Psoriasiform dermatitis is driven by IL-36-mediated DC-keratinocyte crosstalk

Abstract

Psoriasis is a chronic inflammatory disorder of the skin affecting approximately 2% of the world's population. Accumulating evidence has revealed that the IL-23/IL-17/IL-22 pathway is key for development of skin immunopathology. However, the role of keratinocytes and their crosstalk with immune cells at the onset of disease remains poorly understood. Here, we show that IL-36R-deficient (Il36r-/-) mice were protected from imiquimod-induced expansion of dermal IL-17-producing γδ T cells and psoriasiform dermatitis. Furthermore, IL-36R antagonist-deficient (Il36rn-/-) mice showed exacerbated pathology. TLR7 ligation on DCs induced IL-36-mediated crosstalk with keratinocytes and dermal mesenchymal cells that was crucial for control of the pathological IL-23/IL-17/IL-22 axis and disease development. Notably, mice lacking IL-23, IL-17, or IL-22 were less well protected from disease compared with Il36r-/- mice, indicating an additional distinct activity of IL-36 beyond induction of the pathological IL-23 axis. Moreover, while the absence of IL-1R1 prevented neutrophil infiltration, it did not protect from acanthosis and hyperkeratosis, demonstrating that neutrophils are dispensable for disease manifestation. These results highlight a central and unique IL-1-independent role for IL-36 in control of the IL-23/IL-17/IL-22 pathway and development of psoriasiform dermatitis.

Figure 1. IL-36R and IL-36R antagonist play critical roles in the development of IMQ-induced psoriasiform dermatitis.

Ears of both Il36r^–/– and Il36rn^–/– mice and WT mice (n ≥ 4 per group) were topically treated with an IMQ-containing cream (Aldara, MEDA Pharma) for 7 consecutive days. (A) Ear swelling was measured daily before treatment. Values show averages ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001. (B) At day 7, photographs were taken from whole ears or (C) H&E-stained ear sections representing the thickest epidermal region of each ear from individual mice representative of indicated groups. Epidermis (e), dermis (d), and hair follicle (f) are indicated. Data are representative of 4 experiments. Scale bar: 100 μm.

Figure 2. IL-36 regulates the recruitment of inflammatory cells and the expansion of IL-17A–producing γδ T cells in the skin.

Mice were treated as described in the legend to Figure 1. At days indicated, mice (n = 3/group) were sacrificed and immune cells in the ear were characterized by flow cytometry. Graphs show total numbers (± SEM) of (A) CD45⁺ hematopoietic cells, (B) Gr-1^hiCD11b⁺ neutrophils, (C) CD11b⁺MHCII^–/lo macrophages, (D) MHCII^hi DCs, (E) γδ T cells, and (F) αβ T cells per ear. (G) Dot plots show expression of TCRγδ and TCRαβ on CD45⁺ cells. The TCRγδ^hi population represents dendritic epidermal T cells. Values represent average percentages ± SD of the gated populations from groups of mice. (H) Total numbers of IL-17A–producing cells in the ears at indicated days. (I) Characterization of cell types producing IL-17A in WT mice throughout the course of IMQ treatment. (J) Percentages of CD4⁺ T cells, γδ T cells, and CD4^–γδ^– cells among IL-17A producers in ears of indicated groups of mice at day 7 after treatment. (K) Percentages of IL-17A producers among γδ T cells in ears of indicated groups of mice. Data are representative of 4 independent experiments. (H–K) Values show averages ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 3. IL-36R on radioresistant resident cells is crucial for the expression of IL-17A in the skin and development of IMQ-induced skin inflammation.

Reconstituted bone marrow chimeras, Il36r^–/– CD45.2 BM → WT CD45.1 mice (KO→WT), WT CD45.1 BM → Il36r^–/– CD45.2 mice (WT→KO), and WT CD45.1 BM → WT CD45.1 mice (WT→WT), were treated with Aldara cream, as described in the legend to Figure 1. (A) Ear thickness was monitored daily. (B–F) At day 7, cell populations in the ear were characterized by flow cytometry. Total numbers of (B) CD45⁺ hematopoietic cells and neutrophils, (C) γδ and αβ T cells, and (D) IL-17A–producing cells in ears of indicated chimeras are shown. (E) Characterization of IL-17A⁺ cells. (F) Percentage of host-derived remaining hematopoietic cell populations that survived the irradiation and were not replaced. (A–C, E, and F) Values indicate averages ± SEM of groups. (D) Symbols represent individual mice, and horizontal lines indicate averages of groups. *P < 0.05; **P < 0.01.

Figure 4. IL-17A production and skin disease can be partially mediated by αβ T cells in γδ T cell–deficient mice.

Mouse ears were treated with Aldara as described above. (A) Ear thickness was monitored daily. Values show averages ± SEM (n ≥ 3). (B–D) At day 7, cell populations in the ear were characterized by flow cytometry. (B) Total numbers of indicated populations and (C) IL-17A–producing cells per ear are shown. Symbols represent individual mice, and horizontal lines indicate averages of groups. Values indicate averages ± SEM. (D) Percentages of αβ T cells, γδ T cells, and non–T cells among IL-17A–producing cells in ears. *P < 0.05.

Figure 5. Il36r^–/– mice are better protected than mice lacking IL-23, IL-17A, or IL-22.

(A, G, and J) Ears of indicated groups of mice were treated with Aldara, and ear thickness was analyzed as described above. Values show averages ± SEM. Cell populations in the ear were characterized by flow cytometry at day 7. Total numbers of (B, H, and K) CD45⁺ hematopoietic cells and neutrophils, (C and L) γδ and αβ T cells, and (D and I) IL-17A–producing cells in ears of indicated mice. (E) Percentage of IL-17A producers among γδ and αβ T cells in ear lesions. (F) Percentage of IL-17A–producing cells gated on lymphocytes in dLNs. (J–L) Il17a^–/– mice treated with neutralizing anti–IL-22 or control (ctrl) mAb compared with Il36r^–/– mice treated with control mAb. (B–F, H, I, K, and L) Symbols represent individual mice, and horizontal lines indicate averages of groups. *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 6. IL-36 promotes IL-17A expression in the skin and IMQ-induced psoriasis independent of IL-1R1 signaling.

Il1r1^–/–, Il36r^–/–, and WT mice were treated with Aldara, as described in the legend to Figure 1. (A) Ear thickness was monitored daily. Values show averages ± SEM (n ≥ 3). (B) At day 7, photographs were taken from whole ears (top row) and H&E-stained ear sections, representing the ear tip (middle row) and the thickest epidermal region (bottom row) from individual mice representative of indicated groups. Scale bar: 100 μm. On day 7, cells prepared from ears were analyzed by flow cytometry. The graphs show the total numbers of (C) infiltrating CD45⁺ hematopoietic cells and neutrophils, (D) αβ and γδ T cells, and (E) IL-17A–producing cells per ear. (C–E) Symbols represent individual mice, and horizontal lines indicate averages of groups. (F) Dot plots showing IL-17A–producing cells and γδ T cells. Values indicate frequencies of IL-17⁺γδ⁺ T cells in adjacent gated fields. *P < 0.05; **P < 0.01.

Figure 7. Autocrine IL-36 production by skin-resident cells promotes expression of chemokines and growth factors involved in disease manifestation.

Ear dermis and epidermis were separated. CD45⁺ hematopoietic cells and CD45^– dermal and epidermal cells, consisting almost exclusively of dermal mesenchymal cells and epidermal keratinocytes, respectively, were purified (>99%) by cell sorting. CD45^– dermal and epidermal cells were cultured for 8 hours in the absence (Ctrl) or presence of IMQ or recombinant IL-36β. (A) Expression of indicated genes was measured by real-time PCR. (B) Tlr7 expression in CD45⁺ and CD45^– cells from dermis and epidermis stimulated as indicated. Values show mean ± SEM normalized to G6PD.

Figure 8. DCs produce IL-36 upon IMQ treatment and are pivotal for IMQ-induced cutaneous pathology.

CD11c-DTR transgenic and WT control mice were injected with diphtheria toxin 6 hours prior to daily treatment with Aldara. (A) Ear thickness was monitored daily. Values show averages ± SEM (n ≥ 4). (B–E) On day 4, cells prepared from ears were analyzed by flow cytometry. Numbers of (B) CD11c⁺ DCs, (C) CD45⁺ cells and neutrophils, (D) αβ and γδ T cells, and (E) IL-17A–producing cells per ear are shown. Symbols represent individual mice, and horizontal lines indicate averages of groups. (F) Il36a, Il36g, and Il23a mRNA expression in BMDCs cultured for 6 hours in the absence (Ctrl) or presence of IMQ, recombinant IL-36β, or IMQ and IL-36β, measured by real-time PCR. Values show mean ± SEM normalized to HPRT. (G) Scheme illustrating development of psoriasiform dermatitis driven by DC-keratinocyte crosstalk via autocrine and paracrine IL-36 loops. Neu, neutrophil. *P < 0.05; **P < 0.01; ***P < 0.001.