The Pharmacology of the SLC15A4‐TASL Complex, an Emerging Target for the Treatment of Lupus

ABSTRACT

SLC15A4 is a member of the solute carrier superfamily that has been strongly linked to the pathogenesis of the auto‐immune disease SLE. The endo‐lysosomally localized SLC15A4 has traditionally been considered a proton‐dependent histidine and peptide transporter, but has also been shown to facilitate TLR7, 8, and 9 induced IFN‐alpha signaling in B cells and pDCs. New research has shown that SLC15A4 facilitates endo‐lysosomal TLR signaling through a scaffolding interaction with a newly identified signal transducer, termed TASL. This interaction enables TLR7, 8, and 9 induced IRF5 activation, which is a crucial pathway in the pathogenesis of lupus. SLC15A4 is therefore an attractive target for the development of drugs for the treatment of SLE, and several small molecule SLC15A4 binders have now been reported which are capable of inhibiting TLR7, 8, or 9 signaling. This review summarizes the SLC15A4:TASL complex, its link to SLE, existing evidence for the transporter activity of SLC15A4, the current small molecule binders, and tools available for further drug discovery efforts.

1. Introduction

SLC15A4 (Solute Carrier Family 15 member 4) is an endo‐lysosomally located protein from the solute carrier family that has been heavily implicated in auto‐immune disease, particularly systemic lupus erythematosus (SLE), through human genome wide association studies (GWAS) and mouse knockout (KO) studies [1]. As solute carriers are known to be tractable targets for small molecule modulation, SLC15A4 has been proposed as a novel target for the treatment of SLE [2]. Despite increased interest in the protein, the pharmacology of SLC15A4 remains poorly understood. Originally identified as a proton‐histidine co‐transporter [3], early reports linked SLC15A4 to NOD1/2 signaling and suggested that the protein acted to transport bacterially‐derived peptides from the lysosome to the cytosol [4]. Mouse KO studies subsequently demonstrated that SLC15A4 was essential for toll‐like receptor 7 (TLR7)‐mediated interferon‐alpha (IFN‐alpha) production in plasmacytoid dendritic cells (pDCs) [5] along with B cell class switch recombination and auto‐antibody production [6]. More recently a role of SLC15A4 in tethering a novel signal transducer TASL (TLR adaptor interacting with SLC15A4 on the lysosome) to the endo‐lysosome to enable TLR7, 8 and 9 signaling has been established [7, 8]. The relevance of this interaction to lupus was strengthened by the identification of TASL as an SLE risk gene in GWAS studies [9] and protection from disease following TASL KO in lupus mouse models [10]. Several small molecule SLC15A4 binders have now been reported, with a subset of these compounds reported to prevent TASL complex formation [11, 12]. This review will summarize the known functions of SLC15A4 and the SLC15A4:TASL complex, outline the emerging class of small molecule SLC15A4 inhibitors and summarize learnings for future small molecule drug discovery.

1.1. SLC15A4 and SLE

SLE is characterized by an overproduction of type I IFN and production of auto‐antibodies which target double‐stranded DNA and ribonucleoproteins. The disease has a diverse set of manifestations and can affect the skin, joints, kidneys, cardiovascular system, and central nervous system [13]. The pathogenesis of the disease is likely heterogeneous; however, aberrant clearance of auto‐antigens, chronic production of pro‐inflammatory cytokines, and dysregulation of B cell tolerance have all been implicated [13, 14].

Mainstay therapies for SLE include anti‐malarials, glucocorticoids, and other broad immunosuppressants [14, 15]. With a greater understanding of the pathogenesis of SLE, there has been a push for newer biological agents targeting particular arms of the disease—namely the inhibition of autoantibodies via B cell inhibition or depletion, or targeting of the IFN pathway [13, 14, 16]. Currently, two main agents are used to target the dysregulation of B‐cells in SLE. The BAFF (B‐cell activating factor) targeting monoclonal antibody Belimumab was initially approved in 2011, and rituximab, a monoclonal antibody targeting CD20, is currently used off‐label [17, 18]. Anifrolumab, a biologic targeting the Type I IFN receptor, was also recently approved [19, 20, 21]. Despite the emergence of these targeted approaches, SLE therapies remain constrained by limited efficacy and adverse effects [21].

There is strong genetic evidence implicating SLC15A4 and other endosomal TLR pathway signaling components in the development and progression of lupus. In 2009, two SNPs were identified in the Chinese Han population (rs10847697, odds ratio 1.26 and rs1385374, odds ratio 1.26) which increased susceptibility to lupus [22]. Since this study, numerous other SNPs in SLC15A4 have been implicated in lupus susceptibility from a wide range of ethnic groups [9, 23]. The wider TLR pathway is also heavily implicated in SLE (Figure 1). TLR7, one of the receptors upstream of SLC15A4, is strongly linked to SLE with gain‐of‐function mutations being shown to drive human disease [24]. Downstream of SLC15A4, both TASL and IRF5 have been strongly linked to SLE through GWAS studies [9].

FIGURE 1.

Activation of the SLC15A4‐TASL‐IRF5 axis in pDCs and B cells is central to lupus pathophysiology. Sensing of nucleic acids (red) by endo‐lysosomal toll‐like receptors (TLR7, 8 & 9) activates multiple kinases through adaptor molecule MyD88. This leads to phosphorylation of IRF5 following its recruitment to the endo‐lysosomal membrane by the SLC15A4:TASL complex which acts as a scaffold for IRF5 activation. TLR7, 8 and 9 activation also leads to NF‐κB phosphorylation. These transcription factors then translocate to the nucleus where they drive expression of pro‐inflammatory genes. Endo‐lysosomal TLRs, SLC15A4 and TASL are highly expressed in pDCs and B cells, both of which are key drivers of lupus pathology. The SLC15A4‐TASL‐IRF5 axis is central to disease‐relevant effector functions, including production of pro‐inflammatory cytokines and type I interferon (IFN) which results in an interferon‐stimulated gene (ISG) signature, B cell activation and differentiation, class‐switch recombination (CSR) and auto‐antibody production.

In mice, SLC15A4 was first linked to pDC effector functions in a study that screened 4500 genetic mutants for defects in IFN‐alpha production following challenge with a TLR9 agonist [5]. Using this screen, the researchers identified a non‐responding mouse with a mutation in SLC15A4, which they termed feeble. Despite having no defect in pDC frequency compared to WT mice, pDC cultures from feeble mice had impaired type I IFN, TNF, IL‐6, and IL12p40 responses to TLR7, 9, or viral stimuli. The same group then went on to show that the feeble mouse was unable to effectively control LCMV infection [25].

Several groups have tested the effect of SLC15A4 KO in genetically or chemically induced murine models of SLE, demonstrating that SLC15A4 KO reduces pathogenic auto‐antibody production amongst other disease symptoms [6, 8, 26, 27, 28]. Murine KO models have also demonstrated the key role of SLC15A4 in class‐switch recombination within B cells, with multiple studies showing the crucial role of SLC15A4 in IgG2a/c production [6, 29, 30]. SLC15A4 KO has also been shown to lower disease severity in murine models of colitis and psoriasis [31, 32].

1.2. The Interaction of SLC15A4 With TASL

Historically, the link between SLC15A4 and auto‐immune disease was attributed to the protein's transport function, either through the direct transport of NOD ligands or by modulation of TLR signaling through changes in pH (covered in more detail in the following section). The significance of the transporter activity of SLC15A4 in auto‐immune disease has since been called into question by the discovery of TASL [7]. The protein, formerly known as CXorf21, had previously been linked to SLE in GWAS studies [9], and was shown to be a crucial adaptor molecule, binding to the cytoplasmic interface of SLC15A4 and facilitating IRF5 activation (Figure 1). The study by Heinz and colleagues demonstrated that TASL facilitates the phosphorylation and dimerization of IRF5 following endosomal TLR‐induced MyD88 activation. The interaction of TASL and IRF5 occurs via the highly conserved pLxIS motif, which is common amongst other IRF adaptor proteins. Heinz and colleagues also showed that genetic deletion of TASL impairs IRF signaling whilst leaving NF‐kB/MAPK signaling intact. Whilst the exact mechanism for SLC15A4:TASL induced IRF5 activation remains unclear, the authors demonstrated that the kinases IKKBeta, IRAK4 and TAK1 were necessary for CpG‐B induced IRF5 phosphorylation [7].

Mutagenesis was used to demonstrate that the lysosomal localisation and function of TASL is dependent on its interaction with SLC15A4, specifically an interaction that was predicted by Alphafold to be mediated through the insertion of the N terminus of TASL into the inward open binding site of SLC15A4 [7]. Supporting the essential role of SLC15A4 as a scaffold protein in TASL recruitment, a subsequent study by Zhang and colleagues showed that TLR7, 8 or 9 signaling can be restored in the absence of SLC15A4 if TASL is artificially anchored to the endo‐lysosomal membrane using a fusion of TASL to a LAMTOR truncate and other lysosomally localized proteins [8].

Structural studies utilizing cryogenic electron microscopy (cryo‐EM) have since demonstrated that TASL recruitment induces a conformational switch in SLC15A4 from the apo lysosomal lumen‐exposed outward‐open dimeric protein to the monomeric inward‐open TASL bound conformation [33, 34]. The structure of the TASL:SLC15A4 complex confirmed the binding mode predicted previously by Alphafold (Figure 2), demonstrating that E465, a residue that was previously thought to mediate peptide transport [35], is a key mediator of TASL binding and subsequent TLR signaling [7, 8, 33].

FIGURE 2.

Apo (left) and TASL‐bound (right) SLC15A4 structure (adapted from Chen et al., 2023). In its apo state, SLC15A4 (orange) resides in an outward‐open (lysosomal lumen facing) conformation which does not allow TASL binding. Transition to an inward‐open (cytoplasmic facing) conformation exposes a cavity which permits insertion of TASL's N‐terminal alpha‐helices (cyan), stabilizing the SLC15A4:TASL complex and allowing for downstream signal transduction.

Recently, two independent studies were reported in which TASL KO mouse models were created [10, 30]. Drobek and colleagues observed that the inhibition of IRF5 phosphorylation in both B cells and pDCs was lower than that observed in the equivalent feeble mouse model [30]. The authors then identified a novel murine specific paralogue of TASL which they termed TASL2, showing that double knock‐out (DKO) of both TASL and TASL2 mice resulted in complete p‐IRF5 inhibition. It was observed that DKO mice phenocopied feeble mice, with reduced circulating IgG2c at steady state and inhibition of type I IFN and ISG response following TLR7 or 9 agonist stimulation. Like feeble mice, DKO mice failed to resolve LCMV infection; however, both TASL and TASL2 KO mice were able to resolve infection, indicating a level of redundancy between the proteins. Finally, the authors demonstrated that, like feeble mice, both DKO and TASL KO were protected from auto‐antibody induction in a pristane‐induced SLE model. The observation that TASL KO was sufficient to protect against auto‐antibody production in chemically induced auto‐immune disease was confirmed using both pristane and topical imiquimod in a further mouse TASL KO study [10].

A recent pre‐print has reported a third TASL KO mouse study in which a TASL KO mouse was shown to be protective in the MRL/lpr model of lupus. This study also demonstrated that TASL KO reduced the frequency of disease‐linked age‐associated B cells [36].

The discovery of TASL revealed a key mechanistic link between SLC15A4 and TLR7, 8 and 9 signaling, giving a persuasive rationale for how the protein can mediate pathogenic TLR signaling in SLE. In addition, the complementary genetic links of both SLC15A4 and TASL with SLE suggest that the TASL dependent functions of SLC15A4 are responsible for the protein's role in this disease.

It is possible that SLC15A4 has scaffolding roles beyond its interaction with TASL. A study conducted in HEK293 cells recombinantly expressing SLC15A4 used BioID to suggest an interaction with LAMTOR1 and GTPases RagA, C, and D [37]. Whilst these interactions were not observed when a Mass Spectrometry (MS) based affinity capture methodology was applied to THP‐1 cells [7], co‐immunoprecipitation experiments conducted in HEK293 and WEHI231 cells have demonstrated interactions with components of the Mechanistic Target of Rapamycin (mTOR) machinery [35, 37]. There is some evidence that SLC15A4 KO may modulate mTOR signaling in specific contexts, for example after salmonella infection of murine dendritic cells [38] or IFN‐alpha stimulation of murine B cells [6], however, more work is needed to further define how SLC15A4 interacts with mTOR. The recent structural elucidation of the interaction between the lysosomally localized solute carrier SLC38A9 and the mTOR signaling complex demonstrates that solute carriers can directly interact with this pathway [39]. It has also been suggested that SLC15A4 plays a role in metabolic processes in macrophages possibly through a scaffolding interaction with AMPK that was demonstrated through co‐immunoprecipitation in HEK293 and WEHI231 cells [37].

1.3. SLC15A4 as a Transporter

The SLC15 family consists of five proteins, each containing 12 transmembrane domains. SLC15A1 (PEPT1) and SLC15A2 (PEPT2) are localised to the plasma membrane and SLC15A3 (PHT2) and SLC15A4 (PHT1) to the endo‐lysosome. The subcellular localisation of SLC15A5 is unknown. The structure and function of the other SLC15 family members have been reviewed previously [40, 41]. The cell surface localized transporters SLC15A1 and SLC15A2 are the most well characterized members of the SLC15 family and are known to mediate the uptake of di‐ and tripeptides as well as various drugs [42]. These proteins mediate transport by switching from outward open (extracellular facing) to inward open (cytoplasm facing) states, using the pH gradient as a driving force. These are analogous to the Apo and TASL‐bound states of SLC15A4, respectively [40, 41]. SLC15A3 is the member of the family most closely related to SLC15A4, however the protein dose not interact with TASL [7].

SLC15A4 was first labeled as a histidine and di‐peptide transporter in the original study that cloned the gene from rat brain and retinal extracts [3]. The authors identified the gene as a possible peptide transporter due to homology with SLC15A1, 2 and NTR1 ( A. thaliana ) and observed that recombinant expression of rat SLC15A4 in Xenopus laevis oocytes increased transport of radiolabelled histidine 5‐fold. The researchers found that the histidine transport activity was strongest at pH 5.5 and that at this pH the histidine K _M was 17 μM. The study went on to demonstrate inhibition of histidine transport by a range of unlabelled di‐ and tri‐peptides and to show the transport of radiolabelled carnosine (beta‐alanyl‐L‐histidine). A later study attempted to characterize the transporter activity of human SLC15A4 recombinantly expressed in COS‐7 cells. In contrast to the initial rat experiment, this study identified only a minor increase in histidine and carnosine transport in cells expressing SLC15A4 as compared to the control [43].

The interpretation of both of these initial studies was complicated by the subsequent discovery that SLC15A4 is localized to the endo‐lysosome. A study attempted to reproduce the transport experiments, using a plasma membrane localized SLC15A4 construct recombinantly expressed in MDCK cells, and measuring substrate internalization with HPLC‐MS/MS [44]. In the hands of these researchers, d3‐histidine transport could only be observed with the mutated plasma membrane localized and not wildtype lysosomal localized protein. The researchers recorded a histidine transport K _M of 16.3 μM, although interpretation of these data was complicated by a significant non‐specific component of a similar K _M, indicating that much of the histidine transport was mediated through a non‐SLC15A4 dependent mechanism. The study presented more convincing evidence of Gly‐Sar dipeptide transport (K _M = 1.7 mM) and measured a 2–4 fold increase in the transport of carnosine and Gly‐Gly‐Gly when mutant SLC15A4 was expressed. Furthermore, the researchers demonstrated a significant increase in the internalization of NOD agonists MDP and Tri‐DAP in the presence of the mutant SLC15A4. Competition studies with labeled MDP were then conducted, which confirmed the Gly‐Sar affinity measurement but showed an unexpectedly weak affinity for histidine (> 1 mM).

More recently a wide range of amino acids, di‐peptides and tri‐peptides were screened for binding to chicken SLC15A4 using a Nano‐DSF thermo‐stabilization assay [34]. This analysis demonstrated pH dependent binding of several di‐ and tri‐peptides, notably Lys‐Val (K _D = 63 μM) and Arg‐Phe (73 μM), however no interactions with MDP, histidine or other amino acids were observed at pH 7.5 or 5.5. A structure of human SLC15A4 bound to a Lys‐Leu peptide was subsequently solved using cryo‐EM, building further confidence in the interaction of SLC15A4 with di‐peptides [45].

Characterization of SLC15A4 transport activity was recently attempted using microscale thermophoresis (MST) and solid supported membrane electrophysiology (SSME) [46]. The group used MST to measure the interaction of histidine with GFP tagged SLC15A4 in HEK293 membrane preparations, measuring an EC₅₀ of 3.8 mM. It should be noted that MST relies on the fluorescent molecule of interest changing size, charge or hydration state to alter movement following laser pulse, and it was unclear how histidine binding modulated the properties of SLC15A4‐GFP in this scenario. The researchers then used SSME to study the change in membrane potential after treatment of lysosomal fractions from HEK293 cells overexpressing SLC15A4‐GFP. Whilst the researchers observed changes in membrane potential in the presence of several substrates and inferred the changes to be the result of proton co‐transport, significant non‐specific signal complicated interpretation of the results. As SLC15A4 is known to be expressed in HEK293 cells [4, 47], the results would have been easier to interpret if the researchers had conducted the experiments in a SLC15A4 KO background [4, 47].

The use of fluorescently labeled di‐peptides has been a useful tool in the characterization of the closely related PEPT2 transporter [48]. There has been a report of fluorescent FAM‐His‐Leu peptide transport by surface localized SLC15A4 expressed in HEK293 cells [35], however, SLC15A4 mediated transport of the peptide was not thoroughly characterized and the transport has not been reproduced in any further studies.

In vitro knock down, KO, over‐expression and in vivo KO studies have been used to link SLC15A4 to NOD signaling [4, 6, 32, 47, 49]. These studies were conducted before the role of the SLC15A4:TASL:IRF5 signaling axis was discovered. The authors hypothesized that SLC15A4 influences NOD activation through the transport of NOD ligands from the endosome into the cytosol, after NOD ligands had been internalized through endocytosis [4, 50]. One study used microscopy to demonstrate that the expression of surface localized SLC15A4 or SLC15A3 could increase the rate of MDP‐rhodamine uptake in HEK293 cells [47]. It should, however, be noted that no detrimental effect on NOD signaling was observed in SLC15A4 KO THP‐1 cells [7], and intraperitoneal stimulation of SLC15A4 KO mice with Tri‐DAP or MDP only showed a defect in NOD1 (Tri‐DAP) signaling [32]. Since these initial observations, NOD ligand transport by SLC15A4 has been studied directly alongside peptide transport, and these studies generated conflicting results with MDP transport observed in MDCK cells using an MS read out, but no MDP binding observed to chicken SLC15A4 in Nano‐DSF (as has been discussed earlier in this section) [44]. Several further studies have used NF‐κB activation downstream of NOD agonist application in cells expressing surface localized SLC15A4 to make inferences about transport [4, 12, 35]. However, these experiments do not control for possible modulation of NF‐κB signal transduction by endo‐lysosomal SLC15A4 [4, 6, 32, 47, 49].

As SLC15A4 was reported to be a lysosomal transporter for the basic amino acid histidine, it has been hypothesized that the channel might play a role in lysosomal pH regulation. The first study to put forward this theory demonstrated that SLC15A4 KO blocked TLR7 agonist induced lysosomal acidification in murine DCs and B cells [6]. Subsequent studies demonstrated that even in the absence of stimulation SLC15A4 KO could increase lysosomal acidity in mouse B cells [27], mouse pDCs [28] and HL‐60 cells [51]. However, contradictory observations in which the KO of SLC15A4 had no effect on lysosomal pH were made in murine mast cells [52], mouse DCs [32], and THP‐1 cells [7]. It has also been suggested that TASL expression can lower lysosomal pH [53]. Taken together, the current relationship between SLC15A4 and lysosomal pH regulation is unclear and more work is needed to establish if modulation occurs.

The characterization of the transporter function of SLC15A4 is an ongoing process with definitive identification of SLC15A4 substrates and relative affinities hampered by conflicting literature reports. Taken together, the current evidence strongly indicates that SLC15A4 can bind certain di‐peptides with micro to millimolar affinity; however, data on histidine transport are more variable, with most studies complicated by high non‐specific transport. Whilst there is strong evidence that SLC15A4 can influence NOD signaling in certain situations, there is only limited data suggesting that this is due solely to the transporter function of SLC15A4. Recent reports have demonstrated that the SLC15A4 TASL adaptor, but not transport function, are required for endosomal TLR signal transduction [8]. It has also been reported that pDCs isolated from TASL KO mice have reduced NOD1 and 2 responses [10], therefore, it is possible that the SLC15A4 adaptor function may influence further immune pathways. Furthermore, it is unclear if the interaction with TASL is transitory enough to allow transport in cells expressing both proteins. Despite this, SLC15A4 appears to be more widely expressed than TASL (which is limited to immune cell populations) and so likely has functions beyond its interaction with TASL (www.proteinatlas.org) [54]. It has also been reported that the closely related endo‐lysosomal transporter SLC15A3, which has an expression pattern that closely tracks TASL, can transport fluorescent MDP [47].

More research is needed to conclusively characterize SLC15A4 substrates and to demonstrate the biological role of the transport function of SLC15A4. Future studies will need to ensure that non‐specific signals are adequately controlled for, especially those from endogenously expressed SLC15A4. More sensitive methodologies and quantitative descriptions of transport rates under different pH conditions would be of great help, especially in the case of NOD agonists and histidine. The effect of receptor dimerisation on transport also remains an open question.

1.4. Discovery of Small Molecule Probes for SLC15A4

The strong genetic evidence for the role of SLC15A4 in the pathogenesis of SLE, the recent structural and functional characterization of the SLC15A4:TASL interaction and the proven tractability of the wider SLC15 family [42], have made the protein an attractive target for small molecule modulation. Current targeted therapies for SLE often only modulate a single arm of this heterogeneous disease i.e., only B cell autoantibody production, or only inhibiting Type I IFN receptor. This approach may allow ‘leakage’ of the disease along alternate pathways [21]. SLC15A4 inhibition offers the opportunity to target both arms of the disease, as well as offering the prospect for a less invasive treatment option, as an orally‐ administered small molecule therapy.

Boeszoermenyi and colleagues have reported a conformation‐locking inhibitor of SLC15A4 (C5 or feeblin), which binds a lysosomal outward‐open conformation incompatible with TASL binding (Figure 2). The group utilized a microscopy‐based screen to identify compounds that engaged SLC15A4, displaced TASL, and reduced overall TASL levels due to stronger degradation of the unstable unbound form. A reporter construct was engineered to encode GFP‐tagged TASL followed by a self‐cleaving peptide P2A and mCherry (TGC reporter), allowing ratiometric measurement of TASL‐GFP stability. The group screened ~90 000 compounds, with 154 of these decreasing the GFP/mCherry ratio > 15% after a 48 h incubation. Of these, 5 compounds were shown to inhibit IRF5 nuclear translocation in Cal‐1 cells following R848 stimulation, whilst leaving the NF‐κB pathway untouched [11].

C5 (feeblin) inhibited ISRE reporter activity in THP‐1 cells in response to R848, whereas inhibition of Pam3CSK4‐mediated ISRE was ~20‐fold less potent, suggesting this compound possesses moderate selectivity [11]. The researchers solved a cryo‐EM structure of C5 bound to SLC15A4 and demonstrated that the compound did not inhibit IRF5 activation in SLC15A4‐deficient Cal‐1 cells expressing a LAMTOR‐TASL fusion protein, providing convincing evidence that the primary target of C5 is SLC15A4. Furthermore, C5 inhibited R848‐mediated TNFα production and IRF5 translocation in both healthy and SLE patient PBMCs and B cells, respectively. This demonstrated translational efficacy of C5 in primary cell models in both normal and diseased contexts.

A recent patent published by Solgate GmbH (Patent ID: WO 2025/068487) [55] has disclosed a wider set of amide quinoline‐based SLC15A4 inhibitors which are proposed to work through a similar mechanism to C5. Compound activity was measured using inhibition of R848‐induced cytokine release in PBMCs and the TGC reporter system used to identify C5 [11]. Furthermore, a novel NanoBRET target engagement assay demonstrated compound binding to cell membrane‐localized SLC15A4, as measured by competition with a fluorescently labeled probe (Figure 3). Despite the strong target engagement data included in the patent, it was difficult to assess the functional selectivity of the compounds due to the lack of a counterscreen.

FIGURE 3.

Chemical structures of literature SLC15A4 inhibitors. (A) C5 (feeblin) discovered by Boeszoermenyi et al. 2023. (B) AJ2‐30 identified by Chiu et al. 2024. (C) NBD conjugated fluorescent probe reported in Solgate GmbH patent (Patent ID: WO 2025/068487) used to determine compound binding to SLC15A4.

Chemo‐proteomic approaches have also been used to identify small molecule SLC15A4 inhibitors. Chiu and colleagues reported the discovery and optimisation of chemical scaffolds with affinity for SLC15A4 and demonstrated that such compounds suppress inflammatory signaling in human and mouse immune cells. An initial hit (FFF‐21) was identified through the comparison of MS‐based detection of SLC15A4‐ligand interactions in PBMCs and inhibition of CpG‐A induced IFN‐alpha release from pDCs. The researchers then developed this fragment into AJ2‐30, a compound which blocked CpG‐A induced IFN‐alpha from pDCs with an IC₅₀ of 1.8 μM [12].

Further profiling of AJ2‐30 revealed inhibition of endosomal signaling and downstream SLC15A4‐mediated functions in multiple relevant cell types [12]. For instance, AJ2‐30 also perturbed TLR7‐mediated IFN‐alpha production in human pDCs, although discrepancies were observed between inhibition of TLR9‐ vs. TLR7‐driven responses with respect to pDC‐derived IL‐6 and TNFα. AJ2‐30 was also shown to inhibit TLR7‐mediated TNFα production by human monocytes, along with B cell activation and production of cytokines (TNF‐alpha/IL‐6) and IgG in response to TLR7 or 9 agonism. Furthermore, AJ2‐30 blocked the IFN signature in mice treated with DOTAP‐complexed CpGA and TLR7, 8 and 9‐stimulated human PBMCs from lupus patients. Unlike C5, immunoprecipitation experiments suggested that AJ2‐30 does not disrupt the SLC15A4‐TASL interaction. Instead, the proposed mechanism of action was the induction of SLC15A4 destabilization and subsequent lysosomal degradation, which occurs following 4–6 h of treatment. However, inhibition of IFN‐beta gene induction and STAT1 phosphorylation is observed as early as 1 h post‐treatment, suggesting functional SLC15A4 inhibition precedes protein degradation. The exact mechanism of AJ2‐30 remains to be elucidated. Although Chiu and colleagues reported an extensive dataset generated in human primary cells, there was limited counter‐screening and target engagement data to demonstrate that the compounds were acting directly through SLC15A4.

1.5. Tools for Studying SLC15A4

There is now persuasive evidence supporting both that SLC15A4 is a key mediator in SLE and that the protein may be tractable for small molecule modulation. Recent studies reviewed here have presented a variety of techniques to study SLC15A4 and to quantify the effects of small molecule intervention, and the pros and cons of these methodologies are briefly summarized in Table 1.

TABLE 1.

Summary of literature assays that could be used to characterize novel small molecule SLC15A4 inhibitors.

Assay type	Assay (reference)	Key limitations	Key advantage
Target engagement	NanoBRET FL competition [55]	SLC15A4 is re‐localized to cell surface.	High throughput. Quantitative. Protein within intact membrane.
	Nano‐DSF [34]	Only established for chicken protein. Low throughput.	Quantitative.
	Microscale thermophoresis (MST) [34]	Unclear why diffusion of protein is modulated by ligand engagement.	Quantitative.
	Bilayer interferometry (BLI) [34]	Only validated for Nanobody interactions with SLC15A4.	Quantitative.
	Cryo‐EM [11]	Very low throughput. Not quantitative.	Information on binding pose and key interactions.
	Cellular thermal shift assay (CETSA) [12]	AJ2‐30 did not stabilise protein.	Semi‐quantitative.
	Proteomics (reactive fragment or TMT isobaric labelling based) [12]	Liable to identify binders with off target activity.	Can identify low affinity binders. High throughput.
Mechanistic	TASL degradation imaging assay [11]	Constrained to TASL displacement and degradation MoA.	High throughput. Quantitative.
	TASL degradation western blot [11]	Low throughput. Poorly quantitative.	Transferable (between cell types).
	Transport assays (e.g., MS or radiolabelled substrate)	SLC15A4 liable to transport assays are not well validated in the literature.	Quantitative.
Proximal signaling	IRF5 phostag gels [11]	Laborious and low throughput. Not quantitative.	Well validated in literature. Transferable (between cell types).
	IRF5 nuclear translocation [12]	Requires confocal microscopy.	Ability to distinguish between IRF5 and NF‐κB pathways. Transferable.
Functional (on target)	Primary cell cytokine inhibition (e.g., IFN‐alpha inhibition in pDC) [12].	Liable to donor to donor variability. Results may be influenced by para‐autocrine signaling.	Quantitative. Demonstrates translation into disease relevant cell subsets.
	ISRE reporter assays (e.g., THP‐1 dual) [12]	Results may be influenced by auto‐crine signaling Liable to off‐target activity.	Quantitative. Reproducible. High throughput. Well validated.
	Interferon stimulated gene (ISG) inhibition	Results may be influenced by para/autocrine signaling.	Quantitative. Well validated. Transferable.
Functional (counterscreen)	Pam3CSK4/LPS inhibition assays [11, 12]	TLR specific off‐target inhibitors are not identified.	Transferable between cell types.
	LAMTOR‐TASL expressing SLC15A4 KO ISRE reporter assays [11]	Requires an engineered cell line.	Allows identification of off‐target TLR7, 8 or 9 inhibitors.

Key considerations for novel drug discovery efforts at poorly characterized proteins are quantifying target engagement and mechanism of action. As discussed earlier, a recent patent reported the use of a NanoBRET based fluorescent probe competition assay utilizing HAP1 cells recombinantly expressing a cell surface localized SLC15A4 construct with a luciferase inserted into a lysosomal/extracellular loop [55]. This assay allowed the researchers to profile the binding of unlabelled competitor molecules in the plasma membrane of living cells. Future studies could be improved with the use of a higher affinity probe and assessment of compound binding in an acidic extracellular environment, allowing re‐capitulation of protein dynamics under the lysosome: cytosol pH gradient. This type of assay could also be used to define endogenous substrates for SLC15A4.

Other potential target engagement assays for drug discovery include Nano‐DSF or microscale thermophoresis (MST), which have been employed to quantify substrate engagement to purified protein or membrane preparations, respectively [34, 46]. Biolayer interferometry (BLI) has also been used to quantify synthetic nanobody interactions with purified chicken SLC15A4 [34]. At this early stage in the development of chemical probes for SLC15A4, target engagement assays are vital to demonstrate that functional inhibition is driven by compound binding.

The use of mechanistic assays is also key to gain an understanding of how SLC15A4 engagement leads to functional inhibition. Chui and colleagues screened for compounds that bound to SLC15A4 and blocked IFN‐alpha signaling, postulating that the compounds identified from this effort disrupted endosomal TLR signaling by binding to and destabilizing SLC15A4 and also blocked NOD agonist transport [12]. Whilst the group demonstrated that their compound induced SLC15A4 destabilization using MS and CETSA, it is unclear how target engagement led to destabilization. Interpretation of their compound's inhibitory effect on transport was also complicated by the lack of a MS or radio/fluorescently labeled substrate transport assay. In contrast, Boeszoermenyi and colleagues screened for compounds with a desired TASL displacement and degradation‐based mechanism of action [11]. Once the group had identified a weak binder they then confirmed target engagement with cryo‐EM.

As there are many proteins with proven tractability in the TLR7, 8 or 9 signaling pathways, driving small molecule development through functional assays, for example ISRE reporter cell lines, is likely to increase the risk of developing off‐target activity. This risk can be reduced by focusing on proximal read outs, for example IRF5 nuclear translocation, but ideally functional assays should be paired with an appropriate counter‐screen. Both published studies to identify small molecules for SLC15A4 have utilized Pam3CSK4 or LPS based counter‐screens to demonstrate their compounds are more potent in inhibiting TLR7, 8 or 9 stimuli; however, these efforts would not be able to identify off‐target TLR7, 8 or 9 interacting compounds. Boeszoermenyi and colleagues demonstrated a particularly well‐designed counter‐screen in which a LAMTOR‐TASL fusion protein was expressed in SLC15A4 KO CAL‐1 cells allowing compound activity to be tested in an artificially reconstituted TLR7/8 signaling pathway in the absence of SLC15A4 [11]. The group did not assess if C5 interacted with closely related solute carriers such as SLC15A3 [11], the functional consequences of such a cross reactivity are currently unclear.

2. Conclusion

Mouse models have established that TASL KO mice recapitulate the relevant pDC and B cell phenotypes induced in SLC15A4 KO mice and yield the resultant abrogation of SLE‐like disease [30]. This evidence strongly indicates that the interaction with TASL is the key function of SLC15A4 in the context of SLE. The emerging class of TASL‐displacing SLC15A4 binders highlights the tractability of SLC15A4 as a drug target and presents a clear mechanism of action through which to drive functional activity [11]. Despite these exciting developments, there are still key unanswered questions about the pharmacology of SLC15A4. Future studies will be needed to establish the relevance of SLC15A4 as a transporter, the function of the protein in cell types that do not express TASL, and the mechanism through which the protein modulates other immune pathways such as NOD signaling. The initial small molecule probes for SLC15A4 have been weak binders, and if stronger, more selective compounds can be developed they will offer the chance to gain a better understanding of the relevance of SLC15A4 in different cell types and possibly treat patients suffering with auto‐immune disease.

Author Contributions

Miguel C. Silva: writing – original draft, writing – review and editing, project administration. Steven J. Charlton: writing – review and editing, project administration. Charles S. Lay: conceptualization, writing – original draft, writing – review and editing. Kathy Sengmany: writing – original draft, writing – review and editing. Alex L. Wilkinson: writing – original draft, conceptualization, writing – review and editing, visualization. Dillon Popat: conceptualization, writing – original draft, writing – review and editing.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.