IL-17 mediates essential inflammatory responses in host defense and autoimmunity. The IL-17A–IL-17F signaling complex is composed of IL-17RA and IL-17RC, both of which are necessary for signal transduction. To date, the specific contribution of IL-17RC to downstream signaling remains poorly understood. To define the regions within the IL-17RC cytoplasmic tail required for signal transduction, we assayed signaling by a panel of IL-17RC deletion mutants. These findings reveal that IL-17RC inducibly associates with a specific glycosylated IL-17RA isoform, in a manner independent of the IL-17RC cytoplasmic tail. Using expression of the IL-17 target genes IL-6 and 24p3/lipocalin-2 as a readout, functional reconstitution of signaling in IL-17RC−/− fibroblasts required the SEF/IL-17R signaling domain (SEFIR), a conserved motif common to IL-17R family members. Unexpectedly, the IL-17RC SEFIR alone was not sufficient to reconstitute IL-17–dependent signaling. Rather, an additional sequence downstream of the SEFIR was also necessary. We further found that IL-17RC interacts directly with the adaptor/E3 ubiquitin ligase Act1, and that the functional IL-17RC isoforms containing the extended SEFIR region interact specifically with a phosphorylated isoform of Act1. Finally, we show that IL-17RC is required for in vivo IL-17–dependent responses during oral mucosal infections caused by the human commensal fungus Candida albicans. These results indicate that IL-17RC is vital for IL-17–dependent signaling both in vitro and in vivo. Insight into the mechanisms by which IL-17RC signals helps shed light on IL-17–dependent inflammatory responses and may ultimately provide an avenue for therapeutic intervention in IL-17–mediated diseases.
IL-17 (IL-17A) plays a major host protective function at mucosal surfaces by driving inflammatory responses in mesenchymal and epithelial cell types (1, 2). The inflammatory program initiated by IL-17A leads to induction of IL-6, CXC chemokines, matrix metalloproteinases, acute phase reactants, and antimicrobial peptides, such as β-defensins and lipocalin-2/24p3 (3, 4). The potent inflammatory properties of IL-17A establish its critical role in host defense against extracellular pathogens. Conversely, aberrations in the IL-17A signaling axis correlate with a number of autoimmune diseases, including rheumatoid arthritis, systemic lupus erythematosus, and psoriasis (5).
The fundamental mechanisms by which IL-17A transduces signals are still poorly understood (6). IL-17A and the closely related family member IL-17F mediate their inflammatory activities via the IL-17R complex, composed of IL-17RA and IL-17RC. IL-17RA was the first receptor to be indentified for this cytokine family (7). Bioinformatic studies identified a domain encoded within the IL-17RA intracellular tail that is found in all IL-17R family members, termed “similar expression to FGF/IL-17R” (SEFIR) (8). The SEFIR domain was discovered based on homology with a functional domain within Toll-/IL-1R family members (TIR). The SEFIR domain is also found in the signaling adaptor/E3 ubiquitin ligase Act1/cytokine inducer of NF-κB signaling (8), a downstream signaling intermediate in the IL-17R pathway. To recruit Act1 and promote downstream signaling, IL-17RA depends on homotypic interactions between its SEFIR domain and the Act1 SEFIR (9–11). Structure-function studies subsequently showed that the IL-17RA SEFIR motif is critical for activation of IL-17A–dependent downstream effector pathways including the MAPK, NF-κB, and C/EBP pathways (10, 12). Although the SEFIR domain is essential for IL-17RA–mediated signaling, it is not sufficient; rather, a region extending beyond the conserved SEFIR is also required for activation of NF-κB and downstream target gene expression (12).
Whereas IL-17RA is required for IL-17 signaling, genetic evidence (13) and the recent crystal structure of the IL-17F–bound IL-17RA receptor subunit (14) indicate that the ligand-bound IL-17R complex is an IL-17RA and IL-17RC heterodimer (15). However, the precise stoichiometry and the relative biochemical contributions of the IL-17RA and IL-17RC subunits to downstream signaling are still not well defined (14). Notwithstanding the requirement of both IL-17RA and IL-17RC to transduce IL-17 signals, IL-17RA, and IL-17RC have strikingly distinct tissue expression patterns. IL-17RA is expressed in most cell types, with high levels in hematopoietic cells relative to other cell types. In contrast, IL-17RC is expressed at high levels in glandular tissues, such as the adrenal gland, prostate, liver, and thyroid, with little expression in hematopoietic tissues (16, 17). Recent reports, however, have identified IL-17–dependent responses in hematopoietic cell types (17–19). Therefore, it is not certain whether IL-17RC is indeed needed in all tissues to mediate IL-17–dependent signal transduction.
The specific contributions of the IL-17RC subunit to IL-17 signaling remain unknown, while previous studies indicated that the cytoplasmic tail of IL-17RC is essential for functional IL-17A–dependent responses in fibroblast cells (13, 20). Because IL-17RA uses sequences beyond the SEFIR domain to mediate signaling, we aimed to define regions within the IL-17RC cytoplasmic tail that are integral to IL-17A and IL-17F signaling and to determine IL-17RC function in IL-17–dependent immune responses in vivo. In this study, we show that IL-17A and IL-17F promote the association of IL-17RC with a glycosylated IL-17RA isoform. This association is not dependent on the IL-17RC cytoplasmic tail. Moreover, the IL-17RC SEFIR domain is necessary but not sufficient to mediate IL-17 dependent signaling. Rather, an additional ∼20–30 aa extension downstream of the IL-17RC SEFIR (termed SEFEX) is necessary for activation of NF-κB–dependent genes, such as IL-6 and 24p3. We also found that this SEFEX region is required for ligand-dependent interaction with a phosphorylated Act1 species. These data demonstrate for the first time that Act1 is phosphorylated and that the phospho-Act1 isoform preferentially associates with the IL-17R complex upon IL-17 stimulation. Finally, we show that both IL-17RC and IL-17RA are required in vivo for immunity to oral infection with the commensal yeast Candida albicans, and that IL-17RC functions indistinguishably from IL-17RA in this regard.
Materials and Methods
Cell cultures and luciferase assays
Primary IL-17RC−/− fibroblasts were isolated from IL-17RC−/− adult tail-tips (20). IL-17RA−/− fibroblasts from IL-17RA−/− mice were immortalized with the SV40 T Ag as described (12). ST-2 murine stromal cells, HEK293T cells, and fibroblasts were cultured with α-MEM (Sigma, St. Louis, MO) with 10% FBS, L-glutamine, and antibiotics (Invitrogen, Carlsbad, CA). HEK293T cells were transfected using Fugene 6 per the manufactuer’s instructions (Roche, Indianapolis, IN). IL-17RC−/− and IL-17RA−/− fibroblasts were transfected with the Amaxa MEF2 Nucleofector Kit per the manufacturer’s instructions (Lonza, Germany), and luciferase assays were performed as previously described (21). Cytokines were from Peprotech (Rocky Hill, NJ).
IL-17RC cDNA (encoding full length, unspliced IL-17RC) was generated from ST-2 cells by RT-PCR. The IL-17RC truncations were generated by PCR with a C-terminal Myc tag and confirmed by sequencing. The murine IL-17RA construct was described previously (12). Act1 cDNA was obtained by RT-PCR from ST2 cells and fused to an N-terminal CFP tag in pcDNA3.1.Zeo (Invitrogen). The 24p3 promoter luciferase reporter was described previously (21).
ELISA and flow cytometry
IL-6 ELISA kits were from eBioscience (San Diego, CA). Flow cytometry was performed on a FACSCalibur with Cell Quest Software (BD Biosciences, San Diego, CA). Cells were stained with an anti–IL-17RC Ab followed by a PE-anti-goat Ig Ab (R&D Systems, Minneapolis, MN).
Western blotting, EMSA, and immunoprecipitation
Western blotting, EMSA, and immunoprecipitation experiments were performed as described (12). Immunoprecipitates were incubated with shrimp alkaline phosphatase (Fermentas, Glen Burnie, MD) at 37°C for 1 h. Anti–mIL-17RA Abs were from R&D Systems. Anti-Act1 Abs were from Santa Cruz Technologies (Santa Cruz, CA) and anti-MYC Abs were from Cell Signaling (Beverly, MA). Mouse TrueBlot ULTRA HRP anti-mouse IgG was from eBioscience (San Diego, CA). Densitometry was performed using Image J software (from the NIH, www.rsbweb.nih.gov/ij/) on scanned gels.
Mouse model of oropharyngeal candidiasis
IL-23p19−/− and IL-17RC−/− mice were provided by Genentech, produced in collaboration by Genentech and Lexicon Pharmaceuticals to analyze the function of 500 secreted and transmembrane proteins (20). IL-17RA−/− mice were provided by Amgen (Seattle, WA). Age- and sex-matched wild type (WT) control mice were from The Jackson Laboratory (Bar Harbor, ME). Mice were infected under anesthesia by placing a 0.0025-g cotton ball saturated with 2 × 107 CFU C. albicans (strain CAF2-1) sublingually for 75 min, as previously described (22, 23). If indicated, 225 mg/kg cortisone acetate (Sigma-Aldrich) was injected days −1, 1, and 3 relative to infection. Tongue was homogenized and analyzed for CFU per gram of tissue, and paraffin-embedded tongue sections were stained with periodic-acid Schiff by the University at Buffalo Histology Core Facility or the University of Pittsburgh Research Histology Services. Protocols were approved by the SUNY Buffalo and University of Pittsburgh IACUC.
An experimental system for evaluating IL-17RC functional signaling domains
To delineate motifs within the IL-17RC intracellular domain required for functional signaling responses, we established a system to study IL-17 signaling in murine IL-17RC−/− tail-tip fibroblasts and HEK293T cells. Because of the requirement of IL-17RC for IL-17 signaling and the failure of the human and murine receptor subunits to complement one another (13), IL-17RC−/− fibroblasts and HEK293T cells lacking mIL-17RC are deficient in IL-17 responses, and therefore provide us with a useful experimental platform to perform IL-17RC signaling analyses (13, 20). Accordingly, we created a series of murine carboxyl terminally truncated IL-17RC mutants (Fig 1A). The IL-17RC truncations included deletions that lack the SEFIR signaling domain (amino acids 495–645), which is uniquely found on IL-17R family members and is known to be critical for IL-17RA signaling (6, 8). Cell surface expression of these mutants was verified by flow cytometry (Fig 1B).
IL-17RC association with IL-17RA does not require the IL-17RC intracellular domain
The ligand-bound IL-17R complex is reported to be composed of both IL-17RA and IL-17RC, and previous fluorescence resonance energy transfer studies suggested that IL-17RA forms homodimers, at least in the absence of ligand (13, 14, 24). We thus questioned whether the association of IL-17RC and IL-17RA occurs in a ligand-dependent manner and whether this interaction requires any portion of the IL-17RC intracellular domain. Accordingly, HEK293T cells were cotransfected with a plasmid encoding full-length murine IL-17RA together with various IL-17RC receptor truncations. There was baseline association of IL-17RA and IL-17RC, which was enhanced by treatment with either IL-17A or IL-17F (Fig 2A). Unlike the TLRs (25), the association of IL-17RA with IL-17RC appeared to be independent of the IL-17RC cytoplasmic tail, because none of the IL-17RC cytoplasmic truncations were defective in association with IL-17RA (Fig 2A, Supplemental Fig. 1).
Interestingly, IL-17A or IL-17F treatment caused the association with a slower-migrating IL-17RA isoform, although IL-17A was more potent than IL-17F (Fig. 2A, lanes 6–11). Several differentially glycosylated forms of IL-17RA have been reported (7, 19, 26), but the biochemical nature of the specific IL-17RA molecule that is pulled down with IL-17RC was unclear. To assess whether glycosylation accounted for the larger IL-17RA isoform, we pretreated IL-17RA–transfected cells with tunicamycin to deglycosylate IL-17RA before immunoprecipitation. Upon 24 h tunicamycin treatment, the larger IL-17RA isoform resolved to a single band (Fig. 2B). We then showed that pretreatment of cells with tunicamycin prior to immunoprecipitation led to coassociation of IL-17RC with the faster-migrating band of IL-17RA (Fig. 2C). These results indicate that IL-17RC associates with non-glycosylated IL-17RA constitutively, and that IL-17A and IL-17F binding induces association of IL-17RC with a specific glycosylated isoform of IL-17RA.
An extended region beyond the SEFIR domain is required for functional IL-17RC signaling
To delineate motifs within the IL-17RC cytoplasmic domain necessary for functional responses, primary fibroblasts from IL-17RC−/− mice (20) were prepared and assayed for IL-17–dependent signaling. Treatment of these cells with IL-17A or IL-17F alone or in combination with TNF-α (with which IL-17 exhibits potent synergistic responses) (27, 28), did not induce secretion of IL-6 or activation of NF-κB compared with ST-2 control cells or WT fibroblasts (Fig. 3A, 3B, and data not shown), whereas a TNF-α control induced NF-κB (Fig. 3C) and an IL-1β control induced IL-6 secretion (Fig. 3B). Induction of other target genes, such as 24p3/lipocalin 2 and CXCL5, in response to IL-17A signaling was also impaired in IL-17RC−/− fibroblasts (data not shown). Therefore, IL-17RC is essential in fibroblasts for IL-17A–dependent signaling responses.
Because IL-17RC is required for signaling in these cells, we transfected IL-17RC−/− cells with the panel of IL-17RC receptor truncations in attempt to reconstitute signaling. Cells were then treated with IL-17 alone or in combination with TNF-α and assessed for cytokine-dependent induction of prototypical IL-17 target genes, including 24p3/lipocalin 2 and IL-6 (12, 29). Cells expressing the IL-17RCΔ668, IL-17RCΔ678, and IL-17RC-FL receptors activated the 24p3 promoter reporter (Fig. 3D) and produced IL-6 significantly in response to IL-17A alone and in combination with TNF-α (Fig. 3E). In the cells expressing IL-17RCΔ658, there was no IL-6 production in the synergistic condition or in IL-17-mediated activation of the 24p3 promoter; however, there was a small but statistically significant induction of IL-6 with the IL-17RCΔ658 mutant following IL-17 treatment. This result is likely due to the low signal-to-noise ratio of this assay, not because the IL-17RCΔ658 mutant recapitulates full-length IL-17RC function. Intriguingly, IL-17RC mutants, such as IL-17RCΔ648, failed to respond to IL-17A, although these deletions retain the entire conserved SEFIR domain (Figs. 1A, 3D, 3E). This result suggests that, similar to IL-17RA, the IL-17RC SEFIR domain is not sufficient for functional IL-17–mediated signaling. Rather, the SEFEX additional region downstream of the IL-17RC SEFIR is also necessary, the terminus of which extends to between amino acids 658 and 668.
The adaptor molecule Act1 is an E3 ubiquitin ligase required to link IL-17RA with downstream signaling pathways, such as TRAF6 and NF-κB (30), although Act1 may not be fully necessary for ERK activation (10). Act1 has been shown to co-immunoprecipitate (IP) with IL-17RA (9, 10). To determine whether Act1 also associates with IL-17RC, HEK 293T cells were cotransfected with a CFP-tagged murine Act1 together with the panel of IL-17RC truncations. The association of Act1 with the IL-17RC mutants was evaluated by co-IP in the presence of either IL-17A or IL-17F. A low baseline level of Act1 was pulled down with all IL-17RC deletion mutants tested (Fig. 4A, lanes 1–7). Interestingly, this constitutively associated Act1 represented the smaller of two isoforms of this protein (Fig. 4A, bottom arrow). Notably, a larger Act1 isoform was inducibly associated with IL-17RC (Fig. 4A, top arrow, lanes 11–13; Fig. 4B, lanes 4–5). Moreover, the larger isoform of Act1 associated only with IL-17RC variants that were shown to be capable of mediating signaling; namely, IL-17RCΔ668, IL-17RCΔ678, and IL-17RC-FL, suggesting that this Act1 isoform may be the version of this molecule competent for signaling in the IL-17 pathway.
To better characterize the isoform of Act1 that associates with the functional IL-17RC mutants, we examined whether this Act1 isoform was phosphorylated. To that end, HEK293T cells were cotransfected with Act1 and IL-17RC and stimulated with IL-17A and IL-17F for 20 min. Immunoprecipitates were then treated with shrimp alkaline phosphatase (SAP; Fig. 4B). SAP treatment resulted in the complete disappearance of the larger Act1 isoform and a diminution of both isoforms, confirming that the functional Act1 species represents a phosphorylated moiety. To characterize the kinetics of phospho-Act1 association with IL-17RC, a time course was performed. Phospho-Act1 coassociation with IL-17RC showed a reproducibly bisphasic profile, rising to an ∼6-fold induction 20 min after IL-17A stimulation, diminishing after 1 h, and reappearing to high levels at 2–12 h (Fig. 4C).
Although there is evidence that murine IL-17RC does not partner with human IL-17RA (13), it is possible that the expression of endogenous human IL-17RA in HEK293T cells is contributing to the observed interactions between murine IL-17RC and murine Act1 transfected in these cells. Therefore, to determine whether IL-17RA helps bridge Act1 and IL-17RC, cotransfection and pulldown experiments with IL-17RC and Act1 were performed in IL-17RA−/− fibroblasts stimulated with IL-17A (Fig. 4D). As shown, murine IL-17RC could still immunoprecipitate Act1 in the absence of human and murine IL-17RA, indicating that IL-17RC does not require IL-17RA to associate with downstream signaling intermediates.
IL-17RC−/− mice are susceptible to infection with Candida albicans
We have previously shown that IL-17RA−/− mice are highly prone to infection with Candida albicans, a commensal yeast of the human oral cavity that causes oropharyngeal candidiasis (OPC), or thrush, in immunocompromised individuals (22). To determine whether IL-17RC−/− mice show a similar susceptibility and to establish whether IL-17RC is required for in vivo IL-17–dependent immune responses, mice were infected orally with C. albicans, and 5 d later the tongue was evaluated for fungal load and histologic evidence of pathology. As a positive control, WT and IL-17RC−/− mice were immunosuppressed with cortisone. As shown, cortisone-treated mice had a high fungal load after infection (Fig. 5A). There was no detectable colonization in sham-infected IL-17RC−/− mice, which was expected because C. albicans is not a normal commensal organism in rodents. However, IL-17RC−/− mice showed a dramatic increase in fungal burden in the oral cavity, whereas heterozygous IL-17RC+/− mice fully cleared the infection with no residual fungal load by day 5 (Fig. 5A). Consistent with the fungal burden in the oral cavity, invasion of hyphal and psuedohyphal forms of C. albicans was visible on the tongue in cortisone-treated mice (data not shown) and infected IL-17RC−/− mice (Supplemental Fig. 2). The surface epithelium of the tongues in these mice was damaged by the invading yeast. In contrast, sham-infected mice and infected IL-17RC+/− mice showed no evidence of C. albicans, and the surface epithelium of the tongue was undamaged. These results show that IL-17RC plays an important role in mediating host defense against C. albicans.
To determine whether there was any detectable difference in function between IL-17RC and IL-17RA in the context of OPC, we performed a side-by-side comparison of IL-17RC−/− and IL-17RA−/− mice. As an additional control, we included IL-23p19−/− mice, which are deficient in IL-17-producing Th17 cells (31). As shown, the IL-17RC−/−, IL-17RA−/−, and IL-23p19−/− mice were all susceptible to OPC, and there was no statistically significant difference in fungal load between any of these cohorts (Fig. 5B). Furthermore, both the IL-17RA−/− and IL-17RC−/− mice exhibited defective polymorphonuclear cell recruitment to sites of fungal infection (Fig. 5C). Therefore, these data support the concept that IL-17RC is an integral part of the IL-17 signaling complex, and its absence predisposes mice to similar infections as IL-17RA−/− mice (20), underscoring the functional overlap of IL-17RA and IL-17RC.
IL-17 has come into prominence with its identification as the signature cytokine of the Th17 lineage (32–34). However, to date the signaling mechanisms mediated by IL-17 are surprisingly poorly defined. The IL-17R family is striking in that it bears little resemblance to better known cytokine families, and thus it has not been possible to infer how this receptor mediates signals based merely on homology with other systems (6). An important clue came with the discovery of the SEFIR domain common to IL-17R family members. The homology of the SEFIR to TIR (8) was consistent with the similar panels of proinflammatory gene targets activated by IL-17 and TIR/IL-1R ligands (27). Despite this homology, the SEFIR domain is functionally and structurally distinct from TIRs. First, unlike TIR-containing receptors, IL-17R does not require the prototypical innate immune adaptors MyD88 or TRIF (9, 12). Second, SEFIR domains do not encode the BB-loop structure found in TIR domains that is required to mediate specificity of homotypic TIR–TIR interactions (25). Furthermore, empirical structure-function studies of IL-17RA showed that, while the SEFIR is necessary for signaling, an extended region beyond the SEFIR is also required (12). Our work shows that IL-17RC SEFIR also requires an additional downstream region to mediate downstream signaling.
When IL-17RC was shown to be a key component of the IL-17R signaling cascade, this raised important questions about its mechanisms of action with respect to signal transduction. Given the genetic requirement for IL-17RC in transducing IL-17 signals (13) and the IL-17F-bound IL-17RA crystal structure suggesting a IL-17RA/IL-17RC heterodimeric receptor (14), we wished to determine the functional significance of the IL-17RC receptor in the IL-17 signaling axis. Using IL-17RC truncations, our data indicate that the IL-17RC cytoplasmic tail is not required for its association with IL-17RA (Fig. 2A). Furthermore, the IL-17RC SEFIR and the SEFEX additional downstream 23 amino acids are required for IL-17–dependent IL-6 and 24p3 expression, and for a ligand-dependent association with a phosphorylated Act1 isoform (Figs. 3, 4). We further show that IL-17RC is required for IL-17–dependent host defense against oropharyngeal candidiasis (Fig. 5). These results demonstrate that IL-17RC is indispensable for IL-17–dependent immune responses, and they identify key structural and biochemical requirements responsible for signaling downstream of the IL-17RC receptor.
The stoichiometry of how IL-17RC associates with IL-17RA is currently undefined. Fluorescence resonance energy transfer studies of IL-17RA suggested that preassembled homodimers exist on the cell surface. However, the addition of IL-17 causes a conformational shift to occur, which could be explained by dissociation of IL-17RA homodimers or recruitment of additional subunits (24, 35). Subsequent genetic data, demonstrating that IL-17RC is required for the IL-17–dependent production of CXCL1, supported the idea that association with IL-17RC is involved in this process (13). Our present data indicate that a basal level of IL-17RA and IL-17RC may exist as preassociated complexes, but that addition of IL-17A or IL-17F enhances this association (Fig. 2A). Although our results must be interpreted in the context of an overexpression system, these findings are supported by recent biochemical studies of IL-17RA, which indicate that ligand binding increases the affinity of the IL-17RA–IL-17RC interaction (14). Notably, we also find that IL-17 treatment induces IL-17RC to preferentially associate with a larger IL-17RA isoform. Tunicamycin treatment leads to the disappearance of the larger IL-17RA isoform, suggesting that this is a glycosolated IL-17RA isoform (Fig 2B-C). Various N-linked glycosylated forms of IL-17RA have been reported, and the E3 ubiquitin ligases Act1 and TRAF6 have been shown to be associated with ubiquitinated forms of IL-17RA (7, 26, 30, 36). The relevance of these glycosylated species of IL-17RA is unclear at present, but its ligand-dependent association argues that it may be critical for signaling.
IL-17RC has a higher affinity for IL-17F than IL-17A (16). However, our data and others (13) demonstrate that IL-17RC is nonetheless required for IL-17A–mediated signaling. IL-17RC also exists in multiple splice forms, some of which have differential recognition of IL-17 family ligands (16). In these studies we evaluated only the unspliced (full-length) IL-17RC isoform, but it will be interesting to determine in the future whether there are differences in how other splice forms of IL-17RC participate in signaling.
Previous reports indicate that the IL-17RC cytoplasmic tail, which contains a SEFIR domain, is integral for functional IL-17 responses (8, 13, 37). We previously found that IL-17RA uses a SEFIR extension to mediate signaling (12). Analogously, we show here that IL-17RC uses both its SEFIR domain and an additional downstream sequence of 23 amino acids to activate signals such as IL-6 and 24p3 expression (Fig. 3). Consistent with this result, Hu et al. (37) find that the IL-17RC SEFIR is required for functional IL-17 signal transduction both in vitro and in vivo. Bioinformatic database searches and sequence alignments, however, indicate that the extended sequence downstream of the IL-17RC SEFIR lacks homology to the sequence downstream of the IL-17RA SEFIR, the Act1 SEFIR, or other known receptors or signaling intermediates (A.W. Ho, unpublished observations). Thus, the extended IL-17RC SEFEX domain may contribute a novel signaling function. Indeed, a chimeric receptor construct composed of the IL-17RA extracellular domain fused to the complete IL-17RC cytoplasmic tail cannot rescue IL-17–dependent signaling in IL-17RA−/− cells (R. Onishi and J. Park, unpublished data), implying that the role of IL-17RC is probably not simply to supply another SEFIR domain to more efficiently recruit additional Act1 molecules. Therefore, despite the fact that the IL-17RC cytoplasmic tail is much shorter than the IL-17RA intracellular domain (214 versus 521 amino acids), IL-17RC may possess novel signaling functions distinct from IL-17RA.
Act1 is an essential signaling component downstream of IL-17RA and is required for IL-17–dependent immune responses (9, 10). We now show that IL-17RC also associates with Act1 even in the absence of IL-17RA (Fig. 5). Interestingly, IL-17 treatment leads to the association of IL-17RC with a larger phosphorylated form of Act1. This result indicates for the first time that Act1 is phosphorylated, and that phospho-Act1 associates with IL-17RC in a ligand-dependent manner. Even more striking, the only IL-17RC deletions that associate with this phospho-Act1 isoform are those that retain function (i.e., mutants that contain the IL-17RC SEFEX motif). It is possible that the SEFEX domain recruits other signaling intermediates needed to create a stable signaling scaffold to permit Act1 to efficiently transduce downstream signals, analogous to other systems such as the coreceptors of the TCR (38).
Although good biochemical and genetic evidence supports a role for IL-17RC within the IL-17R complex, it is still not fully established whether IL-17RC is required for all aspects of IL-17 signal transduction in vivo. To evaluate the physiologic function of IL-17RC, we used an oral fungal infection model (23) that we and others previously showed to be strongly IL-17/IL-17RA-dependent (22, 39). In this study, we demonstrate that IL-17RC is required for host defense against infection with the yeast C. albicans (Fig. 5). Moreover, susceptibility to OPC was identical between IL-17RC−/− and IL-17RA−/− mice, in addition to IL-23p19−/− mice (22). This finding also suggests, albeit indirectly, that there is no additional role for IL-25 in this process, because IL-17RA but not IL-17RC participates in the IL-25 receptor complex (40). In line with these findings, IL-17RC was recently shown to participate in the pathogenesis of autoimmune inflammation of the CNS, an event highly dependent on the IL-17 signaling axis (37).
Our results show that IL-17RC is critically important for IL-17–dependent signaling and immune responses. IL-17RC mediates signaling via an extended SEFIR domain, which is required for a ligand-dependent association with a phosphorylated Act1 isoform to promote downstream signaling. These studies provide the first report of a signaling intermediate directly downstream of the IL-17RC receptor and are the first to define important structural sequence elements within this receptor. Lastly, like IL-17RA, IL-17RC is required for host defense against oral fungal infections caused by C. albicans. It will be interesting in future studies to link IL-17RC signaling pathways directly to biologic signals and ultimately exploit this knowledge to improve treatments for a host of human diseases.
We thank Genentech for providing IL-17RC−/− and IL-23p19−/−mice, Amgen for providing IL-17RA−/− mice, and C. Coyne and B. Tomson for critical reading and helpful suggestions.
Disclosures S.L.G. received a research grant from Amgen. F.S. and W.O. are full-time employees of Genentech.
This work was supported by National Institutes of Health Grant AR054389 and the Alliance for Lupus Research (to S.L.G.), National Institutes of Health Grants AI067544 (to L.P.K.) and HL079142 (to J.K.K.), the Medical Scientist Training Program at the University at Buffalo, State University of New York (to A.W.H.), and National Institutes of Health Training Grants DE007034 to the Department of Oral Biology at the University at Buffalo (to H.R.C.) and CA082084 to the University of Pittsburgh (to N.H.S.).
The online version of this article contains supplemental material.
Abbreviations used in this paper:
free probe alone
shrimp alkaline phosphatase
SEF/IL-17R signaling domain extension
SEF/IL-17R signaling domain
TLR/IL-1 receptor signaling domain