phylogenetic tree, divided into subgroups by substrate type or function, where known, as indicated in the tree diagram.

Tricarboxylate

SLC25A1: The citrate carrier CIC (SLC25A1), also known as the tricarboxylate carrier, is responsible for the electroneutral exchange of a tricarboxylate (e.g., citrate, isocitrate) for either another tricarboxylate, a dicarboxylate (e.g., malate), or phosphoenolpyruvate (PEP) across the inner mitochondrial membrane (285, 2037). CIC is essential for fatty acid and sterol biosynthesis by transporting citrate from the mitochondria to the cytosol (285). Thus, citrate plays a key role in mitochondrial metabolism and respiration and also has fundamental functions in the cytosol as a metabolic substrate, an allosteric regulator of strategic enzymes of the catabolic and anabolic pathways, and as a source of acetyl-CoA. CIC (SLC25A1) provides the citrate transport function for all these vital pathways and is the only known human mitochondrial citrate transporter to fulfil this function.

It is therefore not surprising that loss of CIC is pathogenic, and indeed mutations or deletions of the SLC25A1 gene have been associated with a complex and heterogeneous spectrum of developmental disorders (2038). Missense mutations of SLC25A1 cause an autosomal recessive neurometabolic disorder characterized by neonatal-onset encephalopathy with severe muscle weakness, intractable seizures, respiratory distress, and failure of psychomotor development, often resulting in early death (2039). Biallelic germline mutations in SLC25A1 have been shown to cause D/L-2-hydroxyglutaric aciduria, a fatal systemic disease characterized by accumulation of both enantiomers of 2-hydroxyglutaric acid as a result of mitochondrial respiratory deficit and metabolic remodeling (2040).

On the other hand, upregulation of CIC activity has been reported in Behçet syndrome (2041) and Down syndrome (2042). Increased expression of CIC (SLC25A1) has also been shown to cause an autistic-like phenotype with altered neuronal morphology (2043). Moreover, upregulation of SLC25A1 expression is a hallmark of several cancers and metabolic disorders (2044). In order to combat these pathologies, CTPI-2 and other CIC inhibitors are being considered as a novel class of promising therapeutic agents (2038). In studies in mice, inhibition of the mitochondrial citrate carrier SLC25A1 with the specific inhibitor CTPI-2 was shown to reverse steatosis, glucose intolerance and inflammation in preclinical models of non-alcoholic fatty liver disease (NAFLD) and its progression to inflammatory responses in NASH (2045). The study showed that CTPI-2 treatment enabled mice to tolerate a high-fat diet without developing liver damage, obesity or impaired glucose homeostasis (2045). This suggests that CTPI-2 may act as a glucose-restriction mimetic, potentially extending its use to pathological conditions associated with impaired glucose homeostasis. The biological activity of CTPI-2 was shown to be potent and broad, providing a rationale for further exploitation of pharmacological inhibition of SLC25A1.

SLC25A21: The oxodicarboxylate carrier ODC (SLC25A21) imports 2-oxoadipate and exports 2-oxoglutarate as part of the catabolism of lysine, hydroxylysine, and tryptophan (2037, 2046). ODC (SLC25A21) transports 2-oxoadipate and 2-aminoadipate, which are generated from the breakdown of tryptophan and lysine, respectively, from the cytosol into the mitochondrial matrix where they are oxidized and fed into the TCA cycle. A homozygous pathogenic variant of SLC25A21 in a patient has been shown to lead to mitochondrial dysfunction with urinary excretion of pipecolic and quinolinic acids, which are by-products of the lysine and tryptophan catabolic pathways (2047, 2048). Accumulation of these compounds in turn leads to reduction of mitochondrial DNA, causing toxicity in spinal motor neurons and resulting in spinal muscular atrophy-like disease (2047).

Glutamate and aspartate

SLC25A12, SLC25A13: AGC1/aralar (SLC25A12) and AGC2/citrin (SLC25A13) are Ca²⁺-stimulated mitochondrial aspartate/glutamate transporters with EF-hand Ca²⁺-binding motifs in their N-terminal domains (2049). They export aspartate from mitochondria in exchange for cytosolic glutamate and a proton. AGCs are involved in antioxidant defense and Ca²⁺-dependent control of mitochondrial respiration via the malate-aspartate-NADH shuttle. Because of their association with glycolysis and the NAD⁺/NADH ratio, a role in carcinogenesis has been suggested, and indeed, targeting citrin with inhibitors has been found to be a promising approach in cancer therapy (2050). AGC1/aralar (SLC25A12) is a significant factor in neuronal physiology and is believed to contribute to glutamate-induced excitotoxicity. Accordingly, mutations in AGC1/aralar (SLC25A12) cause “early infantile epileptic encephalopathy”, a rare human disease (2051). Mutations in SLC25A13 (AGC2/citrin) lead to type II citrullinemia, a condition mainly observed in the Japanese population. It inhibits the urea cycle, causing the buildup of ammonia and other harmful substances and disturbs the production of proteins and nucleotides. The condition can cause neonatal intrahepatic cholestasis and affects the central nervous system in adulthood (2052, 2053).

SLC25A18, SLC25A22: GC2 (SLC25A18) and GC1 (SLC25A22) import glutamate together with a H⁺ but do not export aspartate (2054). Glutamate is then converted by glutamate dehydrogenase into α-ketoglutarate and ammonia, with reduction of NAD(P)⁺ to NAD(P)H, which then enters the respiratory chain via complex I. GC1 (SLC25A22) is a low affinity, high capacity glutamate transporter and GC2 (SLC25A18) is a high affinity, low capacity glutamate transporter (2054). These glutamate carriers do not have the EF-hand Ca²⁺ binding motifs found in aspartate/glutamate carriers. They serve as the main gate for glutamate to entry mitochondria and are ubiquitously expressed, including in astrocytes. Mutations in SLC25A22 have been shown to cause neonatal epileptic encephalopathy (2055). The reason for this condition is that lack of GC1 (SLC25A22) function leads to intracellular accumulation of glutamate in astrocytes, whose critical role is to prevent extracellular glutamate from accumulating to excitotoxic levels (2056).

Glycine

SLC25A38: GLYC (SLC25A38) facilitates the import of glycine into the mitochondria, where it reacts enzymatically with succinyl-CoA to form δ-aminolevulinic acid (ALA) as part of the heme biosynthesis pathway (2057). ALA is then transported by an unknown transporter to the cytosol where it is used as a precursor for porphyrin synthesis (2057). The resulting coproporphyrinogen III is then transported into mitochondria and the final step is the insertion of ferrous iron (Fe²⁺), transported by mitoferrin MFRN1 (SLC25A37) or MFRN2 (SLC25A28), into the protoporphyrin IX (PPIX) ring, a reaction which is catalyzed by ferrochelatase. The inability to form ALA due to a defect in the enzyme ALA-synthase that catalyzes the reaction leads to heme deficiency and sideroblastic anemia caused by the inability of erythroid cells to synthesize heme. Consistent with this, a defect in the glycine transporter GLYC (SLC25A38) also results in sideroblastic anemia (2058, 2059).

S-adenosyl homocysteine

SLC25A26: The mitochondrial S-adenosylmethionine carrier SAMC (SLC25A26) imports S-adenosylmethionine (SAM) synthesized in the cytosol into the mitochondria in exchange for mitochondrial S-adenosylhomocysteine. SAM is required for methylation reactions of DNA, RNA and protein in the mitochondrial matrix with generation of S-adenosylhomocysteine, which is returned to the cytosol via the transporter. A missense mutation in SLC25A26 causes intramitochondrial methylation deficiency consistent with its function, leading to oxidative phosphorylation deficiency, i.e., respiratory insufficiency, lactic acidosis, acute episodes of cardiopulmonary failure, and progressive muscle weakness (2060).

Iron

SLC25A28, SLC25A37: The mitoferrins MFRN2 (SLC25A28) and MFRN1 (SLC25A37) transport iron into mitochondria (2061). There, iron is essential for heme biosynthesis, hemoglobin production, and Fe-S cluster protein assembly during erythroid development. MFRN1 (SLC25A37) is highly expressed in differentiating erythroid cells, whereas MFRN2 (SLC25A28) is ubiquitously expressed in non-erythroid tissues. Iron is transported into mitochondria by mitoferrins, presumably as Fe²⁺, as shown for the yeast homologs of mitoferrins, Mrs3p and Mrs4p (2062). It is then delivered to ferrochelatase to catalyze the incorporation of Fe²⁺ into protoprophyrin IX for heme production in erythroid cells, or made available via frataxin for the synthesis of Fe-S clusters (in any cell type), which are essential for the proper function of complexes I, II, and III of the electron transport chain, the citric acid cycle enzyme aconitase, and others. Frataxin acts either as an iron chaperone or as an iron storage protein. Friedreich ataxia (FRDA) is an autosomal recessive neurodegenerative and cardiac disorder caused by defective frataxin expression (2063) for which novel discoveries are opening therapeutic treatment strategies (2064, 2065).

Pyrimidine

SLC25A33, SLC25A36: The PNC1 (SLC25A33) and PNC2 (SLC25A36) pyrimidine nucleotide transporters play a vital role in mitochondrial DNA and RNA synthesis and degradation (292, 2066). PNC1 (SLC25A33) utilizes an antiport mechanism to transport uracil, thymine, and cytosine (deoxy)nucleoside di- and tri-phosphates, whereas PNC2 (SLC25A36) employs a uniport or antiport mechanism to transport cytosine and uracil (deoxy)nucleoside mono-, di-, and tri-phosphates. Both carriers transport guanine nucleotides, but they do not transport adenine (deoxy) nucleotides. A separate mitochondrial GTP/GDP carrier has been identified in yeast (2067), but an equivalent transporter has not been found in humans. The two human proteins with the highest amino acid sequence similarity to yeast GTP/GDP carriers are MCART1 (SLC25A51) and MCART2 (SLC25A52), however, whether they indeed transport GTP or GDP is yet unknown (292, 2068).

Nucleotide and flavine derivatives, coenzyme A

SLC25A17: SLC25A17 is a peroxisomal transporter of coenzyme A, FAD and NAD⁺ (2069, 2070). Its main function is thought to be the transport of free CoA, FAD and NAD⁺ into peroxisomes in exchange for intraperoxisomally generated PAP (adenosine 3ʹ,5ʹ-diphosphate), FMN and AMP. SLC25A17 mRNA was found to be ubiquitously expressed, with particularly prominent expression in the testis (2069).

SLC25A32: SLC25A32 transports FAD/NAD-like substrates including riboflavin (2071, 2072). Consistent with this, SLC25A32 expression is increased in a variety of human tumors to sustain cancer cell proliferation by promoting flavin adenine nucleotide (FAD) metabolism (2073). Folate transport by SLC25A32 has also been proposed because in a study with Chinese hamster ovary cells, a mutation inactivating the mitochondrial inner membrane folate transporter created a glycine requirement for the survival of these cells (2074). However, the folate transporter function remains to be experimentally validated in an independent manner. In support of flavine transport, the substrate binding site has the typical features of an adenine binding pocket, consistent with the FAD/riboflavin transport assignment. A pathological mutation of this transporter causes exercise intolerance, and riboflavin supplementation has been shown to benefit patients, suggesting that the most likely substrates of this carrier are flavin-related compounds rather than folate (2075).

Thiamine pyrophosphate

SLC25A19: TPC (SLC25A19) is a thiamine pyrophosphate transporter, an important cofactor of three mitochondrial enzymes: pyruvate dehydrogenase, branched chain ketoacid dehydrogenase and 2-oxoglutarate dehydrogenase (2076). SLC25A19 mutations cause Amish lethal microcephaly (MCPHA), which retards brain development and leads to α-ketoglutaric aciduria (2077). Consistent with this, knockout of Slc25a19 in mice causes mitochondrial thiamine pyrophosphate depletion, embryonic lethality, CNS malformations, and anemia (2076).

ADP/ATP

SLC25A4, SLC25A5, SLC25A6 and SLC25A31: As highlighted in the Discovery paragraph, the primary structure of the SLC25A4 ADP/ATP carrier was determined by classical amino acid sequencing of purified transporters from bovine heart mitochondria and represented the first reported sequence of a member of the SLC25 family (118). SLC25A4 functions as an adenine nucleotide translocase (ANT). Human has four closely related ANTs, ANT1 (SLC25A4), ANT2 (SLC25A5), ANT3 (SLC25A6) and ANT4 (SLC25A31). The ANTs exchange mitochondrial ATP for cytosolic ADP, thereby providing the mitochondria with new ADP for conversion to ATP by ATP synthase, and they export the newly synthesized ATP to the cytosol to fuel energy-demanding metabolic processes (285, 2078–2080). The different paralogs function in a tissue-specific manner (292). For example, ANT4 (SLC25A31), also known as SFEC, is a testis-specific adenine nucleotide transporter that mediates distal flagellar energy-related processes important for protein phosphorylation and motility (2081).

Mutations in SLC25A4 cause well-characterized mitochondrial diseases (2082). In particular, heterozygous SLC25A4 mutations cause adult-onset autosomal dominant progressive external ophthalmoplegia with multiple mitochondrial DNA deletions, whereas recessive mutations cause childhood-onset mitochondrial myopathy and cardiomyopathy.

Orphan transporter subgroup

SLC25A43 - Orphan transporter: Genome-wide screening in the haploid system revealed mouse Slc25a43 as a target gene of oxidative toxicity (2083). Specifically, Slc25a43 knockout of mouse haploid embryonic stem cells resulted in reduced reactive oxygen species damage and increased cell viability when exposed to H₂O₂. However, the transport function of SLC24A43 remains unknown.

SLC25A16 - Orphan transporter GDC (SLC15A16) encodes a protein belonging to the mitochondrial metabolite carrier family, the so-called Grave disease carrier protein (GDC). SLC15A16 was originally cloned from a thyroid library generated from the serum of a patient with Grave disease (2084) and it was proposed to be associated with Grave disease, although this has not yet been verified (292). In addition, GDC (SLC25A16) has been proposed to transport coenzyme A (CoA) (2085). The closely related SLC25A42 has directly been shown to transport CoA, but it is unclear whether SLC25A16 also contributes to this function, as its transport function has not been demonstrated. Only a single homozygous mutation in the SLC25A16 gene has been reported, causing a nail disorder of the hand with varying degrees of onychodystrophy (2086), which does not appear to be consistent with the function of this transporter as a CoA transporter.

Coenzyme A

SLC25A42: SLC25A42 is a ubiquitously expressed coenzyme A (CoA) transporter (292, 2085). Many reactions in the mitochondrial matrix, such as dehydrogenase activities, require CoA as co-factor which is synthesized outside of mitochondria and transported into the mitochondrial matrix via these transporters. The function and kinetic parameters of SLC25A42 were determined in transport assays with substrate specificities restricted to CoA and PAP (2087). A disease variant of SLC25A42 was shown to cause mitochondrial myopathy with muscle weakness, lactic acidosis and encephalomyopathy (2088) (2089).

ATP/phosphate

SLC25A23, SLC25A24, SLC25A25 and SLC25A41: APC2 (SLC25A23), APC1 (SLC25A24), APC3 (SLC25A25) and APC4 (SLC25A41) are ATP-Mg/P_i carriers that facilitate net import and export of adenine nucleotides by exchanging phosphate for adenine nucleotides, coupled to magnesium or protons, in an electroneutral way. This allows mitochondria to respond to changes in energetic demand and to replenish adenine nucleotide pools after mitochondrial division and macromolecular synthesis (286, 292, 2090, 2091). The three human paralogs APC2 (SLC25A23), APC1 (SLC25A24) and APC3 (SLC25A25) are calcium-dependent, while a fourth paralog, APC4 (SLC25A41), is not (2092, 2093).

Branched-chain amino acids

SLC25A44: SLC25A44 is a branched-chain amino acid (BCAA) transporter (2094). The BCAAs valine, leucine, and isoleucine can be broken down to provide metabolic energy and are required for protein synthesis in the mitochondria. The carrier was discovered in brown adipose tissue upon exposure to cold. Cold stimuli, in addition to glucose and fatty acids, can increase mitochondrial BCAA uptake and oxidation in brown adipose tissue as a fuel for thermogenesis, a process that requires SLC25A44. In turn, defective BCAA catabolism in brown adipose tissue results in impaired BCAA clearance and thermogenesis, leading to the development of diet-induced obesity and glucose intolerance (2094).

Glutathione

SLC25A39, SLC25A40: SLC25A39 and SLC25A40 are mitochondrial carriers required for glutathione (GSH) import (2095). GSH plays a key role in oxidative metabolism, and mitochondria, as the main site of oxidative reactions, must maintain sufficient levels of GSH. Cells deficient in both SLC25A39 and SLC25A40 are characterized by loss of activity and stability of iron-sulfur cluster proteins (2095).

NAD⁺

SLC25A51, SLC25A52: MCART1 (SLC25A51), together with its very close paralog MCART2 (SLC25A52) (96% identical), are mitochondrial NAD⁺ uptake transporters (2096–2099). They are required for NAD⁺ uptake into mitochondria. NAD⁺ is the most widely used cofactor of enzymatic redox reactions in cells, and the discovery helps to understand how NAD⁺ pools are distributed between the cytoplasm and mitochondria (2099).

SLC25A53 - Orphan transporter: Although relatively little is known about MCART3 (SLC25A53), TREM2 (Triggering Receptor Expressed on Myeloid Cells 2) has been shown to decrease expression of MCART3 (SLC25A53) in the inner mitochondrial membrane (2100). TREM2 is expressed by several cell types, including macrophages, immature monocyte-derived dendritic cells, osteoclasts and microglia, where it is essential for metabolic fitness in response to injury and stress. The study showed that macrophages expressing TREM2 exhibited decreased SLC25A53 transcription via the SYK-SMAD4 pathway after efferocytosis, which impairs NAD⁺ transport into mitochondria (2100). This caused a breakpoint in the TCA cycle and subsequently increased itaconate production because pyruvate could still be taken up by mitochondria. The oxidation of pyruvate leads to the formation of citrate and cis-aconitate, which is then converted to itaconate by aconitate decarboxylase (2101, 2102). Itaconate has been shown to promote tissue repair (2103), for example after myocardial infarction (2100). TREM2 is also essential for sustaining the microglial response to stress events, for example in counteracting neurodegeneration in Alzheimer disease (2104).

In microglia, TREM2 is a key transcriptional regulator of cholesterol transport and metabolism via the PI3K-mTOR signaling pathway (see the description of NPC1/SLC65A1). TREM2 localizes in lipid rafts, the altered lipid components and cholesterol concentrations of which are associated with altered TREM2 levels in the plasma membrane of reactive microglia (2105). The latter depends on NPC1-mediated cholesterol exit from lysosomes, as NPC1 deficiency blocks trafficking of cholesterol from lysosomes to the plasma membrane. This disrupts TREM2-dependent metabolic regulation in microglia (2106). Therefore, while inducing TREM2 expression ameliorates pathological phenotypes in Alzheimer disease models by reprogramming the responsiveness of microglia (2107), loss of TREM2 suppresses mTOR activation, increases autophagy and enables the full expression of SLC25A53.

Taken together, these findings support the concept that SLC25A53 functions as an NAD⁺ transporter. This is especially important because mitochondrial NAD⁺ uptake is crucial for energy metabolism, cellular signaling, and autophagy, which controls the cellular level of NAD⁺ to maintain mitochondrial function (2108).

Phosphate

SLC25A3: The phosphate carrier PiC (SLC25A3) is the only known gene product for this function in humans. PiC was first cloned from bovine heart (2109) and then from rat liver (2110) and human heart (2111). PiC (SLC25A3) has two splice variants, PiC-A and PiC-B. PiC-A is present only in heart and muscle, whereas PiC-B is ubiquitous (2112). Both isoforms mediate the H⁺-coupled P_i transport (or P_i/OH^- countertransport) into mitochondria, which is essential for the oxidative phosphorylation of ADP to ATP. PiC-A has a low affinity and PiC-B a high affinity for phosphate. Thus, PiC provides phosphate as a key substrate for ATP production and oxidative phosphorylation. PiC transport activity is also important for effective mitochondrial Ca²⁺ handling. Ca²⁺ uptake occurs via MCU (see above under “Molecular aspects, physiological roles and links to disease”) (2113) and the rapid and intense matrix Ca²⁺ load provided by this pathway is compensated by rapid P_i uptake, primarily via PiC.

Mitochondrial deficiency of PiC (SLC25A3) has a lethal outcome (2114). A homozygous mutation was reported in two patients with hypertrophic cardiomyopathy and muscular hypotonia, highlighting the importance of this carrier for oxidative phosphorylation (2115). Studies of biopsies from these patients revealed a non-canonical mitochondrial network with low mitochondrial proliferation. The absence of this protein may reduce the mitochondrial membrane potential and impair the import of nuclear-encoded mitochondrial precursor proteins (2116).

Outer mitochondrial membrane (SLC25A46, MTCH1 (SLC25A49) and MTCH2 (SLC25A50))

These SLC25 members form a separate group on the phylogenetic tree. It is hypothesized that they lack the conserved sequences to form a pore for small molecule transport and instead function as outer mitochondrial membrane proteins in mitochondrial fission or fusion (SLC25A46) or as gatekeepers of mitochondrial outer membrane biogenesis (MTCH1 (SLC25A49) and MTCH2 (SLC25A50)) (Fig. 36).

SLC25A46 - Orphan transporter: SLC25A46 has been reported to be a mitochondrial outer membrane protein involved in mitochondrial fission and fusion (2094, 2117–2119). Membrane fusion plays an important role in controlling the shape, number and distribution of mitochondria. Mutations in SLC25A46 result in changes in mitochondrial morphology. SLC25A46 mutations also underlie a broad spectrum of neurodegenerative diseases. Studies of the pathogenicity of three genetic variants of SLC25A46 in a human fibroblast cell line revealed that mitochondria in SLC25A46 knockout cells are fragmented, whereas mitochondria in cells expressing the pathogenic variants are hyperfused. Loss of SLC25A46 function resulted in altered mitochondrial lipid composition. It was hypothesized that SLC25A46 may facilitate interorganellar lipid flux or play a role in membrane remodeling associated with mitochondrial fusion and fission.

Variants of SLC25A46 have been found to be the cause of optic atrophy in association with peripheral neuropathy and congenital pontocerebellar hypoplasia (2120, 2121).

MTCH1 (SLC25A49) and MTCH2 (SLC25A50) - Orphan transporters: MTCH1 (SLC25A49) is a mitochondrial carrier that has been identified as a mitochondrial anti-ferroptosis factor in cervical cancers, allowing cervical cancer growth (2122). The combination of MTCH1 deficiency with the clinical antitumor drug Sorafenib effectively and synergistically induced ferroptosis and suppressed cervical cancer growth in a nude mouse xenograft model. In the absence of MTCH1 (SLC25A49), mitochondrial NAD⁺-concentration was decreased, ATP production was reduced, reactive oxygen species (ROS) levels were increased, thereby triggering ferroptosis. When NAD⁺ was added to MTCH1 (SLC25A49)-deficient cells, a significant reduction in ROS was observed. Thus, ROS accumulate when NAD⁺ is depleted, and rescue of MTCH1 expression restores mitochondrial NAD⁺ and ATP content and reduces ROS levels. Therefore, MTCH1 deficiency resulted in increased ROS production, presumably due to mitochondrial NAD⁺ depletion. However, how exactly MTCH1 regulates ferroptosis and NAD⁺ levels and whether it transports NAD⁺ itself requires further investigation.

MTCH2 (SLC25A50) is a mitochondrial outer membrane protein that regulates apoptosis. MTCH2 has also been reported to be a mitochondrial outer membrane protein that stimulates mitochondrial fusion (2123). Previously, MTCH2 has been reported to regulate apoptosis (2124).

New insights into the functional roles of MTCH1 and MTCH2 in the mitochondrial outer membrane were provided by a genome-wide CRISPR screen that identified MTCH2 and its paralog MTCH1 (293). In this study, MTCH2 was shown to be required for the insertion of biophysically distinct tail-anchored, signal-anchored, and multipass proteins, but not for outer membrane β-barrel proteins (293). Depletion of the close paralog MTCH1, which also localizes to the outer mitochondrial membrane, had an additive effect to the loss of MTCH2 on the biogenesis of mitochondrial tail-anchored proteins with a single C-terminal TMH. Based on these findings, it has been postulated that MTCH1 and MTCH2 define a unique class of membrane protein insertases that utilize the SLC25 transporter fold (293). As highlighted (293), the identification of MTCH2 as an insertase provides a mechanistic explanation for phenotypes previously linked to its dysfunction such as dysregulated mitophagy and mitochondrial fragmentation (2123), as well as for disease associations including Alzheimer disease (2125). MTCH proteins have therefore been proposed to act as gatekeepers of mitochondrial outer membrane biogenesis, with their disruption accounting for both their pleiotropic phenotypes and their links to human disease (Fig. 36).

Interestingly, insertases such as MTCH1 and MTCH2 also appear to possess lipid scramblase activity, using the same structural groove involved in insertase activity to mediate a “credit card”-like shuffling of lipid molecules between membrane leaflets (294). Notably, lipid scrambling and protein insertion can occur independently, indicating that scramblase activity represents an additional, distinct function of these proteins. Loss of scramblase activity could explain the mitochondrial morphological defects observed in MTCH2 deficiency (2123).

Moreover, proteomics data show that MTCH2 assembles with mitochondrial bridge-like lipid transfer proteins VPS13A and VPS13D, indicating its potentially broader role in organellar lipid transport and homeostasis, a function that remains to be fully characterized (2126, 2127).

Dicarboxylates

SLC25A10: DIC (SLC25A10) is a dicarboxylate carrier involved in the electroneutral exchange of dicarboxylates (e.g., malonate, malate, succinate) and inorganic phosphate and inorganic sulfur-containing compounds (e.g., sulfite, sulfate, and thiosulfate) (2128, 2129). It is involved in gluconeogenesis and ureogenesis, metabolism of sulfur compounds, and de novo fatty acid synthesis (2037). Inhibiting DIC (SLC25A10) has been reported to cause decreased mitochondrial glutathione levels and impaired complex I activity in rat neurons (2130). A mutation that abolishes DIC (SLC25A10) function was reported to cause a progressive form of epileptic encephalopathy and severe hypotonia associated with complex I deficiency in an affected patient, and it was proposed that loss-of-function causes pathological disturbances in respiration demanding conditions and susceptibility to oxidative stress (2131).

SLC25A11: OGC (SLC25A11) transports 2-oxoglutarate from the mitochondrial matrix across the inner mitochondrial membrane in an electroneutral exchange for malate or other dicarboxylates (2132, 2133). It plays an important role in the malate-aspartate shuttle, the oxoglutarate-isocitrate shuttle, and gluconeogenesis (2037).

OGC (SLC25A11) has also been reported to transport glutathione into mitochondria to limit ROS production and NADH for ATP production, factors that play an important role in liver cancer progression (2134). SLC25A11 has been shown to be a prognostic marker in liver cancer (2134).

Orphan transporters

SLC25A34 - Orphan transporter: SLC25A34 is involved in shaping lipid and glucose homeostasis (2135). NAFLD is a condition associated with insulin resistance. Fat builds up in the liver due to defective lipid metabolism and there is mitochondrial dysfunction. SLC25A34 is a major repressive target of miR-122, a microRNA that is conserved among vertebrate species that has a central role in NAFLD and liver cancer. While the transport function of SLC25A34 is unknown, in vitro experiments with hepatocytes depleted or overexpressing Slc25A34 and in vivo experiments with SLC25A34 knockout mice have shown that SLC25A34 plays a role in mitochondrial respiration and bioenergetics during NAFLD (2135). However, as its transport function remains elusive, further studies are needed to determine exactly how SLC25A34 affects the pathogenesis of NAFLD.

SLC25A35 - Orphan transporter: SLC25A35 was found to be significantly associated with the prognosis of pancreatic ductal adenocarcinoma (2136).

Protons/uncoupling

UCP1 (SLC25A7): The uncoupling protein UCP1 (SLC25A7), whose primary structure has been determined by amino acid sequencing (2137), is predominantly expressed in brown adipose tissue (2138, 2139). UCP1 dissipates the proton motive force, thereby short-circuiting the mitochondrion, resulting in heat production (Fig. 36). UCP1 is activated by fatty acids and inhibited by purine nucleotides in brown adipose tissue (2140). UCP1 is a monomer that binds three cardiolipins and a single purine nucleotide. Among other closely related proteins (see below), UCP1 is thought to be the only one involved in thermogenesis.

The structural mechanisms of UCP1 (SLC25A7) in thermogenesis regulation have been evaluated based on new insights provided by cryo-EM structures of this transporter (2141–2144). The structures of human UCP1 in the nucleotide-bound and free states were compared to address the molecular mechanism of UCP1 inhibition in non-shivering thermogenesis. The studies showed how purine nucleotides inhibit UCP1 in a pH-dependent manner and identified several elements within the UCP1 structure involved in the activation mechanism (2144). However, questions remain, such as how UCP1 is activated in vivo by free fatty acids and how fatty acids initiate proton leak across the inner mitochondrial membrane via UCP1, information that would help clarify the molecular basis of non-shivering thermogenesis in humans.

UCP2 (SLC25A8), UCP3 (SLC25A9), SLC25A14, SLC25A27, SLC25A30 - Orphan transporters: There are several closely related homologs to the uncoupling protein UCP1 (SLC25A7), such as UCP2 (SLC25A8), UCP3 (SLC25A9), UCP4 (SLC25A27), UCP5 (SLC25A14), and UCP6 (SLC25A30) (292), but their biological function is not clear and they are probably not true “uncoupling proteins”. UCP2 (SLC25A8) and UCP3 (SLC25A9) may prevent oxidative stress (2145), but the mechanism by which they do so remains elusive. UCP5 (SLC25A14) and UCP6 (SLC25A30) have been proposed to catalyze the export of sulfite and thiosulfate (the degradation products of H₂S) from the mitochondria as their main physiological roles (2146), thereby modulating the level of the important signaling molecule and cytoprotectant H₂S (2147). The mitochondrial uncoupling process has been reported to be a critical player in neurodegenerative diseases, and it has been reported that a non-coding variant (rs9472817) in the last intron of the SLC25A27 (UCP4) gene influences the risk of developing sporadic frontotemporal dementia (2148).

Carnitine/acylcarnitine

SLC25A20: CAC (SLC25A20) is a carnitine-acylcarnitine carrier, a key component of the carnitine cycle, importing acyl-carnitine into the mitochondria for fatty acid β-oxidation and exporting carnitine (2149, 2150). SLC25A20 mutations cause carnitine/acylcarnitine carrier deficiency, an autosomal recessive disorder with two clinical manifestations: 1) a severe neonatal onset with cardiomyopathy; and 2) a milder phenotype with hypoglycemia but no cardiomyopathy (2150). The inability to transport fatty acid chains into the mitochondria makes patients dependent on carbohydrates and amino acids for energy metabolism.

Ornithine/citrulline/lysine/arginine

SLC25A2, SLC25A15, SLC25A29: The ornithine carriers ORC2/ORNT2 (SLC25A2) and ORC1/ORNT1 (SLC25A15) exchange cytosolic ornithine and intramitochondrial citrulline, an important step in the urea cycle, which requires activities in both cytosol and mitochondria (2151) (for a review see (2037, 2114)). The first mitochondrial ornithine carrier was identified and cloned in S. cerevisiae (2152) and ORC1/ORNT1 (SLC25A15) was the first human mitochondrial ornithine transporter identified and cloned (2153), followed by ORC2/ORNT2 (SLC25A2) (2154, 2155). ORC1/ORNT1 (SLC25A15) is expressed in most tissues, with the highest levels in liver, pancreas, lung and kidney, while the expression of its closely related paralog ORC2/ORNT2 (SLC25A2) is more restricted to liver, testis, spleen, lung and pancreas (2155). Both transporters have been shown to facilitate the import of ornithine into the mitochondrial matrix in exchange for the export of citrulline and H⁺ (2156).

Impaired transport of ornithine into mitochondria due to genetic defects of ornithine carriers disrupts the urea cycle, which is required for the deamination of amino acids, leading to ornithine accumulation in the cytoplasm and causing impaired ureagenesis and hyperammonemia (Fig. 36). Thus, a defect in ornithine transport results in accumulation of ammonium and ornithine. Consistent with this, dysfunction of ORC1/ORNT1 (SLC25A15) is the cause of hyperornithinemia-hyperammonemia-homocitrullinuria (HHH) syndrome, a rare autosomal recessive disorder leading to growth retardation, seizures, and spasticity, associated with persistent hyperornithinemia and episodic hyperammonemia, which often present at several months or even several years of age (2157, 2158). Homocitrulline is thought to originate from carbamylation of lysine. It remains to be determined whether there are mutations in ORC2/ORNT2 (SLC25A2) that cause HHH syndrome.

The somewhat more distantly related basic amino acid carrier BAC (SLC25A29) mainly transports arginine, lysine, homoarginine, methylarginine (2159). Accordingly, its main physiological role is to import basic amino acids into mitochondria for mitochondrial protein synthesis and amino acid catabolism. In lung adenocarcinoma, one of the deadliest cancers, SLC25A29, along with SLC2A1 encoding the GLUT1 glucose transporter, and SLC27A4 encoding a fatty acid transport protein, were identified as key genes associated with survival (2160). The study provided new insights into the pathogenesis of lung adenocarcinoma and the development of potential therapeutic strategies.

SLC25A45 - Orphan transporter: In a study of rare mutations associated with serum creatinine, a missense mutation in SLC25A45 was identified and it was hypothesized that SLC25A45 may play a role in the biosynthesis of arginine, which is involved in the synthesis of creatine, the precursor of creatinine (2161).

SLC25A47: SLC25A47 is a liver-specific mitochondrial inner membrane carrier of NAD⁺ required for hepatic gluconeogenesis and energy homeostasis. GWAS revealed significant associations between SLC25A47 and fasting glucose, HbA1c, and cholesterol levels in humans (2162). SLC25A47 depletion resulted in decreased hepatic pyruvate flux and mitochondrial malate accumulation, thereby limiting hepatic gluconeogenesis. The study identifies a critical node in hepatic mitochondria that regulates fasting-induced gluconeogenesis and energy homeostasis (2162).

A subsequent study identified NAD⁺ as an endogenous substrate for SLC25A47 (2163). This discovery was made using a combination of methods, including homology-based modeling of SLC25A47 and virtual screening of the human metabolome database, followed by uptake experiments. The study further showed that SLC25A47 is involved in the pharmacological action of metformin through activation of the cellular energy sensor AMP-activated protein kinase α (AMPKα) in hepatocytes. Specifically, metformin was shown to increase the expression level of Slc25a47 in mice to deliver NAD⁺ into mitochondria. SIRT3, a deacetylase belonging to the sirtuin family, requires NAD⁺ as a cofactor. SIRT3 activation then activates AMPKα in hepatocytes. AMPKα activity was greatly reduced in Slc25a47 knockout mice. Based on these findings, it was concluded that SLC25A47 is a hepatocyte-specific mitochondrial NAD⁺ transporter and one of the pharmacological targets of metformin. SLC25A47 regulates lipid homeostasis through AMPKα and may serve as a potential drug target for treating NAFLD and hepatocellular carcinoma (2163).

SLC25A48: SLC25A48 has been reported to be highly expressed in brown adipose tissue, where it functions as a choline transporter in the inner mitochondrial membrane, which is essential for whole-body cold tolerance, thermogenesis, and mitochondrial respiration (2164, 2165). Choline uptake into the mitochondrial matrix via SLC25A48 facilitates betaine synthesis and one-carbon metabolism (Fig. 36). Cells lacking SLC25A48 showed reduced synthesis of purine nucleotides and failed to initiate the G1-S phase transition, leading to cell death. Thus, SLC25A48 plays a critical role in mitochondrial respiratory capacity, purine nucleotide synthesis and cell survival. Human loss-of-function mutations in SLC25A48 lead to impaired choline transport into mitochondria and are associated with elevated urine and plasma choline levels (2165).

Orphan transporter family members (14)

UCP2 (SLC25A8), UCP3 (SLC25A9), SLC25A14 (UCP5), SLC25A16 (GDC), SLC25A27 (UCP4), SLC25A30 (UCP6), SLC25A34, SLC25A35, SLC25A43, SLC25A45, SLC25A46, MTCH1 (SLC25A49), MTCH2 (SLC25A50), SLC25A53 (MCART3)

SLC26 Multifunctional anion exchanger family (2.A.53.2/Sulfate_transp/NAT)

Discovery: Slc26a1 (SAT-1) was the first member of this family to be discovered. It was identified by expression cloning from rat liver and shown to be a Na⁺-independent SO₄^2- transporter (151).

Gene family members (10 + 1 pseudogene):
SLC26A1 (SAT1)	SLC26A4 (Pendrin)	SLC26A7 (SUT2)	SLC26A10P (pseudogene)
SLC26A2 (DTDST)	SLC26A5 (Prestin)	SLC26A8 (TAT1)	SLC26A11 (SUT1, KBAT)
SLC26A3 (DRA)	SLC26A6 (PAT1, CFEX)	SLC26A9

Molecular aspects, physiological roles and links to disease

The 11 members of the SLC26 family include multifunctional anion exchangers and anion channels that transport a wide range of substrates, including bicarbonate (HCO₃^-), chloride (Cl^-), iodide (I^-), formate, and divalent ions such as sulfate (SO₄^2-) and oxalate (2166–2169) (Fig. 37). The citric acid cycle waste product bicarbonate (HCO₃^-) forms a crucial buffering system in intracellular and extracellular fluids, requiring specific membrane transport proteins to fulfill its biological function. In addition to the members of the SLC4 family, the SLC26 transporters play an essential role in this process.

Fig. 37. Phylogenetic tree of the SLC26 family.

The SLC26 family is part of a class of anion transporters that are found in all kingdoms of life. While homologs in prokaryotes and plants function as Na⁺- or H⁺-coupled transporters of mono- and divalent anions (2170, 2171), SLC26 members in animals show considerable mechanistic differences, as they function as either coupled anion exchangers, anion transporters with channel-like properties or in the case of SLC26A5 (prestin), as a motor protein in cochlear outer hair cells (2166). The molecular basis for the distinction between these different functional modes of mammalian SLC26 family members have been thoroughly studied (228). Whereas most these operate as exchangers of different monovalent and divalent ions (2166), SLC26A9 functions as a fast passive Cl^- transporter (2172, 2173).

SLC26 transporters belong to the Sulfate Permease (SulP) family (TC 2.A.53) which is part of the NAT fold family, whose members share a 7+7 TMH inverted repeat architecture with an N-terminal transmembrane domain and a C-terminal cytoplasmic domain called STAS (Sulfate Transporter and Anti-Sigma factor antagonist; see also Section 8) (2174). Structural and functional studies of SLC26A9 revealed a homodimeric architecture that is representative of the entire SLC26 family (2173, 2175). In general, each subunit consists of a transport domain, also called the core domain (TMHs 1-4 and 8−11), a scaffold domain, also called the gate domain (TMHs 5-7 and 12−14) and a STAS domain which has terminal PDZ motif known to commonly interact with multi-domain scaffolding proteins (2176–2178).

While there are still many unanswered questions regarding the function of the STAS domain, it is known to be important for dimerization, interacting with other proteins, having regulatory functions, and influencing anion transport. Cryo-EM structural studies of SLC26A4, SLC26A6 and SLC26A9 have provided additional insight (2173, 2176, 2177). Studies showed that the intracellular STAS domain, which starts after TMH 14, extends to the neighboring protomer without interacting with the TMHs of the same protomer. The dimer interface mainly involves interactions occurring at the two STAS domains exchanged between the two protomers (see the description of pendrin (SLC26A4) for more information).

Structural studies of SLC26 and also SLC4 family members have led to a consensus that the transport domain undergoes a rigid-body motion, while the scaffold domain together with the cytoplasmic STAS domains (which constitute the bulk of the dimer interface), form a contiguous rigid scaffold of the dimeric protein that presumably remains static during transport to expose the anion-binding site on either side of the membrane, in accordance with the elevator-type transport mechanism that facilitates alternate access of substrate transport across the membrane (2179).

The Cl^- transporter SLC26A9 mediates channel-like chloride currents that saturate at high mM concentration and efficiently discriminates against HCO₃^- (2173). On the other hand, several other SLC26 family members are Cl^-/HCO₃^- exchangers (2166). Structural evidence has emerged to explain this difference: SLC26A6, one such example of a Cl^-/HCO₃^- exchanger, has an arginine residue at position 404 (R404) that stabilizes bound anions, with its guanidium group potentially engaging in direct interactions with bicarbonate or oxalate. In the Cl^- transporter SLC26A9, this residue is replaced by an alanine residue (A390). Together with the presence of additional residues at the binding site of SLC26A6, the captured Cl^- can rely on a greater density of polar interactions, further providing a suitable environment for interaction with HCO₃^- and oxalate, neither of which are substrates of SLC26A9. A similar anion interaction is found in the paralogs pendrin (SLC26A4) (229) and prestin (SLC26A5) (226, 227), both of which have a comparable substrate preference. There is also a remarkable similarity between the core domains of SLC26A6 and prestin (SLC26A5), despite the altered function of the latter, which is no longer able to transport but instead acts as a motor protein in mammalian cochlear outer hair cells (2180), probably due to structural changes that prevent the transition from the inward-facing to the outward-facing conformation to release the bound anion into the extracellular environment as the final step of the transport cycle (see description of prestin/SLC26A5 below).

The STAS domains and other structural domains of the SLC26 family members are thought to be important for interacting with other proteins as well, including CFTR (2181) and SLC13A2 (1405). As highlighted in the SLC13 family description, due to the ability of oxalate transporter SLC26A6 to interact with the citrate transporter NAC1/NADC1 (SLC13A2), SLC26A6 plays a crucial role in renal stone formation (1405, 1406). In the case of CFTR, this interaction markedly activates Cl^-/OH^- exchange mediated by SLC26A3, SLC26A4, and SLC26A6 (2182). Further studies of combined overexpression of SLC26A3 and CFTR in a heterologous system showed that each protein stimulates the activity of the other through direct interaction of the SLC26A3 STAS domain with the R domain of CFTR, as well as indirectly through mediation of PDZ domain proteins (2183).

SLC26A1: SAT1 (SLC26A1) is a ubiquitously expressed pH-sensitive anion exchanger for sulfate and bicarbonate, which also mediates oxalate transport, with highest expression in liver, bone marrow and testis, small intestine and kidney (151, 2166, 2184). It is expressed on the basolateral membranes of intestinal epithelial cells and renal proximal tubules. In the latter case, after apical uptake of filtered sulfate by NaS1 (SLC13A1; see the SLC13 family description), SAT1 (SLC26A1) completes sulfate reabsorption at the basolateral membrane to maintain plasma sulfate levels. SAT1 is a major determinant of sulfate homeostasis in humans (2185). However, while rare variants of SLC13A1 have been found to be associated with intervertebral disc disease (1396), the pathophysiological role of genetic variation in SLC26A1 in sulfate homeostasis is less clear. Hyposulfatemia and hypersulfaturia have been reported in Slc26a1 knockout mice (2186), whereas the impact of SLC26A1 on oxalate homeostasis leading to hyperoxaluria and urolithiasis is less clear (2185). SLC26A1 has been identified as a putative target for modulation of musculoskeletal health (2185).

SLC26A2: DTDST (SLC26A2) is a Na⁺-independent sulfate/chloride exchanger with strong expression in intestine (colon, rectum), developing and mature cartilage, eccrine sweat glands, bronchial glands, and placental villi (2187) (2166). The transporter is important for the uptake of sulfate into chondrocytes, in order to maintain adequate sulfation of proteoglycans and has an important role in endochondral bone formation (2188). Biallelic mutations in SLC26A2 are the cause of diastrophic dysplasia, a rare autosomal recessive chondrodysplasia, resulting in impaired sulfate transport through cell membranes, intracellular sulfate depletion and inadequate sulfation of proteoglycans.

In the human colon, DTDST (SLC26A2) has been localized to the upper third of the crypts, where it is directed toward the apical membrane (2189).

SLC26A2 has been shown to be downregulated in the intestinal mucosa of patients with active ulcerative colitis compared to healthy controls (2190, 2191). The results indicate that SLC26A2 is negatively correlated with the IL-17 signaling pathway and positively associated with tight junctions. This leads to abnormal immune cell infiltration and inflammatory injuries during the pathogenesis of ulcerative colitis. Thus, SLC26A2 may serve as a protective candidate and a putative drug target for ulcerative colitis treatment.

SLC26A3: SLC26A3, also known as downregulated in adenoma (DRA), facilitates the cellular uptake of Cl^- or oxalate in exchange for intracellular HCO₃^- at the apical membrane of enterocytes in the distal ileum and the large intestine, where it is highly expressed (2167, 2184, 2192). The exchange of Cl^- for intracellular HCO₃^- formed from CO₂ by carbonic anhydrase is important because HCO₃^- makes the stool alkaline. In addition, the parallel exchange of Na⁺/H⁺ via NHE3 (SLC9A3) and the Cl^-/HCO₃^- antiport causes electroneutral NaCl absorption, an important digestive mechanism of sodium absorption.

The DRA (SLC26A3) protein was also detected in the glandular region of the stomach by Western blot analysis, where it is proposed to play a role in HCO₃^- secretion together with the Na⁺/H⁺ exchanger NHE8 (SLC9A8) (see the SLC9A8 description) (1262).

Loss of function mutations of SLC26A3 in humans (2193) and mice (2192) have been shown to cause chloride-losing diarrhea and reduce urinary excretion of oxalate, a major component of kidney stones. Slc26a3 knockout mice showed a significant reduction in urinary oxalate excretion compared to wildtype (2192, 2194). Mutations in SLC26A3 cause congenital chloride diarrhea (2193), the clinical presentation of which is lifelong, potentially fatal diarrhea with high chloride content (2193). These findings highlight the therapeutic utility of inhibiting this transporter for the treatment of hyperoxaluria and constipation (2195). Potent substituted 4-methylcoumarin inhibitors of SLC26A2 have now been developed that can be used for further preclinical development for the treatment of constipation and hyperoxaluria (2195, 2196).

SLC26A4: Pendrin (SLC26A4) is an electroneutral anion exchanger that facilitates HCO₃⁻ exchange for Cl⁻ and is crucial for maintaining pH and salt homeostasis in the kidney, lung, and cochlea. In the thyroid gland, it also exports iodide (I⁻). Specifically, at the apical side of the thyroid follicular cells, pendrin, together with an additional anion exchanger (SLC26A7; see below), mediates the efflux of iodide that has been taken up from the blood circulation by the basolateral Na⁺-iodide cotransporter NIS (SLC5A5). This is followed by thyroid hormone production in the follicular lumen.

In the airway surface epithelia of the lung, pendrin mediates bicarbonate secretion and enhances CFTR (ABCC7) function (2197) (see Fig. 33, bottom part). In the β-intercalated cells of the cortical collecting duct of the kidney, pendrin facilitates HCO₃^- secretion together with the CFTR anion channel in the apical membrane (2198) (see Fig. 12). The secretion of bicarbonate via pendrin also drives the reabsorption of chloride. CFTR is required for the activity of pendrin, but it remains to be determined whether CFTR stimulates pendrin by molecular interaction or other mechanisms, or whether it simply provides a pathway for chloride recycling across the apical membrane (2198).

In the inner ear, pendrin is found in the apical membrane of epithelial cells of the cochlea, vestibular system, and endolymphatic sac and duct. At this site, pendrin-driven chloride reabsorption and bicarbonate secretion control the intracellular and luminal fluids (2199, 2200). Pendrin in the inner ear therefore functions mainly as a Cl^-/HCO₃^- exchanger, secreting HCO₃^- into the endolymph and thereby increasing the endolymphatic pH.

Mutations in SLC26A4 have been identified in patients with autosomal recessive nonsyndromic deafness (DFNB4) and Pendred syndrome, a disorder associated with hearing loss and goiter (2200–2203). Pendred syndrome-associated mutations are found in the anion-binding site residues as well as in the STAS domain (2176).

The structures of pendrin from the wild swine Sus scrofa has been determined in the presence of either Cl⁻, I⁻,HCO₃⁻ or in the apo-state (2176). These findings provided new insights into how anions are selected and exchanged. They also provided new insight into the structural and functional role the cytosolic STAS domain of pendrin. STAS consists of five β-strands (β1-5) sandwiched between four α-helices. It functions as part of the dimer interface of pendrin, with the two STAS domains forming a domain-swapped dimer. Also contributing to the dimer interface is part of the N-terminus, residues 19 to 25 (βn), where the βns from adjacent protomers form an antiparallel beta sheet that interacts with the dimeric STAS domain. The relative angle between the STAS domain and the TM domains in SLC26 family members may influence the anion transport properties. Consistent with this, several of the Pendred syndrome mutations are located within the STAS domain, highlighting the structural and functional importance of the STAS domain, findings that may serve as the basis for therapeutic interventions to treat Pendred syndrome.

SLC26A5: Prestin (SLC26A5) is abundantly expressed in the outer hair cells that mediate cochlear amplification (2180). There it confers voltage-dependent somatic elongation and contraction, referred to as electromotility, which acts to amplify sound levels. It is an incomplete anion transporter that does not allow anions to cross the cell membrane, but instead undergoes a conformational change in response to changes in intracellular Cl^- levels, resulting in a change in cell length. Mutations in SLC26A5 have been associated with non-syndromic hearing loss (2204).

SLC26A6: SLC26A6, also known as PAT1 (putative anion transporter 1) or CFEX (chloride/formate exchanger), is an anion exchanger expressed in the apical membrane of the renal proximal tubule, the apical membranes of the pancreatic duct cells, and the brush border membrane of the duodenum. Note that the name PAT1 stands for “putative anion transporter 1” and should not be confused with the H⁺-coupled amino acid transporter encoded by SLC36A1.

SLC26A6 transports chloride, oxalate, sulfate, and bicarbonate. In the intestine, SLC26A6 and SLC26A3 play important roles in Cl^--dependent HCO₃^- secretion (2205).

In the apical membrane of kidney proximal tubule cells, SLC26A6 facilitates oxalate secretion, and phenotypic and functional analysis of SLC26A6 genetic variants in patients with familial hyperoxaluria and calcium oxalate nephrolithiasis confirmed that they disrupt the balance between citrate and oxalate excretion, thereby promoting kidney stone disease (2206). As mentioned in the SLC13 family description, SLC26A6 interacts with the citrate transporter NAC1/NADC1 (SLC13A2), which has an impact on kidney stone formation (1405, 1406).

In the pancreas, SLC26A6 together with SLC26A3 facilitates the excretion of basal HCO₃^-. The secretion of HCO₃^- and fluid is an essential function of the pancreatic ductal epithelium (2205, 2207). The bicarbonate-rich fluid, along with the digestive enzymes secreted by acinar cells, makes up pancreatic juice required for digestion in the small intestine (2208).

SLC26A7: SLC26A7 is a Cl^-/HCO₃^- exchanger that is most highly expressed in the thyroid and to a lesser extent in the kidney and stomach. As mentioned above (see description of SLC26A4), circulating iodide is concentrated in thyrocytes via the Na⁺-coupled iodide transporter NIS (SLC5A5) expressed on the basolateral membrane of thyrocytes, pendrin (SLC26A4) expressed on the opposite apical membrane contributes to iodide efflux into the follicular lumen, and genetic defects of SLC26A4 cause Pendred syndrome characterized by congenital deafness and thyroid goiter. However, iodide incorporation (i.e., incorporation of iodine into thyroglobulin and subsequent production of thyroid hormone) is only partially impaired in Pendred syndrome. This is due to the presence of an additional iodide transporter, SLC26A7. In fact, SLC26A7, also located on the apical membrane like SLC26A4, originally identified as a Cl^-/HCO₃^- exchanger in the stomach and kidney, fulfills this role as the most abundantly expressed iodide transporter in the thyroid.

In the stomach, SLC26A7 functions as a basolateral Cl^-/HCO₃^- exchanger in gastric parietal cells where it plays an important role in gastric acid secretion (2209). In the kidney, SLC26A4 is localized to the basolateral membrane of acid-secreting intercalated cells of the collecting duct and its expression is osmolarity- and pH-dependent (2210). Interestingly, while mice lacking Slc26a7 develop distal renal tubular acidosis (2211), individuals carrying a mutated SLC26A7 gene have normal acid-base status but develop goitrous congenital hypothyroidism (2212, 2213).

SLC26A8: TAT1 (SLC26A8) is a sperm-specific member of the SLC26 family of anion exchangers that associates with the CFTR channel and strongly stimulates its activity (2214). Ion fluxes play an essential role in the control of sperm motility, and homozygous deletion of Slc26A8 in mice results in male sterility due to a complete lack of sperm motility. TAT1 is a sperm-specific activator of CFTR and the two proteins work together to regulate the anion fluxes required for proper sperm motility and capacitation. TAT1 and CFTR form a molecular complex thought to be involved in the regulation of Cl^- and HCO₃^- fluxes during sperm capacitation. Mutations of SLC26A8 impair the formation of the SLC26A8-CFTR complex, consistent with an impairment of CFTR-dependent sperm activation events (2215).

SLC26A9: SLC26A9 is an epithelial anion transporter that is highly expressed in stomach, salivary gland and prostate, and it is also expressed at somewhat lower levels in lung and other tissues (2166). Functional studies in HEK293 cells revealed that SLC26A9 not only mediates Cl⁻/HCO₃^- exchange but is also capable of Cl⁻-independent HCO₃^- extrusion (2216). Subsequently, based on transport studies in Xenopus oocytes, HEK-293 cells and CHO cells, mouse SLC26A9 was shown to exhibit three distinct transport modes: 1) electrogenic nCl⁻/HCO₃^- exchange; 2) electrogenic Na⁺/nAnion^- cotransport; and 3) anion channel activity leading to constitutive Cl^- secretion (“n” represents the number of ions of a given type that are transported per cycle of the exchanger) (2217, 2218).

HCO₃^- secretion by gastric mucosal cells is essential for protection against acid injury and peptic ulcer. SLC26A9 has been identified as an apical HCO₃^- transporter in gastric surface epithelial cells (2216). In particular, in gastric parietal cells, SLC26A9 is important for acid secretion, and the absence of the Cl^- transporter in knockout mice caused loss of tubulovesicles in parietal cells and thus impaired acid secretion in the stomach (2219).

SLC26A9 has been suggested to contribute to airway surface fluid hydration via constitutive and regulated Cl^- transport, and several studies show functional and regulatory interactions between SLC26A9 and CFTR in airway epithelial cells (2220, 2221) (see Fig. 33, bottom part).

As highlighted above under “Physiological roles and links to disease”, cryo-EM structural studies have provided new insights into the gating mechanism of SLC26A9 with potential pharmacological applications in the treatment of disease-related dysfunction (2173, 2222). Subsequently, a potent and selective inhibitor of SLC26A9, S9-A13, was developed and the contribution of SLC26A9 and CFTR to airway transport was reported (2223). Surprisingly, it showed that SLC26A9 does not play a role in airway chloride secretion, but confirmed its role in gastric acid secretion. Nevertheless, the role of SLC26A9 in lung disease should not be overlooked, as previous studies have shown that SNPs of SLC26A9 can alter or mimic the CF phenotype, supporting the concept that SLC26A9 is a valuable therapeutic target to improve lung function in patients with CF or other lung diseases (2224–2226).

SLC26A10P – Pseudogene: SLC26A10P is an unprocessed transcribed (unitary) pseudogene. Protein-coding orthologs for this gene are annotated in other species including guinea pig where the SLC26A10 protein was found to be localized to the luminal membrane of pancreatic duct cells (2227).

SLC26A11: SLC26A11 (also known as SUT1 or KBAT) is ubiquitously expressed. In the brain, it is highly expressed in Purkinje cells (PCs) of the cerebellum, where it functions as a Cl^- transporter, possibly regulating acid translocation by H⁺-ATPase across the plasma membrane and in intracellular compartments (2228). Subsequently, based on studies in transgenic mice, it was shown that SLC26A11 plays a critical role in modulating chloride homeostasis and neuronal activity in the cerebellum, serving as a Cl^- transporter that regulates intracellular chloride in PCs, which is important for inhibitory neurotransmission and locomotor coordination (2229). SLC26A11 is suggested to be highly expressed in lymphoid tissues according to the HPA. Consistent with this, SLC26A11 was first characterized as an SO₄^2- transporter cloned from human lymphoid high endothelial venules (2230). The exact role of SLC26A11 in different tissues and whether it functions as a SO₄^2- or Cl^- transporter remains to be elucidated.

It is worth noting that, similar to SLC26A9, it has been shown that two independent modes of Cl⁻ transport exist, the classical electroneutral Cl^-/HCO₃^- exchange mode and the electrogenic Cl^- conductance mode, depending on the transmembrane potential (2231). Indeed, in the kidney, SLC26A11 colocalizes with the vacuolar H⁺-ATPase in intercalated cells, where SLC26A9 may function as a Cl⁻ channel or as a Cl^-/HCO₃^- exchanger on the apical membrane (2232).

Clinical relevance and pharmaceutical aspects

The study of SLC26-mediated molecular pathogenesis not only helps elucidate disease mechanisms, but also facilitates the discovery of therapeutic targets. Diseases associated with genetic mutations in SLC26 family members have been reviewed (2184) and the following is a summary of them: SLC26A1, calcium oxalate nephrolithiasis, hyperoxalemia; SLC26A2, skeletal deformities and abnormal cartilage development; SLC26A3, chloride diarrhea, alkalosis; SLC26A4, Pendred syndrome, deafness, EVA (enlarged vestibular aqueduct) syndrome, and thyroid lesions; SLC26A5, non-comprehensive deafness; SLC26A6, diseases related to bicarbonate ion metabolism; SLC26A7, hypothyroidism; SLC26A8, asthenozoospermia (weak sperm disease); SLC26A9, cystic fibrosis; SLC26A11, dysregulation of chloride homeostasis and neuroactivity. In addition, altered expression of SLC26 members due to inflammation or surgery has important consequences for intestinal transport and barrier function in common diseases such as IBD or bariatric surgery (2169).

Disorders of oxalate homeostasis cause hyperoxalemia and hyperoxaluria, leading to kidney stone disease with formation of calcium oxalate stones. SLC26 proteins have been shown to be aberrantly expressed during nephrolithiasis and thus may represent therapeutic targets, e.g., SLC26A3 (2184).

Orphan transporter family members: N/A

SLC27 Fatty acid transporter family (4.C.1/AMP-binding/single TMH)

Discovery: The SLC27 FATP fatty acid transport proteins function in the plasma membrane or at intracellular membrane junctions with the endoplasmic reticulum as gates in the regulated cellular uptake of saturated, monounsaturated or polyunsaturated long-chain (LCFAs; containing 13-21 carbons) or very long-chain (VLCFAs, containing at least 22 carbons) fatty acids. FATPs were originally identified based on their ability to increase long-chain fatty acid (LCFA) uptake when expressed in cells (163). LCFAs and VLCFAs are important energy substrates for cardiomyocytes and other cells and are involved in numerous intracellular signaling pathways. An expression cloning strategy using a cDNA library from 3T3-Ll adipocytes was employed to screen for LCFA uptake when expressed in cultured cells. The identified fatty acid transport protein (FATP) was reported to function as a plasma membrane transporter for LCFAs despite having only a single TMH domain, since it was identified by expression cloning based on its ability to increase the translocation of fatty acids, such that when overexpressed, host cells exhibited FATP-triggered accumulation of radioactive or fluorescent fatty acids (2233).

Gene family members (6)
SLC27A1 (FATP1)	SLC27A3 (FATP3)	SLC27A5 (FATP5)
SLC27A2 (FATP2)	SLC27A4 (FATP4)	SLC27A6 (FATP6)

Molecular aspects, physiological roles and links to disease

The SLC27 family has 6 members (Fig. 38) and belongs to the Fatty Acid Group Translocation (FAT) family (TC 4.C.1), which includes acyl-CoA synthetases such as fatty acyl-CoA synthetases, as well as the FATPs, which also have acyl-CoA synthetase activity (2234). FATPs have a short extracellular N-terminal segment and a longer cytosolic C-terminal segment containing both an AMP-binding region and a long-chain acyl-CoA synthetase (ACSL) catalytic domain.

Fig. 38. Phylogenetic tree of the SLC27 family.

FATPs facilitate the transport of LCFAs and VLCFAs across cell membranes and convert unesterified fatty acids into fatty acyl-CoA. To perform this function, FATPs must be anchored to the cell membrane. However, single TMH membrane proteins typically do not have direct membrane transport functions (2235), and it is unclear whether LCFAs or VLCFAs require a transport mechanism or can simply diffuse across the membrane and be acylated intracellularly. A fatty acid flip-flop model has been proposed. In this model, LCFA is transferred from albumin to the outer leaflet of the plasma membrane. Protonation then results in a lipid-soluble fatty acid (2236). This fatty acid is then translocated to the inner leaflet. However, it has been questioned whether this mechanism is fast enough to allow cellular uptake. At present, it is not known whether FATPs directly facilitate LCFA and VLCFA or whether other membrane proteins are involved. It is also not clear whether a flip-flop of LCFAs would be sufficient for translocation. Regardless of how LCFA or VLCFA enter the cell, intracellular lipid thioesterification with coenzyme A through FATP will safely block the lipid from leaving the cell and ensure unidirectional transport (2237, 2238). FATP-mediated transport of LCFAs or VLCFAs may simply involve vectorial acylation of fatty acids, leading to their intracellular capture to form fatty acyl-CoA esters in the cytosol.

In addition to FATPs, several other proteins are involved in cellular fatty acid uptake (2239, 2240), including fatty acid translocase CD36/SR-B2 (CD36) (see SLC2A4 description), which is expressed in cardiomyocytes, adipocytes, erythroid precursors and macrophages (2241) (2242, 2243), low-density lipoprotein receptor (LDLR) that mediates LDL transport across endothelial cells via caveolae (2244, 2245), and fatty acid binding proteins (FABPs) (2246). Moreover, MFSD2A (SLC59A1) (see the SLC59 family description), which is expressed in the blood-brain barrier endothelium, is responsible for the deposition of omega-3 fatty acid docosahexaenoic acid (ω3-DHA) in the brain. Note that in this review the abbreviation ω3-DHA is used for this omega-3 fatty acid rather than DHA. This is to distinguish it from DHA, which stands for dehydroascorbic acid.

The uptake and metabolism of LCFAs and VLCFAs via FATPs is critical for many physiological and cellular processes, including membrane synthesis, intracellular signaling, energy metabolism, post-translational modifications, and transcriptional regulation of metabolic genes (2247). In addition, FATPs represent potential therapeutic targets to prevent the acquisition of fatty acids, particularly saturated and trans-unsaturated fatty acids, that contribute to disease, as in the case of fatty acid overload (2238). A great number of obesity-related diseases are due to an abnormal influx of LCFA from adipose stores into highly metabolic tissues such as heart, liver, and muscle, where the abnormal accumulation of lipids leads to insulin resistance, endoplasmic reticulum stress, and cell death (2233).

Since cellular fatty acids are usually derived from uptake from the extracellular milieu rather than de novo synthesis, fatty acids must be taken up by cells across the plasma membrane, which is aided by FATPs with immediate intracellular activation to their CoA thioesters to make them available for further metabolism in different tissues. Much of what we know about the physiological role of protein-mediated fatty acid uptake comes from the characterization of FATP knockout and transgenic mouse strains, human data, and in vitro studies (2240, 2247, 2248).

SLC27A1: FATP1 (SLC27A1) is abundantly expressed in skeletal muscle, heart, adipose tissue and brain (163, 2234, 2249).

Analogous to other FATPs, FATP1 is a single-pass type I transmembrane protein with the N-terminal portion on the extracellular side and the C-terminal portion on the cytosolic side. In FATP1, the latter contains an 11 amino acid motif (IYTSGTTGXPK) characteristic of proteins that either have an interaction side with ATP or catalyze reactions such as acyl-CoA synthetases (2250).

In terms of its acyl-CoA synthetase activity, FATP1 shows activity towards saturated fatty acids such as palmitic acid, but can also activate polyunsaturated fatty acids (2234, 2251).

Insulin has been shown to be a regulator of FATP1 (2247). Analogous to the insulin-sensitive glucose transporter GLUT4, a model has been proposed in which insulin counteracts the postprandial rise in dietary lipids by increasing the expression of FATP1 on the plasma membrane of adipocytes and muscle cells. According to this model, the basal LCFA uptake is mediated by other proteins, including FATP4 and CD36 (2252).

While adipocytes are the main suppliers of fatty acids, cancer-associated fibroblasts also act as hubs of fatty acids to meet the needs of cancer cells, and it has been shown that FATP1 plays a pivotal role in fatty acid transfer between breast cancer cells and non-cancerous cells in the microenvironment (2253). The results suggest that FATP1 is a putative therapeutic target for breast cancer treatment (2253).

FATP1 is also highly expressed throughout the brain, including at the BBB, where it facilitates the transport of fatty acids, including the essential polyunsaturated fatty acid ω3-DHA, which it does together with FATP4 (SLC27A4) (2249, 2254, 2255). FATP1 and FATP4 can trap fatty acids, including ω3-DHA, in the endothelial cell by forming acyl-CoA. It has also been shown that insulin rapidly increases the supply of ω3-DHA to the brain by promoting the translocation of FATP1 to the cell membrane. Thus, FATP1 also facilitates the transport of neurosupportive substances to the brain such as ω3-DHA (2249). Note that as mentioned above, MFSD2A (SLC59A1) also plays an important role in ω3-DHA delivery to the brain, but unlike FATPs, it only facilitates the transport of esterified ω3-DHA.

According to the HPA, FATP1 (SLC27A1) is also highly expressed in the choroid plexus. The role of the choroid plexus as an alternative pathway to supply brain tissue with very long unsaturated chains has been discussed previously (2256). However, there is no further information available on this subject, such as the cellular and subcellular localization of FATP1 in choroid plexus epithelial cells (2256).

SLC27A2: FATP2 (SLC27A2) is primarily expressed in the liver and kidney. It has been shown to function as both a fatty acid transporter and an acyl-CoA synthetase (ACSL) (2247).

FATP2 has also been shown to be a useful therapeutic target for the treatment of a variety of cancers, including thyroid cancer, because it provides cancer cells with increased exogenous fatty acids for lipid metabolism (2257).

SLC27A3: Little is known about FATP3 (SLC27A3), which is encoded by the human SLC27A3 gene. Expression analysis has shown that mouse Slc27a3 is highly expressed in the mouse adrenal gland, testis, ovary and lung (2258). In addition, while SLC27A3 is weakly expressed in the neonatal and adult brain, it is highly expressed in the embryonic brain (2258).

SLC27A4: FATP4 (SLC27A4) is the primary FATP expressed in enterocytes and is specifically localized on the apical side of intestinal epithelial cells, suggesting that it is involved in intestinal LCFA absorption (2259). However, mice lacking the transporter showed no appreciable protection against high-fat diet-induced weight gain, suggesting that other lipid transport proteins are involved (2260).

FATP4 is also highly expressed in skin where it plays an important role in maintaining skin barrier function through ceramide metabolism, contributing to epidermal barrier function. Specifically, it is required for the incorporation of saturated VLCFAs into epidermal ceramides and monoacylglycerols. Mutations that disrupt human FATP4 are found in patients with ichthyosis prematurity syndrome, a rare autosomal recessive disorder that manifests with premature birth, respiratory symptoms, and swollen skin with severe caseosa-like scaling (2261, 2262).

FATP4 (SLC27A4) is also highly expressed in the brain, where it works together with FATP1 (SLC27A1) to transport fatty acids across the BBB (2255, 2263).

SLC27A5: FATP5 (SLC27A5) is unlikely to play a significant role in fatty acid transport, but instead plays a major role in bile acid recycling. It activates the primary unconjugated bile acid (cholic acid) to its CoA thioester derivative (cholate), as well as secondary bile acids via its bile acid-CoA ligase activity (2264). In this function, FATP5 is required for bile acid reconjugation but not for de novo synthesis. During bile acid biosynthesis, the bile acid-CoA thioester intermediate is synthesized by FATP2 (SLC27A2). However, a specific bile acid-CoA ligase, FATP5 (SLC27A5), also known as bile acyl-CoA synthetase (BACS), is used for previously synthesized unconjugated bile acids that are returned to the liver in the enterohepatic circulation (2265). Regardless of the source of the bile acid-CoA thioester, bile acids are then conjugated to taurine or glycine by bile acid-CoA amino acid N-acyltransferase.

Interestingly, the expression of SLC27A5 was found to be decreased in the livers of cirrhotic patients and mice with liver fibrosis (2266). In addition, in the same study, Slc27a5^-/- mice showed enhanced progression of liver fibrosis due to activation of hepatic stellate cells. Mechanistically, SLC27A5 deficiency led to the accumulation of unconjugated bile acids, particularly cholic acid, which induced liver fibrosis through activation of hepatic stellate cells mediated by the transcriptional regulator EGR3 (early growth response protein 3). The study not only provides new mechanistic insights into the role of SLC27A5 in the regulation of liver fibrosis, but also highlights a therapeutic strategy for the treatment of liver fibrosis involving the restoration of SLC27A5 expression.

SLC27A6: FATP6 (SLC27A6) is expressed primarily in the heart, specifically in the sarcolemma of cardiomyocytes and in plasma membranes adjacent to the blood vessels of the heart. FATP6 has been shown to function as a fatty acid transporter in the heart with preference for palmitic acid (2267). It likely plays an important role in lipid-related heart disease (2247).

Orphan transporter family members: N/A

SLC28 Na⁺-coupled nucleoside transport family (2.A.41.2/Gate/CNT)

Discovery: The rat intestinal pyrimidine nucleoside transporter CNT1 (SLC28A1) was identified by expression cloning using Xenopus oocytes (148).

Gene family members (3):
SLC28A1 (CNT1)	SLC28A2 (CNT2)	SLC28A3 (CNT3).

Molecular aspects, physiological roles and links to disease

The SLC28 family has three members, SLC28A1 (CNT1), SLC28A2 (CNT2), and SLC28A3 (CNT3), which function as Na⁺-coupled nucleoside transporters. The SLC28 transporters belongs to the Concentrative Nucleoside Transporter (CNT) family (TC 2.A.41), which harbors the CNT fold. The cryo-EM structure of CNT3 (SLC28A3) has been reported, showing a trimeric structure with each protomer consisting of 11 TMHs (298).

SLC28A1, SLC28A2, SLC28A3: CNTs mediate unidirectional uptake of nucleosides coupled to the influx of Na⁺. The Na⁺/nucleoside coupling ratio is 1:1 for CNT1 (SLC28A1) and CNT2 (SLC28A2), and 2:1 for CNT3 (SLC28A3). A distinguishing feature of CNT3 (SLC28A3) among CNTs is the acceptance of H⁺ in its ion-binding pocket, allowing H⁺-coupled nucleoside transport, which is most likely relevant to ER-related functions of this protein. The three members also differ in their substrate selectivity, except for uridine, which can be transported by all subtypes. The detailed biochemical properties such as substrate selectivity and specificity of the CNT nucleoside transporters have been reviewed elsewhere (296, 2268, 2269).

Nucleoside transporters have important functions in nucleoside homeostasis, providing nucleosides and nucleobases derived from the diet or produced by tissues such as liver for salvage pathways of nucleotide synthesis in tissues and cells deficient in de novo biosynthetic pathways. Nucleoside transporters also regulate many cellular processes including neurotransmission, vascular tone, adenosine concentration near cell surface receptors and nucleoside drug transport and metabolism. While human CNT1 (SLC28A1) and CNT2 (SLC28A2) are selective for pyrimidine and purine nucleosides, respectively, human CNT3 shows a broad acceptance for different pyrimidine and purine nucleosides (2270).

The CNTs of the SLC28 family and the ENTs of the SLC29 family work together to enable transepithelial uptake of nucleosides across the intestinal and renal epithelial layers, transport into and out of hepatocytes and transport across endothelial cells of the BBB and the blood-cerebrospinal fluid barrier of the choroid plexus.

In the intestine, transepithelial transport of nucleosides is enabled by CNT1 (SLC28A1), CNT2 (SLC28A2) and CNT3 (SLC28A3) in the apical membrane and ENT1 (SLC29A1), possibly also ENT2 (SLC29A2), in the basolateral membrane (Fig. 33). In the renal proximal tubules, CNT3 (SLC28A3) in the apical membrane and ENT2 (SLC29A2) in the basolateral membrane allow reabsorption of filtered nucleosides such as adenosine. The liver plays a key role in systemic nucleoside and nucleoside drug homeostasis. In hepatocytes, CNT1 (SLC28A1) and CNT2 (SLC28A2) in the blood-facing sinusoidal membranes facilitate hepatic uptake of nucleosides, and at the bile-facing canalicular membranes, CNT1 (SLC28A1) and CNT2 (SLC28A2) enable hepatic reuptake of nucleosides into hepatocytes.

CNT2 and CNT3 are also relevant to purinergic signaling. In the kidney they are thought to play an important role in adenosine-mediated tubulo-glomerular feedback regulation (2271).

CNT2 has been identified on the luminal side of the BBB endothelium and on the apical side of the choroid plexus epithelium (2272) (Fig. 11). The evidence suggests a role for CNT2 in the removal of adenosine from brain extracellular fluids. Whether the exit into the blood in these endothelial or epithelial cells is facilitated by an ENT/SLC29 transporter or another SLC solute carrier remains to be determined.

As indicated above, cryo-EM structural studies of CNT3 revealed new insight into the molecular architecture of CNTs. CNT3 forms a trimer and each protomer consists of 11 TMHs (298). The studies show that each protomer has two reentrant hairpin loops and three interfacial helices (IHs), consistent with previous findings (297). The N-terminal region of CNT3, which comprises TMHs 1 to 3 and IH1, shares high sequence conservation with the paralogs CNT1 and CNT2. The scaffold domain consists of TMH4, TMH5, TMH6, TMH9, and IH2. The transporter domain (IH3, HP1, TMH7, TMH8, IH4, HP2, TMH10, and TMH11) has two structurally inverted repeats that are linked by TMH9. Structural investigations of bacterial CNT homologs showed that CNT transporters operate according to an elevator mechanism, where the transport domain is displaced relative to the scaffold domain during the transport cycle (297, 299).

Several polymorphisms have been described in CNT and ENT proteins that likely affect nucleoside homeostasis, adenosine signaling events, or the cytotoxicity or pharmacokinetics of nucleoside drugs (2273).

Clinical relevance and pharmaceutical aspects

SLC28 family CNTs and SLC29 family ENTs mediate the uptake of a variety of nucleoside drugs, many of which are used in anticancer therapy (296). Relevant drug substrates are listed elsewhere (2273). Among the CNTs, CNT3 (SLC28A3) is particularly suitable for drug delivery because it has a relatively broad substrate specificity, accepting both purine and pyrimidine nucleosides as well as a variety of anticancer and antiviral nucleoside-derived drugs. Therefore, CNT3 (SLC28A3) is also an obvious choice for routinely testing against novel anticancer and antiviral agents.

The cryo-EM structural studies of CNT3 (SLC28A3) not only provide the molecular determinants for the transport mechanism of CNTs, but also facilitates the design of nucleoside drugs (298).

Orphan transporter family members: N/A

SLC29 Facilitative nucleoside transporter family (2.A.57.1/Nucleoside_tran/MFS)

Discovery: Purification and N-terminal sequencing of the isolated protein responsible for Es-type (see below) nucleoside transport activity from human erythrocytes allowed the cloning of a human placental cDNA encoding the corresponding transporter, designated hENT1 (human equilibrative nucleoside transporter 1, SLC29A1) (2274).

Gene family members (5):
SLC29A1 (ENT1)	SLC29A2 (ENT2)	SLC29A3 (ENT3)
SLC29A4 (ENT4/PMAT)	CLN3 (SLC29B1)

Molecular aspects, physiological roles and links to disease

The SLC29 family contains three equilibrative nucleoside transporters (ENTs), an organic cation transporter (ENT4/PMAT) and the more distantly related lysosomal membrane protein, CLN3 (SLC29B1), which is required for the lysosomal clearance of glycerophosphodiesters (GPDs) (Fig. 39).

Fig. 39. Phylogenetic tree of the SLC29 family.

The SLC29 family belongs to the widespread Equilibrative Nucleoside Transporter (ENT) family (TC 2.A.57.1), which is part of the MFS superfamily. The cryo-EM structure of ENT1 in complex with adenosine reuptake inhibitors has been reported (2275). It has 11 TMHs with the N-terminus on the cytosolic side and the C-terminus on the extracellular side, and a glycosylated extracellular loop between TMHs 1 and 2, consistent with previous studies (2276). ENT1 exhibits a pseudo-symmetric 6+5 topology in which the first 6 TMH and the remaining 5 TMH form separated bundles (2275). Interestingly, the structure of ENT1 shows distinct structural deviations from the classical architecture of MFS transporters, which follows a strict 12-TMH topology.

SLC29A1, SLC29A2: ENT1 (SLC29A1) and ENT2 (SLC29A2) are equilibrative nucleoside transporters (296, 2268) and nicotinamide transporters (2277).

The ENT family has been divided into two types of transporters, Es and Ei, based on their sensitivity to inhibition by nitrobenzylthioinosine (NBMPR), with Es transport sensitive to NBMPR and Ei transport insensitive. Based on sequence similarity, rat ENT1 (Slc29a1), which mediates Es transport, and rat ENT2 (Slc29a2), which mediates Ei transport, have been identified by molecular cloning (2278). ENT1 and ENT2 are both expressed in the basolateral membranes of enterocytes in the small intestine, where they mediate cellular exit following uptake via the concentrative Na⁺-coupled counterparts CNT1 (SLC28A1) and CNT2 (SLC28A2) expressed in the apical membranes, thereby facilitating the absorption of dietary nucleosides. According to the HPA, ENT1 is ubiquitously expressed, whereas ENT2 is most highly expressed in skeletal muscle cells.

Purine and pyrimidine nucleosides play important physiological roles and their derivatives offer a wide range of pharmacological applications. In addition, these nucleosides can be converted to nucleotides, the energy-rich elements of intermediary metabolism, precursors of nucleic acids, and key players in signaling (296, 2268). ENTs play critically important roles in adenosine signaling, cellular uptake of nucleoside for DNA and RNA synthesis, and nucleoside-derived anticancer and antiviral drug delivery.

The drug dipyridamole, which is used to dilate blood vessels in patients with peripheral arterial disease and coronary artery disease, is thought to potentiate the action of endogenous adenosine by blocking its cellular uptake via the nucleoside transporter ENT1 (SLC29A1) through which adenosine enters erythrocytes and endothelial cells (80).

As noted in the SLC28 family description, the concentrative nucleoside transporters of the CNT/SLC28 family are Na⁺-coupled transporters predominantly found in epithelial cells of the intestine and kidney, as well as in the liver, the BBB and the choroid plexus (Fig. 11). ENTs in contrast are ubiquitous in most, if not all, cell types and mediate bidirectional fluxes of purine and pyrimidine nucleosides down their concentration gradients. These nucleoside transporters are also expressed in the liver (canalicular membrane, sinusoidal membranes), BBB endothelial cells and the choroid plexus (2268, 2279). Importantly, ENT1 on erythrocytes is the major ENT responsible for the uptake of extracellular adenosine, which is why extracellular adenosine has a short half-life (see below under “Clinical and Pharmaceutical Aspects section”) (2280).

The cryo-EM structural studies of ENT1 (SLC29A1) (see above) together with mutagenesis studies unveils how the non-nucleoside vasodilator dilazep can bind and inhibit ENT1 (2275). This work offers great opportunities for the rational design of improved therapeutic drugs modulating nucleoside transport.

It should be noted that there is another nucleobase transporter as part of the SLC43 family (see EEG1 in the SLC43 family description) that plays an important role in hepatocytes.

A subsequent study surprisingly revealed that the human ENT1 (SLC29A1) and ENT2 (SLC29A2) transporters also function as nicotinamide cell membrane transporters (2277). In addition to nicotinic acid (also known as niacin or vitamin B3), nicotinamide is a key precursor in the synthesis of NAD⁺ in bodily tissues. Transporters for other NAD⁺ precursors have previously been identified, including those for nicotinic acid [SMCT1 (SLC5A8) and OAT2 (SLC22A7)], and NMN [CCC9 (SLC12A8)]. Additionally, the mitochondrial carriers SLC25A47 and SLC25A51 transport NAD⁺ into mitochondria, triggering biological responses (see the corresponding SLC25 summaries). The discovery that ENT1 (SLC29A1) and ENT2 (SLC29A2) are nicotinamide transporters provides a crucial piece of the puzzle for the metabolic network of nicotinamide and NAD⁺ (2277). The findings lay the groundwork for creating more effective NAD⁺-boosting strategies to enhance the anti-aging, metabolism-regulating, and organ-protecting/repairing effects of nicotinamide.

SLC29A3: ENT3 (SLC29A3) is a widely distributed lysosomal transporter that controls nucleoside flow from the lysosome to the cytoplasm, including adenine, adenosine, and uridine (2281, 2282). ENT3 facilitates the release of nucleosides and nucleobases produced by nucleic acid breakdown in the lysosomal interior (2281). ENT3 differs from ENT1/2 in possessing a very long (51 residues), hydrophilic N-terminus (2281). This region contains a dileucine motif characteristic of endosomal/lysosomal targeting sequences. A mitochondrial localization has also been reported for ENT3, where it likely functions in the uptake of nucleosides and nucleoside drugs (2283).

ENT3 is critical for nucleoside transport in cells that cannot synthesize their nucleosides de novo, is functional in intracellular organelles such as lysosomes, and transports purine and pyrimidine nucleosides in a pH-dependent manner. The optimal pH is 5.5, reflecting the location of the transporter in acidic intracellular compartments such as lysosomes (2284, 2285). ENT3 maintains the availability of the cytoplasmic nucleotide pool required for several essential processes in energy metabolism such as ATP/GTP generation, signal transduction pathways, and the nucleoside salvage pathway (2284).

ENT3 not only transports hydrophilic nucleosides and nucleobases, but also shows broad selectivity towards hydrophilic antiviral and anticancer nucleoside drugs (2285). Thus, ENT3 is recognized as a vital player in nucleosides, nucleobases, hydrophilic anticancer and antiviral nucleoside drug transport, energy metabolism and signal transduction, and its deactivation due to pathological mutations is associated with the onset, progression and prognosis of hereditary disorders and tumors as indicated above (2285).

SLC29A3 mutations and alterations in expression lead to various inherited diseases and cancers (2285): One of the inherited diseases is H syndrome, an autosomal recessive disease cluster characterized by cutaneous hyperpigmentation, hypertrichosis, hepatosplenomegaly, cardiac anomalies, hearing loss, hypogonadism, short stature, hallux valgus, among others (2286–2288).

Another inherited disease associated with SLC29A3 mutations is pigmentary hypertrichosis and non-autoimmune insulin-dependent diabetes mellitus (PHID) syndrome, an allelic variant of H syndrome that has large overlaps with Rosai-Dorfman disease (2289), and Faisalabad histiocytosis (FHC), an autosomal recessive form of histiocytosis (2285, 2290–2292). A progressive and spontaneous macrophage-dominated histiocytosis was observed in a mouse model lacking ENT3 expression (2293).

The loss of function of ENT3 (SLC29A3) is also closely associated with the onset, development, and prognosis of a variety of human cancers (2285). Inhibition of ENT3 may serve as an effective strategy to potentiate the anticancer activity of chemotherapy.

As reviewed elsewhere (2285), the possible reasons why different pathogenic mutations of SLC29A3 lead to different pathologies are: 1) Subcellular localization, nucleoside transport, protein stability and pH sensing may be differentially affected by different mutations in SLC29A3; 2) deficiency in nucleoside transport due to different SLC29A3 mutations may lead to different intracellular nucleoside accumulation; 3) because ENT3 functions in the membranes of both lysosomes and mitochondria, different SLC29A3 mutations may alter the nucleoside pools in these organelles in different ways, potentially disrupting lysosomal and mitochondrial homeostasis, which depends on ENT3 substrates such as nucleosides and nucleobases; 4) aberrant transport of nucleosides into the cytosolic pool due to SLC29A3 mutations increases lysosomal pH, thereby blocking critical cellular pathways; 5) SLC29A3 deficiency interferes with the proper clearance of apoptotic cells and increases macrophage colony-stimulating factor activation, thereby affecting the lysosomal system and leading to increased macrophage numbers and histiocytosis.

SLC29A4: ENT4/PMAT (SLC29A4) has been identified as a novel human plasma membrane monoamine transporter called PMAT, that is highly expressed throughout the CNS but is not homologous to any of the previously known neurotransmitter transporters and only exhibits low sequence identity to the other members of the SLC29 equilibrative nucleoside transporter family (2294). PMAT is a multi-specific organic cation transporter rather than a prototypic ENT low-affinity adenosine transporter. Specifically, it efficiently transports serotonin (K_m = 114 μM) and dopamine (K_m = 329 μM). Transport is not Na⁺ or Cl^– dependent but appears to be sensitive to changes in membrane potential.

Uptake of released monoamines into presynaptic neurons is mainly carried out by a family of Na⁺- and Cl^-- dependent high affinity plasma membrane transporters, which includes the norepinephrine transporter NET (SLC6A2), the dopamine transporter DAT (SLC6A3) and the serotonin transporter SERT (SLC6A4), but several lines of evidence including from studies of Slc6a3 and Slc6a4 knock-out studies in mice provide evidence for the existence of alternative monoamine transporters. Several studies highlight PMAT as an important additional contributor regulating monoamine neurotransmitter levels in the brain (2294, 2295).

Data also suggest that PMAT is expressed on the brush border membrane of enterocytes (2296), where it is predicted to contribute to the uptake of the oral anti-diabetic drug metformin, together with OCT1 (SLC22A1), OCT3 (SLC22A3) and SERT (SLC6A4) (2297).

The HPA suggests that PMAT is highly expressed in horizontal and bipolar cells of the retina, enteroendocrine cells, female germ cells, at somewhat lower levels in pancreatic endocrine cells, inhibitory neurons and astrocytes, and at lower levels in enterocytes. At the blood-retinal barrier, PMAT has been shown to transport of MPP⁺, the dopamine neurotoxin leading to parkinsonism (2298).

CLN3 (SLC29B1): CLN3 (SLC29B1), also known as battenin (BTN1), is a widely expressed lysosomal/endosomal transmembrane protein (179, 2299–2301). CLN3 is required for clearance of lysosomal glycerophosphodiesters (GPDs) (2302). CLN3 is somewhat more distantly related to the SLC29 family members and has therefore been classified within the SLC29 family in a new subgroup B as SLC29B1, an alias for CLN3.

Batten disease, also known as neuronal ceroid lipofuscinoses, is one of the most devastating forms of neurodegenerative lysosomal storage disorders. It is caused by mutations in CLN3, and loss of the Batten disease protein CLN3 leads to mannose-6-phosphate receptor (M6PR) mistrafficking and defective autophagic-lysosomal reformation (2300).

Clinical and pharmaceutical aspects

ENT1 and ENT2 transport a wide range of therapeutically important anticancer nucleosides and nucleobases (2268). The crystal structures of hENT1 in complex with adenosine reuptake inhibitors (2275), combined with mutagenesis studies (2303), reveal the inhibitory mechanisms of human ENT1 and provide insight into adenosine recognition and transport as well as the design of ENT subtype-specific inhibitors or nucleoside analogues for drug delivery via ENT1. These advances are important for future structure-based drug design, as ENT subtype-specific inhibitors are still lacking, and for the development of new drug delivery strategies, as ENT1 and ENT2 are already known to be involved in the transport of many clinically important nucleoside/nucleotide analogues. These include the antiviral drugs remdesivir and EIDD-1931 used to treat COVID-19, fialuridine, a nucleoside analog that has been investigated as a potential therapy for hepatitis B virus infection, and gemcitabine, a chemotherapeutic drug used to treat various types of cancer (2304).

In addition, adenosine is known to exert beneficial effects through its signaling pathways by regulating physiological and pathological processes, thereby protecting cells from damage caused by increased metabolism and protecting organ dysfunction due to pathological conditions. Adenosine is released during hypoxia, ischemia, β-adrenergic stimulation, and inflammation. It also affects cardiac rhythm and causes vasodilation in the systemic, coronary or pulmonary vasculature (2305). As a result, adenosine helps to control the body through various signaling pathways triggered by adenosine membrane receptors called A1R, A2AR, A2BR and A3R (2305). The main sources of adenosine in the blood are endothelial and muscle cells, where it is generated by dephosphorylation of AMP by specific nucleotidases. Adenosine release also occurs after adrenergic stimulation. At the extracellular level, adenosine is generated by the dephosphorylation of ATP and AMP by the cell surface enzymes CD39 and CD73. Intracellular adenosine leaves cells via ENT transporters. In the extracellular space, adenosine has a short half-life due to its uptake by erythrocytes via ENT transporters. During hypoxia, ischemia, or inflammation, the release of adenylyl nucleotides increases, and the concentration of adenosine increases both intracellularly and extracellularly. Because of its short half-life, adenosine allows very rapid adaptation of the cardiovascular system. However, the effects of adenosine on the cardiovascular system are sometimes beneficial and sometimes detrimental. The development of novel modulators of adenosine receptors and ENT transporters may be beneficial to slow down or conversely enhance the adenosinergic response according to the occurrence of different pathological conditions (2305–2307).

As mentioned above, targeting ENT3 may serve as an effective strategy to potentiate the anticancer activity of chemotherapy (2285).

The cryo-EM structural work of ENT1 (SLC29A1) (2275) opens the door to the rational design of advanced therapeutic drugs that specifically modulate nucleoside transport.

In addition, the cryo-EM structure of the Plasmodium falciparum nucleoside transporter PfENT1, which is essential for purine nucleoside uptake from the host, has been reported in apo-, inosine-, and inhibitor-bound states (2308). Malaria, caused by infection with Plasmodium parasites, is a global infectious disease threat to human health. Specific inhibitors of PfENT1 prevent the proliferation of P. falciparum at submicromolar concentrations. The elucidation of the precise substrate recognition and inhibitory mechanism of PfENT1 is of great importance (2308).

Orphan transporter family members: N/A

HGNC update

SLC29B1 is a new alias for CLN3.

SLC30 Zinc efflux family (2.A.4/Cation_efflux/CDF)

Discovery: With respect to the identification of the SLC30 family, baby hamster kidney (BHK) cells that had been transfected with a zinc-responsive reporter gene (MRE-bGeo) were extensively mutagenized and screened for variants with high basal expression of the reporter gene. One recessive clone was identified that not only had a high basal expression of the reporter gene but was also more sensitive to zinc toxicity (2309). This revealed an easy strategy to select for genes that could confer resistance to zinc toxicity. Rat Slc30a1 (ZnT1) and Slc30a2 (ZnT2) cDNAs were recovered from this screen (2310) and sequencing revealed that they were homologous to yeast genes ZRC1 and COT1, which had been shown to confer resistance to zinc toxicity.

Gene family members 11:
SLC30A1 (ZnT1)	SLC30A4 (ZnT4)	SLC30A7 (ZnT7)	SLC30A10 (ZnT10)
SLC30A2 (ZnT2)	SLC30A5 (ZnT5)	SLC30A8 (ZnT8)	TMEM163 (SLC30A11)
SLC30A3 (ZnT3)	SLC30A6 (ZnT6)	SLC30A9 (ZnT9)

Molecular aspects, physiological roles and links to disease

Among its many effects, Zn²⁺ accelerates cell proliferation, regulates wound healing, affects multiple aspects of the immune system, plays an essential role in epithelial physiology, affects the gastrointestinal system, is essential for spermatogenesis, and is involved in secretory organs, including the pancreas, salivary glands, and prostate. Zn²⁺ is also involved in protein folding, DNA and RNA synthesis, and the function of many enzymes. In the mammary gland, Zn²⁺ accumulation in the maternal milk is essential to support infant growth during the neonatal period. ZnT transporters contribute to the cytoplasmic zinc balance by exporting zinc to the extracellular space or by sequestering cytoplasmic zinc in intracellular compartments when cellular zinc levels are elevated. In contrast, metal ion transporters of the ZIP (SLC39) family function to increase cytoplasmic zinc concentrations when cellular zinc is depleted.

The SLC30 family belongs to the Cation Diffusion Facilitator (CDF) family (TC 2.A.4) which is a member of the CDF superfamily. As members of the CDF family, ZnTs contain a transmembrane core structure consisting of 6 TMHs (except ZnT5).

Structural information on mammalian ZnT transporters has long been lacking, and therefore the crystal structure of the bacterial homologue YiiP has been studied in great detail (232, 236, 237, 2311, 2312). This gap was successfully filled with the reports of the first high-resolution structures of human SLC30 zinc transporters, ZnT8 (SLC30A8) (15) and ZnT7 (SLC30A7) (233).

An integrative analysis of the cryo-EM structure and site-specific mutagenesis of human SLC30A1 has subsequently been performed, identifying a zinc transport mechanism unique to SLC30A1 within the SLC30 family (2313). Furthermore, the study shows that human SLC30A1 forms a homodimer with four zinc-binding sites and identifies H43 as crucial for zinc selectivity.

Based on the phylogenetic tree the SLC30 family can be divided into the following subfamilies (2314) (Fig. 40):

Fig. 40. Phylogenetic tree of the SLC30 family.

• Subfamily A: ZnT1 (SLC30A1) and ZnT10 (SLC30A10) - Plasma membrane targeted

• Subfamily B: ZnT2 (SLC30A2), ZnT3 (SLC30A3), ZnT4 (SLC30A4), and ZnT8 (SLC30A8) - Zn ²⁺ /H ⁺ antiporters, sequestering zinc in acidic compartments and vesicles

• Subfamily C: ZnT5 (SLC30A5), ZnT6 (SLC30A6), ZnT7 (SLC30A7) - Heterodimers and homodimers that play a critical role in the activation of zinc ectoenzymes in the ER/Golgi

• Subfamily D: ZnT9 (SLC30A9), TMEM163 (SLC30A11) - Mitochondrial zinc export

Description of the individual members of each subfamily:

Subfamily A: ZnT1 (SLC30A1) and ZnT10 (SLC30A10) - Plasma membrane targeted

Members of this subfamily are the only plasma membrane targeted members of the SLC30 family. They have been shown to be divalent metal ion/Ca²⁺ exchangers (2315, 2316). While ZnT1 functions as a Zn²⁺/Ca²⁺ exchanger, ZnT10 turned out to be a manganese transporter rather than a zinc transporter, as it has a distinct metal binding site. Specifically, the His-Asp-His-Asp motif of the Zn²⁺ metal binding site, which is conserved in most members of the ZnT family, is altered in ZnT10 by substitution of Asn for His (2315). ZnT1 is widely expressed, whereas ZnT10 expression is restricted to liver, brain and retina.

SLC30A1: ZnT1 (SLC30A1) is a widely expressed metal ion transporter that functions primarily as a plasma membrane Zn²⁺ exporter. It thus contributes to cellular zinc homeostasis by lowering cytosolic Zn²⁺ concentrations to protect cells from zinc toxicity (2317, 2318). Interestingly, ZnT1 has been shown to function as a Zn²⁺/Ca²⁺ exchanger in neuronal cells (2315) and as a Zn²⁺/H⁺ exchanger in HEK293 kidney cells (2319).

As mentioned above under “Discovery”, ZnT1 (SLC30A1) was cloned in 1995 as the predominant surface-expressed Zn²⁺ exporter in synaptic neurons and glia (2309). In the CNS, Zn²⁺ is an essential micronutrient and impaired regulation of Zn²⁺ levels affects cognitive function (2320). ZnT1 function has been associated with reduction of Zn²⁺ neurotoxicity under neuropathological conditions of prolonged exposure to Zn²⁺ (2315). In neurons, ZnT1 has been reported to modulate NMDA receptor function and neuronal signaling through its function as a Zn²⁺/Ca²⁺ exchanger.

ZnT1 also has an essential function in the transport of maternal zinc into the embryonic environment during the oocyte cylinder stage of development, and the absence of Slc30a1 has been shown to be embryonic lethal for this reason (2321).

In enterocytes, ZnT1 is located on the basolateral membrane (2322). After dietary zinc is absorbed across the apical membrane via ZIP4 (SLC39A4) and ZIP8 (SLC39A8) (Fig. 22), ZnT1 exports zinc across the basolateral membrane into circulation to maintain systemic zinc homeostasis (2317, 2323, 2324).

In the heart, ZnT1 has been shown to interact with the β-subunit of the L-type calcium channel in cardiomyocytes, resulting in a decrease in surface expression of the pore-forming α₁-subunit of the channel after upregulation of ZnT1 expression (2325). Zinc has a protective role in cardiovascular diseases, e.g., in the redox signaling pathway, lack of zinc under oxidative stress leads to the degradation of critical proteins (2326) and it plays an important role in excitation-contraction coupling (2327). In brief, during atrial fibrillation, cardiac myocytes exhibit increased electrical activity and significant cellular calcium overload, and the reduction in cell surface expression of the pore-forming α₁ subunit of the L-type calcium channel in fibrillating atria likely contributes to the self-perpetuating nature of the arrhythmia (2328). Rapid pacing of cultured cardiomyocytes has been shown to increase ZnT1 expression and it is proposed that ZnT1 binds the β-subunits of L-type calcium channels to sequester them, rendering them unavailable to chaperone the α₁-subunit to the plasma membrane as a protective mechanism (2325).

ZnT transporters such as ZnT1 (SLC30A1) also play important roles in the pancreas, thyroid, and adrenal glands (2329). In addition, somatic mutations in SLC30A1 have been shown to be the cause of excessive aldosterone production in aldosterone-producing adenomas in the adrenal cortex, causing primary aldosteronism (2330). In the healthy state, aldosterone-producing cells of the zona glomerulosa of the adrenal cortex express both voltage-gated calcium channels (2330–2332) and SLC30A1, and it has been shown that ZnT1 (SLC30A1) downregulates the calcium channel activity. In contrast, the somatic mutation of SLC30A1 causes an abnormal Na⁺ influx. The latter could be inhibited by the Na⁺ channel inhibitor ethylisopropyl amiloride (EIPA). The effect of EIPA could be due to direct inhibition of mutant ZnT1 or indirectly via effects on intracellular pH homeostasis, as EIPA also inhibits NHEs. Aberrant Na⁺ influx then leads to cell depolarization, which in turn activates the voltage-gated calcium channel, stimulating calcium influx, which serves as a signal leading to increased expression of nuclear receptor NR4A2, followed by expression of CYP11B2. The latter encodes aldosterone synthase, thereby catalyzing the renin-independent production of aldosterone in the adenomas (2330).

ZnT1 expression is tightly regulated in response to changes in intracellular levels of Zn²⁺, both at the transcriptional and post-translational levels (2333). In terms of transcriptional regulation, cytosolic Zn²⁺ is a potent transcriptional inducer of the expression of the zinc finger-containing transcription factor metallothionein 1 (encoded in human by MT1A) (2334). After binding Zn²⁺ in the cytosol, metallothionein 1 translocates to the nucleus where it binds to metal response elements in the promoter regions of genes to be activated, including SLC30A1 (2335). Thus, the induction of the expression of metallothioneins and ZnT1 is a mechanism that acts against elevated zinc levels.

In terms of post-translational regulation, under zinc-sufficient conditions, ZnT1 accumulates at the plasma membrane, consistent with its zinc efflux function. In contrast, under zinc-deficient conditions, ZnT1 molecules at the plasma membrane are endocytosed and degraded by both proteasomal and lysosomal pathways (2333). Thus, metallothionein-induced transcriptional regulation and posttranslational regulation of ZnT1 cooperatively regulate cellular zinc homeostasis.

The proposed function of ZnT1 in neurons, involving Ca²⁺ uptake in exchange for Mn²⁺, may be reminiscent of the divalent cation selectivity of Orai Ca²⁺-release-activated Ca²⁺ (CRAC) channels (2315), although these channels have a much higher permeability for Ca²⁺ than for Mn²⁺ (2336, 2337). However, phylogenetic analysis does suggest that CRAC channels and SLC30 transporters share a common origin within the CDF family, as mentioned in Section 8 (2338).

Another study reveals that SLC30A1 plays a protective role against Salmonella infection in macrophages, highlighting its importance in innate immune responses (2339). When Salmonella invades the body through contaminated food and water, they are rapidly engulfed and killed by resident macrophages. Studies in mice have shown that Slc30a1 expression in macrophages is upregulated during Salmonella infection and contributes to controlling the infection. Through its expression in the plasma membrane as a zinc exporter, this results in a short-term decrease in the concentration of zinc in the cytosol of macrophages, and through its expression in the membrane of Salmonella-containing phagosomes, it causes zinc toxicity in the phagosomes. Furthermore, studies in transgenic mice showed that loss of Slc30a1 leads to an accumulation of intracellular zinc. This upregulates metallothionein 1 expression and reduces iNOS and nitric oxide production via reduced NF-κB signaling. Consequently, the bacterial clearance capacity of the cell is reduced (2339). Experiments using human monocyte-derived macrophages revealed that human SLC30A1 is similarly upregulated, thus confirming its contribution to protection against Salmonella infection in humans. As the investigators of this study highlight, SLC30A1 expression is induced by a wide range of pathogens, including bacteria, fungi, and viruses (2339).

SLC30A10: ZnT10 (SLC30A10) is an apical membrane protein that functions as a manganese (Mn²⁺) exporter (2316). It exports manganese from hepatocytes into bile and from enterocytes into the gastrointestinal tract (2340). ZnT10 works closely with the Mn²⁺ transporter ZIP14 (SLC39A14) to regulate manganese homeostasis (Fig. 22).

Manganese is an essential trace nutrient that acts as a cofactor in many enzymatic reactions, including those related to protein glycosylation and antioxidant defense (2341). However, excess manganese, such as from environmental exposure through contaminated drinking water, is toxic, leading to oxidative stress and mitochondrial dysfunction, among other issues. Chronic exposure primarily affects the central nervous system, leading to neuropsychiatric disturbances and Parkinson disease-like motor dysfunctions (2342).

Evidence that SLC30A10 (ZnT10) localizes to the apical membrane of hepatocytes and enterocytes, facilitating the export of manganese (Mn²⁺) into bile and the intestinal lumen, comes from a study in which tissue-specific knockout mouse models were used to investigate the role of ZnT10 (Slc30a10) in manganese homeostasis (2340). The study revealed that mice lacking ZnT10 in the liver exhibit impaired biliary manganese excretion, whereas mice lacking the transporter in the intestine demonstrate reduced export of manganese into the intestinal lumen. These results highlight the essential role of SLC30A10 (ZnT10) in maintaining manganese balance by regulating its excretion through hepatic and intestinal pathways.

The transport mechanism and Mn²⁺ specificity of ZnT10 have been investigated using heterologous expression systems (2316). The experiments revealed that ZnT10 mediates Mn²⁺ efflux via a Ca²⁺-coupled exchange mechanism. Importantly, the study shows that ZnT10 does not facilitate Zn²⁺ transport under the tested conditions. Based on the findings, it was concluded that the physiological role of ZnT10 is specific to manganese homeostasis and that it does not significantly contribute to zinc transport in vivo (2316). ZnT10 shares similar cation selectivity with ZnT1. Additionally, pH has been shown to strongly regulate Mn²⁺ transport by ZnT10.

Inherited causes of Mn²⁺ excess are rare, but the study of these diseases has greatly advanced our molecular understanding of Mn²⁺ homeostasis in the human body. Mutations in the SLC30A10 gene cause manganese excess, which leads to a syndrome known as hypermanganesemia with dystonia 1 (see below) (2343–2345).

Mutations in the SLC39A14 divalent metal ion transporter gene (see the SLC39 family description) lead to a similar hypermanganesemia syndrome. This syndrome causes childhood-onset parkinsonism-dystonia and is distinguished from ZnT10 (SLC30A10) deficiency by the absence of liver involvement and polycythemia (2346). Conversely, mutations in the SLC39A8 gene, which encodes the divalent metal ion uptake transporter ZIP8, lead to decreased blood Mn²⁺ levels. This results in diminished activity of Mn²⁺-dependent enzymes, such as β-1,4-galactosyltransferase and Mn superoxide dismutase, leading to dysglycosylation (2347, 2348). This condition is referred to as congenital disorder of glycosylation type IIb (CDG2b). Affected individuals present with developmental delay, short stature, dwarfism, seizures, hypotonia, and dystonia as early as infancy.

As mentioned above, genetic defects in the SLC30A10 gene lead to a syndrome called hypermanganesemia with dystonia 1. This condition is characterized by excessive accumulation of manganese in the body, which leads to neurological dysfunction, liver cirrhosis, and polycythemia. Neurologic and liver disease are attributed to manganese toxicity. Hepatic hypoxia-inducible factor HIF2α (encoded by the EPAS1 gene) is a key signaling determinant of the outcome of manganese excess in liver in SLC30A10 deficiency and explains the cause of polycythemia, which is due to erythropoietin excess (2349): In case of systemic excess of Mn²⁺ due to mutations in SLC30A10, Mn²⁺ is imported into the liver by SLC39A14. Liver Mn²⁺ excess leads to increased HIF2α-dependent erythropoietin (EPO) expression. EPO excess results in polycythemia in the bone marrow and suppresses liver expression of hepcidin (the peptide hormone that plays a crucial role in iron homeostasis; see the SLC11 family description), leading to increased dietary iron absorption via the brush border divalent metal ion transporter DMT1 (SLC11A2) and the basolateral iron transporter ferroportin (SLC40A1) (see the description of SLC40A1), with the majority of excess iron consumed by erythropoiesis (2349).

In addition to liver, HIF signaling has also been shown to be activated in the brain (i.e., basal ganglia) when brain Mn²⁺ levels increase, and that HIF activation drives transcriptomic changes in the basal ganglia to protect against manganese neurotoxicity. Note that HIFs are heterodimeric transcription factors formed by the dimerization of a labile α-subunit with a common and stable β-subunit to encounter insufficient oxygen availability during hypoxia (2350). Under normoxic conditions, the α-subunit undergoes prolyl hydroxylation, which targets it for degradation. The two HIFα paralogs HIF1α (HIF1A) and HIF2α (EPAS1) activate transcription after dimerization with the β-subunit. Mn²⁺ activates HIF signaling by inhibiting prolyl hydroxylation of the α-subunit, and HIF1α and HIF2α are redundant in controlling Mn²⁺ homeostasis in cells (2351). Whether HIF1α or HIF2α is responsible for protection against manganese-induced neurotoxicity has not been shown (2349).

It has also been reported that ZnT10 regulates Mn²⁺ levels directly in specific regions of the brain by transporting Mn²⁺ out of brain parenchymal cells, and that especially during periods of early postnatal life, ZnT10 protects against lasting deficits in neuromotor function and dopaminergic neurotransmission. A deficit in dopamine release may be a likely cause of early-life Mn²⁺-induced motor disease (2352). The role of ZnT10 in adulthood was proposed to be to lower brain Mn²⁺ levels when the body burden of Mn²⁺ increases and Mn²⁺ excretion capacity is overwhelmed.

The HPA also suggests expression of ZnT10 in spermatids where Zn²⁺ is an essential trace element for spermatogenesis and spermatid differentiation (2353–2355).

Subfamily B: ZnT2 (SLC30A2), ZnT3 (SLC30A3), ZnT4 (SLC30A4), and ZnT8 (SLC30A8) - Zn²⁺/H⁺ antiporters, sequestering zinc in acidic compartments and vesicles

These are Zn²⁺/H⁺ antiporters sequestering zinc in acidic compartments, including vesicles such as endosomes/lysosomes, synaptic vesicles and insulin granules (2314).

SLC30A2: ZnT2 (SLC30A2) is abundantly expressed in tissues with high Zn²⁺ requirements such as the mammary and prostate glands (2356). ZnT2 is expressed in the secreting mammary epithelium where it is required to ensure adequate Zn²⁺ levels in breast milk. Loss-of-function mutations in SLC30A2 result in impaired zinc secretion into breast milk, causing transient neonatal zinc deficiency in exclusively breastfed infants (2357). Human ZnT2 expression is highly upregulated in the lactating mammary gland via stimulation of its gene expression by prolactin.

Loss-of-function mutants of SLC30A2 have also been associated with intestinal dysbiosis and alterations in intestinal gene expression in preterm infants (2358). Altered microbial environment may result from defects in Paneth cell function and/or granule activity. Paneth cells are the secretory cells located in the crypts of Lieberkühn, and the granules of these cells contain antimicrobial proteins such as lysozyme, α-defensins, and phospholipase. Lysozyme inhibits bacterial growth by attacking and hydrolyzing glycosidic linkages in bacterial cell wall peptidoglycans. Loss of ZnT2 function reduces lysozyme function and antibacterial activity as shown in Slc30a2 null mice.

SLC30A3: ZnT3 (SLC30A3) serves as a critical transporter of zinc into synaptic vesicles of a subset of glutamatergic zinc-enriched neurons (2359, 2360). ZnT3 (SLC30A3) is responsible for moving Zn²⁺ into the synaptic vesicles of these neurons. Zn²⁺, released together with glutamate at the synaptic cleft, inhibits NMDA receptors and thus the excitability of the hippocampal neuronal circuit. A rare variant of SLC30A3 leads to febrile seizures due to loss of function and reduced synaptic Zn²⁺ release, resulting in increased susceptibility to neuronal excitability and seizures (2361). ZnT3 expression is thought to contribute to the prevention of age-related cognitive loss because ZnT3 expression levels decrease with age (2362) and in patients with Alzheimer disease or Parkinson disease (2363, 2364).

Slc30a3 knockout in mice decreased zinc levels in the hippocampus and cortex, which was associated with progressive cognitive impairment (2365). ZnT3 deficiency and the resulting reduced brain zinc levels decreased the density of mature dendritic spines. The deficiency may lead to impaired glucose metabolism, which may ultimately be the cause of cognitive impairment (2365).

ZnT3 has been shown to colocalize with insulin in the rat insulinoma cell line INS-1 (2366), but appears to be absent in β-cells from mouse pancreatic islets (2367).

According to the HPA, ZnT3 is by far most highly expressed in spermatids. There it may play a crucial role in intracellular homeostasis of Zn²⁺, which is known to be essential for optimal spermatogenesis and sperm motility (2353, 2368, 2369).

SLC30A4: Mouse Slc30a4 (Znt4) was first isolated as the gene responsible for inherited zinc deficiency in the lethal milk mouse (2370). It has been shown to be critical for the zinc secretory function of the mammary gland, moving Zn²⁺ into the trans-Golgi apparatus for lactose synthesis and across the apical cell membrane for efflux from mouse mammary epithelial cells into milk. Once inside the trans-Golgi network, zinc is made available to zinc-dependent enzymes such as galactosyltransferase, the catalytic component of the lactose synthetase system, and carbonic anhydrase VI. “Lethal milk” mice exhibit low milk Zn²⁺ concentrations, smaller mammary glands, decreased milk volume, and lactation failure (2371, 2372).

ZnT4 (SLC30A4) is not expressed at significant levels in the mammary gland, but is almost exclusively expressed in the prostate, as suggested by the HPA. The first study to demonstrate the expression of ZnT4 in a human tissue is part of a microarray search for novel genes of potential clinical and biological significance in prostate cancer. The study demonstrated the expression of ZnT4 in the human prostate and that expression decreased during the progression from early prostate cancer to invasive prostate cancer (2373). The function of ZnT4 in the prostate is unknown, but it is known that normal and hyperplastic prostate tissues accumulate the highest levels of zinc of any soft tissue in the body (see below, Clinical and Pharmaceutical Aspects) and that zinc levels are markedly reduced in prostate cancer. Whether ZnT4 functions as a specialized plasma membrane zinc uptake transporter in the human prostate or in intracellular compartments remains to be determined.

SLC30A8: ZnT8 (SLC30A8) is a Zn²⁺/H⁺ exchanger selectively expressed in pancreatic β-cells, where Zn²⁺ transporters play an important role.

ZnT8 is expressed almost exclusively in β cells of pancreatic islets, where it is involved in insulin secretion in the insulin secretory granules (2374). In β-cells, the highest concentration of Zn²⁺ is in the insulin secretory granules. ZnT8 transports zinc into these granules, which is critical for proper insulin crystallization and contributes to optimal packaging efficiency of the insulin stored in the granules (2375). While protons are pumped into the granules by the V-ATPase containing the α3 subunit (encoded by TCIRG1), ZnT8 facilitates the transport of Zn²⁺ ions into the lumen of the secretory granules in exchange for H⁺ (2376). From there, Zn²⁺ is cosecreted with insulin. Studies of β-cell-specific Slc30a8 knockout mice showed a marked reduction in β-cell zinc content (2377).

In T1D, a chronic autoimmune disease, lymphocytes attack pancreatic islets and destroy insulin-producing β-cells. ZnT8 is a major autoantigen on β-cells and offers potential protection from autoimmune attack (2378, 2379). A monoclonal antibody against the cell surface antigen ZnT8 was shown to mask β-cell antigen exposure and suppress the immunologic cascade, providing a novel islet-targeted immunotherapy to prevent and reverse clinical T1D (2378).

Polymorphisms in SLC30A8 have been shown to reduce the risk of T2D in humans (2380–2384). Specifically, loss-of-function variation in SLC30A8 has been associated with lower glucose levels and a 65% reduction in T2D risk, resulting from enhanced insulin responsiveness to glucose and increased pro-insulin processing (2381, 2382). However, the precise mechanism linking these mutations to protection against T2D risk is still unclear. It may be related to increased insulin secretion, insulin clearance deficiency in the liver, reduced glucagon secretion, greater sensitivity to reactive oxygen species or involvement of nearby genes at the SLC30A8 locus (2380, 2381, 2383, 2384). Nevertheless, there is consistency in clinical data, with loss-of-function mutations of SLC30A8 being associated with increased insulin responsiveness and a lower risk of developing T2D, and thus ZnT8 has emerged as an interesting target for antidiabetic therapies (2381, 2383).

To investigate the molecular basis of the Zn²⁺/H⁺ exchange mechanism of ZnT8, the cryo-EM structures of human ZnT8 were determined in both outward- and inward-facing conformations (15). ZnT8 forms a dimeric structure with four Zn²⁺ binding sites within each subunit (15) (Fig. 4 shows the right half of a subunit in the CDF structure). Comparison of the outward- and inward-facing structures revealed that the transmembrane domains of each subunit, which accommodate the Zn²⁺ substrate, undergo a major structural rearrangement, allowing alternate access to the primary Zn²⁺ site during the transport cycle. The studies provided structural insights into the Zn²⁺/H⁺ exchange mechanism.

Prior to or in parallel to this, a density map of ZnT8 was reported using negative stain electron microscopy and single particle image analysis, and structural models were generated based on the bacterial homologue of the Zn²⁺/H⁺ antiporter YiiP (2385).

A novel live-cell imaging assay has been developed to measure SLC30A8 activity, facilitating the screening of small molecules that modulate its function (2386). Upon further optimization, the assay is proposed for use in future drug discovery campaigns to identify SLC30A8 modulators, given its emerging recognition as a therapeutic target for diabetes.

Subfamily C: ZnT5 (SLC30A5), ZnT6 (SLC30A6), ZnT7 (SLC30A7) - Heterodimers and homodimers that play a critical role in the activation of zinc ectoenzymes in the ER/Golgi

ZnT5 and ZnT6 form heterodimers and ZnT7 homodimers. They play a critical role in the activation of zinc ectoenzymes, such as alkaline phosphatases, which must be metalated by zinc and activated in the early secretory pathway compartments before reaching their destination (2387). Their expression is widespread but shows relatively high levels in certain tissues (2314).

SLC30A5, SLC30A6 and SLC30A7: ZnT5 (SLC30A5), ZnT6 (SLC30A6) and ZnT7 (SLC30A7) are widely expressed and present in most cell types, providing zinc supply to the ER/Golgi lumen (2388–2390). ZnT6 (SLC30A6) alone cannot transport zinc as it forms a ZnT5-ZnT6 heterodimer, where ZnT5 is the active zinc transporter. ZnT6 is a putative auxiliary protomer in the heterodimer (2391), wherein ZnT5 recruits ZnT6 to the Golgi apparatus to form the heterodimeric complex (2387). In contrast to ZnT5 and ZnT6, ZnT7 forms a homodimer.

The ZnT5-6 heterodimer and the ZnT7 homodimer are both closely associated with activation of zinc-requiring ectoenzymes, which are membrane-bound secretory and organelle-resident enzymes (2392)). These enzymes become active in the ER/Golgi lumen by coordinating zinc at their active site during the early secretory process, which is important because zinc ectoenzymes are associated with cell fate and cellular activity (2389, 2390). Activation of zinc-requiring ectoenzymes requires control of zinc mobilization for which the ZnT5-6 and ZnT7 zinc transporters are responsible. One such example of an ectoenzyme is the zinc-dependent tissue-nonspecific alkaline phosphatase TNAP, which hydrolyzes phosphate groups from a wide variety of substrates, including hydrolysis of pyrophosphate to inorganic phosphate. The absence of ZnT5-6 or ZnT7 dimers exacerbates the unfolded protein response, likely due to an increase in misfolded apo-zinc ectoenzymes and a decrease in chaperone activity, which is essential to facilitate zinc metalation of nascent zinc ectoenzymes, as a number of chaperones function in quality control of the ER and secretory pathway, some of which are regulated by zinc, such as the pH-sensitive chaperone ERp44 (2393). Thus, in the absence of Zn²⁺ in the ER/Golgi complex, the resulting unfolded protein response disrupts the quality control mechanisms of the early secretory pathway.

ZnT5 has been reported to play an important role in mast cell-mediated delayed-type hypersensitivity reactions, as mice deficient in Slc30a5 have been shown to exhibit severe osteopenia and male-specific sudden death from bradyarrhythmia during their reproductive period (2394–2396). FceR?, the high-affinity receptor for IgE, serves as a key receptor on the surface of mast cells responsible for their activation, and Zn²⁺ is required for FcεRI-triggered translocation of PKC to the plasma membrane. ZnT5 (i.e., the ZnT5-6 heterodimer) has been proposed to act to deliver zinc to the zinc finger-like domains in PKC, followed by translocation of PKC to the plasma membrane. This in turn activates the NF-κB signaling pathway, leading to NF-κB-dependent cytokine production such as IL-6 and TNFα. The involvement of Zn²⁺ delivered by ZnT5 to the ER/Golgi in FcεRI-triggered activation of PKC with subsequent translocation of PKC to the plasma membrane is as follows: Within minutes of FcεRI stimulation of mast cells, the ER releases zinc as part of specific “zinc waves” consisting of transient and transcription-independent increases in cytosolic zinc. The release of zinc from the ER is likely mediated by ZIP7 (SLC39A7), which is predominantly located in the ER, as knockdown of this transporter prevented the zinc waves (2397) (see the description of SLC39A7). Thus, ZnT5 is required for the immune response and plays a selective role in the allergic response (2396). ZIP7 (SLC39A7) also plays an essential role in this process. In mice, loss of Slc39a7 function is embryonic lethal, and ZIP7 has been shown to be essential for B cell development in both mice and humans (2398), as detailed in the SLC39A7 section. These findings once again highlight the importance of the precise regulation of ER/Golgi and cytosolic Zn²⁺ levels via the zinc transporters ZnT5 and ZIP7 in immune responses.

A splice variant of ZnT5 (SLC30A5) called ZNT5B or hZTL1 (GenBank accession number AF439324) has been described, which lacks a long cytoplasmic N-terminus and is proposed to localize to the plasma membrane, unlike full-length ZnT5, which mainly localizes to the Golgi (2399–2401). Studies in Caco-2 cells have shown that ZNT5B localizes to the apical membrane, where it functions bidirectionally and is regulated by zinc through transcription and mRNA stability (2400, 2401). Therefore, after Zn²⁺ is transported across the basolateral membrane of enterocytes via ZIP5 (SLC39A5) during endogenous release of excessive Zn²⁺, this splice variant could enable Zn²⁺ to leave the enterocyte through the apical membrane and enter the intestinal lumen (2324, 2401, 2402) (see Fig. 22 and the SLC39A5 description). However, this role of ZNT5B has not been conclusively established in vivo, so further research is needed to demonstrate its expression in the apical membrane of enterocytes and how its zinc exit mechanism would function. Furthermore, unless ZNT5B functions as a Zn²⁺/H⁺ exchanger, it is unclear what would drive the exit of positively charged Zn²⁺ ions, given that the cytosolic free Zn²⁺ concentration is extremely low and the intestinal lumen may have high Zn²⁺ levels, particularly during dietary intake. Alternatively, Zn²⁺ taken up by ZIP5 (SLC39A5) at the basolateral membrane of enterocytes may be sequestered in metallothioneins (2403) and released when enterocytes slough off. In this case, an apical Zn²⁺ exit mechanism would not be necessary.

ZnT7 (SLC30A7) has an essential function in dietary zinc absorption and regulation of body adiposity based on studies with Slc30a7 knockout mice (2404). The study suggests that reduced zinc absorption in the gut results in reduced zinc accumulation in other organs of the body. Transport of Zn²⁺ to the Golgi apparatus in the absorptive enterocytes of the gastrointestinal tract has been proposed to be an essential step in the absorption of dietary zinc. ZnT7 deficiency is likely to result in a disruption of zinc sequestration in the Golgi apparatus of the enterocyte. However, how exactly this would affect the transepithelial uptake of Zn²⁺ is still unclear.

Furthermore, ZnT7 deficiency in adipocytes was reported to reduce lipid synthesis by inhibiting insulin-dependent Akt activation and glucose uptake (2405). As a result, ZnT7 inactivation reduced body weight gain and body fat accumulation in mice. Lack of ZnT7 in 3T3-L1 adipocytes had a negative impact on insulin sensitivity, glucose uptake and lipogenesis, and it is concluded that ZnT7 is an important regulator of lipid synthesis in adipocytes. However, how exactly ZnT7-mediated Zn²⁺ transport affects insulin-dependent Akt activation was not determined in this study. In this regard, it is interesting to note that the insulin-sensitizing effect of zinc that stimulates lipogenesis and glucose uptake in isolated adipocytes has previously been attributed to inhibition of the tyrosine phosphatase activity of protein tyrosine phosphatase 1B (PTP1B), which in its active form inhibits the insulin signaling pathway, including inhibition of insulin-dependent Akt activation and glucose uptake (2406). Elevated cytosolic Zn²⁺, possibly influenced by ZnT7, has been proposed to inactivate PTP1B by non-covalent binding to its cysteine residues, thereby abolishing inactivation of the insulin pathway (2406).

The cryo-EM structure of human ZnT7 reveals the mechanism of Zn²⁺ uptake into the Golgi apparatus (233). Analysis of the cryo-EM structure shows that ZnT7 exists as a dimer via tight interactions in both the cytosolic and transmembrane (TM) domains of two protomers, each containing a single Zn²⁺-binding site in its transmembrane domain. ZnT7 was found to adopt a “mushroom” dimeric architecture with the transmembrane domains (TMD) and cytosolic domains (CTDs) of two protomers tightly packed (Fig. 4 shows the right half of the protomer in CDF cation diffusion facilitator structures which corresponds to human ZnT8 (15)). The TMDs of each ZnT7 protomer contains six transmembrane helices (TM1-TM6) with both the amino and carboxyl termini exposed to the cytosol, forming a Zn²⁺ transport pathway. The TM helices are tightly bundled at the cytosolic half of the TMD in both protomers, whereas they are widely opened to encompass a deep cavity at the luminal half of the Golgi, forming a solvent-accessible passageway from the luminal side. A likely scenario for ZnT7-mediated Zn²⁺ transport from the cytosol to the Golgi lumen is proposed (233).

Subfamily D: ZnT9 (SLC30A9), TMEM163 (SLC30A11) - Mitochondrial zinc export

ZnT9 (SLC30A9) is a mitochondrial zinc exporter. It is highly expressed in human brain, skeletal muscle and kidney. TMEM163 is also highly expressed in the brain and to a lesser extent in other tissues, and its function is presumably also zinc transport.

SLC30A9:ZnT9 (SLC30A9) is a mitochondrial zinc exporter that functions as a Zn²⁺/H⁺ antiporter (2407, 2408). Mitochondrial zinc is a cofactor for several mitochondrial enzymes. Moreover, mitochondrial Ca²⁺-activated Mg²⁺-ATP carrier SLC25A25 (APC3) was identified as an important regulator of mitochondrial Zn²⁺ import, loss of which suppresses the accumulation of mitochondrial Zn²⁺ (2408). Thus, the results suggest that ZnT9 (SLC30A9), together with APC3 (SLC25A25) represent a pair of mitochondrial carriers that control mitochondrial Zn²⁺ import and export, respectively, to maintain proper mitochondrial Zn²⁺ levels (2408). Mitochondrial Zn²⁺ export via ZnT9 (SLC30A9) is considered important for maintaining healthy mitochondrial Zn²⁺ homeostasis. This transport direction may seem to contradict that of other members of the SLC30 family, many of which translocate zinc to acidic organelles such as lysosomes and the Golgi apparatus. However, the mitochondrial proton gradient works in the opposite direction, and thus the reversed mitochondrial proton gradient well explains the outward direction of ZnT9 transport.

SLC30A9 mutations affecting intracellular zinc homeostasis have been described to cause Birk-Landau-Perez syndrome (BILAPES), an autosomal recessive cerebro-renal syndrome associated with genetic defects in the SLC30A9 gene (2409), first reported in six individuals from a large Bedouin kindred (2410). The disease is characterized by both neuronal dysfunction (i.e., early-onset neurological deterioration, intellectual disability, hearing loss) and chronic kidney disease. It is believed to be caused by increased production of ROS leading to increased oxidative stress. In addition, based on slc39a9 knockout studies in Caenorhabditis elegans, mitochondria in neurons were swollen due to lack of zinc export and thus had lower distribution in axons and dendrites, which likely contributes to the neuronal dysfunction in BILAPES (2411).

Developmental studies in flies and mouse embryos confirmed the role of ZnT9 (SLC30A9) in controlling mitochondrial zinc homeostasis and highlighted its importance for early mammalian embryonic development, findings that may help to better understand the pathogenesis of BILAPES (2412).

TMEM163 (SLC30A11) – Orphan transporter: SLC30A11 (TMEM163) has been added to the SLC30 family, although it has low sequence identity to other members of the SLC30 family. But low sequence identity was also the case for SLC30A9, which actually resides on the same phylogenetic branch as TMEM163 (2413). TMEM163 was originally identified as a synaptic vesicle membrane protein and potential transporter (2414). It has the “cation efflux” Pfam family motif, which is characteristic of SLC30 members (2413). Initial experiments with PC12 cells transiently or stably expressing rodent TMEM163 (also called “synaptic vesicle 31” SV31) resulted in intracellular zinc accumulation, suggesting that it has some Zn²⁺ transport activity (2414, 2415). TMEM163 is highly expressed in the brain, especially in the cerebellum, but also in the pancreas, and at somewhat lower levels in lymphoid tissues, endocrine tissues, respiratory system, intestine and kidney, as suggested by the HPA. One aspect of controversy about TMEM163 zinc transport function is whether it acts as an influx or efflux transporter (2413). Subsequently, TMEM163 genetic variants were identified in two unrelated patients with hypomyelinating leukodystrophy (HLD), which is part of a group of rare genetic disorders characterized by a persistent myelin deficit. The phenotypic effects of these variants have been studied using functional assays on cultured cells and the zebrafish model, and it was suggested that TMEM163 plays a role in oligodendrocyte development (2416). However, the precise biological role of TMEM163 remains unclear and further work is needed to elucidate its transport function and subcellular localization. In addition, TMEM163 was found to be required for ATP-evoked currents in cerebellar granule cells and dorsal root ganglion neurons, including those that relay pain-related information from peripheral tissues. Thus, it has been suggested that TMEM163 is a critical element in the regulation of P2XRs and ATP-evoked behavior by modulating the potency with which ATP acts on P2XRs (2417).

Clinical and pharmaceutical aspects

In the epithelial cells of the prostate, zinc plays a special role in the production of citrate (2418, 2419). A major function of the prostate is high citrate production and secretion. This is achieved by zinc-induced inhibition of m-aconitase, thereby preventing oxidation of citrate via the Krebs cycle for ATP production and freeing it for secretion. Thus, normal prostate epithelial cells accumulate remarkably high levels of zinc via ZIP/SLC39-mediated uptake with limited efflux or sequestration of zinc via ZnT/SLC30 exporters. However, in malignant prostate cancer, metabolic changes are initiated to favor energy production. This includes a reduction in cellular zinc levels, which has been linked to dysregulation of ZnT/SLC30 and ZIP/SLC39 transporters, as an early event in prostate cancer development (2420–2422). Therefore, a potential strategy for the treatment of prostate cancer is the inhibition or activation of specific zinc transporters.

As mentioned above, a monoclonal antibody to cell surface ZnT8 masks β-cell antigen exposure and suppresses the immunological cascade in T1D, providing a novel islet-targeted immunotherapy (2378).

Orphan transporter family member (1)

TMEM163 (SLC30A11)

HGNC update

SLC30A11 is a new alias for TMEM163.

SLC31 Copper transporter family (1.A.56/Ctr/Ctr)

Discovery: The (Ctr) copper transporter family (SLC31), which is essential for many copper-dependent processes, including mitochondrial oxidative phosphorylation, free radical detoxification, pigmentation, neurotransmitter synthesis and iron metabolism was identified by genetic studies in yeast. This allowed the identification of the first proteins involved in high-affinity copper uptake at the plasma membrane of eukaryote cells, known as yCtr1 and yCtr3 (2423). Complementation of the defective copper uptake phenotype of a yeast ctr1/ctr3 double mutant allowed the isolation of Ctr homologues from human (2424).

Gene family members (2):
SLC31A1 (CTR1)	SLC31A2 (CTR2)

Molecular aspects, physiological roles and links to disease

The SLC31 family belongs to the Copper Transporter (Ctr) family (TC 1.A.56). The X-ray structures of the high-affinity copper transporter Ctr1 from Salmo salar have been reported in both Cu⁺-free and Cu⁺-bound states, revealing a homotrimeric Cu⁺-selective ion channel-like architecture (see below) (33).

Copper is an essential cofactor in many biological processes, including mitochondrial oxidative phosphorylation, neurotransmitter synthesis, pigment formation, peptide biogenesis, antioxidant defense, and iron metabolism. However, excess copper can be toxic or even lethal to cells, so there is an elaborate scheme to maintain appropriate copper levels that includes cellular uptake, export transporters, along with intracellular compartmentalization and buffering systems.

While the two copper transport-related genes, ATP7A and ATP7B, responsible for the human diseases Menkes syndrome and Wilson disease, respectively (2425), are involved in copper export, copper uptake in humans is mediated by CTR1 (SLC31A1) and CTR2 (SLC31A2) (2426). Thus, cellular copper homeostasis is primarily regulated by these transporters.

SLC31A1: CTR1 (SLC31A1) is a major, high-affinity copper uptake transporter in mammalian cells (Zhou, 1997, #592). It is a crucial protein for cellular copper uptake and is essential for maintaining copper homeostasis. Its function is therefore tightly regulated, and retromer-dependent recycling plays a role in responding to fluctuating extracellular copper levels (see below).

Copper exists in two oxidation states in living organisms, reduced Cu⁺ and oxidized Cu²⁺ (2427). CTR1 is thought to be highly selective for Cu⁺ over Cu²⁺ (33).

CTR1 is ubiquitously expressed, with the highest levels in the liver (2426), allowing hepatic copper acquisition followed by biliary excretion.

CTR1 has 3 TMHs, which is atypical for a membrane transporter. However, its function as a Cu⁺ transporter was confirmed by structural studies, which revealed a homotrimer giving rise to 9 TMs that form a Cu⁺-selective ion channel-like architecture (33). Functional studies of human CTR1 expressed in SF9 and HEK293 cells using either ⁶⁴Cu or ⁶⁷Cu radioisotope uptake studies revealed that CTR1 mediates saturable copper uptake with a K_m of approximately 3.5 μM (2428, 2429). The transporter was found to be copper specific, energy independent, and stimulated by extracellular acidic pH and high K⁺ concentrations. Taken together, the structural and functional data provide a detailed picture of how Cu⁺ is imported across cell membranes via CTR1.

The steps of cellular copper uptake, based on current knowledge, are as follows (2429):

• 1)Human CTR1 acquires Cu²⁺ from the blood, e.g., from albumin, which binds to Cu²⁺ and has been shown to transfer copper to the N-terminus of CTR1 in the presence of vitamin C. Cu²⁺ interacts with the extracellular amino-terminal region of CTR1, which contains two methionine-histidine clusters and adjacent aspartates. These regions distinctively bind Cu⁺ and Cu²⁺ prior to copper uptake.
• 2)The N-terminal region facilitates the reduction of Cu²⁺ to Cu⁺, possibly in conjunction with external reducing agents such as vitamin C or a STEAP copper reductase.
• 3)After reduction, uptake of Cu⁺ occurs down its electrochemical gradient.
• 4)Cu⁺ enters the cytosol with concurrent binding to metallochaperones such as ATOX1 (named Atx1 in yeast), which protect the cell from reactive oxygen species. Cu⁺ then undergoes exchange reactions between cysteine residues of ATOX1 and those of the trans-Golgi network copper transporters ATP7A or ATP7B (2430, 2431).

CTR1 function is regulated via retromer-dependent recycling and responds to fluctuating extracellular copper levels as follows:

• 1)Cu⁺ entering CTR1 from the extracellular side triggers endocytosis of CTR1 via its second N-terminal His-Met-Asp cluster, which regulates subsequent levels of copper uptake (2429). This step is important to cope with fluctuating extracellular copper levels.
• 2)After endocytosis, CTR1 delivery back to the cell surface is controlled by the retromer complex, a key endosomal protein sorting machinery that recognizes the specific cargo membrane proteins for recycling (2432–2434).
• 3)Retromer is tightly regulated in response to extracellular cues, including changes in copper concentration (2435). As mentioned above, upon copper exposure, CTR1 is endocytosed to prevent excessive copper uptake.
• 4)Internalized CTR1 localizes to retromer-positive endosomes, and in response to decreased extracellular copper, retromer controls the recycling of CTR1 back to the cell surface to maintain copper homeostasis (2432).

A rare autosomal recessive disease caused by mutations of the CTR1 gene has been described, which results in profound copper deficiency in the central nervous system, infantile seizures, and neurodegeneration (2436). Furthermore, inactivation of the Ctr1 gene by targeted mutagenesis of Ctr1 in mice resulted in embryonic lethality in homozygous mutant embryos and a deficiency of copper uptake in the brains of heterozygous animals (2437).

Clinical relevance and pharmaceutical aspect

CTR1 (SLC31A1) is the primary platinum uptake transporter in platinum-based cancer chemotherapy. The efficacy of platinum-based anticancer drugs has been shown to correlate with the expression of CTR1. In line with this, retromer-deficient cells have been shown to have reduced sensitivity to the platinum-based drug cisplatin (2432). For a brief overview of the SLC solute carriers that are involved in platinum-based cancer chemotherapy, see the description of SLC22A2/OCT2.

While CTR1 (SLC31A1) is essential for cellular copper uptake and maintaining copper homeostasis, increased SLC31A1 expression has been observed in cardiomyocytes of diabetic patients, leading to excess copper. This leads to functional damage in these cells and may be related to cuproptosis in diabetic cardiomyopathy (2386). The findings could open new directions in diabetes research and treatment (2438).

SLC31A2 - Orphan transporter: CTR2 (SLC31A2) is a low-affinity copper transporter closely related to yeast Ctr2. The expression of CTR2 is ubiquitous, in the plasma membrane, late endosomes and lysosomes (2426). In mammalian cells, studies have suggested a role for CTR2, analogous to yeast Ctr2, as a vacuolar copper exporter, mobilizing copper from the vacuolar lumen into the cytosol during periods of extracellular copper scarcity (2439). But transport activity has not yet been demonstrated for human CTR2.

CTR2 also plays a role in platinum resistance, possibly by facilitating platinum efflux (2440): Higher CTR2 levels correlated with platinum resistance in ovarian cancer cell lines (2441) and in a human epithelial cancer cell model (2442). Knockdown of CTR2 was found to be associated with increased platinum cellular accumulation and efficacy (2441). Since CTR2 is mainly expressed in lysosomes and late endosomal formations, regulation of endocytosis may be involved in cisplatin efflux by CTR2 (2443). To better understand the role of CTR2 in cisplatin resistance in human cancers, further studies are needed.

Orphan transporter family member (1)

SLC31A2 (CTR2)

SLC32 Vesicular inhibitory amino acid transporter family (2.A.18.5/Aa_trans/APC)

Discovery: Screening of genomic clones for their ability to rescue the phenotype of unc-47 gene of C. elegans that was implicated in the release of GABA in GABAergic neurons resulted in the isolation of the vesicular GABA transporter (2444). In addition, an independent search for transporter sequences in a region of the genome comprising unc-47 (2445) led to the identification of UNC-47 and of a rodent ortholog. The transporter encoded by SLC32A1 is known as VGAT (vesicular GABA transporter), or VIAAT (vesicular inhibitory amino acid transporter).

Gene family member (1)

SLC32A1 (VIAAT)

Molecular aspects, physiological roles and links to disease

The SLC32 family belongs to Amino Acid/Auxin Permease (AAAP) family (TC 2.A.18.5) which is part of the MFS superfamily.

SLC32A1: VIAAT/VGAT (SLC32A1) is a vesicular Cl^--cotransporter of GABA or glycine (2446). 2 Cl⁻ and one GABA are cotransported per transport cycle. VIAAT/VGAT is the only known protein that loads the inhibitory neurotransmitters GABA and glycine into synaptic vesicles (2447).

Slc32a1^-/- knockout in mice is embryonically lethal, whereas heterozygotes showed no abnormalities in behavioral assays (2448). However, since disruption of GABAergic neurotransmission is an established cause of epilepsy and neurodevelopmental disorders, impaired VIAAT/VGAT function is a plausible cause for a developmental and epileptic encephalopathy.

In line with this concept, rare genetic missense variants of SLC32A1 have been shown to lead to a novel genetic etiology in neurodevelopmental disorders with epilepsy (2447). Functional analyses demonstrated that these variants impair GABAergic neurotransmission due to impaired filling of synaptic vesicles with GABA, leading to early-onset epilepsy and developmental delays.

The antiepileptic drug vigabatrin inhibits VIAAT/VGAT, and thus shows affinity similar to that of GABA (2449).

According to the HPA, VIAAT/VGAT (SLC32A1) is highly expressed in bipolar cells of the retina. Indeed, the vesicular release of GABA from mammalian horizontal cells is known to play an essential role in horizontal cell synaptic transmission (2450). The localization of numerous synaptic vesicle proteins, including VIAAT/VGAT, to horizontal cell processes and endings is consistent with the idea that synaptic vesicles are present in horizontal cells and participate in calcium-triggered exocytosis (2450). VIAAT/VGAT-containing synaptic vesicles fuse with the plasma membrane in a depolarization- and calcium-dependent manner, which is characteristic of the exocytosis of neurotransmitters.

In conclusion, based on current knowledge, VIAAT/VGAT (SLC32A1) functions only in neurons and some neuroendocrine cells, where it packs GABA and glycine into synaptic vesicles for inhibitory neurotransmission. No other functions have been ascribed to this protein outside the nervous system. Yet, the HPA suggests high VIAAT/VGAT (SLC32A1) expression in dendritic cells. However, this remains speculative because there are currently no experimental reports of VIAAT/SLC32A1 being expressed in dendritic cells or participating in vesicular inhibitory signaling in immune cells.

Orphan transporter family members: N/A

SLC33 Acetyl-CoA transporter family (2.A.1.25/Acatn/MFS)

Discovery: The transporter was identified by expression cloning from human melanoma cells using a specially developed protocol (166). Briefly, based on the previous demonstration that a single α-2,8-sialyltransferase enzyme, ST8Sia I, can catalyze the formation of the precursor gangliosides GD3 and GT3, a recipient cell line was generated by transfection of ST8Sia I cDNA into COS-1 cells. Using this cell line, a cDNA encoding the novel acetyl-CoA transporter required for the expression of O-acetylated gangliosides was identified.

Gene family members (2):
SLC33A1 (AT-1/ACATN1)	SLC33A2 (MFSD3)

Molecular aspects, physiological roles and links to disease

The SLC33 family belongs to the Peptide/Acetyl-Coenzyme A/Drug Transporter (PAT) family (TC 2.A.1.25), which is part of the MFS superfamily. It has at least 10 putative TMHs based on hydropathy analysis (166), but AlphaFold structural predictions suggest that all 6+6 helices of the MFS fold are present.

Cryo-EM unveiled the structure of human SLC33A1 in complex with acetyl-CoA. SLC33A1 exhibits a typical fold characteristic of the MFS, adopting a cytoplasm-facing conformation (2451). Two lobes of the protein, TMH 1-6 (N domain) and TMH 7-12 (C domain), are oriented towards the cytoplasm between TMH 4–5 and TMH 10–11. The research highlighted a “rigid anchoring–flexible adaptation” binding mode, where the acetyl-CoA molecule is securely held within the central cavity of the transporter, ensuring efficient transport into the ER (2451).

Although numerous homologs of this family of proteins have been identified in lower organisms, only two members of the SLC33 family exist in mammals:

SLC33A1: ACATN (SLC33A1) is ubiquitously expressed and transports acetyl-CoA into the lumen of the endoplasmic reticulum/Golgi apparatus. There, acetyl-CoA serves as a substrate for acetyltransferases that modify a variety of molecules, including sialic acid residues of gangliosides and lysine residues of membrane proteins (2452). ACATN (SLC33A1) is essential for the viability of motor neurons and is associated with neurodegenerative diseases such as sporadic amyotrophic lateral sclerosis. The transporter has also been shown to act as a key metabolic regulator that maintains acetyl-CoA homeostasis by promoting metabolic crosstalk between different intracellular organelles (2453).

In terms of clinical implications, studies have demonstrated that lung adenocarcinomas with Kelch-like ECH-associated protein 1 (KEAP1) mutations depend on SLC33A1 (2454). KEAP1 is a negative regulator of the antioxidant transcription factor, nuclear factor erythroid 2-related factor 2 (NRF2). Loss of KEAP1 function leads to chronic NRF2 activation. Among other effects, this elevates the burden on the ER folding machinery, making the tumor more reliant on systems that maintain ER function, such as SLC33A1. These findings suggest that SLC33A1 could serve as a potential therapeutic target in specific cancer subtypes.

As mentioned above, cryo-EM structural insights clarify our understanding of the physiological functions of SLC33A1, specifically the interaction between the 3’-phosphorylated ADP group of acetyl-CoA and SLC33A1 (2451). This research also holds potential for developing drugs to treat various diseases, including neurodegenerative disorders, age-related conditions, and different types of cancer (2451).

SLC33A2 - Orphan transporter: SLC33A2 (MFSD3) is a ubiquitously expressed putative transporter involved in altered nutrient uptake (2455). In the central nervous system, it is expressed on the plasma membrane of neurons. After starvation of mice, Slc33a2 (Mfsd3) was specifically upregulated in the brainstem, the gate that transmits signals from peripheral circulating hormones to the rest of the brain, whereas after high fat consumption, Mfsd3 expression was extensively reduced. It remains to be determined whether SLC33A2 is also an acetyl-CoA transporter.

Orphan transporter family member (1)

SLC33A2 (MFSD3)

HGNC update

MFSD3 has been renamed as SLC33A2.

SLC34 Type II Na⁺-phosphate cotransporter family (2.A.58/Na_Pi_cotrans/unknown)

Discovery: The human and rat renal Na⁺/PO₄^3- cotransporter NaPi-IIa (SLC34A1) was the first member of the SLC34 family to be identified (146). This family is also referred to as “type II sodium-phosphate cotransporters”.

Gene family members (3):
SLC34A1 (NaPi-IIa)	SLC34A2 (NaPi-IIb)	SLC34A3 (NaPi-IIc)

Note: There are also the so-called type I phosphate transporters of the SLC17 family, which later turned out to be organic anion transporters (see the SLC17 family description above and the article on the molecular identification of this transporter (142)), as well as the ubiquitously expressed phosphate transporters of the type III/SLC20/PiT family (1739, 2456, 2457).

Molecular aspects, physiological roles and links to disease

The SLC34 family belongs to the Phosphate:Na⁺ Symporter (PNaS) family (TC 2.A.58). SLC34 proteins are a unique class of membrane transport proteins with no apparent homology to other solute carrier families and there is currently no 3D structure of mammalian SLC34 proteins or their bacterial homologues. While SLC34 transporters are still structural orphans, the AlphaFold structural model suggests 8 TMHs plus two intramembrane hairpins (see Section 8 and Fig. 2), consistent with previous predictions (2456). The three carboxyl-terminal residues of NaPi-IIa (SLC34A1) form a PDZ-binding domain that binds to the PDZ scaffold protein NHERF-1 (2458, 2459).

Phosphate plays an essential role in skeletal mineralization and its chronic deficiency leads to rickets and osteomalacia (1738). Bone is the main storage site for phosphate. Phosphate also plays an important role in the kidney as a titratable acid that can be secreted. Regulation of phosphate homeostasis is accomplished by a balance between intestinal absorption, renal excretion, and transport into and out of bone. This balance is controlled by the coordinated effects of 1,25-dihydroxyvitamin D, PTH and FGF23 on SLC34 family members expressed on the apical membranes of intestinal and renal epithelia (1738, 1739, 2456, 2457, 2460–2462).

Calcium and phosphate are both critical components of bone, forming the core crystalline material, hydroxyapatite. Calcium also plays a role in blood clotting, serves as a critical second messenger in countless cell signaling pathways, and is involved in nerve conduction. Phosphate, on the other hand, is a component of phospholipids, nucleotides, and the cellular energy molecule ATP. Plasma levels of both calcium and phosphate are tightly regulated by the aforementioned phospho-calciotropic hormones.

SLC34A2: NaPi-IIb (SLC34A2) facilitates intestinal absorption of Na⁺-coupled phosphate across the apical membrane. The exact mechanism of basolateral exit, however, is still unknown. Intestinal absorption of phosphate is stimulated by 1,25-dihydroxyvitamin D via upregulation of NaPi-IIb expression (2463). NaPi-IIb (SLC34A2) also exhibits prominent expression in the lung and biallelic inactivating mutations of SLC34A2 are found in patients with pulmonary alveolar microlithiasis, a lung disease characterized by the deposition of microcrystals (2464–2466) (see Fig. 33, bottom part). In contrast, no evidence of impaired systemic P_i homeostasis has been reported in these patients. Nevertheless, NaPi-IIb-mediated intestinal P_i absorption may be a target for pharmaceutical intervention in patients with chronic kidney disease and P_i overload (2465).

NaPi-IIb has also been reported to be involved in the reabsorption of P_i from primary rat hepatic bile and has been proposed to play an important role in the regulation of biliary P_i concentration (1650).

SLC34A1, SLC34A3: In the proximal tubule of the kidney, the phosphate transporters NaPi-IIa (SLC34A1) and NaPi-IIc (SLC34A3) are expressed at the apical membranes where they are responsible for the reabsorption of filtered phosphate. NaPi-IIa accounts for approximately 70-80% of total renal phosphate reabsorption. It is abundantly expressed in the early renal proximal tubule segments, especially S1, whereas NaPi-IIc (SLC34A3) expression gradually increases in the later S2 and S3 proximal tubule segments (2461, 2467, 2468). The mechanism of basolateral P_i efflux has been elusive, but recent data established that it is carried out by XPR1 (see the SLC53 family description).

Renal phosphate reabsorption via NaPi-IIa and NaPi-IIc is intricately regulated to maintain the body’s calcium and phosphate balance. An increase in serum phosphate triggers the release of PTH from the parathyroid gland due to decreased stimulation of the Ca-sensitive receptor (CaSR) because increased serum phosphate causes it to complex with serum calcium to form calcium phosphate, resulting in decreased stimulation of CaSR and increased release of PTH (2469). PTH then travels to the renal tubules where it binds to PTH1R receptors. It reduces urinary calcium excretion by stimulating the apical calcium channel TRPV5 in the distal convoluted tubules and connecting tubules, thereby increasing calcium reabsorption (2470, 2471). In addition, it reduces urinary phosphate excretion by inhibiting NaPi-IIa. Here, PTH-PTH1R interaction on the basolateral membrane of renal proximal tubule cells stimulates the PKA pathway, which phosphorylates the PDZ-containing protein NHERF1 (Na⁺/H⁺ exchange regulatory cofactor 1). This scaffold protein tethers NaPi-IIa in the apical membrane and phosphorylation releases it for receptor-mediated endocytosis through clathrin-coated pits and subsequent degradation (2472). PTH1R is also expressed in the apical membrane of the proximal tubule, whereupon PTH-binding NaPi-IIa is also internalized. However, apical PTH1R signaling is mediated by PKC. Importantly, both the apical and basolateral signaling pathways require PLC coupling and converge on the ERK1/2 MAP kinase pathway, which is also modulated by FGF23 (2472) (see below).

PTH signaling in the renal proximal tubule cells also increases the expression of the 25-hydroxyvitamin D 1-alpha-hydroxylase CYP27B1, to generate the active form of vitamin D, 1,25-dihydroxyvitamin D, which then acts on the intestine, kidney, and bone to stimulate calcium and phosphate absorption, reabsorption, and resorption, respectively (2473).

Fibroblast growth factor FGF23 is another important regulator of NaPi-IIa independent of PTH (2461, 2462, 2474). It is produced primarily in osteoblasts and osteocytes in response to elevated dietary phosphate, elevated serum phosphate, and 1,25-dihydroxyvitamin D and binds to parathyroid fibroblast growth factor receptors (FGFR) to suppress PTH release (2475). It acts on the proximal renal tubules by suppressing phosphate reabsorption through Klotho-dependent activation of ERK1/2 (extracellular signal-regulated kinase-1/2) and SGK1 (serum/glucocorticoid-regulated kinase-1), leading to phosphorylation of NHERF-1 and subsequent internalization and degradation of NaPi-IIa (2461, 2476). FGF23 also indirectly suppresses intestinal phosphate absorption by downregulating the production of 1,25-dihydroxyvitamin D (2474, 2477). Thus, FGF23 inhibits PTH release and then takes its place, helping to inhibit phosphate reabsorption in the kidney without the phosphate-releasing effect on bone.

Given the important role of NaPi-IIa in the renal reabsorption of inorganic phosphate (P_i), it is not too surprising that patients with biallelic inactivating mutations in SLC34A1 develop hypophosphatemia, hypercalcemia, hypercalciuria and nephrocalcinosis, and nephrolithiasis in early childhood (for a review see (2465)). In addition, a critical role for NaPi-IIa in intrauterine and postnatal renal phosphate handling has been demonstrated in many families with biallelic inactivating mutations in SLC34A1 (2465).

Mutations in both SLC34A1 and SLC34A3 cause renal phosphate wasting, nephrocalcinosis, and nephrolithiasis. However, the clinical presentations differ in that rickets is a primary symptom of SLC34A3 mutations, but not of SLC34A1 mutations (2478). A multicenter retrospective study subsequently analyzed data from 113 individuals with pathogenic variants in SLC34A1 and SLC34A3 (2479). The findings revealed that individuals with biallelic SLC34A1 mutations often present with symptoms such as polyuria, failure to thrive, vomiting, constipation, hypercalcemia, and nephrocalcinosis in infancy. In contrast, individuals with biallelic SLC34A3 mutations typically present during childhood or adulthood with symptoms like rickets/osteomalacia, hypophosphatemia, and, less frequently, nephrocalcinosis. The prevalence of kidney stones was comparable between the two groups. Moreover, adult biallelic SLC34A3 carriers had a sixfold increase in chronic kidney disease prevalence compared to the general population. Heterozygous carriers of mutations in either the SLC34A1 or SLC34A3 genes often exhibit variable phenotypes. For individuals with SLC34A1 or SLC34A3 mutations, phosphate treatment may help promote kidney phosphate loss and boost 1,25-dihydroxyvitamin D synthesis by increasing PTH production.

Orphan transporter family members: N/A

SLC35 Nucleoside-sugar transporter family (2.A.7/Nuc_sug_transp, UAA, TPT, SLC35F/NST)

Discovery: Nucleotide sugars are transported from the cytoplasm to the endoplasmic reticulum and the Golgi apparatus where they serve as substrates for glycosylation. The molecular identities of the first group of NST nucleoside-sugar transporters (SLC35 family) were revealed by complementation cloning using NST-deficient mutant cells. The yeast UDP-GlcNAc transporter was cloned using a mutant strain of Kluyveromyces lactis that lacked a functional UDP-GlcNAc transporter (2480).

Gene family members (31 + 1 pseudogene):
SLC35A1 (CST)	SLC35D2 (HFRC1)	SLC35F4
SLC35A2 (UGT)	SLC35D3 (FRCL1)	SLC35F5
SLC35A3 (NGT)	SLC35D4 (TMEM241)	SLC35F6 (C2orf18)
SLC35A4	SLC35E1	SLC35G1 (POST/TMEM20)
SLC35A5	SLC35E2A (pseudogene)	SLC35G2 (AMAC1/TMEM22)
SLC35B1	SLC35E2B	SLC35G3 (TMEM21A)
SLC35B2 (PAPST1)	SLC35E3	SLC35G4 (AMAC1L1)
SLC35B3 (PAPST2)	SLC35E4	SLC35G5 (AMAC/AMAC1L2)
SLC35B4 (YEA)	SLC35F1	SLC35G6 (AMAC1L3/TMEM21B)
SLC35C1 (FUCT1)	SLC35F2	SLC35H1 (OVCOV1)
SLC35D1 (UGTREL7)	SLC35F3

Molecular aspects, physiological roles and links to disease

Currently there are 32 nucleoside sugar transporters in the SLC35 family, but many have not yet been fully characterized (2481).

The SLC35 family belongs to drug/metabolite transporter (DMT) superfamily (TC 2.A.7) and its family members bear the Nucleoside-sugar transporter (NST) structural fold. Members of the DMT superfamily share a similar structure of 10 TMHs that exhibit an inverted repeat architecture with two 5-TMH bundles inserted into the membrane in opposite orientations (see Section 8).

The family is broadly divided into seven subfamilies (A through H) that represent different branches on the phylogenetic tree (2481). Members of subfamily G differ from those of the other subfamilies in that they encode drug and metabolite transporters (see below).

SLC35 family members transport a wide range of nucleotide sugars, including CMP-sialic acid, UDP-glucose, UDP-galactose, etc. These transporters transport nucleotide sugars pooled in the cytosol into the lumen of these organelles where they are utilized as sugar donors by glycosyltransferases for the synthesis of sugar chains of glycoproteins, glycolipids and polysaccharides. Many of them act as exchangers in the ER or Golgi, exchanging nucleoside monophosphates for the corresponding nucleoside diphosphate conjugated sugar. A number of SLC35 transporters have not yet been assigned specific substrates.

As the scientists identified the members of the SLC35 family, they asked for subgroups, indicated by capital letters, to create subfamilies. The grouping was done sequentially as the transporters were discovered. However, now that all SLC35 genes have been identified, this grouping has proven to be partially incorrect. Therefore, to best describe the SLC35 members, the family has been divided into branches 1 to 12, independent of its capitalized subgrouping (see Fig. 41).

Fig. 41. Phylogenetic tree of the SLC35 family.

Branch 1 (subfamily A) – Nucleotide sugar transporters

SLC35A1: The cytidine 5’-monophosphate (CMP)-sialic acid transporter CST (SLC35A1) is a classical nucleotide-sugar transporter localized to Golgi organelles that transports sialic acid conjugated to a nucleotide monophosphate. It catalyzes the exchange of cytoplasmic CMP-sialic acid for Golgi luminal CMP. SLC35A1 is the only mammalian CMP-sialic acid transporter (CST) identified to date (2482). The crystal structure of mouse Slc35a1 in complex with CMP-sialic acid indicates that SLC35A1 possesses 10 transmembrane helices with the N- and C-termini oriented towards the cytosol. Additionally, it suggests that TMD5 and TMD10 contribute to the formation of dimers, whereas other domains participate in creating transport bundles (2483, 2484). CRISPR-Cas9 knockout of SLC35A1 in HEK293T cells revealed the presence of an additional unknown Golgi CMP-sialic acid uptake transporter (2485). SLC35A1-CDG (congenital disorder of glycosylation) is a rare inherited disorder that mainly affects the vascular systems of the body (2486). The mutants were found to be deleterious for the function of the transporter and/or to impair the ability of CST variants to form dimers.

Studies with proteoliposomes suggested three possible transport modes for SLC35A1: 1) NS (nucleotide sugar) exchange for NDP (nucleotide diphosphate); 2) NS exchange for another NS; and 3) NS transport fin the absence of the antiported molecule (passive transport).

SLC35A2: The UDP-galactose transporter UGT (SLC35A2) is an X-linked transporter that carries uridine diphosphate (UDP)-galactose from the cytosol to the lumen of the Golgi apparatus and the endoplasmic reticulum (2487). Interestingly, two splice variants of SLC35A2 were identified encoding two proteins UGT1 and UGT2, which differ in 3 amino acids in the C-terminus. UGT1 is localized only in the Golgi apparatus, whereas the UGT2 C-terminus contains a dilysine motif that is responsible for dual localization in the Golgi and endoplasmic reticulum (2485, 2488). Since the SLC35A2 gene is located on the X-chromosome, most patients with SLC35A2-CDG are female. SLC35A2-CDG patients with this rare disease have profound neurological and developmental impairments, most with epilepsy and/or skeletal abnormalities.

SLC35A3: NGT (SLC35A3) is a uridine diphosphate (UDP-N-acetylglucosamine; UDP-GlcNAc) transporter in mammals that regulates the branching of N-glycans. Compound heterozygous mutations in SLC35A3 were identified in patients from three unrelated families with arthrogryposis, impaired intellectual development, and seizures and a homozygous missense mutation in SLC35A3 was identified in a patient with severe vertebral anomalies (2487). Moreover, mice lacking Slc35a3 exhibit lethal chondrodysplasia with vertebral anomalies and impaired glycosaminoglycan biosynthesis. It has been proposed that the function of UGT (SLC35A2) and NGT (SLC35A3) in galactosylation is coupled through heterologous complex formation in the Golgi membrane so that their function is combined via mutual interaction (2489).

SLC35A4 - Orphan transporter: SLC35A4 has been proposed to be an UDP-Gal transporter (2484) although others have suggested its role may not be in glycosylation but rather have a modulatory or regulatory role related to UGT (SLC35A2) and NGT (SLC35A3).

SLC35A5 - Orphan transporter: SLC35A5 is a putative UDP-sugar transporter. Inactivation of the SLC35A5 gene resulted in decreased Golgi uptake of uridine diphosphate (UDP)-glucuronic acid, UDP-N-acetylglucosamine, and UDP-N-acetylgalactosamine, with no effect on UDP-galactose transport (2490). Further studies showed that SLC35A5 localizes exclusively to the Golgi apparatus. The C-terminus of this protein is extremely acidic and contains distinctive motifs, namely DXEE, DXD, and DXXD (the C-terminus is directed towards the cytosol). SLC35A5 has also been shown to form homomers as well as heteromers with other members of the SLC35A protein subfamily. Thus, the SLC35A5 protein may be a Golgi-resident multiprotein complex member involved in nucleotide sugar transport (2490).

Branch 2 (part of subfamily F)

SLC35F6 - Orphan transporter: In an effort to identify novel molecular targets and develop molecular therapy for pancreas ductal adenocarcinoma (PDAC), genome-wide gene expression profiling of microdissected PDAC cells identified a novel gene, C2orf18 (SLC35F6), as a molecular target for PDAC treatment (2491). Its overexpression in PDAC cells and limited expression in normal adult organs was validated by transcriptional and immunohistochemical analysis. siRNA knockdown of C2orf18 (SLC35F6) in PDAC cell lines induced apoptosis and suppressed cancer cell growth, suggesting its essential role in maintaining PDAC cell viability. C2orf18 showed mitochondrial localization and interaction with adenine nucleotide translocase ANT2 (SLC25A5), which is involved in maintaining mitochondrial membrane potential and energy homeostasis, suggesting a role in apoptosis. The results suggest that SLC35F6, also known as ANT2-binding protein (ANT2BP), may be a candidate molecular target for pancreatic cancer treatment. By interacting with ANT2, SLC35F6 may play a role in glycolysis-related ATP production that renders pancreatic cancer cells resistant to hypoxia and chemotherapy. Some small molecule inhibitors of an ANT transporter are already established as anticancer drugs (2491). In addition, the analysis of lncRNA UCA1-related downstream pathways of cisplatin resistance in lung adenocarcinoma led to the identification of SLC35F6 among the top ten up-regulated mRNAs (2492). Thus, SLC35F6 may also contribute to cisplatin resistance in lung adenocarcinoma.

Branch 3 (part of subfamily F) – Orphan transporters

SLC35F1 - Orphan transporter: SLC35F1 was identified as a candidate gene for neurodevelopmental disorders resembling Rett syndrome (2493). SLC35F1 is mainly expressed in the brain and was thought to be a nucleotide sugar transporter (2481). However, subsequent studies showed that deletions in a chromosomal region containing regulatory sequences of SLC35F1 (6q22.1q22. 31) are associated with childhood epilepsy, and that mouse SLC35F1 colocalizes with RAB11, a protein fundamental to dendritic spine formation, mutations in which are associated with developmental and epileptic encephalopathy in humans, suggesting a neurodevelopmental or synaptic plasticity role for SLC35F1 (2494).

The subcellular localization of the SLC35F1 and SLC35F6 orphan transporters, overexpressed in HeLa cells or expressed endogenously in the neuroblastoma cell line SH-SY5Y, has been investigated (2495). Unlike other SLC35 family members, which typically localize to the ER/Golgi, SLC35F1 was found in recycling endosomes and SLC35F6 was found in lysosomes. The study identified specific sorting signals in the cytoplasmic tails of these transporters that are responsible for their distinct localizations (2495). The presence of these signals suggests that the transporters have specialized functions in endosomal and lysosomal pathways.

SLC35F2: A haploid genetic screen in human cells identified SLC35F2 as the absolute dependent of the clinically tested anticancer drug YM155 (2496). While it was previously described that the uptake of YM155 is transporter-mediated, the identity of this uptake transporter remained unclear (2497, 2498). YM155 (sepantronium bromide) is a small imidazolium-based proapoptotic agent with anti-tumor activity. SLC35F2 was found to be highly expressed in a variety of human cancers (2496). SLC35F2 expression has been shown to be required for YM155-mediated DNA damage, and it has been proposed that SLC35F2 is expressed on the cell membrane where it transports YM155 into the cell.

SLC35F2 was also found to be highly expressed in aggressive and invasive tumors and to promote bladder cancer progression. The results suggest new therapeutic opportunities in combination with the anticancer drug YM155 (2499). The ubiquitin-specific peptidase USP32 was found to confer cancer cell resistance to YM155 by promoting ER-associated degradation of SLC35F2 (2500), and inhibition of USP32 was proposed as an additional strategy to enhance chemotherapeutic efficacy by increasing SLC35F2 expression for maximal uptake of the anticancer drug YM155.

A later study provided clarification of the transport activity and revealed the physiological substrates of this transporter (2501). It was shown that SLC35F2 is a highly specific transporter for the micronutrients queuine and queuosine. Queuine and its nucleoside derivative queuosine are exclusively synthesized by bacteria and recovered by eukaryotic organisms, where queuine is subsequently incorporated into tRNAs encoding His, Tyr, Asp, and Asn codons. Since it is used by all eukaryotes, but is produced exclusively by bacteria, it is considered to be a putative vitamin. The study showed that YM155 competitively inhibits the uptake of these micronutrients and that YM155 itself is a transport substrate (2501). According to the HPA, SLC35F2 is ubiquitously expressed, particularly in the salivary gland, small intestine and prostate. It has been proposed that queuine is absorbed from the small intestine by SLC35F2 and delivered to the liver for cleavage by QNG1, which then serves to release the queuine base for subsequent release into the serum and further distribution throughout the body (2501).

Branch 4 (subfamily B) – Transporter of ATP, PAPS, etc

SLC35B1: SLC35B1 is an ATP/ADP exchanger in the ER membrane. It was initially reported to mediate counter-transport of UDP-glucuronic acid (UDP-GlcUA) in exchange for UDP-N-acetylglucosamine (UDP-GlcNAc) across the ER membrane (2502). It was later proposed to act as a UDP-galactose transporter-related protein (UGTrel1) supporting uptake of UDP-galactose for ER-localized glucuronidation reactions (2503).

Subsequent studies revealed a central role in ER energy metabolism, showing that SLC35B1 mediates ATP/ADP exchange and is therefore also known as AXER (ATP/ADP exchanger in the ER membrane) (2504). SLC35B1 mRNA is broadly expressed with highest levels in energy-demanding tissues such as skeletal muscle (2484, 2505), consistent with the need for robust ER ATP supply.

Subsequent studies have refined this view and demonstrated that SLC35B1 does not transport nucleotide sugars directly but instead exchanges ATP with di- and tri-nucleotides to fuel ER processes (2506). Most notably, a 2025 cryo-EM study revealed stepwise ATP translocation by human SLC35B1, providing a previously undescribed model for substrate translocation by an SLC transporter and offering a detailed mechanistic framework for ATP import into the ER lumen (2507). This mechanistic insight strengthens the link between SLC35B1 activity and ER protein folding, quality control, and energy homeostasis (2504, 2508).

The HPA shows ubiquitous SLC35B1 expression with particularly high levels in muscle, epididymis, mammary gland, kidney, intestine, liver, thyroid, parathyroid, and choroid plexus, suggesting its importance for ER function in diverse tissues.

SLC35B2: PAPST1 (SLC35B2) is a transporter of the nucleotide sulfate 3’-phosphoadenosine 5’-phosphosulfate (PAPS), a universal sulfuryl donor for sulfation modification of glycoproteins, glycolipids, and proteoglycans (2509). Sulfation involves the transfer of a sulfate group from the activated form of PAPS to a substrate, a reaction catalyzed by several sulfotransferase enzymes. PAPS is required as a sulfate donor for all sulfotransferase enzymes and is transported by PAPS transporters to the Golgi apparatus for sulfation. The PAPST1 (SLC35B2) transporter is ubiquitously expressed and a role for SLC35B2 in the regulation of total sulfation, including heparan sulfate and protein tyrosine sulfation, has been reported (2510).

SLC35B3: PAPST2 (SLC35B3) is another PAPS transporter that is ubiquitously expressed in addition to PAPST1 (SLC35B2) (2511). Similar to SLC35B2, the SLC35B3 protein also shows a Golgi localization. RNAi knockdown of SLC35B2 or SLC35B3 in colon cancer cells resulted in a decrease in the sulfation level of cellular proteins and a suppression of cell proliferation (2512). It has also been shown that SLC35B2 and SLC35B3 are involved in the maintenance and differentiation of mouse embryonic stem cells (2513). Whether SLC35B2 and SLC35B3 serve as redundant PAPS transporters to ensure proper sulfation modification or whether they have some distinct roles has yet to be determined.

SLC35B4: SLC35B4 has been reported to function as a nucleotide sugar transporter that transports UDP-N-acetylgalactosamine (UDP-GlcNAc) and UDP-xylose (UDP-Xyl) into the endoplasmic reticulum or Golgi apparatus for glycosylation (2514). UDP-GlcNAc is a sugar substrate required for O-GlcNAcylation of proteins. SLC35B4 is required for hepatocellular carcinoma (HCC) tumorigenesis by enabling the O-GlcNAc modification of c-Myc, which stabilizes c-Myc and drives HCC tumorigenesis (2515). Therefore, SLC35B4 is a promising therapeutic target for the treatment of HCC.

The function of SLC35B4 as a UDP-Xyl transporter is crucial for the biosynthesis of proteoglycans in mammals. UDP-Xyl is found in a linker (–GlcA–Gal–Gal–Xyl–Ser/Thr) that connects glycosaminoglycan chains to core proteins in proteoglycans (Maszczak-Seneczko, 2022, #1343).

SLC35B4 has two splice variants: a longer version encoding a protein of 331 amino acids and a shorter version encoding a protein of 231 amino acids. A C-terminal dilysine motif KDSKKN is crucial for ER localization (2485).

Studies have also revealed a link between SLC35B4 function and glucose homeostasis. Reduced hepatic Slc35b4 expression was found to increase hepatic gluconeogenesis (2516). On the other hand, SLC35B4 protein expression was increased by glucose stimulation and SLC35B4 was shown to act as an inhibitor of gluconeogenesis in response to glucose stimulation. It has been postulated that SLC35B4-mediated uptake of UDP-GlcNAc into the ER/Golgi reduces cytosolic O-linked glycosylation by depleting substrate for glycosyltransferases, leading to decreased gluconeogenesis (2517).

Branch 5 (part of subfamily F)

SLC35F3: SLC35F3 has been shown to function as a thiamine transporter (2518, 2519). It is abundantly expressed in neurons of the brain (cerebellum, cerebral cortex, basal ganglia, etc.), with subcellular expression in the Golgi and nucleoli.

Sequence homology to a putative yeast thiamine (vitamin B1) transporter led to the expression of human SLC35F3 in E. coli, which confirmed [³H]-thiamine uptake via SLC35F3 (2519). Consistent with this transport activity, homozygotes for the SLC35F3 risk allele had decreased erythrocyte thiamine content (2519). In addition, in twin pairs, the SLC35F3 risk allele predicted heritable cardiovascular traits previously associated with thiamine deficiency, including increased cardiac stroke volume with decreased vascular resistance and increased pressor responses to environmental stress, including cold stress (2519).

A genome-wide association study revealed the association between SLC35F3 and the risk of metabolic syndrome (MetS) (2520). SLC35F3 expression was found to be associated with MetS risk factors such as blood pressure, insulin, and visceral fat (2521, 2522). A risk allele of SLC35F3, rs10910387, was shown to increase the incidence of MetS (2523). Thiamine transport is associated with carbohydrate metabolism and risk of MetS because thiamine is a coenzyme of pyruvate dehydrogenase, the rate-limiting enzyme for pyruvate entry into the tricarboxylic acid cycle. There are four types of genetic defects in SLC19A2, SLC19A3, SLC25A19, and thiamine pyrophosphokinase (TPK1), which are involved in the transport and metabolism of thiamine (2518). SLC35F3 also likely influences MetS through its role in carbohydrate metabolism.

SLC35F4 - Orphan transporter: SLC35F4 is a paralog of SLC35F3 with 92% amino acid sequence similarity. It is proposed that SLC35F4 also functions as a thiamine transporter, although experimental validation is yet to be performed (2524). SLC35F4 has a similar tissue distribution as SLC35A3 according to the HPA, but with more specific expression in the cerebellum. It is also highly expressed in the retina.

SLC35F5 - Orphan transporter: SLC35F5 is a ubiquitously expressed plasma membrane protein. Its transport function is currently unknown. SLC35F5 has been shown to affect cellular chemosensitivity to nucleoside analogues, suggesting a potential role in nucleoside transport and metabolism (2525). A recent study in human myeloid leukemia cells suggests that SLC35F5 serves as a cell surface nucleoside transporter that affects purine levels and thereby chromatin states that are dependent on BRD4, an epigenetic reader that recognizes histone proteins and acts as a transcriptional regulator (2526). Loss of transporters thought to be involved in purine transport, such as SLC35F5, led to dysfunction of BRD4-dependent transcriptional regulation, consistent with the concept of a specific role of purine/adenine metabolism in modulating BRD4-dependent epigenetic states (2526).

Branch 6 (subfamily G)

Members of this branch are considered to be part of the drug/metabolite superfamily (see below). Some of them are likely to be plasma membrane transporters.

SLC35G1 - Orphan transporter: TMEM20, also referred to as POST (SLC35G1), was discovered as a membrane protein with unknown function in the search for a “binding partner of STIM1” (POST) (2527). The identified POST (TMEM20; SLC35G1) is located in the plasma membrane and in the endoplasmic reticulum. The calcium sensor STIM1 (stromal interaction molecule 1) is known to trigger cellular Ca²⁺ influx after depletion of ER Ca²⁺ stores by opening the plasma membrane store-operated Ca²⁺ channel Orai1, allowing calcium to enter the cytoplasm, followed by refilling of ER Ca²⁺ stores by the SERCA calcium pump. To fine-tune this process, POST/TMEM20 was found to bind STIM1, and the complex then moves within the ER to near the cell membrane surface, allowing STIM1 to bind to and inhibit the plasma membrane calcium exporter pump PMCA (2527). Thus, while POST/TMEM20 (SLC35G1) does not affect store-operated calcium entry, it reduces the activity of the plasma membrane Ca²⁺ efflux pump, resulting in increased intracellular Ca²⁺ levels. Additionally, studies have demonstrated the ability of microRNA-150 (miR-150) to suppress the expression of POST/TMEM20 (SLC35G1), thereby suppressing the increase in intracellular calcium levels (2528). This fine-tuning by miR-150 is important for the initiation of an immune response. It has been demonstrated that after loss of miR-150, naïve CD8⁺ T cells cannot be activated because miR-150 deficiency increases TMEM20 expression, which leads to increased intracellular Ca²⁺ levels in naïve CD8⁺ T cells and failure to respond to their specific antigen as this triggers the expression of anergy-inducing genes. Nevertheless, CD8⁺ T cell-specific suppression of miR-150 expression may be a novel approach for the treatment of autoimmune diseases (2528). But how exactly the SLC35G1/POST/TMEM20 transporter in conjunction with STIM1 can trigger the binding of STIM1 to calcium pumps and what the transport function of this transporter would have to be in this role remains a mystery.

SLC35G2 - Orphan transporter: TMEM22 (SLC35G2) was identified in an attempt to search for the expression of candidate TMEMs with predicted ER localization as potential classifiers of clear cell renal cell carcinoma (ccRCC) (2529). TMEM22 is predicted to localize in the endoplasmic reticulum, and it contains a domain similar to the E. coli multidrug resistance antiporter EmrE (2529). Specifically, it has been identified as a member of the EamA drug/metabolite transporter-like family with two copies of the EamA domain. Differential regulation of SLC35G2 expression in intrahepatic cholangiocarcinoma (2530), promoter hypermethylation of SLC35G2 in melanoma (2531), upregulation of SLC35G2 in Caki-1 and Caki-2 renal cell carcinoma cell lines and tumor samples (2529), decreased cell growth upon siRNA silencing of SLC35G2 in these cells indicate its involvement in cancer progression and development (2532).

SLC35G3 (AMAC1/TMEM21A), SLC35G4 (AMAC1L1), SLC35G5 (AMAC/AMAC1L2), SLC35G6 (AMAC1L3/TMEM21B) - Orphan Transporters: These are additional members of the drug/metabolite transporter superfamily, which includes a variety of protein domain families with multiple functions that were identified based on a phylogenetic analysis of nucleotide sugar transporters and drug/metabolite transporters using a combination strategy of Hidden Markov Models and maximum likelihood estimation (2533). Based on these results, it was proposed that the original annotation of SLC35G sequences as members of the AMAC (acyl-malonyl condensing enzyme) subfamily was likely incorrect (2533). Subsequently, organic cation uptake activity has been attributed to several SLC35G members, specifically choline and nicotine for SLC35G4 and SLC35G3, respectively, as quantified in stably transfected HEK293 cells using HPLC-MS/MS analysis (2534). Nevertheless, the physiological relevance of this finding warrants further investigation.

Branch 7 (subfamily H)

SLC35H1 (updated from SLC35C2) - Orphan transporter: OVCOV1 (SLC35H1) has been proposed to function as an additional Golgi transporter that transports GDP-fucose, similar to SLC35C1 (see branch 11), although experimental validation is lacking (2508). The gene was identified as a suppressor that inhibits the increased fucosylation phenotype in a gain-of-function LEC11B Chinese hamster ovary (CHO) mutant that upregulates α1,3-fucosyltransferase activity (2535). OVCOV1 (SLC35H1) is ubiquitously expressed and has been reported to be regulated by oxygen tension, suggesting a potential role in the cellular response to tissue hypoxia (2536) (reviewed in (2481)).

SLC35C2 has been reclassified into a new subfamily H as SLC35H1 because the encoded protein has a slightly higher percentage identity with SLC35E3 and SLC35E4 than SLC35C1, and a mouse knockout suggests that, unlike SLC35C1, it does not transport GDP-fucose (2537). It was also excluded that SLC35H1 is an alternative supplier of GDP-fucose for N-glycan fucosylation (2538).

Branch 8 (part of Subfamily E)

SLC35E3 - Orphan transporter: SLC35E3 was identified as a target of a novel microRNA via microRNA profiling of patients with cardiovascular disease (2539). According to the HPA, SLC35E3 mRNA is expressed at low levels with relatively low tissue specificity, while at the single cell level, mRNA expression is highest in monocytes. Protein expression could not be estimated.

Branch 9 (part of subfamily E)

SLC35E4 - Orphan transporter: Genome-wide transcriptome analysis of vaginal tissue samples from women of reproductive age undergoing gynecologic surgery revealed a subset of 10 highly expressed solute carriers, including SLC35E4 (2540). The FDA-approved antiretroviral drugs currently being evaluated for pre-exposure HIV chemoprophylaxis include the nucleoside reverse transcriptase inhibitors tenofovir and emtricitabine and the CCR5 antagonist maraviroc, which are administered as vaginal gels. Membrane transporters important for the disposition of these drugs were found to be consistently over- or under-expressed in all vaginal tissue samples. As a result of such individual expression differences, these drugs are likely to have different therapeutic efficacy. Nevertheless, whether SLC35E4 contributes to the uptake and disposition of these drugs remains to be shown.

Branch 10 (part of subfamily E)

SLC35E1 - Orphan transporter: SLC35E1 was identified in a screen for cellular proteins involved in herpes simplex virus 1 nuclear export (2541). However, a suitable substrate for transport has yet to be identified.

SLC35E2A - Orphan transporter, gene/pseudogene: The promoter region of SLC35E2A (previously known as SLC35E2) contains mutation hotspots, which are significantly associated with worse prognosis in patients with esophageal squamous cell carcinoma (ESCC). However, the biological function and the transport substrate of SLC35E2A are still unclear (2542). Genome annotators at Ensembl and NCBI Gene classify SLC35E2A as a pseudogene.

SLC35E2B - Orphan transporter: SLC35E2B was identified by exon sequencing in a search for genetic mutations that cause high myopia, an eye disease with environmental and genetic factors. The SLC35E2B protein is ubiquitously expressed, including in the retina, but its biological function and transport substrate remain unknown (2543).

Branch 11 (part of subfamily C)

SLC35C1: SLC35C1 is a GDP-fucose (GDP-Fuc) transporter. Fucose is a terminal residue found in N- and O-glycans. The transporter was discovered via complementation cloning, in order to identify the gene responsible for a genetic defect in GDP-fucose import of an individual with congenital glycosylation disorder characterized by a general deficiency of fucosyl residues in glycoproteins (2544).

Mutations in the SLC35C1 gene lead to congenital disorder of glycosylation (CDG) type IIc (also known as leukocyte adhesion deficiency-2, LAD2). SLC35C1-CDG is a rare inherited condition that primarily affects the immune system, the endocrine system and the nervous system. The patients lack fucose-containing antigens such as the sialyl Lewis X epitope, the ligand for selectins on leukocytes.

Branch 12 (subfamily D)

SLC35D1: SLC35D1 transports UDP-glucuronic acid and UDP-N-acetylgalactosamine (UDP-GlcNAc), which are the substrates for the synthesis of chondroitin sulfate (CS). Chondroitin sulfate is required for cartilage proteoglycans to function. SLC35D1 pathogenic variants block the biosynthesis of CS and cause autosomal recessive Schneckenbecken dysplasia, a severe fatal skeletal dysplasia (2545).

SLC35D2: HFRC1 (SLC35D2) is the human ortholog of the fruit fly Fringe connection (Frc) transporter and has been shown to transport UDP-glucose (UDP-Glc) to the Golgi in addition to UDP-GlcNAc (2546). It has been hypothesized that ER/Golgi nucleotide sugar transporters may contribute to the cellular release of UDP-Glc as a signaling molecule. UDP-Glc would be released from cells (e.g., by exocytosis) to enable UDP-Glc induced purinergic receptor signaling.

SLC35D3: SLC35D3, also known as fringe connection-like protein (FRCL1), is localized to both the ER and early endosomes and functions as a UDP-Glc transporter (2508, 2547). It is involved in platelet dense granule formation, synaptic vesicular neurotransmitter transport, and adipocyte differentiation (2547, 2548). It also regulates dopamine signaling in striatal D1 neurons (2549). SLC35D3 is involved in the pathogenesis of the recessive metabolic disorders Hermansky-Pudlak syndrome, a disease of malformation of lysosome-related organelles such as pigment cell melanosomes (2548, 2550), and Chediak-Higashi syndrome (CHS), a rare inherited disease caused by mutations in the lysosomal trafficking regulator (LYST) gene (2551). Hermansky-Pudlak syndrome (HPS) is characterized by oculocutaneous albinism, which causes abnormally light pigmentation of the skin, hair, and eyes, and bleeding diathesis due to deficiency of the platelet δ-storage pool caused by malformation of platelet dense granules and pigment cell melanosomes (2508, 2548).

Slc35d3 is mutated in the Roswell (Slc35d3 ^ros/_ros) mouse, which leads to dysfunction of lysosome-related organelles and platelet dysfunction, thus serving as an animal model for HPS and CHS. Platelet functions are largely mediated by soluble factors released from membrane-bound storage organelles, including dense granules (2552). The ros mutation results in defective dense granule biogenesis, causing a δ storage pool deficiency with reduced or undetectable dense core structures (2548).

The Slc35d3 ros mutant mice were also shown to have MetS due to impaired dopamine signaling in striatal D1 neurons (2549). The mice showed impaired dopamine signaling in striatal neurons, which are involved in metabolic control in the central nervous system by regulating dopamine signaling. Specifically, in ros striatal D1 neurons, the absence of SLC35D3 leads to the accumulation of D1R on the ER, impairing its exit from the ER. In addition, two mutations in SLC35D3 have been identified in patients with MetS, resulting in altered subcellular localization of SLC35D3 and impaired dopamine signaling in striatal neurons (2549).

SLC35D4: SLC35D4, also known as TMEM241 and C18orf45, is a ubiquitously expressed transporter localized to the cis-Golgi network (2508, 2553). Its Hidden Markov Model (HMM) fingerprint shows similarity to the TC #2.A.7.13 subfamily (7), whose proteins are Golgi GDP-mannose:GMP antiporters from plants, yeast and other organisms (2554, 2555). According to the HPA, SLC35D4 is ubiquitously expressed, with highest expression in excitatory neurons at the single cell level.

SLC35D4 has been shown to function as a UDP-N-acetylglucosamine (UDP-GlcNAc) transporter localized to the Golgi apparatus (2553). Furthermore, SLC35D4 was shown to be required for the mannose-6-phosphate (Man-6P) modification of the Niemann-Pick C2 intralysosomal cholesterol-binding protein NPC2 (2508, 2553). This modification is required for lysosomal targeting of NPC2, which mediates cholesterol exit from lysosomes, as shown in Fig. 55. In addition, knockout of the SLC35D4 gene in HeLa cells resulted in impaired sorting of the NPC2 protein, which in turn caused cholesterol accumulation in lysosomes. The SLC35D4 protein shows similarity to SLC35A3, and both proteins localize to the cis-Golgi network. Cells lacking either SLC35A3 or SLC35D4 showed a significant reduction in UDP-GlcNAc levels in the Golgi apparatus compared to wild-type cells. SLC35A3 knockout cells also showed a reduction in Man-6P modification of total cellular proteins, which was associated with cholesterol accumulation in lysosomes. Overexpression of SLC35A3 rescued the intracellular cholesterol accumulation and NPC2 sorting abnormalities caused by SLC35D4 knockout. Mice deficient in Slc35d4 exhibited cholesterol accumulation in lung cells, lung injury and hypokinesia, highlighting a critical role of this transporter in lung function (2553).

Based on a Mexican genome-wide association study, it has been reported that SLC35D4 mRNA levels are regulated by hepatocyte nuclear factor 4 α (HNF4A) and that decreased SLC35D4 mRNA levels due to disruption of the nuclear factor contribute to elevated triglyceride levels and thus increased risk of cardiovascular disease (2556). Whether this is the result of reduced Man-6P modification remains to be determined.

Orphan transporter family members (18)

SLC35A4, SLC35A5, SLC35E1, SLC35E2A (gene/pseudogene), SLC35E2B, SLC35E3, SLC35E4, SLC35F1, SLC35F4, SLC35F5, SLC35F6, SLC35G1 (POST/TMEM20), SLC35G2 (TMEM22), SLC35G3 (AMAC1/TMEM21A), SLC35G4 (AMAC1L1), SLC35G5 (AMAC/AMAC1L2), SLC35G6 (AMAC1L3/TMEM21B), SLC35H1

HGNC update

SLC35H1 has been renamed SLC35C2. The reasons for this change are explained in the SLC35H1 description. TMEM241 has also been updated to SLC35D4 in line with its discovery as a UDP-GlcNAc transporter, cited above.

SLC36 Proton-coupled amino acid transporter family (2.A.18.8/Aa_trans/APC)

Discovery: Members of this family are part of the amino acid/auxin permease (AAAP) superfamily which also includes the SLC32 and SLC38 families. SLC36 members are H⁺-driven amino acid transporters that were initially discovered in plants by expression cloning in yeast (161) or functional complementation (162). Later on, mammalian members of the SLC36 transporters were identified by homology-based cloning, using either the vesicular neurotransmitter transporter VIAAT/VGAT or the yeast amino acid vacuolar transporter AVT3 cDNAs as a probe (2557) (132).

Gene family members (4):
SLC36A1 (PAT1/LYAAT1)	SLC36A3 (PAT3)
SLC36A2 (PAT2)	SLC36A4 (PAT4/LYAAT2)

Note that the name PAT1 stands for “proton-coupled amino acid transporter” and should not be confused with the “putative anion transporter 1” PAT1 (SLC26A6).

Molecular aspects, physiological roles and links to disease

The SLC36 members identified include the H⁺-coupled lysosomal amino acid transporter PAT1/LYAAT1 (SLC36A1) that actively exports neutral amino acids from lysosomes.

The SLC36 family constitute a group of four lysosomal transporters (Fig. 42), belonging to the Amino acid/auxin permease (AAAP) family (TC 2.A.18.8) which is part of the MFS superfamily (2558). PAT1 and PAT2 are each predicted to have 11 TMHs, each with a cytosolic N-terminus (2559). The C-terminus of each transport protein is located extracellularly if expressed at the plasma membrane and intravesicularly if expressed at the lysosomal or endosomal membrane.

Fig. 42. Phylogenetic tree of the SLC36 family.

PAT transporters regulate intracellular amino acid concentrations and mTORC1 signaling in lysosomes (2560). From there, PAT family members can act as amino acid-stimulated receptors, or so-called “transceptors” that link amino acids to mTORC1 activation (2561).

SLC36A1: PAT1/LYAAT1 (SLC36A1) contributes to amino acid uptake across the apical membrane of enterocytes in the small intestine (2559) and to lysosomal export of neutral amino acids from lysosomes, for example in neurons (132). The HPA suggests that it is highly expressed in the parathyroid gland, brain (cerebellum, cortex), bone, and small intestine.

SLC36A1 transports a wide range of amino acids as well as a variety of amino acid-based drugs and GABA-and proline-related compounds in a H⁺-coupled manner (2559).

In the intestine, it is found on the apical membrane of epithelial cells (2559) where it contributes to the uptake of amino acids from nutrients. However, the extent of its contribution relative to uptake by other systems like B⁰ (SLC6A19) and IMINO/SIT (SLC6A20) is not fully understood.

In lysosomes of neurons, it has been reported to mediate efflux of amino acids produced during intralysosomal proteolysis (2562).

As indicated above, SLC36A1 has also been reported to function as a sensor together with SLC38A9 in lysosomes to help sense intracellular amino acids and thereby regulate the mTORC1 pathway to respond appropriately to intracellular nutrient levels (2563) (see the SLC38 family description).

SLC36A2: PAT2 (SLC36A2) is prominently expressed in kidney, skeletal muscle and testis and is localized on the plasma membrane. In the kidney, it plays a role in the reabsorption of glycine, proline, and hydroxyproline. Its function is in part shared with B⁰AT1 (SLC6A19), XT2 (SLC6A18), and SIT1 (SLC6A20) (2558).

Iminoglycinuria is an autosomal recessive renal tubular transport disorder caused by defects in SLC36A2 (1014, 2564) (and possibly also by defects of SLC36A1/PAT1 expressed in the intestine). The disease affects the H⁺-coupled renal reabsorption of the amino acids glycine, proline and hydroxyproline. The primary characteristic of the disease is increased urinary excretion of these amino acids, but otherwise the disorder is relatively benign.

SLC36A2 was found to be the major gene responsible for iminoglycinuria and its inheritance was consistent with a classical semidominant pattern in which two inherited nonfunctional alleles conferred the disease phenotype of iminoglycinuria, while one nonfunctional allele was sufficient to confer the related disorder, hyperglycinuria without iminoglycinuria (1014). Mutations in SLC36A2 that retained residual transport activity resulted in the iminoglycinuria phenotype when combined with mutations in the gene encoding the imino acid transporter SIT (SLC6A20), also known as system IMINO. Additional mutations have been identified in the genes encoding the neutral amino acid transporters B⁰AT3 (SLC6A18) and B⁰AT1 (SLC6A19) that result in either iminoglycinuria or hyperglycinuria.

SLC36A3 - Orphan transporter: PAT3 (SLC36A3) is testis-specific but its function remains unknown. Studies in mice also showed high expression in brown adipose tissue and results from Slc36a3 knockout mice revealed a role for SLC36A3 in systemic glucose and lipid metabolism (2565). However, the exact physiological role and transport function of PAT3 (SLC36A3) is still unclear.

SLC36A4: PAT4/LYAAT2 (SLC36A4) is ubiquitously expressed, with strong expression in the retina. It is a lysosomal amino acid sensor linked to mTORC1 activity. PAT4, expressed in retinal pigmented epithelial cells, mediates the amino acid-sensing mechanism that regulates mTORC1 activation inside the cell (2566), similar to the amino acid sensor SNAT9 (SLC38A9) (see the SLC38 family description) (1654) and PAT1/LYAAT1 (SLC36A1) (see above). It might be a target for the treatment of certain cases of macular degeneration.

Orphan transporter family member (1)

SLC36A3 (PAT3)

SLC37 Sugar-phosphate/phosphate exchanger family (2.A.1.4/MFS_1/MFS)

Discovery: The first member of the sugar-phosphate/phosphate exchanger family (SLC37) was human SLC37A1. Its mRNA was identified as part of an exon trapping strategy to identify chromosome 21 genes involved in Down syndrome (132), which resulted in the cloning and characterization of a human glycerol 3-phosphate permease gene.

Gene family members (4):
SLC37A1 (SPX1)	SLC37A3 (SPX3)
SLC37A2 (SPX2)	SLC37A4 (SPX4/G6PT)

Molecular aspects, physiological roles and links to disease

The SLC37 transporters are members of the Organophosphate:P_i Antiporter (OPA) family (TC 2.A.1.4) due to their homology to the bacterial hexose-6-phosphate and glycerol-3-phosphate (G3P) translocases and thus belong to the major facilitator superfamily (MFS) (2567–2569). OPA family members are responsible for the transport of specific organophosphates or sugar phosphates across biological membranes with the simultaneous translocation of inorganic phosphate (P_i) in the opposite direction.

SLC37A1, SLC37A2 and SLC37A3 are predicted to contain 10-12 TMHs (2569, 2570). 10 TMHs were predicted for SLC37A4 (2569, 2571, 2572).

The SLC37 family consists of four sugar-phosphate exchangers which are anchored in the endoplasmic reticulum or lysosome membranes (Fig. 43). The best-characterized member of this family is SPX4 (SLC37A4), better known as the glucose-6-phosphate (G6P) transporter (2568). SPX1 (SLC37A1), SPX2 (SLC37A2) and SPX4/G6PT (SLC37A4) have all been shown to function as P_i-linked G6P antiporters capable of both homologous (P_i:P_i) and heterologous (G6P:P_i) exchange, whereas SLC37A3 lacks antiporter activity (2573, 2574). The physiological roles of SPX1 (SLC37A1), SPX2 (SLC37A2), and SPX3 (SLC37A3), are less well characterized compared to SPX4/G6PT (SLC37A4).

Fig. 43. Phylogenetic tree of the SLC37 family.

SLC37A1: SPX1 (SLC37A1) is a G6P/P_i exchanger that is ubiquitously expressed, most strongly in duodenum, cerebellum, skeletal and smooth muscle, and prostate, and at lower levels in pancreas, bone marrow, thymus, and spleen as previously reported (2568) and also according to the HPA. SPX1 (SLC37A1) has been localized to the ER (2575). It is homologous to bacterial glycerol-3-phosphate permeases, suggesting that it may also transport glycerol-3-phosphate, but experimental verification is still required (2576). An association of SPX1 (SLC37A1) with glycolipid metabolism has been suggested (2567, 2577, 2578). Moreover, an important role of SLC37A1 as a G3P exchanger in cancer cells has been postulated (2575). It has been proposed that SLC37A1 is required for lipid biosynthesis in cancer cell lines (2575). Although G3P transport activity has not been demonstrated, SPX1 (SLC37A1) has been proposed to import G3P into the ER lumen to support phospholipid biosynthesis required to promote cancer progression (2575, 2578).

SLC37A2: SPX2 (SLC37A2) displays 59% amino acid sequence identity to SPX1 (SLC37A1) and is also widely expressed (2568). SPX2 is thought to function as a secretory lysosomal sugar transporter that is critical for bone metabolism. Osteoclasts, giant bone-digesting cells, contain specialized lysosome-related organelles called secretory lysosomes. The findings highlight a previously unrecognized plasticity of the specialized lysosome-related organelle of the osteoclast (2579). According to the authors of this study, SPX2 (SLC37A2) corresponds to a previously described transport function of glucose and fructose out of lysosomes by an unknown monosaccharide transporter (2580) and it was suggested that SPX2 (SLC37A2) fulfills this role in osteoclasts because, for example, metabolic profiling showed that loss of function of this transporter led to increased levels of glucose and fructose in osteoclasts.

The HPA also suggests that SPX2 (SLC37A2) is most highly expressed in the salivary gland, whereas expression in bone, adrenal gland, skin and urinary bladder is shown at somewhat lower levels. Whether and how SPX2 is involved in the lysosomal activity in the parenchyma of the salivary gland remains to be determined.

SLC37A3: SPX3 (SLC37A3) is a ubiquitously expressed lysosomal transporter that plays an important role in the treatment of osteoporosis (2568). Briefly, nitrogen-containing bisphosphonates (N-BPs) are a class of drugs widely prescribed for the treatment of osteoporosis, which are dependent on the expression of SLC37A3 in lysosomes (2581). SLC37A3 forms a complex with ATRAID (all-trans retinoic acid-induced differentiation factor). SLC37A3 and ATRAID localize to lysosomes and are required for the release of N-BP molecules, which have been trafficked to lysosomes by fluid-phase endocytosis into the cytosol. Thus, SLC37A3 opens a gate in the cell membrane that allows N-BPs to enter osteoclasts.

SLC37A4: SPX4/G6PT (SLC37A4), also known as glucose-6-phosphate transporter (G6PT), transports glucose-6-phosphate from the cytoplasm into the ER lumen (2568, 2569). It is widely expressed, particularly in the liver and kidney, where it is part of a multicomponent system with glucose-6-phosphatase to maintain blood glucose homeostasis (2567, 2568). G6PT consists of 10 TMHs with both N- and C-termini facing the cytoplasm (2569, 2572, 2582). At the C-terminus it has a KKXX endoplasmic retention signal (2571).

In a healthy individual, maintenance of fasting blood glucose at approximately 5 mM is achieved primarily by glycogenolysis and gluconeogenesis in the liver and gluconeogenesis in the kidney. For glucose to be released from the liver or kidney, glucose-6-phosphate must first be imported from the cytosol into the ER, where it is cleaved into glucose and phosphate. Glucose is then exported to the cytosol from where it can be released into the blood via GLUT2 (SLC2A2) in the liver and via GLUT1 (SLC2A1) or GLUT2 (SLC2A2) in the proximal tubules of the kidney (2568, 2583). Interestingly, the transporter mediating glucose export from the ER is unknown (2584). It has been referred to as hepatocyte glucose-6-phosphatase subcomponent T3, which was thought to be similar to GLUT2 (SLC2A2) (2585), yet alternative glucose pathways have been discussed (2586).

Glycogen storage disease type I (GSD-I) is an autosomal recessive disease. It is caused by abnormalities in two different genes encoding glucose-6-phosphatase G6Pase-α (G6PC1) and the glucose-6-phosphate transporter SPX4/G6PT (SLC37A4), which form the G6PT/G6Pase-α complex in the ER. This complex is essential for maintaining interprandial glucose homeostasis by catalyzing glucose production as the final step of both glycogenolysis and gluconeogenesis (2569). Mutations in the G6PC1 gene result in GSD-Ia, while mutations in the glucose-6-phosphate transporter SPX4/G6PT (SLC37A4) result in GSD-Ib. Both deficiencies cause excessive accumulation of glycogen and fat in the liver, kidneys, and intestinal mucosa, resulting in liver enlargement and hypoglycemia.

To maintain normal glucose levels, prevent hypoglycemia, and maximize growth and development, GSD-I is treated with a special diet consisting of small, frequent portions of carbohydrates. Because fructose and galactose cannot be metabolized to glucose via G6P in GSD-I, their intake contributes to the metabolic abnormalities. Therefore, a diet low in sucrose and fructose and high in complex carbohydrates, such as whole grains, with limited intake of galactose and lactose is recommended (2587, 2588).

Among the various loss-of-function mutations that cause GSD-Ib, many are single point mutations affecting residues throughout the SPX4/G6PT amino acid sequence, except for the C-terminus (2569, 2572). Individuals with the disease also suffer from neutropenia, neutrophil dysfunction, and inflammatory bowel disease (IBD), in addition to hypoglycemia and accumulation of glycogen in the liver and kidneys (2589). The reason for the additional symptoms is that SPX4/G6PT (SLC37A4) is also expressed in neutrophils, along with glucose-6-phosphatase-β (G6PC3), with this analogous G6PT/G6Pase-β complex playing a critical role in neutrophil maintenance. Loss of function of SPX4/G6PT (SLC37A4) in this complex impairs neutrophil survival and blunts their metabolic burst associated with phagocytosis (2590). The resulting immunodeficiency, if untreated, makes GSD-Ib patients susceptible to infection. The main treatment for this condition of GSD-Ib involves the use of filgrastim, an engineered type of colony stimulating factor that stimulates white blood cell production, especially neutrophil production (2591). Enterocolitis due to an IBD-like disorder is also common in individuals with GSD-Ib (2592).

A second disease caused by a single autosomal dominant mutation in the SLC37A4 gene has also been described (2571), which affects the C-terminal region of SPX4/G6PT, abolishing the above-mentioned endoplasmic retention signal and instead exposing a weak Golgi retention signal (2593). SPX4/G6PT derived from this mutant allele relocalizes to the Golgi where it interferes with the processing of protein N-glycans leading to a congenital disorder of glycosylation rather than glycogen storage disease (2571).

Orphan transporter family members: N/A

SLC38 System A and System N sodium-coupled neutral amino acid transporter family (2.A.18.6 and 2.A.18.9/Aa_trans/APC)

Discovery: The first members of the SLC38 family of Na⁺-coupled neutral amino acid transporters, SNAT1 (System A/SLC38A1) (2594) and SNAT3 (system N/SLC38A3) (2595) (2596) were cloned by virtue of their sequence homology with the vesicular inhibitory amino acid transporter (VIAAT, also known as the vesicular GABA transporter, VGAT, or SLC32A1), a transporter responsible for vesicular packaging of neurotransmitter GABA and glycine in central inhibitory neurons (2597).

Gene family members (12):
SLC38A1 (SNAT1)	SLC38A4 (SNAT4)	SLC38A7 (SNAT7)	SLC38A10
SLC38A2 (SNAT2)	SLC38A5 (SNAT5)	SLC38A8 (SNAT8)	SLC38A11
SLC38A3 (SNAT3)	SLC38A6 (SNAT6)	SLC38A9	SLC38A12 (TMEM104)

Molecular aspects, physiological roles and links to disease

The SLC38 family belongs to the Amino Acid/Auxin Permease (AAAP) family (TC 2.A.18.6), which is part of the APC superfamily and possesses the LeuT fold. In general, members of the SLC38 family mediate Na⁺-dependent transport of amino acids. Certain SLC38 transporters, such as SLC38A9, are also involved in amino acid sensing and signaling. The structure of SLC38 transporters contains the 5 + 5 inverted repeat core characteristic of the LeuT fold. The family can be further divided into the following subfamilies based on the phylogenetic relationships (Fig. 44):

Fig. 44. Phylogenetic tree of the SLC38 family.

Subfamily 1 - System A: SLC38A1, SLC38A2 and SLC38A4

Subfamily 2 - System N: SLC38A3, SLC38A5, SLC38A6

Subfamily 3 - SLC38A7 and SLC38A8

Subfamily 4 - SLC38A10 and SLC38A11

Subfamily 5 - SLC38A9 and SLC38A12

Description of the members of each subfamily

Subfamily 1 - System A (SLC38A1, SLC38A2 and SLC38A4) 

The system A transporters mediate the transport of a variety of small neutral amino acids such as alanine, are inhibited by the substrate analogue 2-methylamino-isobutyric acid (MeAIB), and are regulated by amino acid supplementation or depletion (1079).

SLC38A1: SNAT1 (SLC38A1), previously referred to as ATA1, GlnT, SA2, SAT1 or NAT2, mediates the transport of aliphatic and zwitterionic amino acids, with a preference for glutamine (K_0.5 approximately 0.3 mM), alanine, asparagine, cysteine, histidine, and serine (2594, 2597–2599). It also transports the System A-specific analog MeAIB. Transport is electrogenic and pH-sensitive and it is coupled to the cotransport of one Na⁺ ion, creating a unidirectional flow, unlike the System N transporters (2597).

SNAT1 (SLC38A1) is mainly expressed in brain, retina, muscle, endocrine tissues, placenta, testis, bone marrow, lymphoid tissues, intestine and kidney (2597). In brain it is present in the cerebral cortex in pyramidal and GABAergic neurons (2599, 2600) (see Fig. 16). The localization of SNAT1 to certain dopaminergic neurons in the substantia nigra and cholinergic motoneurons suggests that it may have additional specialized functional roles. For example, SNAT1 may provide metabolic fuel, such as α-ketoglutarate, or precursors like cysteine and glycine for glutathione synthesis (2600).

SLC38A2: SNAT2 (SLC38A2), also known as SAT2, SA1 or ATA2, transports neutral amino acids, including glutamine, alanine, and serine, among others (2597, 2601, 2602). Transport is voltage- and Na⁺-dependent (1:1 stoichiometry) and pH-sensitive. It is inhibited by MeAIB. The transport mechanism of SNAT2 is similar to that of SNAT1, but SNAT2 is expressed more widely than SNAT1. However, its expression in the retina and brain is lower than that of SNAT1 (2601, 2602).

Excess amino acids in our body are primarily metabolized in the liver. Alanine acts as the main amino acid for gluconeogenesis. The transport of amino acids is facilitated by systems A and N, predominantly through SNAT2 (SLC38A2). In the diabetic condition, studies have shown that SNAT2 (Slc38a2) and SNAT4 (Slc38a4) are upregulated in streptozotocin-induced diabetic rats, likely as a direct consequence of insulin deficiency, thereby providing amino acids such as alanine to periportal (SNAT2) or perivenous (SNAT4) hepatocytes for gluconeogenesis. Additionally, glucagon also increased upregulation (2603).

SNAT2 (SLC38A2) facilitates the uptake of glutamine in excitatory neurons, with SNAT1 (SLC38A1) predominantly mediating this process in inhibitory neurons (2604). In short, neuronal L-Glu can be synthesized from L-Gln supplied by astrocytes (see Fig. 6) (365). L-Gln exiting from astrocytes via SLC38A3 and SLC38A5 (system N) is subsequently taken up by presynaptic neurons through SLC38A2, and also SLC38A7, where glutaminase GLS2 in the mitochondrial intermembrane space hydrolyzes L-Gln into L-Glu, which is then transported into synaptic vesicles via vesicular glutamate transporters such as SLC17A6.

SNAT2 (SLC38A2) has also been implicated as a transceptor for the activation of mTOR known to regulate cell proliferation, autophagy, and apoptosis, similar to SLC38A9 (2605, 2606) (see below). Oncogenic KRAS mutations have been found to enhance the expression of various glutamine transporters, including the transceptor SLC38A2, in colorectal cancer cells, according to recent experiments conducted in our laboratory (2607). Oncogenic KRAS mutations increased the expression of amino acid transporters via the Hippo effector YAP1, leading to mTOR activation and colon cancer cell proliferation.

SLC38A4: SNAT4 (SLC38A4) expression is unusual in that it is almost exclusively expressed in the liver in perivenous but not periportal hepatocytes (2597), although expression in the placenta has also been reported (2608). As previously stated, SNAT4 (SLC38A4) is upregulated in the diabetic condition, allowing amino acids such as alanine to be used for gluconeogenesis in perivenous hepatocytes. However, under healthy conditions, gluconeogenesis is mainly confined to periportal hepatocytes (2609). Additionally, the presence of glutamine uptake function in perivenous hepatocytes may seem unnecessary since these cells are known for producing glutamine. But glutamine is not one of the preferred substrates for SNAT4. As demonstrated by two separate studies, SNAT4 has a high affinity for transporting cationic amino acids like lysine and arginine, regardless of Na⁺. (2610) (2608). Thus, SNAT4 (SLC38A4) is a Na⁺-dependent transporter for neutral amino acids and a Na⁺-independent transporter for cationic amino acids, similar to CAT-1 (SLC7A1) and y⁺LAT1 (encoded by SLC3A2 and SLC7A7). L-arginine supplementation is known to protect the liver from injury that may be mediated by nitric oxide (2611, 2612) and it is tempting to speculate that L-arginine is taken up by SNAT4 (SLC38A4) since the liver does not express any of the other known cationic amino acid transporters such as b^0,+ (encoded by SLC3A1 and SLC7A9) or y⁺LAT1, which are mainly expressed in the small intestine and kidney, neither B^0,+ (SLC6A14) (988), whose expression is mostly restricted to the lung and mammary gland. The only cationic amino acid transporter that has been shown to be expressed in the liver is CAT-2a, an alternative isoform of CAT-2 (SLC7A2) (1087). However, it is a low affinity transporter, whereas SNAT4 (SLC38A4), which is highly expressed in the liver, has about 10-fold higher affinity for arginine than CAT-2a (2610). Therefore, it is likely that SNAT4 (SLC38A4) is the main mechanism for arginine entry into hepatocytes under physiological conditions.

Expression of SNAT4 has also been detected in syncytiotrophoblast microvillous plasma membrane vesicles of human placenta during the first trimester and also at term, suggesting that this transporter transports amino acids, including arginine, from maternal blood into the intervillous space, into the placenta and to the fetus (2608). In line with this, it is well documented that L-arginine plays an important role in the placenta during pregnancy (2613) and prenatal oral L-arginine may have beneficial effects on birth outcomes in women with a history of poor pregnancy outcomes (2614). Furthermore, as pregnancy progresses, placental endothelial nitric oxide synthase (eNOS) expression in syncytiotrophoblasts increases and produces nitric oxide from L-arginine, which plays an important role in placental development (2615).

Subfamily 2 – System N (SLC38A3, SLC38A5, SLC38A6) 

This transport system is designated as system N to indicate the presence of nitrogen in the side chain of its substrates, such as glutamine, asparagine, and histidine. In contrast to system A, system N transporters countertransport H⁺ while also being Na⁺-coupled, which could be crucial for their reverse transport in certain cell types such as astrocytes (2597).

SLC38A3: SNAT3 (SLC38A3), also known as SN1, mediates amino acid transport that is coupled to the cotransport of 1 Na⁺ and the countertransport of 1 H⁺, making the transport electroneutral (2595, 2597). According to the HPA and other sources, it is highly expressed in liver and retina, and at lower levels in brain, muscle and pancreas (2595, 2616). It is responsible for the transport of glutamine, asparagine, histidine, and alanine (2595). SNAT3 can mediate both glutamine efflux and uptake under physiological conditions.

Expression in the brain is largely restricted to astrocytes (2595), especially perisynaptic astrocytes, where it serves as a glutamine exporter as part of the glutamate-glutamine cycle (2617, 2618). In brief, glutamate is released from glutamatergic presynaptic terminals and promptly eliminated from the synaptic cleft to terminate neurotransmission and prevent excitotoxicity. This is done by sequestering it into neighboring astrocytes through glutamate transporter GLT1/EAAT2 (SLC1A2). This transporter combines the absorption of glutamate with the simultaneous transport of 3 Na⁺ ions, which raises the intracellular Na⁺ concentration and promotes the efflux of glutamine via Na⁺-coupled SNAT3 (SLC38A3) in exchange for H⁺, thereby acidifying the cell (2595, 2618). Glutamine enters the presynaptic terminals through a glutamine transporter of the SLC38 family and is converted to glutamate to replenish the presynaptic neurotransmitter pool, as previously described (365, 2619). While it was previously thought that SNAT1 (SLC38A1), SNAT2 (SLC38A2), or SNAT7 (SLC38A7) fulfill this function, it was subsequently proposed that this neuronal glutamine transporter is SNAT6 (SLC38A6) (2620) (see below).

An analogous situation can be observed in GABAergic synapses, where the neurotransmitter GABA is released and absorbed by the neighboring astrocytes, in addition to being transported back to the presynaptic terminal via GAT1 (SLC6A1). The GABA within the astrocytes is then metabolized into α-ketoglutarate, which is subsequently converted to glutamate, and finally to glutamine, which then exits the astrocyte via SNAT3 (SLC38A3). Glutamine is transported back into the presynaptic GABAergic neuron via SNAT1 (SLC38A1) as mentioned above, where it is converted to glutamate and then to GABA. Thus, in the brain, SNAT3 (SLC38A3) plays an important role in replenishing the glutamate and GABA neurotransmitter pools. In agreement with this, Slc38a3^-/- mice show a reduction of the glutamate and GABA neurotransmitter pools in the brain (2616).

In the retina, SNAT3 (SLC38A3) likewise plays an important role in the recycling of glutamate, the major neurotransmitter for the photoreceptor-bipolar-ganglion cell circuit (2621). Glutamate, once released at the synapse, is cleared from the synaptic space by GLT1/EAAT2 (SLC1A2) found in retinal Müller glial cells. There, glutamate is converted to glutamine, which is then released and taken up by ganglion cells.

The glutamate/glutamine cycle also plays an important role in the liver, where it facilitates ammonia detoxification from the portal blood. Periportal hepatocytes convert glutamine, taken up by System A or System N amino acid transporters (including SNAT3), into glutamate. The formation of ammonium can be toxic to the CNS if not cleared. Therefore, ammonium produced exits together with glutamate, followed by entry of these molecules into perivenous hepatocytes, where glutamate uptake occurs via the glutamate transporter SLC1A2 (GLT1/EAAT2) (408), which is coupled to the cotransport of 3 Na⁺ ions. Glutamine is synthesized again within perivenous hepatocytes via glutamine synthetase and is transported out of the cell into the bloodstream with the Na⁺ ions from SLC1A2 transport via SNAT3 (SLC38A3) into the blood.

During metabolic acidosis, there is a significant increase in glutamine production in the liver, and the liver becomes the primary site for net synthesis of glutamine in perivenous hepatocytes (2622). The resulting glutamine is transported into renal proximal tubule cells for ammoniagenesis during metabolic acidosis through the basolateral transporter SLC38A3, whose expression is induced in the kidney during metabolic acidosis (2623).

The clinical features of ten subjects from seven unrelated families who had biallelic deleterious variants of SLC38A3 have been analyzed (2624). The results align well with the above-described physiological roles of SNAT3 (SLC38A3). These features consist of 1) epileptic encephalopathy, which correlates with the reduction of GABA signaling, 2) visual impairment, resulting from the defective recycling of glutamate in the retina, 3) elevated plasma ammonia, consistent with a role for SNAT3 (SLC38A3) in ammonia detoxification in hepatocytes, and 4) metabolic acidosis, likely due to failure of renal proximal tubule cells to produce ammonia due to lack of SNAT3 (SLC38A3) transport of glutamine into renal proximal tubule cells. SNAT5 (SLC38A5) performs many of the same functions as SNAT3 (SLC38A3). However, the abnormalities observed in the subjects above suggest that SNAT5 (SLC38A5) is unable to adequately compensate for the function of SNAT3 (SLC38A3).

A significant association with T2D was found for the rs1858828-G/T variant of the SLC38A3 gene, highlighting the role of SNAT3 in the pancreas where it is expressed in β-cells (2625). It has been reported that elevated plasma levels of glutamine induce the uptake of this amino acid by SNAT3 (SLC38A3) with consequent stimulation of insulin secretion through the conversion of glutamine to glutamate (2626). However, validation of T2D association in an independent group of patients is still needed, and it remains to be seen whether this variant also affects GABA signaling, the visual system, and metabolic acidosis.

SLC38A5: SNAT5 (SLC38A5), also known as SN2, facilitates the electroneutral, Na⁺-dependent cotransport of neutral amino acids with H⁺ antiport. It transports glutamine, asparagine, histidine, serine, glycine, and alanine, with a stoichiometry of 1 Na⁺ in, 1 amino acid in, and 1 H⁺ out (2627–2629). Like SNAT3 (SLC38A3), SNAT5 can act as an amino acid-dependent Na⁺/H⁺ exchanger. This highlights the unique role of system N amino acid transport in cellular pH regulation. According to the HPA, and in agreement with the findings following the cloning of SLC38A5 (2627, 2628), this gene is highly expressed in the pancreas and at lower levels in the brain, stomach, intestinal tract, respiratory system, skin, bone marrow, among others.

SNAT5 (SLC38A5) has been reported to be regulated at mRNA and protein levels by mTORC1 and WNT/β-catenin pathways, and to be sensitive to pH, nutrient stress, inflammation, and hypoxia, as well as being overexpressed in several types of cancer cells (2630).

In the pancreas, SNAT5 (SLC38A5) plays a crucial role in the amino acid sensing machinery (2631). The α-cells of the islets of Langerhans in the pancreas detect elevated levels of plasma amino acids by uptake via SNAT5 (SLC38A5), which stimulates glucagon secretion and mTOR-dependent proliferation. mTOR also stimulates the expression of SLC38A5 to promote further amino acid uptake.

A study focusing on psoriatic skin inflammation demonstrated that SNAT5 (SLC38A5) localizes to lysosomes in dermal dendritic cells, facilitating lysosomal acidification (2632). This activity enhances TLR7 signaling and the production of pro-inflammatory cytokines, thereby exacerbating psoriatic skin inflammation. Mechanistically, SNAT5 (SLC38A5) potentiates lysosomal acidification in dendritic cells, which dictates the cleavage and activation of TLR7 with ensuing production of pro-inflammatory cytokines such as IL-23 and IL-1β and eventually aggravates psoriatic inflammation. Thus, this study uncovered a novel supplemental mechanism in driving lysosomal acidification, and revealed SNAT5 (SLC38A5) as a putative therapeutic target for treating psoriasis.

SLC38A6: SNAT6 (SLC38A6) is a ubiquitously expressed transport protein. In the brain, it has been reported to be exclusively expressed in excitatory neurons (2620). Although SNAT6 has long been considered an orphan transporter with an unknown substrate profile, subsequent studies have shown that its substrates are glutamine and glutamate (2620). Furthermore, it has been proposed that this transporter facilitates the uptake of glutamine produced by neighboring astrocytes from glutamate, as part of the glutamate-glutamine cycle, into presynaptic neurons in complex with caveolin (2620). The presence of these amino acid substrates enables the formation of SNAT6-caveolin complexes, which aid in the sodium-dependent trafficking of SNAT6 off the plasma membrane(2620).

In the lungs, SNAT6 is an essential protein for sepsis-related pulmonary inflammation. Some research suggests that SNAT6 causes inflammation by directly maintaining interleukin IL-1β in macrophages and affecting kinase activity, which impacts the expression of inflammatory genes in macrophages (2633). It has been proposed that the transporter represents a promising pharmaceutical target for the management of bacterial pneumonia and pulmonary inflammation associated with sepsis (2633). However, it is still unclear how exactly its transport function affects pulmonary inflammation and whether glutamine transport is involved.

Subfamily 3 (SLC38A7 and SLC38A8) 

The two members of this subfamily play crucial roles in maintaining proper retinal function and other functions, as discussed below.

SLC38A7: SNAT7 (SLC38A7) is an Na⁺-coupled amino acid transporter with glutamine as its preferred substrate. It shows a rather widespread tissue distribution (2634).

Strong expression has been reported in both glutamatergic and GABAergic neurons, where SNAT7 (SLC38A7) may contribute to the recycling of GABA and glutamate (2634). As shown in Fig. 6 and Fig. 16, glutamate and GABA are recycled between neurons and astrocytes, involving their conversion to glutamine in the astrocyte and subsequent delivery of glutamine to neurons. Neuronal localization of the SNAT7 (SLC38A7) transporter has been demonstrated, with high expression at the cell body membrane and in axons. Based on Xenopus oocyte expression studies, the preferred substrate was L-glutamine and the transport was Na⁺-coupled.

Genetic variability of SLC38A7 has been shown to influence susceptibility to T2D and diabetic retinopathy (2635). In particular, the intronic SNP rs9806843 in SLC38A7 was found to be protective against retinopathy (2635). The protective role has been explained in the context of glutamate acting as an excitatory neurotransmitter in the retina, while excess glutamate may be toxic to retinal neurons. Exactly how SNAT7 (SLC38A7) is involved in glutamatergic and GABAergic neurotransmission and how genetic variability in SLC38A7 affects susceptibility to T2D and diabetic retinopathy still requires further investigation.

In contrast, other studies have identified SNAT7 (SLC38A7) as a major lysosomal glutamine transporter. In one study, it was shown to be required for extracellular protein-dependent growth of cancer cells (2636). SNAT7 has been proposed to be the primary permeation pathway for glutamine across the lysosomal membrane required for cancer cell growth in a low free glutamine environment when macropinocytosis and lysosomal degradation of extracellular proteins are used as an alternative source of amino acids. Therefore, SNAT7 has been proposed as a novel target for glutamine-based anticancer therapies.

Subsequently, SNAT7 was shown to be an important regulator of mTORC1 following macropinocytosis of extracellular proteins such as albumin (2637). In support of this, SNAT7 depletion inhibited albumin-induced mTORC1 lysosomal localization and subsequent activation. SNAT7 was found to be essential for maintaining KRAS-driven pancreatic cancer cell growth via mTORC1. It was concluded that SNAT7 links glutamine signaling from extracellular proteins to mTORC1, independent of the heterodimeric Rag GTPase complex, in contrast to SLC38A9 (see below). This type of glutamine signaling has been shown to be required for macropinocytosis-mediated mTORC1 activation in pancreatic cancer cell growth, and deletion of SNAT7 significantly hindered macropinocytosis-mediated activation of mTORC1, cell size, and cell proliferation of pancreatic cancer (2637).

It will be an interesting for future studies to clarify what determines the ability of SNAT7 (SLC38A7) to function at the plasma membrane or in lysosomes in different tissues and cell types.

SLC38A8 - Orphan transporter: SLC38A8 has a limited tissue distribution, being expressed in both the retina (2638) and amygdala. SLC38A8 is a putative glutamine transporter with strong expression in the photoreceptor layer of the retina. SLC38A8 mutations lead to an arrest of retinal development at an earlier stage, resulting in arrested retinal development with loss of cone photoreceptor specialization, a rare medical condition called foveal (or macular) hypoplasia (2638) (2639, 2640). The precise functional mechanism by which SLC38A8 leads to retinal developmental arrest and whether it is expressed in the plasma membrane or in intracellular compartments has not yet been elucidated.

Subfamily 4 (SLC38A10 and SLC38A11)

Subfamily 4 is related to bidirectional amino acid transporters expressed in subcellular compartments.

SLC38A10 - Orphan transporter (in the sense that its physiological function is not yet clearly understood)

SLC38A10 is widely expressed in human tissues (2641). It has been shown to facilitate bidirectional transport of L-alanine, L-glutamate, L-glutamine, and D-aspartate and efflux of L-serine when expressed in Xenopus laevis oocytes (2642). Bidirectional transport of amino acids is a characteristic of system N transporters. On the other hand, SLC38A10 has a characteristic of system A transporters in that it is inhibited by MeAIB (2642). In the brain, it is expressed in both astrocytes and neurons (2642) and localized to the Golgi and endoplasmic reticulum (2643). There, it may play a role in neurotransmission of neurons and astrocytes by transporting glutamine, glutamate and aspartate. SLC38A10 deficiency in mice affects plasma threonine and histidine levels based on initial characterization of Slc38a10^-/- mice (2644, 2645). Slc38a10 knockdown resulted in decreased body and plasma levels of the essential amino acids threonine and histidine. The widespread expression of this transporter suggests an important role in the general maintenance of amino acid homeostasis in cells, so that its deficiency has an impact on body weight homeostasis. However, the underlying mechanism for this needs to be further clarified. Interestingly, Slc38a10-deficient mice have increased exploratory behavior, suggesting a functional role of SLC38A10 in the brain as well (2645). Overall, the exact biological function of SLC38A10 still needs to be elucidated in more detail.

SLC38A11 - Orphan transporter: There is currently no information on the functional properties of human SLC38A11 and whether it is expressed on the plasma membrane or in intracellular compartments. However, studies on the knockdown of the Drosophila SLC38A11 ortholog, called CG13743, which is mainly expressed in the salivary gland and brain, revealed a metabolic relevance of the encoded transporter in the Drosophila fruit fly in terms of maintaining general metabolic pathways and behavior (2646). According to the HPA, SLC38A11 is widely expressed in human tissues, with the highest expression observed in kidney, choroid plexus, stomach, gallbladder, epididymis, endometrium, cervix, and smooth muscle, but not in salivary gland. Whether SLC38A11 function is related to nutrient sensing network as part of the mTOR pathway remains to be determined.

Subfamily 5: SLC38A9 and SLC38A12

Subfamily 5 includes lysosomal transporters.

SLC38A9: SLC38A9 is broadly expressed in human tissues, with highest expression levels in duodenum, placenta, liver, adrenal gland, and testis. It mediates the lysosomal efflux of nonpolar essential amino acids – including phenylalanine, leucine, isoleucine, tryptophan, and methionine – as well as tyrosine, via a Na⁺-coupled transport mechanism (2641).

In addition, SLC38A9 functions as an arginine-sensitive high affinity transporter for leucine, linking amino acid availability to mTORC1 signaling (1654, 2641). The K_m for leucine was found to be ~90 μM, compatible with lysosomal leucine concentrations of 60-80 μM, while that for arginine was much higher (~4 mM). SLC38A9 transports amino acids, such as leucine, in an arginine-regulated manner (2641, 2647, 2648). First, arginine binds to SLC38A9 on the luminal side of the lysosomal membrane. This binding induces a conformational change that causes SLC38A9 to export leucine to the cytosol. The crystal structure of the zebrafish SLC38A9 protein revealed the L-arginine binding mechanism in the transport cycle (2647, 2648).

SLC38A9 functions as a lysosomal transceptor that senses luminal arginine, a key regulator of mTORC1 signaling and cellular metabolism (1654). Later studies identified additional amino acid sensors involved in mTORC1 regulation (2649, 2650). Below is a brief overview of the metabolic role of lysosomes, current understanding of the mTORC1 pathway, and the role of SLC38A9.

Lysosomes degrade proteins, lipids, carbohydrates, and nucleic acids, maintaining an acidic environment via the vacuolar-type H⁺-ATPase (V-ATPase), which optimizes conditions for hydrolytic enzymes. Beyond degradation, lysosomes play a central role in metabolic adaptation, nutrient sensing, and the regulation of inflammatory responses (2651–2654). Dysfunction in lysosomal pathways contributes to a wide range of diseases, including lysosomal storage disorders, neurodegeneration, and immune-related conditions (1613, 2300, 2655–2659). The mTORC1 pathway is a key metabolic regulator that governs cellular homeostasis, growth and autophagy (2660–2662). Depending on nutrient availability and in response to growth factor signals, it facilitates anabolic or catabolic processes. Under nutrient-rich conditions, mTORC1 activation stimulates cell growth and suppresses autophagy. Conversely, during nutrient deprivation or stress, mTORC1 is inhibited, triggering autophagy and a shift toward catabolism (23, 2660, 2663–2665). Dysregulation of mTORC1 is implicated in aging, cancer, and metabolic disorders such as diabetes (2666).

Nutrients such as amino acids generated by lysosomal degradation activate the mTORC1 complex through a process involving Rag GTPases and the Ragulator complex (see Fig. 45). Conversely, insulin activates mTORC1 through a pathway involving the tuberous sclerosis complex (TSC1 and TSC2) and the small GTPase Rheb. Specifically, insulin inhibits the TSC1-TSC2 complex via the PI3K-Akt signaling pathway, thereby releasing Rheb, a direct activator of mTORC1 (2667). These nutrient- and growth factor-dependent signaling cascades converge at the lysosomal surface, ensuring that mTORC1 is fully activated when these inputs are present (23, 2667, 2668).

Fig. 45. The role of SLC38A9 in nutrient sensing.

The figure illustrates the mTORC1 signaling pathway, which promotes anabolism and cell growth when active, and drives catabolism and autophagy when inhibited. Green highlights processes that activate mTORC1 and support growth, while red indicates steps leading to mTORC1 inhibition and induction of autophagy. The activity of GATOR2 is regulated primarily by two amino acids: leucine and arginine. These amino acids bind to Sestrin2 and CASTOR1, respectively. In the absence of leucine, Sestrin2 binds to and inhibits GATOR2 activity. This results in GATOR1 activation, Rag GTP hydrolysis, and mTORC1 inhibition. As shown in the figure, the presence of leucine disrupts the interaction between Sestrin2 and GATOR2. This leads to GATOR1 inhibition and mTORC1 activation (4). Similarly, arginine binds to CASTOR1, activating GATOR2 and mTORC1 signaling. LED, lysosomal export domain of GPR155. KICSTOR is a four-protein complex that anchors the GATOR1 complex to the lysosomal membrane, facilitating its function in inhibiting mTORC1 under nutrient-poor conditions (8). The nucleotide state of RagA/B controls the lysosomal association of GATOR in a manner that is competitively antagonized by the N-terminus of the amino acid transporter SLC38A9, which is shown here as “N” surrounded by a circle (23). Figure created de novo; elements of the conceptual framework are based on multiple sources, including (4), (8), (23) and (26).

The key step in the nutrient-dependent activation of mTORC1 involves its nutrient-regulated recruitment to the lysosomal membrane by the active Rag GTPase-Ragulator complex (Fig. 45) (2669–2672). The Ragulator complex includes RagA or RagB GTPases heterodimerized with RagC or RagD and its activity is regulated by the GTP/GDP loading state of these GTPases. The heterodimer is anchored to the lysosomal membrane via the Ragulator scaffold, which is composed of five subunits: LAMTOR1-5. Among these, LAMTOR1 is lipid-modified by myristoylation and palmitoylation, which tethers the complex to the membrane (2669, 2673). In response to nutrients such as arginine, leucine, and cholesterol, the Rag GTPases transition between inactive and active nucleotide states. When nutrients are abundant, RagA/B is GTP-bound and RagC/D is GDP-bound, forming the active Ragulator complex. This conformation enables the recruitment of mTORC1 to the lysosomal membrane, a critical step for its activation (1656, 2674, 2675) (Fig. 45).

When cytosolic amino acid levels are low, the Rag-Ragulator complex is inactivated by GATOR1, which functions as a GTPase-activating protein (GAP) toward the Rag complex (2676). Specifically, under amino acid deprivation, GATOR1 promotes GTP hydrolysis by the RagA subunit, thereby suppressing mTORC1 signaling (2677, 2678). The activity of GATOR1 is negatively regulated by the GATOR2 complex, which includes the KICSTOR complex (8, 2676). The entire signaling machinery is localized to the lysosomal surface through interaction with the Rag-Ragulator complex (23) (see Fig. 45).

Importantly, GATOR1 and GATOR2 do not directly sense amino acids. Instead, amino acid sensors such as Sestrin2 (SESN2) and CASTOR1 transmit nutrient availability signals by modulating the activity of GATOR complexes (2649, 2679, 2680). The structural mechanisms by which these sensors influence GATOR signaling are central to metabolic regulation. Cryo-EM studies have shown that under low arginine conditions, CASTOR1 binds to GATOR2, enabling GATOR1 to suppress mTORC1 activity (2681) (Fig. 45).

Another critical component of lysosomal amino acid sensing is SLC38A9, which has been reported to act as a lysosomal arginine sensor (2682). Upon binding arginine, SLC38A9 promotes the conversion of RagA from its GDP- to GTP-bound form, thereby activating the Rag-GTPase heterodimer and contributing to mTORC1 activation. This function was found to be mediated by the N-terminal domain of SLC38A9 (labeled “N” in Fig. 45) (2647).

However, subsequent studies disrupting endogenous SLC38A9 have shown that mTORC1 can still respond to arginine availability in its absence, suggesting that SLC38A9 is not essential for amino acid-dependent mTORC1 activation. In contrast, disruption of the interaction between the cytosolic arginine sensor CASTOR1 and the GATOR2 complex was found to abolish mTORC1 signaling in response to arginine completely abolished the response of mTORC1 signaling to arginine availability (23).

Thus, SLC38A9 is now thought to functions more as an effector than a primary regulator of Rag GTPases. As previously proposed (2641), its main role may lie in facilitating the arginine-dependent lysosomal efflux of essential amino acids, such as leucine, produced by lysosomal proteolysis. These amino acids then activate mTORC1 via the cytosolic Sestrin1/2-GATOR2 axis (Fig. 45) (23).

Once activated in response to nutrient-rich conditions, mTORC1 phosphorylates the translation regulator eukaryotic translation initiation factor 4E (eIF4E) binding protein 1 (4E-BP1, EIF4EBP1) and S6 kinase 1 (S6K1, RPS6KB1), thereby promoting protein synthesis, lipid biogenesis and metabolism to induce cell growth and proliferation, and in turn reduce cellular catabolism by repressing autophagy (2683).

Several additional mechanisms have been shown to contribute to nutrient-dependent mTORC1 regulation:

As shown in Fig. 45, mTORC1 can sense lysosomal amino acids through an inside-out mechanism that requires V-ATPase (2684). Studies have shown that V-ATPase binds to the Ragulator complex on the lysosomal surface, and that amino acid availability affects the interaction between V-ATPase and Ragulator. Amino acids within the lysosomal lumen trigger conformational changes in V-ATPase that modulate its interaction with the Ragulator complex, thereby initiating the translocation of mTORC1 to the lysosome. Specifically, the study revealed that ATP hydrolysis and the associated V-ATPase rotation are essential for relaying an amino acid signal from the lysosomal lumen to the Rag GTPases, while the ability of V-ATPase to establish the lysosomal proton gradient proved to be dispensable.

The SLC3A2/4F2-SLC7A5/LAT1 heterodimers can deliver leucine directly into lysosomes, triggering the activation of mTORC1 via V-ATPase, as described in the SLC3A2 and SLC7A5 descriptions and illustrated in Fig. 9C. This is enabled after the heterodimer relocates from the plasma membrane to the lysosomal membrane via LAPTM4B (18). The process is particularly important in cancer cells because it increases anabolism and tumor growth.

PAT1/LYAAT1 (SLC36A1) has been reported to physically interact with Rag GTPases (2563, 2685–2687). Similar to SLC38A9, its interaction with Rag GTPase is enhanced by the addition of arginine following amino acid starvation (2563). Leucine, previously shown to activate mTORC1 through a Rag GTPase-dependent mechanism (2688), also increased the expression of both SLC38A9 and SLC36A1, contributing to mTORC1 activation. However, whether PAT1/LYAAT1 (SLC 36A1) functions as a direct amino acid sensor, beyond its role as a neutral amino acid transporter, remains unclear (2563).

SNAT2 (SLC38A2) has been identified as a sensor involved in activating mTORC1, despite being primarily localized to the plasma membrane (2605, 2606). Moreover, as mentioned in the SLC38A7 description, its encoded protein SNAT7 has been reported to regulate mTORC1 following macropinocytosis of extracellular proteins (2637). Together, these findings suggest an expanded version of the nutrisome, which involves the coordinated action of amino acid transporters and regulatory proteins at both the lysosomal and plasma membranes. The nutrisome concept has been discussed in the context of different amino acid transporters, such as SLC38A9 and PAT1 (2689, 2690). PAT1 and SLC38A9 work together at the lysosomal surface to mediate the recruitment and activation of mTORC1 in a manner dependent on amino acids, while SNAT2 (SLC38A2, System A) acts as at the plasma membrane, conveying extracellular amino acid status to the mTORC1 pathway (2606).

The SLC-like GPR155 (LYCHOS) functions as a cholesterol sensor in lysosomes, as discussed in Section 10. When cholesterol is abundant, it binds to the N-terminal domain of GPR155 (Fig. 45) (2691). This triggers GPR155 to sequester GATOR1 via its cytosolic LED (lysosomal export domain). This results in GATOR1 inhibition and favors RagA/B activation, which promotes mTORC1 recruitment to the lysosome. Thus, GPR155 (LYCHOS) is a lysosomal cholesterol sensor that is directly linked to mTORC1 regulation. Additionally, NPC1 (SLC65A1) plays an important role in ensuring that lysosomes are not overloaded with cholesterol (see Fig. 45). As described in the SLC65 family description, loss of NPC1 function results in Niemann-Pick disease type C (NPC), a devastating neurodegenerative disease caused by the accumulation of unesterified cholesterol in late endosomes/lysosomes due to defective lysosomal cholesterol efflux caused by NPC1 mutations.

Fig. 45 illustrates how the different nutrient-sensing mechanisms converge to regulate the mTORC1 signaling pathway, promoting anabolic growth under nutrient-rich conditions and inducing catabolism and autophagy during nutrient scarcity. The following is a list of the individual steps and the additional details associated with each one:

1. The cytosolic amino acid sensors include Sestrin2 (for leucine) and CASTOR1 (for arginine). When amino acid levels are sufficient, these sensors bind their respective ligands and release their inhibitory hold on GATOR2, thereby activating it. For example, under leucine deprivation, hypophosphorylated Sestrin2 binds to and inhibits GATOR2. Upon leucine binding, Sestrin2 undergoes a conformational change and dissociates from GATOR2, relieving the inhibition and allowing GATOR2 to activate downstream signaling (2692, 2693).

2. Activated GATOR2 represses the GAP activity of GATOR1 toward RagA/B (4). With GATOR1 inhibited, RagA/B accumulate in the GTP-bound form, while RagC/D becomes GDP-bound, thereby creating the active Rag heterodimer that recruits mTORC1 to the lysosomal surface.

3. The GTP-bound RagA/B also acts as a platform enabling Rheb-GTP (a small GTPase anchored on the lysosome) to access and activate mTORC1’s kinase function (2694).

4. SLC38A9 binds arginine in the lysosome lumen and mediates arginine-gated leucine export (23, 2682). This increases cytosolic leucine levels, which feed into Sestrin2–GATOR2–GATOR1 regulation. SLC38A9 also interacts physically with Ragulator/Rag GTPases to facilitate mTORC1 recruitment.

5. Cholesterol sensing occurs via GPR155 (LYCHOS), which binds to the N-terminal domain of GPR155 when cholesterol is abundant (2691). This triggers GPR155 to sequester GATOR1 via its cytosolic LED loop, thereby eliminating GATOR1 inhibition. This, in turn, favors RagA/B activation and promotes the recruitment of mTORC1 to the lysosome.

6. Lysosomal luminal amino acid sensing occurs in a process mediated by V-ATPase (2684). Presence of amino acids such as leucine inside the lysosome results in a conformational change in V-ATPase that modifies its interaction with Ragulator, facilitating mTORC1 docking, which is then subsequently activated by Rheb-GTP.

A variant of SLC38A9, rs4865615-C (S182T), has been identified and found to be significantly associated with a reduced risk of chronic kidney disease (2635). The kidney plays a central role in arginine metabolism as the proximal tubule is a major site of arginine production from citrulline (2695). However, the mechanism by which this missense mutation affects SLC38A9 function and reduces the risk of chronic kidney disease remains unclear.

SLC38A12 - Orphan transporter: SLC38A12 (TMEM104) is widely expressed in human tissues. Considering its distant phylogenetic relationship with the SLC38 family, it may represent an amino acid transporter (7). In a recent study, it was proposed to be a lysosomal transporter (2696). Further experimental validation of these findings as well as a detailed functional characterization will be important.

Orphan transporters family members (4)

SLC38A8, SLC38A10, SLC38A11, SLC38A12

HGNC update

TMEM104 has been renamed as SLC38A12.

SLC39 Metal ion transporter family (2.A.5/Zip/ZIP)

Discovery: The first members of the ZIP (Zrt, Irt-like proteins; SLC39 family) identified were the zinc-regulated transporter Zrt1 from Saccharomyces cerevisiae (2697) and the iron-regulated transporter Irt1 from Arabidopsis thaliana (165). To identify Irt1, a cDNA library from Arabidopsis thaliana was screened for clones that could restore iron-limited growth to a yeast strain lacking iron uptake (165).

Gene family members (14):
SLC39A1 (ZIP1)	SLC39A5 (ZIP5)	SLC39A9 (ZIP9)	SLC39A13 (ZIP13)
SLC39A2 (ZIP2)	SLC39A6 (ZIP6)	SLC39A10 (ZIP10)	SLC39A14 (ZIP14)
SLC39A3 (ZIP3)	SLC39A7 (ZIP7)	SLC39A11 (ZIP11)
SLC39A4 (ZIP4)	SLC39A8 (ZIP8)	SLC39A12 (ZIP12)

Molecular aspects, physiological roles and links to disease

Members of this family transport divalent metal ions such as Zn²⁺, Fe²⁺, Mn²⁺, and/or Cd²⁺. Several members of the ZIP/SLC39 family primarily transport Zn²⁺. The SLC39 family has 14 members (Fig. 46) and belongs to the zinc (Zn²⁺)-iron (Fe²⁺) permease (ZIP) family (TC 2.A.5), which incorporates the ZIP structural fold. The transporter core of ZIP/SLC39 family members consists of 8 TMHs, where the first 4 TMHs are symmetrically related to the last 4 TMHs by a pseudo-two fold axis. N-terminus and C-terminus are outside and there is a large histidine-rich intracellular loop between TMH3 and TMH4, known as the L2 loop (see Section 8 for further details).

Fig. 46. Phylogenetic tree of the SLC39 family.

As previously mentioned in the SLC30 family description, zinc is an essential trace metal necessary for various functions in living organisms, including growth, development, immune response, as well as neurological and endocrine functions. Excessive amounts of zinc are toxic, hence it is crucial for cellular zinc levels to be tightly regulated by the two transporter families, namely SLC30, which primarily exports zinc, and SLC39, which facilitates zinc uptake or the exit of zinc from intracellular organelles (324).

The 14 human ZIP family proteins have diverse functions in various cellular processes and are found in a wide range of tissues. Many of these proteins are situated on the plasma membrane and initiate metal ion uptake in cells. Their levels on the cell surface are often regulated by protein trafficking in response to the extracellular concentration of the divalent ions they transport. For example, this mechanism regulates the surface level of ZIP1 (SLC39A1) in cells, and depletion of zinc results in increased ZIP1 surface expression due to reduced endocytosis and degradation (2698). Some members of the ZIP/SLC39 family are regulated by hormones such as prolactin and testosterone (ZIP1/SLC39A1), and one of them, ZIP9 (SLC39A9), is a hormone receptor itself.

The SLC39/ZIP family can be divided into 4 subfamilies based on phylogenetic analysis and functional criteria (2699).

Subfamily I: SLC39A9 (ZIP9).

Subfamily II: SLC39A1 (ZIP1), SLC39A2 (ZIP2) and SLC39A3 (ZIP3).

Subfamily III – the GufA subfamily: SLC39A11 (ZIP11) is the only human member of this subfamily, named GufA after a prokaryotic protein with unknown function from Myxococcus xanthus (MxGufA).

Subfamily IV -the LIV-1 subfamily: SLC39A4 (ZIP4) SLC39A5 (ZIP5), SLC39A6 (ZIP6), SLC39A7 (ZIP7), SLC39A8 (ZIP8), SLC39A10 (ZIP10), SLC39A12 (ZIP12), SLC39A13 (ZIP13), SLC39A14 (ZIP14). The name LIV-1 originates from the discovery of genes at the University of Liverpool with a potential role in breast cancer (2700).

Description of the members of each subfamily:

Subfamily I

Members in this subfamily are mainly from fungal and plant sources and SLC39A9 (ZIP9) is the sole human member of subfamily I.

SLC39A9: Interestingly, ZIP9 (SLC39A9) functions as a membrane androgen receptor coupled to G proteins that directly regulates zinc homeostasis by controlling zinc transport (2701, 2702). Thus, ZIP9 (SLC39A9) functions both as a membrane androgen receptor that signals through G proteins and as an androgen-dependent zinc transporter. It has all the properties of a specific membrane androgen receptor and initiates testosterone-induced apoptosis in several nuclear androgen receptor-negative cancer cell lines by activating an inhibitory G-protein and intracellular signaling pathways and by increasing intracellular zinc levels. The intracellular L2 loop is significantly shorter than in other ZIP proteins and does not possess any expected endocytosis motifs (2698). The diastereomer (-)-epicatechin, a flavonoid found in high concentrations in foods such as green tea and cocoa, exerts beneficial anti-tumorigenic and apoptotic effects in breast and prostate cancer cells by binding to ZIP9 (SLC39A9) (2703).

Subfamily II

Three zinc transporters belong to this subfamily in humans: SLC39A1 (ZIP1), SLC39A2 (ZIP2) and SLC39A3 (ZIP3). Like members of subfamily I, the cytoplasmic L2 loop regions of these proteins are variable. However, there is a conserved His-rich part in this region. Moreover, between TMH 4 and 5, there are two highly conserved regions that contain conserved His residues in all family members. These two regions are believed to play an important role in zinc transport. The mechanism of zinc transport used by SLC39 proteins is still unclear. Zinc uptake by ZIP1 (SLC39A1) and ZIP2 (SLC39A2) was found to be energy-independent and zinc uptake was not dependent on K⁺ or Na⁺ gradients either (2704–2706). But ZIP2 (SLC39A2) activity was stimulated by HCO₃^- suggesting a Zn²⁺/HCO₃^- symport mechanism. Interestingly, the functional activity of ZIP2 increases at acidic pH (2706). As discussed below under ZIP8 (SLC39A8), there may be other coupling mechanisms for ZIP transporters.

SLC39A1: ZIP1 (SLC39A1) is ubiquitously expressed and traffics between intracellular organelles and the cell surface in response to extracellular zinc. The mechanisms for this trafficking process have been extensively reviewed in detail (2698, 2707). Briefly, most ZIP proteins, with the exception of ZIP7 and ZIP13, reside on the cell surface for at least part of their time. To regulate ZIP transporter expression on the cell surface based on extracellular levels of divalent metal ions, endosomal trafficking is a crucial factor. Thus, the cell surface levels rely on both the rate of recycling from Golgi compartments and/or endosomes back to the surface, and the rate of degradation. However, even though the regulatory mechanisms based on zinc availability are well known, our knowledge of the molecular mechanisms underlying the regulation and whether it also takes place on the basis of intracellular Zn²⁺ levels is still incomplete. In prostate epithelial cells, for example, functionality depends on high levels of intracellular zinc, which requires prominent expression of ZIP1 (SLC39A1) on the cell surface. This inhibits mitochondrial aconitase (2708), which limits citrate oxidation and promotes the secretion of high levels of citrate to ensure sperm viability (2418, 2419). This is why in prostate cancer, ZIP1 functions as a tumor suppressor gene (2709). In other types of cancer such as glioma, the situation is the opposite, and ZIP1 is upregulated to promote tumor progression (2710). Additionally, the expression of ZIP1 contributes to the progression and immunosuppression of hepatocellular carcinoma (2711). Thus, in prostate cancer, drugs that upregulate ZIP1 would be beneficial, whereas in other cancers, drugs that inhibit ZIP1 would be needed.

ZIP1 (SLC39A1) together with ZIP3 (SLC39A3) also play an important role in the hippocampus in the brain where they are expressed on distinct neuronal populations in the CA3 region (2712). As already noted in the SLC30 family description, ZnT3 (SLC30A3) serves as an important transporter of zinc into synaptic vesicles of a subset of glutamatergic zinc-enriched neurons and Zn²⁺ released together with glutamate at the synaptic cleft inhibits NMDA receptors and thus the excitability of the hippocampal neuronal circuit. To this end, ZIP1 mediates zinc influx of released Zn²⁺ into postsynaptic cells and ZIP3 (SLC39A3) is responsible for zinc reuptake from the synapse into dentate granule cells. Thus, neuronal zinc toxicity and degeneration can be modulated by regulation of the function of specific zinc transporters (2712).

SLC39A2: ZIP2 (SLC39A2) is relatively widely expressed, with highest expression in skin and male tissues. While ZIP2 may be regulated by protein trafficking similar to ZIP1, its expression may also be regulated at the transcriptional level in response to cellular zinc depletion. For example, in the mouse heart, in response to ischemia/reperfusion injury, Zip2 expression increases due to phosphorylation of the transcription factor STAT3, compensating for cellular zinc loss (2713). The enhanced zinc uptake by Zip2 (Slc39a2) is important to protect the heart from cardiac damage caused by ischemia/reperfusion injury.

As already alluded to, ZIP2 (SLC39A2), similar to ZIP1 (SLC39A1), were found to be energy independent. Studies in our laboratory further showed that the Zn²⁺ transport process is electroneutral, independent of ATP hydrolysis, independent of Na⁺ and K⁺ gradients, stimulated in the presence of extracellular HCO₃⁻ and inhibited by lowering the extracellular pH (2706). Additional studies provided new insight into the transport mechanism (2714), whereby E179 & H175 (TMH4), H202 (TMH5), E276 (TMH7) were shown to directly bind the divalent metal ion substrate. This single divalent metal ion binding site is a unique property of SLC39 subfamily II (SLC39A1, SLC39A2 and SLC39A3), as opposed to SLC39 subfamily IV / LIV-1. Apart from the residues near the proposed substrate-binding site, there are several other residues that affect the pH sensitivity of transport in ZIP2, the most prominent one being His-63 residing in a bulge along TMH2, the loss of which renders the transporter completely pH-insensitive. The findings agree with an alternating-access transport mechanism.

SLC39A3: ZIP3 (SLC39A3) is ubiquitously expressed. Similar to ZIP1, ZIP3 has been shown to traffic in a zinc-dependent manner, thereby controlling plasma membrane expression (2715). In addition, as highlighted in the description of ZIP1 (SLC39A1), it plays an important role in neurons (2712). It also plays an important role in the mammary glands. There, ZIP3 is not involved in the acquisition of Zn²⁺ from the maternal circulation for secretion into milk, but rather facilitates the reuptake and cellular retention of Zn²⁺ in the mammary gland from the previously secreted milk pool, thereby regulating cellular function.

Subfamily III - the GufA subfamily

This subfamily is also called GufA (gene of unknown function A), a group of prokaryote and eukaryote proteins related to the originally identified GufA gene of Myxococcus xanthus (MxGufA) (2699). Only one human zinc transporter belongs to this subfamily: ZIP11 (SLC39A11). It is primarily found in the nucleus and the Golgi apparatus.

SLC39A11: ZIP11 (SLC39A11) is a member of the GufA subgroup of ZIP transporters. It is expressed at high levels in the testes, stomach, ileum, and cecum, with lower expression found in the liver, duodenum, jejunum, and colon (2716). Analysis of isolated stomach and colon tissues from mice revealed that ZIP11 (Slc39a11) localizes to the nucleus in mammalian cells (2717, 2718). ZIP11 (SLC39A11) is partially regulated by dietary zinc intake and plays a vital role in zinc homeostasis, ensuring proper mucosal integrity and function. Adequate nuclear Zn²⁺ levels are necessary for proper gene expression, which is facilitated by ZIP11 (SLC39A11). The SLC39A11 gene has metal responsive elements (MREs) that allow its expression to be regulated by Zn²⁺ via the zinc finger protein MTF-1 (metal-responsive transcription factor-1), which binds to the MREs (2719). In the gastrointestinal tract, Zn²⁺ deficiency modestly downregulates ZIP11, which subsequently induces the expression of ZIP4, taking over Zn²⁺ uptake from the colon in response to Zn²⁺ deficiency (2718). Accumulating evidence suggests that ZIP11 (SLC39A11) is linked to the development of various cancer types (2717). For instance, studies have revealed that genetic variations of the nuclear zinc transporter ZIP11 (SLC39A11) influence the carcinogenic potential of ovarian cancer cells (2716).

Subfamily IV - the LIV-1 subfamily਀

This subfamily consists of nine members that distinguish themselves from other ZIPs for the following reasons (2700): a) The proteins feature a metalloproteinase-like motif “HEXPHEXGD” within TMH5 and numerous other histidine-rich regions (2720). In metalloproteinases, the corresponding consensus sequence of this motif (HEXXHXXG) represents the catalytic zinc-binding site that is crucial for their proteolytic activity (2720, 2721). Additionally, the N-terminal ectodomain and extracellular loop between TMH2 and TMH3 contain additional histidine residues, some of which bear resemblance to prions (see below). The name of the gene group “LIV-1” originates from its first identified gene, which is estrogen-induced in MCF-7 and ZR-75 breast cancer cells (2722). Later on, it was discovered that LIV-1 is equivalent to the zinc transporter SLC39A6 (2723).

Based on the phylogenetic trees of the SLC39 family, LIV-1 proteins can be categorized into the following branches.

LIV-1 branch a

This branch consists of SLC39A8 and SLC39A14, which play an important role in maintaining Mn²⁺ homeostasis in the body. These transporters also accept Zn²⁺ and Cd²⁺ as a transport substrate. In addition, they can transport Fe²⁺, which is likely to be relevant in iron overload diseases where non-transferrin bound iron occurs in the plasma. Under these conditions, ZIP8 (SLC39A8) or ZIP14 (SLC39A14) can take up iron into tissues, leading to toxic accumulation of iron. Compared to other members of the ZIP/SLC39 family, ZIP8 and ZIP14 have a glutamic acid substitution for the initial histidine of the “HEXPHEXGD” motif and a glutamine substitution for the conserved histidine in TM2 (2700). These sequence changes may account for the extended range of divalent metal ion substrates of ZIP8 and ZIP14.

SLC39A8: ZIP8 (SLC39A8) is a metal ion transporter that transports multiple divalent metal ions, including Mn²⁺, Zn²⁺, Cd²⁺, Fe²⁺, and Co²⁺ (2724–2726). ZIP8 is abundantly expressed in lung, placenta, salivary gland, thymus, kidney, intestine, and brain (2725, 2726). ZIP8 is typically found as an N-glycosylated protein on the cell surface (2727), though it has also been reported to be expressed in the membranes of intracellular organelles, such as lysosomes (2728). In the case of polarized cells, such as intestinal (ileum), kidney proximal tubule, and lung epithelial cells, it is found on the apical side (2729–2731).

ZIP8 was initially identified as a Zn²⁺ transporter subsequent to its induction in monocytes by microbial challenge, resulting in intracellular zinc accumulation (2732). Bacteria-mediated induction of ZIP8 in human monocytes facilitates zinc influx from the extracellular environment (Pyle, 2017, #4658). ZIP8 has been shown to regulate the innate immune system through NF-κB activity in macrophages and monocytes, thereby influencing immune function during mycobacterial infection and inflammation (2727). ZIP8 has also been reported to be expressed in human T cell lysosomes, where it releases stored zinc into the cytosol, thereby inhibiting calcineurin and increasing interferon-gamma (IFN-γ) expression (Aydemir, 2009, #4341).

ZIP8-mediated metal transport has been shown to be electroneutral by coupling with the cotransport of HCO₃^- (2347, 2733). In ZIP8-expressing Xenopus oocytes (2734), electrogenicity studies showed an influx of two HCO₃^- anions for each divalent metal ion. Selenite (HSeO₃^-), an inorganic form of selenium of pharmaceutical importance, has also been shown to serve as a co-transported anion for ZIP8 (2735). Studies investigating the physiological role of ZIP8 (SLC39A8) have used Slc39a8 inducible KO mice because constitutive inactivation of Slc39a8 is embryonic lethal (2736).

Based on observations in patients with rare SLC39A8 loss-of-function mutations (2737, 2738) and transgenic mice lacking ZIP8 function (see below), it was found that Mn²⁺ is a physiologically important transport substrate of ZIP8 (SLC39A8) (2739–2741). Consistent with this, patients with SLC39A8 mutations and Slc39a8 iKO mice exhibit low blood levels of Mn²⁺. In addition, they have reduced Mn²⁺ levels in liver, kidney, brain and heart, indicating systemic whole-body Mn²⁺ deficiency (2739). Hepatocyte-specific inactivation of Slc39a8 replicated the systemic Mn²⁺ deficiency and highlighted the essential role of hepatocyte SLC39A8 in whole-body Mn²⁺ homeostasis (2739). In intestinal epithelial cell-specific knockout (Slc39a8-IEC KO) mice, SLC39A8 was shown to play an important role in Mn²⁺ uptake across the apical membrane of enterocytes, particularly in ileum and colon (2742), followed by exit at the basolateral membrane via ferroportin (SLC40A1) (2743).

Enterocytes and hepatocytes work together to maintain systemic manganese homeostasis. This is achieved through the functions of ZIP8 (SLC39A8), ZIP14 (SLC39A14) and ZnT10 (SLC30A10) (2744). As mentioned above, in enterocytes, manganese is taken up apically via ZIP8 and exits basolaterally, presumably via ferroportin (SLC40A1), although the details remain to be clarified (Fig. 22). Conversely, ZIP14 facilitates the reuptake of manganese from the blood into the enterocytes at the basolateral membrane. At the luminal side, manganese can then be released into the intestinal lumen via ZnT10 or remain inside the enterocytes until they are sloughed off (Fig. 22). In hepatocytes, ZIP14 imports circulating manganese into the hepatocyte at the basolateral (sinusoidal) membrane, while ZnT10 exports manganese into the bile at the apical canalicular membrane (2744). ZIP8 retrieves manganese from the bile, thereby increasing hepatic manganese stores.

In the kidney, ZIP8 is highly expressed in the S3 proximal tubule segment where it contributes to the reabsorption of filtered divalent metal ions (2745). A clinical study showed that the urinary excretion of Mn²⁺ in patients with rare SLC39A8 loss-of-function mutations during magnesium therapy was much higher than in normal individuals (2746). This observation is consistent with Mn²⁺ uptake across the apical membrane of renal proximal tubule S3 segments playing an important role in maintaining systemic Mn²⁺ homeostasis.

While Mn²⁺ is an essential micronutrient required for basic cell functions and vital physiological processes, elevated levels of Mn²⁺ are toxic to cells because they increase oxidative stress, impair mitochondrial function, and promote cell death. Exposure to high levels of manganese has been shown to lead to Mn²⁺ accumulation in the brain and a Parkinson-like disease. For these reasons, manganese homeostasis must be tightly controlled to avoid both excess and insufficiency.

Dysfunction of ZIP8 (SLC39A8) has been shown to lead to a number of fatal diseases and unique traits. Several rare loss-of-function mutations of SLC39A8 have been shown to cause systemic manganese deficiency (2737, 2738) due to impaired ZIP8 protein expression (2727, 2747, 2748).

In addition, GWAS and other studies have shown that a genetic variant of SLC39A8, rs13107325 (A391T), is associated with numerous pathological conditions such as schizophrenia, Crohn disease, scoliosis, and obesity, many of which are reported to be the result of defective manganese homeostasis (2725, 2738, 2740, 2742, 2744, 2749–2752). The transport properties of this variant have been investigated by several groups using rs13107325 knock-in (KI) mice. These mice showed alterations in metal ion homeostasis consistent with GWAS and reduced arterial blood pressure (2725). In addition, the Slc39a8 rs13107325 KI mice exhibited remarkable insulin resistance and were protected from elevated blood glucose when challenged with dietary sucrose supplementation (2725).

Another study with Slc39a8 rs13107325 KI mice revealed abnormal Mn²⁺ homeostasis with reduced liver and kidney Mn²⁺, reduced blood Mn²⁺ and increased biliary Mn²⁺ excretion (2740). Impaired Mn²⁺-dependent glycosyltransferase activity has been shown to be a reason for increased susceptibility to the development of Crohn’s disease (2740, 2742, 2751).

Further studies using the same type of knock-in mouse model have demonstrated an association of this variant with schizophrenia due to changes in protein glycosylation in the brain (2750). The glycosylation of Asn residues in glycoproteins (N-glycosylation) was most significantly affected, resulting in a substantial proportion of cortical glycoproteins being abnormally N-glycosylated, particularly glycoproteins previously implicated in schizophrenia development, such as cell adhesion molecules and neurotransmitter receptors.

To further investigate the role of ZIP8 (SLC39A8) in the brain, additional transgenic studies in mice have been performed (2726). Since the use of Slc39a8-inducible KO mice is not suitable to screen for extrahepatic roles of ZIP8 (SLC39A8) in tissue Mn²⁺ uptake/accumulation because, as mentioned above, loss of SLC39A8 in hepatocytes results in systemic Mn²⁺ deficiency, Slc39a8-inducible KO mice were crossed with Slc39a14 KO mice (2726). ZIP8 normally facilitates the uptake of Mn²⁺ from plasma into the cell, whereas in the hepatocytes, it recovers Mn²⁺ from the bile. Therefore, the combined Slc39a14 KO / Slc39a8 iKO mice do not develop systemic Mn²⁺ deficiency due to loss of SLC39A8 in hepatocytes, but instead exhibit hypermanganesemia and Mn²⁺ overload in extrahepatic tissues. These mice exhibited hypermanganesemia with increased Mn²⁺ levels in bone and kidney while decreased Mn²⁺ levels were found in the brain (2726). The study demonstrates the important role for ZIP8 in the accumulation of Mn²⁺ in the brain. Subsequent studies showed that SLC39A8 is required for Mn²⁺ uptake by the brain across the BBB (2726). The studies also suggest that ZIP8 (SLC39A8) is a potential target for reducing Mn²⁺ uptake and accumulation in the brain.

Although ZIP8 plays a critical role in whole-body Mn²⁺ homeostasis, its ability to transport Zn²⁺ is also of fundamental importance in tissues. For example, in the lung, where ZIP8 is highly expressed, it has been reported to play an important role in alveolar epithelial repair after injury by mediating zinc transport into alveolar type 2 epithelial cells (AT2s) (2753) (see Fig. 33, bottom part). These cells are required for alveolar epithelial regeneration to maintain alveolar architecture and function. The repair process requires the proliferation of AT2s, which then differentiate into alveolar type 1 epithelial cells (AT1s). For this purpose, Zn²⁺ is taken up via ZIP8 in AT2s to facilitate sirtuin signaling. SIRT1 is a NAD⁺-dependent deacetylase that regulates inflammation and stress resistance, as well as age-related lung disease, which is associated with downregulated expression of SIRT1 (2753). Failure of AT2 regeneration leads to progressive lung fibrosis, a key feature of idiopathic pulmonary fibrosis (IPF). Consistent with this, deficiency of ZIP8 (SLC39A8) expression was identified in AT2s from both idiopathic pulmonary fibrosis lungs and lungs of aged mice, which was associated with impaired AT2 renewal capacity and increased lung fibrosis. It has also been shown that in the distal air space of the lung, lipopolysaccharide-induced inflammation induces ZIP8 to mediate metal ion uptake into lung tissue (2754).

Both ZIP8 (SLC39A8) and ZIP14 (SLC39A14) (see below) can transport Fe²⁺ (2755). Therefore, they are thought to play a role in iron overload diseases, as these transporters can transport non-transferrin bound iron (NTBI) from the circulating blood to major organs in an uncontrolled manner, resulting in damage to organs such as the liver, heart and pancreas (2756). Using the global inducible ZIP8 knockout mouse model, the role of ZIP8 in steady-state iron homeostasis was investigated, and an unexpected phenotype of elevated spleen iron levels and decreased serum iron was observed in SLC39A8 KO mice, suggesting that ZIP8 plays an important role in iron recycling (2754).

ZIP8 (SLC39A8) has been shown to mediate microglial ferroptosis through its role in iron uptake (2757). The role of the iron reductase STEAP3 and the metal ion transporter ZIP8 (SLC39A8) in mediating ferroptosis in microglia was investigated in rats after neurotoxicity was induced by exposure to lead and cadmium. Exposure to these heavy metals significantly increased iron accumulation, oxidative stress, and inflammatory responses in rat brain tissue, accompanied by abnormal activation of microglia. Inhibiting STEAP3 and ZIP8 effectively reduced inflammation, suggesting that they may play a crucial role in microglia-mediated neurotoxicity (2757). This indicates that SLC39A8 could be a target for treating neurodegenerative diseases.

SLC39A14: ZIP14 (SLC39A14) is a divalent metal ion transporter with a substrate specificity similar to that of ZIP8 (SLC39A8) (2758, 2759). Again, the importance of ZIP14 (SLC39A14) in the regulation of body manganese levels was recognized by the discovery of mutations in humans (2760) (see below).

ZIP14 is expressed in the intestine, liver, and kidneys, where it plays a key role in regulating systemic Mn²⁺ homeostasis. Studies in Slc39a14 knockout mice revealed that ZIP14 is critical for the regulation of systemic Mn²⁺ metal ion homeostasis (2741, 2761).

ZIP14 (SLC39A14) expression is somewhat tissue-specific with the highest abundance in jejunum, liver, heart and kidney according to studies in mice (2761). According to the HPA, the highest expression is in liver, small intestine and pancreas. As mentioned in the description of ZIP8 (SLC39A8), in the intestine and liver, ZIP14 is expressed on the basolateral and sinusoidal membranes, respectively, where it transports Mn²⁺ from the blood into the enterocytes and hepatocytes. ZnT10 (SLC30A10) is an apical metal ion transporter that exports Mn²⁺ from enterocytes into the intestinal lumen and from hepatocytes into the bile (2744). ZIP14 has also been reported to be expressed in renal proximal tubule segments S1, S2 and S3 where it likely contributes to the renal uptake of divalent metal ions from the blood across the basolateral membrane into the renal proximal tubule cells.

Thus, loss of ZIP14 function leads to systemic manganese overload, which primarily affects the central nervous system and causes neurological disorders. In support of this, patients carrying loss-of-function mutations in SLC39A14 develop manganese toxicity and early-onset dystonia due to manganese hyperaccumulation in the brain (2346, 2762–2765) and Slc39a14^-/- mice exhibited blood and brain manganese loads that were more than 10 times the normal level (2766–2768). To further investigate the role of ZIP8 in Mn²⁺ homeostasis, studies were performed in mice with combined inactivation of intestinal and hepatic Slc39a14 (2769). The studies showed that although deletion of intestinal ZIP14 only moderately increased systemic manganese burden and liver-specific Slc39a14 knockout (Slc39a14-L-KO) mice did not show manganese hyperaccumulation in blood or brain despite significantly reduced liver manganese (2741, 2770), deletion of both intestinal and hepatic Slc39a8 greatly exacerbated the body’s manganese burden.

It has also been reported that ZIP14 mediates the uptake of non-transferrin bound iron (NTBI) into cells and that it may play a role in iron metabolism in hepatocytes, where this transporter is abundantly expressed. Since NTBIs are frequently found in the plasma of patients with hemochromatosis and transfusional iron overload, ZIP14 uptake of NTBI may contribute to the hepatic iron overload characteristic of hemochromatosis (2771).

ZIP14 (SLC39A14) has been shown to play a role in esophageal squamous cell carcinoma (ESCC). Increased expression of ZIP14 correlates with tumor progression and poor prognosis in ESCC tissues (2772) These results suggest that ZIP14 (SLC39A14) could be used as a biomarker and a potential therapeutic target in esophageal squamous cell carcinoma (ESCC). Mechanistic analyses have shown that the procarcinogenic effects of SLC39A14 are mediated by activation of the PI3K/Akt/mTOR signaling pathway.

In conclusion, despite the critical roles of ZIP8 and ZIP14 in maintaining metal ion balance and the far-reaching implications of their dysfunction in various pathologies, limited efforts have been made to develop drugs that specifically target them. Nevertheless, a first class of inhibitors for ZIP8 has been identified (2773).

LIV-1 branch b

ZIP5 (SLC39A5), ZIP6 (SLC39A6) and ZIP10 (SLC39A10) are the members of this branch. They exhibit prion-like amino acid sequences in their N-termini (2774). Prions are GPI (glycosylphosphatidylinositol)-anchored glycoproteins that are abundant on the surface of neurons (2775). Upon binding to extracellular Cu²⁺ or Zn²⁺ (but not Mn²⁺), they undergo rapid endocytosis (2776). The N-termini of ZIP6 and ZIP10 are predicted to have a structure similar to prion proteins (2777), with a corresponding set of essential disulfide bonds. Furthermore, the CPALLY pattern in the LIV-1 family (see note below) is located in the same region as the disulfide bonds in prion proteins (2774), further suggesting common functional aspects between ZIP transporters and prions.

Members of all LIV-1 branches, except LIV-1 branch c, have the distinctive “CPALLY” motif at their N-terminus (2700, 2720). This motif consists of 40 residues and aligns with the consensus sequence “C (X26) CPALLYQ (X5) C”, with three conserved cysteine residues.

SLC39A5: ZIP5 (SLC39A5) is a zinc transporter that plays a crucial role in zinc homeostasis and protects against zinc toxicity in the pancreas (2778). It is mainly expressed in small intestine, kidney, liver and pancreas.

ZIP5 plays an important role in the pancreas as it protects this organ from zinc toxicity by functioning in acinar cells and helping to prevent zinc-induced acute pancreatitis (2778). Furthermore, because pancreatic secretions that empty into the intestinal lumen contain significant amounts of zinc, the pancreas likely plays an important role in zinc homeostasis by serving as a primary route of zinc excretion (2779).

The other potential route for the release of endogenous zinc into the gastrointestinal tract is through the serosal-to-mucosal transport of Zn²⁺, which eventually releases it into the intestinal lumen (2779). ZIP5 (SLC39A5) is localized to the basolateral membrane of enterocytes and may contribute to intestinal zinc excretion by facilitating zinc uptake from the serosal side (2780). Thus, while apical ZIP4 (SLC39A4) and basolateral ZnT1 (SLC30A1) play important roles in intestinal zinc uptake (Fig. 22), ZIP5 is localized to the basolateral side of enterocytes, where it may be involved in zinc excretion. It remains to be determined whether zinc (Zn²⁺) taken up by ZnT1 across the basolateral membrane exits into the intestinal lumen via the reported splice variant of ZnT5 (SLC30A5), referred to as ZNT5B (2400, 2401), or whether it is sequestered in metallothioneins and eliminated in sloughed enterocytes (see the SLC30A5 description). Alternatively, ZIP5 residing within the basolateral membrane of enterocytes could monitor the body’s zinc status (2781). ZIP5 function could activate the metal response element-binding transcription factor 1 (MTF-1), which regulates the expression of genes such as metallothioneins that help sequester and store zinc (see Fig. 22).

A subsequent study demonstrated that genetic inactivation of SLC39A5 leads to elevated serum Zn²⁺ levels and improved liver metabolism and hyperglycemia in obesogenic settings (2782). The study has used genetic analysis of large human cohorts, followed by comprehensive analysis of a mouse Slc39a5 knockout mutant, to show that SLC39A5 affects hepatic lipid handling through AMPK signaling. This alteration has been shown to be associated with a reduced risk of T2D.

Specifically, in mice, loss of Slc39a5 resulted in elevated hepatic zinc, lower glucose levels, and has protective effects in models of congenital and diet-induced obesity. The effects are proposed to be mediated by the activation of hepatic AMPK and AKT signaling, in part due to Zn²⁺-mediated inhibition of hepatic protein phosphatase activity. The study therefore uncovers a mechanistic basis for Zn²⁺-induced liver protection and indicates that ZIP5 (SLC39A5) inhibition may hold therapeutic potential for the treatment of patients with T2D as well as NAFLD, whereby possible adverse effect of the pancreatic response to ZIP5 inhibition would need to be considered.

SLC39A6: ZIP6 (SLC39A6) is ubiquitously expressed, with highest mRNA levels in male and female reproductive tissues, according to the HPA. It was reported to be an important component of the lymphocyte activation machinery, acting as the transporter responsible for zinc entry into lymphocytes, ensuring the proper function of the cellular activation machinery (2783), as well as playing a critical role in T-cell development (2784).

SLC39A10: ZIP10 (SLC39A10) is prominently expressed in cerebral cortex, thyroid, respiratory system, prostate, endometrium, oviduct and placenta and at lower levels in kidney and liver.

The zinc transporter ZIP10 (SLC39A10) was originally identified as a putative zinc importer in the apical membrane of rat kidney and its expression was shown to be regulated by zinc (2785). Subsequently, it was also found to play an important role in embryonic hematopoiesis (2786) by protecting hematopoiesis from zinc deficiency-induced necroptosis. Thus, as a zinc transporter, SLC39A10 promotes fetal hematopoietic stem/progenitor cell development and survival, whereas loss of the transporter results in impaired hematopoiesis that can be rescued by zinc supplementation.

LIV-1 branch c

SLC39A7: ZIP7 (SLC39A7) is ubiquitously expressed, has a highly conserved histidine-rich N-terminal region, and is involved in zinc transport in the endoplasmic reticulum (2397). There it serves as a gatekeeper in controlling the release of Zn²⁺ from intracellular ER Zn²⁺ stores into the cytosol after post-translational activation by phosphorylation at residues S275 and S276 (2787). Release of Zn²⁺ into the cytosol has been proposed to drive key signaling pathways such as MAPK, mTOR and PI3K/AKT, which play an important role in regulating cell growth and ensuring survival and are often over-activated in cancer (2787).

ZIP7 (SLC39A7) is a unique member of the ZIP family because of its localization to the endoplasmic reticulum membrane where it is implicated in the mobilization of zinc from the endoplasmic reticulum, the storage site for zinc, to the cytosol. ZIP7 contains a highly conserved potential metalloprotease motif (HEXPHEXGD) in TMH5 (2788).

An autosomal recessive disease characterized by absence of B lineage cells, agammaglobulinemia, and early-onset infections has been reported in five unrelated families (2398). The resulting immunodeficiency was shown to be due to hypomorphic mutations of SLC39A7. Follow-up studies in transgenic mice showed that while homozygosity for a null allele is embryonic lethal, hypomorphic alleles reproduce the failed B cell development seen in patients. B cells from mutant mice exhibited decreased levels of cytoplasmic free Zn²⁺, increased phosphatase activity, and decreased phosphorylation of signaling molecules. During development, after rearranging heavy and then light chain immunoglobulin genes, B cells must pass quality control checkpoints that signal the expression of the pre-B cell receptor (BCR) and then the BCR. The results show that ZIP7 (SLC39A7) is essential for B cell development and highlight a specific role for cytosolic Zn²⁺, delivered from the ER via ZIP7, in modulating B cell receptor signaling strength and positive selection (2398).

SLC39A13: ZIP13 (SLC39A13) is ubiquitously expressed, including at relatively high levels in skin and muscle and it is able to mediate zinc influx (2789). ZIP13 (SLC39A13) is important for connective tissue development because loss-of-function variants in SLC39A13 are causative for the rare spondylo-dysplastic form of the connective tissue disorder Ehlers-Danlos syndrome, which is characterized by loose joints, joint pain, stretchy, velvety skin, and abnormal scarring (2789, 2790). How the lack of ZIP13 function leads to the disease pathology is still not clear. Proper cross-linking of collagen is crucial, which requires adequate hydroxylation by lysyl hydroxylase and prolyl 4-hydroxylase located in the lumen of the ER. These enzymes are critical for collagen hydroxylation during collagen maturation as reduced cross-linking is present in Ehlers-Danlos syndrome. Both enzymes require iron and vitamin C as cofactors. One possible explanation for how a zinc ER exporter could be involved in the disease mechanism is that the lack of Zn²⁺ export from the ER to the cytosol leads to the accumulation of Zn²⁺ in endosomes, which interferes with ER iron metabolism and competes with iron as a cofactor. Alternatively, ZIP13 (SLC39A13) could itself function as an iron transporter, as found in its ortholog in Drosophila (2791).

LIV-1 branch d

SLC39A4: ZIP4 (SLC39A4) is expressed almost exclusively in the small intestine, colon, stomach and kidney (2792). In the intestine, it is responsible for zinc uptake across the brush border membrane (see Fig. 22), as evidenced by the finding that its loss-of-function mutations drastically reduce zinc absorption and cause a life-threatening autosomal recessive disorder called acrodermatitis enteropathica (2792, 2793). The mutations occur in the conserved transmembrane zinc transport machinery as well as in the extracellular domain, which is present in only a fraction of mammalian ZIP family members (2793).

ZIP4 plays a tumor-promoting role in many types of cancer. These include hepatocellular carcinoma (2794), pancreatic cancer (2795), ovarian cancer (2796) (2797) and oral squamous cell carcinoma (2798).

It has also been reported that ZIP4 is expressed in excitatory synapses and that it plays a functional role at excitatory postsynapses (2799).

SLC39A12: ZIP12 (SLC39A12) is a zinc transporter that mediates cellular zinc uptake in the brain. It is expressed at high levels in the choroid plexus and in glial cells throughout the brain, including Müller glial cells, according to the HPA. ZIP12 (SLC39A12) has been shown to be important for neurite outgrowth, neuronal differentiation, and microtubule polymerization and stability in mouse neuronal models (2800). In this study, the kinetic properties of mouse ZIP12 were investigated in transfected CHO cells using ⁶⁵Zn²⁺ radioisotope transport studies. The results showed that ZIP12 (SLC39A12) can transport zinc ions from the outside to the inside of a cell. The possible roles of ZIP12 (SLC39A12) in the nervous system, including protection against oxidative stress and neurodegeneration, and its putative involvement in the neuropathology of schizophrenia have been extensively reviewed (2801). Yet, the precise biological function of this transporter is not as well understood as that of the other SLC39 members, and its potential role in the choroid plexus remains elusive.

Orphan transporter family members: N/A

SLC40 Basolateral iron transporter (2.A.100/FPN1/MFS)

Discovery: The basolateral iron transporter IREG1/ferroportin/MTP1 (SLC40A1) was identified in parallel by three different groups using three different approaches: 1) By a subtractive cloning approach in hypotransferrinemic (hpx) mice that absorb iron at a high rate (2802); 2) by positional cloning, leading to the identification of ferroportin as the causative gene of the severely anemic zebrafish phenotype weissherbst (weh) (2803); and 3) using an iron-responsive protein (IRP) affinity column to fish out iron-responsive element (IRE)-containing mRNAs, leading to the identification of the metal transporter protein-1 (MTP1) (2804).

Gene family member (1):

SLC40A1 (ferroportin/FPN)

Molecular aspects, physiological roles and links to disease

SLC40A1 belongs to the ferroportin (Fpn) family (TC 2.A.100), which is part of the MFS superfamily.

SLC40A1: Ferroportin/Fpn (SLC40A1) is a cellular ferrous iron (Fe²⁺) efflux transporter with a narrow substrate spectrum that also includes manganese, cobalt and zinc (2805, 2806).

The cryo-EM structure of human Fpn has been reported, showing 12 TMHs and two potential metal ion binding sites S1 and S2 (2807). Fpn is inhibited by the peptide hormone hepcidin, which was shown to bind to the outward-facing conformation of Fpn, from where it reaches the S2 site for inhibition. The study also showed that Fpn functions as an electroneutral 2H⁺/Fe²⁺ antiporter. The structures of ferroportin in complex with its specific inhibitor vamifeport have also been reported (2808) (see below, Clinical Relevance and Pharmaceutical Aspects).

Knockout studies of Fpn (Slc40a1) in mice have shown that this iron exporter is essential for iron homeostasis (2809). While orthologs and paralogs are distributed in a variety of plants (2810), FPN (SLC40A1) is the only cellular iron exporter in humans and acts as a unique regulatory checkpoint for iron homeostasis. Thus, the primary function of Fpn is cellular exit of Fe²⁺.

Fpn is highly expressed in duodenal enterocytes, macrophages and hepatocytes. In enterocytes, it is expressed in the basolateral membrane where it facilitates cellular exit of iron into the blood following Fe²⁺ uptake by DMT1 (SLC11A2) in the apical membrane (Fig. 22). In splenic red pulp macrophages, Fpn facilitates iron exit as part of iron recycling from senescent erythrocytes via erythrophagocytosis, a process that is impaired during aging (2811). Fpn is also expressed in hepatocytes, where excess iron is stored and can be mobilized when needed. Ca²⁺ has been found to be an essential cofactor for metal efflux by Fpn so that iron efflux is stimulated by extracellular Ca²⁺ even though Ca²⁺ is not transported (2812). The regulation of Fpn is mediated by the iron regulatory proteins (IRPs such as IRP1 and IRP2) and the aforementioned peptide hormone hepcidin.

With respect to the IRP regulatory mechanism, the iron-binding elements (IREs) of mRNAs encoding iron-responsive proteins play an essential role. Based on studies in mice, Slc40a1 has two transcripts, the originally described transcript, Fpn1a, which contains an IRE and is expressed in various organs, particularly in spleen, liver and duodenum, and an additional transcript, Fpn1b, which is conserved in humans, is generated from an alternative upstream promoter, lacks the IRE and is expressed in erythroid precursors and duodenum (2813).

Similarly, there are different splice variants of DMT1 (SLC11A2), and DMT1-IRE is the variant that contains an IRE (see SLC11A2 description). IREs are highly conserved hairpin structures of 26-30 nucleotides to which IRPs bind in the absence of iron. Thus, when intracellular free iron levels are low, IRPs bind to IREs at two specific sites of the corresponding mRNAs (1329, 2814). One is at the 5’ untranslated region (5’ UTR) of mRNAs such as that of FPN1A, where IRPs inhibit ribosome binding and thus mRNA translation, and the other is at the 3’ untranslated region (3’ UTR) of mRNAs such as those of DMT1-IRE and the transferrin receptor TFRC, where the IRPs stabilize the mRNAs by binding to the hairpin structure, thereby protecting them from endonucleolytic degradation. At high intracellular iron concentrations, IRP-IRE binding is inhibited, allowing the 5’ IRE Fpn1 (SLC40A1) mRNA to be translated while the 3’ IRE DMT1 (SLC11A2) mRNA is degraded. In the intestine, this implies that once DMT1 takes up Fe²⁺ across the brush border membrane, its mRNA is degraded due to accumulation of intracellular iron with the dissociation of IRP, while the translation of FPN on the basolateral side is activated after iron-triggered dissociation of IRP to allow basolateral exit of iron (Fig. 22).

Hepcidin, on the other hand, inhibits Fe²⁺ export by Fpn by inducing its internalization (2815). Dysregulation of Fpn internalization leads to iron overload disorders. Therefore, pharmacological agents that inhibit Fpn-mediated iron transport are of great clinical interest. The translation of ferroportin is also downregulated post-transcriptionally by the micro-RNA miR-485-3p, which is produced in response to iron deficiency, a finding which may offer a novel potential therapeutic mechanism, circumventing hepcidin-resistant mechanisms, e.g., due to Fpn gain-of-function mutations (2816).

Whole body iron-dependent regulation of Fpn in duodenal enterocytes can be described as follows (2813): Under the iron-replete condition, the liver senses body iron stores as reflected by transferrin saturation levels, and hepatocytes secrete hepcidin. In the duodenum, hepcidin binds FPN and induces FPN internalization and degradation. Intracellularly, IRPs are inactivated due to high intracellular iron levels, and both FPN1A and FPN1B can be freely translated. In the iron-deficient condition, hepcidin-dependent degradation of FPN (SLC40A1) isoforms is absent. IRPs bind to the 5’ IRE of FPN1A and repress FPN protein synthesis, while FPN1B translation is not repressed by IRPs, allowing sufficient iron export to meet the systemic iron requirements during the iron-deficient state.

FPN has been shown to be essential for normal embryonic development, as mouse embryos lacking Slc40a1 abort before gastrulation, and FPN has been implicated in neural tube closure and forebrain patterning (2809, 2817). FPN is expressed in syncytiotrophoblast cells in the mouse placenta and visceral endoderm at embryonic day 7.5 (2803, 2809).

Macrophages play a crucial role in recycling iron from senescent red blood cells, and ferroportin/FPN (SLC40A1) plays an important role in this process (2809, 2818). During erythrophagocytosis, tissue-resident macrophages recognize and engulf aged or damaged red blood cells. After heme degradation via heme oxygenase, the resulting Fe²⁺ can be stored in ferritin or exported from the macrophage into the plasma by FPN (SLC40A1), which is the only known mammalian iron exporter. During inflammation, the function of macrophages shifts and they become more involved in iron storage, which leads to hypoferremia.

Anemia of inflammation is a common comorbidity in patients with chronic inflammatory conditions, autoimmune diseases, and cancer (2819). Hypoferremia is an innate immune response that limits the proliferation of extracellular pathogens requiring sufficient iron supplies. However, if inflammatory diseases are left untreated, they impair erythropoiesis, which leads to anemia of inflammation.

The mechanistic link between inflammation and iron metabolism has been elucidated, highlighting how inflammatory signals can lead to iron sequestration in macrophages (2820). During inflammation, hypoferremia is caused by two mechanisms: 1) decreased ferroportin on the macrophage cell surface due to hepcidin-induced degradation and 2) repression of SLC40A1 transcription. The study shows that, during bacterial infection or inflammation due to cytokine release by immune cells, downstream activation of NF-κB recruits the histone deacetylases HDAC1 and HDAC3. These proteins act on the antioxidant response element of the SLC40A1 promoter, decreasing its mRNA levels. This results in cellular iron retention and hypoferremia. Thus, the study elucidates the mechanistic link between inflammation and iron metabolism, demonstrating how inflammatory signals lead to iron sequestration in macrophages (2820). Understanding this pathway provides insight into conditions such as anemia of inflammation and may reveal therapeutic strategies that target HDACs, for example, to modulate iron homeostasis.

Clinical relevance and pharmaceutical aspects

Pharmaceutical agents that inhibit FPN-mediated iron transport have been developed and tested. Cryo-EM structures of human FPN in complex with synthetic nanobodies and vamifeport (VIT-2763), the first oral FPN inhibitor in clinical development, show competition with hepcidin for FPN binding (2808). The compound is currently in clinical development for β-thalassemia and sickle cell disease. The structures show two distinct conformations of FPN, representing the open and closed states of the transporter. The vamifeport site is located in the center of the protein, where the overlap with hepcidin interactions underlies the competitive relationship between the two molecules. The work provides insight into the pharmacological targeting of the FPN iron.

Orphan transporter family members: N/A

SLC41 MgtE-like magnesium transporter family (1.A.26/MgtE/MgtE)

Discovery: MgtE magnesium transporter orthologs from the Gram-negative bacterium Providencia stuartii and the Gram-positive bacterium Bacillus firmus were identified as novel Mg²⁺ transporters by means of complementation approaches (2821, 2822). Human SLC41A1 was then identified as an ortholog of prokaryotic MgtE using a bioinformatics approach (2823). SLC41A1 has 10 TMHs and shows significant sequence similarity to a specific region identified in the membrane-spanning part of the prokaryotic orthologs of MgtE (2823).

Gene family members (3):
SLC41A1 (MgtE)	SLC41A2	SLC41A3

Molecular aspects, physiological roles and links to disease

Members of the human SLC41 family have been reported to play an important role in maintaining cellular Mg²⁺ homeostasis (2824, 2825). As mentioned above, the SLC41 family belongs to the Mg²⁺ transporter-E (MgtE) family (TC 1.A.26), which harbors the MgtE fold. Several prokaryotic members of the MgtE/SLC41 family have been characterized, including MgtE from the Gram-negative bacterium Providencia stuartii (2821) and MgtE from the Gram-positive bacterium Bacillus firmus (2822), revealing that these proteins are capable of transporting Mg²⁺ and Co²⁺. The crystal structures of MgtE derived from highly thermophilic bacteria have been reported (see SLC41A1 description below).

SLC41A1: MgtE (SLC41A1) is ubiquitously expressed in human tissues, such as muscle, intestine, kidney, and pancreas, according to the HPA and other findings (2826, 2827). It functions as a Na⁺/Mg²⁺ exchanger in plasma membranes and is important for maintaining cellular Mg²⁺ levels (2828, 2829). Since intracellular Mg²⁺ is essential for energy production and mitochondrial activity in cells, MgtE (SLC41A1) is particularly important in highly energy consuming tissues such as the heart and brain. Consistent with this, SLC41A1 deletion reduced the activities of tricarboxylic acid cycle components and electron transport chain complexes under low dietary Mg²⁺ conditions in cardiac cells from mice (2830).

In line with the association of neurodegenerative disorders with dysregulated Mg²⁺ homeostasis, a number of mutations in the SLC41A1 gene have been identified as being associated with Parkinson’s disease (2831).

SLC41A1 expression in the kidney was found to be localized to the basolateral membrane of the thick ascending limb and distal convoluted tubules, where it may contribute to renal Mg²⁺ extrusion (2832). The TRPM6 and TRPM7 channels in the renal distal convoluted tubule facilitate apical entry, and MgtE (SLC41A1) is thought to be responsible for basolateral exit (2830).

In the intestine, transcellular Mg²⁺ uptake occurs via apical TRPM6 and TRPM7 channels and basolateral exit is primarily mediated by CNNM4 (SLC70A4) (2833). The extent to which MgtE (SLC41A1) is involved, if at all, is not fully understood.

MgtE was proposed to have a channel-like transport mechanism. The high-resolution crystal structure of MgtE from Thermus thermophilus, bound to Mg²⁺, has visualized the MgtE selectivity filter (M1 site) that recognizes a Mg²⁺ ion in its fully hydrated state (39). Interestingly, this is in sharp contrast to the K⁺-channels, which remove the hydration shell from K⁺ before it enters the selectivity filter (2834). The structure revealed that the M1 site, which contains Asp432, recognizes the exact size and geometry of the Mg²⁺ hydration shell, allowing it to prevent the entry of hydration shells of other cations such as Na⁺, K⁺, and Ca²⁺ (39).

The role of MgtE (SLC41A1) in hepatocellular carcinoma (HCC) has been studied (2835). MgtE (SLC41A1) is significantly upregulated in HCC tissues compared to normal liver tissues, and high levels correlate with unfavorable clinical features and reduced overall survival. Elevated MgtE (SLC41A1) expression has been linked to increased infiltration of various immune cells within the tumor microenvironment, suggesting that it plays a role in modulating immune responses. The study also identifies DNA methylation patterns in the SLC41A1 gene that may influence its expression and impact patient outcomes (2835).

SLC41A2: SLC41A2 has been reported to mediate Mg²⁺ transport across the plasma membrane or organellar membranes such as the Golgi apparatus (2824, 2836). The Mg²⁺ currents of human SLC41A2, as well as those of other divalent cations, were studied using the Xenopus oocyte expression system (2837). SLC41A2 was found to mediate voltage-dependent, saturable Mg²⁺ uptake with a K_m value of 0.34 mM. SLC41A2 was also found to transport various other divalent cations, including Ba²⁺, Ni²⁺, Co²⁺, Fe²⁺, and Mn²⁺. Overall, however, the studies suggest that SLC41A2 is primarily a Mg²⁺ transporter that is responsible for maintaining magnesium homeostasis in epithelial cells.

SLC41A2 has been reported to be highly expressed in the lymph node, stomach, lung, testis, and skin, and moderately expressed in the spleen, intestine, heart, and kidney (2824).

Whole-genome studies of lymphoblastoid cell lines revealed that SLC41A2 is positively associated with cellular susceptibility to paclitaxel, a chemotherapeutic drug belonging to the taxane family that inhibits cancer cell division by interfering with microtubules (2838).

SLC41A3: SLC41A3 is a Mg²⁺ transporter that is widely expressed, according to the HPA. Initially, it was predicted that SLC41A3 functions as a plasma membrane Mg²⁺ transporter (2839), possibly operating as a Na⁺/Mg²⁺ exchanger (2840). However, subsequent characterization of SLC41A3-mediated Mg²⁺ transport has indicated that this protein localizes to the mitochondria, despite the absence of a mitochondrial localization sequence (2841).

SLC41A3 is highly expressed in the renal distal convoluted tubule (DCT), and its transcript is upregulated in mice with low magnesium intake (2842). Mice deficient in Slc41a3 (Slc41a3^-/-) exhibit hypomagnesemia, with serum Mg²⁺ levels 29% lower than those of wild-type animals (2839). The knockout also resulted in increased expression of magnesiotropic genes such as Trpm6 and Slc41a1. Based on double knockout studies, SLC41A3 has been shown to play a more significant role than SLC41A1 in regulating systemic Mg²⁺ levels (2843).

Subsequently, two alternative transcripts of Slc41a3 in mice have been identified that extrude Mg²⁺ specifically in the distal convoluted tubules (2840). This finding answers the longstanding question of renal Mg²⁺ transport by revealing the molecular identity of the basolateral Na⁺/ Mg²⁺ exchanger in the distal convoluted tubule (DCT). This is significant because the DCT plays a crucial role in fine-tuning magnesium reabsorption in the kidney.

Furthermore, the data reveal that mitochondria are not the primary site of SLC41A3 expression (2840).

Studies using conditional Slc41a3-homodeficient mice revealed that these animals exhibit abnormal locomotor coordination. This suggests that SLC41A3 also plays a role in muscle contraction and coordinating movement (2825).

Orphan transporter family members: N/A

SLC42 Rh type glycoprotein family of ammonium transporters (1.A.11.4/Ammonium_transp/AmtB)

Discovery: The first ammonium transporter was cloned in 1991 as the product of the E. coli amtA gene by complementation of a mutant deficient in the transport of methylamine, a commonly used surrogate for NH₄⁺ (2844). In 1994, using complementation approaches, the first sequences for ammonium transport proteins were reported for Amt-1 from Arabidopsis thaliana (161) and MEP1 from Saccharomyces cerevisiae (2845). The first structure of an ammonia channel, AmtB, from the Amt/MEP/Rh protein family has been determined at 1.35 Å resolution. The structure, determined with and without ammonia or methylammonia, shows a channel with a vestibule that recruits NH₄⁺/NH₃ through favorable interactions with conserved histidines (62).

RH (Rhesus)-glycoproteins are blood group antigens. Next to the ABO blood group system, the Rh blood group system is most likely to be involved in transfusion reactions. The molecular identification of RH proteins started with the cDNA cloning of erythrocyte membrane protein associated with Rh-blood-group-antigens (2846–2849). As members of the ammonium transporter (Amt) family, it was recognized that the RH-glycoproteins have a channel-like architecture that is related, at least in part, to the transport of ammonia/ammonium (2850–2853).

Gene family members (5):
RHAG (SLC42A1)	RHCG (SLC42A3)	RHD (SLC42A5)
RHBG (SLC42A2)	RHCE (SLC42A4)

Molecular aspects, physiological roles and links to disease

In E. coli, ammonium (NH₄⁺) serves as the sole nitrogen source under nitrogen-limited conditions, requiring a high-affinity ammonium transport system capable of generating large concentration gradients (2844). In plants, ionic nitrogen is taken up from the soil as nitrate or ammonium, with nitrogen-fixing bacteria thought to provide nitrogen from ammonium to legumes. In animals, ammonium transport systems are responsible for the active resorption of ammonium in renal tubule epithelial cells of the thick ascending limb (2854). In addition, the liver is known to be an efficient scavenger of ammonium from the blood and is therefore thought to contain highly active ammonium transport systems (2855).

The mammalian SLC42 RH-glycoprotein family consists of five members (Fig. 47) (2847, 2850). It belongs to the Ammonium Channel Transporter (Amt) family (TC 1.A.11) which exhibits the AmtB fold. In E. coli, the only ammonia transport protein is AmtB, which acts as a channel selective for the transport of uncharged NH₃. For this purpose, it uses a transport mechanism for ammonium ions by reducing their pK_a to allow the passage of the uncharged substrate NH₃. Indeed, structural data of E. coli AmtB indicated that these transporters function as ammonia gas channels (2856). AmtB forms a trimer, and each monomer consists of a central hydrophobic channel surrounded by a right-handed helical bundle of 12 TMHs (2857). Within the bundle there are two five-helix motifs that are similar in structure but of opposite polarity with respect to the membrane. The mammalian SLC42 family members are predicted to have 12 TMHs (2850).

Fig. 47. Phylogenetic tree of the SLC42 family.

RHAG (SLC42A1), RHBG (SLC42A2) and RHCG (SLC42A3): The mechanisms of ammonia and ammonium (NH₃/NH₄⁺) transport by the rhesus-associated (RH) glycoproteins RHAG (SLC42A1), RHBG (SLC42A2), and RHCG (SLC42A3) expressed in Xenopus oocytes were studied using ion-selective microelectrodes and two-electrode voltage clamping to measure changes in intracellular pH and currents (2858). The data shows that RHAG (SLC42A1) and RHBG (SLC42A2) transport both the ionic NH₄⁺ and neutral NH₃ species. Additionally, the transport of NH₄⁺ is electrogenic, and RHCG (SLC42A3) is likely predominantly a NH₃ transporter. RHAG (SLC42A1) and RHBG (SLC42A2) do not appear to be NH₄⁺/H⁺ exchangers.

RHAG (SLC42A1) is predominantly expressed in red blood cells where it exists as a hetero-oligomeric “RH complex” of membrane polypeptides. This complex has the well-known antigenic effect and contributes to the stability of the erythrocyte membrane (2859).

RHBG (SLC42A2) and RHCG (SLC42A3) are expressed only in non-erythroid tissues. The RH glycoproteins have been shown to be related to NH₄⁺ transporters of yeast (MEP proteins) and bacteria (Amt) (2860). A follow-up study (2861) identified a kidney homolog, RHCG (SLC42A3), with close amino acid sequence identity to RHAG (SLC42A1 (50%) and sequence identity (24%) to the MEP/Amt ammonium transporters. The cloning and biochemical characterization of RHBG (SLC42A2) and RHCG (SLC42A3) were subsequently completed (2862).

In the kidney, the non-erythroid Rh proteins RHBG (SLC42A2) and RHCG (SLC42A3) contribute to ammonium transport, with RHBG (SLC42A2) in the basolateral membrane and RHCG (SLC42A3) in the apical membrane (2863) (see Fig. 12).

RHCE (SLC42A4) and RHD (SLC42A5) - Orphan transporters: RHCE (SLC42A4) and RHD (SLC42A5) are additional erythrocyte membrane proteins structurally related to ammonium transporters (Amt) (2864), but their physiological function remains unclear. Interestingly, these non-glycosylated erythrocyte proteins have different residues in the channel-like regions compared to the ammonium transporters RHAG (SLC42A1), RHBG (SLC42A2) and RHCG (SLC42A3), suggesting a different function.

Orphan transporter family members (2)

RHCE (SLC42A4) and RHD (SLC42A5)

HGNC updates

SLC42A4 is a new alias for RHCE; SLC42A5 is a new alias for RHD.

SLC43 Na⁺-independent, system-L-like amino acid transporter family (2.A.1.44/MFS_1/MFS)

Discovery: The cDNA of human LAT3 was identified by expression cloning of L-leucine uptake in Xenopus oocytes (156). The LAT3 gene was found to be identical to POV1, a prostate cancer upregulated gene (2865) and was reclassified as SLC43A1. Homology searches were used to clone human LAT4 cDNA (6), which was named SLC43A2. Finally, mouse Eeg1 was identified as a gene expressed in a cellular model of renal tubulogenesis (2866) and the human ortholog was assigned SLC43A3.

Gene family members (3):
SLC43A1 (LAT3)	SLC43A2 (LAT4)	SLC43A3 (EEG1)

Molecular aspects, physiological roles and links to disease

Human LAT3 and LAT4 share 57% amino acid sequence identity, whereas human EEG1 (embryonic epithelia gene 1) is a distant member of the family, sharing only 27% amino acid sequence identity with the other two members.

SLC43A1, SLC43A2: LAT3 (SLC43A1) and LAT4 (SLC43A2) are members of the system L family of neutral amino acid transporters but are structurally distinct from the heterodimeric LAT1 (SLC7A5) and LAT2 (SLC7A8). Both LAT3 and LAT4 are facilitative amino acid exchangers with 12 TMHs and cytosolic N- and C-termini, mediating Na⁺- and Cl⁻-independent transport of neutral amino acids (6, 156).

According to the HPA, LAT3 is highly expressed in pancreas, skeletal muscle, liver (including fetal liver), placenta, and seminal vesicle, whereas LAT4 shows highest levels in stomach, skeletal muscle, adrenal gland, placenta, kidney, lung, duodenum, and spleen. Based on substrate selectivity, affinity, and N-ethylmaleimide sensitivity, LAT3 was proposed to correspond to the long-recognized “system L2” transporter first described in primary hepatocytes (156, 2867).

Immunofluorescence studies of mouse tissues revealed strong basolateral LAT4 staining in small-intestinal enterocytes, as well as in kidney proximal tubules and thick ascending limb cells, but no detectable signal in liver or skeletal muscle (2868).

LAT4 is an essential amino acid transporter required for early nutrition during mouse development (2868). According to the HPA, LAT4 is highly expressed at the single cell level in syncytiotrophoblasts, consistent with a role in placental amino acid transfer.

In the intestine, LAT4 is enriched in crypt cells where it may support enterocyte amino acid homeostasis (6).

A recent SLC superfamily interactome screen identified PDZ-domain scaffold proteins LIN7C and MPP1 as LAT4 (SLC43A2) interactors that promote plasma-membrane trafficking and stability (1575). These interactions suggest that PDZ-dependent anchoring contributes to LAT4 localization and function at basolateral membranes.

SLC43A3: EEG1 (SLC43A3), also known as ENBT1, is Na⁺-independent purine-selective nucleobase transporter (2869, 2870). It is highly expressed in liver, thyroid, cerebellum, lung, mammary gland, bone marrow, esophagus, skin, urinary bladder, placenta, testis and seminal vesicle according to the HPA. A high level of expression of Slc43a3 in the lung and liver has also been reported in mouse tissues (2871). Expression was also detected in vascular endothelial cells (2872). Functional analysis revealed that EEG1/ENBT1 (SLC43A3) is a nucleobase transporter that facilitates adenine uptake in hepatocytes (Fig. 33) and has distinct inhibitory properties compared to the equilibrative nucleoside transporters ENT1 (SLC29A1) and ENT2 (SLC29A2) (2869, 2870). Model analysis of the kinetics of purine nucleobase uptake in cells revealed that the uptake of adenine is rate-limited by the intracellular enzyme adenine phosphoribosyltransferase (2873). A splice variant (SLC43A3_2) encoding a protein with 13 additional amino acids in the first extracellular loop was also identified, and both variants of EEG1/ENBT (the original variant SLC43A3_1 and the variant SLC43A3_2) were shown to encode proteins with similar functional properties (2874). Both were also shown to mediate the transport of 6-mercaptopurine, a nucleobase analog used in the treatment of acute lymphoblastic leukemia and IBD, and increased SLC43A3 expression enhanced the ability of 6-mercaptopurine to induce cell death (2871, 2874).

Orphan transporter family members: N/A

SLC44 Putative choline transporter CTL1 family (2.A.92/Choline_transpo/SLC44)

Discovery: CTL1 was originally cloned from the marbled electric ray (Torpedo marmorata) and first characterized as a suppressor of a yeast choline transport mutation (ctr) derived from a torpedo electric ray yeast expression library (2875) (2876). Na⁺-independent high affinity choline uptake was demonstrated in ctr mutant yeast transformed with torpedo ctl1 (2875). Rat Slc44a1, which shares 69% identity with its Torpedo counterpart, was then cloned from a brain cDNA library (2875). Interestingly, the human choline transporter-like protein 1 was found to be 99.1% identical to the sequence of the human antigen CDw92, a 70 kDa surface protein widely expressed on leukocytes and endothelial cells (2877).

Gene family members (5):
SLC44A1 (CTL1)	SLC44A3 (CTL3)	SLC44A5 (CTL5)
SLC44A2 (CTL2)	SLC44A4 (CTL4/TPPT)

Molecular aspects, physiological roles and links to disease

The SLC44 family belongs to the Choline Transporter-like (CTL) family (TC 2.A.92), which carries the SLC44 structural fold with 10 TMHs (31, 2878) (Fig. 4).

Several members of this family are reported to be involved in choline transport. For CTL1 (SLC44A1), the choline transport function has been confirmed by several groups (2879–2883). Furthermore, sRNA knockdown studies in the H82 human small cell lung cancer cell line suggested a role in choline transport not only for CTL1 but also for CTL2 (SLC44A2) and CTL5 (SLC44A5) (2884). Also, knockdowns in A549 lung adenocarcinoma cells (328) and neuroblastoma cells (2885) are consistent with a role of CTL1 and CTL2 in choline transport. However, knockdown of CTL4 (SLC44A4) in H82 cells did not affect choline uptake, yet it was the only CTL family member associated with ACh synthesis in a non-neuronal cell line (2884). Subsequent studies revealed that CTL4 functions as a thiamine pyrophosphate (TPP) transporter, which was reported to be responsible for the uptake of microbiota-generated TPP in the colon (2878, 2886).

In another study, CTL1 (SLC44A1) was reported to function as both a choline and an ethanolamine transporter (2887). Therefore, it was concluded that the membrane phospholipids phosphatidylcholine and phosphatidylethanolamine are synthesized de novo via the Kennedy pathway in which extracellular substrates choline and ethanolamine are transported into the cell via CTL1, then phosphorylated and coupled with diacylglycerol to form the final phospholipid products.

Due to the lack of information on the transport function and physiological roles of CTL3 (SLC44A3) and CTL5 (SLC44A5), these transporters have been classified as orphan transporters.

According to the phylogenetic tree (Fig. 48), the SLC44 transporters can be divided into group 1, which includes SLC44A1 (CTL1) and SLC44A3 (CTL3), and group 2, which includes SLC44A4 (CTL4) as well as SLC44A2 (CTL2) and SLC44A5 (CTL5).

Fig. 48. Phylogenetic tree of the SLC44 family.

Group 1 - SLC44A1 (CTL1) and SLC44A3 (CTL3):

SLC44A1: CTL1 (SLC44A1) has been reported to be Na⁺-independent plasma membrane and mitochondrial transporter of choline and ethanolamine with important roles in de novo phosphatidylcholine and phosphatidylethanolamine synthesis via the Kennedy pathway (2887–2889).

FLVCR1 (SLC49A1) has also been shown to function as a transporter of choline and ethanolamine (2890). The substrate affinities of FLVCR1 have been reported to be even more physiologically relevant than those of CTL1/SLC44A1 (2890) (see the description of SLC49A1).

In human colon carcinoma cells (2882) and small cell lung carcinoma cells (2891), acidification of the extracellular milieu by inhibition of CTL1 was shown to block choline transport. Therefore, it was suggested that CTL1 (SLC44A1) functions as an H⁺ antiporter.

In the mitochondria, it has been suggested that CTL1 maintains the intracellular pools of choline (2888, 2889).

CTL1 has widespread expression throughout the nervous system (2880). The HPA suggests that it is most highly expressed in oligodendrocytes, especially in the spinal cord. Expression of CTL1 has also been reported in keratinocytes where it is reported to be responsible for the uptake of choline and organic cations (2881).

Choline is essential for healthy brain development, and its intake is often limited during the high-demand periods of pregnancy. Therefore, supplemental choline can significantly improve offspring cognition and memory, which is particularly relevant for alcohol-exposed offspring (2892–2894). Two functional alleles of SLC44A1 have been reported to increase susceptibility to choline deficiency and have been shown to be associated with impaired cognition in children, probably due to reduced choline transport function. Increasing choline through supplementation counteracts the deleterious effects of this transporter variant. The findings highlight that choline benefits cognitive development in normotypic children and in children exposed to prenatal alcohol.

Interestingly, CTL1 (SLC44A1) deficiency has been shown to impede myelin development in the central nervous system (2895) (preprint). The study reveals that CTL1 (Slc44a1) is essential for myelin development in the central nervous systems of zebrafish and rodents. CTL1 (Slc44a1) deficiency impairs the oligodendroglial choline metabolic pathway. This results in reduced levels of the choline metabolite citicoline, disrupted phosphatidylcholine biogenesis, and altered myelin sheath composition. These changes lead to deficits in oligodendrocyte maturation and myelinogenesis during development. Administering citicoline ameliorates hypomyelination in Slc44a1 knockout animals, suggesting it could be an effective treatment for childhood-onset neurodegeneration in patients with SLC44A1 deficiency or variations (2893, 2894).

SLC44A3 - Orphan transporters: CTL3 (SLC44A3) is a widely expressed but poorly characterized member of the family (2887). As indicated above, it may function as a choline transporter, but its physiological role remains unknown.

Group 2 - SLC44A4 (CTL4) and the closely related SLC44A2 (CTL2) and SLC44A5 (CTL5):

SLC44A4: CTL4/TPPT (SLC44A4) expression is quite widespread. In epithelia, where it is predominantly found, expression was generally at the apical surface based on immunohistochemistry (2896). Specific staining was observed in epithelia of the prostate, lung bronchioles, gastrointestinal tract, a subset of tubules in the renal cortex, fallopian tubes, bladder, ureter, and uterine endometrium. CTL4 expression was also seen in the ductal epithelium of some samples of liver (bile ducts), breast, salivary gland, esophagus, pancreas, sweat glands. The HPA also suggests expression in the CNS.

As mentioned above, CTL4 (SLC44A4) was the only CTL family member associated with acetylcholine (ACh) synthesis in a non-neuronal cell line (2884), although knockdown of CTL4 (SLC44A4) in H82 cells did not affect choline uptake, possibly due to low expression. In contrast, the choline transporters CHT1 (SLC5A17), CTL1 (SLC44A1), CTL2 (SLC44A2) and CTL5 (SLC44A5) did not appear to be required for non-neuronal acetylcholine synthesis and/or secretion (2884).

Independent of the studies on SLC44A4, in an effort to identify the human colonic thiamine pyrophosphate transporter, BLAST searches for a mammalian homologs of the previously identified TPP ABC transporter permease of the oral spirochete T. denticola (2897) led to the identification of SLC44A4, which shows significant similarity to the partial amino acid sequence of the TPP ABC transporter permease in regions predicted to be in substrate-binding domains. CTL4 (SLC44A4) was shown to function as a thiamine pyrophosphate (TPP) transporter responsible for the uptake of microbiota-generated TPP in the colon (2878, 2886).

In the colon, it has been shown to be present at the apical membrane (2878). It has been proposed that colonic microbiota-generated thiamine, present in the phosphorylated form of TPP, is taken up by SLC44A4, whereas the dietary source of thiamine (vitamin B1) is absorbed in the small intestine via thiamine transporters THTR1 (SLC19A2) and THTR2 (SLC19A3) (see the SLC19 family description). TPP-mediated uptake by SLC44A4 was shown to be Na⁺-independent, slightly higher at acidic pH, and saturable, with an apparent K_m of 0.17 μM (2878). The transport was highly specific for TPP and was not affected by free thiamine, thiamine monophosphate, or choline. It has been proposed that colonocytes take up TPP to meet their metabolic needs, while the remainder of TPP is metabolized to TMP and thiamine, for the latter to enter the portal vein via the basolateral membrane transporter THTR1 (SLC19A2), which is also expressed in colonocytes (2878).

The role of SLC44A4 may not be limited to TPP absorption in the colon as it is expressed throughout the gastrointestinal tract (2880, 2896) and the HPA indicates prominent expression in the small intestine, particularly the duodenum, in addition to the colon. Most dietary sources of vitamin B1, including wheat and sunflower seeds, consist largely of the phosphorylated form of thiamine, TPP. TPP is converted to thiamine by gastrointestinal phosphatases prior to intestinal absorption via THTR1 (SLC19A2) and THTR2 (SLC19A3) (2898). However, some may be taken up directly by SLC44A4.

In beriberi disease, which is known to result from dietary thiamine/vitamin B1 deficiency, the source of gut microbial TPP was apparently not sufficient to prevent vitamin B1 deficiency. There could be two different reasons for this: 1) that the contribution of SLC44A4-mediated thiamin supply in the colon is negligible; and 2) that there is a disruption of the gut microbiome in beriberi disease due to the consumption of refined carbohydrate foods, which may, for example, increase susceptibility to microbial infection by thiaminase-producing bacteria, thereby preventing any significant production of TPP in the colon for uptake via SLC44A4 (2899). This would be consistent with the original hypothesis that beriberi may actually result from microbial infection (2900).

Further studies will be necessary to clarify whether SLC44A4 has different functional roles in non-neuronal acetylcholine metabolism and intestinal absorption of TPP. As mentioned above, SLC44A4 has not been shown to be involved in choline transport in lung H82 cells, but has been associated with acetylcholine synthesis (2884). It is tempting to speculate that TPP taken up by SLC44A4 directly increases acetylcholine synthesis (2901, 2902).

CTL4 has also been shown to play a critical role in the choline-acetylcholine system involved in inner ear hair growth. In particular, a mutation in SLC44A4 was shown to be the cause of an autosomal dominant hereditary postlingual non-syndromic mid-frequency hearing loss in a Chinese family (2903). This led to the conclusion that mutant SLC44A4 causes defects in the choline-acetylcholine system, which is crucial for the efferent innervation of hair cells in the olivocochlear bundle to maintain the normal function of outer hair cells and to protect them from acoustic insult, leading to hearing loss.

CTL4 has also been shown to play an important role in tumorigenesis and is significantly upregulated in a variety of epithelial tumors, most notably prostate and pancreatic cancer. An antibody-drug conjugate with potential antineoplastic activity has been developed (2896, 2904). The antibody-drug conjugate comprises a fully human monoclonal antibody directed against an epitope of SLC44A4 termed “AGS-5” linked via a valine-citrulline-maleimidocaproyl linker to the antimicrotubulin drug monomethyl auristatin E (MMAE). The monoclonal antibody moiety of ASG-5ME was shown to selectively bind to AGS-5. After internalization and proteolytic cleavage, MMAE binds to tubulin and inhibits its polymerization, resulting in G2/M phase arrest and tumor cell apoptosis. SLC44A4 has been shown to be overexpressed in more than 80% of samples from patients with pancreatic, prostate and gastric cancer. The first clinical trial of ASG-5ME in prostate cancer has been reported. Although antibody-drug conjugates are intended to increase the efficacy of drugs while reducing off-target side effects, ASG-5ME was associated with significant toxicities (2904).

SLC44A2: CTL2 (SLC44A2), like CLT1 (SLC44A1), has been reported to function as a plasma membrane and mitochondrial choline and ethanolamine transporter involved in phosphatidylcholine and phosphatidylethanolamine synthesis (2887). However, CTL2 (SLC44A2) has only been indirectly implicated in phosphatidylcholine synthesis, and its molecular and cellular functions in the plasma membrane or mitochondria have not yet been fully established (2887, 2889, 2905).

CTL2 (SLC44A2) is expressed in red blood cells, endothelial cells, and neutrophils, as well as in various organs and tissues, such as the kidney, lung, and inner ear (2905, 2906). CTL2 was originally discovered as a supporting cell antigen in the inner ear (2907). Two CTL2 (SLC44A2) isoforms that differ in glycosylation have been identified and shown to exhibit variable expression in the cochlea, tongue, heart, colon, lungs, kidneys, liver, and spleen, suggesting tissue-specific differences (2908).

Since its discovery, human SLC44A2 single nucleotide polymorphisms (SNPs) have been associated with thrombosis (2909) and Ménière disease (2910).

CTL2 (SLC44A2) has been shown to carry a new blood group system in red blood cells (2905). Whole exome sequencing in individuals of Moroccan ancestry identified a novel missense mutation in SLC44A2, and it was shown that their rare blood group phenotype is caused by the P398T substitution in CTL2 (SLC44A2) (2905). However, the biological function of CTL2 (SLC44A2) in blood cells remains unclear since SLC44A2^-/- individuals exhibit normal in vitro erythropoiesis and no apparent hematological disorders (2905, 2911).

As indicated above, CTL2 (SLC44A2) plays a critical role in preserving hearing. Slc44a2^-/- mice have been shown to exhibit hair cell death and hearing loss (2911). Three siblings of European ancestry were found to be homozygous for a large deletion in SLC44A2. This resulted in complete gene deficiency, leading to progressive hearing loss as well as recurrent arterial aneurysms and epilepsy (2905). CTL2 (SLC44A2) is strongly expressed in supporting cells of the human inner ear and plays a role in autoimmune hearing loss (2908, 2912, 2913). Previous studies have shown that antibodies against CTL2/SLC44A2 cause hearing loss in animals and are frequently found in patients with autoimmune hearing loss (2908).

Currently, the relationship between CTL2 (SLC44A2) dysfunction and hearing loss/Ménière disease is unknown. Since CTL2 is expressed in the cochlea and vestibular organs, it could be involved in inner ear homeostasis, perhaps through choline transport or membrane phospholipid metabolism. Due to evidence indicating that autoantibodies against CTL2 (SLC44A2) contribute to the pathogenesis of Ménière disease in certain cases, an autoimmune mechanism may also be considered.

CTL2 (SLC44A2) function has also been found to play a crucial role in counteracting aortic aneurysm, a life-threatening condition with few treatment options. Specifically, the role of CTL2 (SLC44A2) in regulating vascular smooth muscle cells (VSMCs) has been investigated (2914). CTL2 has been found to regulate VSMC phenotypic switching and aortic aneurysm development. Elevated levels of CTL2 have been observed in the aortas of patients with abdominal aortic aneurysms. CTL2 was shown to interact with cell surface proteins such as neuropilin-1 (NRP1) and integrin beta 3 (ITGB3), triggering TGF-β/SMAD signaling. This signaling pathway promotes the expression of genes that maintain smooth muscle cells in a contractile state, thereby preventing these cells from switching to a synthetic/proliferative phenotype associated with vascular diseases, such as aortic aneurysms. Findings reveal the formation of an SLC44A2-NRP1-ITGB3 complex that plays a pivotal role in this VSMC phenotypic switching regulation. Elevated CTL2 (SLC44A2) levels in aortic aneurysms have been shown to be associated with increased RUNX1 expression. The immunomodulatory drug lenalidomide has been shown to promote RUNX1-mediated transcription of SLC44A2 (2915). Thus, this study found that, in aortic aneurysms, upregulated CTL2 (SLC44A2) acts as a scaffolding protein rather than a choline transporter, interacting with NRP1 and ITGB3 and thereby activating TGF-β/SMAD signaling to promote the expression of VSMC contractile genes while inhibiting the expression of VSMC synthetic genes. This process restrains VSMC phenotypic switching in aortic aneurysms. CTL2 (SLC44A2) therefore represents a promising therapeutic target for vascular diseases involving VSMC dysfunction (2914).

SLC44A5 - Orphan transporters CTL5: (SLC44A5) is poorly characterized and its transport function and physiological role remain unknown (2887). According to the HPA it is expressed in inhibitory neurons, excitatory neurons, oligodendrocytes and early spermatids.

Orphan transporter family members (2)

SLC44A3 (CTL3), SLC44A5 (CTL5)

SLC45 H⁺/sugar cotransporter family (2.A.2.4/MFS_1, MFS_2/MFS)

Discovery: The SLC45 family belongs to the major facilitator superfamily, which includes the mammalian facilitative transporters of the SLC2 family, plant H⁺/sucrose transporters, and E. coli lactose permease. A cDNA encoding the H⁺/sucrose transporter SUT1, which is responsible for phloem loading in leaves of higher plants (2916), was identified by complementing an engineered yeast mutant with a cDNA library from spinach leaves. SUT1 contains conserved sugar transporter sequence motifs that were later critical in identifying the rat and then the human ortholog, named PAST-A (proton-associated sugar transporter-A) (SLC45A1), using a unique strategy: a differential display technique was used to identify genes, searching for genes involved in the adaptation of neuronal cells to changes in H⁺/CO₂ concentrations (2917). Conserved sequence motifs previously shown to be critical for sugar transport function were detected, including a sucrose H⁺-transport motif found in all known sucrose H⁺ transporters in plants, such as SUT1, suggesting that PAST-A is a sugar transporter, which was subsequently confirmed by expression studies in COS-7 cells (2917).

Prior to this, the human gene encoding SLC45A1 was identified as a DNB5 (deleted in neuroblastoma-5), encoding a protein of unknown function, in a screen for tumor suppressors on chromosome 1p, and Northern blot analysis revealed strong expression in fetal brain and kidney as well as in adult brain (2918).

In parallel, a fish sugar transporter was identified by positional cloning from the Japanese rice fish medaka, which exhibits many spontaneous pigmentation mutants, one of which, an orange-red variant, is a homozygote of a well-known and common allele, b, encoding a transporter mediating melanin synthesis (2919). The identified protein was predicted to consist of 12 TMHs and was 55% identical to a human expressed sequence tag (EST) of unknown function, which was later shown to correspond to SLC45A2, genetic variants of which are now known to cause oculocutaneous albinism type 4 (OCA4) in humans. The amino acid sequence of the medaka sugar transporter shares ~23% amino acid sequence identity with plant proton-coupled sucrose symporters such as SUT1 and ~55% with SLC45A2 (2919).

Furthermore, positional cloning of the Danio rerio albino mutant revealed that the affected gene encodes slc45a2 (2920). Slc45a2 and V-ATPase were shown to be regulators of melanosomal pH homeostasis in zebrafish, providing insight into human skin pigment variation and pigmentation disorders (2920).

Gene family members (4):
SLC45A1 (PAST-A)	SLC45A2	SLC45A3	SLC45A4

Molecular aspects, physiological roles and links to disease

There are four transporters in this family, SLC45A1, SLC45A2, SLC45A3 and SLC45A4, and all of them are sugar transporters (Fig. 49). The SLC45 family belongs to the Glycoside-Pentoside-Hexuronide (GPH):Cation Symporter family (TC 2.A.2.4), which is part of the MFS superfamily. All SLC45 members harbor 12 predicted TMHs with a large intracellular loop between TMHs 6 and 7 and a signature sequence R-X-G-R-[K/R] between TMHs 2 and 3 that was found to be conserved in plant sucrose transporters including SUT1 (2921).

Fig. 49. Phylogenetic tree of the SLC45 family.

The SLC45 family members are H⁺-coupled sugar transporters. SLC45A1 has been shown to transport glucose and galactose (2917). Heterologous expression studies in Saccharomyces cerevisiae showed that SLC45A2, SLC45A3 and SLC45A4 transport glucose, fructose and sucrose (2922).

SLC45A1: SLC45A1, also known as PAST-A, is an H⁺-coupled glucose and galactose transporter with 12 predicted TMHs that is highly expressed in the brain, especially on the ventral surface of the medulla oblongata (2917, 2923). Expression has also been detected in heart, muscle and kidney (2918, 2924). In addition, the HPA suggests significant expression in male and female tissues, adipose tissue, spleen, muscle and GI tract.

In the brain, SLC45A1 has been implicated in the regulation of glucose homeostasis (2922) and in intellectual development disorder with neuropsychiatric features (2925).

SLC45A1-mediated uptake of glucose into neuronal tissue has been shown to be induced during hypercapnia, a condition with elevated blood CO₂ levels (P_aCO2 >45 mm Hg) (2917). Based on studies in newborn piglets and lambs, it has been reported that during hypercapnia there is a decrease in extracellular and intracellular pH and altered neuronal energy metabolism (2926, 2927). While the role of SLC45A1-mediated H⁺-coupled glucose transport in neurons in the adaptation of neuronal cells to changes in H⁺/CO₂ concentrations still remains to be elucidated (2917), the significantly increased glucose uptake of SLC45A1 at low pH suggests its possible role in providing glucose to neurons under hypercapnia-induced acid stress conditions (2924).

A study showed that autosomal recessive missense variants in SLC45A1 cause intellectual disability, movement disorders, and epilepsy, implicating the gene in neurodevelopmental disorders (2923). In addition, in a single transcript analysis on 400 monozygotic twins used to identify differentially expressed genes and biological pathways involved in cognitive function in the context of aging, SLC45A1 expression was found to be upregulated with increasing cognitive function (2928). In addition, compound heterozygous variants in SLC45A1 have been associated with syndromic intellectual disability, and experiments have shown that the missense mutations of SLC45A1 resulting in the variants V35M and F404C attenuate glucose transport activity through alteration of the tertiary structure and failure of proper intracellular localization (2925).

However, despite these advances, our knowledge of SLC45A1 is still limited and further studies will be needed to elucidate the physiological and pathophysiological roles of this transporter in the CNS and other organs.

SLC45A2: SLC45A2 is an H⁺-coupled glucose and fructose transporter that has been shown to play an important role in melanogenesis (2922, 2924, 2929). As described above under “Discovery”, the corresponding orthologs have been identified by positional cloning in the Japanese rice fish medaka (2919) and zebrafish (2920) as genes associated with pigmentation anomalies.

In humans, genetic defects in the SLC45A2 gene have been shown to cause autosomal recessive oculocutaneous albinism (OCA) (2930–2932). Studies have shown that 10% of all OCA cases from East and South Indian ethnic groups carry pathogenic mutations in SLC45A2. Specifically, SLC45A2 mutations cause OCA type 4 with hypopigmentation of the eyelid and skin (2929, 2933).

A single genetic variant in SLC45A2 SNP rs16891982 resulting in a missense mutation (L374F) has been implicated in differences in skin and hair pigmentation among European, Chinese, South American, and South Asian human populations (2931, 2934, 2935).

SLC45A2, which localizes to acidic organelles such as melanosomes, is thought to support neutralization by enabling melanosomes to export a H⁺ in cotransport with monosaccharides, possibly released from glycoproteins by lysosomal glycosidases.

Consistent with a function of SLC45A2 in H⁺ export and neutralization of the acidic organelle, the zebrafish slc45a2 pigmentation mutant was rescued upon inhibition of endolysosomal and melanosomal acidification by treatment with the V-ATPase inhibitor bafilomycin A1 or by knockdown of the atp6v1 subunit of the vacuolar ATPase (2920). Knockdown of SLC45A2 in a pigmented melanoma cell line was also reported to result in increased acidification of early-stage melanosomes (2936). The localization of SLC45A2 in melanocytes was confirmed and it was shown that SLC45A2 localizes to a cohort of mature melanosomes that only partially overlaps with the cohort expressing the putative chloride channel OCA2 (SLC13B1) (2929), another important regulator of melanosomal pH, which is phylogenetically associated with the SLC13 family (see the SLC13 family description). OCA type 2, the most common form of OCA, is an autosomal recessive disorder caused by mutations in the OCA2 gene (2937); both SLC45A2 and OCA2 have been shown to increase luminal organellar pH (2929). It was concluded that SLC45A2 maintains the melanosome neutralization initially orchestrated by OCA2 to support melanization at late stages of melanosome maturation (2929).

Furthermore, it was shown that the common light skin-associated SLC45A2 allelic variant L374F has reduced activity due to increased protein instability, and that it restores only moderate pigmentation in SLC45A2-deficient melanocytes due to rapid proteasome degradation resulting in lower protein expression levels in melanosomes than the dark skin-associated allelic variant L374 (2929).

SLC45A3: SLC45A3 is a H⁺-coupled glucose and fructose transporter (2922). It was originally called prostein because of its very high expression in normal and cancerous prostate. Besides the very high and almost exclusive expression in prostate cells, the HPA also suggests low expression in spleen and spinal cord and at the single cell level in granulocytes and oligodendrocytes.

SLC45A3 plays an important role in the development of prostate cancer. The majority of prostate cancers harbor recurrent gene fusions between the hormone-regulated TMPRSS2 and members of the ETS (E-26 transformation-specific) family of transcription factors, most commonly ERG (ETS-related gene) (2938, 2939). ERG is an oncogene that regulates cell proliferation, differentiation, and metastasis. Previous studies have found that ~55% of prostate cancer patients have ERG overexpression driven by fusion of the ERG gene with androgen response genes such as TMPRSS2 (transmembrane serine protease 2). Thus, TMPRSS2::ERG fusion is the most common fusion found in prostate cancer and its oncogenic role and regulatory mechanisms have been well studied. In addition, SLC45A3 is the second most common 5’ partner gene in ERG rearrangements (2939). While TMPRSS2 was the only 5’ partner in about 78% of ERG fusion prostate cancers, SLC45A3 was the only 5’ partner in about 6%. Interestingly, concurrent TMPRSS2 and SLC45A3 fusions to ERG were also found. This occurred in about 11% of ERG fusion-positive cancers (2939). Studies on the molecular pathological features of SLC45A3::ERG fusion-positive prostate cancer have been initiated to uncover cellular signaling pathways and provide insight into potential therapeutic strategies (2940).

A role for the microRNA miR-32 in oligodendrocyte function and development through regulation of SLC45A3 has also been reported (2941). Oligodendrocytes produce large amounts of myelin as an extension of their cell membrane, and lipids are the major components of myelin. Therefore, understanding lipid metabolism involved in maintaining myelin is important. miR-32 is highly expressed in the myelin-enriched regions of the brain and mature oligodendrocytes and promotes myelin protein expression. miR-32 was shown to directly regulate the expression of SLC45A3 by binding to the complementary sequence on the 3’ UTR. As a myelin-enriched putative sugar transporter, SLC45A3 is postulated to enhance intracellular glucose levels and synthesis of long-chain fatty acids, and tight regulation of SLC45A3 expression is necessary for the proper maintenance of myelin proteins and structure (2941).

Expression of the ortholog Slc45a3 has also been reported in the medullary collecting duct of mouse and rat kidney, and in vitro studies have shown that SLC45A3 expression is increased several-fold under hyperosmotic conditions. Therefore, it has been proposed that SLC45A3 functions as a novel osmolyte transporter in rodent kidney, acting as a H⁺-coupled sugar transporter in the apical membrane (2942). Verification whether this osmolyte transporter function of SLC45A3 is also pertinent in human kidney is yet to be demonstrated.

SLC45A4: SLC45A4 is a H⁺-coupled sugar and polyamine transporter that is ubiquitously expressed (2922, 2943, 2944).

SLC45A4 has been shown to be involved in the H⁺-coupled uptake of sugars in spermatozoa, where it has been localized in the principal piece of the spermatozoa. There, it was proposed to play a role in the nutrition of spermatozoa during their maturation in the epididymis by providing fructose and glucose, which are abundant in the male reproductive tract (2944). Since mammalian spermatozoa are exposed to a slightly acidic environment in the epididymis, a H⁺-coupled transport mechanism makes perfect sense to provide nutritional support to spermatozoa in the epididymis (2944).

Subsequently, a metabolic mapping approach of the SLC-ome revealed that SLC45A4 functions primarily as a polyamine transporter (1894). This makes sense, since polyamines play an important role in spermatogenesis and in the motility of sperm (2945).

In pancreatic cancer, SLC45A4 has been reported to affect cancer progression by facilitating glucose uptake and glycolysis of tumor cells (2946). Specifically, SLC45A4 was shown to promote glycolysis and prevent AMPK (AMP-activated protein kinase)/ULK1 (serine/threonine kinase)-induced autophagy in TP53-mutant pancreatic ductal adenocarcinoma. Knockdown of SLC45A4 reduced glucose uptake and ATP production, which led to the activation of autophagy via the AMPK/ULK1 pathway, providing an alternative energy source for cancer cells to rapidly resist nutritional stress (2946). In addition, high level of SLC45A4 expression was closely associated with poor clinical outcome in PDA patients.

SLC45A4 also plays an important role in epithelial ovarian cancer, where it shows remarkable upregulation compared to normal ovarian tissue (2943). Among the SLC45 family members, SLC45A4 was reported to have the highest expression in ovarian cancer, and its overexpression correlated with an unfavorable prognosis in patients diagnosed with ovarian cancer. SLC45A4 knockdown inhibited glucose uptake and production of ATP and lactic acid, affected glycolytic metabolism of ovarian cancer, and inhibited malignant behavior of tumor cells (2943).

SLC45A4 may provide promising therapeutic strategies for the treatment of pancreatic and ovarian cancer given its role in glycolytic metabolism.

Orphan transporter family members: N/A

SLC46 Folate transporter family (2.A.1.50/MFS_1/MFS)

Discovery: The molecular basis for the proton-dependent folate transporter was established with the previous cloning of the gene encoding PCFT, designated SLC46A1 (1686, 2947).

Gene family members (3):
SLC46A1 (PCFT)	SLC46A2 (TSCOT)	SLC46A3

Molecular aspects, physiological roles and links to disease

The SLC46 family belongs to the Proton-Coupled Folate Transporter/Heme Carrier Protein (PCFT/HCP) family (TC 2.A.1.50), which is part of the MFS superfamily. PCFT (SLC46A1) is an H⁺-coupled plasma membrane folate transporter (Qiu, 2006, #206), while TSCOT (SLC46A2) and SLC46A3 are likely H⁺-coupled endosomal or lysosomal exporters (1496, 2948, 2949). However, SLC46A2 and SLC46A3 are still “semi-deorphanized” transporters in the sense that their functional properties, including H⁺ coupling, are incompletely characterized, and they export chemically and structurally diverse groups of compounds, which requires further experimental clarification. The cryo-EM structure of PCFT (SLC46A1) has been reported and reveals the classical MFS structural architecture with 12 TMHs (see below)(2950).

SLC46A1: PCFT (SLC46A1) was originally reported to be a pH-independent heme carrier protein (HCP1) (2951). However, a careful functional reassessment revealed that PCFT (SLC46A1) is a H⁺-coupled folate transporter required for folate homeostasis in humans rather than a heme transporter (Qiu, 2006, #206).

As highlighted in the description of the SLC19 family, folate is crucial for numerous bodily functions. During early pregnancy, it is crucial for the development of the neural tube and insufficient folate intake can increase the risk of neural tube defects like spina bifida and anencephaly. The active form of folate, tetrahydrofolate (THF), and its derivatives act as one-carbon donors in several key metabolic pathways. These pathways are required for processes such as nucleic acid synthesis and the formation of activated methyl groups necessary for DNA methylation.

There are three folate transport systems in humans: 1) endocytosis via folate receptor alpha FRα (FOLR1); 2) transport via reduced folate carrier RFC (SLC19A1); and 3) transport via the H⁺-coupled folate transporter PCFT (SLC46A1) (1686, 1687). PCFT (SLC46A1) is required for H⁺-coupled folate uptake across the brush border membrane of the small intestine. It has a pH optimum of 5.0–5.5 (Qiu, 2006, #206). This is consistent with the slightly acidic pH of the unstirred layer at the extracellular surface of the brush border membrane. In contrast, RFC operates as a folate anion antiporter at neutral pH (2952).

PCFT (SLC46A1) is expressed in most tissues but is present at the highest levels in the intestine, where it is localized to the apical brush border membrane of enterocytes in the duodenum and proximal jejunum (Fig. 30), as well as in the choroid plexus, where it is present at the basolateral membrane as well as in intracellular compartments (Fig. 11) (1701, 2953). Thus, folates absorbed in the small intestine by PCFT (SLC46A1) can also be delivered into the cerebrospinal fluid using the same transporter. Other major sites of PCFT expression include liver, placenta, spleen, and retinal pigment epithelium (RPE). In the liver, it is expressed in the sinusoidal membrane, and in the placenta at the microvillous plasma membrane of syncytiotrophoblasts (1686, 2947, 2954).

Loss-of-function mutations in the SLC46A1 gene have been shown to cause hereditary folate malabsorption, an autosomal recessive disorder resulting in severe folate deficiency. This deficiency manifests as anemia, as well as immunological and neurological abnormalities (1701) (2947) (2955). An Slc46a1-null mouse was shown to exhibit a folate deficiency phenotype similar to that observed in humans with loss-of-function mutations in the SLC46A1 gene (2956). In contrast, loss-of-function of the folate receptor FRα (FOLR1), which is almost exclusively expressed in the choroid plexus, results in cerebral folate deficiency only (1701, 2957).

The H⁺-coupled folate transporter, PCFT (SLC46A1), also transports antifolates such as methotrexate and pemetrexed, which are chemotherapeutic agents that block the effects of folic acid. PCFT is therefore a pharmacologically important transporter that works in parallel with the reduced folate carrier (RFC/SLC19A1) and the high-affinity folate receptors (FR) α and β. These transport systems all mediate cellular accumulation of folate and its derivatives/antifolates, albeit with different substrate specificities and mechanisms (2958).

While antifolate drugs, such as methotrexate and pemetrexed, are established treatments for cancer and autoimmune diseases, their efficacy is unfortunately limited, since antifolate drugs are rapidly transported by RFC into normal tissues as well, causing toxicity (2959) (1689, 2960). In contrast, given that the acidic microenvironment of cancer cells increases PCFT-mediated transport (2961), targeting PCFT (SLC46A1) selectively would be a promising approach for developing new antifolate agents. Until recently, however, the molecular basis by which PCFT or RFC distinguish between folates and antifolates remained unclear.

To advance this field, cryo-electron microscopy structures of chicken PCFT in a substrate-free state and in complex with the antifolate drug pemetrexed have been determined (2950). The transporter exhibits the characteristic MFS architecture of transporters, comprising 12 TMHs. The results from this study provide a structural basis for understanding antifolate recognition and offer insights into the H⁺-coupled mechanism, which is particularly relevant in acidic environments such as the tumor microenvironment (2950). Understanding the structural basis of antifolate recognition by PCFT is an important step toward designing new, effective antifolate drugs transported into cancer cells by the H⁺-coupled folate transporter PCFT (SLC46A1).

Building on this structural progress, a detailed analysis has been reported on how specific mutations in the human SLC46A1 gene cause hereditary folate malabsorption (2962). The study used molecular dynamics simulations to examine PCFT (SLC46A1). Significant structural alterations in PCFT were found to be caused by certain pathogenic mutations. The study also explored compensatory mutations that can restore the structural stability and function of PCFT, offering potential therapeutic avenues for hereditary folate malabsorption.

SLC46A2: TSCOT, short for “thymic stromal cotransporter” (SLC46A2), is a lysosomal transporter reported to be highly expressed in mouse thymic cortical epithelial cells (1496, 2948, 2963). The HPA also reports strong expression in the thymus, but additionally shows high expression in the skin and the cervix, and lower levels in the renal tubules and the epididymis. Tracheal cytotoxin (TCT) is produced by Bordetella pertussis and Neisseria gonorrhoeae, and SLC46A2 was shown to promote TCT-triggered NOD1 activation in human epithelial cell lines, which suggested that SLC46A2 is a peptidoglycan transporter that contributes to cytosolic immune recognition (1496).

Both SLC46A2 and SLC46A3 have been implicated in bacterial cell wall peptidoglycan uptake and NOD signaling, for example in mammalian keratinocytes (1495, 1496) (see also the SLC15A4 description): as reviewed (2964), based on an NF-κB luciferase assay, human and mouse SLC46A2 and mouse SLC46A3 transporters were shown to induce a NOD-dependent NF-κB response in HEK-293T and HCT-116 cells stimulated with the peptidoglycans TCT and muramyl dipeptide (MDP). In particular, SLC46A2 is proposed to play a critical role in the transport of TCT, whereas SLC46A3 is proposed to play a critical role in the transport of MDP. SLC46A2 has been localized to late endosomes (1496). In a mouse model, Slc46a2 and Nod1 deficiency strongly suppressed psoriatic inflammation, while methotrexate, a commonly used psoriasis therapeutic, inhibited SLC46A2-dependent transport of DAP muropeptides (2965). Taken together, these studies identify SLC46A2 as a transporter of NOD1-activating muropeptides with a critical role in the skin barrier. Thus, this transporter may serve as an important target for anti-inflammatory intervention.

SLC46A2 is also required for the cellular import of the tumor-derived cyclic dinucleotide cGAMP (2′3′-cyclic GMP-AMP) into host macrophages and monocytes as part of the Stimulator of Interferon Genes (STING, STING1) sensing mechanism of pathogens, i.e., sensing the tumor-derived cyclic dinucleotide that acts as a danger signal (1495, 2966). STING is an ER-associated membrane protein which is critical for innate immune sensing of pathogens. Its activation by cGAMP is followed by the activation of TBK1 (TANK-binding kinase 1) and IRF3 (interferon regulatory factor 3), leading to the expression of inflammatory cytokines such as IFN-I (2967–2970). Thus, cGAMP acts as an immunotransmitter that is transferred from cancer cells to cGAMP-sensing cells in the host, such as macrophages and monocytes, thereby promoting immunity. The cellular import of cGAMP has been reported to be critical for the induction of antitumor immunity, and optimizing dinucleotide-derived therapeutics to specifically target the import mechanism through SLC46A2 may result in more effective anticancer therapeutics. Note that in the above-mentioned study, SLC46A2 is reported to act as a plasma membrane importer of cGAMP, although another study on peptidoglycans localized it to lysosomes (1496). Indeed, it seems more reasonable to predict that cGAMP first enters the endo-lysosomal system via endocytosis, followed by release into the cytosol via SLC46A2, where it then activates the STING dimer from the cytosolic side. Although the STING dimer is located on the ER membrane, its activation side is facing the cytosol (2970, 2971), so endosomal exit via SLC46A2 would be a prerequisite for STING activation from the cytosol.

SLC46A3: In addition to lysosomal peptidoglycan transport (see above), SLC46A3 has been shown to be a lysosomal H⁺-coupled steroid conjugate and bile acid transporter (2949, 2972). It is ubiquitously expressed as suggested by the HPA and has a tyrosine-based lysosomal sorting motif at its C-terminus (2949). SLC46A3 has been described to function as an H⁺-coupled steroid conjugate and bile acid transporter and has been shown to preferentially recognize lipophilic steroid conjugates and bile acids as endogenous substrates (2949).

SLC46A3 has been identified as a critical player in the treatment of patients with HER2-positive breast cancer using the antibody-drug conjugate T-DM1. T-DM1 combines the humanized antibody trastuzumab and the potent anti-microtubule agent emtansine, also known as DM1, a derivative of maytansine, using a unique and highly stable linker (2973). One of the active metabolites of T-DM1 is Lys-SMCC-DM1 (lysine-N_ε-N-succinimidyl-4-(N-maleimidomethyl)cyclohexane-1-carboxylate-DM1), which was shown to be a transported substrate of SLC46A3 (2949). Therefore, it was proposed that this transporter contributes to the delivery of this active metabolite of T-DM1 across the lysosomal membrane into the cytosol. Consistent with this, the cytotoxic effect of T-DM1 was significantly reduced in the presence of SLC46A3 inhibitors.

A first step in the pharmacological action of T-DM1 involves the endocytic pathway with internalization of T-DM1 after binding to the epidermal growth factor receptor HER2 expressed on the surface of breast cancer tumor cells. This is followed by degradation in endosomes/lysosomes to yield the linker-conjugated payload Lys-SMCC-DM1 and its release from lysosomes into the cytosol via SLC46A3. Since Lys-SMCC-DM1 has hydrophilic properties and a large molecular size, lysosomal exit of Lys-SMCC-DM1 is likely to be a rate-limiting step in T-DMT anticancer treatment. Therefore, SLC46A3 is a crucial determinant of the efficacy of T-DM1 anti-cancer treatment (2949).

Using fluorescent compounds as substrates of SLC46A3, a fluorescence-based assay system was developed as a valuable tool to evaluate the interaction of drugs and drug candidates with SLC46A3. This may also be instrumental in developing drug delivery strategies via endocytosis by exploiting SLC46A3 (2972).

In another study, SLC46A3 was implicated in the hepatic toxicity with cytosolic copper deficiency and marked lipid accumulation caused by the environmental contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) (2974). Studies in wild-type and Slc46a3 knockout mice revealed that lysosomal SLC46A3 expression is induced by TCDD through the aryl hydrocarbon receptor (AHR) (2974). AHR is a ligand-activated transcription factor that responds to environmental, dietary, and metabolic signals and TCDD is known to be one of the highest affinity agonists for AHR, causing adverse human health effects (2975, 2976). TCDD-triggered AHR signaling has been implicated as a cause of hepatic lipid accumulation (2974). In particular, it has been reported that TCDD-induced expression of SLC46A3 leads to an increase in lysosomal copper, which would then be released into the bile, resulting in hepatic copper deficiency, and it has further been proposed that SLC46A3 itself is a copper transporter. Copper deficiency would then lead to mitochondrial dysfunction with reduced lipid catabolism, resulting in hepatic lipid accumulation (2974). However, the reported substrate preference of SLC46A3 for lipophilic steroid conjugates and bile acids (see above) seems inconsistent with SLC46A3 functioning as a copper transporter. As an alternative, the effects of TCDD on SLC46A3 expression might lead to activation of the Wilson disease copper exporter ATP7B, which is known to be involved in copper export from liver cells into the bile (2431, 2977).

Orphan transporter family members: N/A

SLC47 Multidrug and Toxin Extrusion (MATE) family (2.A.66.1/MatE/MATE)

Discovery: MATE (multidrug and toxic compound extrusion) proteins belong to a large family of secondary active transporters involved in the transfer of various compounds across cellular and organellar membranes (2978). The mammalian SLC47 family members have been identified as homologs of the NorM Na⁺/multidrug antiporter of Vibrio parahaemolyticus. NorM was originally identified as a drug efflux protein from the chromosomal DNA of V. parahaemolyticus by using an E. coli mutant lacking the major multidrug efflux system AcrAB as the host, screening for plasmids that carry a gene responsible for norfloxacin efflux (2979, 2980). The SLC47A1 and SLC47A2 genes encoding MATE1 and MATE2 were then identified in 2005 (2981).

Gene family members (2):
SLC47A1 (MATE1)	SLC47A2 (MATE2)

Molecular aspects, physiological roles and links to disease

MATE1 (SLC47A1) and MATE2 (SLC47A2) belong to the Multi Antimicrobial Extrusion/Multidrug and Toxin Extrusion (MATE) family (TC 2.A.66.1), which is part of the Multidrug/Oligosaccharidyl-lipid/Polysaccharide (MOP) Flippase superfamily. MATE transporters typically contain 12 TMHs arranged in two pseudosymmetric six-helix bundles that represent the MATE fold (see below under “Structural Insights and Pharmaceutical Perspectives”). The MATE transporters catalyze the active efflux of a wide range of chemically and structurally diverse compounds, including the anti-diabetic drug metformin, the antimalarial compound chloroquine, antimicrobials, and chemotherapeutics such as cisplatin and oxaliplatin (302, 2978, 2982).

SLC47A1: MATE1 (SLC47A1) is a H⁺-dependent antiporter that mediates bidirectional plasma membrane transport of primarily cationic endogenous substrates and xenobiotics (302). The major physiological role of MATE1 is the excretion of substrates by facilitating their efflux from polarized epithelial cells at the apical membranes of the kidney and liver. In the kidney, MATE1 as well as MATE2 are expressed in the apical membranes of the renal tubules where they participate in the clearance of cationic drugs from the body via H⁺/organic cation antiport (2978). In the liver, MATE1 is expressed on the canalicular membrane of hepatocytes. The HPA also suggests expression in the adrenal gland, intestine and lung.

An important focus of investigation for MATE1 has been on its role in renal drug disposition and excretion, specifically the renal elimination of the widely used oral antidiabetic drug metformin and the renal toxicity of cisplatin (1926). MATE1 is critical for the renal and biliary excretion of metformin, and it turns out that clinical inhibitors of organic cation transporters (OCTs, SLC22 family) are also inhibitors of MATEs, and thus modulation of the activity not only of OCTs but at least as much of MATEs determines drug-drug interactions (DDIs) previously attributed solely to OCTs.

In a study of platinum-acridine agents, which are among the most potent platinum-containing anticancer agents, compounds with high activity in cancers expressing MATE1 were sought (2983). Two compounds were identified that showed a high requirement for MATE1 transport. This observation may lead to therapeutic applications for the treatment of tumors expressing MATE1 to overcome resistance to existing therapies.

SLC47A2: MATE2 (SLC47A2) and its splice isoform MATE2-K are exclusively expressed in the apical membrane of proximal tubule cells (1864, 2984). Like MATE1, it is a H⁺/organic cation antiporter that functions as an efflux transporter of organic cations. The major focus, similar to MATE1, has been on the renal excretion of endogenous and exogenous organic cations, particularly metformin (1926).

Orphan transporter family members: N/A

SLC48 Heme transporter family (2.A.110.1/HRG/unknown)

Discovery: The SLC48 family consists of only one human gene, SLC48A1, which encodes a facilitative transporter, HRG-1 (heme responsive gene-1) (2985). This protein has 4 TMDs and therefore does not appear to have the structural of classical membrane transporter. HRG-1 was identified as part of an effort to exploit auxotrophy in C. elegans. The goal was to identify proteins essential for heme homeostasis and normal development in worms and vertebrates. HRG-1 was identified as a previously unknown transmembrane protein located in distinct intracellular compartments (2986). Transient knockdown of hrg-1 (slc48a1b) in zebrafish resulted in hydrocephalus, yolk tube malformations and profound defects in erythropoiesis—phenotypes that are fully rescued by worm HRG-1 (2986).

Gene family member (1)

SLC48A1 (HRG-1)

Molecular aspects, physiological roles and links to disease

The SLC48 family belongs to the Heme Transporter, Heme-Responsive Gene Protein (HRG) family (TC 2.A.110). As indicated above, it is believed to have 4 TMHs (2985–2987), but its structure is unknown. Residues H56 and H100 and the YAHRY motif in the C-terminus are predicted to be important for heme transport (2986).

SLC48A1: HRG-1 (SLC48A1) is associated with endolysosomal membranes and transports heme from endosomes or lysosomes into the cytosol. It is the long-sought heme transporter for heme iron recycling in macrophages (2986–2988). HRG-1 is highly expressed in macrophages of the reticuloendothelial system and localizes specifically to phagolysosomal membranes during erythrophagocytosis. Depletion of slc48a1b in mouse macrophages attenuated heme transport from the phagolysosomal compartment, and missense polymorphisms in human HRG-1 were found to be defective in heme transport (2988). Given the similarities between heme utilization from senescent erythrocytes in macrophages and heme transport from a red meat-rich diet in enterocytes, it was suggested that HRG-1 may play a role in heme iron absorption in the gut.

According to the HPA, HRG-1 (SLC48A1) is expressed ubiquitously, with most prominent expression in brain areas (especially in the spinal cord), endocrine tissues (especially in the thyroid), kidney, ovary and spleen, and moderate levels in the GI tract (especially stomach and colon) and testis. At the single cell level, it shows highest expression in oligodendrocytes, erythroid cells and kidney collecting duct cells. Further work is needed to clarify the role of HRG-1 in these different tissues and cell types and to elucidate the mechanism by which it transports heme.

Orphan transporter family members: N/A

SLC49 FLVCR-related transporter family (2.A.1.28/MFS_1/MFS)

Discovery: FLVCR1 (SLC49A1), the cell surface receptor for the anemia-inducing subgroup C of the feline leukemia virus, also known as MFSD7B, was originally reported to be a heme exporter in hematopoietic cells (2985). This receptor seemed critical for the development of erythroid progenitors. Subsequently, FLCVR1 was cloned and characterized, revealing its membership of the MFS permease superfamily (2989, 2990). Disruption of FLVCR1 function was found to block erythroid progenitor development (2985), and it was suggested to be due to a lack of mitochondrial heme exit resulting in mitochondrial heme toxicity (2985). While FLVCR1 was long believed to play an important role in exporting heme from mitochondria, especially during erythropoiesis, subsequent studies revealed that FLVCR1 is a plasma membrane choline and ethanolamine transporter rather than a mitochondrial heme exporter (2890, 2991). Instead, studies revealed that TMEM14C likely represents the mitochondrial heme exporter in erythroid cells, where it is strongly expressed (see the TMEM14C description in Section 10, “SLC-like Proteins”).

The true reason that FLVCR1 dysfunction blocks the development of erythroid progenitors has finally come to light. It is not due to a lack of mitochondrial heme exit resulting in heme toxicity. Rather, it is due to the high demand for choline and ethanolamine during the terminal differentiation of erythroblasts when they reorganize their membranes using phosphatidylcholine (PC) and phosphatidylethanolamine (PE) (2992, 2993).

These findings highlight how redefining the substrates of SLCs can lead to a major shift in our understanding of physiological processes.

Gene family members (4):
FLVCR1 (SLC49A1)	SLC49A3 (MFSD7)
FLVCR2 (SLC49A2)	SLC49A4 (DIRC2)

Molecular aspects, physiological roles and links to disease

The SLC49 family has 4 members (Fig. 50) and belongs to the Feline Leukemia Virus Subgroup C Receptor (FLVCR)/Heme Importer family (TC 2.A.1.28) which is part of the MFS superfamily. The SLC49 family members have 12 TMHs.

Fig. 50. Phylogenetic tree of the SLC49 family.

FLVCR1 (SLC49A1): FLVCR1 (SLC49A1) is a plasma membrane choline and ethanolamine transporter that is ubiquitously expressed with high expression in intestine, liver, kidney, brain, and bone marrow (2890, 2985). Mutations in the FLVCR1 gene have been identified in individuals with posterior column ataxia with retinitis pigmentosa and hereditary sensory and autonomic neuropathies. A pooled CRISPR screen of genes involved in PC metabolism revealed the unexpected role of FLVCR1 as a choline transporter (2991).

Further supporting the role of FLVCR1 as an essential transporter of choline and ethanolamine is the study of biallelic variation in FLVCR1, which underlies a spectrum of severe developmental disorders. The study revealed that FLVCR1 is necessary for both early development and long-term neurological health. Severe loss of function results in developmental failure and early mortality, whereas milder loss increases the risk of later neurodegenerative diseases.

Choline uptake is essential for the survival of mammalian cells, and while the known choline transporter SLC44A2 is constitutively expressed in K562 cells, it is not required for survival, whereas FLVCR1 (SLC49A1) plays a critical role in choline uptake in K562 cells and is essential for cell proliferation.

An integrative genetic analysis further confirmed the role of FLVCR1 as a plasma membrane choline transporter in mammalian cells (2994). In addition, loss of FLVCR1-mediated choline uptake in cells impaired phosphocholine metabolism, resulting in structural defects in mitochondria and upregulation of the integrated stress response by heme-regulated inhibitor, one of the kinases reported to be activated in response to heme deprivation (2994, 2995). These results further demonstrate that the FLVCR1 (SLC49A1) choline transporter is of vital importance in mammalian cells.

A subsequent study addressed the structure, transport mechanism, and molecular basis of lipid head group entry into the Kennedy pathway via FLVCR1 (2890). As part of the MFS superfamily, FLVCR1 has two groups of six TMHs, which are linked by an extended intracellular linker. FLVCR1 uses the conserved rocker-switch, alternating-access transport mechanism, whereby the two six-TMH bundles rock around a central substrate-binding site. This allows access to the substrate from either the extracellular or intracellular side of the membrane. FLVCR1 has been shown to transport extracellular choline as well as ethanolamine into the cytosol, where they can be phosphorylated by the enzymes Choline kinase alpha (CKα) and ethanolamine kinase 1 (CEK1). These enzymes catalyze the initial steps of the choline and ethanolamine branches of the Kennedy pathway, respectively (2890). Unlike other putative mammalian ethanolamine transporters, such as CTL1 (SLC44A1) and CTL2 (SLC44A2), which have affinities far below the ~2 μM circulating plasma ethanolamine concentration, the K_m of FLVCR1 for ethanolamine (2.8 μM) is proposed to be sufficiently high to enable cellular uptake under physiological conditions (2890). Structure-guided mutagenesis identified residues critical for ethanolamine transport yet dispensable for choline transport. This enabled the functional separation of entry points into the two branches of the Kennedy pathway (2890).

Studies have also indicated that FLVCR1-mediated choline uptake plays a key role in human aging (2996). This study found that exceptional longevity is accompanied by significant changes in gut microbiota and metabolite profiles. The genetic link between FLVCR1 and plasma choline levels across ages underscores the pivotal role of the protein in age-related metabolic maintenance.

FLVCR2 (SLC49A2): FLVCR2 (SLC49A2), also known as MFSD7C, is a choline transporter that is responsible for most choline uptake into the brain (2997). Its functional properties are similar to those of FLVCR1.

FLVCR2 is reported to be highly expressed in BBB endothelial cells throughout development and into adulthood (2998, 2999). In humans, FLVCR2 mutations cause proliferative vasculopathy and hydranencephaly hydrocephalus (PVHH), also known as Fowler syndrome. This rare autosomal recessive brain vascular disorder is associated with impaired cerebral angiogenesis, hydrocephalus, and embryonic lethality (3000, 3001). Mice with endothelial cell-specific knockout of Flvcr2 exhibit PVHH-like phenotypes (2999, 3002). These findings demonstrate that endothelial-expressed FLVCR2 is essential for cerebral angiogenesis and normal brain development.

Subsequently, FLVCR2 was reported to be a BBB choline transporter that supplies the brain with choline (2997). Additionally, the cryo-EM structure of choline-bound FLVCR2 in both the inward- and outward-facing states was determined (2997). This provided insight into how FLVCR2 binds choline in an aromatic cage and mediates its uptake. The results of this study are also expected to offer new approaches for delivering therapeutic agents directly to the brain.

In addition to the BBA, the HPA indicates that FLVCR2 (SLC49A2) is abundantly expressed in various other human tissues, with the highest expression occurring in the choroid plexus. However, the role of FLVCR2 in choroid plexus epithelial cells, particularly in the context of brain choline homeostasis and altered cerebrospinal fluid dynamics due to FLVCR2 mutations associated with Fowler syndrome (3003), is unknown.

SLC49A3 - Orphan transporter: MFSD7 (SLC49A3) was identified as 1 of 14 genes that are predictive of time to relapse of ovarian cancer following therapy (3004) (for a review see (2985)). A follow-up study identified a promoter region SNP that correlates with a reduced risk of invasive ovarian cancer (3005). According to the HPA, MFSD7 is widely expressed. However, functional data on this transporter is lacking, and its subcellular localization remains unknown.

SLC49A4 - Orphan transporters: DIRC2 (disrupted in renal carcinoma 2, SLC49A4) was initially identified as a breakpoint-spanning gene in a chromosomal translocation that is potentially associated with renal cancer development (3006) (for a review see (2985)). Several approaches have demonstrated its lysosomal localization (3007). The lysosomal targeting of DIRC2 has been shown to be mediated by an N-terminal dileucine motif. It has been speculated that DIRC2 is a lysosomal metabolite transporter. However, the transport substrate remains unknown (3007). The HPA suggests relatively wide tissue distribution, with the highest expression occurring in syncytiotrophoblasts and monocytes.

Sequence alignments and structural analyses have been performed to compare MFSD7 and DIRC2 with FLVCR1 and FLVCR2, revealing conserved regions of potential functional importance (2890).

Orphan transporter family members (2)

SLC49A3 (MFSD7), SLC49A4 (DIRC2)

SLC50 Sugar efflux transporters (2.A.123.1/MtN3_slv/SWEET)

Discovery: A new class of sugar transporters, called SWEETs (Sugar Will Eventually be Effluxed Transporters), has been identified in HEK293T cells using a fluorescent intracellular glucose sensor to measure glucose transport, which is essential for maintaining blood glucose levels in animals, nectar production in plants, and seed and pollen development in plants (315). The identity of these sugar efflux transporters has long been elusive. The SWEETs have at least six homologs in Arabidopsis, two in rice, two in Caenorhabditis elegans, and only one in humans (SWEET1/SLC50A1).

Gene family member (1)

SLC50A1 (SWEET)

Molecular aspects, physiological roles and links to disease

The SLC50 family belongs to the “Sweet; PQ-loop; Saliva; MtN3 (Sweet)” family (TC 2.A.123), which is part of the SWEET fold family, whose members have 7 TMHs in a 3+1+3 repeat arrangement (see Section 8).

SLC50A1: Vertebrate SWEET (SLC50A1) is ubiquitously expressed and has been proposed to be involved in sugar efflux from cells of the intestine, liver, epididymis, and mammary gland (215, 315). However, the full range of its functions in animals is not yet clear.

Human SWEET did not promote glucose uptake but mediated weak efflux when expressed in yeast or Xenopus oocytes. Upon expression in HEK293T cells, human SWEET was predominantly expressed at the Golgi with minimal expression at the plasma membrane. Microarray analysis showed the highest level of expression in the oviduct, epididymis and intestinal tract, and expression was induced in the mouse mammary gland during lactation, suggesting that SWEET serves to provide glucose for lactose synthesis in the mammary gland (315).

It has also been proposed that SWEET in the basolateral membrane of enterocytes contributes to glucose efflux into the blood in a parallel pathway mediated by GLUT2 (SLC2A2). The SWEET pathway would involve glucose uptake into the Golgi via SWEET as part of a vesicular efflux pathway (Chen, Hou et al. 2010). Further studies are needed to fully elucidate impact of SWEET on human physiology.

Orphan transporter family members: N/A

SLC51 Transporters of steroid-derived molecules (2.A.82/Solute_trans_a/unknown)

Discovery: OSTα (SLC51A) and OSTβ (SLC51B) were identified in 2001 by expression cloning. A liver cDNA library from a primitive marine vertebrate, the little skate Raja erinacea, was screened for taurocholate uptake in Xenopus laevis oocytes (155). In this library, transport activity was identified that required the co-expression of two different gene products, the organic solute carrier Ostα and the ancillary protein Ostβ. However, there is no sequence similarity between the two. Subsequently, Ostα (SLC51A) was characterized in human and mouse (3008, 3009).

Gene family members (4 + 1 auxiliary protein/beta-subunit):
SLC51A (OSTα)	TMEM184A (SLC51C1)
SLC51B (OSTβ)	TMEM184B (SLC51C2)
(auxiliary protein/beta-subunit of OSTa)	TMEM184C (SLC51C3)

Note: The SLC51B assignment represents a nomenclature exception because SLC51B is not related to SLC51A by sequence similarity, but rather functions as an ancillary protein and β-subunit of the actual organic solute carrier α-subunit. To denote new, phylogenetically distant SLC51 subfamily members, the root SLC51C was introduced (e.g., SLC51C1 for TMEM184A).

Molecular aspects, physiological roles and links to disease

As mentioned above, the SLC51 transport system consists of two polypeptide chains, the actual transport protein α and the ancillary protein β (see below) (3009, 3010). SLC51 α (i.e., SLC51A) is a member of the Organic Solute Transporter (OST) family (TC 2.A.82). Orthologs of OSTα (SLC51A) are found in a wide variety of eukaryotes, including animals (both vertebrates and invertebrates), plants, fungi, and slime molds (3011). Orthologs of the ancillary protein OSTβ (SLC51B) are thought to exist only in vertebrates. If true, then invertebrate OSTα orthologs may either not require an additional partner protein or use a different partner protein (3011). OSTα (SLC51A) has 7 putative TMHs (3009, 3010).

The three transmembrane transporters TMEM184A (SLC51C1), TMEM184B (SLC51C2), and TMEM184C (SLC51C3) are orphan transporters that share a domain structure similar to organic solute transporters, specifically Ostα (SLC51A) (3012) (Fig. 51).

Fig. 51. Phylogenetic tree of the SLC51A+C family.

According to UniProt, SLC51A, SLC51C1, SLC51C2, and SLC51C3 are predicted to have 7 TMH. The phylogenetic relationship of these transporters is shown in Fig. 51. Currently, there are no experimentally determined 3D structures available for SLC51 family members.

Interestingly, SLC51C1, SLC51C2, and SLC51C3 have an extended C-terminal region of up to ~100 amino acids that is missing in SLC51A. In SLC51C1, this C-terminal region includes a lysine- and arginine-rich region proposed to serve as a heparin-binding site (see below).

The functional roles of TMEM184A, TMEM184B, and TMEM184C in transport versus receptor function (e.g., as a heparin receptor) and their potential association with ancillary proteins, such as OSTβ, require further investigation.

SLC51A, SLC51B: OSTα (SLC51A), in complex with its obligate partner OSTβ (SLC51B), forms a heteromeric transporter that is primarily involved in the facilitated transport of organic anions such as bile acids.

OSTα (SLC51A) is a membrane glycoprotein with 7 TMHs, while OSTβ (SLC51B) is a type 2 membrane protein with a single TMH located near its N-terminus. OSTα requires interaction with OSTβ to reach the plasma membrane. Therefore, OSTβ functions as an ancillary protein of OSTα. The glycosylation of OSTα is not required for the interaction with the beta subunit, for the membrane localization, or the function of the heteromeric transporter (3013). Xenopus oocytes injected with cRNA for both Ostα (SLC51A) and Ostβ (SLC51B), but not each separately, were able to take up taurocholate, estrone sulfate, digoxin and prostaglandin E₂, but not p-aminohippurate or S-dinitrophenyl glutathione (155). Transport was sodium-independent and saturable.

OSTα (SLC51A) in complex with OSTβ (SLC51B) functions as an organic anion transporter that facilitates electroneutral diffusion (3009). It is trans-stimulated by known substrates, consistent with a facilitated diffusion transport mechanism (3014). Typical substrates include bile acids (taurocholate, glycocholate), steroid conjugates (e.g., estrone-3-sulfate), drug metabolites and other xenobiotics.

OSTα/β proteins are expressed in various tissues including small intestine, colon, liver, biliary tract, kidney and adrenal gland. In polarized epithelial cells, they localize to the basolateral membrane and function in the export or uptake of bile acids and steroids (3011). OSTα/β proteins typically export bile acids from enterocytes across the basolateral membrane into the portal circulation during intestinal reabsorption as part of the enterohepatic circulation.

Ostβ is not only required for heterodimerization and trafficking, but is also required for function (3015). Studies have shown that OSTα/β is upregulated in liver tissue from patients with extrahepatic cholestasis, obstructive cholestasis, and primary biliary cholangitis, conditions characterized by elevated bile acid concentrations in the liver and/or systemic circulation (3016, 3017) (see Fig. 33). This finding highlights clinical relevance of this transporter as non-alcoholic steatohepatitis (NASH) becomes more common with increasing prevalence of obesity. OSTα/β represents an attractive drug target for the treatment of cholestatic liver disease and other bile acid-related metabolic disorders, such as obesity and diabetes, because OSTα/β is closely linked to bile acid homeostasis and is tightly regulated by the nuclear receptor farnesoid X receptor (NR1H4) (3017). The functions of the extracellular, transmembrane, and cytoplasmic domains of the OSTβ subunit have been reported, with only the transmembrane domain plus 15 associated aminoacyl residues essential for activity (3018).

TMEM184A (SLC51C1) – Orphan transporter: TMEM184A (SLC51C1) is a plasma membrane protein that is strongly expressed in the esophagus, small intestine, and skin (keratinocytes) according to the HPA. It is also expressed at lower levels in the liver, pancreas, urinary bladder, prostate, and endocrine tissues.

Studies in zebrafish, rat, and human have shown that TMEM184A (SLC51C1) functions as a heparin receptor in vascular smooth muscle cells (VSMCs)and endothelial cells. The therapeutic administration of heparin induces anti-inflammatory and angiogenesis-related responses in these cells (3019–3021). The proposed TMEM184A heparin-binding domain is characterized by multiple positively charged amino acids, such as lysine and arginine, located near the C-terminus (3019). This region is absent in SLC51A. Heparin decreases tumor necrosis factor-alpha (TNF-α)-induced endothelial stress responses in VSMCs, and these responses require TMEM184A, as well as the induction of dual-specificity phosphatase 1 (DUSP1) (3021). Thus, TMEM184A functions as a heparin receptor that mediates anti-inflammatory responses to heparin treatment.

Based on studies in zebrafish embryos, TMEM184A has also been implicated in angiogenesis modulation (3022). This study suggests that TMEM184A can fine-tune the interactions between vascular endothelial growth factor (VEGF) and heparin sulfate, thereby modulating VEGF-activated VEGF receptor (Vegfr2)-dependent angiogenesis (3022). Knockdown of Tmem184a resulted in a reduction in the number of intact intersegmental vessels in the zebrafish embryo. Knockdown has also been shown to disrupt proper vascular development in an in vivo model, likely due to its synergistic interaction with Vegfr2, thereby increasing cell proliferation and decreasing cell adhesion proteins, such as vascular endothelial cadherin (3022). This process appears to require an interaction between TMEM184A and heparan sulfate proteoglycans (HSPGs). Thus, TMEM184A acts as a receptor for both exogenous heparin and cell-surface HSPGs, playing an important role in vascular biology. The angiogenic role of TMEM184A has been well characterized in zebrafish models. However, the direct role of TMEM184A in human angiogenesis has yet to be validated, as only anti-inflammatory responses to therapeutic heparin via the TMEM184A receptor have been demonstrated in humans (see above). Further understanding the mechanisms by which human TMEM184A modulates angiogenesis may have implications for developing therapies for human angiogenic diseases.

In summary, TMEM184A has been associated with modulation of three natural physiological pathways: TNF-α responses, VEGF signaling, and angiogenesis. Glycosaminoglycans (including heparan sulfate) often modulate these effects, suggesting that heparan sulfate-binding receptors like TMEM184A play an important physiological role in VSMCs and endothelial cells, where heparan sulfate proteoglycans are abundant (3023–3025). Therefore, TMEM184A likely plays a role in vascular remodeling, inflammation modulation, and repair processes in humans. Whether it functions as a receptor or also has transporter function remains to be determined.

TMEM184B (SLC51C2) – Orphan transporter: TMEM184B (SLC51C2), also known as the putative MAPK-activating protein FM08, is widely expressed, according to the HPA. Its strongest expression is in the brain, specifically in astrocytes and oligodendrocytes. It has been reported to play a role in controlling cancer cell migration and invasion (3026). Additionally, it has been demonstrated that this protein is responsible for the uptake of ibuprofen and possibly taurine, and that its gene expression is regulated by the transcription factor Nuclear Factor of Activated T cells 5 (NFAT5), which plays a crucial role in regulating the cellular response to osmotic stress. NFAT5 has been shown to be involved in the hyperosmotic regulation of TMEM184B, which acts as a putative modulator of ibuprofen transport in renal MDCK I cells (3027).

TMEM184C (SLC51C3) – Orphan transporter: TMEM184C (SLC51C3), also known as TMEM34 in humans, may function as a transmembrane transporter, though its specific substrates and mechanisms remain unclear. It is strongly expressed in the parathyroid gland, cardiomyocytes, and early spermatids, as suggested by the HPA. TMEM184C has been reported to play a role in germ cell sex determination (3028). TMEM184C (SLC51C1) has also been found to be downregulated in anaplastic thyroid cancer cell lines, likely acting as a tumor suppressor (3029).

Orphan transporter family members (3)

TMEM184A (SLC51C1), TMEM184B (SLC51C2), TMEM184C (SLC51C3)

HGNC updates

HGNC has added SLC51C aliases for TMEM184A (SLC51C1), TMEM184B (SLC51C2) and TMEM184C (SLC51C3).

SLC52 Riboflavin transporter family (2.A.125/DUF1011/MFS)

Discovery: RFVT1 (SLC52A1) is responsible for the transport of vitamin B2/riboflavin into cells (3030). Its rat cDNA was originally identified from a rat kidney cDNA library by digital expression profiling (3031). This cloning strategy involved the construction of a subtractive mRNA expression database for nephrectomized kidneys reflecting the changes in mRNA expression after subtotal nephrectomy (3032), a strategy useful for elucidating the molecular mechanisms of progressive renal failure. The same approach was used to clone the Na⁺-dependent glucose transporter (rNaGLT1/MFSD4B/SLC60A2) from rat kidney (3033). RFVT1 (SLC52A1) showed no similarity to other known SLC transporters except for a low similarity to the ENT nucleoside transporter family SLC29.

Gene family members (3):
SLC52A1 (RFVT1)	SLC52A2 (RFVT2)	SLC52A3 (RFVT3)

Note: While some early publications referred to this transporter as RFT1, we recommend using the name SLC52A1 (or RFVT1) to avoid any confusion with the RFT1 gene (SLC76A1).

Molecular aspects, physiological roles and links to disease

The SLC52 family belongs to the Eukaryotic Riboflavin Transporter (E-RFT) family (TC 2.A.125), which is part of the MFS superfamily, and thus harbors the MFS structural fold (3030).

Riboflavin (vitamin B2) is essential for cellular energy metabolism and redox balance (3034). It serves as the precursor for two key coenzymes, flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD), which act as electron carriers in redox reactions, particularly in energy metabolism. These coenzymes are crucial for redox reactions in metabolic pathways such as the electron transport chain (complexes I and II), β-oxidation of fatty acids (FAD-dependent dehydrogenases, e.g., acyl-CoA dehydrogenase), and the Krebs cycle (succinate dehydrogenase). Additionally, riboflavin is crucial for FMN-dependent pyridoxine 5′-phosphate oxidase, the rate-limiting step in converting pyridoxine into the biologically active form of vitamin B6, pyridoxal 5′-phosphate, as well as for the FAD-dependent kynurenine 3-monooxygenase, which is required for converting tryptophan into niacin/vitamin B3.

The biosynthesis of riboflavin occurs in bacteria, fungi, and plants but not in animals. Therefore, humans depend on the absorption and distribution of riboflavin by the three members of the SLC52 transporter family (3030). First, RFVT3 (SLC52A3) facilitates the absorption of riboflavin in the intestine across the apical membrane of enterocytes (3035). Next, RFVT1 (SLC52A1) facilitates basolateral exit (3030). Finally, RFVT2 (SLC52A2), which is widely expressed, delivers the vitamin into the cells of various organs (3036). Once inside cells, especially in the liver, riboflavin is converted to FMN by riboflavin kinase and to FAD by FAD synthetase.

Because riboflavin is widely present in various foods, riboflavin hypovitaminosis is uncommon. It can, however, impair energy metabolism and lead to neurological symptoms. Similarly, riboflavin deficiency caused by SLC52 mutations leads to rare genetic disorders, as outlined below. Additionally, the SLC52 riboflavin transporters have been found to be overexpressed in different types of cancers, including melanoma, brain cancer, colorectal cancer, and esophageal cancer (3037, 3038) (3039). Methods of delivering anticancer drugs conjugated with riboflavin have also emerged (3037, 3040).

Cryo-EM has provided new insights into the structure, function, and dynamics of human riboflavin transporters. The structures of RFVT2 (SLC52A2) and RFVT3 (SLC52A3) were reported in complex with riboflavin in the outward-occluded and inward-open states, respectively (3041). Riboflavin is recognized by a conserved binding pocket in the central cavity of RFVTs. Both RFVT2 and RFVT3 consist of 11 TMHs, with TMH1-6 forming the N-domain and TMH7-11 forming the C-domain. These domains exhibit an asymmetric 6+5 topology (3041). Additionally, the interdomain pseudo-symmetries of RFVTs are reminiscent of the structural architecture of MFS proteins, though they adopt an atypical MFS-like fold. The transporters operate via a “rocker-switch” alternating-access mechanism (208). Interestingly, the SLC52 RFVTs were found to share structural similarities with equilibrative nucleoside transporter 1 (ENT1) of the SLC29 family, which contains members that also possess this atypical 6+5 topology and are classified as MFS-like transporters (191, 201).

The cryo-EM study also identified two protonatable residues, D119 on TMH4 and E145 on TMH5, inside the central cavity of RFVT3 (3041). In contrast, RFVT1 and RFVT2 have different residues in these positions (C125 and Q151, respectively, with the same numbering in both proteins), which makes them pH-insensitive. The results suggest that protons are co-transported in RFVT3 at low pH but not in RFVT2. This distinction highlights the unique transport and regulatory mechanisms that govern each transporter. These findings lay the groundwork for future research into therapeutic strategies targeting pathologies related to riboflavin transporters.

SLC52A1: RFVT1 (SLC52A1) exhibits a narrow expression profile. It is highly expressed in the placenta (syncytiotrophoblasts) and in the small intestine, where it is found in the basolateral membrane of enterocytes (3031, 3042). According to the HPA, it is especially prevalent in the duodenum. The HPA also shows expression in skin.

Genetic variants of the SLC52 genes are associated with rare genetic disorders that present with symptoms similar to those of acyl-CoA dehydrogenase deficiency (MADD) (3030, 3043). MADD mainly affects amino acid and fatty acid metabolism, and while most MADD patients have mutations in the electron transfer flavoprotein or electron transfer flavoprotein dehydrogenase genes, a MADD phenotype can also be caused by riboflavin deficiency due to defects of SLC52 transporters. indeed, rare genetic variants of SLC52A1 have been shown to result in an autosomal dominant riboflavin transporter deficiency with MADD-like symptoms (3044). Especially newborns and infants with these variants primarily exhibited symptoms similar to those of MADD, while individuals identified in adulthood were typically asymptomatic, except for a biochemical profile characteristic of the condition.

Another study advanced our understanding of cellular aging mechanisms related to RFVT1 (SLC52A1) function, based on experiments using both the human U-2 OS osteosarcoma cell line and the human Hs68 fibroblast cell line. This study showed that riboflavin, when taken up by RFVT1, suppresses cellular senescence in two ways (3045): First, riboflavin is converted to flavin adenine dinucleotide inside the cell, which promotes energy production via respiratory complex II of the inner mitochondrial membrane. Second, riboflavin prevents p53-dependent senescence when taken up by the cell (3045).

SLC52A2: RRFVT2 (SLC52A2) plays an important role in riboflavin homeostasis because it is expressed throughout the body (3046). Likewise, according to the HPA, it is strongly expressed in almost all tissues, with the highest expression in the brain. In the intestine, it is highly expressed in the colon but at relatively low levels in the duodenum.

The impact of natural mutations in RFVT2 (SLC52A2) and their relevance to human riboflavin transporter deficiency type 2 (RFTD2) has been reported (3047). Riboflavin transporter deficiency syndrome is a rare, childhood-onset, neurodegenerative disorder caused by mutations in the SLC52A2 and SLC52A3 genes. Adeno-associated virus (AAV9)-mediated gene therapy targeting SLC52A2 has been explored in patient-derived motor neurons. This treatment was shown to improve neurite length, suggesting its potential therapeutic application in treating RFVT2 deficiency (3048). The study provided insights into how specific amino acid changes can alter transporter activity and contribute to our understanding of RFVT2 pathogenesis.

SLC52A3: RFVT3 (SLC52A3) is a H⁺-coupled transporter that facilitates riboflavin uptake across the apical membranes of enterocytes in the small intestine (3035, 3041) (see Fig. 30). According to the HPA, it is most highly expressed in the duodenum. Furthermore, RFVT3 is highly expressed in the testes and at moderate levels in the kidneys, brains, and muscles, according to the HPA.

Recessive mutations in the SLC52A3 gene can lead to Brown-Vialetto-Van Laere syndrome (BVVL) and/or Fazio-Londe syndrome (FLS). BVVL is a rare, autosomal recessive neurological disorder characterized by sensorineural hearing loss and various cranial nerve palsies (3049, 3050). FLS is similar to BVVL, except it does not include sensorineural deafness. The SLC52A3 protein and its mutations were analyzed in silico, structurally, and functionally among all reported patients, and a novel mutation was reported (3049).

To study the effects of riboflavin deficiency on cerebral cortex development, Slc52a3 knockout mice were generated (3051). The study revealed the essential function of RFVT3 in embryonic development. Absence of this transporter resulted in cerebral cortex hypoplasia, which was alleviated by riboflavin supplementation. This finding underscores the importance of RFVT3 (SLC52A3) in neurodevelopment.

Orphan transporter family members: N/A

SLC53 XPR1 phosphate exporter (2.A.94/EXS/XRP)

Discovery: XPR1 (SLC53A1) is the sole member of this family. It was originally identified as a retroviral receptor (3052–3054).

Gene family member (1)

XPR1 (SLC53A1)

Molecular aspects, physiological roles and links to disease

The SLC53 family belongs to the Phosphate Permease (Pho1) family (TC 2.A.94) which is part of the Ion Transporter (IT) superfamily.

XPR1 (SLC53A1): XPR1 (SLC53A1) is ubiquitously expressed, and its function has long remained elusive. Studies have reported that XPR1 is a P_i exporter (3055, 3056), but its export activity has long been questioned. As described in Section 8, “Structure-based classification of SLCs”, structural studies of XPR1 have clarified this: Cryo-EM structures of human XPR1 were reported in different conformations, showing XPR1 as a dimer and demonstrating that its 10 TMH core contains a domain with a P_i-binding site and a translocation pathway (334, 335, 3057). The structural architecture of the XPR1 translocation moiety was found to be similar to that of ion-translocating microbial rhodopsins. These observations are confirmed by site-directed mutagenesis and P_i transport assays, providing definitive evidence that XPR1 is a P_i exporter.

XPR1 has an N-terminal domain of 180 amino acid residues called SPX, named after the transporters SYG1, Pho81 and XPR1 known to regulate intracellular phosphate levels in yeast and plants through the binding of inositol polyphosphates (InsPs), allowing SPX to interact with transcription factors (3058–3061). SPX was found to be dispensable for phosphate export, suggesting that this domain plays a regulatory role (3055). A high-affinity ligand for SPX, InsP₈ has been identified in mammalian cells and XPR1-mediated P_i efflux was shown to be inhibited when cellular InsP₈ synthesis was reduced (1761). InsP₈ is an inositol pyrophosphate (PP-InsP), a class of compounds that represent a subset of InsPs (3062); see also the SLC20A1 description.

As described in the SLC20 family description, XPR1 plays an important role in the regulation of cellular P_i uptake together with PiT-1 (SLC20A1) in mammalian cells, where phosphate influx via PiT-1 is balanced by efflux via XPR1 (SLC53A1). This balance is regulated by the signaling molecule InsP₈, which binds to the SPX domain of XPR1 to functionalize this phosphate exporter during periods of intracellular P_i abundance.

To delineate the substrate translocation pathway and how InsP₈ initiates P_i transport, cryo-EM structures of dimeric XPR1 were determined and the protein complex was functionally characterized (337). Binding of InsP₈ to XPR1 rigidifies the intracellular SPX domains. InsP₈ bridges the dimers and the SPX and transmembrane domains. In this state, the C-terminal intracellular tail is sequestered, opening the entrance of the transport pathway. This explains the obligate roles of the SPX domain and InsP₈. In essence, the C-terminal tail acts as a plug that blocks the entrance to the P_i translocation pathway when InsP₈ is absent. However, the presence of InsP₈ causes the SPX domain to become rigidified and the C-terminal tail of XPR1 to seal off. This opens the intracellular gate, allowing P_i to enter. This is followed by the opening of an extracellular gate, which allows P_i to be released. The latter process has been reported by other structural studies (334, 335).

XPR1 was shown to be important for phosphate regulation in ovarian cancer and in patient-derived tumor samples, transcription factor PAX8-dependent overexpression of the phosphate importer NaPi-IIb (SLC34A2) correlates with sensitivity to loss of the phosphate exporter XPR1 (3063). It was also shown that a novel partner protein of XPR1, KIDINS220 (kinase D interacting substrate 220), is required for the proper cellular localization and activity of XPR1, and that disruption of this protein complex results in acidic vacuolar structures that is followed by cell death. The findings suggest a therapeutic strategy that exploits the vulnerability of the XPR1-KIDINS220 export system in ovarian cancer cells with the goal of generating phosphate efflux inhibitors that reduce tumor cell viability due to toxic intracellular accumulation of phosphate (3063).

Excessive cellular accumulation of inorganic phosphate (P_i) can lead to ectopic calcification in soft tissues, which, depending on the localization of the calcific deposits, can cause cardiovascular disease, primary familial brain calcification (PFBC), etc. (3064). XPR1 missense variants have been identified to cause PFBC in seven families and functional analyses have shown that they either exhibit reduced cell surface expression or impaired phosphate export function (3065–3067). Interestingly, several of them were located in the N-terminal cytoplasmic SPX domain of XPR1. For this reason, the SPX domain was examined by cryo-EM analysis. However, while the TMH core of XPR1 was well resolved, the intracellular SPX domain was completely disordered, implying high intrinsic flexibility (3057).

In the kidney, XPR1 likely mediates the basolateral exit of phosphate from the proximal tubules (1759). The proximal tubule is responsible for reabsorption of phosphate via the NaPi-IIa (SLC34A1) and NaPi-IIc (SLC34A3) phosphate transporters expressed at the apical membrane (see the SLC34 family description). In terms of basolateral P_i efflux, the mechanism has been elusive but recent data suggest that XPR1 fulfills this task. In support of this, mice with conditional inactivation of the XPR1 gene in the renal tubule exhibit generalized proximal tubular dysfunction (Fanconi syndrome) with decreased renal P_i reabsorption and hypophosphatemic rickets (3068).

Orphan transporter family members: N/A

SLC54 Mitochondrial pyruvate carrier family (2.A.105/MPC/SemiSWEET)

Discovery: The mitochondrial inner membrane pyruvate carrier MPC has been identified as a heteromeric complex of the two paralogs MPC1 (SLC54A1) and MPC2 (SLC54A2) by two different groups using two different approaches. One (3069) combined yeast genetics, metabolomics and mutational analysis of cells from patients with mitochondrial pyruvate metabolism defects to identify MPC1 and MPC2. The other (3070) characterized mitochondrial membrane proteins from a proteomic analysis of the IMM that are conserved from yeast to mammals to identify MPC1 and MPC2.

Gene family members (3):
MPC1 (SLC54A1)	MPC2 (SLC54A2)	MPC1L (SLC54A3)

Molecular aspects, physiological roles and links to disease

The SLC54 family is a member of the Mitochondrial Pyruvate Carrier (MPC) family (TC 2.A.105), which is part of the SemiSWEET fold family. Importantly, the SLC54 family members are not related to the SLC25 mitochondrial carriers. They have a different architecture, related to the 3-TMH repeat element architecture in the SWEET transporters. The functional unit of the MPCs is a heterodimer in which each protomer has 3 TMHs (see below).

MPC1 (SLC54A1), MPC2 (SLC54A2) and MPC1L (SLC54A3): In humans, MPC1 (SLC54A1) and MPC2 (SLC54A2) co-assemble to form a functional heterodimer called MPC. Each protomer is predicted to form a membrane protein with three transmembrane α-helices (3071). An alternative dimeric complex of MPC1L (SLC54A3) and MPC2 (SLC54A2) is formed in the testis. Beyond this knowledge, little is known about the structure and mechanism of MPC because of technical difficulties with their purification and stability that has hindered progress in functional and structural analysis (3072).

MPC is critical for cellular homeostasis as it is required in the central metabolism for the transport of pyruvate from the cytoplasm, produced during glycolysis, into the mitochondrial matrix (3071).

Because of its central metabolic role, MPC (MPC1-MPC2 heterodimer) has been proposed as a potential drug target for diabetes, non-alcoholic fatty liver disease (NAFLD), neurodegeneration, and cancers that rely on mitochondrial metabolism. With the guidance of structure-function insights, attempts are being made to develop specific MPC inhibitors to suppress the glucose production of hepatocytes for the treatment of diabetes (3072, 3073).

Orphan transporter family members: N/A

SLC55 LETM mitochondrial cation/proton exchanger family (2.A.97/LETM1/single-TMH)

Discovery: The gene encoding LETM1 (SLC55A1), the leucine zipper EF-hand containing transmembrane protein 1, has been identified as a gene that is deleted in patients with Wolf-Hirschhorn syndrome (WHS) (3074, 3075). This is a rare chromosomal disorder, characterized by severe growth and intellectual impairment, hypotonia, and seizures (3074–3076).

Gene family members (3):
LETM1 (SLC55A1)	LETM2 (SLC55A2)	LETMD1 (SLC55A3)

Molecular aspects, physiological roles and links to disease

LETM1 (SLC55A1) - Orphan transporter: LETM1 (SLC55A1) is ubiquitously expressed and phylogenetically well conserved. It is an inner mitochondrial membrane protein with an osmoregulatory function that controls cation homeostasis (288, 3077, 3078). LETM1 has a short hydrophobic N-terminal domain, a conserved proline-rich transmembrane domain, a hydrophilic C-terminal domain in the matrix, with two to four coiled-coil domains (CCDs), and a characteristic leucine zipper motif (3075). The matrix residing carboxyterminal domain contains an EF-hand motif that is highly conserved among orthologs.

LETM1 may function as a K⁺/H⁺ exchanger (3079). In line with this, recent studies have shown that bi-allelic pathogenic LETM1 variants in patients with WHS are associated with defective mitochondrial K⁺ efflux, swollen mitochondrial matrix structures, and a reduction in protein levels and electron transfer chain activity (3080). The observed neurological and mitochondrial pathologies would be consistent with the resulting impaired mitochondrial osmoregulation. In addition, a genome-wide genetic screen indicated that LETM1 alters mitochondrial Ca²⁺ levels (3081–3083). Whether the bi-allelic variants affect mitochondrial Ca²⁺ levels via the EF-hand motif of LETM1 requires further investigation.

The structure of LETM1 has been proposed to be a hexamer (3083). The transport mechanism of the protein remains unknown, although it has been shown that a change in external pH causes conformational changes in LETM1 (3083). LETM1 contains the mitochondrial signal peptide (3083), similar to LETM2 and LETMD1 according to TargetP-2.0 predictions (3084), suggesting that SLC55 proteins are likely translocated to mitochondria via the TIM23 translocation complex.

Studies on homologs of LETM1 (SLC55A1), the yeast mitochondrial cation/H⁺ antiporters Mdm38 and Ylh47f, have also been reported (3085). Mdm38 and Ylh47 were shown to not transport Ca²⁺, but rather to be selective for K⁺ and Na⁺. A detailed characterization of the ion transport activities of Mdm38, Ylh47, and related structural properties offers additional insight into the functional and physiological roles of LETM1 (3085).

Disease phenotypes are associated either with the absence (Wolf-Hirschhorn syndrome, WHS), decreased expression (insulin resistance in obesity), or increased expression (cancer) of LETM1 (SLC55A1) (3076).

LETM2 (SLC55A2) - Orphan transporter: The second paralog, LETM2 (SLC55A2), has been shown to be a crucial oncogene for tumor progression in pancreatic ductal adenocarcinoma (PDAC) which occurs via activating the downstream PI3K/Akt pathway (3086). Thus, LETM2 provides an effective target for the diagnosis and treatment of PDAC.

LETMD1 (SLC55A3) - Orphan transporters: The third paralog, LETMD1 (SLC55A3) was shown to encode a mitochondrial matrix protein that maintains thermogenic capacity of brown adipose tissue in mice (3087). LETMD1 was found to be strongly associated with mitochondrial function in both mouse and human adipose tissues. The results provide strong evidence that LETMD1 acts upstream of mitochondrial uncoupling protein UCP1, although the precise mechanism through which LETMD1 regulates UCP1 expression remains unclear and warrants further investigation. The findings highlight the potential of LETMD1 (SLC55A3) as a therapeutic target for the treatment of obesity and metabolic diseases.

Orphan transporter family members (3)

LETM1 (SLC55A1), LETM2 (SLC55A2), LETMD1 (SLC55A3)

SLC56 Sideroflexins (2.A.54/Mtc/unknown)

Discovery: The first member of this family (SLC56A1) was identified in the rat and proposed to be a mitochondrial citrate transport protein (CTP) (3088). Subsequently, it was rediscovered by positional cloning as a mitochondrial transmembrane protein called sideroflexin (SFXN), which stands for “siderocytic anemia” and “flexed tail phenotype” observed in flexed-tail (f) mutant mice (3089) with evidence linking sideroflexins to iron homeostasis. However, subsequent studies showed that sideroflexins are not citrate transporters and the association with iron homeostasis was also challenged when studies later showed that SFXN1 (SLC56A1) is a mitochondrial transporter of serine, which is essential for one-carbon metabolism (3090). At best, SFXN1 is thought to be related to iron metabolism in the context of iron-sulfur cluster and heme synthesis (see below) (3091).

Gene family members (5):
SFXN1 (SLC56A1)	SFXN3 (SLC56A3)	SFXN5 (SLC56A5)
SFXN2 (SLC56A2)	SFXN4 (SLC56A4)

Molecular aspects, physiological roles and links to disease

There are five sideroflexins in mammals (Fig. 52), with SFXN3 being the closest homologue of SFXN1. All five are ubiquitously expressed according to the HPA, although individual expression varies in specific tissues (see below).

Fig. 52. Phylogenetic tree of the SLC56 family.

The SLC56 family belongs to the Sideroflexin (SFXN) family (TC 2.A.54). Currently, the family is a structural orphan. An AlphaFold structure is presented in Fig. 2E, showing 6 TMHs.

Studies revealed that sideroflexins, like many members of the mitochondrial SLC25 carrier family, are substrates of the TIM22 complex, an inner membrane translocase that mediates the insertion of multi-pass transmembrane proteins into the inner mitochondrial membrane (3092, 3093).

The existence of several sideroflexin paralogs in vertebrates provides significant functional redundancy (3091, 3094). While there is only one sideroflexin in yeast (Fsf1), there are two paralogs in Drosophila (Sfxn1-3 and Sfxn2) and five in vertebrates (SFXN1-5; SLC56A1-A5).

In humans, despite the widespread expression of sideroflexins, there is still significant variation in different tissues, based on the HPA and other evidence (3093): SFXN1 (SLC56A1) is predominantly expressed in fetal and adult brain, liver and kidney; SFXN2 (SLC56A2) is highly expressed in kidney; SFXN3 (SLC56A3) is the only paralog with high expression in retina; SFXN4 (SLC56A4) is highly expressed in fetal and adult heart, ovary, skeletal muscle, adrenal gland, and pancreas, and at moderate levels in skin, adipose tissue, lungs, and retina; and SFXN5 (SLC56A5) is highly expressed in frontal cortex.

SFXN1 (SLC56A1): SFXN1 (SLC56A1) and SFXN3 (SLC56A3), possibly also SFXN2 (SLC56A2), are mitochondrial serine transporters required for one-carbon metabolism (3090). Specifically, cells lacking SFXN1 were found to be defective in glycine and purine synthesis, and purified SFXN1 was shown to transport serine in vitro (3090). SFXN1 and SFXN3 are likely to have other physiologically relevant substrates besides serine, such as alanine or cysteine.

A possible explanation for the frameshift mutation of sideroflexin causing sideroblastic anemia and the flexed tail (f) mutant mouse phenotype may be the following (3091, 3093): Mutations that affect the part of the heme synthesis pathway in the mitochondria that require glycine cause sideroblastic anemia in humans. In Sfxn1 mutant mice, insufficient import of serine into the mitochondria leads to a decrease in glycine and thus heme synthesis.

SFXN1 (SLC56A1) is expressed in many cancers, most strongly in leukemias and lymphomas (3090).

SFXN1 (SLC56A1) is predicted to have five transmembrane domains, but structural information is still missing. Based on computational predictions, further insight has been provided and several amino acid positions have been identified in these proteins that can be helpful to identify functionally relevant hot spots (288).

SFXN2 (SLC56A2), SFXN3 (SLC56A3), SFXN5 (SLC56A5) – Orphan transporters: As mentioned above, some of these transporters may function as mitochondrial transporters of serine or other amino acids.

SFXN4 (SLC56A4) - Orphan transporter: SFXN4 (SLC56A4) is the most divergent member of the SFXN family and has been shown to lack serine transport (3090). However, mutations in SFXN4 cause a mitochondrial disease with clinical features including macrocytic anemia, and SFXN4 has been shown to interact with core components of the mitochondrial complex I intermediate assembly (MCIA) complex (3095).

Subsequent studies of SFXN4 revealed that it affects Fe-S cluster biogenesis, iron metabolism, mitochondrial respiration and heme biosynthetic enzymes (3096). Specifically, knocking down SFXN4 in the erythroleukemic cell line K562 reduced the stability and activity of cellular Fe-S proteins. This affected iron metabolism by influencing the cytosolic aconitase-IRP1 switch and redistributing iron from the cytosol to the mitochondria. It also impacted heme synthesis by reducing ferrochelatase levels and inhibiting the translation of 5′-aminolevulinate synthase 2 (ALAS2), a crucial enzyme for heme production in erythroid cells.

A study aimed at identifying specific inflammatory proteins in chronic obstructive pulmonary disease (COPD) exacerbations found that SFXN4 (SLC56A4) expression was consistently increased (3097). It has been proposed that SFXN4 overexpression leads to mitochondrial dysfunction and iron sequestration in airway macrophages, resulting in oxidative stress during COPD exacerbations.

Orphan transporter family members (4)

SFXN2 (SLC56A2), SFXN3 (SLC56A3), SFXN4 (SLC56A4), SFXN5 (SLC56A5)

SLC57 NIPA-like magnesium transporter family (2.A.7.25/Mg_trans_NIPA/NST)

Discovery: NIPA1 (SLC57A1) was originally identified as a disease-causing gene for hereditary spastic paraplegia (HSP) (131, 3098). NIPA1 stands for “nonimprinted in Prader-Willi/Angelman loci 1” (131) (see below).

Gene family members (6):
NIPA1 (SLC57A1)	NIPAL1 (SLC57A3)	NIPAL3 (SLC57A5)
NIPA2 (SLC57A2)	NIPAL2 (SLC57A4)	NIPAL4 (SLC57A6)

Molecular aspects, physiological roles and links to disease

The SLC57 family has 6 members (Fig. 53) and belongs to the NIPA Mg²⁺ Uptake Permease (NIPA) family (TC 2.A.7.25) which is part of NST fold families. NIPA1 and NIPA2 are predicted to have 9 TMHs (3098). 3D structural information is currently unavailable.

Fig. 53. Phylogenetic tree of the SLC57 family.

Hereditary spastic paraplegia (HSP) is a heterogeneous group of neurodegenerative disorders characterized by spasticity, hyperreflexia, weakness, and stiffness of leg muscles due to degeneration of corticospinal axons. Autosomal dominant hereditary spastic paraplegia (ADHSP) is the most common form of HSP, of which a very rare type is autosomal dominant spastic paraplegia 6 (SPG6) (131, 3099). In an attempt to understand the molecular pathology of ADHSP, a disease-specific mutation in NIPA1 (nonimprinted in Prader-Willi/Angelman loci 1, SLC57A1) was identified (131), which explains the cause of SPG6. Reduced magnesium levels increased the expression of NIPA1, suggesting a role of NIPA1 in cellular magnesium metabolism. NIPA1 mutations have also been identified and characterized clinically in a Taiwanese cohort with HSP (3100).

NIPA1 (SLC57A1): NIPA1 (SLC57A1) exhibited magnesium transport when expressed in Xenopus oocytes (3101). Specifically, NIPA1 mediated Mg²⁺ uptake that was electrogenic, voltage-dependent and saturable, with a K_m of 0.69 mM, and the transport was relatively selective for Mg²⁺. According to the HPA, NIPA1 (SLC57A1) is most strongly expressed in oligodendrocytes, as well as in excitatory and inhibitory neurons. Moderate expression is also found in cholangiocytes, alveolar type 1 cells, specialized epithelial cells of the prostate gland, and dendritic cells. Subcellular localization revealed that the NIPA1 protein associates with early endosomes and the cell surface in a variety of neuronal and epithelial cells (3102). The protein may play a role in nervous system development and maintenance. HSP-associated mutations in NIPA1 were shown to trigger neural degeneration through a gain-of-function mechanism (3102).

NIPA2 (SLC57A2): Unlike other reported Mg²⁺ transporters, the paralog NIPA2 (SLC57A2) was shown to be very selective for the Mg²⁺ cation (3103). The expression of NIPA2 is particularly enriched in renal cells, where it is normally localized in early endosomes. From there, it may be recruited to the plasma membrane in response to low extracellular magnesium to support renal magnesium conservation.

NIPAL1 (SLC57A3): NIPAL1/NIPA3 (SLC57A3) is a Mg²⁺ uptake transporter that is specifically enriched in pancreatic islets, where it influences insulin secretion (3104). This is of particular importance given the impact of magnesium deficiency on insulin resistance in T2D (3105).

NIPAL2 (SLC57A4) - Orphan transporter: NIPAL2 (SLC57A4) is widely expressed, with the strongest expression observed in excitatory neurons according to the HPA. It is predicted to be involved in Mg²⁺ transport (7, 3106).

NIPAL3 (SLC57A5) - Orphan transporter: According to the HPA, NIPA32 (SLC57A5) is widely expressed, with the strongest expression observed in the brain and prostate. The function of NIPAL3 (SLC57A5) is currently unknown.

NIPAL4 (SLC57A6): NIPAL4 (SLC57A6), also known as NIPA4, is most strongly expressed in keratinocytes and oligodendrocytes according to the HPA. An allelic variant of NIPAL4 was identified that is predicted to be pathogenic and probably the cause of an erythrokeratodermia variabilis (EKV)-like autosomal recessive congenital ichthyosis. It results in a defective Mg²⁺ transporter, which would normally play a role in the development and maintenance of the barrier function of the epidermis (3104).

Several lipids, including specialized ceramides called 1-O-acylceramides, were found to have altered levels and composition in Nipal4 knockout mouse epidermis compared to wild-type (wt) mouse epidermis (3107). It was concluded that elevated Mg²⁺ concentrations in differentiated keratinocytes affect the production of various lipids, resulting in the lipid composition necessary for skin barrier formation.

Orphan transporter family members (2)

NIPAL2 (SLC57A4), NIPAL3 (SLC57A5)

SLC58 MagT-like magnesium transporter family (1.A.76/OST3_OST6/MagT)

Discovery: The first member of this family, MagT1 (SLC58A1), was identified by differential gene expression using microarray analysis to screen for genes upregulated in mouse kidney distal convoluted tubule cells under low Mg²⁺ conditions (65). Subsequently, MagT1 was identified together with its closely related paralog TUSC3 (SLC58A2) in a yeast complementation screen seeking to rescue the loss of the yeast ALR1 Mg²⁺ transporter in the S. cerevisiae mutant alr1Δ (3108).

Gene family members (2):
MAGT1 (SLC58A1)	TUSC3 (SLC58A2)

Molecular aspects, physiological roles and links to disease

The SLC58 family belongs to the Magnesium Transporter 1 (MagT1) family (TC 1.A.76). Both MAGT1 (SLC58A1) and TUSC3 (SLC58A2) have 4 TMHs and an N-terminal cleavable signal peptide (3108). MagT1 and TUSC3 have been proposed to be Mg²⁺ transporters (66), yet they are also part of the OST-B glycosylation complex (34, 332). Specifically, MagT1 and TUSC3 are the human orthologs of OST3 or OST6 oligosaccharyltransferase in S. cerevisiae, respectively (332, 333).

Oligosaccharyltransferases (OSTs) are responsible for catalyzing the transfer of a high-mannose glycan to secretory proteins in the ER. Humans have two different OST complexes, OST-A and OST-B, and their complexes contain multiple subunits with different catalytic subunits, STT3A in OST-A and STT3B in OST-B (3109–3111). In addition, OST-A contains the adaptor protein DC2 (also called OSTC), while OST-B instead contains MagT1 or TUSC3 (34, 332, 3112).

As highlighted in Section 8 in the description of the MagT fold, the cryo-EM structures of the human oligosaccharyltransferase complexes OST-A and OST-B have been determined (34), and as a result, the structures of human MagT1 and its paralogs, DC2/OSTC, are also revealed (34) (Fig. 4), showing a high degree of similarity, with 3 of the 4 TMHs organized in a linear manner (34). In contrast to the initial proposal that MagT1 and TUSC3 function as Mg²⁺ transporters, the experimentally resolved structures did not reveal a transporter-like or channel-like transport mechanism, suggesting that these proteins might affect Mg²⁺ homeostasis in an indirect manner.

MAGT1 (SLC58A1) and TUSC3 (SLC58A2): MAGT1 is widely distributed among tissues, with a particular abundance in epithelial tissues (65, 3113). Functional expression of MagT1 in Xenopus oocytes elicited large Mg²⁺-evoked voltage-dependent currents with low permeability to other divalent metal ions (65, 66), a property suggesting a channel-like transport mechanism. MagT1 was found to be more widely expressed than TUSC3 (3114).

Knockdown of either or both genes in Jurkat cells similarly reduced Mg²⁺ uptake, suggesting that the two gene products function cooperatively (3114). Thus, MagT1 and TUSC3 have been proposed to be vertebrate plasma membrane Mg²⁺ transport systems. In support of this, magnetic fields have been shown to enhance the efficiency of Mg²⁺ transport via MagT1 from poly-l-lactic acid (PLLA) bone scaffolds (3115). Specifically, the activities of the Mg²⁺ channel protein MagT1 on the membrane of bone marrow mesenchymal stem cells could be enhanced via magnetic torque effect (via integrin αV β3/actin) under the action of static magnetic field, which promoted the bone marrow mesenchymal stem cells to capture Mg²⁺ in the microenvironment, thus inducing osteogenesis.

While MagT1 has been proposed to be a Mg²⁺ transporter or channel (3113), studies have shown that it is located in the ER membrane and regulates N-glycosylation within the oligosaccharide transferase complex, and that MAGT1 deficiency in human patients leads to a congenital error of glycosylation (332, 3116). As highlighted above, MagT1 serves as a non-catalytic accessory protein of the oligosaccharide transferase OST-B complex, which in turn is responsible for N-linked glycosylation. Whether MagT1 still has a Mg²⁺ transport function or acts indirectly as a regulator of intracellular Mg²⁺ levels awaits further clarification.

X linked immunodeficiency with magnesium defect, Epstein-Barr virus (EBV) infection and neoplasia (XMEN) disease is caused by hemizygous loss-of-function gene variants in MAGT1 (3117, 3118). Since immune cells do not express TUSC3, N-linked glycosylation of OSTB/STT3B-dependent glycoproteins is entirely dependent on MAGT1. XMEN is a disorder characterized by CD4 lymphopenia, severe chronic viral infections and defective T-lymphocyte activation, which render XMEN patients susceptible to Epstein-Barr virus infections and lead to persistently low levels of intracellular Mg²⁺. XMEN can therefore be considered a congenital disorder of glycosylation that presents as a combined immune deficiency with platelet dysfunction. Different variants of the human MAGT1 gene have been reported that may affect natural killer (NK) cell and platelet glycome composition, resulting in variations in clinical outcomes (3119).

MAGT1 (SLC58A1) deficiency was also found to dysregulate platelet cation homeostasis and accelerate arterial thrombosis and ischemic stroke in mice. The study suggests that MagT1 and the cation channel TRPC6 are functionally linked (3116).

Orphan transporter family member: N/A

SLC59 Sodium-dependent lysophosphatidylcholine symporter family (2.A.2.3/MFS_2/MFS)

Discovery: The major facilitator superfamily domain-containing protein MFSD2A (SLC59A1), previously considered an orphan transporter, is the first identified member of the major facilitator superfamily that transports lipids (3120). In addition, it has been shown that MFSD2A (SLC59A1) is strongly induced in brown adipose tissue during fasting and cold-induced thermogenesis through beta-adrenergic signaling (3120).

Gene family members (3):
MFSD2A (SLC59A1)	MFSD2B (SLC59A2)	MFSD12 (SLC59B1)

Molecular aspects, physiological roles and links to disease

The SLC59 family belongs to the Glycoside-Pentoside-Hexuronide (GPH):Cation Symporter family (TC 2.A.2.3), which is part of the MFS superfamily. MFSD2A (SLC59A1) was shown to have 12 TMHs (see below), consistent with the MFS fold.

MFSD2A (SLC59A1): MFSD2A (SLC59A1) has unique structural features and a special transport mechanism (3121). It is a Na⁺-dependent lysophosphatidylcholine (LPC) transporter expressed at the endothelium of the BBB (see Fig. 33). There it serves as the primary route for delivery of the essential omega-3 fatty acid ω3-DHA and other long-chain fatty acids into the fetal and adult brain (3121, 3122). It has been shown to transport lipids across the BBB when they are attached to a LPC head group. It is essential for mouse and human brain growth and function. Importantly, MFSD2A (SLC59A1) does not transport unesterified ω3-DHA, but rather ω3-DHA in the form of an ester with LPC, which is synthesized by the liver and circulates largely on albumin.

Loss-of-function mutations in the human MFSD2A gene result in autosomal recessive microcephaly 15, a disorder characterized by progressive microcephaly, impaired intellectual development, poor speech, and spasticity (3123, 3124). Furthermore, MFSD2A deficiency has been shown to lead to abnormal oligodendrocyte lineage development and hypomyelination (3125). Studies using Mfsd2a knockout mice have replicated many of these phenotypes (3126).

3D structural models of the human MFSD2A (SLC59A1) protein were generated using homology modeling with MelB- and LacY-based crystal structures. Further refinement was achieved through biochemical analysis. These models show the 12-TMH protein in its outward-open, outward-partially occluded, and inward-open states during the transport cycle. They also revealed a conserved Na⁺-binding site (3127). Subsequently, in vitro studies were conducted using recombinant MFSD2A that was reconstituted in liposomes. These studies examined the ability of MFSD2A to transport lysophosphatidylserine (LPS), which was coupled to an LPS-binding fluorophore. This assay demonstrated that MFSD2A flips LPS from the outer to the inner leaflet of a membrane bilayer in a Na⁺-dependent manner. Furthermore, cryo-EM-guided mutagenesis revealed amino acid residues that are important for the flippase activity of MFSD2A (3128).

Because MFSD2A (SLC59A1) is present at the BBB, it is an attractive target for novel drug delivery strategies to the brain (3122, 3129).

MFSD2B (SLC59A2): MFSD2B (SLC59A2) is an erythrocyte exit pathway for the signaling lipid S1P (sphingosine-1-phosphate). According to the HPA, it is expressed almost exclusively and at a very high level in erythroid cells. S1P plays important biological functions by binding to sphingosine-1-phosphate G protein-coupled receptors that regulate fundamental biological processes such as the immune system and blood vessel integrity.

Of note, there are two different MFS transporters that are involved in S1P export: the aforementioned SLC59A2/MFSD2B, which is essential for S1P export from red blood cells and platelets (3130), and SLC63A2/SPNS2 (Spinster homolog 2; see the SLC63 family description), which primarily exports S1P in endothelial cells. The concentration of S1P in the blood is maintained at micromolar concentrations, mainly associated with high-density lipoprotein and albumin. Endothelial cells and red blood cells are the major secretors of S1P to replenish the rapid turnover rate in the blood.

MFSD12 (SLC59B1): MFSD12 (SLC59B1) is a ubiquitously expressed lysosomal cysteine/H⁺ exchanger (3131, 3132). According to the HPA, it is expressed at particularly high levels in the kidney, skin, brain, intestine, testis, and bone marrow. Moderate expression is also reported in the retina and other tissues.

SLC59B1 (MFSD12) has been shown to facilitate the import of cysteine into melanosomes and, in non-pigmented cells, into lysosomes (3131). MFSD12 is required to maintain normal levels of cystine, the oxidized dimer of cysteine, in melanosomes and to produce cysteinyldopas, the precursors of pheomelanin synthesis, which are made in melanosomes via cysteine oxidation (3131). Reduced function of MFSD12 leads to decreased pheomelanin synthesis, resulting in lighter pigmentation. The Y182H variant in MFSD12 has been shown to be associated with skin pigmentation differences in various human populations, including Latin American, East Asian, and African groups (3133, 3134). Its frequency correlates with levels of sun exposure, suggesting a role in adaptation to different UV environments.

In the lysosomes of kidney proximal tubules, MFSD12 functions as a lysosomal cysteine importer that directly affects cystine levels in the lysosomal lumen. In contrast, cystinosin (CTNS, SLC66A4) is a lysosomal cystine exporter (see the SLC66A4 description) (3135).

The physiological roles of MFSD12 in other tissues have also been investigated (3132). Furthermore, MFSD12 has been demonstrated to play a crucial role in glycosphingolipid (GSL) metabolism (3132). GSL catabolism, the process by which complex lipids are broken down mainly in lysosomes, is regulated in a sequential manner by lysosomal hydrolases such as β-galactosidase. To identify lysosomal transporters involved in GSL metabolism, the effects of lysosomal membrane-localized transporters on β-galactosidase expression were examined after siRNA knockdown. Surprisingly, knocking down MFSD12 increased β-galactosidase expression severalfold, suggesting that MFSD12 modulates GSL metabolism.

Next, intestine-specific Mfsd12 knockout mice were generated. Interestingly, embryonic lethality was exhibited by these mice after embryonic day 14.5, with only around 20% surviving to adulthood. Compared to wt mice, these mice had reduced body weight and size, as well as lower fat mass. MFSD12 deficiency also increased mouse movement, energy expenditure, and respiratory exchange ratio. These findings suggest that MFSD12 is crucial for embryonic development and metabolism. Furthermore, Mfsd12 knockout in the retina has shown that this transporter is necessary for eye development and function (3132).

Investigating the morphology and function of lysosomes from intestine-specific Mfsd12 KO mice revealed grossly swollen lysosomes containing abundant vacuolar structures, indicating lysosomal dysfunction (3132). Additionally, the lysosomes of the knockout mice were more acidic. The importance of the acidic internal environment for lysosomal function is well-documented (3136). MFSD12 plays a crucial role in this process by facilitating the transport of cysteine into lysosomes, where it is exchanged for H⁺. This increase in luminal pH is expected to support normal lysosome function, which is regulated by the mTOR-TFEB pathway (3132). Conversely, the absence of MFSD12 disrupts the mTOR-TFEB pathway due to the instability of the lysosomal acidic environment, which leads to lysosomal dysfunction.

Cysteine in the cytosol is known to regulate mTOR activity, as evidenced by the following observations: Increasing the cytosolic content of cysteine via the cystine-glutamate exchanger SLC7A11 has been shown to promote mTOR activation (3137). In the intestinal cells of Mfsd12-deficient mice, transport of cysteine into the lysosomes was blocked. This resulted in elevated cytosolic cysteine levels, which likely affects the mTOR-TFEB-mediated pathways that regulate lysosomal hydrolases. This leads to impaired catabolic pathways, including compromised GSL catabolism. This, in turn, causes lysosomal storage diseases, which are characterized by the abnormal accumulation of glycosphingolipids (3138). Analysis of lysosomes from wt and intestine-specific knockout mice revealed that Mfsd12 deficiency results in the aberrant accumulation of GSL derivatives in lysosomes (3132).

In summary, MFSD12 maintains the balance of lipids in the lysosome, which is critical for proper mTORC1 activity and TFEB regulation. Disruption of MFSD12 impairs the lysosomal environment, alters glycosphingolipid metabolism, and dysregulates the mTOR-TFEB signaling axis. Since the mTOR-TFEB pathway is vital for cellular adaptation to metabolic stress, MFSD12 may affect autophagic flux, cell survival, and metabolic reprogramming, particularly in conditions such as lysosomal storage disease and cancer.

Indeed, emerging evidence links MFSD12 to various cancers. Altered MFSD12 expression levels have been observed in certain tumors, such as melanoma, breast cancer, and lung cancer. This suggests a potential role for MFSD12 in tumorigenesis and cancer progression (3139). MFSD12 has also been reported to function as an oncogene in lung adenocarcinoma, promoting tumor growth and progression (3140). Overexpression of MFSD12 is associated with adverse clinical outcomes, highlighting its potential as a prognostic biomarker and therapeutic target.

Orphan transporter family members: N/A

HGNC update

SLC59B1 is a new alias for MFSD12.

SLC60 Glucose transporters (2.A.1.7/MFS_1/MFS)

Discovery: The SLC60 family comprises two MFS-like proteins, MFSD4A (SLC60A1) and MFSD4B (SLC60A2). MFSD4B (rNaGLT1, SLC60A2) was first identified from rat kidney and shown to be a novel Na⁺-dependent glucose transporter with low substrate affinity that mediates renal tubular reabsorption of glucose (3033). To identify this novel transporter in the kidney, an mRNA database of 1000 total clones was generated by random sequencing of a male rat kidney cDNA library. After a BLAST search, ~40% of the clones were unknown and/or unannotated and were screened by measuring the uptake of various compounds using Xenopus oocytes. In addition, SLC60A2/MFSD4B has been proposed to be a potential channel for urea in the inner medulla of the kidney, which could potentially contribute to the high urea permeability in the thin limbs (3141).

Gene family members (2):
SLC60A1 (MFSD4A)	SLC60A2 (MFSD4B)

Molecular aspects, physiological roles and links to disease

The SLC60 family belongs to the Fucose:H⁺ Symporter (FHS) family (TC 2.A.1.7), which is part of the MFS superfamily. MFSD4A (SLC60A1) and MFSD4B (SLC60A2) are MFS-like proteins and MFSD4A (SLC60A1) was predicted to have 12 TMHs (3142).

SLC60A1: MFSD4A (SLC60A1) is most prominently expressed in stomach, kidney and in brain areas and at a lower level in intestine based on the HPA and other evidence (3127, 3143). It is involved in the regulation of food intake and energy production, suggesting that it is involved in nutrient regulation or glucose sensing (3142, 3144).

As mentioned above, MFSD4A (SLC60A1) has also been proposed to play a role in urea excretion (3145). More specifically, it has been reported to be involved in the opening of glucose-dependent sodium channels in the renal medulla, thereby affecting the osmotic gradient between the tubule and the interstitium. This process has been proposed to promote urea reabsorption in the kidney, which is dependent on both the osmotic gradient and the opening of sodium channels.

MFSD4A (SLC60A1) also has anti-tumor activities and has been shown to inhibit the malignant progression of nasopharyngeal carcinoma by targeting EPHA2 (ephrin-type A receptor 2) (3146) (3143). In this study, MFSD4A expression was found to be regulated by methylation of its promoter region in nasopharyngeal carcinoma, and it was proposed that MFSD4A could bind to and degrade EPHA2 by recruiting ring finger protein 149 (RNF149).

The non-coding RNA SLC60A1-AS1 was shown to act as an oncogene in papillary thyroid cancer by promoting lymphangiogenesis and facilitating lymphatic metastasis (3147). Its overexpression is associated with aggressive tumor behavior, making SLC60A1-AS1 a potential biomarker for prognosis and a target for therapeutic intervention in papillary thyroid cancer (3147). Although the exact molecular mechanisms are not fully understood, the study suggests that SLC60A1-AS1 may regulate genes and pathways associated with lymphangiogenesis and metastasis.

SLC60A2: MFSD4B (SLC60A2) is ubiquitously expressed based on the HPA. As mentioned above, it is proposed to function as a renal low-affinity Na⁺-dependent glucose transporter. However, it is unclear how this finding relates to the function of the kidney-specific low-affinity Na⁺-glucose transporter SGLT2 (SLC5A2), the main renal glucose reabsorption pathway and target for diabetes treatment (214). In addition, familial renal glycosuria, an inherited disorder resulting in glucose excretion in the urine, is most commonly due to mutations in the SLC5A2 gene, which is expressed in the early proximal tubule of the kidney (3148), and not MFSD4B (SLC60A2), which is widely expressed in many tissues.

Orphan transporter family members: N/A

HGNC update

MFSD4A and MFSD4B have been renamed as SLC60A1 and SLC60A2, respectively.

SLC61 Molybdate transporter family (2.A.1.40/MFS_5/MFS)

Discovery: The major facilitator superfamily domain-containing protein MFSD5 (SLC61A1), also known as MOT2, has been identified as an ortholog of a molybdate transporter in the alga Chlamydomonas reinhardtii MoT2 (3149).

Gene family member (1)

SLC61A1 (MFSD5)

Molecular aspects, physiological roles and links to disease

The SLC61 family belongs to the Major Facilitator Superfamily Domain-containing Protein (MFS-DP) family (TC 2.A.1.40), which is part of the MFS superfamily. MFSD5 (SLC61A1) has 12 putative TMHs according to a study of a green alga ortholog called MoT2, which has been shown to transport molybdate (3149).

Molybdenum (Mo) is an essential micronutrient for most living organisms and plays an important role in metalloenzymes. Mo exists predominantly in its most oxidized form, molybdate oxyanion (MoO₄²⁻), thought to be the major form taken up by plant roots (3150). In humans, four molybdenum-requiring enzymes have been identified to date: sulfite oxidase, xanthine oxidase, aldehyde oxidase, and the mitochondrial amidoxime-reducing component, but nutritional deficiencies induced by low dietary molybdenum have not been observed thus far (3151). Studies of MFSD5 orthologs from algae and plants confirmed their ability to transport molybdate (3149). Thus, MFSD5 (SLC61A1) in higher organisms was proposed to be a high-affinity Mo uptake transporter, probably functioning as an oxyanion molybdate (e.g., MoO₄²⁻)/anion uptake carrier (3149).

SLC61A1 - Orphan transporter: MFSD5 (SLC61A1) is widely expressed in human tissues according to the HPA. In brain regions, it has been implicated in energy homeostasis (3152), because it was shown to be regulated by dietary status and reported to interact with the GLP-1 receptor (GLP1R) (3152, 3153).

The precise transport function of MFSD5 (SLC61A1) is still unknown. Due to its homology with the algal molybdate transporter, it seems reasonable to hypothesize that it transports molybdate, and possibly other oxyanions such as sulfate, selenate, and chromate (3149, 3154).

While the transport function of MFSD5 (SLC61A1) remains unclear, MFSD5 (SLC61A1) emerged in a search for lipoprotein(a) [Lp(a)] target receptors on the surface of valvular interstitial cells (VICs) (see below), thus providing an interesting link between MFSD5 (SLC61A1) function, Lp(a) blood levels, and cardiovascular disease (3155).

This study was prompted by the observation that Lp(a) is a causal risk factor for calcific aortic valve stenosis by promoting the development of VIC calcification. Lp(a) is primarily synthesized in the liver and structurally resembles LDL, consisting of a lipid core of cholesteryl esters and triacylglycerols that is surrounded by phospholipids, free cholesterol, and apolipoprotein B-100 (apoB100). Its distinguishing feature compared to LDL is the presence of the polymorphic apolipoprotein(a) [apo(a)], which is linked to apoB-100 via a disulfide bond. Apo(a) plays a crucial role in the unique pathophysiological functions of Lp(a). Lp(a) is a major risk factor for atherosclerosis, cardiovascular disease, and stroke (3156–3158). Blood levels of Lp(a) are predominantly determined by genetics and are not substantially influenced by dietary modifications (3156), although some evidence suggests that non-genetic factors may also influence Lp(a) levels (3159).

It was anticipated in one study that preventing the uptake of Lp(a) in cardiovascular tissue would be clinically advantageous (3155). To identify the Lp(a) receptor related to valvular heart disease (VHD), researchers employed Ligand Receptor Capture Mass Spectrometry (LRC-MS) (3155), a crosslinking-based affinity purification technique for identifying cell surface receptors and their ligand interactions (3160). The study revealed elevated MFSD5 expression levels in calcified aortic valves from patients with VHD, suggesting its potential role in disease progression. Furthermore, the study demonstrates that MFSD5 promotes the absorption of Lp(a) into valvular interstitial cells, and siRNA-mediated silencing of MFSD5 significantly decreases Lp(a) absorption by valvular cells, thereby reducing calcification. In contrast, siRNA-mediated silencing of the low-density lipoprotein receptor (LDLR) gene did not reduce Lp(a) absorption. A genetic variant of MFSD5 that enhances the effect of Lp(a) on aortic stenosis was identified in the UK Biobank genetic dataset. Collectively, these findings imply that MFSD5 is directly linked to the calcific remodeling observed in VHD and plays a critical role in the pathogenesis of valvular heart disease by mediating Lp(a) uptake and contributing to valvular calcification. Targeting MFSD5 therapeutically could provide a new way to treat calcification in patients with VHD.

It remains to be determined whether MFSD5 interacts with Lp(a) directly or indirectly through a separate Lp(a) receptor. As was previously mentioned, interaction of MFSD5 with the GLP-1 receptor is known to occur. Therefore, it is possible that MFSD5 functions as a molybdate transporter regulated by interaction with such receptors. This would fit well with the pathophysiology of cardiovascular disease, as the molybdenum cofactor, a metal-sulfur complex, plays a crucial role in xanthine oxidase (3161, 3162). Under pathological conditions, increased xanthine oxidase activity is known to contribute to oxidative stress in vascular tissues, which plays a key role in endothelial dysfunction and the development of cardiovascular diseases, such as atherosclerosis (3163, 3164). Therefore, the major function of MFSD5 in the cardiovascular system may be to transport molybdate into valvular endothelial cells, enabling cytosolic xanthine to function properly. The potential therapeutic benefits of inhibiting xanthine oxidase have been discussed (3164). If MFSD5 indeed functions as a molybdate transporter, then inhibiting it would be at least as effective for the treatment of cardiovascular diseases.

Orphan transporter family member (1)

SLC61A1 (MFSD5)

HGNC update

MFSD5 has been renamed as SLC61A1.

SLC62 ATP exporter (2.A.66.9/ANKH/MATE)

Discovery: ANKH (SLC62A1) is the only human member of this family. The progressive ankylosis allele (Ank^ank) was originally identified in mice as a spontaneous recessive mutation causing a severe joint calcification phenotype (3165).

Gene family member (1)

ANKH (SLC62A1)

Molecular aspects, physiological roles and links to disease

ANKH (SLC62A1) is a member of the Multidrug/Oligosaccharidyl-lipid/Polysaccharide (MOP) flippase superfamily (TC 2.A.66) and possesses the MATE fold, giving rise to 12 TMHs (see Fig. 30).

In-depth characterization of the Ank^ank allele showed that it causes a generalized, progressive form of arthritis accompanied by mineral deposition, bony outgrowth formation, and joint destruction (3166). Subsequently, the Ank^ank locus was identified and shown to encode a multi-pass transmembrane protein expressed in joints and other tissues that controls cellular pyrophosphate (PP_i) levels as a possible mechanism regulating tissue calcification and susceptibility to arthritis in higher animals (3166).

ANKH (SLC62A1): ANKH (SLC62A1) is highly expressed in various human tissues including muscle, brain, retina, prostate, tongue, pituitary, liver, skin and intestine, as shown by the HPA.

ANKH (SLC62A1) causes craniometaphyseal bone dysplasia, an extremely rare genetic disorder characterized by progressive thickening of the craniofacial bones and abnormal development of the metaphyses in long bones (3167).

A follow-up report showed that the ANKH (SLC62A1) transporter is an exporter of ATP (305). The enzyme ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) then converts extracellular ATP to adenosine monophosphate and PP_i, thereby determining PP_i levels in the systemic circulation. Thus, while MRP6 (ABCC6) mediates the release of ATP from hepatocytes into the circulation, ANKH (SLC62A1) releases ATP from peripheral tissues into the circulation (305). An overview of ANKH function, binding partners, regulators of ANKH, and loss- and gain-of-function mutations of ANKH has been reported previously (3168).

Orphan transporter family members: N/A

SLC63 Spinster sphingosine-phosphate transporters (2.A.1.49/MFS_1/MFS)

Discovery: The spinster (SLC63) gene family encodes phylogenetically conserved proteins belonging to the major facilitator superfamily. Spinster (spin, also known as benchwarmer/bnch and Spns) was originally identified in Drosophila as a mutation that disrupts programmed cell death and causes neuronal degeneration (3169). Subsequently, in another study, loss of bnch (spinster) function in Drosophila resulted in disruption of a permease of the MFS super family, which was semi-lethal because it led to lysosomal storage problems in yolk globules during oogenesis and resulted in widespread accumulation of enlarged lysosomal and late endosomal inclusions (3170). In flies, spin mutants accumulate both carbohydrates and lipids based on histological staining methods (3170). There are homologs of the spin gene across species, including humans (3169), and based on sequence homology, the spin protein has been proposed to be a sugar transporter. However, there was also evidence that spin acts as a sphingolipid transporter from studies in zebrafish, where spin mutants resulted in impaired sphingolipid signaling (3171). While Drosophila contains one spinster family member, mammals have three Spns homologs, SPNS1 (SLC63A1), SPNS2 (SLC63A2) and SPNS3 (SLC63A3).

Gene family members (3):
SPNS1 (SLC63A1)	SPNS2 (SLC63A2)	SPNS3 (SLC63A3)

Molecular aspects, physiological roles and links to disease

The SLC63 family belongs to the Endosomal Spinster (Spinster) family (TC 2.A.1.49), which is part of the MFS superfamily.

SPNS1 (SLC63A1): SPNS1 (SLC63A1) is ubiquitously expressed according to the HPA. It has been unambiguously shown to function as a lysophospholipid transporter that mediates lysosomal phospholipid exit (3172). One of the key functions of lysosomes is to mediate hydrolysis of macromolecules, which requires degraded products to exit the lysosome. Failure of this process will result in lysosomal storage disease. Lysophospholipids are the degradation products of phosphatidylcholine and phosphatidylethanolamine, and through a combination of a cell-based screen and targeted lipidomics it was shown that SPNS1 (SLC63A1) serves as the rate-limiting step in the lysosomal efflux of lysophospholipids (3172).

Later, it was shown that SPNS1-dependent lysosomal lipid trafficking also enables cell survival under choline limitation (3173) and that loss of SPNS1 leads to intralysosomal accumulation of lysophosphatidylcholine (LPC) and lysophosphatidylethanolamine (LPE).

SPNS1 (SLC63A1) was shown to facilitate H⁺ gradient-dependent exit of LPC species from the lysosome for their reesterification into phosphatidylcholine in the cytoplasm (3173).

In kidney proximal tubules, SPNS1 was found to play an essential role in megalin-dependent endocytosis (3174). Knockout of Spns1 resulted in the arrest of megalin receptor-mediated endocytosis. Knockout was also associated with renal iron overload, probably because iron can enter the proximal tubule via megalin-cubilin by apical capture of iron bound to transferrin.

Lack of SPNS1 has been reported to result in accumulation of lysolipids and lysosomal storage disease (2658). The study provides functional evidence linking SPNS1 deficiency to lysosomal lipid accumulation in both human cells and mouse models. The affected patients exhibited symptoms including developmental delay, neurological impairment, intellectual disability, and cerebellar hypoplasia, which are reminiscent of lysosomal storage diseases. In Spns1 knockout mice, accumulation of these lysolipids in the lysosomes affected liver functions and altered the PI3K/AKT signaling pathway (2658). This work underscores the importance of SPNS1 in lysosomal lipid transport and storage disease pathology (2658).

Cryo-EM resolved the structure of human SPNS1 in an LPC-bound, lumen-facing conformation (3175). SPNS1 adopts the classical MFS fold consisting of 12 TMs that are divided into two pseudosymmetric domains: The N-domain includes TMs 1 to 6, and the C-domain includes TMs 7 to 12. The luminal opening surrounded by TMs 5 and 8 accommodates the polar head of LPC, indicating a specific binding site crucial for substrate recognition and transport.

Interestingly, the binding mode in the initial stage of the transport cycle was found to resembles that of MFSD2A (SLC59A1), which cotransports Na⁺ and LPC across the lipid bilayer. A structural comparison revealed that the inward-opening state of MFSD2A bound to LPC aligns well with the AlphaFold-predicted inward-opening conformation of SPNS1 (3175), suggesting a similar translocation path in these two structural homologues. The study furthermore shows that SPNS1 operates via a rocker-switch mechanism, characteristic of MFS transporters, involving alternating conformational changes that allow substrate translocation across the lysosomal membrane. Specific residues in TM5 and TM8 are pivotal in facilitating these conformational shifts, ensuring efficient LPC transport.

Interestingly, a unique five-residue network on the luminal side of SPNS1 was identified that functions as a crucial H⁺ sensing mechanism, enabling the transporter to respond to the acidic environment of the lysosome. SPNS2 does not have this H⁺ sensing mechanism, which is likely due to a lack of a complete network of H⁺ sensing residues.

In summary, SPNS1 represents a specialized lysosomal transporter that facilitates the efflux of lysophospholipids, such as LPC, from the lysosome to the cytosol. This efflux mechanism is finely tuned by a H⁺-sensing network responsive to lysosomal pH. Disruption of SPNS1 activity leads to lysolipid accumulation, contributing to lysosomal storage disorders and highlighting its importance in cellular lipid regulation and potential as a therapeutic target.

SPNS2 (SLC63A2): The paralog SPNS2 (SLC63A2) has been shown to be a sphingosine-1-phosphate (S1P) transporter. Two MFS transporters are involved in S1P export: MFSD2B (SLC59A2) exports S1P from red blood cells and platelets (3130), and SPNS2 (SLC62A2) exports S1P from vascular and lymphatic endothelial cells (3176). Experiments in zebrafish first suggested the role of spns2 in S1P secretion (3171, 3177).

As mentioned earlier (see the SLC59 family description), S1P is a signaling lipid that performs critical biological functions by binding to sphingosine-1-phosphate G protein-coupled receptors. It regulates the immune system, angiogenesis, auditory function, and epithelial and endothelial barrier integrity (3178–3180), as well as plays an important role in cancer, including cell growth and survival, metastasis, and chemoresistance (3181).

The physiological importance of SPNS2 is further supported by studies in transgenic mice. Spns2 knockout mice exhibit impaired lymphocyte trafficking, resulting in lymphopenia (3182), along with vascular abnormalities. These include defects in retinal vascular development (3183) and hearing loss due to cochlear dysfunction (3184).

SPNS2 (SLC63A2) plays a critical role in inflammatory and autoimmune diseases (3185), and SPNS2 deficiency protects mice from developing multiple sclerosis and other autoimmune diseases (3186). SPNS2 also reduces pulmonary metastasis (van der Weyden, 2017, #4802). Pharmacological modulation of SPNS2 is considered to have significant therapeutic potential.

Cryo-EM was used to resolve the structures of human SPNS2 in lipid nanodiscs, capturing multiple conformational states, including inward-facing, outward-facing, and intermediate forms (3187). These structures reveal the dynamic nature of SPNS2 during the S1P transport cycle. SPNS2 adopts the canonical MFS fold, comprising 12 TMHs that are divided into two pseudo-symmetric domains, the N-domain (TMHs 1–6) and the C-domain includes (TMHs 7–12). The results demonstrate how hydrophobic lipid cargo is transported via facilitated diffusion in a rocker-switch, alternating-access mechanism. The study also investigated the binding of the SPNS2 inhibitor 16d (SLF1081851) (3188). Structural data indicate that 16d locks SPNS2 in an inward-facing conformation, which effectively reduces its transport activity.

Subsequently, additional cryo-EM structures of human SPNS2 in multiple conformational states were resolved, focusing on pharmacological modulators (3189). The study elucidated the binding modes of SPNS2 with its native substrate, S1P, the therapeutic analog, FTY720-P (fingolimod), and the potent, selective inhibitor, 33p (SLB1122168). The findings provide pharmacological tools to modulate extracellular S1P levels, which has potential for therapeutic applications in autoimmune diseases, cancer, and vascular pathologies (3189).

The cryo-EM study also identified how specific SPNS2 mutations disrupt export activity, leading to physiological consequences such as hearing loss and reduced white blood cell counts, conditions that are reminiscent of lysosomal storage diseases (3189). This work provides a molecular framework for understanding how S1P is exported from cells and offers a novel strategy for therapeutically modulating S1P signaling, particularly in diseases involving pathologically elevated extracellular S1P.

SPNS3 (SLC63A3) - Orphan transporter: SPNS3 (SLC63A3) is an orphan MFS transporter with an unknown substrate and function. Although its homology to SPNS1 and SPNS2 suggests that it may be involved in lipid or metabolite transport, there is currently no direct evidence to support a specific role. Overexpression of SPNS3 has been associated with a poor prognosis in patients with acute myeloid leukemia (AML) undergoing chemotherapy or allogeneic hematopoietic stem cell transplantation. It has been suggested that SPNS3 overexpression may regulate the proliferation and differentiation of AML cells through autophagy (3190). However, there is currently no consensus on any diseases associated with SPNS3, and its expression patterns are not well understood. According to public databases, including the HPA, there is a possibility of expression in immune tissues, among others.

Orphan transporter family member (1)

SPNS3 (SLC63A3)

SLC64 Golgi Ca²⁺/H⁺ and/or Mn²⁺/H⁺ antiporter (2.A.106.2/UPF0016/SLC64)

Discovery: TMEM165 (SLC64A1) is the sole member of this family and was discovered as a potential biomarker for invasive ductal carcinoma in a glycoproteomic study using comparative membrane proteomics (3191).

Gene family member (1)

TMEM165 (SLC64A1)

Molecular aspects, physiological roles and links to disease

TMEM165 (SLC64A1) belongs to the Ca²⁺:H⁺ antiporter family CaCA2 (TC 2.A.106.2), a small family of proteins found in bacteria, archaea, yeast, plants and animals involved in ion homeostasis. They usually have 5-7 transmembrane domains, and their primary sequence is defined by the presence of 1 or 2 copies of the E-ϕ-G-D-[KR]-[TS] consensus motif in the transmembrane domains. A TMEM165 structural model has been constructed using AlphaFold 2 (see below).

TMEM165 (SLC64A1) - Orphan transporter: TMEM165 (SLC64A1) is a widely expressed endomembrane Mn²⁺ and/or Ca²⁺ transporter. The function of TMEM165 (SLC64A1) has been the subject of ever-evolving research. Historically, it has been characterized as a Golgi-localized Mn²⁺/H⁺ antiporter essential for proper glycosylation processes. However, subsequent studies have suggested an additional role for TMEM165 in lysosomal Ca²⁺ transport. These functions are not mutually exclusive and may reflect the versatility of the protein in different cellular compartments. Further research is necessary to elucidate the mechanisms that govern the dual roles of TMEM165 and their implications for health and disease.

TMEM165 (SLC64A1) has been reported to play an important role in Mn²⁺ homeostasis in the Golgi, which is critical for proper glycosylation (3192). TMEM165 is thought to serve either as an importer of Mn²⁺, acting as a Mn²⁺/H⁺ antiporter into the Golgi, where this cation is a mandatory cofactor for many glycosyltransferases (3193). However, it has also been considered a Golgi Ca²⁺/H⁺ antiporter that modifies the Golgi’s Ca²⁺ and pH balance, as several homologs from bacteria, plants, and eukaryotes have been shown to be involved in Ca²⁺ and Mn²⁺ homeostasis (3194).

Defects in human TMEM165 (SLC64A1) cause congenital disorder of glycosylation type 2K (CDG2K), an autosomal recessive disorder with variable phenotypes (3195). This finding underscores the importance of TMEM165 in Mn²⁺ transport and glycosylation pathways.

Moreover, TMEM165 (SLC64A1) expression has been shown to be elevated in all types of breast cancer and has been shown to alter the expression levels of key glycoproteins involved in regulating the epithelial-to-mesenchymal transition, such as E-cadherin, thereby promoting invasion and growth (3196).

On the other hand, patch-clamp analysis of HeLa cells stably expressing TMEM165 tagged with RFP at the C-terminus supported the concept that TMEM165 may function as a Ca²⁺/H⁺ antiporter (3197). It has been suggested that alteration of the Golgi Ca²⁺ and pH balance could explain the glycosylation defects observed in TMEM165-deficient patients. Thus, it is still an open question whether TMEM165 primarily transports Mn²⁺ or Ca²⁺, and the precise transport function of TMEM165 (SLC64A1) remains unclear.

A recent structural model has shed new light on the effect of TMEM165 transport mutants found in patients, but could not resolve the question of divalent metal ion selectivity (3193). A TMEM165 structural model was constructed using AlphaFold 2 and refined by molecular dynamics simulation with membrane lipids and water, yielding a 3D protein scaffold consisting of a two-fold repeat of three transmembrane helices/domains with the consensus motifs (108-ELGDKT-113 and 248-EWGDRS-253) facing each other to form a putative acidic cation binding site on the cytoplasmic side of the protein (3193). Putative acidic divalent metal ion binding sites (E108 and E248) may be involved in stabilizing the binding of Mn²⁺, although they could bind other cations such as Ca²⁺ as well.

Follow-up reports have revealed a new function of TMEM165 as a lysosomal Ca²⁺ importer (3198, 3199). Although TMEM165 has been shown to localize to the Golgi apparatus, a recent study confirmed that a significant portion is localized to the lysosome. In lysosomes, TMEM165 has been reported to be a pH-activated Ca²⁺ importer rather than a Mn²⁺/H⁺ antiporter. This finding is particularly important because lysosomal transporters that import calcium into lysosomes have been difficult to identify. The research demonstrated that TMEM165 facilitates Ca²⁺ uptake into lysosomes, particularly under acidic conditions, suggesting a role in lysosomal calcium homeostasis (3198).

Whether TMEM165 has dual localization and function, involving its roles in Golgi manganese transport and lysosomal calcium import, remains to be determined.

A review of the biology of TMEM165 discussed its roles in cellular ion homeostasis and glycosylation, as well as its potential dual localization and function within the Golgi apparatus and lysosomes (3200). In brief, the following is stated: TMEM165 is predominantly recognized as a Ca²⁺/Mn²⁺:H⁺ antiporter that is localized in the medial and trans-Golgi network, where it is crucial for importing Mn²⁺ into the Golgi lumen. This process is vital for the activity of glycosyltransferases that are involved in protein and lipid glycosylation. Disruptions in TMEM165 activity lead to CDG2K due to impaired Mn²⁺ transport. Follow-up studies have proposed an additional role for TMEM165 as a pH-activated calcium importer in lysosomes. Under conditions of elevated cytosolic Mn²⁺, TMEM165 relocates from the Golgi to lysosomes, where it undergoes degradation. This translocation suggests a regulatory mechanism in which TMEM165 helps maintain lysosomal calcium homeostasis. However, the exact mechanisms and implications of this function are still under investigation.

Thus, more studies are needed to clarify the dual roles of TMEM165. These will be essential for understanding its precise contributions to cellular physiology and the pathogenesis of related disorders.

In further support of TMEM165-mediated Mn²⁺ transport, there has been a report of the first successful application of combined oral Mn²⁺ and D-galactose therapy in a patient with a newly identified TMEM165 mutation (3201). The patient had a previously unreported homozygous missense mutation in TMEM165, which resulted in a protein variant (A310P) that was functional but unstable, leading to glycosylation defects. The patient was diagnosed at two months old and exhibited a predominant bone phenotype and combined defects in N-, O-, and glycosaminoglycan glycosylation. A combined oral therapy regimen was initiated, involving daily doses of D-galactose and Mn²⁺. This regimen led to complete normalization of all assessed glycosylation pathways (3201).

Orphan transporter family member (1)

TMEM165 (SLC64A1)

SLC65 NPC-type cholesterol transporters (2.A.6/Patched/RND)

Discovery: NPC1 (SLC65A1) was identified by positional cloning in search of the gene defective in patients with Niemann-Pick type C (NP-C) disease, a fatal neurovisceral disorder that is characterized by lysosomal accumulation of low-density lipoprotein (LDL)-derived cholesterol (3202).

Gene family members (7):
NPC1 (SLC65A1)	PTCH2 (SLC65B2)	PTCHD3 (SLC65C3)
NPC1L1 (SLC65A2)	PTCHD1 (SLC65C1)
PTCH1 (SLC65B1)	PTCHD4 (SLC65C2)

Molecular aspects, physiological roles and links to disease

The SLC65 family has 7 members (Fig. 54) and belongs to the Resistance-Nodulation-Cell Division (RND) Superfamily (TC 2.A.6) (see human RND fold family description in Section 8). It includes the eukaryotic branch of sterol transporters: The Niemann-Pick type C1 protein NPC1 (SLC65A1) and the intestinal hepatic cholesterol transporter NPC1L1 (SLC65A2) (Fig. 55), as well as the closely related proteins PTCH1 (SLC65B1) and PTCH2 (SLC65B2), which serve as hedgehog (Hh) receptors, and finally, the orphan transporters PTCHD1 (SLC65C1), PTCHD4 (SLC65C2) and PTCHD3 (SLC65C3).

Fig. 54. Phylogenetic tree of the SLC65 family.

Fig. 55. Cholesterol transported via NPC1L1 and NPC1 in the intestine followed by delivery to steroidogenic tissues such as adrenal cortex or liver.

a) Cholesterol transport in the intestinal brush border membrane via NPC1L1 (SLC65A2). b) After binding to NPC2 in the endosomal/lysosomal lumen, cholesterol is released into the cytosol via NPC1 (SLC65A1). From there, it is delivered to other cellular compartments such as ER or mitochondria via specialized mechanisms. Inset: Close-up of NPC2/NPC1-mediated cholesterol transport. NTD, N-terminal luminal domain. SSD, sterol-sensing domain. The green spheres in the inset refer to docked cholesterol. Figure created de novo; elements of the conceptual framework are based on multiple sources, including (3).

RND transporters typically contain 12 TMHs and two large external loops between TMHs 1 and 2 and between 7 and 8 (329), as shown in Fig. 4. In addition, the structures of the transmembrane regions of NPC1 (SLC65A1) and NPC1L1 (SLC65A2) harbor the sterol-sensing domain (SSD) that is present in several eukaryotic transmembrane proteins related to the regulation of cholesterol levels (3203, 3204). Fig. 55 shows human NPC1 with 12 TMHs, the SSD, and two distinct luminal domains. Structurally, the SSD resembles the membrane-spanning domain of prokaryotic transporters of the RND family whereas the two 6-TMH halves of the transporter core structure are organized in reverse order on the polypeptide chain (3). Certain multidrug efflux RND transporters are believed to extract their substrates directly from the membrane bilayer (3205, 3206). In contrast, NPC1 picks up cholesterol from the soluble carrier protein NPC2 via its N-terminal domain, and NPC1L1 picks up cholesterol from mixed micelles at the apical surface of enterocytes via its N-terminal domain (3207). After binding, the cholesterol is believed to travel through an internal channel to the sterol-sensing domain (SSD), where it is released into the membrane bilayer (Fig. 55) (32, 3207).

It has been proposed that the two domains of NPC2 facing the endosomal lumen form a pocket-relay system for cholesterol transport in tandem with the N-terminal domains and SSD, the activity of which is regulated by the cholesterol concentration of the adjacent lipid bilayer (3, 3208).

Interestingly, the SSD is also present in proteins involved in cholesterol metabolism and Hh signaling, proteins that were not necessarily thought to be involved in cholesterol transport. These include the cation-driven Hh signaling-related proteins PTCH1 (SLC65B1; TC 2.A.6.6.13) and DISP1 (TC 2.A.6.9.2) that are important in embryonic development and tumorigenesis (3209, 3210), the SREBP cleavage activating protein (SCAP; TC 2.A.6.6.4), which is required for lipid synthesis in the liver in response to cholesterol deficiency (3211–3213), and the 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase (HMGCR; TC 2.A.6.6.5), the rate-limiting enzyme for cholesterol synthesis (3214). Currently, DISP1, SCAP, and HMGCR are not included in the SLC65 nomenclature system. However, they are included in the SLC-like protein category (see Section 10).

Cryo-EM structures of human PTCH1 confirmed the structural similarity of NPC1 and PTCH1 (3215–3218). Both proteins have the 12 TMH α-helices arranged as two groups of six TMHs, and the two homologous extracellular domains form a twofold pseudosymmetric structure (see Fig. 4 and Fig. 55). As reviewed (3218), cholesterol-like densities were seen at two locations in PTCH1 in the cryo-EM structure. One is sandwiched between the two extracellular domains, and the other occupies a membrane-facing cavity in its SSD of the transmembrane region, analogous to a cholesterol-sized pocket seen in the NPC1 cholesterol transporter (3) (Fig. 55). These cholesterol densities appear to be connected by a tunnel across PTCH1 through the membrane, which is large enough to accommodate cholesterol (3218). The tunnel has been proposed to allow the transfer of sterol-like molecules from the SSD within the membrane to the extracellular space. Consistent with this, previous studies have directly implicated PTCH in cholesterol transport, with the direction of cholesterol transport depending on the concentration gradient (3219). The possibility that PTCHD1 also has cholesterol binding and/or transport activity is raised by the fact that both PTCH1 and NPC1 are involved in cholesterol transport (3220).

NPC1 (SLC65A1): Originally named “Niemann-Pick disease, type C1”, NPC1 (SLC65A1) is a ubiquitously expressed lysosomal cholesterol transporter (3). After LDL receptor-triggered endocytosis, cholesterol is delivered in esterified form to the lysosome where lysosomal acid lipase hydrolyzes cholesterol esters to release free cholesterol (Fig. 55). Given the water-insoluble nature of free cholesterol, it must be immediately bound to proteins with hydrophobic sterol-binding domains, or derivatized to make it more water-soluble. Free cholesterol is bound by NPC2, a small soluble lysosomal luminal protein of 132 amino acids. The NPC2 protein then transfers bound cholesterol to NPC1 to facilitate delivery of cholesterol to the cytoplasm, from where it is transported to other cellular compartments such as ER or mitochondria via specialized pathways (3221–3223). In the ER, cholesterol is monitored by sensors such as the sterol regulatory element-binding protein cleavage-activating protein (SCAP), and when cholesterol levels are reduced, this ultimately leads to increased expression of 3-hydroxy-3-methylglataryl coenzyme A reductase (HMGCR), the rate-limiting enzyme for cholesterol synthesis and the target of “statin” inhibitors of cholesterol synthesis (see Section 10 for SLC-like proteins such as SCAP and HMGCR) (Fig. 55).

Niemann-Pick type C (NPC) is a devastating neurodegenerative disease caused by massive accumulation of unesterified cholesterol in the late endosomes/lysosomes of cells due to defective lysosomal cholesterol efflux caused by mutations in NPC1. Accumulation of cholesterol in lysosomes leads to mTORC1 hyperactivation, disrupted mitochondrial function, and neurodegeneration (2106, 2659, 3224).

While lysosomes are major degradative organelles, they also contain critical signaling compartments that support mTORC1 activity. Intracellular nutrients drive mTORC1 translocation from the cytosol to the lysosomal membrane. Like amino acids (see SLC38A9 description), lysosomal cholesterol enables mTORC1 activation by relocalizing it from the cytosol to the lysosomal membrane. This process requires the G protein-coupled receptor GPR155, also known as LYCHOS, a chimera between an SLC transporter and a G protein-coupled receptor that senses lysosomal cholesterol (261) (see Section 10, description of GPR155, for more details). During nutrient deprivation, GPR155 is downregulated, which reduces mTORC1 anabolic signaling, while NPC1 is upregulated, which promotes the export of cholesterol from the lysosomal surface and inhibits mTORC1 signaling (261) (Fig. 55). Interestingly, the lysosomal amino acid transporter SLC38A9 (see SLC38A9 description) was found to be required for cholesterol-dependent activation of mTORC1 under nutrient-rich conditions through conserved cholesterol-responsive motifs (3225).

Microglia, the immune cells of the brain, play a major role in the pathogenesis of NPC. Loss of the NPC1 protein in microglia leads to altered morphology, impaired lipid trafficking, and enhanced phagocytosis in NPC, potentially contributing to neuronal damage (2106, 3226). As discussed in the description of SLC25A53, TREM2 is a key transcriptional regulator of cholesterol transport and metabolism via the PI3K-mTOR signaling pathway. TREM2 localizes in lipid rafts; the altered lipid components and cholesterol concentrations of these rafts are associated with altered TREM2 levels in the plasma membrane of reactive microglia (2105). This depends on NPC1-mediated cholesterol exit from lysosomes because NPC1 deficiency blocks the trafficking of cholesterol from lysosomes to the plasma membrane. This disrupts TREM2-dependent metabolic regulation in microglia, including the dysregulation of key signaling pathways like the TREM2-mTOR pathway, further impacting their function and contributing to neurodegeneration. A further study examining the loss of NPC1 specifically in myeloid cells highlights the significant role of microglial NPC1 in maintaining lipid homeostasis and preventing neuroinflammation (3227). Thus, NPC1 is essential for proper microglial development and function, and genetic deletion of the Npc1 gene alters cholesterol regulation and impairs microglial development (2106, 3226, 3227).

NPC1L1 (SLC65A2): The second member of the SLC65 family, NPC1L1 (SLC65A2) also transports cholesterol. It is highly and specifically expressed in the brush border membrane of the duodenum as well as in the liver (3228, 3229) (Fig. 55). It mediates intestinal cholesterol absorption and can limit hepatobiliary cholesterol excretion, thereby protecting the body from excessive loss of cholesterol (3229). NPC1L1 (SLC65A2) is the target of ezetimibe, a drug that inhibits cholesterol absorption and effectively lowers blood cholesterol levels in humans (81) (Fig. 55). In addition, the combination of ezetimibe and a statin was found to be effective in producing additional LDL cholesterol reduction on top of statins and also in reducing cardiovascular events (3230).

PTCH1 (SLC65B1): PTCH1 (SLC65B1), which stands for “patched homolog 1”, is a ubiquitously expressed Hh receptor, with a particularly high expression in testis and cervix according to the HPA. PTCH1 has been reported to function as a cation-powered cholesterol transporter as part of the Hh pathway (3231). Molecular dynamics simulations and free energy calculations support the function of PTCH1 as a cholesterol transporter, according to which one to three Na⁺ or two to three K⁺ couple to cholesterol export (3232).

In mammals, there are three hedgehog (Hh) proteins: Sonic hedgehog (SHH), Indian hedgehog (IHH) and desert hedgehog (DHH). These proteins act as Hh ligands that play distinct and crucial roles in embryonic development and various physiological processes (3233). SHH and IHH have important functions in several tissues, which sometimes coincide (3234). SHH plays a particularly significant role in the specification of cell types in the nervous system and the patterning of limbs, digits and eyes, whereas IHH plays an important role in skeletal development, primarily in endochondral ossification. DHH is found only in the gonads, including the granulosa cells of the ovaries and the Sertoli cells of the testes. SHH is the most potent of the three ligands and is expressed most widely in embryos and adult tissues. Furthermore, aberrant SHH signaling has been linked to endodermally derived human cancers known as carcinomas (3234).

The Hedgehog (Hh) signaling pathway plays a crucial role in the development of many organs and tissues, including the central nervous system, skeleton, musculature, gastrointestinal tract, lungs, limbs (including digits), and eyes. It regulates cell proliferation and the differentiation of cells during organ development (3209, 3210, 3235). Below is a brief summary of the Hh pathway, with a focus on the roles of PTCH1 and DISP1, an SLC-like homolog (3210, 3236) (see also the description of SLC-like transporters in Section 10):

1. At the plasma membrane, the cholesterol transporter PTCH1 normally inhibits the G protein-coupled receptor Smoothened (SMO) by limiting its access to cholesterol or oxysterols in the membrane through its cholesterol transport activity (3236–3239) (Fig. 56).

2. Activation of the Shh pathway is triggered by the SHH signaling protein. SHH is produced and secreted by different cell types at various developmental stages and plays a role in organ development, regulating cell growth and differentiation as well as the formation of body structures, as indicated above (3239, 3240). Due to post-translational modifications (palmitoylation, cholesterylation, myristoylation) SHH is highly hydrophobic (3241). SHH can signal in an autocrine or paracrine manner. In the latter case, SHH secretion is required, mainly through the Dispatched 1 (DISP1) protein. Production and secretion of SHH in secreting cells involves activation of SHH in the endoplasmic reticulum, where the SHH precursor is processed to form SHH-N, the C-terminus of which is then attached to cholesterol and the N-terminus to palmitic acid. The resulting SHH derivative can then be transported out of the secretory cells with the help of DISP1 to reach its target cells (3239). It has been proposed that the SHH-cholesterol derivative is transported directly across DISP1 along a pathway similar to that of its homologue NPC1, where cholesterol is thought to be transported across its hydrophobic extracellular conduit, the ECD interfaces (3239) (Fig. 55; for DISP1, the transport would be in the opposite direction to that shown for NPC1 in the inset). SCUBE2, a secreted glycoprotein, assists in the transport of the SHH-cholesterol derivative from the SHH producing cell to the target cell.

Fig. 56. Roles of PTCH1 and DISP1 in the Hedgehog (HH) signaling pathway.

SHH-producing cells secrete the sonic hedgehog (SHH) protein, a signaling molecule crucial for development and homeostasis (top part of the figure). The HH pathway plays a role in the development of many tissues during embryogenesis and has also been shown to function in self-renewing adult tissues (1). SHH-target cells (bottom part of the figure) respond to the SHH protein by activating downstream signaling pathways that lead to changes in gene expression and cellular behavior. These cells play crucial roles in embryonic development, tissue regeneration, and diseases, including cancer. Figure created de novo; elements of the conceptual framework are based on multiple sources, including (5).

3. Once SHH-cholesterol has reached its target cell, it binds to and inhibits PTCH1 (3209).

4. This increases the accessibility of cholesterol in the extracellular leaflet of the membrane, allowing cholesterol to bind and activate SMO (3242).

5. SMO then transduces the signal to cytoplasmic effectors, and this in turn leads to the activation of signal to glioma transcription factors (Gli), which move to the nucleus and activate the specific genes needed to promote development.

Genetic variations in the PTCH1 gene have been investigated in the context of reproductive cancers, including ovarian and endometrial cancers. One notable finding is that the PTCH1 rs357564 variant may have a protective effect against ovarian cancer, as the minor allele (T) was more frequent in control individuals, whereas the major allele was associated with increased cancer risk. Since PTCH1 is a key component of the Hedgehog (Hh) signaling pathway, which regulates cell growth and differentiation, variants in this gene may disrupt normal Hh signaling and contribute to tumorigenesis in reproductive tissues (3243).

PTCH2 (SLC65B2): PTCH2 is a homolog of the PTCH1 Hh receptor (3239). It is also a Hh receptor like PTCH1. According to the HPA, PTCH2 is highly expressed in parathyroid glands, salivary glands, ovaries, and testes (Leydig cells), and at lower levels in the brain and other tissues.

A dual germline mutation in PTCH2 – S391* (a truncating mutation) and L104P (a missense mutation) – was identified in a family with a history of glioma, as reported in a case study, suggesting PTCH2 as a potential genetic susceptibility factor for glioma (3244).

PTCHD1 (SLC65C1) - Orphan transporter: PTCHD1 (SLC65C1), which stands for “patched domain containing 1”, is predominantly expressed in the developing and adult brain, and is involved in synaptic function and neuronal signaling. PTCHD1 is a susceptibility gene for autism spectrum disorder and intellectual disability. (3245, 3246). It is particularly highly expressed in astrocytes and excitatory neurons in the cerebellum and other brain areas, according to the HPA. PTCHD1 binds cholesterol but not SHH, suggesting a distinct cellular function (3220).

Detailed clinical insights into the neurodevelopmental impact of PTCHD1 mutations in four affected individuals (two children and two adults) were reported (3246). All four individuals exhibited psychomotor developmental delays and varying degrees of intellectual disability due to PTCHD1 deletions or mutations. In early childhood, intellectual disability was associated with autistic-like behaviors. However, these features were not observed in the adult subjects, suggesting a potential attenuation with age. The patients did not display any distinctive dysmorphic features, congenital abnormalities or epilepsy, suggesting that PTCHD1-related disorders primarily affect cognitive and behavioral domains rather than manifesting as physical symptoms. Thus, PTCHD1 encodes a protein with a patched domain that is predominantly expressed in the developing and adult brain and implicated in synaptic function and neuronal signaling.

Another study on the neurodevelopmental impact of PTCHD1 mutations provided significant insight into the regulatory mechanisms controlling PTCHD1 expression (3245). The study identified a conserved enhancer region located downstream of the PTCHD1 coding sequence. This region is found in an open chromatin region that is unique to the forebrain and is inaccessible in non-neuronal tissues, indicating neuronal-specific regulatory activity. The enhancer contains binding sites for transcription factors that are known to play a role in the regulation of neuronal genes, suggesting that they are involved in modulating PTCHD1 expression. By identifying an enhancer that specifically regulates PTCHD1 transcription in neurons, the study establishes a potential link between genetic variants in regulatory regions and the manifestation of autism spectrum disorder and intellectual disability.

PTCHD4 (SLC65C2) – Orphan transporter: PTCHD4 (SLC65C2), which stands for “patched domain containing 4”, is a transmembrane protein that has been implicated as an integral component of the cellular membrane, and as a protein receptor in the Hh pathway. According to the HPA, it is almost exclusively expressed in neurons and microglial cells.

PTCHD4 (also known as PTCH53) was identified as a homolog of PTCH1 and has been found to be transcriptionally activated by p53 (TP53) in various human cells and tissues (3247). PTCHD4 functions as a repressor of the canonical Hh pathway by inhibiting SMO. PTCHD4 expression was found to be significantly reduced in TP53 mutant human tumors, suggesting that loss of p53 function leads to diminished PTCHD4 levels and aberrant activation of Hh signaling. The study highlights the direct link between p53 activity and Hh pathway regulation through PTCHD4 induction. Thus, PTCHD4 emerges as a potential mediator of tumor-suppressive effects of p53 by restraining oncogenic Hh signals.

A subsequent study investigating the regulation of PTCHD4 through post-transcriptional N⁶-methyladenine (m6A) RNA methylation provided insights into the role of PTCHD4 in cellular senescence (3248). The study demonstrates that the METTL3/METTL14 m6A methyltransferase complex (3249) adds m6A modifications to PTCHD4 mRNA, particularly in senescent human fibroblasts (3248). These modifications increase the stability of PTCHD4 mRNA, leading to elevated protein levels. The m6A reader protein IGF2BP1 then binds to the methylated PTCHD4 mRNA, further stabilizing it and promoting translation. This interaction highlights the importance of m6A readers in regulating gene expression during senescence. Functionally, increased PTCHD4 expression enabled senescent cells to resist apoptosis. Silencing PTCHD4 in pre-senescent cells has been shown to lead to enhanced growth arrest and increased DNA damage, indicating its protective role in senescent cells. The study also addresses the role of PTCHD4 in senolytic treatment strategies aimed at suppressing cell ageing by removing senescent cells (3248). Overall, this study unveils a novel function of PTCHD4 in promoting the survival of senescent cells via m6A-mediated post-transcriptional regulation. By stabilizing PTCHD4 mRNA, cells enhance the expression of a protein that helps them withstand senescence-associated stresses, which may contribute to the accumulation of senescent cells in ageing tissues and age-related diseases.

These findings enhance our understanding of the roles of PTCHD4 in cancer biology and senescence and may inspire future therapeutic strategies targeting PTCHD4 and its associated pathways.

PTCHD3 (SLC65C3) – Orphan transporter: PTCHD3 (SLC65C3), which stands for “patched domain containing 3”, is almost exclusively expressed in lymphatic endothelial cells, spermatocytes, and to a lesser extent in adipocytes, according to the HPA. Mouse PTCHD3 was previously hypothesized to function as a receptor in the Hedgehog (Hh) signaling pathway during spermatogenesis (3250). However, knockout studies in mice showed that Ptchd3 is a non-essential gene in mouse development and spermatogenesis and Ptchd3 knockout mice were born at expected Mendelian ratios and exhibited normal growth, development, and fertility (3251). Whether this is also true for human spermatogenesis remains to be determined.

PTCHD3 has been identified as a potential biomarker for lead-induced neurodevelopmental toxicity. In a DNA methylation profiling study of blood samples from 333 children in southern China, PTCHD3 was identified as a significant epigenetic mediator of the relationship between lead exposure and reduced intelligence in children (3252). A follow-up study has illuminated how environmental lead exposure impacts cognitive development by altering the epigenetic modifications of the PTCHD3 gene (3253). The study established dose-response relationships between lead exposure and PTCHD3 DNA methylation, suggesting that increased lead exposure is associated with altered PTCHD3 methylation patterns. Moreover, PTCHD3 methylation levels were found to correlate with variations in intelligence quotient scores. These results imply that PTCHD3 may influence neurodevelopmental processes affected by toxic environmental factors.

It is interesting to hypothesize that the high expression of PTCHD3 (SLC65C3) in lymphatic endothelial cells, as indicated by the HPA, contributes to the neurodevelopmental process associated with toxic environmental factors. These cells play a critical role in brain development and function as they are part of the immune surveillance and waste removal system of the brain. In particular, lymphatic endothelial cells are found in the leptomeninges – the arachnoid and pia mater – which play essential roles in cerebrospinal fluid circulation and central nervous system protection (3254).

Thus, monitoring environmental pollutants by examining their epigenetic effects on genes like PTCHD3 could help to better understand and mitigate risks to children’s neurodevelopment.

Orphan transporter family members (3)

PTCHD1 (SLC65C1), PTCHD4 (SLC65C2), PTCHD3 (SLC65C3)

HGNC updates

SLC65B1 is a new alias for PTCH1, SLC65B2 is a new alias for PTCH2, SLC65C1 is a new alias for PTCHD1, SLC65C2 is a new alias for PTCHD4, and SLC65C3 is a new alias for PTCHD3.

SLC66 PQ-loop amino acid transporter family (2.A.43/PQ-loop/SWEET)

Discovery: Cystinosin, the founding member of the SLC66 family, is a lysosomal cystine exporter that is expressed in renal proximal tubule epithelial cells, among other tissues, and is defective in patients with cystinosis. It was mapped to chromosome 17p13 and the locus that is deleted in cystinosis was identified, leading to the identification of this novel protein called cystinosin (CTNS/SLC66A4) (3255), followed by the confirmation of its function as an H⁺-driven lysosomal cystine transporter (3256).

Gene family members (5 + 1 pseudogene):
SLC66A1 (PQLC2)	SLC66A2 (PQLC1)	SLC66A3 (PQLC3)
CTNS (SLC66A4)	MPDU1 (SLC66A5)	SLC66A1LP (pseudogene)

Molecular aspects, physiological roles and links to disease

The SLC66 family comprises 5 protein coding genes (Fig. 57) and a pseudogene. It belongs to the lysosomal cystine transporter (LCT) family (TC 2.A.43), which is part of the SWEET fold families. PQ-loop proteins are 7 TMH membrane proteins that are distantly related to bacteriorhodopsin and characterized by a duplicated motif called the PQ-loop, which often contains the amino acid doublet PQ (3257, 3258). The duplicated motif encompasses part of the second and fifth transmembrane helices, including the connecting cytoplasmic loops N-terminal to these helices, and serves as a key functional element (3257).

Fig. 57. Phylogenetic tree of the SLC66 family.

Among several S. cerevisiae PQ-loop proteins of unknown function, Ypq1–3 were shown to be involved in homeostasis in a vacuolar export mechanism of cationic amino acids (3257), with distant similarity to SWEET sugar transporters and to the mitochondrial pyruvate carrier (SLC54 family).

SLC66A1: PQLC2 (SLC66A1) was originally identified as a lysosomal cationic amino acid transporter (3257). Based on the HPA, it is expressed throughout the body, most abundantly in the cerebellum and liver, and also prominently in the retina.

Interestingly, PQLC2 (SLC66A1) is more closely related to the yeast proteins Ypq1-3 than to human cystinosin (CTNS/SLC66A4) which is associated with cystinosis (3257) (see below). Nevertheless, PQLC2 may serve as a target for the treatment of cystinosis caused by the genetic defect of cystinosin, since PQLC2 transports a key chemical intermediate in cysteamine therapy of cystinosis (3257).

PQLC2 (SLC66A1) has a multifaceted role in lysosomes, functioning dually as a transporter and a sensor (i.e., transceptor)(3259–3261):

• 1)It mediates the export of the cationic amino acids arginine, lysine, and histidine from the lysosomal lumen to the cytosol.
• 2)Beyond its transport function, PQLC2 also acts as a receptor that recruits the C9orf72-SMCR8-WDR41 complex to lysosomes during amino acid starvation (3259, 3260). (C9orf72-SMCR8-WDR41 is a heterotrimeric protein complex that plays a critical role in lysosomal function (3259, 3260)).
• 3)The transporter is regulated by cytosolic arginine, which binds to PQLC2 and induces conformational changes that inhibit its transport activity, forming a feedback loop that links cytosolic amino acid levels to lysosomal function.
• 4)The interaction with the C9orf72 complex, mediated via WDR41, is sensitive to lysosomal cationic amino acid content and helps signal amino acid sufficiency or deficiency to downstream pathways.

Through this dual function, PQLC2 serves as a key regulator of lysosomal amino acid homeostasis and signaling, linking nutrient sensing to the localization and activity of the C9orf72 complex. Dysregulation of this axis has implications for neurodegenerative disorders, including amyotrophic lateral sclerosis and frontotemporal dementia (3262, 3263).

Through a comprehensive analysis of exome sequencing data from 913 individuals with inherited retinal diseases, homozygous pathogenic variants in several SLCs were identified (3264). Notably, two distinct variants in SLC66A1 were found to cause autosomal recessive retinitis pigmentosa. Understanding the specific substrates and pathways involving PQLC2 (SLC66A1) could provide insights into the molecular mechanisms of retinal degeneration and potential therapeutic targets.

SLC66A2 - Orphan transporter: PQLC1 (SLC66A2) is ubiquitously expressed, most abundantly in liver, as suggested by the HPA.

A study of the yeast orthologue of PQLC1 (SLC66A2), known as Any1, revealed that it acts as a phospholipid scramblase involved in endosome biogenesis (3265). Any1 was shown to facilitate the bidirectional movement of phospholipids across the bilayer and thereby modulate membrane asymmetry. This activity is crucial for the proper formation of intraluminal vesicles within multivesicular bodies (MVBs). Any1 works alongside Vps13 (3265), a lipid transfer protein, at endosome-ER contact sites (3266). This partnership ensures an adequate lipid supply for MVB biogenesis (3265). Any1 cycles between early endosomes and the trans-Golgi network, suggesting a role in membrane trafficking and organelle communication.

As Any1 in Saccharomyces cerevisiae is orthologous to human SLC66A2 and both belong to the PQ-loop protein family, it is plausible that SLC66A2 performs similar functions in human cells, contributing to endosomal trafficking and membrane homeostasis.

SLC66A3 - Orphan transporter: PQLC3 (SLC66A3) is also ubiquitously expressed, as suggested by the HPA.

CTNS (SLC66A4): CTNS (SLC66A4), also known as cystinosin, is an H⁺-driven lysosomal cystine transporter. It is ubiquitously expressed (3267). Cystinosis is a rare autosomal recessive disorder discovered in 1998 (3255) that involves lysosomal storage of the amino acid cystine and is caused by defects in cystinosin. Cystinosin deficiency leads to intra-lysosomal cystine accumulation and crystal formation in the acidic lysosome of almost all cells and tissues including cornea, conjunctiva, kidney, liver, spleen, muscle, brain, thyroid, intestine, rectal mucosa, lymph nodes, macrophages and bone marrow (3268). Clinical symptoms are variable and renal involvement is characteristic and occurs early. In children, cystinosis is the most common inherited cause of renal Fanconi syndrome, a functional defect of renal proximal tubule cells, impairing absorption of electrolytes, carbohydrates, amino acids and other substances.

A subsequent study elucidates a critical link between lysosomal cystine accumulation and aberrant mTORC1 signaling in kidney proximal tubule epithelial cells, highlighting a novel pathogenic mechanism in cystinosis (2657). The specific findings are as follows:

• 1)Mutations in the CTNS gene lead to cystine accumulation within the lysosomes of proximal tubule epithelial cells in the kidney, stimulating the Ragulator-Rag GTPase complex.
• 2)This results in constitutive activation of mTORC1 on the lysosomal surface, shifting cellular programs from catabolic to anabolic and promoting growth and proliferation over differentiation.
• 3)This, in turn, disrupts proximal tubule epithelial cell function because altered signaling impairs differentiation and function, reducing reabsorptive capacity and contributing to renal manifestations of cystinosis.
• 4)Pharmacological inhibition of mTORC1 using low doses of rapamycin restored lysosomal proteolysis and redirected proximal tubule cells toward proper differentiation and homeostasis.

These findings suggest that cystine acts as a lysosomal fasting signal that modulates mTORC1 signaling to direct cellular fate decisions in the renal proximal tubule epithelium (2657). In addition, this work identifies mTORC1 as a viable therapeutic target and suggests that modulation of this pathway could ameliorate renal dysfunction in affected individuals.

MPDU1 (SLC66A5) - Orphan transporter: MPDU1 (SLC66A5) is ubiquitously expressed. A genetic defect in MPDU1 has been shown to be the cause of a congenital disorder of glycosylation (3269). The mutation affects the use of donor substrates for lipid-linked oligosaccharides. The hamster homolog, Lec35, has been implicated in the proper presentation or “flipping” of dolichol phosphate-sugars in the ER membrane.

A subsequent study provided evidence linking a novel homozygous missense variant in the MPDU1 gene to a severe clinical phenotype resembling ciliopathies (Darouich, 2023, #4820). The affected female infant exhibited a range of symptoms characteristic of ciliopathies. The authors conclude that MPDU1 plays a key role in the utilization of mannose-phosphate-dolichol, which is essential for glycosylation processes, and that defects in glycosylation can disrupt the function of ciliary proteins, resulting in ciliopathy-like manifestations.

SLC66A1LP - Pseudogene: Originally named SLC66A1L based on homology to SLC66A1, SLC66A1LP is now classed as a transcribed unitary pseudogene in human. Coding versions of this gene are present in most primates, chicken, Xenopus and fish but these remain uncharacterized.

Orphan transporter family members (3)

SLC66A2 (PQLC1), SLC66A3 (PQLC3), MPDU1 (SLC66A5)

SLC67 Organic cation transporter-like family (2.A.1.2.53/PF07690/MFS)

Discovery: A transcript encoding a putative membrane transporter called organic cation transporter-like 2 (ORCTL2) was identified in an effort to identify genes at 11p15.5 that contribute to the etiology of Beckwith-Wiedemann syndrome (3270). ORCTL2 is homologous to a family of proteins studied as drug efflux pumps in bacteria, such as the multidrug resistance protein of Bacillus subtilis and the tetracycline transporter of E. coli. The homology of ORCTL2 to bacterial polyspecific cation transporters and its high expression in kidney, liver and intestine suggest that it might play a role in cation transport in these organs.

Gene family members (2):
SLC67A1 (ORCTL2)	SLC67A2 (MFSD9)

Molecular aspects, physiological roles and links to disease

The SLC67 family belongs the Drug:H⁺ Antiporter-1 (12 Spanner) (DHA1) family (TC 2.A.1.2), which is part of the MFS superfamily. While most members of the MFS superfamily have 12 TMHs, SLC67A1 (ORCTL2) encodes a membrane protein with 10 putative TMHs (3271).

SLC67A1 (formerly SLC22A18) - Orphan transporter: ORCTL2 (SLC67A1) is a putative organic cation transporter-like transporter (ORCTL). While it is widely expressed at moderate levels according to the HPA, its most prominent expression is in the duodenum.

Studies of the transport properties of ORCTL2 (SLC67A1) revealed that this protein confers resistance to chloroquine and quinidine when overexpressed in bacteria (3272). Immunohistochemistry showed that ORCTL2 localizes to the apical membrane of the proximal tubules (3272). The results suggest that ORCTL2 plays a role in the transport of chloroquine and quinidine-related compounds in the kidney. On the contrary, Western blot analysis in a subsequent study showed that ORCTL2 (SLC67A1) is predominantly expressed at intracellular organelle membranes (3271).

ORCTL2 (SLC67A1) has been found to be overexpressed in non-small cell lung cancer (NSCLC) (3273). Moreover, it has been shown to function as a tumor suppressor in colorectal cancer. Its gene is located on chromosome 11p15.5, a region of the chromosome that is frequently deleted. It also has frequent missense mutations in a variety of cancers. Its expression is negatively regulated by KRAS (3274).

In addition, the expression of ORCTL2 (SLC67A1) has been shown to regulate oxaliplatin resistance by modulating the ERK signaling pathway in colorectal cancer (3275). It shares homology with tetracycline resistance proteins and bacterial multidrug resistance proteins (3272).

SLC67A2 – Orphan transporter: MFSD9 (SLC67A2) has been reported to be a transporter expressed in food regulatory brain areas where it is located to neurons in mouse brain, and its mRNA expression was affected by diet (3142). MFSD9 is widely expressed according to the HPA. The transport function of this protein is unknown.

Orphan transporter family members (2)

SLC67A1 (ORCTL2), SLC67A2 (MFSD9)

HGNC updates

SLC22A18 has been renamed SLC67A1 based on revised phylogenetic analysis placing it in a new family (SLC67). SLC67A2 is a new symbol for MFSD9.

SLC68 Cation symporter family (2.A.2/MFS_2/MFS)

Discovery: MFSD13A is one of the 29 atypical SLCs of the Major Facilitator Superfamily (MFS) (174) and was previously named TMEM180, reflecting its status as a poorly characterized transmembrane protein. MFSD13B (SLC68A2P) is a unitary pseudogene in humans but has coding orthologs in other vertebrates.

Gene family member (2):
SLC68A1 (MFSD13A/TMEM180)	SLC68A2P (MFSD13B)

Molecular aspects, physiological roles and links to disease

The SLC68 family belongs to the Glycoside-Pentoside-Hexuronide (GPH):Cation Symporter family (TC 2.A.2), which is part of the MFS superfamily. MFSD13A (SLC68A1/TMEM180) has 12 TMHs, consistent with the MFS structural fold (3276).

SLC68A1 – Orphan transporter: As noted above, MFSD13A/TMEM180 (SLC68A1) belongs to the cation symporter family, whose members are H⁺ or Na⁺ sugar symporters. The protein has the putative cation-binding site conserved among orthologs, with a position similar to that of the Na⁺-coupled melibiose transporter MelB, and it was therefore proposed that MFSD13A/TMEM180 acts as a cation-coupled symporter (174, 3276). According to the HPA, MFSD13A (SLC68A1/TMEM180) is highly expressed in different areas of the brain (especially in excitatory neurons), in the retina (rod photoreceptor cells and horizontal cells) and in the lungs (alveolar type 2 cells), while at lower levels it is relatively widely expressed.

SLC68A1 is highly expressed in colorectal cancer (CRC) (3277) and is associated with poor survival in stage III CRC (3278). Moreover, it has been reported to be a risk gene for schizophrenia, possibly through effects on neurodevelopment and schizophrenia-associated pathological pathways (3279).

SLC68A2P - Pseudogene: This is a unitary pseudogene in human. Coding versions of this gene are present in other vertebrates but these remain uncharacterized.

Orphan transporter family member (1)

MFSD13A (SLC68A1/TMEM180)

HGNC updates

The SLC68 family has been created to accommodate this new SLC assignment, and MFSD13A has been updated to SLC68A1 and the unitary pseudogene MFSD13B updated to SLC68A2P.

SLC69 Vitamin A receptor/transporter family (2.A.90/RBP_receptor/STRA6)

Discovery: Stra6 (stimulated by retinoic acid 6) was identified as a retinoic acid-inducible gene in mouse P19 embryonal carcinoma cells using subtractive hybridization cDNA cloning (3280). STRA6 is a retinol-binding protein (RBP) with nine predicted TMHs. At the time of cloning, it showed no sequence similarity to any known transporter, channel, or receptor.

Gene family member (1)

STRA6 (SLC69A1)

Molecular aspects, physiological roles and links to disease

The SLC69 family belongs to the Vitamin A Receptor/Transporter (STRA6) Family (TC 2.A.90). STRA6 has 9 TMHs (3280). The cryo-EM structure of a zebrafish ortholog has been reported (36) (see below).

STRA6 (SLC69A1): STRA6 (SLC69A1) is a vitamin A transporter that facilitates the cellular uptake and exit of retinol (36, 3281). The homeostasis of vitamin A is critical for normal cellular function, vision, immune function, and overall growth and development. Retinoids, particularly retinoic acid, play key roles in cellular proliferation and differentiation, as well as in development. Vitamin A is also crucial for vision as it is required for the visual cycle, converting all-trans-retinal back into 11-cis-retinal, thereby regenerating visual pigments like rhodopsin. The hydrophobic retinol is the major form in which vitamin A is transported, and RBP is the only specific carrier in the bloodstream for retinol. STRA6 mediates the cellular uptake of vitamin A in certain tissues by recognizing RBP-retinol to trigger the release and internalization of retinol.

As described in the Discovery section, mouse STRA6 was identified as a retinol-binding, multi-pass membrane receptor predicted to contain 9 TMHs. It displays a specific pattern of expression during development and in adult tissues. In the mouse testis, STRA6 expression in Sertoli cells has been shown to vary with the spermatogenic cycle. Immunohistochemical analysis also demonstrated strong expression of STRA6 in the mouse eye and periocular mesenchyme during development. Additionally, Stra6 was detected in multiple cell types forming distinct blood-organ barriers. A subsequent study in bovine tissues reported that STRA6 is highly expressed in the eye, localizing predominantly to the basolateral membrane of retinal pigment epithelial (RPE) cells (3280). Expression in female reproductive organs and the placenta has also been reported (3280, 3282).

According to the HPA, STRA6 expression in human tissues exhibits notable differences compared to bovine and murine models. In humans, STRA6 is most prominently expressed in female reproductive tissues, including endometrial stromal cells of the cervix and syncytiotrophoblasts in the placenta. Moderate expression is observed in the lung (bronchiolar epithelial cells, basal respiratory cells, and glandular epithelial cells), kidney (proximal tubule cells), and testis (Leydig cells).

A striking discrepancy highlighted by the HPA is the apparent absence of STRA6 expression in the human and porcine retina, despite strong expression in the mouse retina and in bovine RPE cells, where it is localized to the basolateral membrane (3280). If the HPA data accurately reflect true biological expression patterns, this would suggest that humans rely on an alternative mechanism for vitamin A uptake in the RPE, possibly involving an as yet unidentified transporter.

STRA6 (SLC69A1) is likely to play an important developmental role in the human visual system (3282), as evidenced by its link to Matthew-Wood syndrome (MWS), a severe congenital disorder caused by mutations in the human STRA6 gene (3283, 3284). MWS is characterized by bilateral ocular malformations at birth, including microphthalmia (abnormally small eyes), anophthalmia (complete absence of the eyes), and coloboma (defects in ocular tissue development). These abnormalities are frequently accompanied by pulmonary hypoplasia and congenital heart defects. However, the ocular phenotypes in MWS do not typically exhibit hallmarks of RPE dysfunction, such as progressive retinal degeneration or classical signs of visual cycle impairment. Instead, the visual impairment, often profound or complete from birth, stems from developmental malformations, such as anophthalmia, microphthalmia, or coloboma. Given the high perinatal mortality and limited clinical follow-up in surviving patients, it remains unclear whether any additional postnatal retinal degeneration occurs. Thus, the current evidence suggests that STRA6 deficiency primarily disrupts morphogenesis, including ocular morphogenesis. Accordingly, STRA6 function appears to be particularly critical during prenatal development in humans.

Interestingly, according to the HPA, STRA6 is not expressed in the intestine or liver, despite the central role these organs play in vitamin A absorption, storage, and metabolism. This raises the question of how vitamin A is transported across these barriers and highlights the need to better understand where STRA6 may fit into the overall vitamin A transport network.

In the intestine, dietary vitamin A is obtained primarily as retinyl esters from animal sources and β-carotene from plant sources. β-carotene is believed to be absorbed by enterocytes from mixed micelles, potentially via the long-chain fatty acid translocase CD36 and/or its paralog SCARB1 (also known as SR-B1) (2241, 3285, 3286). Once internalized, β-carotene is enzymatically cleaved by β-carotene oxygenase 1 into retinal, which is then reduced to retinol. Retinyl esters, on the other hand, are hydrolyzed in the intestinal lumen by pancreatic and brush-border lipases, liberating retinol that can enter enterocytes, likely by passive diffusion. There is ongoing speculation that carotenoids may also be taken up via NPC1L1 (SLC65A2), the cholesterol transporter that also transports vitamin E, but this has yet to be conclusively demonstrated (3287, 3288).

Once inside enterocytes, retinol is re-esterified by acyltransferases and incorporated into chylomicrons along with dietary lipids (3289). These chylomicrons are then secreted into the lymphatic circulation, enter the bloodstream, and are eventually cleared by the liver via receptor-mediated uptake, primarily involving LDLR (low-density lipoprotein receptor). Thus, based on current knowledge, STRA6 does not appear to play a role in the intestinal absorption of vitamin A or in its direct uptake by hepatocytes.

In the liver, vitamin A is stored predominantly as retinyl esters in hepatic stellate cells. When needed, it is mobilized as retinol bound to retinol-binding protein 4 (RBP4) for systemic distribution. Despite the liver’s crucial role in retinoid storage and mobilization, STRA6 expression is absent in human hepatic tissue (3290). In mice, this apparent gap is filled by STRA6L (also known as Rbpr2), a structurally related RBP4-binding protein involved in hepatic vitamin A uptake (3291). However, the human ortholog of Stra6l, called STRA6LP, is a pseudogene. STRA6L appears to provide a species-specific, liver-based RBP4-retinol uptake pathway in mice, likely compensating for reduced reliance on chylomicron remnant uptake in this species and others (3292–3294). Thus, the loss of STRA6L in humans may reflect a shift toward more efficient hepatic clearance of chylomicron remnants.

In humans, retinol reenters the circulation from hepatic stellate cells bound to RBP4, destined for uptake in STRA6-expressing extrahepatic tissues. However, the physiological route through which vitamin A exits hepatocytes and enters the bloodstream remains uncharacterized, and there is little evidence implicating STRA6 in this process.

In the setting of liver injury and fibrosis, STRA6 expression is upregulated in hepatic stellate cells (3295). This upregulation appears to facilitate retinol export into the extracellular space, where it is subsequently taken up by hepatocytes, despite the absence of STRA6 expression in hepatocytes, suggesting an alternative uptake mechanism. Once inside hepatocytes, retinol promotes lipogenesis, thereby contributing to fibrotic progression. Blocking STRA6 disrupts this pathological retinoid-mediated cross-talk. Therefore, targeting STRA6 in liver disease may help prevent excessive hepatic fat accumulation and mitigate fibrosis.

Vitamin A is essential for fetal development, particularly for eye formation, but also for the development of the lung and heart, and thus STRA6 is believed to play a key role in maternal-fetal transfer of vitamin A transfer via the placenta (3280, 3282, 3296). This is consistent with the phenotypic spectrum of MWS, where STRA6 mutations impair retinoid delivery to the developing fetus. However, specific localization, e.g., to the apical membrane of human syncytiotrophoblasts, has not been conclusively demonstrated.

A study on the role of STRA6 in human cardiac development highlights the critical developmental importance of STRA6 (3297). Using human embryonic stem cell models, STRA6 was shown to be essential for the retinoic acid-dependent induction of vascular smooth muscle cells (VSMCs) from cardiac neural crest cells, a process essential for proper outflow tract development. Loss of STRA6 function has been shown to disrupt retinoic acid signaling, leading to failure of smooth muscle differentiation and contributing to congenital heart defects in patients with MWS due to a STRA6 mutation. Heart defects occur in approximately half of MWS cases, and their severity can vary. Interestingly, this contrasts with findings in Stra6-deficient mice, which do not exhibit obvious cardiac malformations. This confirms the species-specific role of STRA6 in human heart development (3298–3300) and supports the growing evidence that the tissue distribution and functional roles of STRA6 differ between humans and mice.

A study on the role of STRA6 in human heart development highlights a critical developmental role for STRA6 (3297). Human embryonic stem cell models were used to show that STRA6 is required for the retinoic acid-dependent induction of vascular smooth muscle cells (VSMCs) from cardiac neural crest cells as part of a process vital for proper outflow tract development. Loss of STRA6 function was shown to disrupts retinoic acid signaling, leading to failure in smooth muscle differentiation and potentially contributing to congenital heart defects seen in patients with MWS caused by STRA6 mutation. Heart defects occur in approximately half of MWS cases, and their severity can vary. Interestingly, this contrasts with findings in Stra6-deficient mice, which do not display obvious cardiac malformations. This confirms that there is a species-specific role of STRA6 in human heart development (3298–3300). This adds to growing evidence that tissue distribution and functional roles of STRA6 differ between humans and mice.

In the kidney, STRA6 is expressed at low levels in proximal tubule cells, where it may contribute to renal vitamin A reabsorption. While its precise subcellular localization in these cells remains unclear, one hypothesis is that STRA6 is present on the basolateral membrane, where it could facilitate the efflux of reabsorbed retinol back into the bloodstream, thereby contributing to systemic vitamin A homeostasis.

The cryo-EM structure of the zebrafish stra6 ortholog was used to construct a model of the retinol transport process. According to this model, retinol is released from the RBP into a lipid-filled outer cleft and subsequently transported into the membrane by lateral diffusion through a side-facing window (36). CaM has been shown to be tightly bound to a sequence motif of Stra6 at the intracellular C terminus of zebrafish, which consists of a short α-helical stretch with conserved hydrophobic anchor residues. The motif is conserved among all vertebrates and has a corresponding region in human STRA6 from residues 635 to 640. A functionally important arginine residue within this motif is also conserved and corresponds to R638 in human STRA6. These results suggest a common CaM-dependent regulatory mechanism.

STRA6 has been shown to facilitate both the uptake and exit of retinol in cells (3281). Furthermore, the activity and transport direction of STRA6 via the calcium/calmodulin (Ca²⁺ /CaM) complex have been shown to be controlled by intracellular Ca²⁺ levels, which in turn are regulated by extracellular signals or other mechanisms (3281). Elevated Ca²⁺ /CaM levels have been shown to increase apo-RBP binding to the extracellular portion of STRA6 to facilitate vitamin A efflux from the cell while inhibiting vitamin A uptake. It has been demonstrated that CaM interaction with the CaM-binding site is Ca²⁺-dependent and that the Ca²⁺/CaM complex induces conformational changes that are essential for STRA6 function.

In both mice and cattle, the STRA6 protein is required for the uptake of retinol into the retinal pigment RPE, which plays a role as a precursor in the synthesis of the chromophore 11-cis-retinal (3301). This chromophore then binds to opsins to form functional pigments important for vision. Stra6 knockout mice exhibit significantly reduced retinoid levels in the eye, which cannot be compensated for by alternative vitamin A uptake pathways (3301). As a result, loss of STRA6 function leads to an imbalance between opsins and chromophores. This results in impaired visual function, particularly of the cones, with progressive retinal degeneration. Therefore, STRA6 maintains photoreceptor health and the correct number of visual pigments in mice and cattle. It remains to be clarified whether STRA6 is also expressed in the human RPE or whether another, as yet unidentified transporter fulfills this role in humans.

Orphan transporter family members: N/A

HGNC update

The SLC69 family has been created to accommodate this new SLC assignment, and SLC69A1 was added as a new alias of STRA6.

SLC70 Cyclin M Mg²⁺ exporter family (1.A.112/CNNM/CNNM)

Discovery: The cellular Mg²⁺ homeostasis of prokaryotes is controlled by the Mg²⁺ transporters CorA, MgtA/B, MgtE and CorB/C. As mentioned in the SLC41 family description, the selectivity pore motifs for CorA, MgtE and CorB/C are conserved across species. While the SLC41 proteins are the human orthologs of MgtE, the CNNM proteins are the human orthologs of CorB/C (3302). This indicates that the CNNM proteins are likely to also possess Mg²⁺ transport capacity, especially since the CorA and CorB/C proteins share the overall quaternary structure as well as the functional properties with their respective orthologs, while for SLC41 Na⁺/Mg²⁺ transporters, MgtE only shares the selectivity pore.

Gene family members (4):
CNNM1 (SLC70A1)	CNNM3 (SLC70A3)
CNNM2 (SLC70A2)	CNNM4 (SLC70A4)

Molecular aspects, physiological roles and links to disease

The SLC70 family belongs to the Cyclin M Mg²⁺ Exporter (CNNM) family (TC 1.A.112). Detailed studies of CNNM protein crystal structures have shown that these proteins form homodimers, with each protomer consisting of three TMHs (3303). These form the transmembrane domain, also called the DUF21 (domain of unknown function 21) domain. Structurally, CNNMs contain the N-terminal extracellular domain (ECD), the DUF21 transmembrane domain, and a large cytosolic region that contains both a cystathionine-β-synthase (CBS) domain and a putative cyclic nucleotide-binding homology (CNBH) domain (3304).

Mg²⁺ is a cofactor for many enzymes and is essential for protein synthesis, muscle and nerve function, energy production, oxidative phosphorylation, and DNA stability. The body maintains magnesium levels via tightly regulated systems involving intestinal absorption, renal excretion, bone metabolism, and the parathyroid gland (3305). Magnesium transporters from several SLC families, such as SLC41 (MgtE), SLC57 (NIPA), SLC58 (MAGT), and SLC70 (CNNM), along with the TRPM6 and TRPM7 magnesium channels, are important contributors that provide balanced Mg²⁺ influx and efflux to maintain physiological Mg²⁺ levels.

In general, CNNM proteins are presumed to have Mg²⁺ transport activity because they are homologous to prokaryotic Mg²⁺ transporters (3302). One exception is CNNM1, which has been reported to function as a cytosolic copper chaperone (3306). Consistent with Mg²⁺ transport function, the fish cnnm3 ortholog, which is approximately 30% similar to human CNNM3 (SLC70A3) in protein sequence, has been shown to facilitate Mg²⁺ efflux when expressed in Xenopus oocytes (3307).

Of the four members of the CNNM family (Fig. 58), CNNM4 is the most well-established as a bona fide Na⁺/Mg²⁺ exchanger. This is supported by functional assays in intestinal cells and structural modeling consistent with an alternating-access transport mechanism (40, 3303, 3308, 3309). In contrast, the precise nature of the transport activity of other CNNMs, particularly CNNM3, is debated. They may play a regulatory role rather than a transport function and may act as transceptors. Their transport function may be achieved by interacting with channels such as TRPM7 (3310, 3311). These distinctions may reflect functional specialization within the CNNM family (3303).

Fig. 58. Phylogenetic tree of the SLC70 family.

CNNM proteins contain cytoplasmic C-terminal CBS domains that form pairs of tandem CBS domains, also known as a Bateman module. These domains fold together into a single functional module (3303, 3312, 3313), which binds ATP in a Mg²⁺ -dependent manner, inducing structural changes that modulate transport activity. Thus, the CBS domain pair acts as a regulatory hub, binding Mg²⁺ -ATP and regulating Mg²⁺ transport and homeostasis (3314). Interestingly, only the CBS domain pairs of CNNM2 and CNNM4 bind to ATP in the presence of Mg²⁺; the pairs of CNNM1 and CNNM3 do not bind at significant levels (3303, 3313). Understanding these regulatory mechanisms sheds light on the pathophysiology of diseases associated with CNNM dysfunction (347, 3303, 3313).

CNNM1 (SLC70A1) – Orphan transporter: CNNM1 (SLC70A1), also known as ACDP1 (ancient conserved domain protein 1), is a relatively understudied transporter proposed to function as a cytosolic copper chaperone (3306). Based on studies in mice, CNNM1 is highly expressed in the brain and at moderate levels in the testes and kidneys (3315). According to the HPA, CNNM1 is highly expressed in the brain (specifically, in neurons), the retina (bipolar and horizontal cells), the testis (early and late spermatids), and at lower levels in endocrine tissues, the kidney, and female tissues. Based on studies of mouse testes, CNNM1 has been shown to function as a regulator of germ cell division and differentiation (3316).

CNNM1 is a protein associated with tumor development in prostate cancer and hepatocellular carcinoma. It acts as a cell cycle regulator and represents a potential therapeutic target (3317). CNNM1 regulates the cell cycle and proliferation as soon as the cholesterol-induced regulator of metabolism RNA (CHROMR) increases CNNM1 expression by adsorbing the microRNA miR-1299. This promotes the cycle progress of Diffuse large B-cell lymphoma (DLBCL) cells and makes them resistant to the monoclonal antibody medication rituximab (3317).

The mechanism by which CNNM1 induces rituximab resistance remains unknown, as does its proposed function as a copper chaperone. Further research is necessary to address these issues. Additionally, the potential functional roles of CNNM1 in specific brain areas, including the retina, as suggested by the HPA, are currently unknown.

CNNM2 (SLC70A2) – Orphan transporter: CNNM2 (SLC70A2), also known as ACDP2, is considered essential for renal Mg²⁺ reabsorption (3318–3320).

Cnnm2 was cloned from mouse distal convoluted tubule cells and characterized in Xenopus laevis oocytes using two-electrode voltage clamping (3319). When expressed in oocytes, Mg²⁺ triggered large currents that were saturable with a Michaelis-Menten K_m value of 0.56 mM. However, subsequent electrophysiology experiments using human CNNM2 expressed in HEK293 cells produced different results (Stuiver, 2011, #4877). Rather than mediating Mg²⁺ currents, CNNM2 was found to generate Mg²⁺ -sensitive Na⁺ currents. Further research is required to clarify the functional properties of CNNM2 and its role in basolateral cation exit in kidney distal tubules.

According to the HPA, CNNM2 (SLC70A2) is present at high levels in the choroid plexus and in neurons and glial cells throughout the brain. In addition, CNNM2 is expressed at moderate levels in the kidney, gastrointestinal tract, and placenta. In the human kidney, CNNM2 has been localized to the basolateral membrane of the thick ascending limb and the distal tubule, which are the major nephron segments involved in Mg²⁺ reabsorption (3320).

Mutations in CNNM2 have been identified as the cause of dominant hypomagnesemia in two unrelated families (3320). Affected individuals exhibited markedly reduced serum Mg²⁺ levels but maintained inappropriately normal urinary Mg²⁺ excretion, indicating a defect in renal tubular reabsorption, as the kidneys failed to conserve magnesium despite systemic deficiency (3320). In addition, CNNM2 mutations have been associated with impaired brain development, seizures, intellectual disability syndrome, and brain malformations (3321, 3322). It has been concluded that CNNM2 is essential for brain development, neurological function, and Mg²⁺ homeostasis.

To further assess the role of CNNM2 in Mg²⁺ homeostasis, Cnnm2 knockout mice were generated (3318). Only four Cnnm2^-/- pups were born alive, and they had significantly lower serum Mg²⁺ concentrations than their wild-type littermates. Adult Cnnm2^+/- mice exhibited mild hypomagnesemia and increased serum calcium levels, independent of dietary Mg²⁺ intake, compared to wild-type mice. Based on these studies, it has been concluded that CNNM2 is essential for embryonic development and Mg²⁺ homeostasis in mice as well.

The brain is highly dependent on proper magnesium homeostasis, as reduced magnesium levels have been associated with neurological conditions such as migraine, depression, and epilepsy (3323). Magnesium can enter the brain across both the BBB and the blood-cerebrospinal fluid (CSF) barrier. CNNM2 is also expressed in the choroid plexus, where it may contribute to the previously described mechanism for magnesium transfer across the choroid plexus epithelium, which actively maintains CSF Mg²⁺ concentrations above plasma levels in species such as sheep, dog, and human (3324–3327).

Fig. 11 shows CNNM2 hypothetically localized to the apical membrane of choroid plexus epithelial cells, functioning as a Mg²⁺ exporter in exchange for Na⁺. This hypothesis is based on the observation that CNNM2 is the closest paralog of CNNM4 (SLC70A4; see the phylogenetic tree in Fig. 58), the transport mechanism of which is well established (see below). However, the proposed localization CNNM2 in the choroid plexus and its functional properties still require experimental investigation.

Similar to CNNM4 (see below), the proper folding and dimerization of the extracellular domain of CNNM2 is necessary for its functional localization in the plasma membrane. This domain undergoes N-glycosylation at N112, and the glycosylation is necessary for CNNM2 stability in the plasma membrane (3328). Thus, this domain is important for the biological function of CNNM2, and disease-causing mutations that result in hypomagnesemia occur at this site (3321).

CNNM3 (SLC70A3) – Orphan transporter: CNNM3 (SLC70A3) is widely expressed, with the strongest expression occurring in muscle tissue, the gastrointestinal tract, exocrine tissues, the liver, the brain, and the retina of the eye, according the HPA. CNNM3 has been reported to be involved in a special regulatory circuit, through which it modulates Mg²⁺ influx via the ubiquitously expressed TRPM7 channel (3310). CNNM3 has been shown to directly interact with TRPM7, thereby suppressing its ion channel activity and limiting Mg²⁺ entry into cells. This interaction is dynamically regulated by the following two opposing mechanisms:

• 1)The small GTPase ARL15 stabilizes the CNNM3-TRPM7 complex. This results in further inhibition of TRPM7 activity.
• 2)The phosphatases PRL-1 and PRL-2 disrupt the CNNM3-TRPM7 interaction. This reactivates TRPM7 and promotes Mg²⁺ uptake.

The balance of these two systems is sensitive to the concentration of Mg²⁺ inside cells and PRL expression levels. This allows for precise regulation of magnesium levels in the body in response to changes in physiological needs. This kind of dynamic modulation is essential for sustaining cellular bioenergetics and metabolic adaptability.

The discoveries reveal that CNNM3 acts as a regulator of Mg²⁺ -dependent signaling pathways rather than as an Mg²⁺ transporter. The findings have important implications for renal and neurological disorders, metabolic diseases, and cancer biology (3310).

CNNM4 (SLC70A4): CNNM4 (SLC70A4) is highly expressed in the gastrointestinal tract, eyes, brain, muscle tissue, and bone marrow and at lower levels in the lungs, kidneys, liver, skin, pancreas, and reproductive organs, according to the HPA. Studies in mice have shown that Cnnm4 plays an important role in basolateral Mg²⁺ extrusion in the intestine (3308). Additionally, an electroneutral Na⁺/Mg²⁺ exchanger activity has been demonstrated for CNNM4 (SLC70A4) (3308). Corresponding structural work hypothesizes an alternating-access transport mechanism (40, 3303, 3309). Thus, following apical uptake of Mg²⁺ via the Trpm6 Mg²⁺ channel, CNNM4 expressed on the basolateral membrane facilitates Mg²⁺ exit in exchange for Na⁺ (3308, 3329). Consistent with this, Cnnm4 knockout mice exhibited hypomagnesemia due to an inability to absorb magnesium properly, which highlights the role of the protein in Mg²⁺ extrusion.

A rare genetic disorder known as Jalili syndrome is caused by pathogenic variants in human CNNM4 (SLC70A4). The syndrome is characterized by a combination of cone-rod dystrophy and amelogenesis imperfecta (3330). Functional and pathogenic insights into CNNM4 variants associated with Jalili syndrome have been provided (3331). Two disease-causing missense mutations, G492C and G492D, were identified in the CBS domain of CNNM4. Although these mutant CNNM4 proteins maintained correct membrane localization and preserved their ability to bind Mg²⁺, they exhibited markedly reduced magnesium extrusion activity. This indicates that the primary defect lies in protein function rather than trafficking (3331). These pathogenic mutations likely cause a loss of function through structural destabilization, resulting in the retinal and dental abnormalities characteristic of Jalili syndrome. The findings provide a basis for future therapeutic strategies aimed at restoring Mg²⁺ homeostasis in affected tissues.

The importance of dimerization of the extracellular domain (ECD) of CNNM4, which forms an immunoglobulin-like fold dimer with three N-glycosylation sites, has also been reported (3332). It has been shown that the mutations in the extracellular domain of human CNNM4 prevent its dimerization. The ECD domain was found to be essential for proper dimerization and transport function not only in CNNM4 but also in CNNM2 (see above) (3332). It enables proper assembly and activity of the full-length transporter.

Mg²⁺ plays an important role in regulating energy metabolism and tumor progression (3333). The majority of the Mg²⁺ inside cells is bound to other molecules, especially ATP (3334). An increase in intracellular Mg²⁺ levels through overexpression of PRL-3 or CNNM4 knockdown in cultured cells leads to an increase in the intracellular ATP level to an extent similar to that of Mg²⁺ (3335, 3336). PRL-3 phosphatase is frequently overexpressed in malignant human cancers (3337). It binds to CNNMs at the CBS domain (3338). Elevated ATP levels driven by PRL-3 upregulation are beneficial for tumor growth and metastasis (3337, 3339, 3340). Under these conditions, AMP-activated protein kinase (AMPK), an energy sensor, is deactivated, thus resulting in the abolishment of the suppression of the mTOR pathway by AMPK. The result is uncontrolled activation of the mTOR pathway, which promotes tumor progression. Conversely, mTOR inhibition by rapamycin reverses tumorigenicity (3333). Disrupting the PRL-CNNM interaction prevents the elevation of intracellular ATP levels, thereby suppressing tumor growth and invasiveness (3333). These findings make PRL-3 an attractive therapeutic target for developing novel anti-cancer agents (3337).

Orphan transporter family members (3)

CNNM1 (SLC70A1), CNNM2 (SLC70A2), CNNM3 (SLC70A3)

HGNC updates

The SLC70 family has been created to accommodate these new SLC assignments. SLC70A1, SLC70A2 and SLC70A3 are new aliases for CNNM1, CNNM2 and CNNM3 respectively.

SLC71 putative ammonia transporter family (2.A.1.2/MFS_1/MFS)

Discovery: The genes encoding these proteins were identified as part of a large cDNA sequencing project and initially named hippocampus abundant transcript 1, HIAT1, and hippocampus abundant transcript like 1, HIATL1 (3341, 3342) and later renamed MFSD14A (SLC71A1) and MFSD14B (SLC71A2), respectively. They were first suggested to be involved in nutrient transport (3343) but later reported to play an important role in ammonia transport (3344).

Gene family members (2 + 1 pseudogene):
SLC71A1 (MFSD14A)	SLC71A2 (MFSD14B)	SLC71A3P (MFSD14CP)

Molecular aspects, physiological roles and links to disease

The SLC71 family belongs to the Drug:H⁺ Antiporter-1 (12 Spanner) (DHA1) family (TC 2.A.1), which is part of the MFS superfamily. MFSD14A and MFSD14B have been reported to have 12 putative TMHs (3343), which is a common feature for most MFS members.

SLC71A1 and SLC71A2 - Orphan transporters with new indication for ammonia transport function

MFSD14A/HIAT1 (SLC71A1) and MFSD14B/HIATL1 (SLC71A2) were shown to be expressed in neurons throughout the adult mouse brain (3343). They share an amino acid sequence identity of about 70%. MFSD14A co-localized with the Golgi marker giantin, whereas MFSD14B staining co-localized with a marker for endoplasmic reticulum retention. As a member of the major facilitator superfamily, mammalian MFSD14A was predicted to be a novel sugar transporter due to the presence of the sugar transporter-specific motif D-R/K-X-G-R-R/K between TMH2 and TMH3 (3341). Due to the presence of this sugar transporter motif, the MFSD14 transporters were predicted to have organic substrate profiles and potentially be involved in energy homeostasis based on phylogenetic clustering analyses. Consistent with this, both transporters were temporarily upregulated after amino acid starvation in primary cortical cells subjected to amino acid starvation (3343).

Interestingly, mouse MFSD14A (Slc71a1) was found to be involved in spermatogenesis, as lack of the transporter resulted in failed acrosome formation, sperm head condensation, and faulty mitochondrial localization (3345). MFSD14A expression was localized to the supporting Sertoli cells, where it was proposed to be responsible for sugar uptake (e.g., mannose), important for glycosylation of key molecules required for acrosome formation. While the knockdown of Slc71a1 did not directly cause the death of the animals, the males were infertile, making this transporter essential for reproduction and thus the survival of the species.

Surprisingly, however, MFSD14A has so far not been shown to promote glucose transport (3346), although other monosaccharides that may be involved in glycosylation or energy/carbohydrate metabolism have not yet been tested. In contrast, green crab mfsd14a (Hiat1), which shows 78% amino acid sequence identity to human MFSD14A (SLC71A1), was subsequently shown to contribute to ammonia transport. Specifically, it has been suggested that it mediates a Na⁺-coupled secondary active NH₄⁺ transport against concentration gradients, thereby providing cellular ammonia detoxification (3344). This may be important in ammonia-transporting epithelia, such as nephrons and gills, but also more generally for fertility, considering the sperm malformation in Slc71a1 knockdown mice. Since ammonia is an important acid-base equivalent (i.e., NH₃ as base and NH₄⁺ as acid component), MFSD14A (SLC71A1) may play an important role in cellular and systemic pH regulation and thus be essential for general physiological processes and homeostasis.

Also of note, genome-wide interaction analyses revealed an association between genetic variants of the HIATL1/MFSD14A gene SLC71A1, alcohol consumption, and the risk of developing CRC (3347).

MFSD14B was found to be significantly overexpressed in CRC cancer tissues compared to normal colon tissues, and alcohol further modifies the effects of the MFSD14B transporter on CRC risk through its influence on SLC71A2 gene expression. Interestingly, it has been shown that relatively high levels of ammonia accumulate in CRC tumors that inhibit T-cell growth and response to immunotherapy (3348). Ammonia (e.g., produced by the microbiota) accumulates in CRC tumors, likely due to loss of the ability to detoxify ammonia. Ammonia accumulation is also likely to explain resistance to other cancers, especially since serum ammonia levels are generally elevated in CRC patients (3348). Improving ammonia clearance has been shown to reactivate T cells, reduce tumor growth and prolong survival (3348). Therefore, it seems tempting to predict that these findings provide a potentially novel approach to improve the efficacy of immunotherapies by targeting the expression of the putative ammonia transporter MFSD14B (SLC71A2) to minimize ammonia accumulation in CRC.

SLC71A3P – Pseudogene: SLC71A3P, previously named as MFSD14CP, was originally thought to be protein coding and known as MFSD14C but is now considered to be an unprocessed transcribed pseudogene.

Orphan transporter family members (2)

SLC71A1 (MFSD14A), SLC71A2 (MFSD14B)

HGNC updates

The SLC71 family has been created to accommodate these new SLC assignments. SLC71A1, SLC71A2 and SLC71A3P are new symbols for MFSD14A, MFSD14B and MFSD14CP respectively.

SLC72 lysosomal solute carrier family (MFSD1) (2.A.1.53/MFS_1/MFS)

Discovery: MFSD1 has been identified as a putative SLC solute carrier that is affected by altered nutrient intake (2455).

Gene family member (1)

MFSD1 (SLC72A1)

Molecular aspects, physiological roles and links to disease

The SLC72 family belongs to the Proteobacterial Intraphagosomal Amino Acid Transporter (Pht) family (TC 2.A.1.53), which is part of the MFS superfamily. Based on homology modelling, MFSD1 (SLC72A1) is predicted to have 12 TMHs, a common feature for MFS transporters (2455). The cryo-EM structure of the dipeptide-bound MFSD1 in complex with the glycosylated lysosomal membrane protein GLMP has been reported (3349) (see below).

MFSD1 (SLC72A1): MFSD1 (SLC72A1), also known as SMAP4, is a lysosomal dipeptide transporter that exports lysine, arginine, or histidine-containing dipeptides with a net positive charge from the lysosomal lumen into the cytosol (3350).

MFSD1 (SLC72A1) is ubiquitously expressed (1475, 3351), with particular strong expression in monocytes and macrophages (including Kupffer cells and Hofbauer cells) according to the HPA. At the subcellular level, it was localized to lysosomes, due to a dileucine-based sorting motif in its cytosolic N-terminus (3351).

MFSD1-deficient mice have been shown to develop severe liver disease. The disease is characterized by extravasation of red blood cells, damage to the sinusoids, loss of endothelial cells in the liver sinusoids, and signs of fibrosis (3351).

The MFSD1 (SLC72A1) transport occurs together with its accessory subunit GLMP (179, 3349–3352). The cryo-EM structure of the dipeptide-bound MFSD1-GLMP complex in outward-open conformation characterized the heterodimer interface. In combination with molecular dynamics simulations, the structural basis for the selectivity of MFSD1 towards diverse dipeptides was established (3349). Unlike most lysosomal transmembrane proteins, MFSD1 is not N-glycosylated (3351) but forms a heterodimeric complex with the glycosylated lysosomal membrane protein GLMP. GLMP is a single-pass type I transmembrane protein with a large lysosomal luminal highly N-glycosylated N-terminus. GLMP is thought to have a chaperone function, with a role in protecting against lysosomal proteases and a role in the transport of the complex from the Golgi to the lysosomes (3349, 3351, 3353).

A follow-up study analyzed structure-disrupting non-synonymous SNPs in MFSD1. The findings reveal that these SNPs are deleterious due to their detrimental effects on the stability, conformation, and functionality of the MFSD1 protein (3354). The identified variants affect both the native conformation and the association of MFSD1 with GLMP.

MFSD1 shows a relatively high sequence similarity to members of the proteobacterial intraphagosomal amino acid transporter (Pht) family, transporters found in Legionella pneumophila transporting valine and threonine (3355).

Orphan transporter family members: N/A

HGNC update

The SLC72 family was created to accommodate this new SLC assignment in a separate family after thorough phylogenetic analysis; SLC72A1 is a new alias for MFSD1.

SLC73 Orphan MFSD6 transporter family (2.A.1.65/MFS_1/MFS)

Discovery: The MFSD6 and MFSD6L proteins were identified as putative SLC solute carriers among a collection of 30 atypical candidate proteins (174).

Gene family members (2):
MFSD6 (SLC73A1)	MFSD6L (SLC73A2)

Molecular aspects, physiological roles and links to disease

The SLC73 family belongs to the Uncharacterized Major Facilitator-14 (UMF14) family (TC 2.A.1.65), which is part of the MFS superfamily. A homology model of MFSD6 revealed 12 TMHs (3356) and between TMH 6 and 7 the classical MFS fold (3357).

MFSD6 (SLC73A1) – Orphan transporter: MFSD6 (SLC73A1) is an orphan transporter that has been proposed to have a role in the brain during variable energy consumption (i.e., high fat diet consumption). Specifically, it showed increased expression levels in the brain with increased energy consumption (3356). Based on studies of mouse wildtype brain tissue sections, MFSD6 (SLC73A1) is strongly expressed in brain in neurons but not astrocytes (3356). According to the HPA, it is most highly expressed in oligodendrocytes and neurons, as well as in prostatic basal cell, and it is widely distributed at moderate levels.

Two independent studies have identified MFSD6 as an enterovirus D68 entry receptor. The findings indicate a clear therapeutic potential through decoy receptor strategies (3358, 3359). However, the studies do not provide new data on the transport function or normal physiological role of MFSD6.

MFSD6L (SLC73A2) – Orphan transporter: MFSD6L (SLC73A2) is an acrosome membrane protein. The HPA suggests that it is highly expressed in testis, and at lower levels in intestine, pancreas and endocrine tissues.

MFSD6L plays an important role in the formation of the acrosome by interacting with the inner acrosomal membrane protein SPACA1. MFSD6L deficiency affects male fertility and causes oligoasthenoteratozoospermia in humans and mice due to disruption of the normal formation of acrosome and head shaping (3360). Specifically, men harboring bi-allelic MFSD6L variants displayed reduced sperm motility and abnormal sperm morphologies. Thus, MFSD6L is a new acrosome membrane protein that is required for acrosomal formation, anchoring, and sperm head shaping via interaction with the IAM protein SPACA1, but its transport function is unknown.

Orphan transporter family members (2)

MFSD6 (SLC73A1), MFSD6L (SLC73A2)

HGNC updates

The SLC73 family was created to accommodate these new SLC assignments after thorough phylogenetic analysis. SLC73A1 and SLC73A2 are new aliases for MFSD6 and MFSD6L, respectively.

SLC74 lysosomal chloride channel family (2.A.1.2/MFS_1/MFS)

Discovery: MFSD8 was identified as a novel neuronal ceroid lipofuscinosis gene that encodes a putative lysosomal transporter (3361)

Gene family member (1)

MFSD8 (SLC74A1)

Molecular aspects, physiological roles and links to disease

The SLC74 family belongs to The Drug:H⁺ Antiporter-1 (12 Spanner) (DHA1) family (TC 2.A.1.2), which is part of the MFS superfamily.

MFSD8 (SLC74A1): MFSD8 (SLC74A1), also known as CLN7 (see below), has been reported to function as a lysosomal chloride channel and it has been shown that pathogenic mutations of MFSD8 result in a late-infantile form of neuronal ceroid lipofuscinosis (NCL) called CLN7 disease and lead to a decrease in chloride permeability (3361–3363). CLN7 disease is one of a group of disorders known as NCLs, also collectively referred to as Batten disease, that affect the nervous system and worsen vision, movement and thinking ability. An in-frame deletion in the MFSD8 gene has also been reported to cause CLN7 disease (3364). In a mouse model, deficiency of lysosomal CLN7 (Mfsd8) was reported to lead to impaired constitutive autophagy and neurodegeneration late in the disease (3365).

Based on the HPA and other evidence (3361), MFSD8/CLN7 (SLC74A1) is ubiquitously expressed, but at the single cell level it is most highly expressed in rod photoreceptor cells.

Lysosomes play a central role in retinal cells and are implicated in retinal degenerative diseases (3366). Consistent with this, the recent genetic study from 2024 entitled “Maculopathy and adult-onset ataxia in patients with biallelic MFSD8 variants” strongly supports this concept (3367).

Studies on the role of an MFSD8 ortholog on the secretome of the social amoeba Dictyostelium discoideum provided further insights into the role of MFSD8 in protein secretion, which is altered in CLN7 disease due to pathological mutations of MFSD8 (3368).

Orphan transporter family members: N/A

HGNC update

The SLC74 family was created to accommodate this new SLC assignment after thorough phylogenetic analysis. SLC74A1 is a new alias for MFSD8.

SLC75 Tetracycline transporter-like family (2.A.1.2/ MFS_1/MFS)

Discovery: MFSD10 was identified as a putative SLC solute carrier among a collection of 30 atypical candidate proteins (174).

Gene family member (1)

SLC75A1 (MFSD10)

Molecular aspects, physiological roles and links to disease

The SLC75 family is related to the Drug:H⁺ Antiporter-1 (12 Spanner) (DHA1) family (TC 2.A.1.2), which is part of the MFS superfamily.

SLC75A1 - Orphan transporter: MFSD10 (SLC75A1), also known as tetracycline transporter-like protein (TETRAN), is a human ortholog of the yeast multidrug transporter Tpo1p that has been shown to transport organic anions and is expressed at the luminal membranes of renal proximal tubule cells (3369). Overexpression of TETRAN in cultured has been shown to facilitate the uptake of organic anions such as indomethacin, a non-steroidal anti-inflammatory drug (NSAID), and fluorescein. MFSD10 also has significant similarity (31% identity) with the TetA tetracycline efflux transporter of Gram-negative bacteria (3370). Based on studies in mice, SLC75A1 expression was altered in response to increased or decreased energy consumption (e.g., normal chow, 24 h starvation and high fat diet) (3356). Further studies are needed to fully elucidate the function of this transporter.

Orphan transporter family member (1)

SLC75A1 (MFSD10)

HGNC update

The SLC75 family was created to accommodate this new SLC assignments after rigorous phylogenetic analysis. SLC75A1 is a new symbol for the gene previously approved as MFSD10.

SLC76 Glycolipid translocator family (RFT1) (2.A.66.3/Rft-1/MATE)

Discovery: Using a screen designed to detect cellular defects requiring an intact unfolded protein response, RFT1 was identified as a mutant in Saccharomyces cerevisiae (3371). The mutant strain was shown to be deficient in N-linked glycosylation (3371) and the mutation was mapped to the Rft ER protein (3372). It was subsequently shown that yeast rft1 encodes a protein required for the translocation of the branched oligosaccharide Man₅GlcNAc₂ into the ER (3372). Later, a young patient was diagnosed with a congenital disorder of glycosylation characterized by intracellular accumulation of the lipid-linked oligosaccharide dolichyl pyrophosphoryl (DolPP)-GlcNAc₂Man₅, showing a phenotype reminiscent of the lipid-linked and N-linked oligosaccharide profiles described in yeast lacking the Rft1 protein (3373). These findings suggest that in the absence of RFT1 activity, DolPP-GlcNAc₂Man₅ accumulates on the cytosolic side of the ER membrane, leading to underglycosylation of N-glycoproteins. In the search for a human orthologue of yeast Rft1, a single gene – human RFT1 – was identified, whose predicted protein shares 22% sequence identity with the yeast Rft1 protein. Sequence analysis of RFT1 in fibroblasts from a patient with congenital disorder of glycosylation revealed a point mutation (R67C) in the human RFT1 protein (3373).

Gene family member (1)

RFT1 (SLC76A1)

Molecular aspects, physiological roles and links to disease

The SLC76 family belongs to the Oligosaccharidyl-lipid Flippase (OLF) Family (TC 2.A.66.3), which harbors the MATE fold. N-linked glycosylation is an essential post-translational modification in eukaryotes. The substrate of N-linked glycosylation, DolPP-GlcNAc₂Man₉Glc₃, is synthesized through a series of ordered reactions. A key step in this pathway is the translocation of the intermediate DolPP-GlcNAc₂Man₅ across the endoplasmic reticulum membrane, a process in which RFT1 is believed to be involved (3372, 3373).

RFT1 (SLC76A1): RFT1 (SLC76A1), also known as Man₅GlcNAc₂-PP-dolichol translocation protein, is an ER protein that has 14 TMHs according to AlphaFold2, with N- and C-termini facing the cytoplasm (3374). While there is evidence that RFT1 contributes to N-glycosylation by mediating the translocation of lipid-linked oligosaccharides across the ER membrane (3372, 3375) the significance of this role remains somewhat controversial. Subsequent in vitro studies have shown that Man₅GlcNAc₂-PP-dolichol scramblase activity is retained even in assays lacking RFT1 (3374). Nevertheless, as described above under “Discovery”, genetic defects in RFT1 (SLC76A1) clearly cause inborn errors of glycosylation (3373, 3374).

Orphan transporter family members: N/A

HGNC update

The SLC76 family was created to accommodate this new SLC assignment. SLC76A1 is a new alias for RFT1. Note that SLC52A1 and SLC19A1 have also been published as RFT1 (riboflavin transporter 1) and RFT-1 (reduced folate transporter 1) respectively, but these are not related to RFT1 (SLC76A1).

10. SLC-like proteins

Our recent search for proteins with SLC-like properties identified 129 candidate human proteins that were not previously classified as SLCs (179). The criteria for defining “SLC-like” were reported in detail and used to select protein families from the TCDB and Pfam databases (179). Seventy-seven of the genes encoding these proteins were assigned SLC symbols or, if the original gene nomenclature was already widely used in the literature, SLC aliases. The remaining 52 SLC-like candidates are shown in black on the dendrogram wheel in Fig. 3 and are listed in Table 4 (in clockwise order from the top center of the wheel). The number of putative TMHs, their structural folds and transport modes are given where known in Table 4.

Table 4. SLC-like proteins.

^Fold families where alternating access mechanism has been established; §Fold families where transport mechanism has been suggested.

Family name	UniProt ID	Gene symbol	TCDB family	Pfam family	Protein name	TMHs  (UniProt, AlphaFold)	Structural fold
The Magnesium Transporter1 (MagT1) Family (TC 1.A.76)
SLC58-related	Q9NRP0	OSTC	1.A.76.2	OST3_OST6	Oligosaccharyltransferase complex subunit OSTC, also known as DC2	3	MagT
The Mitochondrial StAR-related lipid transfer protein (StAR) Family (TC 8.A.120)
STAR	Q14849	STARD3	8.A.120.2.1	MENTAL	StAR-related lipid transfer protein 3	4
STAR	O95772	STARD3NL	8.A.120.2.3	MENTAL	STARD3 N-terminal-like protein	4
The Mitochondrial Cholesterol/Porphyrin/5-aminolevulinic acid Uptake Translocator Protein  (TSPO) Family (TC 9.A.24)
TSPO	P30536	TSPO	9.A.24.1	TspO_MBR	Translocator protein	5	TSPO
TSPO	Q5TGU0	TSPO2	9.A.24.1	TspO_MBR	Translocator protein 2	5	TSPO
The Lipid Intermediate Transporter (Arv1) Family (TC 9.A.19)
ARV1	Q9H2C2	ARV1	9.A.19	Arv1	Protein ARV1	3
The Chloride Carrier/Channel (ClC) Family (TC 2.A.49)
CLCN	P35523	CLCN1	2.A.49.2	Voltage_CLC	ClC-1 (Chloride channel)	10	CLC^
CLCN	P51788	CLCN2	2.A.49.2	Voltage_CLC	ClC-2 (Chloride channel)	10	CLC^
CLCN	P51790	CLCN3	2.A.49.2	Voltage_CLC	CLC-3 (H+/Cl- exchanger)	10	CLC^
CLCN	P51793	CLCN4	2.A.49.2	Voltage_CLC	CLC-4 (H+/Cl- exchanger)	10	CLC^
CLCN	P51795	CLCN5	2.A.49.2	Voltage_CLC	CLC-5 (H+/Cl- exchanger)	10	CLC^
CLCN	P51797	CLCN6	2.A.49.3	Voltage_CLC	CLC-6 (H+/Cl- exchanger)	10	CLC^
CLCN	P51798	CLCN7	2.A.49.3	Voltage_CLC	CLC-7 (H+/Cl- exchanger)	10	CLC^
CLCN	P51800	CLCNKA	2.A.49.2	Voltage_CLC	ClC-Ka (Chloride channel)	10	CLC^
CLCN	P51801	CLCNKB	2.A.49.2	Voltage_CLC	ClC-Kb (Chloride channel)	10	CLC^
The Death Effector Domain A (DedA) Family (TC 9.B.27)
TMEM41-64	Q96HV5	TMEM41A	9.B.27.1	SNARE_assoc	Transmembrane protein 41A	5
TMEM41-64	Q5BJD5	TMEM41B	9.B.27.1	SNARE_assoc	Transmembrane protein 41B	6
TMEM41-64	Q6YI46	TMEM64	9.B.27.5	SNARE_assoc	Transmembrane protein 64	6
The Auxin Efflux Carrier (AEC) Family (2.A.69 and NhaA fold - The Monovalent Cation:Proton  Antiporter CPA1 and CPA2 families (2.A.36)
GPR155	Q7Z3F1	GPR155	2.A.69.3	Mem_trans	Integral membrane protein GPR155	17	NhaA^
The Drug/Metabolite Transporter (DMT) Superfamily TC 2.A.7)
TMEM144	Q7Z5S9	TMEM144	2.A.7.8	TMEM144	Transmembrane protein 144	10	NST§
TMEM234	Q8WY98	TMEM234	2.A.7.32	TMEM234	Transmembrane protein 234	3	NST§
The N-Acetylglucosamine Transporter (NAG-T) Family (TC 2.A.1.58)
UNC93	O43934	MFSD11	2.A.1.58	UNC-93	UNC93-like protein MFSD11	12	MFS^
UNC93	Q86WB7	UNC93A	2.A.1.58	UNC-93	Protein unc-93 homolog A	11	MFS^
UNC93	Q9H1C4	UNC93B1	2.A.1.58	UNC-93	Protein unc-93 homolog B1	12	MFS^
The Cholesterol Uptake Protein (ChUP) or Double Stranded RNA Uptake Family (TC 1.A.79)
SIDT	Q9NXL6	SIDT1	1.A.79.1	SID-1_RNA_chan	SID1 transmembrane family member 1	11
SIDT	Q8NBJ9	SIDT2	1.A.79.1	SID-1_RNA_chan	SID1 transmembrane family member 2	10
The Autoinducer-2 Exporter (AI-2E) Family (TC 2.A.86)
TMEM245	Q9H330	TMEM245	2.A.86	AI-2E_transport	Transmembrane protein 245	14
The Lysosomal Cobalamin (B12) Transporter (L-B12T) Family (TC 9.A.54)
LMBR	Q9NUN5	LMBRD1	9.A.54.1	LMBR1	Probable lysosomal cobalamin transporter	9
LMBR	Q8WVP7	LMBR1		LMBR1	Limb region 1 protein homolog	9
LMBR	Q6UX01	LMBR1L	9.A.54.1	LMBR1	Protein LMBR1L	9
LMBR	Q68DH5	LMBRD2	9.A.54.3	LMBR1	LMBR1 domain-containing protein 2	9
The Fatty Acid Exporter (FAX) Family (TC 2.A.126)
TMEM14	Q9Y6G1	TMEM14A	2.A.126.1	Tmemb_14	Transmembrane protein 14A	3	TMEM14
TMEM14	Q9NUH8	TMEM14B	2.A.126.1	Tmemb_14	Transmembrane protein 14B	4	TMEM14
TMEM14	Q9P0S9	TMEM14C	2.A.126.1	Tmemb_14	Transmembrane protein 14C	4	TMEM14
The 4 TMS Multidrug Endosomal Transporter (MET) Family (TC 2.A.74)
LAPTM	Q15012	LAPTM4A	2.A.74.1	Mtp	Lysosomal-associated transmembrane protein 4A	4	Tspn
LAPTM	Q86VI4	LAPTM4B	2.A.74.1	Mtp	Lysosomal-associated transmembrane protein 4B	4
LAPTM	Q13571	LAPTM5	2.A.74.1	Mtp	Lysosomal-associated transmembrane protein 5	5	Tspn
The Resistance-Nodulation-Cell Division (RND) Superfamily (TC 2.A.6)
Dispatched	Q96F81	DISP1	2.A.6.9	Patched	Protein dispatched homolog 1	12	RND^
Dispatched	A7MBM2	DISP2	2.A.6.9	MMPL	Protein dispatched homolog 2	12	RND^
Dispatched	Q9P2K9	DISP3	2.A.6	Patched	Protein dispatched homolog 3	12	RND^
SLC65-related	Q12770	SCAP	2.A.6.6	Patched	Sterol regulatory element-binding protein cleavage-activating protein	8	RND (half)
SLC65-related	P04035	HMGCR	2.A.6.6	Patched	3-hydroxy-3-methylglutaryl-coenzyme A reductase	7	RND (half)
The KX Blood-group Antigen (KXA) Family (TC 2.A.112)
XK		XK	2.A.112.1	XK-related	Membrane transport protein XK	10	XK§
XK		XKR3 (XTES)	2.A.112.1	XK-related	XK-related protein 3	10	XK§
XK		XKR4	2.A.112.1	XK-related	XK-related protein 4	10	XK§
XK		XKR5	2.A.112.1	XK-related	XK-related protein 5	5	XK§
XK		XKR6	2.A.112.1	XK-related	XK-related protein 6	7	XK§
XK		XKR7	2.A.112.1	XK-related	XK-related protein 7	7	XK§
XK		XKR8	2.A.112.1	XK-related	XK-related protein 8	8	XK§
XK		XKR9	2.A.112.1	XK-related	XK-related protein 9	8	XK§
XK		XKRX	2.A.112.1	XK-related	XK-related protein 2	10	XK§
The TMEM205 (TMEM205) Family (TC 9.A.55)
TMEM205		TMEM205	9.A.55.1		Transmembrane protein 205

OSTC (DC2)

Our search for SLC-like transporters identified OSTC/DC2 as “SLC-like” due to its similarity to SLC58 members, as it is a paralog of MAGT1 and TUSC3 (see the description of SLC58 in Section 9). Our search also revealed OSTCP1 (OSTCL) as well as an uncharacterized transcript with UniProt accession “B4DH36” (OSTCP7). As these correspond to two of the 8 processed pseudogenes derived from OSTC and are unlikely to encode functional transporters they are not included in Table 3.

OSTC/DC2 has been shown to be part of the OST-A complex and MAGT1 or TUSC3 part of the Ost-B complex. MAGT1 and TUSC3 have been shown to contribute to Mg²⁺ transport. Whether OSTC/DC2 is also involved in Mg²⁺ transport is unknown.

STARD3 and STARD3NL

STARD3 acts to establish membrane contacts between the ER and late endosomes where it moves cholesterol. It is a lipid trafficking protein expressed in late endosomal membranes where it may be involved in cholesterol export (3376). STARD3 forms a complex with the late endosomal protein STARD3 N-terminal-like protein (STARD3NL) and ER vesicle-associated membrane protein (VAMP)-associated proteins to tether the two organelles together (3377, 3378). Its closest homologue is the steroidogenic acute regulatory protein StAR (StarD1) (3379), which initiates steroid production by translocating cholesterol from the outer to the inner mitochondrial membrane in steroidogenic cells, the rate-limiting step in steroid hormone formation. It is proposed that STARD3 moves cholesterol into the mitochondria under certain conditions to initiate steroidogenesis, such as in the human placenta, which lacks StAR but still produces steroids.

The STARD3 gene is often found to be co-amplified with the human epidermal growth factor receptor 2 (HER2/ERBB2) gene in cases of breast cancer (3380). STARD3 overexpression and the resulting aberrant handling of cholesterol impact overall survival, recurrence-free survival, and non-metastatic survival. A STARD3 inhibitor has recently been developed and tested in various breast and colon cancer cell lines, revealing promising results (3381). STARD3 is considered a biomarker for HER2-positive breast cancer, and it likely plays a role in driving cancer aggressiveness and contributing to the resistance of the anticancer agent trastuzumab in HER2-positive cancers (3382). Therapeutic strategies that target the cholesterol-transfer mechanism of STARD3 could be an attractive treatment approach for this cancer subtype.

The AlphaFold structure of STARD3 is shown in Fig. 2.

TSPO and TSPO2

In steroidogenic cells, cholesterol binds to the STARD1/VDAC1/TSPO complex at the outer mitochondrial membrane from where it is transported to the inner mitochondrial membrane (3383). TSPO is a high-affinity cholesterol- and drug-binding protein that polymerizes upon hormonal stimulation, thereby increasing its binding affinity and stimulating cholesterol transfer to the inner mitochondrial membrane. The cytochrome P450 enzyme 20,22-desmolase (CYP11A1) then initiates steroidogenesis by converting cholesterol to pregnenolone at the inner mitochondrial membrane. Thus, the major function of TSPO is mitochondrial cholesterol trafficking, transporting cholesterol from the outer to the inner mitochondrial membrane. TSPO ligands have potential diagnostic and therapeutic applications ranging from attenuation of cancer cell proliferation to neuroprotection. The structure of TSPO in complex with a diagnostic ligand has been determined (3384). Both TSPO and TSPO2 have five TMHs and a cholesterol-binding motif near the C-terminal region, which is located on the cytoplasmic side. NMR spectroscopy revealed the cholesterol-mediated allosteric regulation of TSPO protein structure (3385).

The subcellular location of its paralog TSPO2 differs from the mitochondrial localization of TSPO. TSPO2 is located on ER and nuclear membranes and its cellular distribution is restricted to specific hematopoietic tissues (3386). During erythropoiesis, organelles are lost and the nucleus is extruded, a process that is inhibited by excess cholesterol, and as erythroblasts mature into erythrocytes, cholesterol levels decrease and the cholesterol biosynthetic pathway is slowed down. The role of TSPO2 has been shown to be the redistribution of free cholesterol within the cell (3387). Therefore, TSPO2 plays an important role during erythroid maturation and it has been demonstrated that cholesterol-binding TSPO2 coordinates maturation and proliferation of terminally differentiating erythroblasts (3388). However, the function of TSPO2 extends beyond cholesterol regulation (3388); it has also been reported to facilitate the transport of the heme analog protoporphyrin IX and ATP in human erythrocytes by forming a complex with the voltage-dependent anion channel and an adenine nucleotide transporter (3389, 3390). This is reminiscent of TSPO, which forms a complex with the voltage-dependent anion channel VDAC and an SLC25 adenine nucleotide transporter (ANT) of the inner mitochondrial membrane to mediate mitochondrial functions, including cholesterol and porphyrin transport (3391).

ARV1

Arv1 is an ER membrane protein that is widely expressed, including in brain and heart muscle, according to the HPA. It is required for normal ER cholesterol and bile acid homeostasis (3392) and is involved in sterol transport out of the ER and allocation to the plasma membrane (3393). In addition, ARV1 is thought to be involved in GPI anchor synthesis (3394). Deletion of ARV1 disrupts sterol distribution, inhibits GPI-anchored protein maturation, and causes protein accumulation in the ER (3395). Studies in yeast have shown that loss of Arv1 primarily induces lipid bilayer stress (3395). Defects in the ARV1 gene cause autosomal recessive epileptic encephalopathy (3396) and dilated cardiomyopathy (3397). The AlphaFold generated structure of ARV1 is presented in Fig. 2, showing this membrane protein with an estimated 3-5 TMHs.

Chloride Carrier/Channel (ClC) Family (CLCN1 to CLCN7, CLCNKA, CLCNKB)

All 9 members of the Chloride Carrier/Channel (ClC) family (TC 2.A.49; approved gene symbols CLCN1 to 7, CLCNKA and CLCNKB) were identified by our approach. This is not too surprising, as several CLCs of the endo/lysosomal system turned out not to be chloride channels, but rather secondary active 2Cl^-/1H⁺ antiporters (67, 3398).

Of the 9 ClC family members, four (ClC-1, ClC-2, ClC-Ka, ClC-Kb) are Cl^- channels located in the plasma membrane and the rest (ClC-3, ClC-4, ClC-5, ClC-6, ClC-7) are Cl^-/H⁺ antiporters located in intracellular organelles (67, 3399). Some of the CLC channels additionally associate with accessory subunits. One of them is Barttin, which improves the stability of the ClC-K channel protein, stimulating its exit from the endoplasmic reticulum and insertion into the plasma membrane (3400). GlialCam is a facultative subunit of ClC-2 that modifies gating and thus increases functional variability within the ClC family (3399).

Important breakthroughs in obtaining cryo-EM structures have revealed the structural details of the ClC channels/transporters (3398, 3401, 3402). All ClC channels/transporters are believed to exist as dimers, with each monomer either forming an independent Cl^- pore in the ClC channels or mediating anion/H⁺ exchange in the CLCN transporters (3403). In addition, each subunit consists of two related, oppositely oriented halves that assemble in an antiparallel fashion to form the anion pore (3402). Overall, the ClC channels/transporters were found to function according to an unconventional transport mechanism that contradicts the basic premises of the alternating-access paradigm for exchange transport and opens new insights into the principles of secondary transport and channel gating (3401).

Diseases resulting from ClC dysfunction include myotonia congenita, leukoencephalopathy, osteopetrosis, epilepsy, and lysosomal storage disorders (67). For example, myotonia congenita is caused by loss-of-function mutations in CLCN1. The disease is characterized by increased excitability in skeletal muscle and delayed recovery after muscle contraction.

Variations in the CLCN4 vesicular 2Cl^-/H⁺ exchanger have been identified as a genetic cause of X-linked neurodevelopmental disorders (3404, 3405).

Variations in the CLCN6 late endosomal Cl⁻/H⁺ exchanger have been reported as a novel cause of neuronal ceroid lipofuscinoses (3406). In addition, ocular manifestations of a gain-of-function mutation in CLCN6 has been reported (3407).

Variations in the lysosomal Cl⁻/H⁺ exchanger CLCN7 are associated with osteopetrosis affecting bone resorptions (3408).

The two ClC-K channels form a subset of the CLC proteins. They are predominantly expressed in the kidney and inner ear and are required for NaCl resorption in the loop of Henle and for K⁺ secretion by the stria vascularis of the of the cochlear duct (3400). The subcellular distribution and function of these channels are tightly regulated by the accessory subunit Barttin. Dysfunction of ClC-K channels results in Bartter syndrome, characterized by impaired urine concentration (see also NKCC2/SLC12A1 description in Section 9) (3400). The reason is that the lack of function of these channels interferes with the urinary countercurrent concentration mechanism in the loop of Henle, which is responsible for the reabsorption of water and electrolytes from the collecting duct, thus resulting in a diuretic effect (3409).

TMEM41A, TMEM41B and TMEM64

The TMEM41A, TMEM41B, and TMEM64 proteins clustered into the same family in our results. These are the only proteins in humans that show any similarity to the “SNARE_assoc” Pfam domain, as well as the TCDB family #9.B.27 called the Death Effector Domain A (DedA) family. The AlphaFold structure of TMEM41A is shown in Fig. 2.

TMEM41A

While little is known about the normal biological function of TMEM41A, it has been shown to be aberrantly expressed in a number of cancers and is associated with poor prognosis, for example, in breast cancer (3410). Overexpression of TMEM41A in cancer cells increases their migration and invasion capacity. TMEM41A overexpression was shown to correlate with poor prognosis and immune alterations in patients with endometrial cancer (3411). TMEM41A was highly expressed in gastric cancer and was associated with lymph node metastasis, distant metastasis, late stage and poor prognosis (3412).

TMEM41B

The best characterized member of the human protein family is TMEM41B. Structural modeling revealed features of this transporter reminiscent of secondary transporters, such as a tandem internal repeat with twofold rotational symmetry, and an H⁺ antiporter activity has been proposed as the mechanism of transport (3413). While the exact function of TMEM41B is still unclear, it has been shown to form a complex with vacuole membrane protein 1 (VMP1), which also harbors the “SNARE_assoc” domain, both of which are required for autophagosome formation (3414).

TMEM41B is widely expressed according to the HPA and localizes to mitochondria-associated ER membranes (3415–3417). TMEM41B-deficient cells exhibit larger lipid droplets, a phenomenon that may be caused by lipid accumulation due to a disruption of the release of free fatty acids from lipid droplets to other organelles such as mitochondria (3414, 3418). It has been suggested that TMEM41B acts as an ER scramblase for lipoprotein biogenesis and lipid homeostasis, shuttling phospholipids between the leaflets of bilayer membranes (3419–3421). However, this interpretation has since been revised (see below) (3422). TMEM41B was also reported to function as a host factor for viral replication in a variety of viruses. TMEM41B appears to be an essential factor for SARS-CoV-2 (3423), and probably also for flaviviral (3424) infections. It may facilitate membrane curvature, which is beneficial for viral replication (3424).

In mammalian cells, the ER passively releases Ca²⁺ under steady state conditions. The channels involved were unclear, until TMEM41B was identified to fulfill this role (3422). As described above, TMEM41B was initially reported to exhibit phospholipid scramblase activity, which suggests its potential involvement in lipid metabolism and viral infection. However, subsequent studies have demonstrated that TMEM41B functions as an endoplasmic reticulum (ER) calcium release channel. This channel plays an indispensable role in cellular processes such as maintaining metabolic quiescence and the responsiveness of naïve T cells. Purified recombinant TMEM41B forms a concentration-dependent Ca²⁺ channel in single-channel electrophysiology assays (3422). At the cellular level, TMEM41B deficiency causes ER Ca²⁺ overload, while TMEM41B overexpression depletes ER Ca²⁺.

Considering the widespread expression of TMEM41B in various tissues, it is reasonable to speculate that TMEM41B-ER Ca²⁺ release extends beyond the immune system to regulate cellular functions. For example, deletion of TMEM41B in the liver has been shown to induce nonalcoholic fatty liver disease in mice (3419).

TMEM64

TMEM64 has been shown to be an ER protein that controls osteoblast and prostate tumor growth via the Wnt/β-catenin signaling pathway (3425–3427) and to serve as a regulator of glioma proliferation and aggressiveness. Further investigation of the mechanism underlying the specific function of TMEM64 indicated a strong correlation between TMEM64-mediated β-catenin nuclear translocation and glioma aggressiveness (3425). TMEM64 may serve as a promising prognostic marker and therapeutic target for glioma treatment.

GPR155 (LYCHOS)

GPR155 was originally identified as a G protein-coupled receptor and was later found to be a lysosomal cholesterol sensor, hence the alternative name “lysosomal cholesterol signaling” (LYCHOS) (2691). For additional information on GPR155, see also “Human NhaA-fold SLC Families” in Section 8 and the description of the SLC65 family in Section 9.

GPR155/LYCHOS belongs to the Auxin Efflux Carrier (AEC) family (TC 2.A.69.3) (3428). The HPA suggests ubiquitous expression with prominent expression in brain (especially choroid plexus), stomach, kidney, skin (melanocytes). GPR155 embodies the characteristic Pfam domain “Mem_trans” as the only human protein analyzed (7). Cryo-EM structures of GPR155 reveal a unique fusion of a plant auxin transporter-like domain and a seven-transmembrane G-protein-coupled receptor (GPCR)-like domain (262, 3429). These structures provide mechanistic insights into the cellular regulation of mTORC1 activity by GPR155.

As highlighted in Section 8, the first 10 TMHs of the protein show a 5+5 TMH arrangement, and TMHs 6-10 show similarity to the N-terminal half of the sodium/bile transporters of ASBT (SLC10A2) (7). On the other hand, the last 7 TMHs of GPR155 (TMHs 11-17) show similarity to the GPCR-fold (7 TMHs) proteins with known structure, with the highest similarity to the structures of the human G protein-coupled receptor Smoothened (SMO), a signal transducer of the developmentally relevant Hh pathway (see SLC65B1/PTCH1 in Section 9) (2691, 3430).

As part of the cholesterol sensing mechanism in lysosomes (2691), cholesterol interacts with TMH1 of the permease core of GPR155 (3428, 3431). It was shown that GPR155 signals cholesterol sufficiency to mTORC and that cholesterol bound to TMH1 at the N-terminal region of GPR155 is required for mTORC1 activation and mTORC1-dependent anabolic signaling. This mTORC1 activation is triggered by the interaction of cholesterol-bound GPR155 with the GATOR1 complex (2691, 3432). In addition, it was shown that GRP155 expression is decreased in fasting animals, whereas expression of the lysosomal transporter NPC1 (SLC65A1), a negative regulator of cholesterol-dependent mTORC1 activation (see SLC65 description in Section 9), is increased in fasting animals (2691).

Thus, GPR155 (LYCHOS) does not function as a classical GPCR. Rather than being localized at the plasma membrane, it is found at the lysosomal membrane, where it acts as an intralysosomal sensor. Its main ligand is cholesterol. There, GPR155 functions as a nutrient-responsive lysosomal cholesterol sensor that complements the SLC38A9–NPC1–mTORC1 axis (see Fig. 45). As reported (2691), the current concept is that GPR155 levels are elevated under conditions of nutrient abundance, which promotes cholesterol-dependent mTORC1 signaling when metabolic building blocks are abundant to trigger mTORC1 anabolic pathways. Under conditions of nutrient deprivation, GPR155 expression decreases, NPC1 levels increase, and lysosomal cholesterol export is increased, shutting down cholesterol-mTORC1 signaling to conserve cellular resources (2691).

GPR155 has been linked to a number of different cancers such as gastric cancer (3433) and hepatocellular carcinoma (3434).

TMEM144 and TMEM234

The search for SLC-like proteins identified the orphan transporters TMEM144 and TMEM234, which show a remote similarity to SLC35 members (7). Both contain the NST fold and thus belong to the Drug/Metabolite Transporter (DMT) Superfamily (TC 2.A.7).

TMEM144 is an orphan transporter reported to be associated with the cholesterol content of bovine milk (7, 3435). TMEM144 is part of the Caenorhabditis elegans ORF (CEO) family (TC 2.A.7.8). The HPA suggests that TMEM144 is expressed almost exclusively in the central nervous system, especially in the spinal cord, with subcellular localization in mitochondria.

TMEM234, also known as C1orf91, is also an orphan transporter. Knockdown experiments have been reported to cause proteinuria in zebrafish (3436). In the TCDB, it is assigned to the “Uncharacterized DMT4 (U-DMT4) family” (TC 2.A.7.32) as part of the Drug/Metabolite Transporter (DMT) superfamily. It contains a corresponding “TMEM234” Pfam domain, which is a member of the “DMT” clan of Pfam domains (7). The HPA suggests that TMEM234 is widely expressed in human tissues.

The cyclophilin PPIH (peptidyl-prolyl cis-trans isomerase H), which is significantly upregulated in hepatocellular carcinoma, showed a strong positive association with the expression of TMEM234/C1orf91 (3437).

MFSD11, UNC93A and UNC93B1

These are orphan transporters that are part of the N-Acetylglucosamine Transporter (NAG-T) family (TC 2.A.1.58) which carry the MFS fold.

MFSD11 is widely expressed, with particularly strong expression in the retina. The fly homolog of MFSD11 has been proposed to be involved in nutrient homeostasis and has a potential role in locomotion, based on an initial characterization of the transporter in Drosophila melanogaster (3438). Based on structural modeling, the predicted structure of MFSD11 (3439) has been shown to overlap well with the crystal structure of the E. coli proton:xylose symporter XylE (493). Thus, it has been proposed that mammalian MFSD11 may be involved in intracellular transport and function as a sugar:H⁺ symporter (3439), a functional prediction consistent with a previous report suggesting that MFSD11 is a membrane protein that transports soluble molecules and is involved in energy regulation (3152).

UNC93A is an orphan transporter that is thought to play a role in controlling potassium flux in neurons to regulate responsiveness to synaptic input (3440). Expression studies in mice showed UNC93A staining in the cerebral cortex, hippocampus and cerebellum at both mRNA and protein levels. The ortholog of UNC93A in C. elegans was reported to be a component of the SUP-9 two-pore K⁺ channel that coordinates muscle contraction and acts as a regulatory protein of the channel (3441). The two-pore K⁺ channel encoded by SUP-9 is similar to human TWIK-related acid-sensitive K⁺ (TASK) channels, which are important for maintaining the resting membrane potential. Studies in fruit flies have shown that UNC93A associates with TASK channels (3442). The HPA suggests that human UNC93A mRNA is most highly expressed in skin and at somewhat lower levels in liver, duodenum and kidney. Interestingly, human UNC93A has been identified as a metabolite-associated locus in patients with chronic kidney disease (3443).

UNC93B1 is an orphan transporter that plays a pivotal role in Toll-like receptor (TLR) 7-dependent autoimmunity, highlighting its importance in the pathophysiology of systemic lupus erythematosus (SLE) (3444). UNC93B1 is thought to direct TLRs from the endoplasmic reticulum to their respective endosomal signaling compartments (3445). According to the HPA, UNC93B1 is widely expressed in human tissues, with particularly high expression in the spleen. Gain-of-function variants in UNC93B1 have been shown to cause SLE (3446). The identified UNC93B1 variants revealed different mechanisms of gain of TLR7 and TLR8 signaling. Large-scale mutational analysis has further identified UNC93B1 variants that drive TLR-mediated autoimmunity in mice and humans (3447).

SIDT1 and SIDT2

The systemic RNA interference defective proteins SIDT1 and SIDT2 are orphan transporters that belong to the cholesterol uptake protein (ChUP) or double-stranded RNA uptake family (TC 1.A.79). SIDT1 is predominantly located at the plasma membrane and promotes the cellular uptake of synthetic small interfering RNA (siRNA) (3448) and plant-derived microRNA (miRNA) (3449), whereas SIDT2 is a lysosomal protein with a variety of functional roles ranging from glucose and lipid metabolism to autophagy and nucleotide transport (3450).

SIDT1 and SIDT2 are annotated in databases as RNA transporters, but also share identity and conserved cholesterol binding (CRAC) domains with the C. elegans ChUP-1 cholesterol transporter. This suggests that they are also involved in cholesterol transport, which was demonstrated by showing that single point mutations targeting disruption of the CRAC domains of both proteins altered cholesterol transport (3451).

The AlphaFold structure of SIDT1 is shown in Fig. 2. Cryo-EM analyses provided further insight into the functional roles of SIDT1 and SIDT2 (3452, 3453). Cryo-EM structures of human SIDT1 were determined as a homodimer in a side-by-side arrangement with two distinct conformations, the cholesterol-bound form and the unbound form (3452). The SIDT1 structures revealed histidine and aspartate residues that coordinate a putative zinc ion in the membrane-spanning region, as well as ceramidase activity that is attenuated by cholesterol binding while acting as an allosteric regulator. The study also provides information on RNA transport by members of the SID-1 family of proteins. Additional structural analysis revealed insights into the inherent conformational dynamics within the lipid binding domain in ChUP family members (3454).

TMEM245

TMEM245 is an orphan transporter for which very little information is available. It has a total of 14 TMHs according to UniProt predictions. According to the HPA, TMEM245 is widely expressed. The C-terminal half of the TMEM245 protein shows weak similarity to members of the autoinducer-2 exporter (AI-2E) family (TC 2.A.86). This suggests that TMEM245 may have a transporter-like domain, at least in the C-terminal part of the protein. From a structural perspective, there is currently no similarity to any known structure, suggesting that TMEM245 has its own structural fold. An AlphaFold structure prediction is shown in Fig. 2.

LMBR1, LMBR1L, LMBRD1 and LMBRD2

Our search for SLC-like proteins identified four human proteins (LMBR1, LMBR1L, LMBD1/LMBRD1, LMBRD2) with the “LMBR1” Pfam domain that clustered into two families (7). These proteins correspond to the Lysosomal Cobalamin (B12) Transporter (L-B12T) family.

UniProt predicts 9 TMHs in a 5+4 arrangement for these proteins, but the tertiary structures are still unknown and no homologs with a known structure have been found. An AlphaFold structure of LMBR1 is presented in Fig. 2.

LMBR1

LMBR1 was the first protein in the family to be identified. The HPA reports that its expression is widespread, with the highest levels observed in neurons, microglia, oligodendrocytes, and astrocytes at the single-cell level.

Initially, the literature suggested that LMBR1 is associated with limb malformations (3455). However, subsequent research has clarified that this is an incorrect interpretation (3456). The critical regulatory element disrupted in this case is the ZRS (Zone of Polarizing Activity Regulatory Sequence) enhancer. ZRS is a long-range, limb-specific enhancer of the Sonic hedgehog (SHH) gene located within an intron of the LMBR1 gene on chromosome 7, just adjacent to the SHH gene. While the ZRS enhancer does not affect LMBR1 protein function, it controls SHH expression in the adjacent gene in the developing limb bud. Mutations in this enhancer zone lead to ectopic SHH expression at the anterior margin of the limb bud, resulting in extra digits (preaxial polydactyly) (3457). Therefore, although LMBR1 and SHH are adjacent, genomically linked genes, the observed polydactyly resulting from mutations in the LMBR1 locus does not reflect a functional role for the LMBR1 protein in limb development, but rather a disruption of the regulatory mechanism of the SHH gene.

It has also been reported that circular RNA (circRNA), derived from LMBR1, inhibits tumor growth in bladder cancer cells by binding directly to the ALDH1A3 aldehyde dehydrogenase protein (3458). However, again, this does not mean that the LMBR1 protein itself plays a role in tumor suppression. Circular RNAs are formed by back-splicing of exons from protein-coding genes and usually function independently of the protein encoded by their parent gene (3459).

Thus, despite these observations, the physiological role of LMBR1 remains unclear.

LMBR1L

The LIMR (lipocalin-1-interacting membrane receptor) protein, which is encoded by the LMBR1L gene, was originally described as having “significant homology” to lipocalin membrane receptors (3460). LIMR binds lipocalin-1 with high affinity (3461, 3462). Lipocalin-1 plays an important role in maintaining the stability of the tear film by eliminating lipids and fatty acids from the corneal surface and it also has antimicrobial and anti-inflammatory properties (3463). LIMR it is widely expressed in the ER fraction of cells and likely acts as an endocytic receptor for lipocalin-1 (3464). However, the link between LMBR1L and lipocalin-1 in the broader context of their biological roles remains unclear.

LIMR has been reported to regulate the proliferation and migration of endothelial cells through Norrin/β-catenin signaling, revealing an essential role for LMBR1L in angiogenesis (3464). Specifically, LIMR was shown to coimmunoprecipitate with numerous components of the Wnt/β-catenin signaling apparatus (3465). Studies using Lmbr1l knockout mice revealed that LIMR is essential for retinal vascular development (3464).

LMBRD1

The LMBRD1 (LMBR1 domain containing 1) gene encodes the LMBD1 protein which is widely expressed according to the HPA. It is believed to function as a lysosomal exporter of cobalamin (vitamin B12). Genetic defects in LMBRD1 were shown to cause the cblF (cobalamin F) defect, a rare inherited disorder of cobalamin (vitamin B12) metabolism that leads to methylmalonic acidemia and homocystinuria (3460). The defects prevent the lysosomal export of vitamin B12, disrupting its conversion into essential cofactors for mitochondrial succinyl-CoA synthesis and cytosolic methionine synthesis. Affected children suffer from heart defects, developmental delay and megaloblastic anemia. LMBD1 was shown to interact with the ABC transporter ABCD4 and assist in its lysosomal trafficking (3466). Interestingly, ABCD4 was later shown to transport vitamin B12 even in the absence of LMBD1 (3467). Therefore, further studies are required to clarify the precise role of LMBD1 in lysosomal cobalamin export.

To investigate the physiological role of LMBD1 further, Lmbrd1 knockout mice were generated (3468). Loss of LMBD1 in mice resulted in early embryonic lethality due to gastrulation failure. Studies have shown that LMBD1 is essential for the initiation of gastrulation in mice (3468). The crucial role of LMBD1 in early embryonic development is likely due to its involvement in transporting cobalamin, which is required for methylation and metabolic pathways that are essential for rapid cell fate transitions and epigenetic gastrulation remodeling during gastrulation. The discrepancy in phenotype between humans and mice resulting from the loss of LMBD1 expression is interesting and requires further investigation.

LMBRD2

LMBRD2 is an SLC-like protein with poorly described function. LMBRD2 protein levels were shown to be strongly upregulated by β2-adrenoceptor signaling agonist (3469). LMBRD2 is widely expressed according to the HPA. It is most strongly expressed in neurons, astrocytes, cardiomyocytes and spermatocytes. De novo missense variants in LMBRD2 have been identified that are associated with developmental and motor delays, brain structure abnormalities and dysmorphic features (3470).

TMEM14A, TMEM14B and TMEM14C

The TMEM14A, TMEM14B and TMEM14C proteins are members of the Fatty Acid Exporter (FAX) family (TE 2.A.126). They are the only human proteins that contain the “Tmemb_14” Pfam domain. While Pfam lists this domain as functionally uncharacterized, a plant fatty acid export protein (FAX1) containing this domain has been suggested to be involved in fatty acid export from the inner envelope of chloroplasts (3471).

TMEM14A

TMEM14A is widely expressed, most strongly in the choroid plexus according to the HPA. In kidney glomeruli it has been localized in podocytes (3472).

TMEM14A has been identified as a suppressor of Bax using a yeast-based functional screening (3473). Specifically, it inhibits Bax-induced apoptosis, blocking the activation and mitochondrial translocation of the pro-apoptotic protein Bax, a key component of the intrinsic apoptotic pathway. The studies showed that TMEM14A functions as a mitochondria-associated membrane protein that stabilizes mitochondrial membrane potential (3473). This highlights that TMEM14A has antiapoptotic activity. The function of vertebrate TMEM14 proteins remains unknown, though a role in mitochondrial lipid metabolism, energy metabolism, or apoptosis seems plausible (3471).

TMEM14A has been reported to accelerate the progression of human ovarian cancer cells by increasing glycolytic activity (3474). TMEM14A inhibits apoptosis in ovarian cancer cells while accelerating energy metabolism, including glycolysis and oxygen respiration. TMEM14A is positively correlated with c-Myc. Thus, these studies reveal that TMEM14A is a critical metabolic facilitator in ovarian cancer, enhancing glycolysis activity partly via c-Myc activation.

In non-small cell lung cancer (NSCLC), activation of the AXL receptor tyrosine kinase has been shown to lead to increased TMEM14A expression, promoting tumor cell proliferation. This finding highlights the role of TMEM14A in AXL-driven oncogenesis. AXL transcriptionally upregulates TMEM14A expression to mediate cell proliferation in NSCLC (3475).

Subsequently, TMEM14A was identified as a novel component of the glomerular filtration barrier (GFB), which is critical in podocytes for preventing protein leakage under normal conditions and in disease states (3472). Knockdown of tmem14a in zebrafish embryos resulted in proteinuria without impairing tubular reabsorption, confirming a podocyte-specific barrier defect. Increased levels of TMEM14A protein were found in the glomeruli of patients with various proteinuric renal diseases such as diabetic nephropathy, suggesting a compensatory or protective response to maintain podocyte function. Thus, TMEM14A acts as a critical regulator of glomerular barrier integrity, particularly in podocytes, where it prevents proteinuria. This likely occurs through mitochondrial protection, which maintains the structure of the GFB barrier and ensures an intact filtration system.

TMEM14B

TMEM14B is widely expressed, most highly in erythroid cells, germ cells and cone photoreceptor cells, according to the HPA. The functional properties are unknown.

TMEM14B has been identified as a poor prognostic biomarker in hepatocellular carcinoma using a specialized single-cell RNA sequencing analysis approach (3476). Specifically, TMEM14B was identified as the gene most strongly associated with survival outcomes. It is thought to affect metabolism and mitochondrial regulation.

TMEM14C

TMEM14C is widely expressed, most highly in erythroid cells, according to the HPA. It has been shown to be required for erythroid mitochondrial heme metabolism (3477) and later on to function as a mitochondrial heme exporter (3478).

The structure of TMEM14C has been solved using nuclear magnetic resonance (NMR) spectrometry (3479). It reveals a bundle of three TMHs and an amphipathic helix. The transport mechanism, however, remains to be clarified.

TMEM14C was originally identified as a putative mitochondrial protein and was found to be consistently co-expressed with proteins of the core machinery of heme biosynthesis (3480). TMEM14C was subsequently shown to be required for erythroid mitochondrial heme metabolism and was proposed to mediate import of protoporphyrinogen IX (PPgenIX) into the mitochondrial matrix.

However, genetic screens have revealed that TMEM14C mediates the exit of mitochondrial heme rather than its uptake. This conclusion stems from studies addressing the molecular basis of artemisinin drug susceptibility (3478). Artemisinins are a class of drugs primarily used as antimalarials that are derived from the Chinese herb Artemisia annua. They are sesquiterpene lactones that contain a unique endoperoxide bridge, which is crucial for their antimalarial activity. Artemisinins are activated by free heme within the mitochondria of malaria-causing Plasmodium parasites, a process involving reductive cleavage of the endoperoxide bridge that generates carbon-centered free radicals (3478). CRISPR screens in Toxoplasma gondii, validated in Plasmodium falciparum, identified Tmem14c disruption as significantly increasing dihydroartemisinin susceptibility. Disruption of this export, shown via Tmem14c knockout, increases mitochondrial heme and thereby sensitizes cells to artemisinin action. Based on these findings, it is currently believed that its human ortholog, TMEM14C, facilitates mitochondrial porphyrin export.

Although TMEM14C was originally reported to be essential for the delivery of porphyrins, especially protoporphyrinogen IX, into mitochondria across the inner membrane for heme synthesis, the newer study based on Plasmodium data (3478) challenges this notion, revealing that TMEM14C likely functions as a mitochondrial heme exporter instead. The importance of a heme exporter from mitochondria in erythropoiesis is well-documented. It clears porphyrins and heme from the mitochondrial matrix and ensures proper delivery to cytosolic heme-utilizing systems. FLVCR1 (SLC49A1) was previously thought to play this role, but it was later shown that FLVCR1 does not function as a mitochondrial heme exporter transporter (see the description of the SLC49 family).

LAPTM4A, LAPTM4B and LAPTM5

The LAPTM4A, LAPTM4B and LAPTM5 proteins belong to the Multidrug Endosomal Transporter (MET) family (TC 2.A.74). They represent a group of lysosome-associated transmembrane proteins that emerged from our search for SLC-like proteins. Originally, the mouse transporter protein (Mtp, ortholog of LAPTM4A) was characterized as a transporter mediating the transport of nucleosides and nucleobases between the cytoplasm and intracellular compartments (3481). But later, it was proposed that they are not transporters per se, but rather regulatory factors that either support the localization and targeting or the function of other transporters (3482).

LAPTM4A

LAPTM4A is a 4 TMH membrane protein found in the lysosome and endosome (3483). The AlphaFold structure of LAPTM4A is shown in Fig. 2. LAPTM4A is widely expressed according to the HPA. The highest expression occurs – in order of decreasing levels – in late spermatids, basal prostatic cells, trophoblasts, airway basal cells, granulocytes, ovarian stromal cells, granulosa cells, Leydig cells, renal proximal tubule cells, fibroblasts, Langerhans cells, and the choroid plexus.

Studies have shown that LAPTM4A is targeted from the Golgi to late endosomes/lysosomes in a manner dependent on the E3 ubiquitin ligase NEDD4 (Nedd4-1) and ESCRT (endosomal sorting complexes required for transport) proteins (3483). LAPTM4A is equipped with YxxΦ and PY motifs within its structure that are crucial for proper lysosomal targeting and the functional regulation of cellular processes. The YxxΦ motif plays a crucial role in lysosomal targeting because it is recognized by adaptor protein complexes (e.g., AP-1 or AP-3), which mediate trafficking to the endo-lysosomal system. The PY motif is commonly found in proteins that interact with WW domain–containing E3 ubiquitin ligases, such as NEDD4. These motifs recruit NEDD4-1 and the ESCRT machinery, which mediates sorting from the Golgi to intralumenal vesicles within late endosomes/lysosomes. Experimental evidence demonstrates that LAPTM4A binds to NEDD4 in a manner dependent on PY motifs (Hirota, 2021, #4943).

LAPTM4A has been identified as a regulator of macrophages in glioblastoma. It promotes immunosuppression and reduces responses to checkpoint therapy (3484). In a mouse glioma model, LAPTM4A was shown to promote the polarization of tumor-associated macrophages towards the M2 phenotype, which contributes to glioma progression by enhancing cell proliferation and invasion. Conversely, LAPTM4A-deficient glioma models demonstrate a shift toward the M1 macrophage phenotypes, resulting in stronger immune activation and heightened sensitivity to anti-programmed cell death (anti-PD-1) therapy. Research has shown that loss of LAPTM4A reprograms macrophages toward an anti-tumor M1 state, thereby improving the efficacy of anti-PD-1 therapy (3484). These findings suggest the potential of LAPTM4A potential as an immunotherapy target in brain tumors.

LAPTM4B

The late endosomal protein LAPTM4B has been shown to be involved in ceramide-dependent cell death and autophagy, while its physiological role lies in lysosomal nutrient signaling (3485). LAPTM4B is a 4-TMH membrane protein. According to the HPA, it is most highly expressed in photoreceptor cells and at moderate levels in most tissues.

Ceramide-mediated regulation of LAPTM4B is enabled by a sphingolipid interaction motif in TMH3 of LAPTM4B, thereby controlling the internalization of amino acid transporters (3485). LAPTM4B was shown to promote the recruitment of the amino acid transporter 4F2hc (SLC3A2)/LAT1 (SLC7A5) to lysosomes, thereby enhancing lysosomal uptake of leucine and other essential amino acids (see the SLC7A5 description in Section 9). This activates the V-ATPase, and hence mTORC1 activation (via the Ragulator, Rag GTPases, and Rheb-GTP) (18). Ceramide has been shown to induce the internalization of several nutrient transporters, including 4F2hc (SLC3A2) (3486, 3487).

LAPTM4B is frequently overexpressed in acute myeloid leukemia (AML), and high expression levels of LAPTM4B correlate with a poor prognosis. LAPTM4B has been shown to interact with ribosomal protein S9 (RPS9), which stabilizes it. Stabilized RPS9, in turn, activates the STAT3 signaling pathway, a well-known driver of cell survival, proliferation, and resistance to apoptosis. Knocking down LAPTM4B has been demonstrated to disrupt the RPS9/STAT3 axis, leading to inhibited AML cell growth (3488). It has been concluded that LAPTM4B acts as a potent oncogenic driver in AML by stabilizing RPS9 and activating STAT3-mediated transcription, which promotes leukemia cell proliferation. This highlights the broad versatility of LAPTM4B, as it can act beyond its lysosomal function by playing a pro-oncogenic role in AML.

LAPTM5

LAPTM5 is a 4-TMH membrane protein that is primarily expressed in hematopoietic cells, particularly B cells. It localizes to late endosomes and lysosomes, using PY/NEDD4 motifs to mediate trafficking via the ESCRT machinery (3489). LAPTM5 plays an important role in controlling protein degradation and immune signaling. It helps internalize and degrade B-cell receptors (BCRs), particularly those that might react strongly to “self-antigens”. It mediates immature B cell apoptosis and B cell tolerance by regulating the WWP2-PTEN-AKT pathway in the following manner: WWP2, an NEDD4 family E3 ubiquitin ligase, tags other proteins for destruction and one of its targets is the tumor suppressor protein PTEN. PTEN normally inhibits the AKT signaling pathway, which promotes cell survival and growth. LAPTM5 degrades WWP2, leading to higher PTEN levels. This results in less AKT activity and more cell death, particularly of “dangerous” immature B cells. Dysfunction of this pathway can contribute to autoimmunity due to the defective elimination of self-reactive B cells (3489).

DISP1, DISP2, DISP3, SCAP and HMGCR

The DISP1-3, SCAP and HMGCR proteins are part of the Resistance-Nodulation-Cell Division (RND) superfamily (TC 2.A.6) and are related to the SLC65 NPC-type cholesterol transporter family (see SLC65 description in Section 9).

As described earlier, the PTCH and PTCHD proteins, sharing about 20% sequence identity with NPC proteins, were classified as SLC65 members based on the finding that PTCH1 mediates cholesterol transport (3232). However, the transport function of DISP, SCAP and HMGCR is less well established, and they are more distantly related to NPC proteins than the PTCH and PTCHD proteins. Therefore, they have not been classified in the SLC65 family, although AlphaFold prediction revealed structural similarity between PTCHD1, NPC1, PTCH1 and DISP1 (3220). Here is a brief summary of the characteristics of these additional SLC-like RND superfamily members:

DISP1

As detailed in the description of the SLC65 family (Section 9), DISP1 facilitates SHH secretion in SHH-producing cells as part of the Hh pathway (Fig. 56). According to the HPA, DISP1 is ubiquitously expressed, with particularly high expression in oligodendrocytes and microglia.

DISP2

According to the HPA, DISP2 is most highly expressed in the brain (cerebellum/neurons), retina (horizontal cells), and enteroendocrine cells. Consistent with this, DISP2 has been reported to be widely localized in neuronal cells and the enteric nervous system (ENS) (3490). DISP2 was identified as a risk locus associated with lung cancer (3491) and diverticular disease (3492). In contrast to DISP1, only limited data are available on the functionality of DISP2.

DISP3

DISP3 has been shown to maintain a progenitor phenotype in neural cells, and the level of DISP3 expression has been shown to influence their cell fate (3493). Furthermore, DISP3 expression is influenced by thyroid hormones. This may link thyroid activity to cholesterol metabolism in the brain (3494).

SCAP

SREBP cleavage activating protein (SCAP) is required for liver lipid synthesis in response to cholesterol deficiency and has sterol-sensing domain (SSD) (3211–3213) (Fig. 55). SCAP is an integral membrane protein located in the endoplasmic reticulum (ER).

HMGCR

3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA reductase; HMGCR; TC 2.A.6.6.5) is a rate-limiting enzyme for cholesterol synthesis (3214). HMG-CoA reductase is anchored in the membrane of the ER and contains an SSD (Fig. 55).

XK, XKR3 (XTES), XKR4, XKR5, XKR6, XKR7, XKR8, XKR9 and XKRX

The XK-related family of proteins are members of the KX Blood-group Antigen (KXA) family (TC 2.A.112). Two membrane proteins express the antigens that make up the Kell blood group system: XK, a 440 amino acid residue protein with 20 TMHs that bears the KX antigen and has characteristics of a membrane transport protein; and Kell, a 93-kDa single-pass type II membrane glycoprotein that bears over twenty antigens (3495).

XK was initially thought to be an Na⁺-dependent neutral amine and/or oligopeptide transporter (3496), but was later discovered to be a scramblase (see below). Absence of the XK protein results in the McLeod phenotype, which is associated with abnormal, spiculated red blood cells called acanthocytes, late-onset muscular dystrophy, and nerve abnormalities (3495). XK was subsequently shown to function as a scramblase in ATP-induced phosphatidylserine exposure in T cells. XK interacts with VPS13A, a lipid transfer protein implicated in necrotic signaling and neuroacanthocytosis, a neurological disease characterized by movement abnormalities and the presence of acanthocytes (3497, 3498).

XKR3, also called XTES, is exclusively expressed in testis. It has 31% amino acid sequence identity to the XK protein and is predicted to have a similar topology to XK (3499). According to the HPA, it is expressed almost exclusively in early spermatids. Germline mutations in XKR3 were found in pituitary adenoma associated with vestibular schwannoma and it was speculated that XKR3 may be a genetic predisposition factor (3500).

XKR4 is expressed specifically in neurons, especially inhibitory neurons, and in oligodendrocytes; cerebral cortex; some lower expression in colon (glandular cells), according to the HPA.

During apoptosis, phosphatidylserine, which is normally restricted to the inner leaflet of the plasma membrane, is exposed on the surface of apoptotic cells and serves as an “eat me” signal to induce phagocytosis. XKR8 has been shown to mediate phosphatidylserine exposure in response to apoptotic stimuli, and cells from Xkr8^-/- mice failed to expose phosphatidylserine during apoptosis and were inefficiently engulfed by phagocytes (3501). Thus, XKR8 appears to act as a scramblase that promotes the exposure of phosphatidylserine on the surface of dying mammalian cells.

The XKR family members are generally expressed in the plasma membrane of cells (3502). Therefore, XKR8-deficient cells, which do not release phosphatidylserine during apoptosis, were transformed with XKR family members. Transformants expressing XKR4, XKR8 or XKR9 responded to apoptotic stimuli by releasing cell surface phosphatidylserine and were efficiently engulfed by macrophages (3502). XKR8, XKR4, and XKR9 were shown to have a caspase-recognition site in the C-terminal region and to require its direct cleavage by caspases for their function. Site-directed mutagenesis studies of these XKR family members identified essential residues in the second TMH and second cytoplasmic regions. It was also shown that in contrast to XKR8, which is ubiquitously expressed, the expression of XKR4 and XKR9 is tissue specific (3502).

Genetic variants in XKR6 were found to be associated with childhood-onset systemic lupus erythematosus in a Korean cohort (3503), elevated serum total cholesterol levels with increased risk of coronary heart disease and ischemic stroke (3504), and neuroticism (3505).

Further structural and functional insights have clarified the physiological roles of members of the KX blood group antigen family, particularly XKR8, XKR9, and XKR4, highlighting their function as lipid scramblases:

• The tertiary structure of XKR8 in complex with basigin (BSG/CD147) and phospholipids in the transmembrane cavity was reported using a combination of cryo-EM and X-ray crystallography (3506). This structure provides insight into the molecular mechanisms underlying phospholipid scrambling.

  Interestingly, BSG is also associated with members of the SLC16 family (i.e., MCT1-4 and MCT11), acting as an ancillary glycoprotein that supports the expression of MCTs in the plasma membrane (see the description of the SLC16 family). Similarly to MCTs, BSG also escorts XKR8 to the plasma membrane, where XKR8 exposes phosphatidylserine during apoptosis (3507). BSG plays a role in various pathophysiological processes, including cancer progression, inflammation, and viral infections (3508). Whether there is coordinated expression of BSG, SLC16 family members, and XKR8 that reflects parallel roles in metabolic adaptation and apoptotic signaling remains to be determined.

• The cryo-EM structure of XKR9 provides functional confirmation of its role as a caspase-activated protein involved in apoptotic lipid scrambling (3509).

• Human XKR4 is an active phospholipid scramblase involved in apoptosis that facilitates bidirectional lipid movement across the membrane. It uses a unique arrangement of conserved acidic residues to thin the membrane locally. This enables efficient lipid movement and the exposure of phosphatidylserine on the cell surface during apoptosis (3510).

• The study of the XK-VPS13A pathway revealed the scrambling mechanism triggered by extracellular ATP in immune cells (3511). In brief, when cells are damaged or infected, they release ATP into the extracellular space, activating P2X7 ATP-gated ion channels. The resulting P2X7 signaling activates the XK scramblase at the plasma membrane. VPS13A (also located at the plasma membrane) supplies lipids that work with XK to transfer and flip lipids, such as phosphatidylserine, to the outer surface of the membrane. The exposure of phosphatidylserine represents the “eat me” signal, marking programmed cell death.

Thus, while motifs hint at transporter ancestry in these KX blood-group antigen family proteins, current evidence supports phospholipid flipping within membranes, which is a type of transport activity. Several family members still remain uncharacterized.

TMEM205

The TMEM205 protein belongs to the TMEM205 family (TC 9.A.55). TMEM205 is a relatively small membrane protein (189 amino acids) with 4 TMHs, according to UniProt annotations. The AlphaFold structure of TMEM205 is shown in Fig. 2.

TMEM205 is associated with cisplatin resistance (3512). It is mainly expressed in the liver, pancreas, and adrenal glands, and it is located at the plasma membrane (3512). TMEM205-mediated resistance is selective for platinum-based drugs such as cisplatin and oxaliplatin but not carboplatin (3513). Additionally, TMEM205 has been found to contribute to platinum resistance development in ovarian cancer through the exosomal efflux of platinum drugs. This finding paves the way for preclinical studies of TMEM205/exosome-targeted therapies (3514).

Although structural information about the protein is unavailable, mutagenesis studies of TMEM205 revealed that mutations in sulfur-containing residues, particularly in TMH2 and TMH4, diminish the effect of cisplatin resistance (3513).

TMEM205 drives tumor-associated macrophages toward an M2-like (immunosuppressive) phenotype, contributing to a pro-tumor environment. Knockdown of TMEM205 makes tumor cells more sensitive to cisplatin, highlighting its potential as a therapeutic target (3515).

Thus, although the natural substrate of TMEM205 remains unknown, its behavior suggests a role in exosomal trafficking, detoxification, and the stress response. In this context, platinum drug export represents a pharmacological exploitation of its native function.

11. Outlook and future perspectives

In this review, we present a comprehensive outline of the approved families of 464 SLCs from the SLC-ome, grouped into 76 SLC families. In addition, we highlight 52 SLC-like proteins with interesting properties, bringing the total number of validated and putative membrane transport proteins to 516. Our aim was to offer up-to-date information that supports future research into the physiological roles of SLCs and their potential as therapeutic targets. With the SLC annotation – conducted in collaboration with the HGNC – now nearly complete, this guide to the SLC-ome serves as a timely roadmap for advancing discovery in the field.

The review presents the molecular, structural, physiological, clinical and pharmacological aspects of all members of the 76 SLC families. In addition, it highlights synergies between transporters from different SLC families. Key examples include:

• 1)the glutamate-glutamine cycle, which maintains adequate supplies of neurotransmitter glutamate in the CNS (Fig. 6);
• 2)the recycling of vitamin C in astrocytes via SLC2A1, SLC2A3 and SLC23A2, ensuring high neuronal concentrations needed for antioxidant defense and as cofactor for a variety of enzymes (Fig. 8);
• 3)the astrocyte-neuron lactate shuttle, involving SLC2A1, SLC16A1, SLC16A3, and SLC16A7, to fuel neuronal activity;
• 4)the intestinal absorption of iron via SLC11A2 and SLC40A1 (Fig. 22), and the recycling of iron in macrophages after erythrophagocytosis, mediated by SLC48A1, SLC11A2 and SLC40A1, for systemic iron homeostasis;
• 5)renal acid-base regulation through SLC4 and SLC42 family members to maintain systemic acid-base balance (Fig. 12);
• 6)the enterohepatic cycling of bile acids, via SLC10A2, SLC51A and SLC51B, to conserve and route the bile acid pool between the intestinal and hepatobiliary compartments (Fig. 33);
• 7)the lysosomal cholesterol sensing, involving the SLC-like sensor GPR155 and the cholesterol transporter NPC1 (SLC65A1), which regulates mTORC1 activity and prevents hyperactivation and neurodegeneration (Fig. 45);
• 8)the lysosomal recruitment of the amino acid transporter heterodimer LAT1-4F2hc (SLC7A5-SLC3A2) by the SLC-like protein LAPTM4B, which enhances mTORC1 activation and anabolic signaling under nutrient-rich conditions (Fig. 9C).

The review provides numerous examples of how SLCs can be exploited for drug delivery or as direct drug targets in diseases such as cancer, diabetes, immunological disorders, and neurological conditions. SLCs have proven to be highly druggable targets (85, 92), and many new therapeutic applications are expected to follow. Continued advances in drug discovery, including innovative high-throughput screening assays (3516), three-dimensional SLC structures that enable computational modeling and virtual compound screening (88, 92), and insights from clinical genetics that help prioritize disease-relevant transporters, are creating powerful opportunities to unlock the therapeutic potential of the SLC-ome. Complementing these approaches, next-generation 3D organoid systems (3517), refined to better reproduce the complexity of human organs, promise to revolutionize preclinical research by enabling realistic testing of hit and lead compounds. This wealth of new information will also facilitate the development of new SLC-mediated drug-delivery strategies, such as the classical prodrug strategy and the transporter-targeted nanocarrier formulation strategy (43).

Approximately 24% of the membrane transporters in the SLC-ome remain functionally uncharacterized (orphan transporters). Uncovering their function and physiological roles is expected to unlock additional new therapeutic opportunities for treating human diseases. The complete deorphanization of SLCs therefore represents a key challenge for the near future.

A major hurdle is that many orphan transporters are localized to intracellular compartments such as mitochondria, lysosomes, and the ER-Golgi network, making them less accessible to standard membrane transport assays. However, techniques such as binding and transport assays in crude membrane extracts or isolated organellar membranes, for example, using microscale thermophoresis or solid-supported membrane electrophysiology, offer promising alternatives for probing their function (1505, 3518).

Additional cutting-edge tools are also advancing this field. A comprehensive review of the current cell-based SLC assay platforms highlights innovative approaches for functional characterization and drug discovery (3519). In addition, as with earlier discoveries, linking genetic variants of orphan SLCs to human diseases will continue to be a powerful strategy for elucidating their physiological roles.

Several recent resources provide roadmaps for future SLC research and deorphanization, including:

• the SLC interactome (1575)

• a protein binding toolbox (3520)

• an SLC metabolic map (3521)

• an SLC genetic interaction map (3522)

• a functionally annotate SLC landscape (3523)

Once SLC transporters are fully characterized, it should be possible to generate steady-state and dynamic flux models of transport substrates and drugs for specific metabolic pathways. In terms of steady-state modeling, Reactome (https://reactome.org/) is a versatile open resource with curated pathways that include many SLC transporters. Reactome also allows users to overlay gene expression, protein-protein interactions, and drug/chemical interaction data onto these pathways. This feature provides insight into dynamic regulation and functional relationships (3524). Similarly, KEGG (Kyoto Encyclopedia of Genes and Genomes) is a comprehensive biological database that integrates genomic, chemical, and functional information and includes SLC transport systems (3525). These resources currently have two limitations. First, they focus on qualitative pathway topology, i.e., how SLCs interact with each other, without incorporating quantitative information, such as the concentration of molecules, reaction rates, kinetics, compartmental fluxes, or gradients and tissue-specific dynamics. Therefore, in terms of membrane transport, these resources define what happens and how components are connected (the roadmap) but not how much, how fast, or under what conditions.

Although a comprehensive, dynamic platform focusing entirely on SLC transporters does not yet exist, several initiatives are moving in this direction. Examples include the Physiome Project of the International Union of Physiological Sciences (https://physiomeproject.org/) and the Virtual Physiological Human, a European initiative (3526). One example of a Physiome project is the energy-based bond graph models of glucose transport via SLC transporters (3527). These models use bond graphs to dynamically represent glucose transport via SLC transporters while taking into account mass, charge, and energy conservation. This study addresses a critical gap by demonstrating that transporter-mediated flux can be modeled dynamically and accurately by integrating physical laws with biological data. These models are expected to underpin physiological systems at a high level in the future.

In conclusion, the information presented here for the largest group of membrane transporters encoded by the human genome, combined with the recent efforts and contributions mentioned above, provide a strong momentum for future progress in this field. It is clear that collaborative efforts from international, interdisciplinary networks such as ReSOLUTE (https://re-solute.eu/) (3528) and NCCR-TransCure (https://www.ibmm.unibe.ch/about_us/nccr_transcure_2010_2022/index_eng.html) (3529) hold significant promise for advancing our understanding of SLC transporters. Looking ahead, new research networks and initiatives will be essential to fully uncover the physiological roles of all SLCs and to realize the untapped therapeutic potential of the human SLC-ome.

Clinical Highlights.

This review aims to enhance our understanding of the physiological, clinical and pharmaceutical roles of human SLC transporters, the largest class of membrane transport proteins. SLC mediate the transmembrane transport of essential compounds, including amino acids, sugars, vitamins, trace minerals, lipids, neurotransmitters, metabolites, and numerous drugs. Dysfunction in SLC transporters contributes to a wide range of human diseases, such as Alzheimer disease, Parkinson disease, schizophrenia, cancer, immune disorders, metabolic and cardiovascular diseases, as well as numerous rare genetic conditions. Many of these disease associations are discussed in detail in this review.

Recent collaborations between academia and the pharmaceutical industry have greatly advanced the discovery of clinical applications for SLC membrane transporters, firmly establishing them as a druggable class of proteins. Several SLCs are currently targeted by approved drugs, for example the glucose transporter SGLT2 (SLC5A2) by antidiabetic gliflozins to reduce renal glucose reabsorption, the serotonin transporter SERT (SLC6A4) by selective serotonin reuptake inhibitors (SSRIs) for the treatment of depression, and the GABA transporter GAT1 (SLC6A1) by the anticonvulsant tiagabine for the treatment of epilepsy. As highlighted throughout this review, many additional SLCs are under investigation as potential targets for a broad range of diseases, opening exciting possibilities for treatment strategies.

Beyond serving as drug targets, many SLCs influence the pharmacokinetics of approved drugs and investigational compounds by modulating ADMET processes (absorption, distribution, metabolism, excretion, and toxicity). Their function and abundance can be altered by factors such as genetic variation, ethnicity, age, sex, diet, and concomitant medications (41). These considerations make SLC status highly relevant to personalized medicine, both for drug selection and dosing optimization, with the goal of maximizing efficacy and minimizing toxicity. SLCs are also emerging as promising targets in precision oncology (42).

In addition, SLCs can be harnessed for drug delivery across barriers such as the small intestine and the blood-brain barrier, or into specific tissues. In the prodrug approach, transporter substrates are covalently linked to drug molecules via cleavable bonds, facilitating transport across barriers and into specific tissues. More recently, nanoparticle-based strategies have been developed, in which nanoparticles carrying therapeutic payloads are functionalized with SLC substrates to direct delivery to specific sites, such as tumors, thereby enhancing efficacy and reducing off-target effects (43).

Notably, approximately one quarter of human SLCs are “orphan” transporters, with unknown substrates. This unexplored subset of the SLC-ome represents a significant reservoir of untapped therapeutic potential. By compiling comprehensive and up-to-date information on all known SLCs, this review aims to accelerate efforts to harness both characterized and orphan SLCs as attractive drug targets for drug development and delivery.

Synopsis.

Why SLCs matter

The coordinated movement of solutes that build cellular structures, sustain organelles, and drive essential physiological processes is fundamental to life. Solute carriers (SLCs), the largest group of membrane transport proteins encoded by the human genome – collectively called the SLC-ome – direct the regulated transport of these solutes across biological membranes. Examples of transported solutes include nutrients, metabolites, ions, lipids, and many other physiologically important compounds, as well as a wide range of pharmacologically active drugs. Through these roles, SLCs influence virtually every aspect of cellular and systemic physiology and determine the pharmacological actions of many therapeutic agents.

The SLC-ome currently comprises nearly 500 genes grouped into 76 families, each defined by shared sequence features, conserved structural folds, and related transport mechanisms. These transporters are expressed across all tissues and organ systems, where they are involved in nearly all vital processes, including nutrient uptake, ion homeostasis, development and growth, intracellular signaling, metabolite exchange, immune defense, and nervous system function such as neurotransmission. In addition, about 50 additional proteins have been identified that exhibit SLC-like features but have not yet been fully characterized or officially included in the SLC classification, representing a rich opportunity for future discovery.

SLCs are also central to clinical medicine. Variations in SLC genes underlie a wide spectrum of inherited disorders, while many SLCs determine drug absorption, distribution, and/or elimination, making them critical players in both pharmacology and therapeutics. Despite their ubiquity and relevance, many SLCs remain less well studied than proteins encoded by other gene families. Addressing this gap in our knowledge will require integrated, cross-disciplinary research efforts to unlock the full biological and therapeutic potential of the SLC-ome.

What this review provides

This review maps the SLC-ome from its molecular discovery to structural, physiological, and clinical insights, providing a framework that connects past progress to future opportunities. It introduces all SLC and SLC-like families (see Sections 9 and 10), outlines their phylogenetic relationships, and highlights the common structural folds that underlie their diverse transport mechanisms (Section 8). Readers will gain in-depth insights into the roles of SLCs within cellular networks and whole-body physiology, and their impact on pharmacology and therapeutic outcomes.

The systematic identification of SLCs spans multiple decades, beginning with molecular cloning of individual transporters in the 1980s. All SLC families have since been assigned names within the SLC nomenclature system, coordinated by the HUGO Gene Nomenclature Committee (HGNC). Readers interested in the historical trajectory and classification principles will find these summarized in Sections 4 and 5 of the full review.

Beyond foundational knowledge, the review emphasizes the translational potential of SLC research. It highlights current and emerging pharmaceutical applications and demonstrates how an integrated understanding of the SLC-ome can guide drug discovery, disease modeling, and precision medicine. Numerous SLCs are already established as drug targets or determinants of drug disposition (see Section 3). For example, neurotransmitter transporters underlie the efficacy of antidepressants, antiepileptics, and psychostimulants; renal transporters control uricosuric drug action; and glucose transporters are central to diabetes therapy and cancer metabolism. Yet, many SLCs and SLC-like transporters remain underexplored, representing an untapped reservoir of pharmacological opportunity. This promise extends beyond orphan SLCs, as even well-studied transporters continue to reveal new physiological roles, regulatory mechanisms, and unexpected substrate repertoires. By covering the spectrum from molecular identification to therapeutic opportunities, this review is intended to serve as a broad reference for researchers, clinicians, and pharmacologists, supporting both fundamental understanding and translational exploration of the SLC-ome.

Navigating the full review

1) Overview and Early-Career Readers: We recommend that non-specialized or early-career readers begin with the overview sections (1–8) to grasp key concepts, particularly on different types of membrane transport proteins in the SLC-ome (Section 1), the SLC nomenclature system (Section 5), orphan transporters (Section 6), SLC-like proteins (Sections 6), and the structural folds underlying diverse transport mechanisms (Sections 7 and 8). Readers can then proceed to the descriptions of individual SLC and SLC-like families (Sections 9 and 10) to gain detailed, system-level insights.

2) Researchers with disease- or pharmaceutical-focused interests: Sections 9 and 10 provide extensive information on clinical relevance, human genetics, disease associations, and therapeutic applications for individual SLC family members. Those developing transporter assays or models may also find Section 4, as well as the “Discovery” descriptions in Section 9, useful.

3) Clinicians: Disease-specific text searches are the easiest way to find relevant sections. The following disease areas are frequently discussed in this review in the context of SLC involvement, yet many additional conditions, including a myriad of rare diseases, are also covered.

• Cancers: Lung, breast, colorectal, gastric, liver (including hepatocellular carcinoma), thyroid, and prostate cancers.

• Neurological disorders: Alzheimer and Parkinson diseases, epilepsy, schizophrenia, amyotrophic lateral sclerosis, and stroke.

• Autoimmune and inflammatory diseases: Rheumatoid arthritis, systemic lupus erythematosus, inflammatory bowel disease (IBD, including Crohn’s disease), and type 1 diabetes (T1D).

• Cardiovascular conditions: Heart failure, coronary heart disease, cardiomyopathy, arrhythmias, and other cardiovascular diseases.

• Metabolic and multisystem disorders: Type 2 diabetes (T2D), metabolic syndrome, hypertension, obesity, non-alcoholic fatty liver disease (NAFLD), chronic obstructive pulmonary disease (COPD), hemochromatosis, lysosomal storage diseases (LSD, including Niemann-Pick disease type C, NPC), congenital disorders of glycosylation (CDG), gout, nephrolithiasis (including cystinuria), retinitis pigmentosa, hereditary deafness, glycogen storage diseases (GSD), osteoporosis, and rickets.

4) Structural, computational, and comparative/phylogenetic biologists: Section 8 provides a detailed discussion of fold families and structure-function paradigms, with additional relevant information in Section 9. Figure 3 in Section 8 presents a circular dendrogram showing hierarchical clustering of the human SLC families.

Translational opportunities and future directions - Despite remarkable progress nearly one-quarter of human SLCs are still orphan transporters, lacking firmly identified substrates, and many lack atomic-resolution structural information. Understanding transporter regulation, tissue-specific expression, and interaction networks is advancing but remains incomplete, limiting the full exploitation of SLCs in drug discovery and precision medicine. Our review highlights advances in understanding SLCs at multiple levels, from physiological and pathological studies that reveal their roles in health and disease, to functional genomics, structural breakthroughs, and emerging clinical applications. Cryo-EM discoveries have illuminated conformational cycles and transport mechanisms, providing detailed views of substrate binding and a foundation for structure-based drug design.

Looking forward, the integration of advanced computational modeling of target SLCs, virtual and high-throughput compound screening, and insights from clinical genetics will create new opportunities to unlock the therapeutic potential of the SLC-ome. Complementing these approaches, next-generation 3D organoid systems, refined to more faithfully recapitulate the complexity of specific human organs, promise to revolutionize preclinical research by enabling realistic testing of hit and lead compounds that target SLCs. Together, these approaches open the door to personalized and transformative therapies.

This review provides a comprehensive, integrated account of the SLC landscape and is intended to guide both early-career and experienced investigators in exploring how solute movement shapes cellular and systemic physiology, thereby paving the way for future discoveries and innovative therapies directed at the SLC-ome.

Glossary

Abbreviations and Glossary

ABC

transporter ATP-Binding Cassette transporter

AbgT

Prokaryotic p-aminobenzoyl-glutamate antimicrobial resistance transporter

ACE

Angiotensin Converting Enzyme

ADHD

Attention-Deficit/Hyperactivity Disorder

ADME

Drug absorption, distribution, metabolism and excretion

AE

Anion Exchanger

AKT signaling pathway

The AKT signaling pathway, also known as the PI3K-Akt pathway, promotes cell survival and growth in response to extracellular signals

Alpha MeGlc / αMeGlc

α-methyl-D-glucopyranoside, a non-metabolized analog of glucose that is transported by the Na⁺/glucose transporters SGLT1 (SLC5A1) and SGLT1 (SLC5A2)

AML

Acute Myeloid Leukemia (see SLC22A16, SLC63A3 and SLC-like LAPTM4B)

AMP

Adenosine Monophosphate

ANLS

Astrocyte-Neuron Lactate Shuttle

APC

Amino acid-Polyamine-organoCation superfamily

APS

Amino acid-Polyamine-organocation Superfamily

ARL15

ADP-ribosylation factor-like 15 is a protein in humans that is encoded by the ARL15 gene. It is a small GTP-binding protein, which is a member of the RAS superfamily of proteins (3530)

Ascorbate

Vitamin C / ascorbic acid

ATS

Arterial Tortuosity Syndrome

Aβ

Amyloid beta peptides; these are peptides of 36 to 43 amino acids that are the major component of amyloid plaques in the brains of people with Alzheimer disease

BBB

Blood-Brain Barrier

BCAA

Branched-Chain Amino Acid

BSG

Basigin, also known as extracellular matrix metalloproteinase inducer (EMMPRIN) or cluster of differentiation 147 (CD147) is a member of the immunoglobulin superfamily

BVVL

Brown-Vialetto-Van Laere syndrome, a rare, inherited neurodegenerative disorder characterized by progressive weakness and paralysis of the facial, tongue, and throat muscles (pontobulbar palsy), along with sensorineural hearing loss that most often manifests in infancy or early childhood. Since mutations in the SLC52A3 riboflavin transporter gene have been identified as the cause of BVVL, BVVL is now also known as riboflavin transporter deficiency (OMIM 614707)

CaM

Calcium-calmodulin

CaMKII

Ca²⁺/calmodulin-dependent protein kinase II

CBS domain

Cyclin and cystathionine β-synthase domain. The CBS domain is named after cystathionine beta-synthase (CBS), a protein in which it was first identified. CBS domains are found in a wide range of proteins across different species and functions, including members of the cyclin and CBS domain magnesium transport mediators (CNNM) and the ClC chloride channel/transporter family (347, 3312, 3314)

CDF

Cation Diffusion Facilitator

CDG

Congenital Disorder of Glycosylation

CDN

Cyclic Dinucleotide

CF

Cystic Fibrosis

CFTR

Cystic Fibrosis Transmembrane conductance Regulator (ABCC7)

CHP1

Calcineurin B-Homologous Protein 1

CHS

Chediak-Higashi Syndrome

Clan

See Pfam

CNS

Central Nervous System

CNT

Concentrative Nucleoside Transporter family

CoA

Coenzyme A

COPD

Chronic Obstructive Pulmonary Disease, a type of progressive lung disease characterized by chronic respiratory symptoms and airflow limitation (3531)

CPA1, CPA2

Cation:Proton Antiporter families 1 and 2

CRAC

Ca²⁺ release-activated Ca²⁺ channel

CRC

Colorectal Cancer

cryo-EM

cryo-electron microscopy

CSF

Cerebrospinal Fluid

C9orf72-SMCR8-WDR41 complex

The C9orf72-SMCR8-WDR41 complex is a heterotrimeric protein complex that plays a crucial role in lysosomal function, particularly in response to amino acid scarcity. This complex is recruited to lysosomes, where it interacts with the cationic amino acid transporter PQLC2 (3260)

DAACS

Dicarboxylate/Amino Acid: Cation (Na⁺ or H⁺) Symporter

DASS

Divalent Anion: Na+ Symporter

DCs

Dendritic Cells

DDIs

Drug-Drug Interactions

DHA

Dehydroascorbic Acid, an oxidized form of vitamin C (ascorbic acid)

DHEAS

Dehydroepiandrosterone Sulfate

DMT

superfamily Drug/Metabolite Transporter superfamily

DMT1

Divalent Metal ion Transporter 1

Dolichol phosphate

Dolichol phosphate is a polyisoprene lipid that serves as a crucial carrier for sugars during protein glycosylation

DUSP1

Dual-Specificity Phosphatase 1, also known as MAPK phosphatase-1 (MKP-1), is an enzyme that dephosphorylates and inactivates members of the MAPK family, especially JNKs, p38, and ERK1/2. Studies with vascular smooth muscle cells (VSMCs) have revealed that the induction of DUSP1 decreases ERK activity, resulting in decreased cell proliferation, which depends on specific heparin binding (3021)

ECD

Extended Extracellular Ectodomain

EGFR

Epidermal Growth Factor Receptor. EGFR and ERBB2 (Epidermal Growth Factor Receptor 2; see below) are both members of the ErbB family of receptors

EPO

Erythropoietin

ERBB2

Epidermal Growth Factor Receptor 2 gene. The human protein is referred to as HER2 (Human Epidermal Growth Factor Receptor 2)

ESCC

Esophageal Squamous Cell Carcinoma ESCRT machinery Endosomal Sorting Complex Required for Transport multi-protein machinery, a group of protein complexes that play a crucial role in various cellular processes. An important function of ESCRT is to sort proteins within endosomes, specifically targeting ubiquitinated proteins for degradation in lysosomes (3532). It facilitates the formation of intralumenal vesicles (ILVs) inside late endosomes. The ESCRT system consists of four main complexes: ESCRT-0, ESCRT-I, ESCRT-II, and ESCRT-III, along with accessory proteins like VPS4. These components work sequentially to carry out their functions

EST

Expressed Sequence Tag

FABPs

Fatty Acid Binding Proteins

FAD

Flavin Adenine Dinucleotide

FDA

U.S. Food and Drug Administration

FLS

Fazio-Londe syndrome, a rare, inherited motor neuron disease characterized by progressive paralysis of muscles innervated by cranial nerves

FMH

Flavin Mononucleotide

FPN

Ferroportin

FRDA

Friedreich Ataxia

FSASDs

Free Sialic Acid Storage Disorders, caused by SLC17A5 dysfunction

GABA

γ-aminobutyric acid

GAP

GTPase-activating protein

GI

Gastrointestinal

GLP-1

Glucagon-Like Peptide-1

GLP1R

Glucagon-Like Peptide-1 Receptor

GLUT1DS

GLUT1 Deficiency Syndrome

GPR155

G-protein coupled receptor 155

GSH

Glutathione

GWAS

Genome-Wide Association Atudies

HCC

Hepatocellular Carcinoma

HCO₃^-

Bicarbonate

HH signaling

Hedgehog signaling. The Hh signaling pathway exerts its biological effects through a complex signaling cascade that plays important roles in embryonic development and tissue homeostasis, in regulating the proliferation and differentiation of adult stem cells, and is associated with increased cancer prevalence, malignant progression and poor prognosis (3239)

HGNC

HUGO Gene Nomenclature Committee (https://www.genenames.org)

HIF2α

Hepatic Hypoxia-Inducible Factor

HMM

Hidden Markov Model: HMM techniques are used to model families of biological sequences (3533, 3534)

HNF1α

Hepatocyte Nuclear Factor 1 alpha

HPA

Human Protein Atlas (https://www.proteinatlas.org/)

HPS

Hermansky-Pudlak syndrome

HSPGs

Heparan Sulfate Proteoglycans

HSV-1

Herpes Simplex Virus 1

IBD

Inflammatory Bowel Disease

IFN-γ

Interferon-gamma, a type II interferon and a crucial cytokine in immune responses

iNOS

inducible isoform of Nitric Oxide Synthase

InsP8

1,5-Bis-diphosphoinositol 2,3,4,6-tetrakisphosphate, a high-affinity ligand for the SYG1/Pho81/XPR1 (SPX) domain of XPR1 (SLC53A1)

InsPs

Inositol polyphosphates regulate diverse cellular processes in eukaryotic cells by interacting with specific protein targets (see XPR1/SLC53A1 description). InsP8 is an inositol pyrophosphate (PP-InsP), a class of compounds that represent a subset of InsPs (3062)

IRBIT

Inositol 1,4,5-trisphosphate receptor-binding protein released with inositol 1,4,5-trisphosphate interacts with a wide array of proteins with distinct biological functions (3535); long-IRBIT is an IRBIT homolog with a unique N-terminal appendage

IRE

Iron Responsive Element, a short stem loop in the untranslated regions of certain mRNAs encoding proteins involved in iron metabolism. It binds iron-responsive proteins (IRPs) in an iron-dependent manner

IRP

Iron Regulatory Protein. Systemic iron homeostasis is maintained by the iron regulatory proteins IRP1 and IRP2 and iron-responsive element (IRE) signaling pathways (see “IRE” for more information)

Isoform

A protein isoform (or variant) is the result of alternative splicing of a single gene or other post-transcriptional modifications. See also “paralog” and “ortholog”

ITC

International Transporter Consortium. This consortium was established to identify a subset of transporters of particular clinical interest and to outline decision trees that should be used to predict the clinical significance of changes in transporter activity (1852) (see: https://www.itc-transporter.org/)

ITGB3

Integrin beta 3, a cell adhesion receptor that binds extracellular matrix proteins like vitronectin, fibronectin, and fibrinogen

KEAP1

Kelch-like ECH-Associated Protein 1. KEAP1 acts as a negative regulator of the transcription factor NRF2. NRF2 activation leads to an antioxidant and anti-inflammatory response. KEAP1 represses NRF2 activation by degrading it when cellular conditions are normal, to prevent excessive antioxidant responses

K_m

Michaelis-Menten constant, the substrate concentration required to achieve half the maximum transport rate (V_max)

LCFA

Long-Chain Fatty Acids

L-Cys

L-cysteine

LDL

Low-Density Lipoprotein

LDLR

Low-Density Lipoprotein Receptor

LeuT

Bacterial leucine transporter

LeuT fold

Paradigm structure for APC superfamily members; identified in the crystal structure of the bacterial Na⁺-coupled amino acid importer LeuT (212)

Lp(a)

Lipoprotein(a), a liver-derived plasma lipoprotein structurally similar to low-density lipoprotein (LDL), consisting of an LDL-like particle bound via a disulfide bond to a unique glycoprotein called apolipoprotein(a) [apo(a)]. Elevated Lp(a) levels are a genetically determined risk factor for atherosclerotic cardiovascular disease and calcific aortic valve stenosis (3156)

lncRNA

Long non-coding RNA. Unlike mRNA, which translates into proteins, lncRNAs primarily regulate gene expression at various levels, including transcriptional, post-transcriptional, and post-translational. They play crucial roles in biological processes like chromatin remodeling, RNA processing, and mRNA translation (3536)

LPC

Lysophosphatidylcholine

LPE

Lysophosphatidylethanolamine

LPI

Lysinuric Protein Intolerance

LPS

Lysophosphatidylserine

MagT1

Magnesium transporter family 1

MATE transporter

Multidrug And Toxin Extrusion transporter

MCF

Mitochondrial Carrier Family

MDP

muramyl dipeptide

MeAIB

2-methylamino-isobutyric acid, a specific substrate for amino acid transport system A

MetS

Metabolic syndrome

MFS

Major Facilitator Superfamily

MFSD

Major Facilitator Superfamily Domain

mk mice

microcytic anemia mice

MPP⁺

MPP⁺ (1-methyl-4-phenylpyridinium), a neurotoxin. MPP⁺ selectively eliminates dopaminergic neurons in the pars compacta of the substantia nigra and induces parkinsonian symptoms in animals

MPTP

1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine, a potent neurotoxin commonly used in animal model of Parkinson disease. In the brain, MPTP is converted by monoamine oxidase B to MPP⁺, which induces parkinsonism by eliminating dopamine-producing neurons in the substantia nigra

mTOR

mammalian Target of Rapamycin, a protein kinase that functions via the multiprotein complexes mTORC1 and mTORC2. While mTORC1 regulates cell growth and metabolism, mTORC2 regulates cell proliferation and survival (2664)

mTORC1

mammalian target of Rapamycin complex 1, a protein complex that acts as a sensor of nutrient availability and overall cellular metabolic status, controlling protein synthesis when sufficient building blocks and energy are available. As such, it is primarily involved in regulating the balance between cellular growth and catabolism (3537)

NAAG

N-Acetylaspartylglutamate, a neuromodulator of glutamatergic synapses acting on NMDA receptors or metabotropic glutamate receptors

NADH/NAD⁺

Nicotinamide adenine dinucleotide (reduced and oxidized form, respectively)

NAFLD

Non-Alcoholic Fatty Liver Disease

NASH

Non-alcoholic steatohepatitis

NBCe1

electrogenic sodium bicarbonate cotransporter 1

NDCBE

Electroneutral transporter of the SLC4 family that exchanges sodium and bicarbonate for chloride

NEDD4

NEDD4 (neural precursor cell-expressed developmentally down-regulated protein 4) is an E3 ubiquitin ligase that functions in the ubiquitin-proteasome system by targeting proteins for ubiquitination and subsequent degradation

NFAT5

Nuclear Factor of Activated T cells 5, a transcription factor that plays an important role in cellular responses to osmotic stress, particularly in hypertonic conditions

NF-κB

A transcription factor that is activated by various intra- and extra-cellular stimuli

NhaA

Major Na⁺/H⁺ antiporter of E. coli. The NhaA structure represents a unique structural fold, the “NhaA fold”

NMDA receptor

N-methyl-D-aspartate receptor

NMN

Nicotinamide Mononucleotide

Norrin

A signaling molecule with structural and functional characteristics of an autocrine and/or paracrine acting growth factor (3538). Norrin interacts with both the CRD domain of Frizzled-4 receptors (FZD4) and ectodomain ECD of LRP5/6 by simulating the finger-like loop of Wnt, thereby activating the Wnt/β-catenin signaling pathway (3539, 3540)

NPC

Niemann-Pick C, a rare progressive genetic disorder characterized by the inability of the body to transport cholesterol

NRAMP

Natural Resistance-Associated Macrophage Protein; transporter of divalent metal ions (e.g., Mn²⁺, Fe²⁺) and member of the SLC11 family

NRF2

Nuclear factor erythroid 2-related factor 2, an antioxidant transcription factor. The Kelch-like ECH-associated protein 1 (KEAP1) acts as a negative regulator of NRF2

NRP1

Neuropilin-1, a non-tyrosine kinase receptor that acts as a co-receptor for several ligands, including VEGF (vascular endothelial growth factor) and TGF-β

NSAIDs

Nonsteroidal anti-inflammatory drugs

NSCLC

Non-Small Cell Lung Cancer, the most common type of lung cancer

NTBI

Non-Transferrin Bound Iron

NTCP

Na⁺-Taurocholate Cotransporting Polypeptide (sodium/bile acid cotransporter)

NTDs

Neural TubeDefects

OCD

Obsessive-Compulsive Disorder

Ortholog

Genes that are separated from each other by speciation are called orthologs, see also “paralog” and “isoform”

PAP

Adenosine 3’,5’-diphosphate

PAPS

3’-phosphoadenosine 5’-phosphosulfate

Paralog

Genes separated by gene duplication events are called paralogs, see also “ortholog” and “isoform”

PAT1

This symbol has been used for two phylogenetically unrelated SLCs: The H⁺-coupled lysosomal amino acid transporter PAT1/LYAAT1 (SLC36A1) and the anion exchanger PAT1 (SLC26A6), also known as CFEX. In the latter case, PAT1 and CFEX are alternative names for the same protein of the SLC26 family, where PAT1 stands for “putative anion transporter 1” and CFEX stands for “chloride/formate exchanger”

PC

Phosphatidylcholine

PDAC

Pancreatic Ductal Adenocarcinoma

PET

Positron Emission Tomography

Pfam

The Pfam database is a collection of multiple sequence alignments and hidden Markov models (HMMs) (3534) covering a large number of common protein domains. Pfam is located at EMBL-EBI, Hinxton, UK (http://pfam.xfam.org/, now hosted by InterPro (https://www.ebi.ac.uk/interpro/). Proteins are generally made up of one or more functional regions, commonly referred to as domains. Different combinations of domains give rise to the diversity of proteins found in nature. Identifying the domains that occur within proteins can therefore provide insight into their function. Pfam also generates higher-level groupings of related entries called clans. A clan is a collection of Pfam entries that are related by similarity of sequence, structure, or profile-HMM

Pfam curator

A Pfam curator reviews submissions for conformance to the Pfam data model

PFBC

Primary Familial Brain Calcifications

P_i

Inorganic phosphate

PtdIns(3,5)P₂

Phosphatidylinositol-3,5-bisphosphate [also known as PI(3,5)P2], an important signaling lipid that plays a crucial role in regulating the trafficking of intracellular organelles, especially within the endolysosomal system

PI3K/Akt

Phosphoinositide 3-kinase (PI3K)/Akt signaling pathway. PI3K activation phosphorylates and activates AKT. This localizes AKT in the plasma membrane. AKT, also known as protein kinase B (PKB), is a group of three serine/threonine-specific protein kinases that play key roles in cellular processes such as glucose metabolism, apoptosis, cell proliferation, transcription, and migration. The PI3K/AKT/mTOR pathway is an intracellular signaling pathway that plays a crucial role in regulating the cell cycle

PIP₂

Phosphatidylinositol 4,5-bisphosphate

PIP₃

Phosphatidyl inositol 3,4,5-trisphosphate

PLC

Phospholipase C

PPAR-γ

Peroxisome Proliferator-Activated Receptor gamma

PQ-loop

PQ-loop proteins contain two well conserved repeat sequences, the so-called PQ-loop motif, including a characteristic proline-glutamine dipeptide (3541, 3542)

PRL phosphatases

The human Phosphatase of Regenerative Liver (PRL) family of phosphatases comprises three members, PRL-1, PRL-2 and PRL-3. They play a role in various cellular processes, including cell signaling and metabolism (3543)

PTEN

Phosphatase and tensin homolog (PTEN) is a tumor suppressor gene that encodes a protein with dual phosphatase activity. It dephosphorylates phosphatidylinositol 3,4,5-trisphosphate (PIP3) to produce phosphatidylinositol 4,5-bisphosphate (PIP2) and also dephosphorylates various protein substrates. PTEN plays a crucial role in regulating cell growth, proliferation, and apoptosis by participating in various signaling pathways, including the PI3K/Akt pathway

PTH

Parathyroid Hormone, a peptide hormone produced by the parathyroid glands that plays a crucial role in regulating calcium and phosphorus levels in the blood

Rag GTPase

Rag GTPases are small GTPases that belong to the family of Ras-related GTPases and play an important role in regulating the mTOR complex 1 (mTORC1). Rags form heterodimers, whereby Rag A or Rag B bind to Rag C or Rag D to recruit mTORC1 to the lysosome where it becomes activated. The activity of Rags is regulated by amino acids. Amino acid depletion leads to hydrolysis of Rag A/B-bound GTP to GDP and GDP to GTP exchange at Rag C/D, resulting in dissociation of mTORC1 into the cytosol (3544)

RFT1

The symbol RFT1 has been used for phylogenetically unrelated SLCs: SLC52A1 and SLC19A1 have been published as RFT1 (riboflavin transporter 1) and RFT-1 (reduced folate transporter 1) respectively, but these are not related to the glycolipid translocator RFT1 (SLC76A1). To avoid confusion with the gene officially approved as RFT1 (SLC76A1), it is recommended to use the symbol SLC52A1 (or RFVT1) for the riboflavin transporter

RND superfamily

Resistance-Nodulation-cell Division superfamily

RPE

Retinal Pigment Epithelium

RBP

Retinol-Binding Protein, a protein that primarily functions as a carrier protein for retinol in the bloodstream

RUNX1

Runt-related transcription factor 1 regulates critical processes in many aspects of hematopoiesis (3545)

SCFAs

Short-Chain Fatty Acids

SGLT1

Sodium Glucose Transporter 1

SHH

Sonic hedgehog, a protein that acts as a signaling molecule during embryonic development in humans and other animals (3233, 3546)

SLC

SoLute Carrier (solute carrier, membrane transporter)

SLE

Systemic Lupus Erythematosus

SNP

Single Nucleotide Polymorphism

SOD

Superoxide Dismutase

SSD

Sterol-Sensing Domain

SSRI

Selective Serotonin Reuptake Inhibitor

STAS domain

Sulphate Transporter and Anti-Sigma factor antagonist, a cytoplasmic domain that plays an important role in SLC26 family transporter in regulating chloride transport across cell membranes

STAT3

Signal Transducer and Activator of Transcription 3, a transcription factor. In response to cytokines and growth factors, STAT3 is phosphorylated by receptor-associated Janus kinases. It then forms homo- or heterodimers and translocates to the cell nucleus, where it acts as a transcription activator (3547). The STAT3 signaling pathway is crucial for cell growth, cell survival, and immune responses. Its dysregulation causes numerous diseases, in particular cancer and autoimmune diseases

STING

Stimulator of Interferon Genes. STING (STING1) is an ER-associated adaptor protein that plays a central role in innate immunity. It is activated by cyclic dinucleotides (CDNs), which are produced by cyclic GMP-AMP (cGAMP) synthase (cGAS) upon the detection of cytosolic double-stranded DNA. Once activated, STING moves from the ER to the Golgi, triggering signaling cascades via the kinase TBK1. TBK1 then phosphorylates the transcription factor IRF3, resulting in the production of type I interferons and other inflammatory cytokines (3548)

SUCNR1

Succinate Receptor, also known as GPR91, a GPCR that is activated by succinate. It is involved in sensing local stress, such as ischemia, hypoxia, and metabolic imbalances (3549)

Superfamily

A large group of genes encoding proteins that contain superimposable 3D structures of common origin, such as the bacterial LeuT fold of the APC superfamily

SWEET family

“Sugars Will Eventually be Exported Transporter” family

T1D

Type 1 Diabetes

T2D

Type 2 Diabetes

TBZ

Tetrabenazine, a drug for the symptomatic treatment of hyperkinetic movement disorders

TC

Transporter Classification Database number (e,g, TC 2.A.69.3)

TCDB

Transporter Classification Database (https://www.tcdb.org)

TFEB

Transcription Factor EB coordinates expression of lysosomal hydrolases, membrane proteins and proteins involved in biogenesis and autophagy in response to sensing related to lysosomal stress and nutritional conditions (1653). EB stands for E box binding (TFEB is a specific type of transcription factor that binds to DNA sequences known as E-boxes, which are crucial for regulating gene expression)

TfR

Transferrin Receptor (TFRC) is a high affinity receptor ubiquitously expressed, whereas TfR2 (TFR2) is predominantly expressed in liver and erythroid precursor cells and not affected by intracellular iron concentration

TGF-β

Transforming Growth Factor beta (TGF-β), a multifunctional cytokine expressed by almost every tissue and cell type. Its signal transduction can stimulate diverse cellular responses, playing a critical role in embryonic development, wound healing, tissue homeostasis, and immune homeostasis (3550)

TGF-β/SMAD

When TGF-β binds to its receptor, it activates SMAD transcription factors, which translocate to the nucleus and turn on specific genes

TGN

Trans-Golgi Network

TLR

Toll-Like Receptor. TLRs are a class of proteins that play a crucial role in the innate immune system by recognizing pathogen-associated molecules derived from microbes such as lipopolysaccharide (LPS) from bacteria, double-stranded RNA from viruses, and various lipids, lipoproteins, and proteins found on microbial surfaces

TMEM

Transmembrane protein. The TMEM naming system was instigated by the HGNC for genes of unknown function, identified from large-scale genome sequencing projects, that were predicted to encode one or more transmembrane domains

TMH

Transmembrane Helix

TMP

Thiamine Monophosphate

TNF-α

Tumor Necrosis Factor alpha, a protein that plays a key role in the inflammatory response and immune system. It is produced by immune cells, such as macrophages and T-lymphocytes

TPP

Thiamine Pyrophosphate

TPP⁺

Tetraphenylphosphonium

TREM2

Triggering Receptor Expressed on Myeloid cells 2. TREM2 plays a crucial role in immune cell function, particularly in the brain. It is expressed on immune cells like microglia and macrophages and is involved in the regulation of inflammation, phagocytosis, and neuronal development, and loss of TREM2 function increase the risk of developing Alzheimer disease (2104)

TrkB

Tyrosine receptor kinase B

UDP-GlcNAc

UDP-N-acetyl-glucosamine

UDP-GlcUA

UDP-glucuronic acid

Uphill transport

Uphill transport of a solute species refers to membrane transport against its concentration gradient by coupling the flux with that of another species, such as Na⁺ or H⁺, in the same direction. This mode of transport is called cotransport. Classic examples are the Na⁺-glucose cotransporter SGLT1 (SLC5A1) and the H⁺-coupled oligopeptide transporter PepT1 (SLC15A1) of the small intestine

VEGF

Vascular Endothelial Growth Factor, a signaling protein that plays a crucial role in angiogenesis

Vegfr2

VEGF receptor

V_max

See K_m

VSMCs

Vascular Smooth Muscle Cells

VRAC

Volume-Regulated Anion Channel

WHS

Wolf-Hirschhorn syndrome

Wnt/β-catenin

The Wnt/β-catenin pathway is composed of a group of proteins that play essential roles in embryonic development and adult tissue homeostasis (3540). Norrin is a signaling molecule that activates the Wnt/β-catenin pathway by binding to the ectodomain (ECD) of the Frizzled-4 (FZD4) receptor and the LRP5/6 receptor (see “Norrin” for more information.)

WW domains

Small protein domains, approximately 40 amino acids in length, characterized by the presence of two conserved tryptophan (W) residues

YM155

Sepantronium bromide, is a small imidazolium-based proapoptotic agent with anti-tumor activity

YxxΦ and PY motifs

YxxΦ and PY motifs are crucial for regulating protein function, particularly in cellular processes involving protein-protein interactions. The YxxΦ motif is a well-characterized tyrosine-based sorting signal where “Y” stands for tyrosine, “x” represents any amino acid, and “Φ” is a bulky hydrophobic residue (like leucine, isoleucine, or phenylalanine). In LAPTM4A, this motif plays a crucial role in lysosomal targeting, as it is recognized by adaptor protein complexes (e.g., AP-1 or AP-3) that mediate trafficking to the endo-lysosomal system. The PY motif refers to a proline-tyrosine (P-Y) sequence, commonly found in proteins that interact with WW-domain–containing E3 ubiquitin ligases, such as NEDD4. In LAPTM4A, the PY motifs are involved in regulating protein-protein interactions and ubiquitin-mediated trafficking or degradation (3483)

ZIP

Zrt/Irt-like Protein (SLC39 zinc transporter family)

ω3-DHA

Omega-3 fatty acid docosahexaenoic acid. Note that this review uses the abbreviation ω3-DHA to distinguish the omega-3 fatty acid DHA from DHA, which stands for dehydroascorbic acid