Premalignant melanocyte outgrowth is accompanied by cutaneous immune infiltration.

(A-C) BPT-TdTomato and Tyr-TdTomato control mice were 4-HT-painted and CD45+ cells in the skin enumerated after 35 days by flow cytometry. (A) Schematic diagram of the experimental approach. Insets show representative melanoproliferation in BPT ears 1 month after 4-HT painting. (B) Quantification of the indicated T cell subsets. (C) Quantification of the indicated myeloid cell subsets. Tom+ = TdTomato positive. In B and C, *, **, and *** denote P ≤ 0.05, P < 0.01, and P < 0.001, respectively, calculated by lognormal Welch’s t-test, except in the case of Tom+ cDC1 and Tom+ cDC2, where Mann-Whitney test was applied. (D-E) BPT-TdTomato mice reconstituted with Cd11c-yfp bone marrow were 4-HT-painted and subjected to weekly two-photon imaging for 35 days. (D) Schematic diagram of the experimental approach. (E) Representative image of mutant melanocytes (red) and DCs (yellow) in the skin, with DC-melanocyte interactions indicated by cyan arrowheads. Scale bar = 50 μm. (F) Quantification of DC-melanocyte interaction frequency, normalized to the total number of DCs in each image. Error bars denote SD.

Treg cell depletion promotes mutant melanocyte outgrowth in the BPT model.

(A-B) BPT and BPT;Foxp3-DTR mice were subjected to DT treatment either 1 week before (early), just after (intermediate), or 2 weeks after (late) 4-HT painting, and melanocyte outgrowth assessed after 1 month. (A) Schematic of the experimental protocol. (B) Photographs showing representative melanocytic darkening at the indicated timepoints. (C) qRT-PCR quantification of Tyrp1 expression in ear skin at the experimental endpoint. N = 11 mice per group. Error bars indicate SD. * denotes P ≤ 0.05, calculated by unpaired t-test. (D-E) Foxp3-DTR or control (C57BL/6J) mice were treated with DT and then injected s.c. with 500 B16F10 cells. Subsequent tumor growth was measured for three weeks. (D) Schematic diagram of the experimental approach. (E) Quantification of tumor growth, with error bars indicating SD. * and ** denote P ≤ 0.05 and P < 0.01, respectively, calculated by two-way ANOVA. N = 7 mice per group. (F-G) Foxp3-DTR mice were treated with DT or vehicle control (PBS), then injected i.v. with 2 × 105 B16F10-Luciferase cells. Subsequent tumor growth was measured for three weeks by IVIS imaging. (F) Schematic diagram of the experimental approach. (G) Quantification of tumor growth, with error bars indicating SD. * denotes P ≤ 0.05, calculated by two-way ANOVA. N = 4 mice per group.

Treg cell depletion drives multimodal inflammation of the skin.

BPT-TdTomato;Foxp3-DTR mice and BPT-TdTomato controls were treated with DT, followed by 4-HT painting, and immune cells in the ear skin enumerated by flow cytometry after 8 days. (A) Schematic of the experimental protocol. (B) Quantification of the indicated T cell subsets. (C) Quantification of the indicated myeloid cell subsets. Tom+ = TdTomato positive. In B and C, *, **, ***, and **** denote P ≤ 0.05, P < 0.01, P < 0.001, and P < 0.0001, respectively, calculated by lognormal Welch’s t-test. N = 2 BPT-TdTomato mice and 4 BPT-TdTomato;Foxp3-DTR mice.

Treg cell depletion leads to monocyte/macrophage inflammation and tissue remodeling.

(A-D). CD45+ cells extracted from the ear skin of Foxp3-DTR mice or C57BL/6J controls were analyzed by scRNA-seq 8 days after DT treatment. (A) Schematic of the experimental protocol. (B) UMAP showing all sequenced cells from both samples, colored by sample on the left and by cell type on the right. (C) UMAP reclustering of the monocyte and macrophage clusters identified by the UMAP analysis of all cells in B. Cells are colored by sample on the left and by Seurat cluster on the right. The identities of individual Seurat clusters are shown in the legend below. (D) Feature plots of the monocyte/macrophage UMAP showing expression of the indicated inflammatory and tissue remodeling genes in the C57BL/6J (B6) and Foxp3-DTR (DTR) samples. (E-F) Foxp3-DTR mice and C57BL/6J controls were treated with DT and qRT-PCR performed on skin homogenates 5 days later. (E) Schematic of the experimental protocol. (F) Quantified expression of the indicated genes, with error bars denoting SD. *, **, and *** denote P ≤ 0.05, P < 0.01, and P < 0.001, respectively, calculated by unpaired t-test. N = 4 C57BL/6J mice and 5 Foxp3-DTR mice.

Treg cell depletion destabilizes vasculature in the skin.

(A-C) Foxp3-DTR mice and C57BL/6J controls were treated with DT and cutaneous vasculature imaged 5 days later by two-photon microscopy after dextran injection. (A) Schematic of the experimental protocol. (B) Representative images of dextran and dermal collagen in C57BL/6J (top) and Foxp3-DTR (bottom) skin. Scale bars = 100 μm. (C) Quantification of vascular leakage by fractional area occupied by fluorescent dextran. N = 6 mice per group. (D-E) Foxp3-DTR mice and C57BL/6J controls were treated with DT and vasculature leakage in the ears assessed 5 days later by Evans Blue injection. (D) Schematic of the experimental protocol. (E) Quantification of Evans Blue leakage into the skin. N = 4 C57BL/6J (B6) mice and 3 Foxp3-DTR mice. (F-H) BPT-TdTomato and BPT-TdTomato;Foxp3-DTR mice were treated with DT, painted with 4-HT, and then subjected to two-photon imaging after dextran injection 30 days later. (F) Schematic of the experimental protocol. (G) Representative images of dextran, dermal collagen, and TdTomato+ melanocytes in BPT-TdTomato (top) and BPT-TdTomato;Foxp3-DTR (bottom) skin. Scale bars = 100 μm. (H) Quantification of vascular leakage by fractional area occupied by fluorescent dextran. N = 3 mice per group. All error bars indicate SD. * and ** denote P ≤ 0.05 and P < 0.01, respectively, calculated by unpaired t-test.

UVB irradiation drives myeloid inflammation and tissue remodeling in the skin.

(A-B) BPT mice were UVB irradiated 3 days after 4-HT painting and melanocyte outgrowth assessed at the 30 day timepoint. (A) Schematic of the experimental protocol. (B) Photographs showing representative melanocytic darkening in unirradiated (Ctrl) and UVB-irradiated mice. (C) qRT-PCR quantification of Tyrp1 expression in ear skin at day 30. N = 3 Ctrl mice and 5 UVB-treated mice. Error bars indicate SD. * denotes P ≤ 0.05, calculated by unpaired t-test. (D-E) C57BL/6J mice were UVB- or mock-irradiated and qRT-PCR performed on skin homogenates one day later. (D) Schematic of the experimental protocol. (E) Quantified expression of the indicated genes. N = 3 mice per group. (F-G) C57BL/6J mice were exposed to UVB or mock irradiation and CD45+ cells from the ear skin analyzed by scRNA-seq 7 days later. (F) UMAP reclustering of the monocyte and macrophage clusters identified by UMAP analysis of all cells (see Fig. 6 - supplement 2). Cells are colored by sample on the left, and by Seurat cluster on the right. The identities of individual Seurat clusters are shown in the legend below. (G) Feature plots of the monocyte/macrophage UMAP showing expression of the indicated inflammatory and tissue remodeling genes in the Ctrl and UVB samples. (H-J) C57BL/6J mice were UVB- or mock-irradiated and cutaneous vasculature integrity assessed 3 days later by Evans Blue injection or 8 days later by two-photon microscopy after dextran injection. (H) Schematic of the experimental protocol. (I) Representative images of dextran in mock- (left) and UVB-irradiated (right) skin. Scale bars = 100 μm. (J) Left, vascular leakage at 8 days post UVB, quantified by fractional area occupied by fluorescent dextran. N = 4 mice per group. Right, quantification of Evans Blue leakage into the skin. N = 4 mice per group. All error bars indicate SD. * denotes P ≤ 0.05, calculated by unpaired t-test.

Contact hypersensitivity drives inflammation dependent melanocyte expansion.

(A-C) C57BL/6J mice were sensitized with DNFB and then rechallenged on one ear in the presence or absence of Dex therapy. qRT-PCR analysis of inflammatory genes and Evans Blue analysis of vascular leakage was performed 2 days after elicitation. (A) Schematic of the experimental protocol. (B) Expression of the indicated inflammatory genes was quantified for the indicated DNFB and Dex treatment conditions. N ≥ 3 mice per group. ** and **** denote P < 0.01 and P < 0.0001, respectively, calculated by paired t-test for +/- DNFB comparisons and unpaired t-test for +/- Dex comparisons. (C) Quantification of Evans Blue leakage into the skin. N ≥ 3 mice per group. * and ** denote P ≤ 0.05 and P < 0.01, respectively, calculated by one-way ANOVA. Error bars indicate SD. (D-E) BPT mice were sensitized to DNFB, 4-HT-painted on both ears, and then rechallenged with DNFB on one ear in the presence or absence of Dex therapy. Melanocyte outgrowth was quantified photographically and by qRT-PCR 30 days later. (D) Schematic of the experimental protocol. (E) Left, photographs showing representative melanocytic darkening in DNFB- and vehicle-treated ears, with elicitation performed in the presence or absence of Dex. Right, qRT-PCR quantification of Tyrp1 expression in ear skin at the experimental endpoint. N ≥ 6 mice per group. ** denotes P < 0.01, calculated by paired t-test for +/- DNFB comparisons and unpaired t-test for +/- Dex comparisons. (F-G) BPT-Het mice were sensitized to DNFB, 4-HT-painted on both ears, and then rechallenged with DNFB on one ear. Melanocyte outgrowth was quantified photographically and by qRT-PCR 55 days later. (F) Schematic of the experimental protocol. (G) Left, photographs showing representative melanocytic darkening in DNFB- and vehicle-treated ears. Right, qRT-PCR quantification of Tyrp1 expression in ear skin at the experimental endpoint. N = 3 mice per group. * denotes P ≤ 0.05, calculated by paired t-test.

List of qRT-PCR Primers

List of Antibodies

Flow cytometric analysis of BPT mice.

(A) Gating strategies for lymphoid (left) and myeloid (right) panels, starting from live, CD45+ singlets. (B-C) BPT-TdTomato and Tyr-TdTomato control mice were 4-HT-painted and CD45+ cells in the dLN enumerated after 35 days by flow cytometry. (B) Schematic diagram of the experimental approach. (C) Quantification of the indicated T cell subsets. (D) Quantification of the indicated myeloid cell subsets. Tom+ = TdTomato positive. * and ** denote P ≤ 0.05 and P < 0.01, respectively, calculated by Mann-Whitney test.

Flow cytometric analysis of Treg cell depletion and Tconv cell infiltration in Foxp3-DTR mice.

(A-B) Foxp3-DTR mice and C57BL/6J controls were DT-treated and Treg cells quantified by flow cytometry in the ear skin and dLN 7 and 21 days after the last DT dose. (A) Schematic diagram of the experimental approach. (B) Absolute count of Treg cells in the ear skin (left) and dLN (right). Error bars indicate SD. *, **, and **** denote P ≤ 0.05, P < 0.01, and P < 0.0001, respectively, calculated by one-way ANOVA. N = 2 C57BL/6J control mice and 4 Foxp3-DTR mice. (C-D) Foxp3-DTR mice and C57BL/6J controls were DT-treated and then injected s.c. with B16F10 cells. 20 days later, tumor infiltrating T cells were quantified by flow cytometry. (C) Schematic diagram of the experimental approach. (D) Quantification of tumor infiltrating CD4+ (left) and CD8+ (right) Tconv cells. Error bars indicate SD. * denotes P ≤ 0.05, calculated by lognormal Welch’s t-test. N = 3 C57BL/6J control mice and 7 Foxp3-DTR mice.

Treg cell depletion drives multimodal inflammation in the dLN.

BPT-TdTomato;Foxp3-DTR mice and BPT-TdTomato controls were treated with DT, followed by 4-HT painting, and immune cells in the dLN enumerated by flow cytometry after 8 days. (A) Schematic of the experimental protocol. (B) Quantification of the indicated immune cell subsets. Tom+ = TdTomato positive. * and ** denote P ≤ 0.05 and P < 0.01, respectively, calculated by lognormal Welch’s t-test. N = 2 BPT-TdTomato mice and 4 BPT-TdTomato;Foxp3-DTR mice. (C) Subset analysis of CD4+ (left) and CD8+ (right) Tconv cells in the dLN at the experimental endpoint. Fractions of total represent mean values derived from N = 2 BPT mice and N = 4 BPT;Foxp3-DTR mice. (D) Intracellular cytokine staining by CD4+ Tconv cells after in vitro stimulation with PMA/ionomycin. Error bars indicate SD. N = 2 BPT mice and N = 4 BPT;Foxp3-DTR mice.

Treg cell depletion drives multimodal inflammation in the skin.

CD45+cells extracted from the ear skin of Foxp3-DTR mice or C57BL/6J controls were analyzed by scRNA-seq 8 days after DT treatment. (A) Feature plots of the all cell UMAP, showing expression of Cd4 and Cd8. (B) Gene expression heat map of Immgen-defined T cells, organized by sample (CD57BL/6J (B6) or Foxp3-DTR (DTR)) and by lineage (CD4 or CD8). (C) UMAP reclustering of the DCs identified in the all cell UMAP. Cells are colored by sample on the left and by Seurat cluster on the right. (D) Dot plot showing the expression of critical DC lineage genes, organized by sample and Seurat cluster.

Inflammatory monocyte/macrophage recruitment and differentiation in Treg cell-depleted skin.

CD45+ cells extracted from the ear skin of Foxp3-DTR mice or C57BL/6J controls were analyzed by scRNA-seq 8 days after DT treatment. Monocyte and macrophage clusters from the all cell analysis were subjected to Seurat reclustering. (A) Heat map showing the differentially expressed genes that define the resulting monocyte/macrophage clusters. (B-C) Pseudotime analysis was performed on clusters 0, 1, 2, and 4. (B) Pseudotime trajectory output mapped onto the monocyte/macrophage UMAP. (C) The pseudotime distribution of each cluster, with central line denoting median, colored boxes denoting the central 50th percentile, and whiskers indicating the range, with outliers (determined by Tukey’s method) shown as individual dots.

UVB induces cutaneous inflammation.

(A-B) C57BL/6J mice were UVB- or mock-irradiated and qRT-PCR performed on skin homogenates 7 days later. (A) Schematic of the experimental protocol. (B) Quantified expression of the indicated genes. * denotes P ≤ 0.05, calculated by unpaired t-test. N ≥ 3 mice per group. (C-E) C57BL/6J mice were UVB- or mock-irradiated and immune cells in the ear skin enumerated by flow cytometry after 8 days. (C) Schematic of the experimental protocol. (D) Quantification of the indicated myeloid cell subsets. (E) Quantification of the indicated T cell subsets. In D and E, error bars indicate SD. * and ** denote P ≤ 0.05 and P < 0.01, respectively, calculated by lognormal Welch’s t-test. N ≥ 3 mice per group.

UVB irradiation drives multimodal inflammation in the skin.

C57BL/6J mice were exposed to UVB or mock irradiation and CD45+ cells from the ear skin analyzed by scRNA-seq 7 days later. (A) Schematic of the experimental protocol. (B) UMAP showing all sequenced cells from both samples, colored by sample on the left and by cell type on the right. (C) Monocyte and macrophage clusters from the all cell analysis were subjected to Seurat reclustering. The heat map shows differentially expressed genes that define the resulting monocyte/macrophage clusters.

DNFB-induced hypersensitivity drives cutaneous inflammation.

(A-B) C57BL/6J mice were sensitized with DNFB and then rechallenged on one ear in the presence or absence of Dex therapy. Ear thickness was measured 2 days after elicitation. (A) Schematic of the experimental protocol. (B) Ear thickness measurements, pairing DNFB-elicited ears with contralateral controls. N = 4 mice per group. * denotes P ≤ 0.05, calculated by unpaired t-test. (C-D) C57BL/6J mice were sensitized with DNFB and then rechallenged 5 days later. Immune cells in the ear were enumerated by flow cytometry 2 days after elicitation. (C) Schematic of the experimental protocol. (D) Infiltration of the indicated immune subsets, pairing DNFB-elicited ears with contralateral controls. N = 4 mice per group. * and ** denote P ≤ 0.05 and P < 0.01, respectively, calculated by paired lognormal t-test.