The IL-17 family cytokines IL-17A and IL-17C drive the pathogenesis of psoriatic skin inflammation, and anti–IL-17A Abs were recently approved to treat human psoriasis. Little is known about mechanisms that restrain IL-17 cytokine-mediated signaling, particularly IL-17C. In this article, we show that the endoribonuclease MCP-1–induced protein 1 (MCPIP1; also known as regnase-1) is markedly upregulated in human psoriatic skin lesions. Similarly, MCPIP1 was overexpressed in the imiquimod (IMQ)-driven mouse model of cutaneous inflammation. Mice with an MCPIP1 deficiency (Zc3h12a+/−) displayed no baseline skin inflammation, but they showed exacerbated pathology following IMQ treatment. Pathology in Zc3h12a+/− mice was associated with elevated expression of IL-17A– and IL-17C–dependent genes, as well as with increased accumulation of neutrophils in skin. However, IL-17A and IL-17C expression was unaltered, suggesting that the increased inflammation in Zc3h12a+/− mice was due to enhanced downstream IL-17R signaling. Radiation chimeras demonstrated that MCPIP1 in nonhematopoietic cells is responsible for controlling skin pathology. Moreover, Zc3h12a+/−Il17ra−/− mice given IMQ showed almost no disease. To identify which IL-17RA ligand was essential, Zc3h12a+/−Il17a−/− and Zc3h12a+/−Il17c−/− mice were given IMQ; these mice had reduced but not fully abrogated pathology, indicating that MCPIP1 inhibits IL-17A and IL-17C signaling. Confirming this hypothesis, Zc3h12a−/− keratinocytes showed increased responsiveness to IL-17A and IL-17C stimulation. Thus, MCPIP1 is a potent negative regulator of psoriatic skin inflammation through IL-17A and IL-17C. Moreover, to our knowledge, MCPIP1 is the first described negative regulator of IL-17C signaling.
In the past decade, IL-17 family members have emerged as drivers of inflammatory and autoimmune conditions, particularly psoriasis (1). Indeed, IL-17A–targeting Abs are approved for treatment of plaque psoriasis, psoriatic arthritis, and ankylosing spondylitis, underscoring the importance of IL-17A in these conditions (2, 3). IL-17A can be produced by a variety of immune cells, including Th17 cells, γδ T cells, invariant NKT cells, “natural” Th17 cells, and type 3 innate lymphoid cells (4–9). However, most studies indicate that IL-17 exerts its actions primarily on nonhematopoietic cells. IL-17A induces expression of a typical gene signature profile in target cells that is characterized by proinflammatory cytokines, chemokines, and antimicrobial peptides (1). Together, these factors provide important host defense against extracellular microbes but, when dysregulated, they can promote inflammation in autoimmunity.
In addition to IL-17A, the less-studied cytokine IL-17C is implicated in psoriasis. IL-17C is predominantly produced by epithelial cells and induces a similar set of downstream genes with inflammatory, antibacterial, and antiapoptotic functions. IL-17C appears to act in hematopoietic and nonhematopoietic cells, including intestinal epithelial cells, keratinocytes (KCs), and Th17 cells (10–13). IL-17C protein concentrations are ∼125-fold higher than IL-17A levels in psoriatic lesions, making it the most abundant IL-17 family member in human psoriasis (14). Additionally, KC-specific overexpression of IL-17C in mice causes a spontaneous psoriasis-like phenotype (14).
IL-17 family members signal through multimeric receptors composed of a common chain, IL-17RA, and a second chain that varies by ligand. IL-17A signals through IL-17RA and IL-17RC, and IL-17C was shown to act via an IL-17RA/IL-17RE receptor complex. Both cytokines use the adaptor and E3 ubiquitin ligase Act1 (also known as CIKS) to drive signaling, but little else is known about the positive activators of IL-17C downstream signaling (1, 15, 16).
Given the prominence of IL-17A in inflammation, numerous mechanisms have evolved to negatively regulate its signaling that are needed to limit collateral tissue damage during inflammatory processes. IL-17A is a key driver of psoriasis, but little is known about the factors that normally restrain IL-17A–dependent signal transduction in the skin. The endoribonuclease and deubiquitinase MCP-1–induced protein 1 (MCPIP1; also known as regnase-1 and encoded by the ZC3H12A gene) is a vital regulator of inflammation. Its expression is induced by proinflammatory stimuli, including MCP-1, TLR ligands, IL-1β, and IL-17A (17–21). MCPIP1 regulates TLR signaling through cleavage of target gene mRNAs, including Il6 (22–24), or deubiquitination of inflammatory mediators (25). In addition, MCPIP1 constitutively restricts TCR signaling, and MCPIP1 in T cells is inducibly degraded following T cell activation. MCPIP1 deficiency in CD4+ cells was shown to enhance Th17 effector function (26, 27). We showed that MCPIP1 negatively regulates IL-17–dependent inflammation through the degradation of IL-17A–induced target gene transcripts and IL-17RA mRNA (18). In this study, we examined the impact of MCPIP1 on IL-17A and IL-17C signaling in a mouse model of acute psoriatic-like skin inflammation and in human psoriasis clinical samples. To our knowledge, our data identify MCPIP1 as the first recognized inhibitor of IL-17C signaling and establish this protein as a common regulator of IL-17 family members during skin inflammation.
Materials and Methods
Nine healthy nonpsoriatic controls and 10 patients with chronic plaque psoriasis were enrolled (psoriasis lesional plaque, psoriasis nonlesional skin, and healthy normal skin). Patients were off systemic treatment for ≥4 wk and off all topical treatments 2 wk prior to enrollment. Two 6-mm punch biopsies from uninvolved skin and two biopsies from lesional skin or two biopsies from normal skin were obtained under local anesthesia. One biopsy was fixed in 4% formaldehyde for immunohistochemistry, whereas the other was snap-frozen in liquid nitrogen and stored at −80°C until processing. Informed consent was obtained from all subjects under a protocol approved by the Institutional Review Board of the University of Michigan Medical School (HUM00087890). This study was conducted according to the Declaration of Helsinki Principles.
C57BL/6 and CD45.1 mice were from the Jackson Laboratory (Bar Harbor, ME). Zc3h12a−/− mice were provided by P. Kolattukudy (University of Central Florida) (28), Il17ra−/− mice were from Amgen (Seattle, WA), Il17a−/− mice were from Y. Iwakura (University of Tokyo) and Il17c−/− were from Genentech via the Mutant Mouse Regional Resource Center (Davis, CA). Experimental mice were female, matched for age, between 6 and 10 wk of age, and used in accordance with approved University of Pittsburgh Institutional Animal Care and Use Committee protocols and the National Institutes of Health’s Guide for the Care and Use of Laboratory Animals.
Dorsal skin was shaved with an electric razor and subsequently epilated with Nair cream to facilitate cutaneous absorption of treatment. Following a 48-h rest, 6.25 mg of 5% imiquimod (IMQ; 3M; Aldara) was applied daily to dorsal and ear skin. Contralateral ears or control groups were treated with a control cream (Vanicream; Pharmaceutical Specialties, Rochester, MN). Ear thickness was measured with a caliper (Mitutoyo). Gross pathology was recorded daily by investigators blinded to the experimental cohorts. Ten 4-mm punch biopsies from random sections of skin were taken after euthanasia; five were used for flow cytometry, and five were snap-frozen for RNA extraction.
To generate bone marrow (BM) chimeras, mice were irradiated with two 500-rad doses delivered 4 h apart. A total of 1 × 107 BM mice cells was delivered i.v. 16 h postirradiation. Mice were given sulfamethoxazole (960 μg/ml) and trimethoprim (192 μg/ml) in drinking water for 2 wk to prevent infections. Chimeras were allowed to engraft for 45 d before experimentation.
RNA isolation and quantitative real-time PCR
Human biopsies were homogenized, and RNA was isolated with an RNeasy Plus Mini Kit (QIAGEN). For quantitative real-time PCR (qPCR), RNA was reverse transcribed using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). qPCR was performed using a 7900HT Fast Real-Time PCR system with TaqMan primers (both from Applied Biosystems). Expression was normalized to RPLP0. Mouse skin was homogenized in QIAzol (QIAGEN), in gentleMACS M Tubes (Miltenyi Biotec), and total RNA was isolated using an RNeasy Lipid Mini Kit (QIAGEN). Cells were lysed in Buffer RLT, and RNA extraction was performed using an RNeasy Mini Kit (QIAGEN). cDNA was prepared using a Superscript III First-Strand Kit (Invitrogen). qPCR was performed with QuantiTect Primer Assays (QIAGEN) and SYBR Green FastMix, ROX (Quanta Biosciences, Gaithersburg, MD). Expression was normalized to Gapdh for mouse skin and Rplp0 for KCs. Samples were analyzed on a 7300 Real-Time instrument (Applied Biosystems). Threshold cycle (Ct) values were obtained for target and housekeeping genes, and ΔCt values were calculated. Expression was normalized to Gapdh for mouse skin and Rplp0 for KCs. Gene induction was transformed to a linear scale via calculation of 2−ΔCt (relative expression).
Histology and immunohistochemistry
Immunohistochemistry was performed on 5-μm paraffin sections from human skin. Paraffin-embedded sections were heated at 60°C for 30 min, deparaffinized, and rehydrated. Slides were placed in Ag-retrieval buffer (pH 6) and heated at 95°C for 20 min. After cooling, slides were treated with 3% H2O2 (5 min) and blocked with 10% goat serum (30 min). Slides were incubated for 2 h at room temperature with MCPIP1 Abs (4.8 μg/ml, 1:100 dilution, catalog number 25009-1-AP; Protein Tech Group). Slides were washed and treated with 2°C Ab and peroxidase (30 min) and diaminobenzidine substrate. Mouse skin samples were fixed in 10% formalin and embedded in paraffin, and H&E staining was performed at the University of Buffalo Histology Core (Buffalo, NY).
Skin biopsies were collected in media + GolgiStop. Samples were lysed in gentleMACS C Tubes (Miltenyi Biotec) with 1.5 ml of digestion mix (Liberase TM, DNase I, and GolgiStop) and incubated at 37°C for 80 min. Homogenates were filtered through 70-μm strainers. Single-cell suspensions were blocked with Fc and surface stained with Abs against CD45 (clone 30-F11), Ly6G (1A8), CD11b (M1/70), CD3 (145-2C11), and γδ TCR (GL3). Cells were permeabilized with Cytofix/Cytoperm (BD) and stained with anti–IL-17A Abs (TC11-18H10) or isotype control. Cells were acquired with a BD Fortessa and analyzed by FlowJo software (TreeStar).
Primary mouse KCs
KCs were prepared as described (29, 30). Briefly, skin was excised from postnatal day 2 skin, stretched in a 6-cm culture dish, and incubated for 2 h at 4°C. Skins were floated on 0.25% trypsin for 16 h at 4°C. Dermis was separated from epidermis and discarded, and undifferentiated cells were retrieved by vortexing in Eagle’s MEM (EMEM) with 8% FBS + 1.4 mM Ca2+. A total of 4 × 105 cells per well was plated in 12-well dishes in EMEM with 8% FBS + 0.3 mM Ca2+. After 16 h, media were replaced by EMEM with 8% FBS and 0.05 mM Ca2+ to promote differentiation. KCs were rested for 3 d and stimulated for 16 h with mouse IL-17A (200 ng/ml), IL-17C (200 ng/ml), or TNF-α (2 ng/ml). Supernatants were collected for ELISA, and cells were scraped, pelleted, and snap-frozen for RNA analysis.
Data are presented as mean ± SEM. Statistical analysis was performed using ANOVA, followed by the Bonferroni post hoc test or Mann–Whitney U test, with GraphPad Prism (La Jolla, CA). All experiments were performed a minimum of twice. For qPCR data, statistical analyses were run on ΔCt values. The p values < 0.05 were considered significant.
MCPIP1 is expressed in psoriasis and limits disease severity in a mouse psoriasis model
To ascertain whether MCPIP1 is linked with psoriasis, we assessed ZC3H12A expression in lesional and nonlesional skin from psoriasis patients or healthy controls (healthy normal skin). ZC3H12A mRNA expression was elevated ∼10-fold in psoriasis lesions compared with nonlesional skin from the same patient or with skin from healthy controls (Fig. 1A). Immunohistochemical staining of skin samples indicated that MCPIP1 was expressed primarily in the epidermis, but it was also detectable in the inflammatory infiltrate (Fig. 1B), consistent with the known expression pattern of this protein (23).
In mice, topical application of IMQ leads to IL-17A– and IL-23–dependent development of skin lesions with the hallmarks of human psoriasis (31, 32). This system is considered a good model of the early events in psoriatic plaque formation (33). Consistent with the clinical samples, Zc3h12a mRNA expression was induced throughout the course of IMQ-driven psoriasiform dermatitis (Fig. 2A). Expression was detectable as early as 1 d following IMQ treatment. As expected, Il17a and Il17c were also upregulated during this time frame (Fig. 2B, 2C). Il17c expression increased rapidly, peaking by day 2 and remaining elevated. Il17a mRNA was induced somewhat later, with progressively increased levels seen over 5 d.
MCPIP1 negatively regulates several inflammatory stimuli, but its role in psoriasis is poorly defined (18, 34). Zc3h12a−/− mice cannot be used for experimentation because they exhibit severely shortened lifespans as a consequence of unrestricted TLR signaling and widespread inflammation (23, 28). To circumvent these confounding issues, we assessed disease in haploinsufficient mice (Zc3h12a+/−) (18). Importantly, Zc3h12a+/− mice, unlike Zc3h12a−/− mice, did not present with exacerbated baseline inflammatory levels in skin, as determined by expression of proinflammatory cytokines, chemokines, and antimicrobial proteins (Il17a, Ifng, Csf2, Lcn2, Cxcl1, Cxcl5) (Fig. 2D). There also was no difference in cutaneous neutrophil infiltration in Zc3h12a+/− mice versus Zc3h12a+/+ littermates (hereafter termed wild-type [WT]) (Fig. 2E). However, after 3 d of IMQ treatment, Zc3h12a+/− mice showed increased disease severity compared with WT, with enhanced erythema, epidermal thickening, skin scaling, and ear swelling (Fig. 2F, 2G). H&E staining of dorsal skin from IMQ-treated Zc3h12a+/− mice revealed increased neutrophil microclusters and parakeratosis compared with WT (Fig. 2H). Therefore, MCPIP1 restricts inflammation in IMQ-driven dermatitis.
MCPIP1 limits IL-17A– and IL-17C–mediated skin inflammation
In T cells, MCPIP1 deficiency is associated with increased TCR signaling and amplified Th17 differentiation (27). Thus, an explanation for the increased skin inflammation in Zc3h12a+/− mice could be increased IL-17A production in lymphocytes following IMQ treatment. Vγ4+ γδ T cells are a major source of IL-17A in IMQ-induced inflammation (35, 36). However, the levels of γδ T cell–produced IL-17A did not differ between IMQ-treated Zc3h12a+/− mice and WT (Fig. 3A). Similarly, Il17a, Il17f, and Il17c mRNA levels were not statistically different in IMQ-treated Zc3h12a+/− mice compared with controls and neither was Il23 or its receptor, Il23r (Fig. 3B).
Because the increased IMQ-induced skin inflammation in Zc3h12a+/− mice did not appear to be due to increased IL-17A or IL-17C expression, we hypothesized that it might instead be through increased IL-17R signaling. A hallmark of psoriasis is IL-17A–dependent neutrophil infiltration (31). Commensurate with their increased inflammation, Zc3h12a+/− mice showed increased neutrophil frequency, accompanied by increased expression of the neutrophil-attractive chemokine Cxcl5 and IL-17A–associated genes Lcn2 and Il6 (Fig. 3C, 3D). To directly test the hypothesis that the enhanced susceptibility to IMQ-induced dermatitis in Zc3h12a+/− mice was due to enhanced downstream IL-17 signaling, Zc3h12a+/− mice were crossed to Il17ra−/− mice and treated with IMQ. Zc3h12a+/−Il17ra−/− mice exhibited markedly reduced inflammation compared with Zc3h12a+/− mice, with milder skin pathology, reduced neutrophil infiltration (Fig. 4A, data not shown), and abrogated expression of Il17a, Il17c, and IL-17–dependent genes associated with psoriasis, such as Il6 and Lcn2 (Fig. 4B–E). In line with previous reports, Il17ra−/− mice also showed elevated IL-17A and IL-17C expression at baseline (Fig. 4B) (37, 38). These data verify that the MCPIP1-dependent phenotype is due to IL-17 signaling and not other cytokines associated with psoriasis.
IL-17RA is a shared subunit used by several IL-17 family cytokines, so the enhanced inflammation that we observed could be due to more than one ligand (11, 39). In particular, IL-17C is reported to signal through IL-17RA paired with IL-17RE, inducing a characteristic subset of inflammatory genes and AMP expression in KCs. To determine which ligands were affected by MCPIP1, Zc3h12a+/− mice were crossed to Il17a−/− and Il17c−/− mice. Following IMQ treatment, Il17a and Il17c expression levels did not differ among WT, Zc3h12a+/−, Zc3h12a+/−Il17a−/−, and Zc3h12a+/−Il17c−/− mice (Fig. 4B, 4C). Notably, Zc3h12a+/−Il17a−/− mice and Zc3h12a+/−Il17c−/− mice exhibited less skin inflammation than did Zc3h12a+/− mice, with decreased neutrophil accumulation and reduced Lcn2, Il6, and Defb4 expression (Fig. 4). In addition, Il17a−/−, Zc3h12a+/−Il17a−/−, Il17c−/,− and Zc3h12a+/−Il17c−/− mice presented comparable levels of inflammation. These results demonstrate that the increased dermal inflammation seen with MCPIP1 deficiency can be partially rescued through elimination of IL-17A or IL-17C signaling and more strongly reversed in the absence of the shared subunit IL-17RA.
MCPIP1-driven regulation of IL-17 signaling is restricted to the stromal compartment
MCPIP1 regulates the effector mechanisms of IL-17A and its production by Th17 cells. To determine the relative contribution of MCPIP1 to the regulation of the immune and stromal compartments, we generated BM chimeric mice with MCPIP1 haploinsufficiency in hematopoietic or radioresistant cells and subjected them to IMQ dermatitis. Mice receiving Zc3h12a+/− BM developed gross skin inflammation that was indistinguishable from that of mice receiving WT BM (data not shown). Consistently, they showed comparable levels of skin neutrophil accumulation (Fig. 5A). In contrast, Zc3h12a+/− mice given WT BM developed exacerbated skin pathology with increased neutrophilia, comparable to those receiving WT BM (Fig. 5A). These data indicate that MCPIP1 regulates the skin phenotype specifically in stromal and radioresistant skin cells.
KCs are the most abundant cell type in the epidermis and serve a key role in the initiation and perpetuation of immune responses. KCs are highly responsive to IL-17A and IL-17C (14, 40, 41). To determine whether MCPIP1 restricts IL-17R signaling in KCs, primary KC cultures from postnatal day 2 WT and Zc3h12a−/− mice were assessed for expression of various genes in response to IL-17A and IL-17C after 16 h of cytokine treatment. Consistent with a role for MCPIP1 as a negative regulator of signaling, Zc3h12a−/− KCs showed enhanced expression of IL-17A and IL-17C target genes, including Cxcl1, Defb3, and S100a7 (Fig. 5B–D). MCPIP1 was shown to induce degradation of mRNA transcripts encoding IL-17RA and IL-17RC (18). Levels of Il17re mRNA were slightly, but not significantly, reduced in Zc3h12a−/− KCs (Fig. 5E). However, Il17rc and Il17ra mRNA levels were modestly, but consistently, elevated in Zc3h12a−/− KCs compared with WT (Fig. 5F, 5G), perhaps contributing to increased cytokine signaling.
IL-17A and IL-17C play important roles in controlling infections, but they drive pathology in inflammatory conditions, especially psoriasis. Elevated levels of IL-17A and IL-17C occur in lesional human psoriasis skin samples (14, 39, 42), and human KCs stimulated with IL-17A upregulate antimicrobial peptides and neutrophil-attracting chemokines (43, 44). Consistently, genome-wide association studies (GWASs) identified psoriasis-associated polymorphisms in genes critical for Th17 differentiation and IL-17A signaling, such as IL23R and TRAF3IP2 (encoding the essential IL-17–signaling intermediate Act1) (45–49). Similarly, mouse preclinical models of psoriasis revealed a role for IL-17 family cytokines in mediating disease. In the IMQ dermatitis model, IL-17RA–deficient mice show dramatically reduced disease development (31). IL-17C intradermal application of mice led to enhanced leukocyte recruitment in skin, and IL-17C–deficient mice develop milder skin inflammation upon IMQ treatment (11). Additionally, a transgenic mouse overexpressing IL-17C in KCs develops spontaneous skin lesions with many features of human psoriasis (14). The importance of IL-17A–mediated inflammation in psoriasis was highlighted more recently by the clinical success of biologic drugs, including the IL-17A–blocking Abs secukinumab and ixekizumab and the IL-17RA–targeting Ab brodalumab (2, 50–53).
Given its prominent role in inflammation, it is not surprising that the IL-17A signaling pathway is tightly controlled. Indeed, recent studies elucidated several mechanisms by which IL-17A signaling is negatively regulated. For example, TRAF3 and TRAF4 interfere with receptor-proximal events by competing with Act1 or TRAF6 for IL-17RA occupancy (54, 55). The deubiquitinase A20 (TNFAIP3) is induced by IL-17A and mediates the removal of K63-linked ubiquitin chains on TRAF6, tempering activation of NF-κB and MAPK pathways (56). Similarly, USP25-mediated deubiquitination of TRAF5 and TRAF6 dampens IL-17A signaling (14). GSK-3β–mediated phosphorylation of the transcription factor C/EBPβ inhibits IL-17–dependent gene expression (57). Finally, MCPIP1 is a feedback inhibitor that degrades IL-17–induced target genes, including Il6, Nfkbiz, and transcripts encoding IL-17R subunits (18, 21, 58). A new report indicates that ABIN-1 (Tnip1) controls IL-17–induced pathology in IMQ-induced dermatitis (59). The concept that multiple inhibitors are needed to adequately control inflammatory cytokines was elegantly reviewed by Carpenter et al. (60). Despite its emerging role in inflammatory diseases (14), IL-17C signaling mechanisms are largely undefined, and no study has focused on negative regulation of IL-17C–dependent signal transduction.
In this article we show that MCPIP1 restricts IL-17A– and IL-17C–driven skin inflammation. Deficiency of one copy of the Zc3h12a gene was sufficient to render mice hypersusceptible to IMQ-driven psoriasis. We reported previously that the strong inflammatory phenotype in Zc3h12a+/− heterozygous mice is not due to impaired baseline levels of MCPIP1 in peripheral organs, such as kidney and lung; rather, haploinsufficient cells failed to show ligand-inducible upregulation of MCPIP1 (18). We verify in this study that Zc3h12a+/− mice also do not show exacerbated baseline inflammation in skin (Fig. 2). Notably, Zc3h12a+/− mice on an Il17a-deficient or Il17c-deficient background exhibited milder dermatitis than did Zc3h12a+/− mice (Fig. 4). However, the absence of Il17ra was required to render Zc3h12a+/− mice fully resistant to psoriasis, indicating a contribution of both IL-17 family members to skin pathology in this model system.
GWAS analysis of human psoriasis revealed associations with known regulators of immune signaling, including TNFAIP3 (A20), TNIP1 (ABIN-1, NAF1), and NFKBIA (IκBα) (61). Eleven nonsynonymous single-nucleotide polymorphisms were reported for the ZC3H12A gene, but none are associated with human disease (62). MCPIP1 belongs to a four-member family of CCCH-type zinc finger proteins. It is intriguing that polymorphisms in ZC3H12C (encoding MCP-1–induced protein 3) were identified in a GWAS study (48, 63). Similar to MCPIP1, MCP-1–induced protein 3 has endoribonuclease activity and curbs inflammatory responses in endothelial cells by suppressing NF-κB signaling (64). In addition, ZC3H12C and ZC3H12A are induced in LPS-treated macrophages (65), although it is not known whether ZC3H12C is an IL-17 target gene. Future studies are needed to establish how different MCP-1–induced protein family members regulate inflammation.
We found that ZC3H12A expression was markedly elevated in human psoriatic lesions compared with normal or uninvolved psoriatic skin (Fig. 1), consistent with published data (20, 66) and our data in mice (Fig. 2). Although this may seem counterintuitive given the role of MCPIP1 in restricting inflammation, it is common for feedback inhibitors like TNFAIP3 and TNIP1 to be associated with inflammation (67). It is likely that the increased ZC3H12A mRNA levels reflect the ongoing inflammatory milieu and particularly the high IL-17A levels found in diseased skin (18, 20, 22, 28).
BM chimeras demonstrated that MCPIP1 haploinsufficiency specifically in resident skin cells, and not in radiation-sensitive hematopoietic cells, is sufficient to exacerbate inflammation in the IMQ model (Fig. 5). This agrees with the restricted pattern of tissue expression of the IL-17RC and IL-17RE subunits, whereas IL-17RA is more broadly expressed. Prior studies in a model of autoimmune arthritis similarly demonstrated a role for IL-17RA only in the nonhematopoietic compartment (68). These results also agree with data showing that Act1 drives the development of dermal and epidermal pathology dominantly in cells of stromal origin (i.e., KCs, endothelial cells, and skin fibroblasts) (32). Previously, MCPIP1 was shown to negatively regulate TCR signaling and IL-17A production by Th17 cells (27). However, we saw that Zc3h12a haploinsufficiency did not cause increased IL-17A production at the mRNA or protein levels during IMQ inflammation (Fig. 2). Of note, γδ T cells are the main IL-17A–producing population in the skin in this model, whereas the frequency of IL-17+ CD4+ T cells was very low in Zc3h12a+/− mice. Accordingly, because this is an acute disease model in which IL-17 is made predominantly by γδ T cells, a role for MCPIP1 in conventional CD4+ T cells may not be apparent.
In summary, we identified a novel role for the endoribonuclease and deubiquitinase MCPIP1 in restricting IL-17A– and IL-17C–mediated skin inflammation. To our knowledge this is the first report of a negative-regulatory element in the IL-17C signaling pathway, although it is not likely to be the last (60). Future studies will lend insight into the degree of conservation of MCPIP1-driven mechanisms of IL-17 cytokine family regulation.
We thank P. Kolattukudy for Zc3h12a+/− mice, Y. Iwakura for Il17a−/− mice, Amgen for Il17ra−/− mice, and Genentech for Il17c−/− mice.
S.L.G. was supported by National Institutes of Health Grants AI107825, DE022550, and AR062546 and has received grants from Novartis and Janssen. N.L.W. was supported by National Institutes of Health Grants AR062546, AR39750, AR063437, and AR063852. J.E.G. was supported by National Institutes of Health Grant AR069071 and by the A. Alfred Taubman Medical Research Institute Frances and Kenneth Eisenberg Emerging Scholar Award. A.M. was supported by National Institutes of Health Grants AR066548 and AR067746.
This work is solely the responsibility of the authors and does not necessarily reflect the views of the National Institutes of Health or other funding bodies.
Abbreviations used in this article:
genome-wide association study
MCP-1–induced protein 1
quantitative real-time PCR
S.L.G. is on the Scientific Advisory Board of Lycera. The other authors have no financial conflicts of interest.