Fig 2: TG2 mediates IMQ-induced IL-6, CXCL8, and CCL20 expression in keratinocytes.a WT and TG2−/− mice (n = 11/group) were treated with IMQ for four consecutive days. Cytokine and chemokine mRNA levels in the back skin epidermis were measured by RT-PCR. b, c Primary keratinocytes prepared from WT and TG2−/− mice were treated with IMQ (200 μM) or Aldara cream (25 μM) for 3 h. Ccl20, Il6, and Cxcl1 mRNA (b) and protein levels in the media (c) were measured by RT-PCR and a multiplex cytometric bead array, respectively (n = 3/group). d, e HaCaT cells stably transfected with control and TG2 shRNA were treated with IMQ (200 μM) or Aldara cream (25 μM). IL6, CXCL8, and CCL20 mRNA (d) and protein levels in the media (e) were measured by RT-PCR after 3 h and a multiplex cytometric bead array after 6 h, respectively. Protein levels were normalized to total cellular soluble protein (n = 3/group). f Human primary keratinocytes stably transfected with control or TG2 shRNA were treated with IMQ (200 μM; n = 3). IL6, CXCL8, and CCL20 mRNA levels were measured by RT-PCR after 3 h. Data represent the mean ± SEM (n = 3/group). *p < 0.05; **p < 0.01.

Fig 3: TG2-dependent chemokine production in keratinocytes is critical for IL-17-producing CCR6+ γδT-cell and neutrophil dermal infiltration.a, b Effect of DMSO- or IMQ-treated WT- or TG2−/−-CM on CCR6+ γδT-cell (a) and CD11b+ Ly-6G+ cell (b) migration according to Transwell migration assays. Flow cytometric analysis of the percentage of cells migrating toward the CM in the presence or absence of anti-CXCL1 (a) and anti-CCL20 (b) antibodies (n = 3/group). c, d WT and TG2−/− mouse-derived CCR6+ γδT-cell (c) and CD11b+ Ly-6G+ cell (d) migration in response to recombinant mouse CCL20 and CXCL1, respectively. Data represent the mean ± SEM (n = 3/group). *p < 0.05; **p < 0.01. e, f Cells were isolated from the back skin of WT and TG2−/− mice treated with IMQ for four consecutive days and analyzed by flow cytometry. Representative flow cytometric profiles of CCR6+ γδTCRlow cells (e) and CD11bhigh Ly-6G+ cells (f). The percentage of cells is shown. Data represent the mean ± SEM (n = 9/group). g Skin sections were immunostained with CCR6 (green)- and IL-17 (red)-specific antibodies. Nuclei were stained with DAPI (blue). Scale bar, 100 μm. h–j Back skin was separated into epidermis and dermis and analyzed by RT-PCR for IL-17-producing γδT-cell markers (h) (Ccr6 and Il23r), IL17 and related cytokines (i) (Il17a, Il17f, and Il22), and myeloid cell-derived cytokines (j) (Il12a and Il23a). Data represent the mean ± SEM (n = 11/group). *p < 0.05; **p < 0.01 vs. WT mice.

Fig 4: IMQ activates TG2 in keratinocytes, leading to CCL20 expression by activating NF-κB.a HaCaT cells stably transfected with control and TG2 shRNA were treated with IMQ, and TG2 mRNA levels were determined by RT-PCR. b HaCaT cells were treated with IMQ in the presence of 5-(biotinamido) pentylamine (BP), and their lysates were subjected to western blot analysis with peroxidase-conjugated streptavidin to detect BP-incorporated proteins. In situ TG activity was measured according to band intensity by ImageJ software (n = 3/group). c HaCaT cells co-transfected with NF-κB-Luc reporters and phRL-TK control vectors were treated with IMQ for 6 h. NF-κB activity was represented by dual luciferase activity normalized to the control (n = 3/group). d Schematic representation of WT and mutated CCL20-promoter reporter plasmids. Transcription factor-binding sites are indicated. e HaCaT cells co-transfected with reporter constructs and phRL-TK control vectors for 48 h were treated with 200 μM IMQ for 12 h. CCL20-promoter activity was estimated by normalizing firefly luciferase luminescence to phRL-TK vector Renilla luciferase. Data are relative to the pGL3 control (n = 4/group) and represent the mean ± SEM. *p < 0.05; **p < 0.01.

Fig 5: TKO Tregs Display Marked Defects in Mobilization In Vitro and In Vivo(A) CD4+CD25− T cells from spleens of WT and TKO mice after 12 weeks of HFD were treated with anti-CD3 antibodies and TGF-β1 for differentiation into in vitro differentiated Tregs (iTregs). Percentage of WT and TKO CD4+CD25+Foxp3+ Tregs were measured by flow cytometry at the indicated time points (n = 6 per group). (B and C) CD4+CD25− T cells from spleens of WT and TKO mice after 12 weeks of HFD were activated by anti-CD3 antibodies for 24 h and subjected to qRT-PCR analysis (B) or ELISA from supernatants (C) for the indicated cytokines, chemokines, and growth factors (n = 5–9 per group).(D and E) Transwell migration study of CD4+CD25+ Tregs isolated from WT and TKO mice after 12 weeks of HFD. Cells were assessed for migration in the presence or absence of CCL19 (D) or CCL20 (E) (n = 3 per group).(F and G) Flow cytometry for CCR7 (F) or CCR6 (G) expression in WT and TKO Tregs (n = 6 per group).(H) Schematic of PKH26-labeled HFD WT and TKO Tregs adoptively transferred to HFD C57BL/6 mice. Flow cytometry shows percentage of PKH26-expressed cells in liver, VAT, and SAT of recipient mice (n = 6 per group).(I and J) Schematic of glucose uptake study of differentiated 3T3-L1 cells co-cultured with HFD WT and TKO iTreg supernatant (supe) (I). (J) Fluorescence intensity of 2-Deoxy-D-glucose (2-DG) uptake by differentiated 3T3-L1 cells co-cultured with supernatants of WT and TKO CD4+ Tregs in the presence or absence of insulin stimulation (n = 4 per group).(K and L) Schematic of glucose production study of mouse primary hepatocytes co-cultured with HFD WT and TKO iTreg supernatants (K). (L) Glucose production by mouse primary hepatocytes co-cultured with supernatants of HFD WT and TKO CD4+ Tregs (n = 6 per group).Statistical differences are indicated as *p<0.05, **p<0.01, and ***p<0.001. Results are reported as mean ± SEM.Related to Figures S3 and S4.