FIGURE LEGENDS
Figure 1 The therapeutic response of psoriasis patients treated with UVB is dictated by a specific cluster of CD4+expressing Treg and TCR-activated markers. (A) UMAP of CD4+ T cells showing a distinct cluster of CD25+FOXP3+CD127-cells (cluster 11) by CyTOF analysis. (B) Heatmap showing the differences among clusters in CD4+ T cells. Each row represents the median arcsinh-transformed intensities in each cluster, and each column represents a maker. (C) Receiver operating characteristic (ROC) curve for the percentage of cluster 11 at baseline predicting UVB treatment response. Models were fit using logistic regression, both unadjusted (blue) and adjusted for gender and age (red). (D) Change in the percentage of cluster 11 CD4+T cells before (pre-UVB) and after (post-UVB) UVB treatment (n = 19). The P value was calculated using the Wilcoxon signed-rank test. PASI: Psoriasis Area Severity Index. (E) Left: Biaxial contour plots illustrating the distribution of HLADR-TIGIT and HLADR-TIM3 expressing cells within cluster 11, pre- and post-UVB treatment. Numbers indicate the percentage of cells in each quadrant. Right: Paired Wilcoxon signed-rank tests were used to compare the frequencies of HLADRhiTIGIThi and HLADRhiTIM3hi cells in cluster 11 before and after UVB treatment.
Figure 2 UVB alleviates OVA-induced inflammatory responses in the skin by increasing OVA-specific Treg cells. (A) Mice were intraperitoneally injected with OVA/alum on days 0, 7, and 14 and epicutaneously treated with OVA patches on day 14. The mice were exposed to UVB every other day three times during the OVA patch application.(B) Images of mouse dorsal skin are shown. (C) Changes in transepidermal water loss (TEWL). n = 5 for the ctrl and O groups; n = 3 for O+U groups. (D) Histological changes within the graft were assessed using hematoxylin and eosin staining. (E-G) Flow cytometry of cells isolated from skin and draining lymph nodes (LNs). Cells isolated from the skin were restimulated. IL-4, IL-5, and IL-22 in CD4+ T cells were analyzed (E). CD25+Foxp3+ Treg cells and the markers associated with TCR activation were analyzed in cells isolated from skin-draining LNs (F). Cells from the skin were incubated with the OVA323-339 tetramers, followed by the antibodies against the indicated markers. OVA-specific Treg cells were analyzed (G). Statistical analyses: one-way ANOVA with Kruskal-Wallis test. Mean ± SEM. *, p<0.05; **, p<0.01; ***, p<0.001.
Figure 3 UVB prolongs allograft survival by inducing allograft-specific Treg cells. (A) Schematic depicting the allograft transplantation experiment. Mice were treated with UVB on alternating days starting on day 5 (d5) for five treatments post-transplantation. n = 9 for the No UVB group; n = 8 for the UVB group. (B) Skin grafts were monitored from d5 post-transplantation. Grafts were considered rejected when less than 20% viable tissue remained.(C) Histological changes within the graft were assessed using hematoxylin and eosin staining. (D) Flow cytometry of cells isolated from the grafts. CD25+Foxp3+ Treg cells and ICOS, CTLA-4, and LAG-3 expression were analyzed. (E) Schematic diagram of the second allograft transplantation. Recipient mice, previously subjected to a first allograft transplant with or without UVB treatment, received a second allograft from either C57BL/6 or C3H mice. (F) Skin graft appearance was monitored until rejection. n = 5 for each of the groups. Statistical analyses: Gehan-Breskiw-Wilcoxon test. *, p<0.05; **, p<0.01; ****, p<0.0001.
Figure 4 Antigen-specific Treg cells affect clinical outcomes of inflammatory skin diseases. (A) Scatter plot illustrating the correlation between the percentage of cluster 11 in CD4+ T cells and PASI score in psoriasis patients. The P value was calculated using Spearman’s correlation. (B) Box-and-whisker plot comparing the percentage of cluster 11 in CD4+ T cells in healthy controls (HC) and psoriasis patients with low (PASI ≤ 4.5), medium (4.5 < PASI < 9), and high (PASI ≥ 9) disease severity. PASI: Psoriasis Area Severity Index. (C) Foxp3DTR-GFP Mice were sensitized with OVA/alum, challenged by the OVA patch, and treated with UVB. CD25+GFP+ Treg cells were sorted from the skin-draining LNs. CD4+GFP-T cells as responder T cells were sorted from the skin-draining LNs of OTII/Foxp3DTR-GFP mice. Treg and responder T cells were mixed at a ratio of 1 to 1 and then were administered to the recipient mice by intravenous injection. The recipient mice were then sensitized with OVA/alum by intraperitoneal injection and challenged with the OVA patches. (D) TEWL was measured every other day. n = 3 for each of the groups. (E) Histological changes within the graft were assessed using hematoxylin and eosin staining. (F) Cells isolated from the skin were restimulated. IL-5 production on CD4+ T cells was analyzed. (G) Sorted CD25+GFP+ Treg cells from mice with OVA stimulation and UVB treatment were intravenously injected into the recipient mice. The ears of the mice were then stimulated with 15 mg of imiquimod-containing cream. (H) Changes in the ear swelling of mice were shown. n = 4 for each of the groups. (I) Cells isolated from ears were restimulated. IL-17 and IL-22 on CD4+ T cells were analyzed. Statistical analyses: one-way ANOVA with Kruskal-Wallis test. P for trend test was calculated by ordinary one-way ANOVA. Mean ± SEM. *, p<0.05; **, p<0.01; ***, p<0.001.
Figure 5 UVB enhances genes involved in cell stability, cell proliferation, and cellular localization, and suppresses genes associated with negative regulation of Treg development. (A) Pathway analysis of upregulated genes in Tregs from skin of mice treated with UVB compared to those of mice without UVB treatment (n = 2 treated and n = 2 untreated, pooled samples). GOBP: Gene Ontology Biological Process; NES: Normalized Enrichment Score. (B) UMAP plot of CD4+ T cells from patients PBMC samples (2 patients, with pre- and post-treatment samples). (C) Dot plot of gene markers for Treg, T memory, T effector, and naive T clusters. (D) Heatmap showing gene expression changes in CD4+ Tregs from mice and human PBMC under UVB treatment. (E) Violin plot of negatively-associated genes (RAB35 , IDH1 , and VAV1 ) with significant gene expression changes (p<0.05) in human PBMC-derived Treg cells under UVB treatment. Statistical significance was tested through the Wilcoxon ranked test.
Figure 6 Treg transition from human CD4+ T cell lineage and clonal expansion under UVB treatment. (A-D) Analyses of CD4+ T cells obtained from the periphery of psoriasis patients treated with UVB and the public dataset generated from healthy human skin. UMAP of CD4+ T cell clusters with an overlaid Slingshot trajectory (A). UMAP plot of 9375 CD4+ T cells in the skin with Slingshot trajectory (B). A chronologically color-coded pseudotime UMAP of CD4+ T cells from violet (least differentiated) to yellow (most differentiated) (C). Expression of key genes for pseudotime trajectory, color coded by the clusters shown in panel A (D). (E) (Left) UMAP visualization of CD4+ T cell subtypes. (Right) Clonal expansion overlay on the UMAP, with higher density indicating greater clonal expansion. (F) Distribution of clonal size across four CD4+ T cell subtypes under UVB treatment. Numbers represent cell counts in each condition. (G) Expansion indices were quantified using the STARTRAC-expa method.