Renal cell markers: lighthouses for managing renal diseases

Abstract

Kidneys, one of the vital organs in our body, are responsible for maintaining whole body homeostasis. The complexity of renal function (e.g., filtration, reabsorption, fluid and electrolyte regulation, and urine production) demands diversity not only at the level of cell types but also in their overall distribution and structural framework within the kidney. To gain an in depth molecular-level understanding of the renal system, it is imperative to discern the components of kidney and the types of cells residing in each of the subregions. Recent developments in labeling, tracing, and imaging techniques have enabled us to mark, monitor, and identify these cells in vivo with high efficiency in a minimally invasive manner. In this review, we summarize different cell types, specific markers that are uniquely associated with those cell types, and their distribution in the kidney, which altogether make kidneys so special and different. Cellular sorting based on the presence of certain proteins on the cell surface allowed for the assignment of multiple markers for each cell type. However, different studies using different techniques have found contradictions in cell type-specific markers. Thus, the term “cell marker” might be imprecise and suboptimal, leading to uncertainty when interpreting the data. Therefore, we strongly believe that there is an unmet need to define the best cell markers for a cell type. Although the compendium of renal-selective marker proteins presented in this review is a resource that may be useful to researchers, we acknowledge that the list may not be necessarily exhaustive.

INTRODUCTION

The human kidney is responsible for maintaining homeostasis in the body by regulating acid-base and electrolyte balance, controlling blood pressure, and excreting metabolic toxins and waste products. The nephron is the basic structural and functional unit of the kidney, which is composed of Bowman’s capsule, the glomerulus, tubules [proximal tubule (PT), loop of Henle, distal convoluted tubule (DCT), and collecting tubules], and the collecting duct (CD). These segments participate in different aspects of the filtration process and contain a plethora of distinct specialized cell types. For example, parietal epithelial cells (PECs) are found in Bowman’s capsule. Within the glomerulus, more than 20 different cells have been found with three major types: endothelial cells (ECs), podocytes, and mesangial cells. Further down the nephron, the PT and DCT are lined with epithelial cells. On the other hand, the connecting tubules (CNTs) and CD possess heterogeneity even within the epithelial cell types, i.e., intercalated cells (ICs) are interspersed with principal cells (PCs). Even within ICs, there is further distinction that imparts unique function to each cell type. The kidney interstitium is further composed of other cell types including ECs of peritubular capillaries (either fenestrated or nonfenestrated), vascular smooth muscle cells, pericytes (covering capillaries), fibroblasts, and resident immune cells. Discoveries made in the last two decades or more, based on advanced techniques including micropuncture, isolation of perfused tubules, and electron microscopy, indicated considerable structural and functional heterogeneity along the nephron (1). Recently, an interactive multimodal atlas encompassing both transcriptomic and epigenomic data generated from a combination of single nucleus assays for transposase-accessible chromatin using sequencing and single nuclear sequencing (snRNA-seq) has fostered, refined, and matured our understanding of unique cell types and heterogeneity in the mature human kidney (2).

Thus, the anatomy of kidney is shaped by the specific physiology of each cell type as well as their remarkable spatial relationship (3). Consequently, glomerular or tubular pathophysiology is not limited to a single cell type, i.e., it is attributed to epithelial, endothelial, mesenchymal, and/or immune cell dysfunction. For example, acidosis and alkalosis are caused by impaired IC function. Therefore, a long-term goal in renal biology is to analyze different kidney cell types in detail and to identify the gene expression profile in those cell types. Although bulk tissue RNA sequencing of different components of the kidney can provide insights into segment-specific transcriptomes, it only represents average RNA expression and not the transcriptome at the single cell level. Moreover, the presence of multiple and complex cell types in kidney tissue obfuscates data interpretation. Additionally, lack of response in a more abundant cell type might mask an apparently strong response in a minority cell type (4). Therefore, next-generation techniques such as single-cell RNA sequencing (scRNA-seq), which exploit unbiased genome-wide RNA profiling of individual cells, are used to provide gene expression data at single cell resolution (5, 6). The limitation of this approach is that the analysis of large populations of cells shadow the rare cell types and obscure miniscule differences between different individual cells. Furthermore, scRNA-seq may not be always feasible for a single cell, especially when the tissue is inflamed or fibrotic, making tissue dissection or single cell dissociation extremely challenging. Thus, snRNA-seq, which uses an easy-to-isolate high-quality nuclear preparation, has become a popular alternative to scRNA-seq. Cells like fibroblasts, which are difficult to detach from the basement membrane, are also easily captured and analyzed more thoroughly in snRNA-seq. On the flip side, immune cells are captured poorly in snRNA-seq. Moreover, to analyze the data generated from snRNA-seq, different alignment methods and parameters are required because most of the nuclear RNA is in the unspliced form (7).

With this background, it is evident that on the one hand, there is a tremendous increase in the technological development that allows for unprecedented cellular measurements and gene expression in individual renal cell types; on the other hand, each technique presents significant pros and cons, which questions the reliability, validity, and completeness of the data output. Therefore, a combined community effort from kidney experts is certainly needed to best define a marker that is specific and exclusive to a renal cell type without any discrepancy. Based on the data available, this review is aimed to provide an overview of different renal cell types, specific guide marker proteins associated with each cell type, and their association with renal pathophysiology. We anticipate that this information will provide a systematic understanding of kidney function and specific dysfunction in renal pathologies.

RENAL CELLS AT THE FINEST LEVEL OF SUBDIVISION

The kidney is a complex organ with various vascular compartments each including morphologically and functionally distinct cell types (Table 1). These cells possess several distinct characteristics that distinguish them from other cell types and bestow them with unique functions. In this review, we outline their localization and dynamics within the kidney. The classification is based on the available knowledge on the identification of these cells by detection of expression of their marker proteins.

Table 1.

Overview of the renal cell markers present in different segments of the nephron

Renal Cells/Marker Proteins	Reference(s)
Bowman’s capsule
CD133	(8)
CD24	(9, 10)
CD73	(11, 12)
CD105	(11, 12)
Platelet-derived growth factor receptor-β	(12)
α-Smooth muscle actin	(12)
CD29	(11, 13, 14)
Vimentin	(14)
Stem cell antigen-1	(14)
Nestin	(14)
Cells of renin lineage	(15, 16)
Glomerulus
Nephrin	(17–19)
Podocin	(20)
CD2-associated protein	(21–23)
Dendrin	(24, 25)
Synaptopodin	(26)
α-Actinin-4	(27)
Wilms’ tumor-1	(28)
Kirre-like nephrin family adhesion molecule 1/Neph1	(29)
Podocalyxin	(30, 31)
Cl− intracellular channel protein 5	(32)
Glomerular epithelial protein 1	(33)
Podoplanin	(34)
p35	(35)
Proximal tubules
Megalin	(36)
Cubilin	(37)
Na+-glucose cotransporter-1	(38–42)
Na+-glucose cotransporter-2
Vimentin	(43, 44)
Kidney injury molecule-1
Na/Pi	(45)
PDZK1	(46)
NaS1	(47)
Stem cell antigen-1	(48)
Aquaporin-1	(49,50)
Loop of Henle
Ca2+-sensing receptor	(51, 52)
Cl− channel-Kb	(53)
Na+-K+-2Cl− cotransporter-2	(54–56)
Distal convoluted tubule
Na+/Ca2+ exchanger isoform 1	(57, 58)
Thiazide-sensitive Na+-Cl− cotransporter	(59)
Calbindin 1	(58)
Receptor for lipocalin-2 (24p3R)	(60)
Collecting duct
Cl−-bicarbonate transporter 1 (anion exchanger 1)	(61)
H+-ATPase, apical/basolateral
Pendrin
Cl−/ anion exchanger and SLC26A11
Na+-driven Cl−/${\text{HCO}}_3^{-}$ exchanger
Arginine vasopressin receptor 2	(62)
Renal inwardly rectifying K+ channel subfamily J 1 (ROMK1)	(46)
P2Y14 receptor (GPR105)	(63)
Urea transporter-A1	(64, 65)
Endothelium
Endothelial nitric oxide synthase	(66)
Platelet/endothelial cell adhesion molecule 1	(67)
von Willebrand factor	(68)
Vascular-endothelial cadherin	(69)
Fibroblasts
Fibroblast-specific protein 1	(70)
Platelet-derived growth factor receptor-β	(71)
Mesangial cells
Transient receptor potential canonical member 1	(72, 73)
Platelet-derived growth factor receptor-α	(74)
Epithelial cells
Epithelial Na+ channel subunits α/β/γ (SCNN1A/B/G)	(75)
Cadherin	(76)
Macula densa
B isoform of Na+-K+-2Cl− cotransporter	(77)
Na+/H+ exchanger (isoforms 2 and 4)	(78)

BOWMAN’S CAPSULE

The parietal epithelium of Bowman’s capsule consists of polygon-shaped epithelial cells that rest on a basement membrane. Historically, these PECs were viewed simply as cells lining the inner layer of Bowman’s capsule, but recently they have been recognized as endogenous stem cells with a potential to self-renew. In humans, a subset of PECs coexpress the common stem cell marker CD133 and renal embryonic cell marker CD24 (8–10). These CD24⁺CD133⁺ PECs are called adult parietal epithelial multipotent progenitors (APEMPs). They do not express any lineage-specific markers but demonstrate multilineage differentiation potential. APEMPs are selectively found on the urinary pole of Bowman’s capsule during nephron development (i.e., in embryonic human kidneys), but their presence decreases as development progresses, and they only represent <2% of whole cells in adult kidneys (13). Indeed, APEMPs exhibit heterogeneous potential for differentiation and regeneration into PT cells (9, 10) and podocytes (8), presumably as a compensation for their loss during renal diseases. Interestingly, gene expression profiles of glomerular and tubular APEMPs were not different from one another, indicating that those stem populations are homogenous (79). This population of cells was shown to coexpress stem cell markers CD29, CD44, CD54, and CD106, as revealed by FACS analysis of cell suspensions obtained from human embryonic kidneys (13). Of note, CD106 (vascular cell adhesion molecule 1) was used to segregate tubular-committed APEMPs from those residing in Bowman’s capsule, i.e., APEMPs lacking CD106 were localized either in the PT or in the DCT (9). Other stromal markers including CD73, CD105, platelet-derived growth factor (PDGF) receptor-β (PDGFR-β), and α-smooth muscle actin (α-SMA) were detected in the adult kidney capsule (and in the perivascular region concurrently) by immunohistochemistry (11, 12). In addition, cells derived from the mouse renal capsule were positive for mesenchymal stem cell markers CD29, vimentin, stem cell antigen-1 (Sca-1), and nestin (14). Nestin, a cytoskeleton-associated filament protein, was found to be expressed in mouse (80), rat (81), and human (82, 83) kidneys during embryogenesis. In humans, this expression is restricted to podocytes in mature glomeruli (82, 83). On the other hand, an intermediate filament protein, vimentin, is not only expressed in podocytes but also in mesangial cells of the glomerulus, the endothelium of renal capillaries, and renal stromal cells (84).

Besides APEMPs that only express stem cell proteins, Bowman’s capsule also harbors committed progenitors that coexpress podocyte markers including α-actinin-4, glomerular epithelial protein 1 (GLEPP1), nephrin, podocin, podocalyxin, synaptopodin, and the transcription factor Wilms’ tumor-1 (WT-1) (85–88). These “transitional” cells were found to be residents of Bowman’s capsule only in young and still growing glomeruli and then extend along the vascular stalk of the glomerulus at later stages of life (88–90). Lineage tracing experiments revealed that claudin-1 (87, 90, 91), p57 (87), PAX-2 (91, 92), and Ki-67-positive PECs (92) could also regenerate podocytes.

Another subset of cells lining Bowman’s capsule comprises cells of renin lineage, which extends the mapping of progenitor cells beyond the glomerular tuft since cells of renin lineage are exclusively of juxtaglomerular origin (15, 16). These landmark studies demonstrating the presence of resident multipotent progenitor cells in the adult human kidney as well as their potency for self-renewal, clonogenicity, differentiation, and migration have opened new avenues for regenerative medicine in patients with renal diseases (93, 94).

THE GLOMERULUS

Glomeruli are the key functional units of the kidney filtration apparatus. Each glomerulus consists of capillary tufts, which are structurally maintained by mesangial cells and a three-layered filtration barrier comprising ECs, the glomerular basement membrane (GBM), and highly specialized podocytes (epithelial cells). Glomerular podocytes comprise a cell body and unique actin-based foot processes (FPs) that give them an arborized morphology (95). Adjacent FPs are connected to each other via specialized intercellular junctions called as slit diaphragms (SDs). While charge selectivity is believed to be a function of the GBM, the SD functions as a size-selective sieve (96). This function of the SD is devoted to its flexible multilayered architecture, which can respond to fluctuating blood pressures (97). Under normal healthy conditions, the filtration barrier allows the passage of water, solutes, and plasma proteins smaller than albumin. However, under diseased conditions, podocytes are reorganized into a flattened shape with effacement of podocyte FPs and loss of SD substructures, culminating in massive proteinuria (>3.5 g albumin/day). Since podocytes are the key cells that are involved not only in glomerular physiology but also in the pathology underlying several nephropathies (98–101), researchers have studied and identified specific podocyte proteins that have a direct association with either the disease outcome or its prognosis and suggested new agents or therapies specific to podocytes (102–104).

Nephrin (encoded by NPHS1) is the adhesion protein that is primarily found in the kidneys (17). It is the first protein that was shown to be expressed on SDs of human podocytes (18, 19). The immunoelectron microscopy-based analysis suggested that nephrin extending from two adjacent podocyte FPs interacts with each other to form a zipper-like arrangement (18). A year later, another important SD-associated podocyte protein, podocin (encoded by NPHS2), was identified (20). The importance of podocin was underscored by the finding that localization of nephrin to the SD depended on its ability to interact with podocin via its R1160 residue at the COOH-terminal cytoplasmic tail (105, 106). This signaling complex has been shown to direct nephrin to the lipid rafts on the cell surface (107) and maintain the structural integrity of podocytes (106, 108). The most common NPHS2 disease-causing variant, the R138Q mutation in podocin, disrupts its folding and causes incomplete glycosylation (109). These events further cause podocin to relocalize from the plasma membrane and interfere with the proper trafficking of nephrin (109). As a result of this point mutation, both R138Q podocin and nephrin are retained in the endoplasmic reticulum (ER) (109), which was reported to be the subcellular destination of missense mutated nephrin molecules (110). As indicated in the later study, NPHS1 mutant proteins are also associated with improper folding and glycosylation, which is required for the plasma membrane localization of nephrin (111), leading to their retention and degradation in the ER (112). These studies demonstrated that the appropriate trafficking of podocin is a prerequisite for the successful localization of nephrin on the cell surface and stabilization of the SD. Vice versa, nephrin is also required for the appropriate localization of podocin to the SD. Recent efforts to produce kidney organoids using human induced pluripotent stem cells from patients with NPHS1 missense mutations (E725D or R460Q) have shown that the expression and overall “basal” distribution of podocin were unaffected but that nephrin was required for the recruitment of podocin to “lateral” SD domains (113, 114). In total, ∼250 distinct genetic mutations in NPHS1 (115) and >125 mutations in NPHS2 (116) have been identified. It is noteworthy that genetic defects in podocin can start at any age, whereas mutations in nephrin cause nephrotic syndrome (NS) within the first 3 mo of life (117). Therefore, while mutations in NPHS1 cause congenital NS of Finnish type in humans (17), mutations in NPHS2 are associated with autosomal recessive familial steroid-resistant NS (20). Moreover, a plethora of studies have reported downregulation of nephrin (118–125) and podocin (122–126) expression in various proteinuric kidney diseases. In fact, recent studies have demonstrated a decline in nephrin expression as an early event independent of podocyte loss, which impairs the ability of podocytes to recover after injury, thereby facilitating disease progression (127).

Another major protein, which was discovered in the late 1990s to be necessary for the SD, is CD2-associated protein (CD2AP) (22, 23). CD2AP has been detected in diverse tissues like the placenta, colon, pancreas, kidney, and thymus (128). In the kidney, it is also expressed in tubules (23); more specifically, there is robust CD2AP expression in cortical ureteric bud epithelial cells and medullary CDs at birth and PTs, DCTs, and CD epithelial cells at maturity as detected by immunofluorescent staining of the mouse kidney (21). Immunohistochemical analysis of normal mouse and human kidneys demonstrated that within the mature glomerulus, CD2AP localization was restricted to podocytes (21). Another study by Lehtonen et al. (129) described CD2AP localization in mouse embryonic and adult kidneys using in situ hybridization and immunofluorescence microscopy. In the embryonic kidney (17 days old), strong expression of CD2AP was seen in mature, medullary branches of CDs and in mature glomerular podocytes. In the adult mouse kidney, CD2AP expression was strongly retained in glomeruli and CDs, with weak but positive staining in a subset of distal tubular segments and no staining in PTs. CD2AP, an 80-kDa protein found at the cytoplasmic side of the SD, has been shown to bind to nephrin (130) and podocin (105). CD2AP functions as an adaptor protein that anchors these two SD proteins to actin filaments of podocyte cytoskeleton and is thus essential for normal podocyte function (131, 132). Furthermore, CD2AP has been shown to bind to actin-associated synaptopodin (133). Through its three adjacent SH3 domains, CD2AP can bind to different structural proteins or signaling components of the cytoskeleton simultaneously (22, 128). Besides being heavily involved in cytoskeletal rearrangements, CD2AP also plays a crucial role in cellular signaling. It interacts with the p85 regulatory subunit of phosphoinositide 3-OH kinase (PI3K) and elicits PI3K-dependent Akt signaling in podocytes (134). This CD2AP-mediated Akt activity regulates a series of biological events including actin dynamics (135) and ER stress and apoptosis (136). However, CD2AP is downregulated under diabetic conditions, which also hampers the PI3K/Akt signaling pathway (137). Furthermore, CD2AP has been shown to be required for activation of PI3K and ERK/MAPK signaling pathways by transforming growth factor-β (TGF-β), and, thus, in the absence of CD2AP, TGF-β receptors failed to engage these antiapoptotic pathways causing hyperactivation of proapoptotic p38 MAPK (138). CD2AP acquires phosphorylation by receptor tyrosine kinases in podocytes (139), which alters its affinity for nephrin. Mice devoid of CD2AP protein exhibit loss of FPs and severe proteinuria (23, 140). Heterozygosity for a defective CD2AP allele is associated with a complex renal phenotype (140), and mutations and polymorphisms in the human gene are correlated with the development of glomerulonephritis and glomerulosclerosis (141–143).

Dendrin, originally identified in the telencephalic dendrites (25), is now recognized as a member of the SD complex, where it directly interacts with nephrin and CD2AP. Immunofluorescence and Western blot analysis using antibodies against dendrin detected the expected 89-kDa and 81-kDa isoforms in the brain, but only the 81-kDa isoform was found in mouse glomeruli (24). Also, dendrin was found to colocalize with the podocyte marker synaptopodin (26, 144). Dendrin, a proline-rich protein with two putative nuclear localization signals and three PPXY motifs, also possesses proapoptotic signaling properties; i.e., in response to glomerular injury, dendrin translocates from the SD to the nucleus of podocytes (145–147). Thus, nuclear localization of dendrin serves as a useful tool or marker to assess the integrity of the SD under pathological conditions.

The SD is considered to be a modified adherens junction (148) since it contains several tight junction proteins including zonula occludens-1 (ZO-1) (149), P-cadherin, α-catenin, β-catenin, γ-catenin (148), FAT (150), p120 catenin, CASK (151), MAGI-2 (152), JAM-A, occludin, and cingulin (153). The link between these proteins and the podocyte actin cytoskeleton is formed by actin-binding proteins such as synaptopodin and α-actinin-4. Synaptopodin is a proline-rich actin-binding protein that is localized in the postsynaptic density region of podocyte FPs and neuronal dendritic spines of the rat brain (26). Functional studies have revealed that mice lacking synaptopodin demonstrated reduced synaptic plasticity due to lower dendritic spine formation in the hippocampus (154) but exhibited no structural differences within the glomeruli (155). When the podocytes were injured via protamine sulfate (PS) or lipopolysaccharide (LPS), the actin bundles (which are complexes of synaptopodin and actin filaments) had delayed reformation at the injury site (155), suggesting that synaptopodin played a role in actin filament regulation. Indeed, synaptopodin could bind to α-actinin-2 (155), α-actinin-4 (155–157), and β-catenin (157) and regulate cellular contraction, elongation, and motility. The observed differences between the cultured hippocampal cells and kidney cells were explained by the existence of different synaptopodin isoforms, e.g., neuronal Synpo-short, renal Synpo-long, and Synpo-T (155). Depending on the isoform, binding of synaptopodin could lead to the inhibition of branching of the α-actinin-induced actin filaments. Calcineurin-dependent degradation of synaptopodin by cathepsin L also causes the loss of actin stress fibers in podocytes (158). The molecular mechanisms underlying synaptopodin and actin cytoskeleton interactions involve the ability of synaptopodin to regulate various signaling molecules including small GTPases RhoA (159), Rac1 (160), and Cdc42 (161), Cdc42:IRSp53:Mena complexes (162), and adapter protein Nck1 (163). Our group has recently shown that synaptopodin could limit the expression of transient receptor potential canonical ion channel member 6 (TRPC6), another SD member protein, controlling the dynamicity of the FP and protecting podocyte function (164). Each of transient receptor potential canonical ion channel member 5 (TRPC5) and TRPC6 acted as antagonistic regulators of synaptopodin by forming molecular complexes with Rac1 and RhoA, respectively, and provided a balance between a motile phenotype and a contractile phenotype in podocytes (165). Despite these protective functions of synaptopodin reported in the past decades, a recent publication showed that any isoform of synaptopodin could be dispensable in vivo but beneficial in the protection against acute podocyte injuries (166). Interestingly, loss of synaptopodin led to lower Rac-1 and RhoA activity, which was podocyte protective (166).

α-Actinin-4 (ACTN4), one of the members of actin-bundling proteins (ACTN1–ACTN4) and known to form ∼100-kDa head-to-tail homodimers, plays an essential role in podocyte FP architecture and adhesion (27). α-Actinin monomers comprise three major distinct domains: an F-actin-binding domain (ABD) at the NH₂ terminus, four spectrin-like repeats (SRs), and EF hand motifs (calmodulin-like domain) at the COOH terminus (167). While ACTN2 and ACTN3 are Ca²⁺ insensitive and show predominant sarcomere-specific expression (168), the other two nonmuscle isoforms, ACTN1 and ACTN4, are highly and moderately sensitive to Ca²⁺, respectively, and are also widely expressed throughout the body (169). However, only ACTN4 expression is detected in human kidneys, and mutations in this gene are associated with an autosomal dominant form of familial focal segmental glomerulosclerosis (FSGS) (170). Interestingly, all the major mutations, including K255E, T259I, and S262P, are located within the evolutionarily conserved ABD of ACTN4 (167). Overall, mutated ACTN4 proteins exhibit higher binding affinity to F-actin causing an alteration in the mechanical characteristics of podocytes such as cytoskeletal rigidity, aberrant localization/aggregation patterns, and significantly diminished half-life (171). Most affected individuals have only mild proteinuria in their early adulthood but a progressive decline in kidney function later. Two more mutations, albeit missense (W59R and I149del), have also been reported within the evolutionary conserved ABD. Individuals carrying these mutations exhibit proteinuria at an early age of 5 yr old that progressed to end-stage renal disease (ESRD) within 3 yr, which is much faster, earlier, and more severe compared with the other ACTN4 mutations. Besides genetic mutations, decreased expression of ACTN4 has been attributed to several glomerulopathies including IgA nephropathy, sporadic FSGS, and minimal change disease (172).

The Wilms’ tumor-1 gene (WT1), a zinc finger protein that acts both as a transcription factor and an RNA-binding protein, is expressed only in podocytes (both mouse and human), undergoes alternative splicing to generate several isoforms, and is essential for kidney development (28). Interestingly, WT-1 controls the expression of many podocyte-specific genes that localize to the SD, such as NPHS1, NPHS2, MAGI2, Kirrel, Plce1, Ptpro, Cldn5, and Nck2, by binding to their promoter regions (173). In addition to SD components, WT1 is a transcription factor and has been shown to regulate the expression of proteins that are involved in cell matrix adhesion, cytoskeletal rearrangement, and polarity maintenance in podocytes (173). Consequently, it is expected that mutations in the WT1 gene will result in a wide spectrum of renal phenotypes like disruption of podocyte development or maintenance (174). These mutations are associated with glomerulopathies such as 1) Denys-Drash syndrome, characterized by early NS with diffuse mesangial sclerosis progressing rapidly to ESRD; 2) Frasier syndrome, presenting as progressive nephropathy with proteinuria and NS accompanied with FSGS progressing to ESRD in adolescence or young adulthood, and 3) steroid-resistant NS (174). Furthermore, mice with podocyte-specific WT1 depletion display defective podocyte differentiation, anuria, kidney failure, and death within a day postbirth (175). A decrease in WT1 staining has been used as a marker to confirm podocyte injury in patients with diabetic nephropathy (DN) (176). Along with desmin, WT1 is regarded as the most sensitive marker for podocyte and glomerular damage since WT1 immunohistochemistry represents a reliable complement to a morphology-driven approach to podocyte disease (177).

Kirre-like nephrin family adhesion molecule 1, known as Kirrel1 or Neph1, shares some homology with nephrin but has a shorter extracellular part, i.e., only five IgG-like domains (29). It was first isolated by a retroviral mutagenesis gene trapping method and found to be expressed in both mouse and human kidneys, specifically in the glomerulus (29). Neph1-deficient mice develop NS at birth causing early mortality (3−8 wk postconception) due to proteinuria and FP effacement (29). In the same year, it was shown that nephrin (NPHS1) knockout mice demonstrated more severe podocyte FP effacement with gross proteinuria and edema, and death occurred much quicker (within 24 h) compared with Neph1 knockout mice (178). Both Neph1 and NPHS1 share a conserved domain for podocin binding in their COOH-termini, suggesting that both proteins are required for nephrin-dependent signaling (179). More evidence was provided in a complementary study showing that the interaction between nephrin and Neph1 was not only specific but also an important determinant of glomerular permeability (180). However, Neph1 is not only involved in maintaining the glomerular permeability barrier. As a type I transmembrane protein, it also regulates outside-in signaling in podocytes, which results in actin polymerization (181, 182). Current studies are investigating various mutations found in the Neph1 gene from human patients and their physiological consequences in NS (183). There are two more Neph1-related proteins, Neph2 (179) and Neph3 (also termed filtrin) (184), and they share a common architecture comprising a transmembrane domain, five IgG-like domains, and a cytoplasmic tail including Grb2‐ and PDZK1‐binding sites.

Podocalyxin (also known as PCLP1, MEP21, gp135, and thrombomucin) is an integral single-pass transmembrane sialoprotein that imparts a negative charge to the glomerular membrane and contributes to podocyte morphogenesis and structural integrity and is found on the apical side of rat podocyte FPs and vascular ECs (30, 31). It is a 150- to 165-kDa protein encompassing a mucin domain, a globular domain, a transmembrane domain, and a highly charged cytoplasmic tail with putative phosphorylation sites for protein kinase C and casein kinase II (185). The PSD-95/Disks-large/ZO-1 (PDZ)-binding motif (DTHL) located at the COOH terminal of podocalyxin facilitates its interactions with Na⁺/H⁺ exchanger (NHE) regulatory factors 1 and 2 (NHERF1 and NHERF2, respectively) (186), the two adaptor proteins implicated in protein trafficking, ion transport, and signaling. Furthermore, podocalyxin has been shown to interact with the actin-binding protein ezrin (187). Thus, podocalyxin performs essential cellular functions through its interaction with NHERF proteins, ezrin, and the actin cytoskeleton. Consequently, deletion of podocalyxin in mice led to defective renal functions with marked anuria and early death. The podocytes in these mice failed to develop FPs and SDs and instead formed impermeable tight junctions (188, 189). Recently, a novel heterozygous missense mutation, c.T1421G (p. L474R), was identified in the podocalyxin encoding gene in an autosomal dominant FSGS pedigree. However, the function of podocalyxin was not affected by this mutation (190). Another study has reported a heterozygous nonsense mutation in the podocalyxin gene (c.C976T; p. Arg326X) that is characterized by proteinuria and renal insufficiency (191). Interestingly, a strong association was found between poor renal outcome and reduction of podocalyxin expression with an increase in its urinary excretion (192). Thus, assessment of podocalyxin levels in renal tissues and urine was proposed as a reliable biomarker to predict the progression of DN. A recent study by Refaeli et al. (193) generated two mouse strains to study the developmental role of podocalyxin: one mouse strain had Podxl deleted from developmentally mature podocytes (Podxl ^Δ Pod) and in the other strain podocalyxin was heterozygous in all tissues (Podxl^+/−). Histological and ultrastructural analyses were performed to gauge kidney development and function in both of these strains. While Podxl^ΔPod mice developed acute congenital NS characterized by FSGS and proteinuria, Podxl^+/− mice had no obvious renal phenotype with a normal lifespan but were extremely susceptible to puromycine aminonucleoside-induced nephrosis. This study indicates that podocalyxin plays a critical role in the formation and maintenance of podocyte structure (193).

Another protein found to be expressed on the apical and basal membranes of mouse FPs and glomerular ECs is Cl⁻ intracellular channel protein 5 (CLIC5) (194). CLIC5 belongs to a family of intracellular Cl⁻ channels that are known to have a putative single transmembrane domain. CLIC5 colocalizes with podocalyxin and ezrin/radixin/moesin (ERM) complex in FPs (194). Mice lacking CLIC5 had a reduced amount of these interacting proteins, and they exhibited FP abnormalities and proteinuria (32). Ezrin, which functions as an actin-membrane linker, is also predominantly expressed in rat podocytes (194). Ezrin has been reported to be a specific marker in both undifferentiated and differentiated glomerular epithelial cells in culture, in the adult rat glomerulus, and also during glomerulogenesis (195). Furthermore, the study was extended to include glomerular disease models where podocytes were injured, and ezrin was found to be elevated in binucleated podocytes or podocytes that were partially or completely detached from the underlying GBM (195).

Glomerular epithelial protein 1 (GLEPP1), also known as protein tyrosine phosphatase receptor type O (Ptpro), is a 132-kDa receptor tyrosine phosphatase present on the apical cell surface of podocyte FPs in rabbits (33), mice (196), and humans (197). The protein contains a transmembrane domain, an intracellular phosphatase domain, and a large extracellular domain consisting of eight fibronectin type III-like repeats. GLEPP1 is believed to play an important role in maintaining podocyte integrity, structure, and function. Targeted depletion of GLEPP1 has been shown to alter podocyte structure; i.e., the typical “octopoid” podocyte structure was overtly simplified to a more “amoeboid” structure and minor FPs exhibited substantial blunting and a broader morphology along with an altered distribution of vimentin, a podocyte intermediate cytoskeletal protein (196). This structural disruption was accompanied by a loss of function, characterized by a reduction in the glomerular filtration rate. Reduction of GLEPP1 is also associated with several podocytopathies like IgA nephropathy and FSGS (198).

Podoplanin (PDPN), also called E11 antigen/GP36/aggrus, is another 43-kDa integral membrane glycoprotein that is localized on the rat podocyte surface. The level of this protein is reduced in puromycin-induced kidney injury (34). Administration of anti-PDPN antibodies to rats led to proteinuria and podocyte FP effacement (199). Furthermore, selective loss of PDPN expression was strongly correlated with enhanced proteinuria in Dahl salt-sensitive rats (200). A serendipitous yet interesting finding by Kasinath et al. (201) revealed fading of glomerular PDPN and its concomitant increase in the tubulointerstitial compartment and in the urine shortly after ischemia-reperfusion injury (IRI) in mice. The role of PDPN as an antiapoptotic factor in ANG II-induced injury in human podocytes was also underscored (202).

Glomerulosclerosis and proteinuria are primarily attributed to podocyte loss resulting from either apoptosis, detachment, or inability of podocytes to adequately proliferate. Unlike other cyclin-dependent kinases (Cdks) that are involved in cell cycle regulation, Cdk5 has been shown to modulate cell maturation, differentiation, migration, and apoptosis (35). Interestingly, in mouse kidneys, Cdk5 shows glomerular expression that is restricted to podocytes. While the levels of Cdk5 were elevated in differentiating conditionally immortalized mouse podocytes in culture and in developing rodent fetal kidneys, they declined markedly in an experimental disease model of GBM nephritis that causes podocyte dedifferentiation/proliferation (203). In contrast to other Cdks that are activated by their cognate cyclins, Cdk5 has been shown to be activated by a noncyclin protein, p35, which is also expressed constitutively in podocytes and forms a complex with Cdk5 (35). This active complex (p35/Cdk5) was found at the plasma membrane (203). Further studies showed that p35-null mice had no kidney irregularities during glomerulogenesis or renal dysfunction during adult life. However, these mice were more susceptible to external injury stimuli and their podocytes underwent apoptosis. This indicates that, although p35 does not affect glomerular development, it certainly impacts podocyte survival postinjury (35).

TUBULES

There are 43 cell types purported to be present in kidney tissue, and PTs make up a substantial portion of the entire kidney tissue (4). PT cells roughly account for 52% of the estimated 200 million tubule epithelial cells. The PT epithelial cell population that is pure, stable, and functional is characterized by CD10 (neutral endopeptidase)/CD13 (aminopeptidase M) double staining and expression of specific markers such as aquaporin-1 (AQP1) and N-cadherin (49). While these CD10⁺/CD13⁺ cells exhibit epithelial characteristics over a long time, those that are positive only for either CD10 or CD13 appear morphologically heterogeneous and do not express PT markers (204).

Some albumin molecules pass through the glomerular filtration barrier under normal physiological conditions but finally are reabsorbed by the PT epithelium (205). Megalin [an ∼600-kDa glycoprotein with a single transmembrane domain, also known as low-density lipoprotein-related protein 2 (Lrp2)], which has been identified as a target antigen in Heymann nephritis (36), and cubilin (a 460-kDa glycoprotein with no transmembrane domain), which has been identified as an albumin-binding protein in rats by affinity chromatography (37), are two large transmembrane proteins expressed on the surface of PT epithelial cells. When megalin was first identified in the proximal convoluted tubule (PCT) (36), it was also detected in glomerular epithelial cells of rats as the target antigen in a model of glomerulonephritis. However, in humans, megalin was only detected in PTs (206). Several years later, using more sensitive techniques like RT-PCR on laser microdissected glomeruli, megalin was shown to be expressed in human podocytes as well (207). Using the same approach, a year later, cubilin was also found in podocytes of rats and humans (208). Both of these glycoproteins play a pivotal role in the endocytic reabsorption of many plasma proteins that are filtered across the glomerular capillary wall (209). Consequently, albumin is absent from the urine of healthy individuals. Thus, because of a highly efficient tubular reabsorption system, the presence of minute amounts of albumin in the urine of a healthy individual is considered as a risk factor for future renal disease (210). Both inherited and acquired dysfunction of megalin and cubilin have been described in humans (211). Mutations in the lrp2 gene causes a rare autosomal recessive disorder, Donnai-Barrow facio-oculo-acousticorenal syndrome, characterized by a distinct facial appearance, developmental delays, high-grade myopia, hypertelorism, sensorineural hearing loss, and congenital diaphragmatic hernia along with proteinuria (212). Similarly, Imerslund-Gräsbeck syndrome, a rare autosomal recessive inherited defect that is characterized by proteinuria, has been ascribed to cubilin deficiency (213). In addition to the above-mentioned inherited disorders, receptor dysfunction or changes in receptor expression have been observed in several acquired disorders associated with proteinuria such as experimental models of acute and chronic renal disease including LPS-induced endotoxemia, Shiga toxin-induced nephropathy, IRI, chronic kidney disease (CKD), and diabetes (214). Furthermore, an increase in the urinary excretion of megalin and its fragments as well as cubilin has been reported in models of Alport syndrome (215), in experimental as well as human diabetes (216, 217), and in IgA nephritis (218).

In the mammalian nephron, filtered hexoses are efficiently reabsorbed, an important physiological process to retain glucose and prevent substantial energy substrate loss into the urine, primarily by a concerted action of Na⁺-dependent (SGLT1 and SGLT2) (38) and Na⁺-independent (GLUT1 and GLUT2) (219) glucose transporters. SGLT2, a low-affinity/high-capacity Na⁺-dependent glucose cotransporter located in the early PT, takes up the bulk of glucose (∼97%) that is filtered by the renal glomerulus (219). The presence of SGLT2 was identified predominantly on the apical side of renal PCTs (S1 and S2 segments) in the mouse and human (39–41). In contrast, using specific antibodies (immunostaining and Western blot analysis) and gene expression studies (Northern blot analysis and in situ hybridization), SGLT1 was found on the brush-border membrane of the straight renal PT (S3 segment) in the human (40), mouse (220), rat (221), rabbit (222), and pig (223). While SGLT1 is responsible for most of the dietary glucose uptake in the intestine, this high-affinity/low-capacity transporter only accounts for the absorption of residual glucose (∼3%) in the S3 segment of the late PT (42). Even though its impact on the kidney is relatively less, the presence of SGLT1 is necessary to absorb the last traces of glucose from the lumen into the blood. These two transporters differ in many aspects: 1) their affinity for glucose and Na⁺ (SGLT1 > SGLT2); 2) their sensitivity to phlorizin inhibitor (SGLT1 > SGLT2); 3) their predilection and selectivity for hexoses (SGLT1 can transport both glucose and galactose equally well, but SGLT2 is shown to transport glucose 10 times better than galactose); and 4) a distinct segmental distribution (224). One of the consequences of diabetes is higher renal SGLT2 expression, which eventually enhances renal glucose reabsorption and increases the glucose transport maximum (from ∼450 to 600 g/day). Drugs that directly block SGLT2 are potent therapeutics that inhibit renal glucose reabsorption, increase glucose excretion, and successively lower hyperglycemia (225, 226), as demonstrated in prospective clinical trials with patients with type 2 diabetes mellitus (227, 228) and corresponding meta-analyses (229). Since SGLT2 reabsorbs Na⁺ along with glucose, SGLT2 blockers promote natriuresis and glucosuria. The resultant osmotic diuresis has been shown to have an antihypertensive effect (230, 231) and cause a compensatory increase in tubuloglomerular feedback (232).

Furthermore, transcellular glucose transport is accomplished by two glucose transporters that are located on the basolateral membrane, and they allow glucose to exit from PT cells. They are low-affinity glucose transporter 2 (GLUT2) present in the S1 segment and high-affinity glucose transporter 1 (GLUT1) found in the S3 segment (233). SGLT2 has been shown to be increased in alloxan-induced diabetic rats (234), and several groups have also shown that GLUT2 in renal PTs is increased in streptozotocin-induced diabetic rats (235). Interestingly, expression of intestinal SGLT1 and GLUT2 has been reported to increase in patients with type 2 diabetes mellitus (236). Renal expression of these glucose transporters has not been investigated.

The PTs and thick ascending limbs (TALs) of the loop of Henle are susceptible to injury caused by episodes of hypoxia or exposure to nephrotoxins. Vimentin and kidney injury molecule-1 are the markers corresponding to tubular injury, and no basal expression of either of these two proteins can be observed in healthy rat kidneys (44). While these acute assaults often lead to loss of tubular cells, these injuries are only transient. This is because renal tubular cells can regenerate after such insults. However, researchers are still unable to completely discern the origin of the tubular cells that replenish the epithelial population after such damage. Whether regeneration depends on an intratubular progenitor or stem cell subpopulation or whether any enduring tubular cell survivor has the potential to divide remains controversial.

The type II Na⁺-coupled phosphate (Na/Pi) transporters are known to perform tubular reabsorption of inorganic phosphate (P_i), which is important for body functions, especially during periods of growth. In 1975, Baumann et al. (43) achieved the first transport assay of its kind to measure the build up of transtubular concentration differences of P_i in the renal PCT of rats and found that the rate of P_i reabsorption occurred with a higher rate in the early part of the tubule compared with the later part. Later, it has been demonstrated that P_i reabsorption is regulated by the apically localized Na/Pi cotransport system in PTs using different experimental systems including primary cultures of renal PTs, isolated tubules, and isolated brush-border membrane vesicles (45). The cloning of NaPi-2a (SLC34A1) (237) and NaPi-2c (SLC34A3) (238) phosphate transporters allowed the design of a new series of experiments to establish the physiological relevance of the isolated genes and characterize the mechanism of P_i reabsorption. While NaPi-2a reabsorbs ∼70% of P_i in the adult kidney, NaPi-2c does the remaining 30% (239). Despite a similar molecular structure, the two transporters exhibit physiological disparities, including how they adapt to dietary changes in phosphate. When P_i intake is high, NaPi-2a decreases within 1 h; however, NaPi-2c levels decrease slowly (4 h). NaPi-2a undergoes internalization in a microtubule-independent way where it is directed to the lysosomes for degradation under high phosphate conditions (240). In contrast, internalization of NaPi-2c takes place through a microtubule-dependent pathway, and instead of being targeted to lysosomes and degraded, it is accumulated in a subapical compartment (241). These differences in their regulation under dietary and hormonal stimuli and stability at the apical membrane are attributed to their differential binding to PDZ proteins, which are scaffolding proteins involved in the regulation of several receptor and transporter proteins (242). NaPi-2a has a class I PDZ-binding site at its COOH terminus (TRL) and is thus susceptible to regulation by several PDZ proteins, including NHERF family members (NHERF1, NHERF2, NHERF3, and NHERF4) (243). NaPi-2c interacts with NHERF1 and NHERF3 (PDZK1) despite the absence of a prototypical PDZ-binding motif in its COOH terminus (46).

PTs are also the site for the reabsorption of inorganic sulfate ($\text{S}{\text{O}}_4^{2-}$), an indispensable and abundant anion in human plasma, which is essential for a variety of cellular processes. Stop-flow experiments in dogs revealed that circulating $\text{S}{\text{O}}_4^{2-}$ concentrations in plasma are mainly regulated by renal PTs and that high plasma values are likely the consequence of compromised kidney function (244). The apical membrane Na⁺-sulfate cotransporter NaS1 (SLC13A1) participates in sulfate reabsorption across PTs and thereby regulates blood sulfate levels (47). NaS1 prefers substrates such as sodium and sulfate, thiosulfate, and selenate. Expression of this transporter is downregulated by hypothyroidism, hypokalemia, metabolic acidosis, a high-sulfate diet, vitamin D depletion, glucocorticoids, and nonsteroidal anti-inflammatory drugs and upregulated vice versa by thyroid hormone, reduced dietary intake of sulfate, growth hormone, vitamin D supplementation, chronic renal failure, and during postnatal growth (47).

Sca-1 (also known as Ly6a) is an 18-kDa glycerophosphatidylinositol-anchored protein and a member of the Ly-6 protein family (48). It is used commonly as a marker for the detection and identification of stem cell and progenitor populations (245, 246). Sca-1 has been shown to be expressed heavily only in specific renal tubular epithelial segments of the adult mouse kidney (247), especially in the PT (but not in the CDs), a nephron segment that is extremely susceptible to IRI, and also in the loop of Henle (248). The study showed that epithelial Sca-1 can interact with TGF-β receptors I and II and inhibit TGF-β-induced canonical Smad signaling, a pathway instrumental in causing renal scarring due to fibrosis. Thus, Sca-1 has been recognized to afford a renoprotective role in IRI (248).

THE LOOP OF HENLE

The loop of Henle has a hairpin-like or a long U-shaped configuration that functions as an intermediate segment connecting the PT to the DCT. It is functionally divided into three distinct regions, e.g., the thin descending limb (DTL), the thin ascending limb (ATL), and the TAL. The glomerular filtrate is further reabsorbed in this region of nephron. Classical understanding from earlier studies is that while the DTL has high water (P_f) and low urea (P_urea) permeabilities, the ATL has minimal P_f but extreme permeability to solutes like NaCl and moderate P_urea. However, detailed studies on limb function, its ultrastructure, three-dimensional organization, and mathematical models have shown that the DTL has two functionally distinct segments (DTL-upper and DTL-lower). The upper 40% of the DTL has been shown to express AQP1, making it highly water permeable, but the lower 60% of the DTL lacks AQP1 channels and is thus impermeable to water (50). Furthermore, expression of the Cl⁻ channel ClC-K1 has been shown to begin from the prebend segment at the terminal end of the DTL that continues all the way up to the ATL, accounting for high Cl⁻ permeability. Both the DTL and ATL are now recognized as highly permeable to urea (249). The thin segments lie entirely within the renal medulla and are connected to the TAL, which is further subdivided into medullary and cortical compartments. Cells of the TAL differ from those residing in the thin limbs since they are metabolically more active and generate a steep electrochemical gradient for apical Na⁺ entry (due to the presence of highly active Na⁺-K⁺-ATPase). Overall, these segments are the key players in maintaining urine osmolarity.

Uromodulin (UMOD), a glycoprotein also known as Tamm-Horsfall protein (THP), is selectively expressed by epithelial cells lining the TAL of the loop of Henle and the early DCT (DCT1) in mammals (250). Mature UMOD, after its synthesis in the ER and extensive glycosylation, undergoes proteolytic cleavage by the serine protease hepsin and is released into the urine, representing the most abundant protein in human urine (250). UMOD has been implicated in concentrating the urine by exerting regulation on the Na⁺-K⁺-2Cl⁻ cotransporter (NKCC2) (251) and renal outer medullary K⁺ channel (ROMK) (252) located in the TAL (253). UMOD also functions as a defense factor against ascending urinary tract infections (254) and kidney stones (255) and regulates innate immunity (256). Mutations in the UMOD gene cause defective maturation and impaired trafficking and retention of UMOD in the ER of cells located in the TAL segment (257), leading to a rare dominant progressive ER storage hereditary disease known as uromodulin-associated kidney disease (258). Hitherto, >70 uromodulin-associated kidney disease-causing UMOD mutations have been reported. Patients with uromodulin-associated kidney disease exhibit compromised urinary concentration ability, hyperuricemia associated with gout, progressive tubulointerstitial fibrosis with moderate infiltration of inflammatory cells, tubular atrophy, and renal cysts in some cases (259). Genome-wide association studies have identified UMOD as a risk factor for CKD, hypertension, and kidney stones and urinary UMOD as a useful biomarker for CKD progression (258).

The extracellular Ca²⁺-sensing receptor (CaSR) is a unique ∼120- to 160-kDa dimeric phospholipase C-sensitive G protein-coupled protein. CaSR, by releasing Ca²⁺ from the stores, functions as the master regulator of Ca²⁺ homeostasis (260). CaSR is expressed in high levels in the parathyroid gland, kidney, gut, and bone, where it regulates secretion of parathyroid hormone, synthesis of vitamin D (via its ability to regulate the expression of 1-α-hydroxylase enzyme), and absorption and resorption of Ca²⁺, respectively (261). Renal CaSR is also known to prevent hypercalcaemia, by promoting Ca²⁺ excretion (262) and by inhibiting 1,25-dihydroxyvitamin D [1,25(OH)₂D] synthesis (263). Thus, gain- and loss-of-function mutations in the CASR gene account for severe disturbances in Ca²⁺ metabolism. For example, heterozygous mutations that inactivate the CASR gene cause benign (asymptomatic) familial hypocalciuric hypercalcemia, but those that are homozygous for inactivating mutations display severe hypercalcemia (264). On the other hand, CASR-activating mutations are the cause of the metabolic wasting disease known as autosomal dominant hypocalcemia or type V Bartter’s syndrome (265). In the rat kidney, abundant expression of CASR has been reported in the medullary and cortical parts of the TAL (266). Several other groups have confirmed these findings in mature mouse and human kidneys along with the developing human kidney (267–270). Besides CaSR expression in the TAL, Riccardi and colleagues (266) reported detection of CASR mRNA in glomeruli and in almost all other tubular segments, including the PCT, proximal straight tubule (PST), DCT, cortical CD (CCD), and inner medullary CD (IMCD) but not in the thin limbs and CNTs. Interestingly, Yang et al. (270) found CASR mRNA only in the DCT and CCD but not in the glomeruli or in any other tubular segments. Furthermore, using antibodies directed against CaSR, Loupy et al. (268) obtained conflicting results with no CaSR protein expression in the PCT, PST, DCT, CNT, or CCD as revealed by immunohistochemistry and immunofluorescence experiments. Thus, expression of CaSR in other parts of the kidney has remained controversial due to variable specificity and sensitivity of detection probes, antibodies, and even species (51, 52). Lately, nephron segment-specific roles of CaSR have been described (271). CaSR-mediated Ca²⁺ regulation is achieved by the interaction between CaSR and several variants of claudins (claudin-16 and claudin-19), which form a channel to reabsorb Ca²⁺ and Mg²⁺ (272). Apart from Ca²⁺ sensing, CaSR also controls ionic homeostasis by regulating the other cotransporters involved in ion reabsorption in the loop of Henle, namely, NKCC2, ATP-sensitive inward rectifier K⁺ channel ROMK1, and Cl⁻ channel ClC-Kb (273). ClC-Kb is a voltage-gated Cl⁻ channel that symports Cl⁻ with Na⁺ from NKCC2 (274). Northern blot analysis revealed that the mRNA for this Cl⁻ channel was expressed predominantly in rat kidneys, especially in the inner medulla, with a faint expression in the bladder (274). A gene expression study in microdissected nephron segments from rats revealed that the ATL of Henle’s loop, which has the highest Cl⁻ permeability among the nephron segments, is the main site of ClC-Kb expression (274). This finding was later confirmed by immunohistochemistry and immunoelectron microscopy (53). Mutations in CLCNKB encoding kidney-specific ClC-Kb lead to Bartter’s syndrome type III, as Cl⁻ is required to keep electroneutrality during salt resorption. Like NKCC2, ClC-Kb is regulated by CaSR (275).

Furosemide-sensitive NKCC2, with type 2 being the major subtype within the kidney, was identified as the major NaCl regulator located in the TALs in the rabbit and mouse (54–56). NKCC2 is negatively regulated by CaSR, by increased production of prostaglandin E₂, which interferes with NKCC2 transport (276). Loss-of-function mutations of the SLC12A1 gene encoding NKCC2 lead to impaired renal salt retention and loss of NaCl via the tubules and TAL. The renal salt retention eventually leads to blood pressure abnormalities.

THE DISTAL CONVOLUTED TUBULE

The DCT is an extension of the TAL and starts shortly after the macula densa (MD). It is characterized by modest solute transport and very low water permeability (277). It further reabsorbs filtered NaCl and facilitates urinary dilution and generation of the osmotic gradient.

Ca²⁺ homeostasis is required for normal physiological functioning of all living cells. This is primarily achieved by three highly regulated systems: renal reabsorption, intestinal absorption, and bone resorption/formation. Any disturbance in these processes causes “hypercalciuria,” which is one of the leading risk factors for kidney stones. In the kidney, ∼45% of plasma Ca²⁺ filters through the glomerulus. However, to maintain a net Ca²⁺ balance, >98% of Ca²⁺ filtered at the glomerulus must be reabsorbed within the nephron, while in the PTs a major portion of filtered Ca²⁺ (∼65%) is passively reabsorbed (278), and the remaining 20% is reabsorbed in the TAL of Henle’s loop (58). Finally, regulation of Ca²⁺ occurs primarily in two segments of the distal nephron: the late part of the DCT (DCT2) (279) and the CNT (57). The DCT has two short segments: DCT1 (early) and DCT2 (late). Protein channels such as Na⁺-K⁺-ATPase, Na⁺/Ca²⁺ exchanger isoform 1 (NCX1), and plasma membrane ATPase type 1 b (PMCA1b) are ubiquitously expressed and found along the DCT2/CNT regions (57) in the human (280), rabbit (281), and mouse (58). However, DCT1 has been shown to express thiazide-sensitive Na⁺/Cl⁻ cotransporter (NCC) and transient receptor potential melastatin subtype 6 (TRPM6) (59). The data demonstrated that the signal for NCC expression was strongest in cells of the DCT in both rats and humans, but it extended into the CNT in humans (59). Although the DCT was an acceptable site of expression for NCC in rodents, the site for the rabbit remained controversial. A combination of in situ hybridization and immunocytochemistry concluded that the DCT and not the CNT is the predominant site of NCC mRNA expression in rabbits as well (282). Furthermore, there are additional similarities between the DCT2 and CNT segments as both express transient receptor potential vanilloid type 5 (TRPV5) and the vitamin D-dependent Ca²⁺-binding protein calbindin-D28k (58). Calbindin-D28k has been shown to be present in the DCT and CNT of rat kidneys (283) and in the DCT and CCDs of rabbits (284) and in mice (285). Since the tight junctions in these segments are impermeable to Ca²⁺, Ca²⁺ reabsorption in the DCT2/CNT is mediated primarily by active transepithelial transport against the electrochemical gradient. Apart from these two segments, a mathematical model of PCs describing the synergistic action of the Na⁺/Ca²⁺ exchanger and Ca²⁺-regulated apical Na⁺ permeability offered important insights into Ca²⁺ dynamics in the CCD (286).

Most of the proteins that are filtered through the glomerulus are reabsorbed in the PT by receptor-mediated endocytosis via concerted action of the multiligand megalin-cubilin high-capacity receptor complex (287). It is believed that, under physiological conditions, the distal parts of the nephron avidly reabsorb filtered proteins, but neither the DCT nor CD involve receptor-mediated endocytosis for protein endocytosis (214). Moreover, the capacity for protein absorption in these segments is augmented when the efficiency of protein reabsorption in the PT is compromised after glomerular or PT damage. A study by Langelueddecke et al. (60) has shown that the receptor for neutrophil gelatinase-associated lipocalin (rodent 24p3/human lipocalin-2) is expressed in the DCT (not the PCT) and IMCDs of rat and mouse kidneys, where it orchestrates high-affinity receptor-mediated endocytosis of low (metallothionein)- and high (transferrin and albumin)-molecular-weight proteins, as tested in transiently transfected cultured cells. Thus, against the existing dogma, this study has cogently demonstrated that when proteins escape reabsorption in the PT due to a loss in its reabsorptive capacity following glomerular or PT damage, a surrogate high-affinity protein reabsorption mechanism in the distal nephron kicks in to compensate for the loss by carrying out exhaustive protein reabsorption, thereby limiting protein losses associated with several inherited or acquired renal diseases. In further support of this, a recent study by Weisz and colleagues (288) demonstrated the remarkable ability of PT cells to retrieve urinary albumin under normal and nephrotic conditions. The study concluded that while cubilin is responsible for the uptake of filtered albumin under normal conditions [a finding bolstered by another study where mutations in the cubilin gene were associated with albuminuria (289)], megalin had a negligible role. However, expression of megalin is essential for both active (cubilin dependent) and passive (incorporation in endocytic vesicles) uptake of albumin under normal and nephrotic states, respectively. Thus, when cubilin-mediated uptake is saturated, a second mechanism involving megalin-dependent passive uptake of albumin via its incorporation into endocytic vesicles accounts for the enormous capacity of PT cells to reclaim filtered albumin under nephrotic conditions (288).

THE COLLECTING DUCT

Originally, the CD was believed to not possess any specialized function except water reabsorption. Lately, essential and highly specific functions have been ascribed to this segment of the nephron as well. The main epithelial cell types present in the CD are PCs and ICs. ICs are specialized epithelial cells responsible for regulating acid-base homeostasis through the exchange of various ions via ion channels and pumps (290). There are three different types of ICs: type A (IC-A), type B (IC-B), and type C (non-A/non-B ICs) (291).

Due to the presence of H⁺-ATPase vacuolar-type proton pumps on the apical/luminal side, IC-As are the acid-secreting cells responsible for urinary acidification. While they also express anion exchanger 1 [Cl⁻/${\text{HCO}}_3^{-}$ exchanger (SLC4A1)], on the basolateral side, IC-As are marked by the absence of pendrin (SLC26A4), another Cl⁻/${\text{HCO}}_3^{-}$ exchanger. Additionally, SLC26A11, an electrogenic Cl⁻ transporter as well as a Cl⁻/${\text{HCO}}_3^{-}$ anion exchanger, are also expressed at the apical pole of the IC-As (61). Loss-of-function mutations in the SLC4A1 gene cause abnormal functioning of kidney anion exchanger 1, leading to either autosomal dominant or recessive type 1 distal renal tubular acidosis syndrome. This results in defective bicarbonate reabsorption through the basolateral side of IC-As, leading to decreased urinary acidification and hyperchloremic, hypokalemic metabolic acidosis (292). On the other hand, inherited forms of distal renal tubular acidosis most often result from dysfunctional IC-As. Due to lack of acid secretion from IC-As, patients with distal renal tubular acidosis exhibit acidemia. The absence of acid secretion translates into an alkaline urine that ultimately leads to nephrocalcinosis and/or nephrolithiasis, along with hypokalemia (293).

IC-Bs, on the other hand, are basic/bicarbonate-secreting cells. Unlike IC-As, IC-Bs express H⁺-ATPase pumps on the basolateral pole and pendrin on the apical pole (61). IC-Bs have an electroneutral NaCl transport/reabsorption pathway that involves pendrin and the Na⁺-driven Cl⁻/${\text{HCO}}_3^{-}$ exchanger at their apical membrane (61).

Type C (non-A/non-B) ICs express both H⁺-ATPase pumps and pendrin on the apical side (61). All these PCs and ICs express major transporter proteins to maintain cellular osmolarity and acid-base homeostasis.

PCs express the epithelial Na⁺ channel (eNaC) and aquaporin 2. Other unique PC-associated markers are arginine vasopressin receptor 2 (AVPR2) and ROMK1. AVPR2 is a member of the G protein-coupled receptors. It has been postulated that in the kidney, AVPR2 is involved in electrolyte and acid-base balances and is located in various cell types within the kidney but primarily in PCs of the CD (62). Earlier studies have reported expression of AVPR2 mRNA in medullary CDs and CCDs in the developing kidneys of 16- and 19-day-old rat embryos (294). Later, evidence was provided for the localization of AVPR2 protein in the TAL, DCT, CNTs, and CDs in rat kidneys (295). While X-linked recessive congenital nephrogenic diabetes insipidus is caused by loss-of-function mutations in the AVPR2 gene, gain-of-function mutations in this gene lead to NS of inappropriate antidiuresis (296). ROMK1 is an ATP-dependent inwardly rectifying K⁺ channel (297) localized to the apical membranes of the distal nephron in the cortex and outer medulla, including the TAL and CNT but not in the glomerulus, PTs, and inner medulla of rats (298, 299). ROMK1 is positively regulated by CaSR. In humans, ROMK is encoded by the KCNJ1 gene, and loss-of-function mutations lead to impairment of salt retention in the kidney and are associated with a metabolic wasting disease termed “Bartter’s syndrome.” Genetically, Bartter syndrome is divided into five subtypes based on the mutations caused in five genes: type 1 (SLC12A1 encoding NKCC2), type 2 (KCNJ1 encoding ROMK1), type 3 (CLCNKB encoding ClC-Kb), type 4 (BSND encoding barttin, a subunit for ClC-Ka and ClC-Kb), and type 5 (CASR encoding CaSR). A closely related disease, termed Gitelman syndrome, is caused by dysfunctional NCC located in the tubules (300–304).

One of the leading causes of kidney failure is uncontrolled inflammation. Interestingly, proinflammatory pathways can be triggered even without an active infection. This process is termed as sterile inflammation. It has been shown that the purinergic receptor P2Y14 (GPR105) is expressed in abundance in ICs of the mouse CD and is responsible for mediating sterile inflammation (63). When cells are injured, they release UDP-glucose, which is a damage-associated molecular pattern molecule, which, in turn, activates P2Y14 (305). Although there is an established link between signaling via purinergic receptors and renal inflammation, the precise mechanism underlying their action is still unknown. Interestingly, copious purinergic receptors are expressed in the kidney, and deregulation of purinergic signaling is correlated with hypertension, CKD, acute kidney injury (AKI), DN, and glomerulonephritis (306, 307).

Urea transporters (UTs) are composed of two main types: UT-A and UT-B. While UT-As are majorly expressed in kidneys, UT-Bs are expressed in other tissues as well, including the brain. UT-A1, UT-A2, and UT-A3 are expressed in the PCs, IMCD, and TAL, respectively. The major function of UTs in the kidney is to concentrate urea. Currently, UT-A and UT-B are targets for diuretic drug treatment (64, 65, 308, 309).

CELL-BASED MARKERS

The kidney is composed of a series of specialized cell types possessing unique permeability characteristics and transport capabilities. Here, we summarize the markers present in major individual renal cells, which are necessary for the establishment of cellular interactions, polarity, filtration, and transport.

Endothelium-Associated Markers

Glomerular capillaries are lined with highly specialized (e.g., flattened and fenestrated) endothelium, which regulates the rapid flux of fluid and small solutes across the glomerular barrier (310). Glomerular ECs, along with the negatively charged luminal glycocalyx layer (consisting of a network of proteoglycans and glycoproteins), interact with both podocytes and mesangial cells and regulate intraglomerular hemodynamics, vascular reactive oxygen species production, prothrombotic and antithrombotic factors, and fibrosis (311, 312).

Nitric oxide (NO), a lipophilic gas has been implicated in many important physiological functions associated with kidneys such as regulation of renal blood flow and glomerular filtration rate, promotes natriuresis and diuresis so that kidneys are equipped to adapt to variations in dietary salt intake and thus help maintain normal blood pressure. NO is produced by a catalytic reaction that involves NO synthase (NOS). Three isoforms of NOS have been identified so far: neuronal NOS (nNOS or NOSI), inducible NOS (iNOS or NOSII), and endothelial NOS (eNOS or NOSIII) (313). Interestingly, expression of all three isoforms has been detected in kidney tissues but with different distribution profiles (66). For example, eNOS was found to be expressed in ECs of glomeruli and peritubular capillaries in the cortex and in the ECs of vascular bundles in the medulla, with a stronger immunoreactivity in cells of the renal medulla than in the cortex (314). Its expression was also found in epithelia of the TAL of the loop of Henle and CD (315). nNOS was detected predominantly in tubular epithelial cells of the MD (314) in the kidneys of rats, mice, guinea pigs, and rabbits; the intensity of the signal was much weaker in humans and pigs (316). Additionally, nNOS was also detected in the IMCD and outer medullary CD, efferent arterioles, Bowman’s capsule, and various cells at the cortical TAL in rats (317). While the mRNA transcript corresponding to iNOS has been detected in the S3 segment of the PT, cortical and medullary TAL, DCT, CCD, and IMCD in normal rat kidneys (318), expression of iNOS protein was observed in the PT, TAL, and ICs of the CD (205). Furthermore, the expression patterns of these isoforms are uniquely distinct in the developing kidneys from that observed in adults (313). Nevertheless, owing to the diffusibility of NO, it is possible that even though it is not produced locally in a nephron segment, it can still have an impact on that segment or neighboring structures. Chronic renal disease is characterized by compromised NO bioactivity. Levels of eNOS mRNA and protein as well as its activity are reduced substantially in human EC cultures exposed to erythrocytes from patients with ESRD (319). Furthermore, eNOS knockout mice develop congenital anomalies, including glomerular hypoplasia (i.e., reduction in the number of nephrons), tubular cell death, and atubular glomeruli (320, 321). eNOS deficiency has also been shown to predispose podocytes to injury in diabetes (322).

Platelet endothelial cell adhesion molecule-1 (PECAM-1 or CD31), a 130-kDa glycoprotein, consists of a single-chain molecule with an extracellular domain (composed of six Ig-like domains), a transmembrane domain, and a relatively long cytoplasmic domain. Since PECAM-1 exists in at least eight isoforms generated by alternative splicing of multiple exons encoding its cytoplasmic tail, differential combinations of these exons impact its interactions with potential intracellular signaling molecules. PECAM-1 expression is found in fenestrated renal glomerular and peritubular ECs (323), which is lost in sclerotic or fibrotic areas (67).

Circulating active levels of von Willebrand factor (vWF; factor VIII-related antigen) have shown to be increased in CKD and ESRD (68). Furthermore, vWF has been shown to contribute to the pathogenesis of acute IRI (324). Although staining for vWF has always been diffusively positive or completely negative in the fenestrated endothelium of glomeruli, its expression on ECs increases in renal glomeruli of patients with hypertension (323).

EC junctions such as tight junctions (zonulae occludens), adherent junctions (zonulae adherens), and gap junctions (desmosomes) play an important role in cellular homeostasis. Vascular-endothelial cadherin (VE-cadherin), a Ca²⁺-dependent transmembrane glycoprotein comprising the adherent junctions between ECs, regulates vascular integrity, endothelial barrier function, and angiogenesis (69). During inflammation, the extracellular domain of VE-cadherin undergoes proteolytic cleavage followed by shedding into the circulation as sVE-cadherin (325). Although the association of tubular epithelial injury with AKI has been exhaustively studied, the contribution of endothelial injury has not been ascertained. Fairly recently, a study by Yu et al. (326) demonstrated a direct correlation between shedding of sVE-cadherin and severe AKI and in patients with severe organ dysfunction and sepsis, suggesting that breakdown of endothelial adherent junctions contributes to renal pathogenesis. Furthermore, uremia has been reported to alter cell-to-cell junctions in ECs and impact the expression of VE-cadherin and ZO-1 (327).

Fibroblast-Associated Markers

In a seminal study on kidney fibrosis by Strutz and coworkers (70), fibroblast-specific protein 1 (FSP-1) was recognized as a marker for cells undergoing epithelial-mesenchymal transition (EMT). Authors showed that FSP-1 protein was expressed by fibroblasts and not epithelial cells and thus coined the term “fibroblast-specific protein-1.” Furthermore, the study demonstrated that while there was a low incidence of FSP-1-positive cells in healthy kidneys, a massive increase was observed in tubular epithelia in renal fibrosis, indicating that under pathological conditions, interstitial fibroblasts originate from tubular cells undergoing EMT. This work led to the widespread use of FSP-1 antibodies for the identification and detection of fibroblasts and tubular epithelium undergoing EMT (328), even though it was a persistent matter of debate in the field.

Nearly all progressive renal diseases funnel into renal fibrosis as an ultimate consequence culminating in ESRD. PDGFs essentially participate in driving processes that ultimately lead to fibrosis (329). The PDGF system consists of four isoforms: PDGF-A, PDGF-B, PDGF-C, and PDGF-D and two receptors (PDGFR-α and/or PDGFR-β chains). These receptors dimerize upon ligand binding. Whereas PDGF-A binds to the PDGFR-αα dimer only, PDGF-B has been shown to function as a ligand for all receptors. PDGF-C, on the other hand, binds to PDGFR-αα and PDGFR-αβ, whereas PDGF-D predominantly binds to PDGFR-ββ. While these two receptor chains are constitutively expressed by mesangial cells, fibroblasts, and vascular smooth muscle cells, they are not expressed by epithelial cells, such as podocytes and tubular cells, under either normal or pathological conditions (71). Upregulated PDGF and its receptor expression have been reported in nearly all rodent models of kidney injury including puromycin-induced FSGS, DN, murine lupus nephritis, IRI, mesangioproliferative anti-Thy 1.1 glomerulonephritis, nephrotoxic nephritis, ANG II-induced renal damage, murine IgA nephropathy, renal transplantation, and their corresponding human renal diseases (330).

For organs such as the kidney, where there is negligible cell turnover, cells from all renal compartments undergo decay and thus need to be replaced. In the absence of such a regeneration process, a progressive loss of the structural and functional integrity is inevitable. PDGF-B and fibroblast PDGFR-β signaling are known to assist wound healing in other organs. In contrast to other organs, little is known about the proregenerative effects of renal fibroblasts in the kidney. In a study on proximal tubular regeneration, Schiessl et al. (331) demonstrated targeted migration of interstitial PDGFR-β-positive cells to replace the cellular loss upon ablation of single tubular cells by focused laser exposure.

Mesangial Cell-Associated Markers

Renal mesangial cells are contractile irregularly shaped cells within the renal corpuscle (332), which are exclusively located in the glomerulus. They maintain the mesangial matrix and provide structural support for glomerular capillary loops. Other functions include communication with other glomerular cells by secreting growth factors and contribution to the regulation of glomerular capillary flow. Therefore, mesangial cell injury can lead to FP effacement and proteinuria (333).

Transient receptor potential canonical member 1 (TRPC1), alongside with other TRPCs (possessing seven isoforms), has been identified in cultured human mesangial cells by Western blot analysis (73) and by an immunofluorescence study in the rat kidney (72). Goel and colleagues (72) showed that TRPC1 is expressed in various parts of the rat kidney such as the tubules and DTL; however, they found that TRPC1 and transient receptor potential canonical member 4 (TRPC4) are the sole ion channels in the mouse mesangial cell line. Sours-Brothers colleagues (334) also showed that TRPC1 could physically interact with TRPC4 and TRPC6. The same group later demonstrated the possibility that TRPC1/TRPC4 interacts with stromal interaction molecule 1, which is a form of the store-operated Ca²⁺ channel, suggesting that TRPC1 could be part of the regulation of store-operated Ca²⁺ channel-mediated Ca²⁺ entry (334).

Renal expression of PDGFs and their receptors (PDGFRs) are well documented, as discussed above. Both PDGFR-α (74) and PDGFR-β (335, 336) are constitutively expressed by glomerular mesangial cells, interstitial fibroblasts, and vascular smooth muscle cells. PDGFR-β has been recognized as a marker for mesenchymal cells, and focal expression of PDGFR-β has also been described on parietal epithelial cells. Although PDGFRs are not expressed on glomerular ECs in vivo, glomerular ECs did express PDGFR-α and responded to PDGF-CC in vitro. Thus, PDGFR-α/PDGF-AA and PDGF-CC might have a role in glomerular capillary healing and angiogenesis (337).

Epithelial Cell-Associated Markers

The general architecture of renal epithelial cells as well as their pattern of interaction with each other and with the surrounding extracellular matrix are important for the extracellular scaffolding and maintenance of the functionally distinct segments of the kidney. Due to their ubiquity and unique characteristics, epithelial cell lines have been used widely in cell-based assays to investigate the alterations in epithelial physiology and signaling and stress pathways. Although cell culture models are widely used, it is imperative to mention that the choice of cell line and their growth conditions can significantly alter the results because they may not fully recapitulate the key aspects of a cell type or its in vivo environment. For instance, a recent study by Khundmiri et al. (338) demonstrated that transcripts expressed in none of the 14 PT cell lines fully matched the transcriptome of native PT cells. The transcripts not only matched partially but also variably with the highest percent match obtained in the opossum kidney (OK cells, 45% of proximal marker genes) followed by the pig kidney (LLC-PK1, 39%), and lower percent matches were seen in human cell lines (HK-2 cells, 26%) and lines from rodent kidneys (NRK-52E cells, 23%) (338). Park et al. (339) also substantiated this by demonstrating that OK cells cultured under continuous orbital shear stress are morphologically and functionally more similar to PT cells in vivo compared with cells maintained under static conditions, necessitating a thorough validation of conclusions derived from cell culture experiments.

ENaC [also known as Na⁺ channel nonneuronal 1 (SCNN1) or amiloride-sensitive Na⁺ channel (ASSC)] is expressed on the apical pole of epithelial cells in the distal nephron (75). The functional channel consists of an assembly of three subunits encoded by three genes: α (SCNN1A), β (SCNN1B), and γ (SCNN1G), with a stoichiometry of 1:1:1. These channels belong to the ENaC/degenerin family of proton-gated cation channels that mediate transport of Na⁺ through the apical membrane from the lumen into the epithelial cell (340). Since ENaCs precisely regulate Na⁺ balance, they play a key role in maintaining blood pressure (341). Mice devoid of any ENaC subunit succumb to kidney dysfunction and do not survive (342). Similarly, germline mutations (frameshift or missense) in an allele of SCNN1A, SCNN1B, or SCNN1G genes causes autosomal dominant Liddle syndrome (pseudoaldosteronism), characterized by an early onset of salt-sensitive hypertension with a low level of K⁺, metabolic alkalosis, inhibition of renin activity, and aldosterone secretion (343).

Cadherins represent one of the most important molecules that participate in renal epithelial cell-cell adhesion and are usually found at the adherens junction of epithelial cells (76). While over 50 cadherins have been described (344), the most exhaustively characterized are the classical type I cadherins: E-cadherin and N-cadherin (345). Cadherins are integral, transmembrane proteins with an extracellular domain containing Ca²⁺-binding sites and regions that impart them adhesive properties (346). The intracellular domain interacts with β-catenin, which is bound to α-catenin, which, in turn, connects the entire complex to the actin cytoskeleton (347). In the adult kidney, the predominant cadherins are classical N-cadherin and E-cadherin (348). The other forms, like OB-cadherin, R-cadherin, and K-cadherin, are transiently expressed during different stages of development (349). Interestingly, atypical kidney-specific cadherin (Ksp-cadherin) has been reported. Ksp-cadherin lacks the cytoplasmic catenin-binding domain, and this is how it differs from classical cadherins (350, 351). However, the functional relevance of this atypical cadherin is still elusive. Furthermore, there is a plethora of evidence that indicates differential expression of these classical cadherins in various segments of the nephron. However, these mapping studies were equivocal and difficult to interpret because of the use of different cadherin panels, different species, and poor techniques to identify the specificity of tubular segments. Therefore, a study by Prozialeck et al. (349) used dual immunofluorescence labeling technique to specifically target E-cadherin or N-cadherin, along with antibodies that serve as markers for specific nephron segments, in the adult rat kidney. The results showed that while N-cadherin is the predominant cadherin present in the PT, E-cadherin is abundant in the DCT, CD, and most medullary segments and is expressed in negligible amounts in the PT (349). In the human kidney, a complex pattern of cadherin expression has been observed (352). Epithelial cells of PTs were found to express N-cadherin and cadherin-6 (K-cadherin) (353), whereas in DCTs, E-cadherin (348) and cadherin-16 (Ksp-cadherin) (354) were present. Additionally, R-cadherin (355) and cadherin-8 (356) have been reported in the differentiating metanephric mesenchyme when kidney development is initiated. A change in E-cadherin expression or cadherin switch (357) in the kidney has been used as a surrogate marker of EMT (358).

Macula Densa Markers

The MD is a unique collection of epithelial cells located at the distal end of the TAL. Cells in MD are known to function as salt sensors that detect changes in distal tubular fluid composition. In response to these changes, MD cells transmit chemical signals to the juxtaglomerular apparatus to control renal hemodynamics, filtration rate, and renin release, a process known as tubuloglomerular feedback (359). Changes in NaCl concentration are fine tuned by a concerted action of ion transport-related intracellular events. NaCl entry into the MD cells is primarily regulated by apically located NKCC2 (BSC1). Interestingly, of the three known splice variants of NKCC2 (A, B, and F), which differ in their ionic affinities and regulation and vary in their distribution along the TAL (360–362), MD cells were shown to express the B isoform of NKCC2 (77), which has the highest affinity for Cl⁻ relative to the other two isoforms (360, 362). With the cotransport of NaCl, the increased cellular Cl⁻ is expelled through a basolateral channel, which results in cell depolarization along with a modest increase in cytosolic Ca²⁺ (78). Using patch-clamp experiments, Lapointe et al. (363) reported that upon an elevation in Ca²⁺ level, a nonselective cation channel, which is like transient receptor potential melastatin subtype 4, with moderate Ca²⁺ permeability gets activated in the basolateral membrane of MD cells. Along with NKCC2, MD cells also possess Na⁺/H⁺ exchange activity that causes cell alkalization. Functional and immunological techniques have identified that this is mediated predominantly by apical NHE2 and basolateral NHE4 isoforms (78). Traditionally, studying live MD cells has been a challenge due to their low number in each nephron. However, in a recent study, Gyarmati et al. (364) developed a method to analyze live MD cells. A chimera of a MD-specific marker, nNOS, and a tamoxifen-inducible membrane targeted green fluorescent protein was generated, and, using this, they were able to visualize the three-dimensional projections (also called the maculopodia) of MD cells (364).

SUMMARY

One of the most outstanding fundamental questions in human physiology is what types of cells form different tissues in an organ so that it can perform the requisite physiological functions. Furthermore, within those specific cell types, what are the defining marker proteins and how can an imbalance (environmental/extrinsic factors) or mutations (genetic/intrinsic factors) in those proteins lead to debilitating disease phenotypes? Historically, many cell markers have been identified through traditional and conventional cell biology experiments (e.g., immunofluorescence or immunohistochemistry). These cell markers were then expansively used to identify, isolate, or classify cell types of interest by exploiting techniques such as FACS, proteomics, confocal, deconvolution, or multiphoton microscopy. However, with the advent of scRNA-seq, more accurate characterization of cell-based markers that are unique to different cell types has become possible. Although scRNA-seq is a promising tool to investigate differential gene expression between control and impacted cells, it does not provide insights into gene regulation. Since understanding the epigenetic program that leads to a unique cell type in the kidney is essential, seminal work from Miao and colleagues (365) providing a chromatin map of the developing and adult mouse kidney based on epigenome classification at single cell resolution is extremely timely. Additionally, a recent study has exploited a FACS-based approach to first isolate and enrich specific nephron segments that have ill-defined boundaries like DCT1/DCT2 and DCT2/CNT and then couple it with scRNA-seq to resolve the cellular composition and transcriptional profiles of minority epithelial cell types in the mouse kidney distal nephron (366). However, even with the large amount of relevant and useful data that emerges from contemporary scRNA-seq and other conventional techniques, researchers find this information on cell-based markers overwhelming and strewn over thousands of publications. Furthermore, the choice of a reliable cell surface marker has always remained a controversy. This review on renal cell-based markers is a small attempt to bridge this big chasm. We first generated a nephron map that provided a cellular landscape and then compiled a list of proteins unique to a specific cell type, which is in a specific segment of nephron in the human kidney (Fig. 1). We believe that this manually curated database will serve as a comprehensive resource for renal cell markers, which can facilitate researchers worldwide to comprehensively and reliably apply the existing information on cell-based markers to their specific study. Furthermore, as we continue to advance our understanding of these proteins and multiprotein assemblies in the kidney, such information on cell surface markers will prove indispensable in the diagnosis and treatment of renal diseases. Nonetheless, considering the speed in which rare cell types and new marker proteins are identified, it is difficult to cope up and scope up with the ever-expanding arsenal of information. Therefore, we acknowledge that our inventory is only comprehensive and not exhaustive.

Figure 1.

Renal cell markers. Shown is a cartoon illustrating the spatial arrangement of a nephron with the various domains segmented and associated markers indicated in a box. The image is not drawn to scale. PDGFR, platelet-derived growth factor receptor; α-SMA, α-smooth muscle actin; TRPC, transient receptor potential canonical ion channel; eNOS, endothelial nitric oxide synthase; vWF, von Willebrand factor; PECAM-1, platelet endothelial cell adhesion molecule-1; NKCC2, Na⁺-K⁺-2Cl⁻ cotransporter; NXC1, Na⁺/Ca²⁺ exchanger isoform 1; FSP-1, fibroblast-specific protein 1; SCNN1, Na⁺ channel nonneuronal 1; Sca-1, stem cell antigen-1; NCC, Na⁺/Cl⁻ cotransporter; GLEPP1; glomerular epithelial protein 1; WT-1, Wilms’ tumor-1; CLIC5, Cl⁻ intracellular channel protein 5; CaSR, Ca²⁺-sensing receptor; NaS1, Na⁺-sulfate cotransporter; NaPi, type II Na⁺-coupled phosphate; AE1, anion exchanger 1; UT-A1, urea transporter A1.

DISCLOSURES

J.R. has patents on novel strategies for kidney therapeutics and stands to gain royalties from their commercialization. He is the co-founder of Walden Biosciences (Cambridge, MA), a biotechnology company in which he has financial interest, including stock. Other authors have nothing to disclose and there are no competing interests.

AUTHOR CONTRIBUTIONS

J.R. and M.M.A. conceived the idea and manuscript design; S.A., Y.R.S., O.K.P., and M.M.A. drafted the manuscript; S.A., Y.R.S., and M.M.A. prepared figures and table; S.A., Y.R.S., O.K.P., and M.M.A. edited and revised the manuscript; S.A., Y.R.S., O.K.P., J.R., and M.M.A. approved the final version of the manuscript.