Skin Activation, and Inflammatory

The Journal of Immunology

IL-36 Promotes Myeloid Cell Inﬁltration, Activation, and Inﬂammatory Activity in Skin

Alexander M. Foster,* Jaymie Baliwag,* Cynthia S. Chen,* Andrew M. Guzman,* Stefan W. Stoll,* Johann E. Gudjonsson,* Nicole L. Ward,† and Andrew Johnston*

The IL-1 family members IL-36a (IL-1F6), IL-36b (IL-1F8), and IL-36g (IL-1F9) and the receptor antagonist IL-36Ra (IL-1F5) constitute a novel signaling system that is poorly understood. We now show that these cytokines have profound effects on the skin immune system. Treatment of human keratinocytes with IL-36 cytokines significantly increased the expression of CXCL1, CXCL8, CCL3, CCL5, and CCL20, potent chemotactic agents for activated leukocytes, and IL-36a injected intradermally resulted in chemokine expression, leukocyte infiltration, and acanthosis of mouse skin. Blood monocytes, myeloid dendritic cells (mDC), and monocyte-derived DC (MO-DC) expressed IL-36R and responded to IL-36. In contrast, no direct effects of IL-36 on resting or activated human CD4+ or CD8+ T cells, or blood neutrophils, could be demonstrated. Monocytes expressed IL-1A, IL- 1B, and IL-6 mRNA and IL-1b and IL-6 protein, and mDC upregulated surface expression of CD83, CD86, and HLA-DR and secretion of IL-1b and IL-6 after treatment with IL-36. Furthermore, IL-36a–treated MO-DC enhanced allogeneic CD4+ T cell proliferation, demonstrating that IL-36 can stimulate the maturation and function of DC and drive T cell proliferation. These data indicate that IL-36 cytokines actively propagate skin inflammation via the activation of keratinocytes, APC, and, indirectly, T cells. The Journal of Immunology, 2014, 192: 6053–6061.

lobal definition of the human transcriptome has revealed effects of IL-36 on KC, T cells, and APC to better understand the many new members of the IL-1 family, including IL-36a role of IL-36 in inflammatory disease. We demonstrate that IL-36 G (formerly known as IL-1F6) (1), IL-36b (IL-1F8) (2, 3), cytokines induce expression of neutrophil, T cell, and myeloid cell IL-36g (IL-1F9) (2, 4), and IL-36Ra (IL-1F5), which, along with chemokines by KC and IL-36a–induced immune cell infiltration the IL-36R (IL-1Rrp2), constitute an independent IL-1 signaling in vivo in mice. Although human T cells or neutrophils did not system analogous to IL-1a,-1b, -1Ra, and IL-1RI. We (5) and express IL-36R, nor respond to exogenous IL-36 cytokines, others (1, 4) have recently demonstrated that IL-36Ra, IL-36a, IL- monocytes, myeloid dendritic cells (mDC), and monocyte-derived 36b, and IL-36g mRNA and protein are elevated in skin plaques DC (MO-DC) expressed both prerequisite receptors for IL-36, IL- of the inflammatory disease psoriasis, and keratinocytes (KC) 1RAcP, and IL-36R, and responded to IL-36 by becoming acti- were identified as the predominant source (4, 5). Emerging evi- vated and producing inflammatory cytokines. When cultured with dence from mouse models indicates a critical role for the IL-36 allogeneic CD4+ T cells, IL-36a–treated MO-DC were also more system in skin inflammation (1, 6, 7), and a crucial role for human potent drivers of allogeneic MLRs, demonstrating that IL-36 can IL-36Ra was recently highlighted, when missense mutations in stimulate the maturation and function of DC. These data are IL36RN that affect the function of IL-36Ra were identified and consistent with the notion that IL-36 contributes to skin inflam- associated with a form of generalized pustular psoriasis (8, 9). mation by acting on KC, DC, and indirectly upon T cells to drive Despite these advances, the regulation and function of these new tissue infiltration, cell proliferation, and keratinocyte hyperpro- IL-1 family members are poorly understood. Given the over- liferation, all hallmarks of lesional psoriatic skin. expression of the IL-36 cytokines in psoriasis, we investigated the Materials and Methods *Department of Dermatology, University of Michigan, Ann Arbor, MI 48109; and Keratinocyte culture †Department of Dermatology, Case Western Reserve University, Cleveland, OH 44106 Normal human keratinocyte (NHK) cultures were established in serum-free medium optimized for high-density keratinocyte growth (medium 154; Received for publication June 5, 2013. Accepted for publication April 18, 2014. Invitrogen/Cascade Biologics, Portland, OR) using sun-protected skin of This work was supported by the American Skin Association, the Dermatology Foun- three healthy adults, as previously described (10). NHK were used for dation, a Babcock Foundation endowment, and National Institutes of Health Grant experiments in the second or third passage. All cells were plated at 5000 K01 AR064765 (to A.J.); National Institutes of Health Grant K08 AR060802, the cells/cm2 and maintained to 4 d postconfluence. Cultures were then starved National Psoriasis Foundation, the Taubman Medical Research Institute (as Kenneth of growth factors in unsupplemented medium M154 for 24 h before use. and Frances Eisenberg Emerging Scholar), and a Babcock Foundation endowment (to Experiments were carried out under low calcium (0.1 mM) conditions. J.E.G.); and a grant from the National Psoriasis Foundation and National Institutes of Health Grants P30AR39750, P50AR05508, R01AR063437, and R01AR062546 Cultures were stimulated with truncated (11) recombinant human 1–2000 (to N.L.W.). ng/ml IL-36a, b,org, IL-36Ra, or IL-1b (R&D Systems, Minneapolis, MN). Informed consent was obtained from all subjects, under protocols Address correspondence and reprint requests to Dr. Andrew Johnston, 6427 Medical approved by the Institutional Review Board of the University of Michigan. Sciences Building 1, University of Michigan, 1301 East Catherine Street, Ann Arbor, MI 48109. E-mail address: [email protected] This study was conducted in compliance with good clinical practice and according to the Declaration of Helsinki Principles. The online version of this article contains supplemental material. Abbreviations used in this article: DC, dendritic cell; GPP, generalized pustular Mouse experiments psoriasis; KC, keratinocyte; mDC, myeloid dendritic cell; MO-DC, monocyte- derived DC; NHK, normal human keratinocyte; QRT-PCR, quantitative RT-PCR. A total of 5 mg murine rIL-36a or BSA was injected intradermally into two separate dorsal regions of skin of 7-wk-old CD1 mice. Animals were Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 treated every other day for 10 d. A circle was drawn around each site after www.jimmunol.org/cgi/doi/10.4049/jimmunol.1301481 6054 IL-36 PROMOTES INFLAMMATORY ACTIVITY IN SKIN every injection to ensure similar location between days. Prior to the first supplemented with GM-CSF (100 ng/ml) and IL-4 (20 ng/ml) (R&D injection, the animal was anesthetized with isoflourane and shaved. Four Systems). Cultures were fed on day 4. On day 8, DC were seeded into poly hours after the last injection, animals were euthanized and each injection (2-hydroxyethyl-methacrylate) (Sigma-Aldrich)-coated 12-well culture site was harvested. During tissue harvesting, each injection site (∼circular plates (Corning Costar) at a density of 1 million cells/well and stimulated in nature) was dissected from the rest of the dorsal skin and then bisected. for 2 d with a mixture containing IL-6 (10 ng/ml), PGE2 (0.1 mM), along Half of each site was put into either 10% neutral buffered formalin and with IL-1b (10 ng/ml), IL-36a,IL-36b,orIL-36g (100 ng/ml) in 500 ml tissue freezing medium for histological and immunostaining analyses, as complete RPMI 1640. DC phenotype was analyzed by flow cytometry, as previously described in detail (12). The other half was placed into tubes described above, using Abs against CD86 (FUN-1; BD Biosciences), HLA- and snap frozen, and total RNA was extracted. Quantitative RT-PCR (QRT- DR (LN3; eBioscience), CD1a (HI149; eBioscience), CD11c (3.9; eBio- PCR) was performed and data were normalized to the housekeeping gene science), CD123 (6H6; BioLegend), and appropriate isotype control Abs. 18S and expressed as fold change over BSA-treated controls (n = 4). H&E staining and immunohistochemistry using Abs specific for CD4, CD8, Allogeneic MLR CD11b, CD11c (BD Biosciences, San Jose, CA), and F4/80 (eBioscience, MO-DC were matured as above with 10 ng/ml IL-6, 0.1 mM PGE2, 10 ng/ San Diego, CA) were also completed, as described previously (12). ml IL-1b, and 100 ng/ml IL-36a, and, then on day 10, they were incubated with 200,000 allogeneic CD4+ T cells at ratios of 1:20 and 1:100 for 5 d in IL-36R expression round-bottom 96-well culture plates (NUNC). In some cases, T cells were CD4+ and CD8+ T cells and CD14+ monocytes were prepared from PBMC prelabeled with CFSE (Invitrogen) before culture, as directed by manu- by negative immunomagnetic selection (Miltenyi Biotec). Cells were facturer. Cells were stained with anti-CD3 (HIT3a; BioLegend) for 20 min typically .85% pure cultures, as determined by flow cytometry using Abs at 4˚C and then treated with 200 ml1mg/ml DAPI (Invitrogen) in PIPES against CD3, CD4, CD8, and CD14, as detailed below. Human mDC were buffer for 10 min at room temperature. Cells were analyzed by flow magnetically isolated from PBMC using two rounds of positive selection cytometry gating on lymphocytes with the DAPI detection channel set for for CD1c+ cells after negatively selecting CD19+ cells using the BDCA-1 linear detection. microbead kit from Miltenyi Biotec. mDC were assessed to be 95% viable lineage (CD3, CD14, CD16, CD19, CD20, CD56) negative, HLA-DR+ Real-time QRT-PCR + 2 CD11c CD123 phenotype by flow cytometry. Total RNA was isolated Total RNA was isolated (RNeasy mini kit; Qiagen) and reverse transcribed + + from CD4 T cells, CD8 T cells, mDC, and monocytes, and QRT-PCR (High Capacity cDNA Transcription kit; Applied Biosystems, Foster City, for IL-1R1, IL-1RAcP, and IL-36R was carried out, as described below. CA), and transcripts were quantified using a 7900HT Fast Real-Time Surface expression of IL-1R1, IL-1RAcP, and IL-36R was detected by PCR system (Applied Biosystems) normalizing to the expression of the incubating cells with biotinylated polyclonal goat anti-human IL-1R1 housekeeping gene ribosomal protein, large, P0 (RPLP0). TaqMan primer (BAF269), IL-1RAcP (BAF676), and IL-36R (IL-1Rrp2, BAF872) Abs sets were purchased from Applied Biosystems. IL1A Hs00174092_m1, (all 50 mg/ml; R&D Systems) on ice for 30 min. After washing in FACS IL1B Hs00174097_m, IL-6 Hs00174131_m1, CXCL1 Hs00236937_m1, buffer, 25 ml allophycocyanin-labeled streptavidin (BD Biosciences), anti- CXCL9 Hs00171065_m1, CXCL10 Hs00171042_m1, CXCL11 CD3 (S4.1; Invitrogen), CD4 (S3.5; Invitrogen), CD8 (SK1; BD Bio- Hs00171138_m1, CCL1 Hs00171072_m1, CCL2 Hs00234140_m1, CCL3 sciences), and CD14 (M5E2; BioLegend, San Diego, CA) were added for Hs00234142_m1, CCL4 Hs99999148_m1, CCL5 Hs00174575_m1, CCL7 30 min at 4˚C in the dark, and, after two more washes, cells were analyzed Hs00171147_m1, CCL20 Hs00355476_m1, IL-1R1 Hs00168392_m1, with a LSR2 flow cytometer (BD Biosciences). IL1RAP Hs00370506_m1, IL1rL2 (IL36R) Hs00909276_m1, and RPLP0 Hs99999902_m1 were used in this study. T lymphocyte stimulation Statistical analysis CD3+ T cells were isolated from PBMC as above and incubated for 24 h with either unstimulated or 1 mg/ml CD3 and CD28 Abs (BD Bio- Data were tested for normality, and statistical significance was calculated sciences), together with 10 ng/ml IL-1b or 100 ng/ml truncated IL-36a, b, using two-way Student t tests, Mann–Whitney U test, or one-way ANOVA or g. Cultures were subsequently prepared for QRT-PCR or flow cytom- with Dunnet’s posttest, as appropriate, using GraphPad Prism version 5.00 etry. Cells were washed in FACS buffer (PBS + 0.5% BSA + 0.1% NaN3) for Windows (GraphPad Software, San Diego, CA). and then stained with Abs against CD3 (clone S4.1; Invitrogen), CD4 (OKT-4; eBioscience), CD8 (SK1; BD Biosciences), CLA (HECA-452; BioLegend), CD103 (Ber-ACT8; BioLegend), CD25 (BC96; BioLegend), Results CD69 (FN50; BioLegend), CD54 (HCD54; BioLegend), and appropriate IL-36 induces chemokine expression by human keratinocytes isotype-matched control Abs for 30 min at 4˚C in the dark. After two washes in FACS buffer, cells were analyzed using a BD LSR2 flow We have previously shown that human keratinocytes are an im- cytometer gating on lymphocytes expressing CD3 and CD4 or CD8. portant source of antimicrobial peptides when exposed to IL-36 family cytokines (5). We now demonstrate that IL-36 cytokines Neutrophil isolation and stimulation induce the robust expression of chemokines that drive immune Neutrophils were isolated from heparinized peripheral blood, as previously cell chemotaxis. Treatment of NHK with IL-36a, IL-36b, or IL- outlined (13). IL-1R1, IL-1RAcP, and IL-36R expression was determined 36g led to significant increases in the macrophage (CCL3, CCL4, by FACS, as described above, gating on small, highly granular cells CCL5, CCL2, CCL17, and CCL22), T cell chemoattractants expressing CD11b. Neutrophil stimulations were performed with 10 ng/ml IL-1b; 100 ng/ml truncated IL-36a, b,org; 100 ng/ml LPS (Sigma- (CCL20, CCL5, CCL2, CCL17, and CCL22), and the neutrophil Aldrich); or 50 ng/ml PMA (Sigma-Aldrich) for 4 h. Conditioned media chemokines (CXCL8, CCL20, and CXCL1) (Fig. 1A). Moreover, were assayed for CXCL-8 and TNF-a by ELISA (Duoset; R&D Systems). all IL-36R agonists, but not IL-36Ra, dose dependently induced CXCL1, CCL5, CXCL8, and CCL20 mRNA expression and APC isolation and culture CXCL8 and CCL20 protein secretion by NHK (Fig. 1B–E), dem- + CD14 monocytes and BDCA-1 mDC were prepared from PBMC as above onstrating that, following IL-36 exposure, KC are potent sources and stimulated for 12 h in round-bottom 96-well culture plates in complete of macrophage, T cell, and neutrophil chemokines. RPMI 1640 plus 10% FCS supplemented with 100 ng/ml IL-36a, b,org. Cells were analyzed by flow cytometry using Abs against CD83 (HB15e; IL-36 induces myeloid cell infiltration of skin concomitant with BioLegend), CD86 (FUN-1; BD Biosciences), HLA-DR (LN3; eBio- science), and appropriate isotype controls. Alternatively, total RNA was chemokine and growth factor induction prepared from cells and assessed for cytokine transcripts using QRT-PCR, Given that IL-36 induced chemokine expression by human KC as detailed below. Conditioned media were analyzed for IL-1b and IL-6 in vitro, we sought to test whether this drove cell chemotaxis in vivo using ELISA Duosets from R&D Systems. by injecting 5 mg murine rIL-36a or BSA intradermally into CD1 DC generation and culture mice every other day for 10 d. By day 10, the mouse skin did not appear erythemic or thickened, but, histologically, a mild acan- Human MO-DC were generated in vitro, as previously described (14). Briefly, CD14+ monocytes were magnetically isolated from PBMC thosis and an increase in eosinophilic dermal collagen were evi- (Miltenyi Biotec) and cultured in RPMI 1640 containing 10% FCS dent along with a very pronounced leukocytic infiltrate (Fig. 2A). The Journal of Immunology 6055

FIGURE 1. IL-36 cytokines induce chemokine expression by keratinocytes. Four-day postconfluent NHK were stimulated for 24 h with recombinant truncated IL-36R ligands or IL-1b. Total RNA was extracted, mRNA transcripts were quantified by QRT-PCR relative to the housekeeping gene RPL-P0, and conditioned medium was assayed by ELISA. A total of 100 ng/ml IL-36a, IL-36b, and IL-36g significantly induced T cell chemokine mRNA expression compared with untreated cells, mean 6 SD (n =3)(A). IL-36a, IL-36b, IL-36g, and IL-1b, but not IL-36Ra, dose dependently induced CXCL1, CCL5, CXCL8, and CCL20 mRNA expression (B–E) and CXCL8 and CCL20 protein secretion (F and G) by keratinocytes. Mean 6 SD (n = 3). Statistical significance indicated by *p , 0.05, **p , 0.01, or ***p , 0.001, Student t test. 6056 IL-36 PROMOTES INFLAMMATORY ACTIVITY IN SKIN

FIGURE 2. IL-36 induces myeloid cell infiltration of skin concomitant with chemokine and growth factor induction. A total of 5 mg murine rIL-36a or BSA was injected intradermally into CD1 mice every other day for 10 d. Back skin was harvested, snap frozen, and processed for RNA and histochemistry. IL-36a treatment led to acanthosis and an increase in eosinophilic dermal collagen (A) and striking infiltration of granulocytes (CD11b) (B), macrophages (F4/80) (C), DC (CD11c) (D), CD4+ cells (E), but not CD8+ cells (F). These changes were accompanied by increases in chemokines (G) and cytokines/ growth factors (H). Mean 6 SEM (n = 4 mice). Statistical significance indicated by *p , 0.05 (two-tailed t test). Scale bar, 100 mm.

The infiltrate was striking in its largely granulocytic character respond to IL-36 cytokines. First, we examined whether isolated (Fig. 2B–D) with few T cells (Fig. 2E, 2F). QRT-PCR revealed blood CD4+ T cells, CD8+ T cells, monocytes, or mDC expressed that IL-36 treatment induced significant fold changes in a number IL-36R. After magnetic separation, giving .85% pure CD4+ of leukocyte chemokines, including CCL3, CCL4, and CXCL12 T cells, CD8+ T cells, monocytes, and .95% mDC from PBMC, (Fig. 2G), as well as IL-1b and HB-EGF (Fig. 2H; all p , 0.05, we isolated RNA and performed QRT-PCR. Along with mono- fold change versus BSA control, n = 4), further supporting a role cytes, mDC, and primary keratinocytes (NHK), both subsets of for IL-36 in facilitating immunocyte recruitment to inflamed skin. T cells expressed the IL-1R1 (Fig. 3A) and IL-1RAcP (Fig. 3B) receptors. IL-36R transcripts were not detectable from CD4+ or APC, but not T cells, express IL-36R CD8+ T cells (Fig. 3C); however, monocytes, mDC, and NHK Given that KC expressed chemokines in response to IL-36 treat- expressed IL-36R mRNA (Fig. 3C). To demonstrate cell surface ment and IL-36 injected intradermally led to leukocyte infiltration, IL-36R expression, we stained T cells, monocytes, and mDC with we questioned whether T cells, neutrophils, and APC would also Abs against IL-36R and used flow cytometry to show that IL-36R The Journal of Immunology 6057

FIGURE 3. Human APC, but not T cells, express the IL-36R. KC, monocytes, and mDC express IL-1R1, IL-1RAcP, and IL-36R mRNA transcripts; however, CD4+ and CD8+ T cells were found not to express IL-36R as determined by QRT-PCR (A–C)(n = 4 donors). Flow cytometric analysis reveals that, in contrast to mDC (D and G) and monocytes (E, F, and H), T cells (E, F, and I) did not express surface IL-36R. Filled histogram: anti–IL-36R; dotted histogram: isotype control Ab. Flow cytometry gating shown in (D)–(F). Flow cytometry data are representative of six donors. was most strongly expressed on the surface of mDC (Fig. 3D), Moreover, when conditioned media were assayed by ELISA, both which was ∼10-fold more than on monocytes (Fig. 3E) and absent IL-1b and IL-6 were significantly elevated in IL-36a– and b– from the surface of T cells (Fig. 3F). In contrast to myeloid cells, treated mDC cultures (Fig. 5D, 5E). Of note is that the responses blood neutrophils showed no expression of IL-36R and failed of mDC to IL-36 were not mediated by the secreted IL-1b or IL-6 to respond to IL-36 treatment (Supplemental Fig. 1). Likewise, (Supplemental Fig. 3). neither resting nor CD3/CD28-activated CD4+ nor CD8+ T cells DC matured in the presence of IL-36 have increased activity responded to IL-36 treatment (Supplemental Fig. 2). Given that primary blood mDC typically compose only 2% of the IL-36 activates and induces APC to secrete IL-1 and IL-6 PBMC population isolated from blood, we sought to use in vitro Given that monocytes and mDC expressed the prerequisite receptors MO-DC as a surrogate APC. We cultured MO-DC and found for IL-36, we stimulated monocyte and mDC cultures with 100 ng/ml IL-36R was expressed 6-fold more abundantly by CD1a+CD142 IL-36a, b,org, and, in contrast to T cells (Supplemental Fig. 2), MO-DC than precursor CD1a2CD14+ monocytes (Fig. 6A). We monocytes were activated and significantly upregulated expression next examined whether IL-36 could alter MO-DC phenotype of IL-1A, IL-1B, and IL-6 mRNA (Fig. 4A–C) after 12 h of IL-36 during maturation. After 6 d in culture, immature MO-DC were treatment, which is supported by significantly increased IL-1b and stimulated for 48 h with a cytokine mixture that promoted DC IL-6 protein secretion into the conditioned medium (Fig. 4D, 4E). maturation in an IL-1–dependent manner. A combination of IL-6 Likewise, mDC treated with IL-36 cytokines for 12 h significantly and PGE2 with IL-1b, IL-36a, IL-36b, or IL-36g significantly increased the proportion of cells with strong CD83, CD86, and HLA- activated DC, as illustrated by 3-fold increases in CD86 with IL- DR expression, as determined flow cytometrically (Fig. 5A–C). 36a (p = 0.014; Fig. 6B), and significantly increased proportions 6058 IL-36 PROMOTES INFLAMMATORY ACTIVITY IN SKIN

FIGURE 4. IL-36 cytokines induce monocyte expression of inflammatory cytokines. Monocytes treated with 100 ng/ml IL-36 cytokines for 12 h upregulate IL-1A, IL-1B, and IL-6, cytokine transcripts (n = 9 donors) (A–C). Significantly elevated levels of IL-1b and IL-6 were detected in the conditioned culture media after 12 h (n = 6 donors) (D and E). Bars, mean 6 SEM. Statistical significance indicated by *p , 0.05, **p , 0.01, ***p , 0.001 (two-tailed t test). of DC strongly expressing HLA-DR compared with IL-6 and increases in T cell proliferation in the allogeneic MLR in terms PGE2 alone (p = 0.03; Fig. 6C). of both CFSE dilution (Fig. 6D, 6E) and increased numbers of Having established that IL-36 could alter DC phenotype, we superdiploid cells (Fig. 6F). began to assess whether this translated into altered DC function. Thus, we cocultured IL-36–matured DC with allogeneic T cells at Discussion DC:T cell ratios of 1:20 (Fig. 6D, 6E) and 1:100 (Fig. 6D) for 5 d Skin inflammation such as that seen in psoriasis results from the and monitored T cell proliferation using CFSE and DAPI label- multipartite interactions of, at least, KC, T cells, APC, fibroblasts, ing of T cells. Compared with basal stimulation with IL-6 and and endothelial cells (15, 16). In this scenario, KC are not passive PGE2 only, DC matured with IL-1b or IL-36a induced significant and may initiate inflammatory cascades following physical stress,

FIGURE 5. IL-36 cytokines facilitate mDC activation and cytokine secretion. Ex vivo blood mDC treated with 100 ng/ml IL-36 cytokines for 12 h upregulate CD83 (A), CD86 (B), and HLA-DR (C) expression (FACS, n = 6 donors) and secretion of IL-1B (D) and IL-6 (E) (ELISA, n = 9). Bars, mean 6 SD. Statistical signiﬁcance indicated by *p , 0.05, ***p , 0.001 (two-tailed t test). The Journal of Immunology 6059

FIGURE 6. MO-DC have enhanced IL-36R expression, and IL-36 cytokines facilitate DC maturation, driving T cell proliferation. MO-DC expressed

6-fold more IL-36R mRNA than their monocyte precursors (QRT-PCR, n =6)(A). When cultured with 10 ng/ml IL-6 + 0.1 mM PGE2 and either 10 ng/ml IL-1b or 100 ng/ml IL-36a, MO-DC signiﬁcantly increased their surface expression of CD86 (B) and HLA-DR (C), indicative of DC maturation (48 h, bars, mean 6 SEM, n = 8 donors). MO-DC matured with 10 ng/ml IL-6 + 0.1 mM PGE2 and 10 ng/ml TNF-a, 10 ng/ml IL-1b, or 100 ng/ml IL-36a drove allogeneic T cell proliferation, as determined by CFSE dye dilution (D and E) or DAPI labeling of DNA (F). No DC (T cells only), 1:100 DC:T cell ratio, 1:20 DC:T cell ratio showing mean 6 SEM (n = 4 donors). Statistical signiﬁcance indicated by *p , 0.05, **p , 0.01, ***p , 0.001 compared with IL-6 +

PGE2 (two-tailed t test).

UV irradiation, or infection (17, 18). KC are a major source of IL- (Supplemental Fig. 3). In the current study, we focused on the 36 cytokines, particularly during inﬂammation (1, 4, 5, 19, 20). use of the recently characterized, highly active truncated IL-36 We now demonstrate that KC, when treated with IL-36 cytokines, cytokines, which have been shown to bind IL-36R and activated are potent sources of chemokines active upon T cells and APC cells in the ng/ml range (11). The truncated IL-36a and b were at (Fig. 1), and this activity was not mediated via IL-1b or IL-6 least 40-fold more potent than their full-length counterparts (11) 6060 IL-36 PROMOTES INFLAMMATORY ACTIVITY IN SKIN

(Supplemental Fig. 4). Interestingly, truncated IL-36g appears to structure and function of IL-36Ra protein, leading to unrestrained be equipotent to IL-36a and b, yet the full-length version showed IL-36 agonist activity. Although GPP has a strong neutrophil com- little to no agonist activity on KC or APC. ponent, we could not demonstrate IL-36R expression by neutro- Many of the inflammatory mediators induced by IL-36 have been phils or direct activity of IL-36 on neutrophils (Supplemental Fig. 1). demonstrated to be overexpressed in psoriatic skin lesions, in- GPP is also associated with increased activation of Th17 and Th22 cluding CCL3, CCL5, CXCL1, CXCL8, and CCL20 (5, 21, 22), cells during the disease flare (32), and, interestingly, Th17 T cells, suggesting that IL-36 may contribute to the chemokine environ- immature DC, gd T cells, and neutrophils all express CCR6 (21, 33, ment of inflamed skin. Indeed, intradermal injection of murine 34); we show that IL-36 is a strong inducer of the CCR6 ligand rIL-36a to mouse skin resulted in chemokine expression, leuko- CCL20 expression by KC (Fig. 1), consistent with a role for the cyte infiltration, and inflammation (Fig. 2). These data are in ac- IL-36–CCL20–CCR6 axis in driving psoriatic inflammation (7). cord with an earlier model of murine IL-36 expression, in which IL-1 family members have also been shown to act synergistically murine IL-36a was overexpressed in basal mouse epidermis under with other cytokines and growth factors (35), and, in this respect, the control of a K14 promoter (1), resulting in a similar pattern of IL-36a, b, and g are no exception, with synergism reported with chemokine induction and cell infiltration. Moreover, IL-36 cyto- IL-17A, TNF-a (36), and IL-1b (37) extending the potential of kines were shown to be essential mediators in the KC-APC IL-36 in the skin, particularly on psoriatic keratinocytes (38). This crosstalk required to drive imiquimod-induced psoriasiform der- may be particularly relevant given that psoriasis is now considered matitis in mice (7). In this case, stimulation of APC via TLR7 a mixed Th1/Th17 disease and that IL-36 upregulates the ex- leads to KC activation and IL-36R–dependent release of chemo- pression of a number of IFN-g–induced chemokines. kines such as CXCL1 and CCL20 and recruitment of macrophages Our data presented in this study, together with our previous and neutrophils to the skin (7). findings (5), data from others using mice (1, 7), and studies of the KC expressing the above chemokines recruits T cells, APC, and IL36RN mutations associated with pustular psoriasis (8, 9), all neutrophils into the skin; thus, we next examined whether IL-36 suggest an important role for the IL-36 system in skin inflam- was also active on these immune cells. It was recently shown mation: IL-36 induces KC expression of antimicrobial peptides, that mouse CD4+ but not CD8+ T cells respond to murine IL-36 in matrix metalloproteinases (5), and chemokines (Fig. 1), which an IL-36R–dependent manner (6) and that IL-36 synergizes with recruit T cells and APC. IL-36 activates APC and biases their IL-12 to drive a potent murine Th1 response (23). Unlike mouse cytokine profile (Figs. 4–6) (27), which further drives the in- cells, however, human CD4+ and CD8+ T cells do not express IL- flammatory response. Taken together, these data are consistent 36R nor respond to IL-36 either when resting or under CD3/CD28 with the notion that IL-36 cytokines, which we have shown to be stimulation (Supplemental Fig. 2). We show, however, that APC overexpressed in psoriasis (5), can influence the phenotype and such as CD14+ monocytes, CD1a+CD142 MO-DC, and CD1c+ function of DC, with subsequent changes in T cell activity. mDC do express IL-36R (Figs. 2, 6). Recently, Mutamba et al. (24) demonstrated expression of IL-36R (IL-1RL2) by MO-DC Acknowledgments and plasmacytoid DC, but not undifferentiated monocytes, mDC, We thank Ann Pero and Sarah Laponsa for subject recruitment and or bulk PBMC, and these cells responded to IL-36R ligation. We Dr. Xianying Xing, Marybeth Riblett, and Candace Matheny for excellent show that MO-DC express approximately six times more IL-36R technical assistance. transcript than their precursor monocytes (Fig. 6A), and this expression is further doubled in blood mDC (Fig. 2C). Disclosures DC activated with a mixture of IL-6, PGE2, and either IL-1b The authors have no financial conflicts of interest. or IL-36a upregulated CD86 and HLA-DR surface expression (Fig. 6B, 6C). This is in accordance with the ability of murine IL- 36 cytokines to activate mouse bone marrow–derived DC (6) and References IL-36b to drive maturation of MO-DC (24). This is a key obser- 1. Blumberg, H., H. Dinh, E. S. Trueblood, J. Pretorius, D. Kugler, N. Weng, + S. T. Kanaly, J. E. Towne, C. R. Willis, M. K. Kuechle, et al. 2007. Opposing vation as HLA-DR presents peptide Ags to CD4 T cells and activities of two novel members of the IL-1 ligand family regulate skin in- CD86 binds to CD28 on the T cell, providing the costimulatory flammation. J. Exp. Med. 204: 2603–2614. second signal essential for priming naive T cells. We next inves- 2. Towne, J. E., K. E. Garka, B. R. Renshaw, G. D. Virca, and J. E. Sims. 2004. Interleukin (IL)-1F6, IL-1F8, and IL-1F9 signal through IL-1Rrp2 and IL- tigated the functional consequences of IL-36 activity on DC 1RAcP to activate the pathway leading to NF-kappaB and MAPKs. J. Biol. + and incubated IL-36–matured DC with allogeneic CD4 T cells. Chem. 279: 13677–13688. DC treated with IL-36 promoted increased T cell proliferation 3. Magne, D., G. Palmer, J. L. Barton, F. Me´zin, D. Talabot-Ayer, S. Bas, T. Duffy, (Fig. 6D–F), which strongly suggests that IL-36 may influence M. Noger, P. A. Guerne, M. J. Nicklin, and C. Gabay. 2006. The new IL-1 family member IL-1F8 stimulates production of inflammatory mediators by synovial T cell function in skin via its effects on APC. Moreover, we fibroblasts and articular chondrocytes. Arthritis Res. Ther. 8: R80. demonstrate that IL-36 cytokines induce APC expression of IL-1b 4. Debets, R., J. C. Timans, B. Homey, S. Zurawski, T. R. Sana, S. Lo, J. Wagner, and IL-6 (Figs. 4, 5), which potentially contributes to a pro-Th17 G. Edwards, T. Clifford, S. Menon, et al. 2001. Two novel IL-1 family members, IL-1 delta and IL-1 epsilon, function as an antagonist and agonist of NF-kappa B activation environment (25). The epidermal hyperproliferation in psoriasis is through the orphan IL-1 receptor-related protein 2. J. Immunol. 167: 1440–1446. now thought to be driven, at least in part, by IL-17– and IL-22– 5. Johnston, A., X. Xing, A. M. Guzman, M. Riblett, C. M. Loyd, N. L. Ward, secreting T cells infiltrating the skin (26). In humans, naive CD4+ C. Wohn, E. P. Prens, F. Wang, L. E. Maier, et al. 2011. IL-1F5, -F6, -F8, and -F9: a novel IL-1 family signaling system that is active in psoriasis and promotes T cells differentiate into mature Th17 cells in vitro in response to keratinocyte antimicrobial peptide expression. J. Immunol. 186: 2613–2622. IL-1b, IL-6, and/or IL-23 (26–28), and we have recently dem- 6. Vigne, S., G. Palmer, C. Lamacchia, P. Martin, D. Talabot-Ayer, E. Rodriguez, onstrated that IL-1b and/or IL-23 can promote the survival and F. Ronchi, F. Sallusto, H. Dinh, J. E. Sims, and C. Gabay. 2011. IL-36R ligands expansion of Th17 cells from the memory T cell pool (29). This is are potent regulators of dendritic and T cells. Blood 118: 5813–5823. 7. Tortola, L., E. Rosenwald, B. Abel, H. Blumberg, M. Scha¨fer, A. J. Coyle, a particularly important observation, as memory T cells are likely J. C. Renauld, S. Werner, J. Kisielow, and M. Kopf. 2012. Psoriasiform der- to be the most important T cell subset for the maintenance of matitis is driven by IL-36-mediated DC-keratinocyte crosstalk. J. Clin. Invest. psoriasis (30, 31). 122: 3965–3976. 8. Marrakchi, S., P. Guigue, B. R. Renshaw, A. Puel, X. Y. Pei, S. Fraitag, J. Zribi, A form of generalized pustular psoriasis (GPP) has recently been E. Bal, C. Cluzeau, M. Chrabieh, et al. 2011. Interleukin-36-receptor antagonist associated with missense mutations in IL36RN (8, 9) that affect the deficiency and generalized pustular psoriasis. N. Engl. J. Med. 365: 620–628. The Journal of Immunology 6061

9. Onoufriadis, A., M. A. Simpson, A. E. Pink, P. Di Meglio, C. H. Smith, 25. Ghoreschi, K., A. Laurence, X. P. Yang, C. M. Tato, M. J. McGeachy, V. Pullabhatla, J. Knight, S. L. Spain, F. O. Nestle, A. D. Burden, et al. 2011. J. E. Konkel, H. L. Ramos, L. Wei, T. S. Davidson, N. Bouladoux, et al. 2010. Mutations in IL36RN/IL1F5 are associated with the severe episodic inflamma- Generation of pathogenic T(H)17 cells in the absence of TGF-b signalling. tory skin disease known as generalized pustular psoriasis. Am. J. Hum. Genet. Nature 467: 967–971. 89: 432–437. 26. Wilson, N. J., K. Boniface, J. R. Chan, B. S. McKenzie, W. M. Blumenschein, 10. Elder, J. T., G. J. Fisher, Q. Y. Zhang, D. Eisen, A. Krust, P. Kastner, J. D. Mattson, B. Basham, K. Smith, T. Chen, F. Morel, et al. 2007. Develop- P. Chambon, and J. J. Voorhees. 1991. Retinoic acid receptor gene expression in ment, cytokine profile and function of human interleukin 17-producing helper human skin. J. Invest. Dermatol. 96: 425–433. T cells. Nat. Immunol. 8: 950–957. 11. Towne, J. E., B. R. Renshaw, J. Douangpanya, B. P. Lipsky, M. Shen, 27. Cosmi, L., R. De Palma, V. Santarlasci, L. Maggi, M. Capone, F. Frosali, C. A. Gabel, and J. E. Sims. 2011. Interleukin-36 (IL-36) ligands require pro- G. Rodolico, V. Querci, G. Abbate, R. Angeli, et al. 2008. Human interleukin 17- cessing for full agonist (IL-36a, IL-36b, and IL-36g) or antagonist (IL-36Ra) producing cells originate from a CD161+CD4+ T cell precursor. J. Exp. Med. activity. J. Biol. Chem. 286: 42594–42602. 205: 1903–1916. 12. Wolfram, J. A., D. Diaconu, D. A. Hatala, J. Rastegar, D. A. Knutsen, 28. Zhou, L., I. I. Ivanov, R. Spolski, R. Min, K. Shenderov, T. Egawa, D. E. Levy, A. Lowther, D. Askew, A. C. Gilliam, T. S. McCormick, and N. L. Ward. 2009. W. J. Leonard, and D. R. Littman. 2007. IL-6 programs T(H)-17 cell differen- Keratinocyte but not endothelial cell-specific overexpression of Tie2 leads to the tiation by promoting sequential engagement of the IL-21 and IL-23 pathways. development of psoriasis. Am. J. Pathol. 174: 1443–1458. Nat. Immunol. 8: 967–974. 13. Clark, R. A., and W. M. Nauseef. 2001. Isolation and functional analysis of 29. Kryczek, I., A. T. Bruce, J. E. Gudjonsson, A. Johnston, A. Aphale, L. Vatan, neutrophils. Curr. Protoc. Immunol. Chapter 7: Unit 7.23. W. Szeliga, Y. Wang, Y. Liu, T. H. Welling, et al. 2008. Induction of IL-17+ 14. Tedder, T. F., and P. J. Jansen. 2001. Isolation and generation of human dendritic T cell trafficking and development by IFN-gamma: mechanism and pathological cells. Curr. Protoc. Immunol. Chapter 7: Unit 7.32. relevance in psoriasis. J. Immunol. 181: 4733–4741. 15. Elder, J. T., A. T. Bruce, J. E. Gudjonsson, A. Johnston, P. E. Stuart, T. Tejasvi, 30. Boyman, O., H. P. Hefti, C. Conrad, B. J. Nickoloff, M. Suter, and F. O. Nestle. J. J. Voorhees, G. R. Abecasis, and R. P. Nair. 2010. Molecular dissection of 2004. Spontaneous development of psoriasis in a new animal model shows an psoriasis: integrating genetics and biology. J. Invest. Dermatol. 130: 1213–1226. essential role for resident T cells and tumor necrosis factor-alpha. J. Exp. Med. 16. Gudjonsson, J. E., A. Johnston, H. Sigmundsdottir, and H. Valdimarsson. 2004. 199: 731–736. Immunopathogenic mechanisms in psoriasis. Clin. Exp. Immunol. 135: 1–8. 31. Johnston, A., J. E. Gudjonsson, H. Sigmundsdottir, T. J. Love, and 17. Nickoloff, B. J. 2006. Keratinocytes regain momentum as instigators of cuta- H. Valdimarsson. 2004. Peripheral blood T cell responses to keratin peptides that neous inflammation. Trends Mol. Med. 12: 102–106. share sequences with streptococcal M proteins are largely restricted to skin- 18. Kennedy-Crispin, M., E. Billick, H. Mitsui, N. Gulati, H. Fujita, P. Gilleaudeau, homing CD8(+) T cells. Clin. Exp. Immunol. 138: 83–93. M. Sullivan-Whalen, L. M. Johnson-Huang, M. Sua´rez-Farinãs, and J. G. Krueger. 32. Teraki, Y., S. Tanaka, K. Hitomi, and S. Izaki. 2010. A case of generalized 2012. Human keratinocytes’ response to injury upregulates CCL20 and other psoriasiform and pustular eruption induced by infliximab: evidence for skin- genes linking innate and adaptive immunity. J. Invest. Dermatol. 132: 105–113. homing Th17 in the pathogenesis. Br.J.Dermatol.163: 1347–1351. 19. Zhou, X., J. G. Krueger, M. C. Kao, E. Lee, F. Du, A. Menter, W. H. Wong, and 33. Greaves, D. R., W. Wang, D. J. Dairaghi, M. C. Dieu, B. Saint-Vis, K. Franz- A. M. Bowcock. 2003. Novel mechanisms of T-cell and dendritic cell activation Bacon, D. Rossi, C. Caux, T. McClanahan, S. Gordon, et al. 1997. CCR6, a CC revealed by profiling of psoriasis on the 63,100-element oligonucleotide array. chemokine receptor that interacts with macrophage inflammatory protein 3alpha Physiol. Genomics 13: 69–78. and is highly expressed in human dendritic cells. J. Exp. Med. 186: 837–844. 20. Theoharides, T. C., B. Zhang, D. Kempuraj, M. Tagen, M. Vasiadi, 34. Yamashiro, S., J. M. Wang, D. Yang, W. H. Gong, H. Kamohara, and A. Angelidou, K. D. Alysandratos, D. Kalogeromitros, S. Asadi, N. Stavrianeas, T. Yoshimura. 2000. Expression of CCR6 and CD83 by cytokine-activated hu- et al. 2010. IL-33 augments substance P-induced VEGF secretion from human man neutrophils. Blood 96: 3958–3963. mast cells and is increased in psoriatic skin. Proc. Natl. Acad. Sci. USA 107: 35. Johnston, A., J. E. Gudjonsson, A. Aphale, A. M. Guzman, S. W. Stoll, and 4448–4453. J. T. Elder. 2011. EGFR and IL-1 signaling synergistically promote keratinocyte 21. Homey, B., M. C. Dieu-Nosjean, A. Wiesenborn, C. Massacrier, J. J. Pin, antimicrobial defenses in a differentiation-dependent manner. J. Invest. Der- E. Oldham, D. Catron, M. E. Buchanan, A. Muller,€ R. deWaal Malefyt, et al. matol. 131: 329–337. 2000. Up-regulation of macrophage inflammatory protein-3 alpha/CCL20 and 36. Carrier, Y., H. L. Ma, H. E. Ramon, L. Napierata, C. Small, M. O’Toole, CC chemokine receptor 6 in psoriasis. J. Immunol. 164: 6621–6632. D. A. Young, L. A. Fouser, C. Nickerson-Nutter, M. Collins, et al. 2011. Inter- 22. Johnston, A., A. M. Guzman, W. R. Swindell, F. Wang, S. Kang, and regulation of Th17 cytokines and the IL-36 cytokines in vitro and in vivo: J. E. Gudjonsson. 2014. Early tissue responses in psoriasis to the anti-TNF-a implications in psoriasis pathogenesis. J. Invest. Dermatol. 131: 2428–2437. biologic etanercept suggest reduced IL-17R expression and signalling. Br. J. 37. Nguyen, T. T., F. Niyonsaba, H. Ushio, T. Akiyama, C. Kiatsurayanon, Dermatol. R. Smithrithee, S. Ikeda, K. Okumura, and H. Ogawa. 2012. Interleukin-36 23. Vigne, S., G. Palmer, P. Martin, C. Lamacchia, D. Strebel, E. Rodriguez, cytokines enhance the production of host defense peptides psoriasin and LL- M. L. Olleros, D. Vesin, I. Garcia, F. Ronchi, et al. 2012. IL-36 signaling 37 by human keratinocytes through activation of MAPKs and NF-kB. J. Der- amplifies Th1 responses by enhancing proliferation and Th1 polarization of matol. Sci. 68: 63–66. naive CD4+ T cells. Blood 120: 3478–3487. 38. Muhr, P., J. Zeitvogel, I. Heitland, T. Werfel, and M. Wittmann. 2011. Expres- 24. Mutamba, S., A. Allison, Y. Mahida, P. Barrow, and N. Foster. 2012. Expression sion of interleukin (IL)-1 family members upon stimulation with IL-17 differs in of IL-1Rrp2 by human myelomonocytic cells is unique to DCs and facilitates DC keratinocytes derived from patients with psoriasis and healthy donors. Br. J. maturation by IL-1F8 and IL-1F9. Eur. J. Immunol. 42: 607–617. Dermatol. 165: 189–193.