Fig 1.
Interaction of LCs and Epi-cDC2s with HSV-1 in the epidermis of inner foreskin explants.
Inner foreskin explants were topically infected with HSV-1 GFP or mock infected for 24 h, snap-frozen and cryo-sectioned prior to immunofluorescent labelling with anti-Langerin (blue) and anti-CD11c (red) antibodies and DAPI staining. (A) Image from a 9 y.o. donor shows LCs (Langerin+ CD11c-, blue cells) and Epi-cDC2s (CD11c+ Langerin+/-, red and red/blue dual labelled cells) in the epidermis of an HSV-1 GFP infected (green) foreskin explant at 20x magnification. Arrows indicate Epi-cDC2s (white arrows) and LCs (yellow arrows) interacting with HSV-1 GFP infected keratinocytes or that appear to be HSV-1 GFP+. This image is representative of the observations in three donors. Scale bar = 50 μm. (B) The same representative donor re-imaged at 60x magnification to generate Z-stacks with 0.1 μm Z spacing. The left panel shows the maximum intensity projection of 49 slices of a region of interest, indicated by the white box in (A). The right panel shows the orthogonal view of the XZ and YZ planes of an individual CD11c+ cell containing HSV-1 GFP+ labelling. Scale bar = 10 μm. (C) A 5 y.o. donor imaged as in (B) with the maximum intensity projection of 50 slices shown on the left and the orthogonal view of a CD11c+ cell containing HSV-1 GFP+ labelling on the right. Scale bar = 10 μm.
Fig 2.
Epi-cDC2s are more extensively infected with HSV-1 than LCs.
A mixed population of epidermal MNPs was isolated from human abdominal epidermis and inoculated with HSV-1 GFP (MOI of 10) or mock treated for 1 h, then washed and incubated for a total infection time of 18 h (B & C) or 6 and 18 h (D & E) in HaCaT-conditioned medium at 37°C. The cells were then washed with PBS and labelled with fixable LIVE/DEAD Near-IR followed by antibodies to HLA-DR, CD45, CD1a, CD11c, and Langerin, then analysed by flow cytometry. HSV-1 infection was detected using GFP expression. (A) Representative gating strategy to distinguish LCs and Epi-cDC2s. CD45+HLA-DR+ cells were gated on and separated into CD1a+CD11c- LCs and CD1alowCD11c+ Epi-cDC2s. Live cells from each subset were then assessed for HSV infection by GFP expression for which a representative donor is shown in (B). (C) The percentage of HSV-1 GFP+ cells at 18 h in LCs (purple) and Epi-cDC2s (orange) is shown in box and whisker plots, where the central bar is the median, the box limits represent the upper and lower quartiles, the whiskers are the minimum and maximum values, and each dot represents an individual donor. A two-tailed paired t-test was used to compare the percentage of HSV-1 GFP+ LCs and Epi-cDC2s, n = 17. (D) Representative histograms of GFP expression at 6 and 18 h in LCs and Epi-cDC2s from one donor. (E) The expression of HSV-1 GFP, shown as gMFI, and the percentage of HSV-1 GFP+ cells for LCs (purple dots) and Epi-cDC2s (orange dots) at 6 and 18 h, time points are paired by donor. Repeated measures two-way ANOVA with Bonferroni’s test for pairwise comparisons were used to compare the gMFI or the % HSV-1 GFP+ LCs and Epi-cDC2s at 6 vs. 18 h, and LCs vs. Epi-cDC2s at each time point, n = 4.
Fig 3.
LCs and Epi-cDC2s express ICP27 and DNA-replication dependent late protein gM.
A mixed population of epidermal MNPs was isolated from human abdominal epidermis and inoculated with HSV-1 GFP (MOI of 10) or mock treated for 1 h, then washed and incubated for a total infection time of 18 h in HaCaT-conditioned medium at 37°C. The cells were then washed with PBS and labelled for flow cytometry as in Fig 2. Additionally, cells were permeabilized then followed by serial intracellular labelling with (A) biotin-conjugated anti-ICP27 then Alexa Fluor-647 (AF647)-conjugated streptavidin, or (B) rabbit anti-HSV-1 gM then donkey-anti-rabbit (DAR) AF647. (A) The percentage of ICP27 AF647+ cells was analysed for each subset at 18 h, with gates set according to the mock condition. The graph shows box and whisker plots of the percentage of LCs (purple) and Epi-cDC2s (orange) expressing total ICP27, a combination of the ICP27+GFP- and ICP27+GFP+ quadrants, indicated by the red boxes. A two-tailed paired t-test was used to compare the percentage of ICP27+ LCs and Epi-cDC2s, n = 4. (B) The percentage of gM AF647+ cells was analysed for each subset at 18 h, with gates set according to the mock condition. The graph shows the percentage of LCs (purple) and Epi-cDC2s (orange) expressing total gM, a combination of the gM+GFP- and gM+GFP+ quadrants, indicated by the red boxes, n = 3.
Fig 4.
The expression of HSV-1(US9)-GFP and gM are significantly reduced by acyclovir, low level extracellular virus is detectable from both LCs and Epi-cDC2s.
(A & B) A mixed population of epidermal MNPs was isolated from human abdominal epidermis and inoculated with HSV-1 GFP (MOI of 10) or mock treated for 1 h, then washed and resuspended in HaCaT-conditioned medium. HSV-1 GFP inoculated cells were split into two tubes, acyclovir (500 μg/ mL) was added to one tube and all cells were incubated for a total infection time of 18 h at 37°C. The cells were then washed with PBS and labelled for flow cytometry as in Fig 2. Additionally, cells were permeabilized then followed by serial intracellular labelling with rabbit anti-HSV-1 gM then donkey-anti-rabbit AF647. The percentage of HSV-1 (US9)-GFP+ and gM AF647+ cells was analysed for each subset at 18 h, with gates set according to the mock condition as shown in the (A) representative plot. (B) HSV-1 infected cells were set as 100% infection and the data are shown as box and whisker plots of the percentage of maximum infection in LCs (purple) and Epi-cDC2s (orange). Repeated measures one-way ANOVA with Tukey’s test for multiple pairwise comparisons was used to compare the percentage of maximum infection of each subset in the presence of acyclovir, HSV-1 (US9)-GFP n = 6, gM n = 2. (C) Mixed epidermal MNPs were isolated from human abdominal epidermis and FACS sorting was performed to isolate each subset. LCs and Epi-cDC2s were separately inoculated with HSV-1 GFP (MOI of 10) for 1 h, then washed, resuspended at 7x103 cells per 100 μL and incubated for a total infection time of 24 h in HaCaT-conditioned medium at 37°C. Supernatants were collected and a fluorescent plaque assay was performed. Data shows the plaque counts of supernatants of HSV-1 infected LCs (purple) and Epi-cDC2s (orange) with PFU normalized to per million cells, n = 3.
Fig 5.
LCs and Epi-cDC2s both express HSV-1 entry receptors HVEM and nectin-1.
Epidermal MNPs were isolated from human abdominal or foreskin epidermis and labelled for flow cytometry as in Fig 2, as well as with antibodies to HVEM or nectin-1. The surface expression of HVEM and nectin-1 on LCs (purple histograms) and Epi-cDC2s (orange histograms) is shown in a representative donor. Mean receptor expression (gMFI) is graphed for LCs (purple) and Epi-cDC2s (orange). Circles indicate abdominal skin, triangles indicate foreskin. Two-tailed paired t-tests were used to compare the gMFI of HVEM or nectin-1 expression in LC and Epi-cDC2s, n = 3.
Fig 6.
HSV-1 entry into MDDCs is pH-dependent and endocytic.
MDDCs, HeLa and Vero cells were pre-treated with bafilomycin A (baf A) or monensin for 30 min at 37°C at the concentrations shown. HSV-1 (strain F) was added (MOI of 3) in the continued presence of the agents and allowed to bind for 1 h at 37°C. Cells were then washed and incubated in the presence of inhibitors at 37°C for a total infection time of 18 h. (A and B) Cells were labelled with anti-HSV-1 gC antibody (HSV-1 gC-FITC), then analysed by flow cytometry. (A) Representative histograms of gC expression in each cell type. (B) Percentage of HSV-1 gC-FITC+ cells with increasing concentrations of baf A or monensin for each of the three cell types at 18 h. HSV-1 infected cells with no inhibitor were set as 100% infection and the data are shown as the percentage of maximum infection. Two-way ANOVA with Tukey’s test for multiple pairwise comparisons was used to compare the percentage of maximum infection between cell types at each concentration of each inhibitor, n = 4 (HeLa n = 3). The ANOVA summary p values for variation in inhibitor concentration and cell type were p<0.0001 for both inhibitors. Significant pairwise p values: 50 nM baf A: HeLa vs. Vero p<0.0001, MDDC vs. Vero p<0.0001, MDDC vs. HeLa p = 0.0390, 100 nM baf A: HeLa vs. Vero p<0.0001, MDDC vs. Vero p<0.0001, 0.1 μM monensin: HeLa vs. Vero p<0.0001, MDDC vs. Vero p<0.0001, 1 μM monensin: HeLa vs. Vero p<0.0001, MDDC vs. Vero p<0.0001. (C) Following infection, MDDCs were washed, lysates were collected, and qPCR was performed to detect mRNA for the HSV-1 immediate early gene ICP4. Fold change relative to mock was calculated, the no inhibitor HSV-1 infected MDDCs were set as 100% infection and the data shows ICP4 expression as a percentage of maximum infection in baf A (left panel) or monensin (right panel) treated MDDCs. Repeated measures one-way ANOVA with Bonferroni’s test for pairwise comparisons were used to compare the percentage of maximum infection of each concentration of baf A or monensin to the no inhibitor (0 nM/μM) control, n = 3.
Fig 7.
HSV-1 entry into LCs is pH-dependent and langerin-dependent.
Isolated abdominal epidermal MNPs were pre-treated with 100 nM bafilomycin A (baf A) for 30 min at 37°C (A and B) or with 5 μg/mL anti-langerin antibody (C) then infected with HSV-1 GFP for 18 h and stained for flow cytometry as in Fig 2, to detect infection using GFP expression. (A) Representative plots of HSV-1 infection in the absence or presence of baf A. The percentage of HSV-1 GFP+ cells was measured with and without (B) baf A (n = 7) or (C) anti-langerin antibody (n = 5) at 18 h. HSV-1 infected cells were set as 100% infection and the data are shown as box and whisker plots of the percentage of maximum infection in LCs (purple) and Epi-cDC2s (orange). Repeated measures one-way ANOVA with Tukey’s test for multiple pairwise comparisons was used to compare the percentage of maximum infection of each subset in the presence of inhibitor.
Fig 8.
HSV-1 uptake into MDDCs and Epi-cDC2s is dependent on actin and cholesterol.
MDDCs or isolated abdominal epidermal MNPs were pre-treated for 30 min at 37°C with mβCD, cytochalasin D (cyto D), latrunculin A (lat A) or no inhibitor at the concentrations shown. HSV-1 GFP was added (MOI of 3 or 10 for MDDCs and epidermal MNPs respectively) in the presence of the inhibitor and allowed to bind for 1 h at 4°C. The cells were washed and incubated for a further 1 h at 37°C in the continued presence of the inhibitor, then trypsinised for 15 min at 37°C to remove surface bound virus. Cells were labelled for flow cytometry as in Fig 2. (A-C) The percentage of HSV-1 GFP+ MDDCs was measured by flow cytometry for (A) mβCD (n = 6) (B) cyto D (n = 6) or (C) lat A (n = 3) treated conditions. HSV-1 infected cells with no inhibitor were set as 100% uptake and the data are shown as box and whisker plots of the percentage of maximum uptake. Repeated measures one-way ANOVA with Tukey’s test for multiple pairwise comparisons was used to compare the percentage of maximum uptake of MDDCs in the presence of each inhibitor. (D) Representative donors showing the percentage of HSV-1 GFP+ cells in the presence of cyto D, lat A or mβCD compared to HSV-1 only in Epi-cDC2s. (E) HSV-1 infected Epi-cDC2s with no inhibitor were set as 100% uptake and the data are shown as box and whisker plots of the percentage of maximum uptake. One-way ANOVA with Tukey’s test for multiple pairwise comparisons was used to compare the percentage of maximum uptake of Epi-cDC2s in the presence of each inhibitor. HSV n = 8, mβCD n = 4, cyto D n = 6, lat A n = 5.
Fig 9.
HSV-1 induces apoptosis of both infected and bystander LCs and Epi-cDC2s.
Epidermal MNPs were isolated from human abdominal epidermis and inoculated with HSV-1 GFP (MOI of 10) or mock treated, then washed and incubated for a total infection time of 18 h. As a positive control, some cells were treated with 1 μM staurosporine for 18 h at 37°C. The cells were labelled for flow cytometry as in Fig 2, as well as with intracellular active Caspase-3 (Casp3) and LIVE/DEAD Near-IR for examining apoptosis. (A) Representative plots showing apoptosis analysis. In the HSV-1 infected conditions, LCs and Epi-cDC2s were first separated into HSV-1 GFP+ and Bystander (HSV-1 GFP-) populations including live and dead cells. Apoptosis was assessed in mock, HSV-1 GFP and staurosporine treated conditions by analyzing the distribution of cells over the four Casp3 vs. Live/Dead quadrants, with gates set according to the mock and staurosporine conditions. Live Casp3-: live cells, Live Casp3+: live apoptotic cells, Dead Casp3+: apoptotic cells, Dead Casp3-: dead cells. (B and C) Grouped analyses of the percentage of LCs and Epi-cDC2s (respectively) in each of the apoptosis quadrants in mock (grey), HSV-1 GFP+ (green) and Bystander (pink) populations shown as box and whisker plots. Repeated measures one-way ANOVA with Tukey’s test for multiple pairwise comparisons was used to compare the percentage of LCs (B) or Epi-cDC2s (C) in each apoptosis quadrant between mock, HSV-1 GFP+ and bystander populations, n = 4. (B) Live cells: Mock vs. HSV-1+ p = 0.0100, Live apoptotic cells: Mock vs. HSV-1+ p = 0.0140, Mock vs. Bystander p = 0.0245, HSV-1+ vs. Bystander p = 0.0006, Apoptotic cells: Mock vs. HSV-1+ p = 0.0100, HSV-1+ vs. Bystander p = 0.0296, Dead cells: Mock vs. HSV-1+ p = 0.0064, HSV-1+ vs. Bystander p = 0.0070. (C) Live cells: Mock vs. HSV-1+ p<0.0001, Mock vs. Bystander p<0.0001, Live apoptotic cells: Mock vs. Bystander p<0.0001, HSV-1+ vs. Bystander p<0.0001, Apoptotic cells: Mock vs. HSV-1+ p = 0.0070, HSV-1+ vs. Bystander p = 0.0158, Dead cells: Mock vs. HSV-1+ p = 0.0172, HSV-1+ vs. Bystander p = 0.0133.
Fig 10.
The HSV viral relay between epidermal and dermal MNPs and HSV entry pathways into them compared to other target cells.
(A) HSV enters skin where the stratum corneum is damaged or absent and infects LCs and Epi-cDC2s, located in the epidermis. Upon HSV infection both cell types undergo apoptosis. HSV infected LCs migrate to the dermis, where they cluster with and are taken up by dermal MNPs, which may go on to present the HSV antigens to T cells. Whether HSV infected apoptotic Epi-cDC2s also migrate to the dermis and are taken up by dermal MNPs, thereby contributing to the HSV viral relay, is being investigated. (B) HSV enters cells via different pathways depending on the host cell type. HSV enters Vero and neuronal cells by neutral fusion at the plasma membrane. A pH-independent endocytic pathway is utilised for C10 mouse melanoma cells and Epi-cDC2s, whereas a pH-dependent endocytic pathway is utilised in epithelial cells such as keratinocytes, HeLa cells and CHO-nectin 1 cells, as well as predominantly in human MDDCs and LCs.