Cutaneous RANK–RANKL Signaling Upregulates CD8-Mediated Antiviral Immunity during Herpes simplex Virus Infection by Preventing Virus-Induced Langerhans Cell Apoptosis  Lars Klenner, Wali Hafezi, Björn E. Clausen, Eva U. Lorentzen, Thomas A. Luger, Stefan Beissert, Joachim E. Kühn, Karin Loser  Journal of Investigative Dermatology  Volume 135, Issue 11, Pages 2676-2687 (November 2015) DOI: 10.1038/jid.2015.225 Copyright © 2015 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 1 Reduced skin lesion size and decreased HSV-1 replication in K14-RANKL TG mice. (a) Skin lesion size of WT and K14-RANKL TG mice after epicutaneous HSV-1 infection (n≥17). (b and c) Representative photographs of skin pathology (b) and immunofluorescence staining (c) of mock-infected contralateral skin (c.s.) at day 8 as well as HSV-1-infected skin from WT and K14-RANKL TG mice at indicated time points post infection. Original magnification × 200, bar=50 μm. Results of statistical analysis of HSV-1-positive area from infected skin of n=5 mice in each group (c). (d) Quantitative real-time PCR of UL54 expression in lesional skin at days 2, 4, 8, and 13 post infection (n≥3). (e) Immunofluorescence staining of mock-infected contralateral skin (c.s.) at day 8 as well as HSV-1-infected skin from WT and K14-RANKL TG mice at indicated time points post infection. Original magnification × 200, bar=50 μm. All the data are shown as mean±SEM; *P<0.05 versus HSV-1-infected WT. Journal of Investigative Dermatology 2015 135, 2676-2687DOI: (10.1038/jid.2015.225) Copyright © 2015 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 2 Increased numbers of cytotoxic T lymphocytes in HSV-1-infected K14-RANKL TG mice. (a and b) Representative dot plots (a) and statistical evaluation (b) of cytolytic markers in CD8+ T cells from regional lymph nodes at day 8 post infection (n≥3). (c) Representative dot plots and statistical evaluation of BrdU incorporation in CD8+ T cells from regional lymph nodes at day 8 post infection (n=5). (d) Representative proliferation assay of CFSE-labeled CD8+ T cells from regional lymph nodes of HSV-1- and mock-infected mice co-cultured with HSV-1-loaded DCs. Numbers of cell divisions are indicated. (e) Representative immunofluorescence staining and statistical evaluation of CD8+ T cells expressing granzyme A (gzmA) in HSV-1-infected skin at day 8 post infection (n=12; original magnification × 400; bar = 20 μm; CD8+gzmA+ cells are highlighted by arrows). (f and g) Representative microscopy (f) and statistical evaluation (g) of target cell lysis in cytotoxicity assays of CD8+ T cells from HSV-1- or mock-infected mice, co-cultured with cell tracker orange–labeled ovalbumin- or HSV-1-loaded target cells (n≥5 visual fields; bar=20 μm; lysed cells are highlighted). All the data are shown as mean±SEM; *, P<0.05 versus HSV-1-infected WT. Journal of Investigative Dermatology 2015 135, 2676-2687DOI: (10.1038/jid.2015.225) Copyright © 2015 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 3 Cutaneous RANK–RANKL signaling upregulates MHC class I-restricted antiviral immunity. (a and b) Adoptive transfer of CD8+ T cells from infected CD45.2+ K14-RANKL TG and WT donors into naive CD45.1+ recipients (WT/TG →CD45.1+) 1 day before HSV-1 infection of recipient mice. Mean skin lesion size from n≥6 mice per group is shown (a). (b) Representative immunofluorescence staining and statistical evaluation of CD8+ and CD8+CD45.2+ cells in lesional skin from HSV-1-infected recipient mice (n=3; original magnification × 400; bar = 50 μm; *P<0.05 versus WT →CD45.1+). (c) Skin lesion size in WT and K14-RANKL TG mice after depletion of CD8+ T cells before epicutaneous HSV-1 infection (n≥6). (d) Quantitative real-time PCR showing the UL54 (ICP27) expression in lesional skin at day 8 post infection (p.i., n=4; *P<0.05 versus IgG control). All the data are shown as mean±SEM. Journal of Investigative Dermatology 2015 135, 2676-2687DOI: (10.1038/jid.2015.225) Copyright © 2015 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 4 Langerhans cell (LC) activation via RANK–RANKL signaling is crucially involved in the induction of cytotoxic T lymphocytes. (a) Representative immunofluorescence staining and statistical evaluation of apoptotic LCs in lesional skin from HSV-1-infected mice at day 4 post infection (n=7; original magnification × 400; bar = 10 μm; double-positive cells are highlighted by arrows). (b) Statistical evaluation of annexin V expression in MHC-II+CD207+ LCs from lesional skin at day 4 post infection (n≥5) as assessed by flow cytometry. (c) Representative histogram overlays showing the expression of CD80, CD86, and IL-12 in LCs from HSV-1-infected skin at day 4 post infection (cells are gated for MHC-II+CD207+). (d–g) LC ablation in K14-RANKL TG × Langerin-DTR and Langerin-DTR mice before HSV-1 infection. Skin lesion size (d, n≥7), quantitative PCR of UL54 expression in lesional skin (e, n=3), representative dot plots showing the expression of cytolytic markers in CD8+ T cells (f), as well as statistical evaluation of BrdU incorporation in CD8+ T cells from regional lymph nodes at day 8 post infection (g, n=4). All the data are shown as mean±SEM; *P<0.05 versus Langerin-DTR mice. Journal of Investigative Dermatology 2015 135, 2676-2687DOI: (10.1038/jid.2015.225) Copyright © 2015 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 5 RANK–RANKL-activated Langerhans cells migrating from lesional skin to regional lymph nodes upregulate TLR3 in lymph node–resident CD8α+ DCs. (a) Statistical evaluation of CD207+ LCs and CD103+ dermal DCs that emigrated from infected skin in regional lymph nodes at day 2 post infection (n≥5; cells are gated for MHC-II+CD326+). (b) Expression of TLR3 in lymph node–resident CD8α+ DCs from HSV-1-infected mice at day 2 post infection; one representative dot plot and the statistical evaluation from n=6 mice in each group are shown. (c) Representative immunofluorescence staining of DCs in lymph nodes draining infected skin using antibodies to MHC-II and TLR3 at day 2 post infection (original magnification × 400; bar = 50 μm; double-positive cells are highlighted by arrows). (d) Representative flow cytometry of TLR3 expression in CD8α+ DCs from regional lymph nodes of Langerin-DTR and K14-RANKL TG × Langerin-DTR mice after LC ablation at day 2 post infection. (e) Skin lesion size in TLR3-/- and K14-RANKL TG × TLR3-/- mice after HSV-1 infection (n=4). (f) Statistical evaluation of cytolytic markers in CD8+ T cells from lymph nodes draining infected skin at day 8 post infection (n=3) as quantified by flow cytometry. All the data are shown as mean±SEM; *P<0.05 versus HSV-1-infected WT or TL3R-/- controls, respectively. Journal of Investigative Dermatology 2015 135, 2676-2687DOI: (10.1038/jid.2015.225) Copyright © 2015 The Society for Investigative Dermatology, Inc Terms and Conditions

Figure 6 Soluble RANKL protects WT mice from HSV-1 infection by inducing cytotoxic T lymphocytes. (a) Skin lesion size after subcutaneous injection of recombinant RANKL protein into lesional skin at day 2 and day 4 post HSV-1 infection (n≥5). (b) Quantitative real-time PCR of UL54 mRNA expression in lesional skin at day 8 post infection (n=3). (c and d) Representative histogram overlays (left) and statistical evaluation (right) of annexin V+ LCs in lesional skin (c) or TLR3-expressing CD8α+ DC in lymph nodes draining infected skin from K14-RANKL TG, RANKL-injected, or phosphate-buffered saline-treated WT mice at day 2 post infection (d). Data from n=3 mice per group are shown in c and d, and TLR3 staining was performed after cell permeabilization. (e–g) Total numbers of CD8+ T cells in lymph nodes draining infected skin (e) as well as representative dot plots or histogram overlays showing CD8+ T cells expressing cytolytic markers as quantified by flow cytometry (f and g). Statistical evaluation of NKG2D+CD8+ T cells from n=3 mice per group are shown (g, right). All the data are presented as mean±SEM; *P<0.05 versus phosphate-buffered saline-treated WT controls. Journal of Investigative Dermatology 2015 135, 2676-2687DOI: (10.1038/jid.2015.225) Copyright © 2015 The Society for Investigative Dermatology, Inc Terms and Conditions