Endogenous retroviruses promote homeostatic and inflammatory responses to the microbiota

Abstract

The microbiota plays a fundamental role in regulating host immunity. However, the processes involved in the initiation and regulation of immunity to the microbiota remain largely unknown. Here, we show that the skin microbiota promotes the discrete expression of defined endogenous retroviruses (ERVs). Keratinocyte-intrinsic responses to ERVs depended on cyclic GMP-AMP synthase (cGAS)/stimulator of interferon genes protein (STING) signaling and promoted the induction of commensal-specific T cells. Inhibition of ERV reverse transcription significantly impacted these responses, resulting in impaired immunity to the microbiota and its associated tissue repair function. Conversely, a lipid-enriched diet primed the skin for heightened ERV- expression in response to commensal colonization, leading to increased immune responses and tissue inflammation. Together, our results support the idea that the host may have co-opted its endogenous virome as a means to communicate with the exogenous microbiota, resulting in a multi-kingdom dialog that controls both tissue homeostasis and inflammation.

Keywords: STING; Staphylococcus epidermidis; T cells; antiretroviral; endogenous retrovirus; high fat diet; keratinocytes; microbiota; skin immunity; tissue repair.

Graphical abstract

Figure 1

S. epidermidis colonization promotes an antiviral program, and responses to S. epidermidis are type I IFN dependent (A) Frequency and absolute number of IL-17A⁺ (Tc17) or IFN-γ⁺ (Tc1) CD8⁺ T cells, IL-17A⁺ (Th17) or IFN-γ⁺ (Th1) CD4⁺ T cells, and IL-17A⁺ γδTCR or MAIT cells in the skin of unassociated or S. epidermidis associated WT mice 14 days post-association. (B) Experimental schematic. (C) Volcano plot for log₂ fold change in gene expression in CD49f⁺Sca-1⁺ keratinocytes isolated from S. epidermidis associated versus unassociated mice. Representative genes associated with chemotaxis (red), antigen presentation (green), and keratinization (blue) are highlighted. (D) Gene ontology (GO) assignments of top 10 GO terms that were enriched in Sca-1⁺CD49f⁺ keratinocytes from S. epidermidis associated versus unassociated WT mice. Upregulated genes are shown for specific pathways of interest. (E) Absolute number of indicated cell subsets in the skin of unassociated or S. epidermidis associated WT mice treated with anti-IFNAR1 neutralizing antibody (anti-IFNAR1) or isotype control. (A and E) ^∗p < 0.05, ^∗∗p <, 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001 as calculated using a Student’s t test (A) or one-way ANOVA with Holm-Šidák multiple comparison test (E). Data are represented as mean ± SEM. In (A), numbers correspond to the frequency of the gated populations ± SEM. Data are representative of four (A) or two (C–E) independent experiments using 4 to 5 mice per group (with the dots in A and E each representing individual mice). See also Figure S1.

Figure S1

Type I IFN is required for S. epidermidis induced T cells responses in the skin, related to Figure 1 (A) Flow cytometry gating strategy used to identify various T cell subsets and MAIT cells. Tc1 and Tc17 cells are defined as live CD45⁺ Thy1.2⁺ TCRβ⁺ γδTCR^– CD8b⁺ IFN-γ⁺ and live CD45⁺ Thy1.2⁺ TCRβ⁺ γδTCR^– CD8b⁺ IL-17A⁺, respectively. Th1 and Th17 cells are defined as live CD45⁺ Thy1.2⁺ TCRβ⁺ γδTCR^– CD4⁺ Foxp3^– IFN-γ⁺and live CD45⁺ Thy1.2⁺ TCRβ⁺ γδTCR^– CD4⁺ Foxp3^– IL-17A⁺, respectively. γδTCR^low cells are defined as live CD45⁺ Thy1.2⁺ TCRβ^– γδTCR^low. MAIT cells are defined as live CD45⁺ Thy1.2⁺ TCRβ⁺ MR1-tetramer⁺. (B) Flow cytometry gating strategy use to identify interfollicular keratinocytes (DAPI^– CD45^– CD31^– CD34^– CD49f⁺ Sca-1⁺) in single cells suspensions isolated from the epidermis of mouse ear pinnae 7 days post daily S. epidermidis association. (C) Principal component analysis of global gene expression of RNA-seq performed on interfollicular keratinocytes sorted from the epidermis at day 7 post S. epidermidis association. Ellipses denote 95% confidence intervals of the mean. Keratinocytes were isolated from ear pinnae. (D) Absolute number of CD8⁺ T cells, CD4⁺ T cells, γδTCR^low and MAIT cells, and frequency of IL-17A⁺ or IFN-γ⁺ CD8⁺ or CD4⁺ T cells, as well as IL-17A⁺γδTCR^low and MAIT cells in the ear pinnae of unassociated mice or S. epidermidis-associated mice treated with anti-IFNAR1 or isotype control antibodies. Cells were stimulated with PMA and ionomycin. ^∗P ˂ 0.05, ^∗∗P ˂ 0.01, ^∗∗∗P ˂ 0.001, ^∗∗∗∗P ˂ 0.0001 as calculated with one-way ANOVA with Holm-Šidák multiple comparison test. Data are represented as mean ± SEM. Data are representative of two independent experiments using 4-5 mice per group (with the symbols in (C) and (D) each representing an individual mouse).

Figure 2

S. epidermidis promotes retroelements expression by keratinocytes (A) Heatmap displaying fold change of differentially expressed retroelement loci from RNA sequencing of Sca-1⁺CD49f⁺ keratinocytes from unassociated (unassoc.) or S. epidermidis (S. epi) associated mice at day 7 post-association. (B) MLV SU expression detected by flow cytometry on the surface of Sca-1⁺CD49f⁺ keratinocytes from unassociated or S. epi associated mice at day 7 post-association. Integrated MFI (iMFI) of MLV SU expression is shown. (C) iMFI of MLV SU expression by Sca-1⁺CD49f⁺ keratinocytes from unassoc. or S. epi associated WT and Tlr2^−/− mice at day 7 post-association. (D and E) WT mice were associated for 7 days with wild type, ΔtagO, or Δlgt S. epi or left unassociated. (D) Transcript levels of the indicated ERVs measured in the ear pinnae by qRT-PCR. ERVs expression was normalized to Gapdh mRNA levels in the same sample. (E) iMFI of MLV SU expression by Sca-1⁺CD49f⁺ keratinocytes. (F) WT mice were treated with vehicle control (vehicle) or antiretroviral by oral gavage, beginning at 1 week before S. epi association for a total of 3 weeks. (G) Representative confocal microscopy images of whole-mount ear pinnae of WT mice treated with vehicle (S. epi + vehicle) or antiretroviral (S. epi + antiretroviral) stained for CD49f (blue), CD4 (yellow), and CD8α (red) at day 14 post-association. Scale bars represent 100 μm. (H) Absolute number of indicated T cell subsets at 14 days post-association from the skin of WT mice treated with either vehicle (S. epi + vehicle) or antiretroviral (S. epi + antiretroviral). (I) Frequency and absolute number of f-MIIINA:H2-M3 tetramer-positive CD8⁺ T cells from ear pinnae of unassoc. or S. epi associated mice treated with vehicle or antiretroviral. (J) Transcript levels of the indicated ERVs measured by qRT-PCR in the ear pinnae of mice after daily association with S. epi, S. aureus, or S. xylosus for 7 days. ERVs expression levels were normalized to Gapdh mRNA levels in the same sample. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001 as calculated with a Student’s t test (B), one-way ANOVA (E, H, and I), or two-way ANOVA (C, D, and J) with Holm-Šidák multiple comparison test. Data are represented as mean ± SEM. Data are representative of two (A, C–E, G, I, and J) or three (B and H) independent experiments using 4 to 5 mice per group. Each dot represents an individual mouse. See also Figure S2.

Figure S2

Endogenous retrovirus activity promotes the accumulation of T cells in the skin following S. epidermidis colonization, related to Figure 2 (A) Volcano plot of expressed loci from sorted Sca-1⁺CD49f⁺ keratinocytes from unassociated or S. epidermidis-associated at day 7 post-association. Underlined denotes locus containing an active reverse transcriptase. (B) Heatmap showing transcript levels of the indicated ERVs measured by qRT-PCR of mRNA isolated from ear pinnae from unassociated or daily S. epidermidis-associated mice at day 7 post association. Values were normalized to Gapdh expression in the same sample. (C) Schematic showing encoded regions of LINE-1 elements in the C57BL/6J (WT) mice genome. Regions highlighted in yellow represent the nucleotide regions encoded in the given locus. (D) Calculated iMFI of MLV SU expression detected by flow cytometry on the surface of CD49f⁺Sca-1⁺ primary keratinocytes isolated from neonatal mice stimulated with TLR agonists. (E-G) WT mice were treated daily with either vehicle control (S. epi + vehicle) or antiretroviral (S. epi + antiretroviral) beginning at 1 week before S. epidermidis association for a total of 3 weeks. Two weeks post association, T cell populations were evaluated by flow cytometry. (E and F) Absolute number of CD8⁺ T cells, CD4⁺ T cells, γδTCR^low T cells, IL-17A⁺ (Tc17) or IFN-γ⁺ (Tc1) CD8⁺ T cells, IL-17A⁺ (Th17) or IFN-γ⁺ (Th1) CD4⁺ T cells, or IL-17A⁺ γδTCR^low T cells (E) in the ear pinnae or (F) the spleen. (G) Absolute number of bead-enriched f-MIIINA:H2-M3 tetramer positive CD8β⁺ T cells in the spleen. Data are represented as mean ± SEM. Each dot represents an individual mouse. (H) WT mice were treated daily with either vehicle control (S. epi + vehicle) or antiretroviral (S. epi + antiretroviral) beginning 1 week before daily S. epidermidis association for a total of 2 weeks. Seven days post-association, RNA was purified from sorted Sca-1⁺CD49f⁺ keratinocytes and sequenced. The expression of the S. epidermidis-induced ERV, Chr5 23.7M is highlighted in red. (I) Enumeration of S. epidermidis colony-forming units (CFU) at day 14 post association in WT mice treated daily with either vehicle control (S. epi + vehicle) or antiretroviral (S. epi + antiretroviral) beginning at 1 week before S. epidermidis association. (J) Growth curve (OD600) of S. epidermidis grown in media treated with vehicle, tenofovir and emtricitabine alone or with a combination of tenofovir and emtricitabine at different concentrations. ^∗P ˂ 0.05, ^∗∗P ˂ 0.01, ^∗∗∗P ˂ 0.001, ^∗∗∗∗P ˂ 0.0001 as calculated with one-way ANOVA with Holm-Šidák multiple comparison test (C, E-H). Data are representative of two (A-D, H-J) or three (E) and (F) independent experiments using 4-5 mice per group (A), (B), E-I). Graphs in (C) and (J) represent the average of a technical triplicate.

Figure S3

Single-cell RNA-seq analysis of interfollicular keratinocytes, related to Figure 3 WT mice were treated daily with either vehicle control (S. epi + vehicle) or antiretroviral (S. epi + antiretroviral) beginning at 1 week before S. epidermidis association for a total of 3 weeks. At day 14 post association, interfollicular keratinocytes were sorted by FACS from the epidermis of ear pinnae and analyzed by single-cell RNA-seq. (A) UMAP projection plots showing the expression profiles of keratinocytes. Colors represent cells clustered together based on similarity of global gene expression. Cells identity was assigned based on the expression level of specific genes: Interfollicular epidermal basal cells: Krt14^hiMt2^hiMt1^hiPostn^low; Interfollicular epidermal differentiated cells subcluster 1 (D1): Krt14^midKrt10^midMt4^hi; Interfollicular epidermal differentiated cells subcluster 2 (D2): Krt14^lowKrt10^hiSbsn^hi;Interfollicular epidermal keratinized cells subcluster 2 (K2): Krt14^lowKrt10^hiLor^hiFlg2^hi; Upper hair follicle: Krt79^hiKrt17^hiDefb^hi; and Infundibular basal cells: Krt14^midMt2^midMt1^midPostn⁺. (B) Schematic illustrating the anatomical localization of distinct populations of keratinocytes. (C) UMAP projection plots depicting expression of the indicated genes that are involved in antigen presentation on MHC class II. Data are representative of one experiment using 5 mice per group. (D) MLV SU expression (shown as integrated MFI) detected by flow cytometry on the surface of Sca-1⁺CD49f⁺MHCII⁺ and Sca-1⁺CD49f⁺MHCII^– keratinocytes from unassociated or S. epidermidis-associated mice at day 7 post-association. ^∗∗∗∗P ˂ 0.0001 as calculated with one-way ANOVA with Holm-Šidák multiple comparison test (D). Data are representative of one (A and C) or three (D) independent experiments using 5 mice per group.

Figure 3

Antiretroviral treatment impairs keratinocyte responses to S. epi and tissue repair WT mice were treated with vehicle control (S. epi + vehicle) or antiretroviral (S. epi + antiretroviral) beginning 1 week before S. epi association for a total of 3 weeks. 2 weeks post-association, keratinocytes responses were analyzed. (A) scRNA-seq data from sorted keratinocytes showing the expression of Postn, B2m, and Ifitm3 in individual keratinocytes for the clusters 0, 1 (interfollicular epidermal basal cells), and 8 (infundibular basal cells). (B) Uniform manifold approximation and projection (UMAP) plot displaying the distribution of the differentially abundant keratinocyte populations. Genes defining cluster 12 (MHCII⁺ keratinocytes) are denoted. (C) Flow cytometry analysis of MHC-II expression and Ki-67 co-expression in keratinocytes at 14 days post-association from S. epi associated WT mice treated with either vehicle (vehicle + S. epi) or antiretroviral (S. epi + antiretroviral). Plots are gated on live CD45⁻CD31⁻CD34⁻Sca-1⁺. ^∗ in the flow plots indicates significant difference between vehicle and antiretroviral treated group. (D and E) WT mice were treated with antiretroviral and associated with S. epi, followed by a back skin punch biopsy 12 days post-association. (D) Representative immunofluorescence images of wounds at day 5 after punch biopsy. Tissue sections are stained for basal keratinocytes (keratin 14 in red) and co-stained with DAPI (blue). Demarcated white dashed lines represent the epidermal tongue length during re-epithelization of the wounds. Scale bars represent 1,000 μm. (E) Quantification of the epidermal tongue length at day 5 post-wounding, with each dot representing the measured length of an individual epidermal tongue. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001 as calculated with one-way ANOVA with Holm-Šidák multiple comparison test. Data are represented as mean ± SEM. Data are representative of one (A and B), two (D and E), or three (C) independent experiments using 4 to 5 mice per group. Each dot represents an individual mouse. See also Figure S3.

Figure S4

cGAS-STING signaling promotes T cells accumulation in the skin in response to S. epidermidis colonization, related to Figure 4 (A) Frequency of CD8⁺ and CD4⁺ (left) T cells, percentage change of the absolute numbers of CD8⁺, CD4⁺, γδTCR^low T cells (middle) and absolute number of IFN-γ⁺ CD8⁺ T cells (Tc1) (right) in the ear pinnae of naive (unassociated) WT, cGas^–/– and Sting^–/– mice. (B) Absolute number of CD8⁺, CD4⁺, γδTCR^low T cells and/or MAIT cells and frequency of IL-17A⁺ or IFN-γ⁺ CD8⁺ or CD4⁺ T cells and IL-17A⁺ γδTCR^low or MAIT cells in the ear pinnae from WT, cGas^–/– and Sting^–/–mice associated with S. epidermidis (day 14) or left unassociated. (C) S. epidermidis CFU enumeration at 14 days post-association in WT and Sting^–/– mice topically associated with S. epidermidis (S. epi) or left unassociated. (D) Expression of Chr5 23.7 locus determined by RNASeq in Sca-1⁺CD49f⁺ keratinocytes purified from unassociated or daily S. epi-associated Krt14^cre-Sting^flox/flox and Krt14^cre+Sting^flox/flox mice at day 7 post-association. ^∗∗ FDR < 0.01, ^∗∗∗∗ FDR < 0.001. (E) Absolute number of CD8⁺, CD4⁺, γδTCR^low T cells and/or MAIT cells and frequency of IL-17A⁺ or IFN-γ⁺ CD8⁺ or CD4⁺ T cells and IL-17A⁺ γδTCR^low or MAIT cells in the ear pinnae from Krt14^cre-Sting^flox/flox and Krt14^cre+Sting^flox/flox mice associated with S. epidermidis (day 14) or left unassociated. (F) Proposed model for ERV control of T cell responses to the microbiota: 1- Discrete sensing of microbiota by keratinocytes (via, in part, TLR2-specific ligands) promotes the expression of defined ERVs; 2- Reverse transcription of ERVs leads to cytosolic accumulation of ERV-derived cDNAs that are sensed by the cGAS/STING pathway. Resulting activation of keratinocytes is associated with an antiviral program and type I IFN production; 3- Discrete keratinocyte “hot spots” could promote an environment favorable to the capture of microbiota-derived antigens by DCs and subsequent migration of these cells to the lymph node; 4- commensal-specific T cells migrate back to the skin where their accumulation and function could be promoted by ERV-activated keratinocytes. In the context of non-classical T cells such as MAIT cells, local responses to ERVs may be sufficient to control their local proliferation and accumulation in the skin. The result of this sequence of events is the accumulation of a network of commensal specific T cells able to broadly promote tissue physiology including tissue repair. All cells were stimulated with PMA and ionomycin. Each dot represents an individual mouse. ^∗P < 0.05, ^∗∗P < 0.01, ^∗∗∗P < 0.001, ^∗∗∗∗P < 0.0001 as calculated with two-way ANOVA with Holm-Šidák multiple comparison test. Data are represented as mean ± SEM. Data are representative of two independent experiments using 4-6 mice per group.

Figure 4

cGAS-STING signaling pathway is required for S. epidermidis induced T cells response WT, cGas^−/−, Sting^−/−, Krt14^Cre-Sting^flox/flox, and Krt14^Cre+Sting^flox/flox mice were topically associated with S. epi or left unassociated. Two weeks post-association, T cell populations were evaluated by flow cytometry. Frequency and/or absolute number of indicated lymphocyte subsets in the ear pinnae from (A) WT, cGas^−/−, and Sting^−/− or (B) Krt14^Cre-Sting^flox/flox and Krt14^Cre+Sting^flox/flox mice. Each dot represents an individual mouse. ^∗ in the flow plots indicates significant difference compared to WT control mice. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001 as calculated with two-way ANOVA with Holm-Šidák multiple comparison test. Data are represented as mean ± SEM. In (A), numbers correspond to the frequencies of gated populations ± SEM. Data are representative of two independent experiments using 4–6 mice per group. See also Figures S4 and S5.

Figure S5

S. epidermidis induced T cells accumulation in the skin is independent of STING signaling in CD11c⁺ cells, related to Figure 4 Frequency and absolute number of CD8⁺ T cells, CD4⁺ T cells, γδTCR^low T cells, IL-17A⁺ (Tc17) or IFN-γ⁺ (Tc1) CD8⁺ T cells, IL-17A⁺ (Th17) or IFN-γ⁺ (Th1) CD4⁺ T cells and IL-17A⁺γδTCR^low cells in the ear pinnae of Cd11c^cre-Sting^flox/flox and Cd11c^cre-Sting^flox/flox mice two weeks post-association with S. epidermidis. Data are represented as mean ± SEM. Each dot represents an individual mouse.

Figure 5

S. epi promotes skin inflammation and heightened retroelement expression in the context of high-fat diet (A) WT mice fed either a control (Ctrl) or a high-fat (HF) diet for 6 weeks were topically associated with S. epi or left unassociated. (B) Ear thickness measurement reported as the change in ear-skin thickness relative to baseline at day 0 (first day of S. epi association). (C) Representative image of hematoxylin-and-eosin staining of the skin from mice fed control diet or HF diet at day 14 after the first topical association with S. epi. Scale bars represent 300 μm or 50 μm (zoom in). (D) Absolute number of indicated lymphocyte subsets from ear pinnae of mice fed control diet or HF diet at day 14 after the first topical association. Each dot represents an individual mouse. (E) GO assignments of top 11 GO terms enriched in Sca-1⁺CD49f⁺ keratinocytes isolated from S. epidermidis associated versus unassoc. WT mice fed a HF diet at day 7 post-association. Upregulated genes related to GO terms keratinization (blue) and antigen presentation via MHC-II (orange) are shown. (F) Volcano plots of expressed retroelement loci from Sca-1⁺CD49f⁺ keratinocytes purified from the epidermis of unassoc. or S. epi associated WT mice fed a control or a HF diet, 7 days post-association. Underlined loci highlight retroelements with active reverse transcriptases. (G) RNA-seq expression levels of gene from differentially expressed ERV loci in sorted keratinocytes from unassoc. or S. epi associated mice fed control or HF diet at 7 days post-association. Multiple biological replicates are shown per condition. (H) Retroelement expression analyzed from previously published clinical cohort in which the transcriptome of normal skin was compared to psoriatic skin within the same patient. Retroelement families enriched in psoriatic lesions are specifically highlighted. ^∗p < 0.05, ^∗∗p< 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001 as calculated with two-way ANOVA with Holm-Šidák multiple comparison test (B and D). Data are represented as mean ± SEM. Data are representative of two (C and E–G) or three (B and D) independent experiments using 4 to 5 mice per group or 54 paired samples (H) from 27 psoriasis patients. See also Figure S6.

Figure S6

High-fat diet induces heightened ERVs expression and T cells responses following S. epidermidis colonization, related to Figure 5 WT mice were fed either a control (Ctrl) or a high-fat (HF) diet for 6 weeks and then topically associated with S. epidermidis (S. epi) or left unassociated. (A) Body weight measurement (g ± SEM) in mice after 6 weeks of diet regimen. (B) Representative image of hematoxylin-and-eosin staining of the ear pinnae from mice after 8 weeks of diet regimen. Scale bars = 300 μm or 50 μm (zoom in). (C) Frequency and absolute number of CD49f⁺Ki-67⁺ keratinocytes in the ear pinnae of unassociated (Unassoc.) and S. epidermidis (S. epi)-associated mice fed a control (Ctrl) or high-fat (HF) diet at day 14 post association. (D) Flow cytometry analysis of MHC-II expression and absolute number of CD49f⁺MHC-II⁺ and CD49f⁺MHC-II⁺Ki-67⁺ keratinocytes from unassociated (Unassoc.) or S. epidermidis (S. epi)-associated mice fed a control (Ctrl) or high-fat (HF) diet at 14 days post association. Plots in (C) and (D) were gated on live CD45^–CD31^–CD34^–Sca-1⁺ cells. ^∗ in the flow plot indicates significant difference between unassociated and S. epi group. (E) Frequency of IL-17A⁺ or IFN-γ⁺ CD8⁺ or CD4⁺ T cells and IL-17A⁺ γδTCR^low or MAIT cells in the ear pinnae from unassociated and S. epidermidis-associated mice fed a control (Ctrl) or a high-fat (HD) diet. All cells were stimulated with PMA and ionomycin. (F) Gene ontology assignments of top 3 or top 2 GO terms that were enriched in Sca-1⁺CD49f⁺ keratinocytes from WT mice fed a high-fat diet versus control diet and from S. epidermidis-associated (day 7) WT fed a high-fat diet versus control diet, respectively. (G) Heatmap showing transcript levels of the indicated ERVs measured by qRT-PCR of mRNA isolated from ear pinnae from unassociated or S. epidermidis-associated mice fed a high-fat (HF) diet at day 7 post association. Values were normalized to Gapdh expression in the same sample. (H) MLV SU expression detected by flow cytometry on the surface of Sca-1⁺CD49f⁺ keratinocytes from unassociated (Unassoc.) or S. epidermidis-associated mice fed a high-fat diet at day 7 post association. (I and J) WT mice were treated with imiquimod (IMQ) cream for 5 consecutive days or not (control). (I) Differential ERV levels analyzed by RNaseq in the ear pinnae 5 days after the beginning of IMQ treatment. Data was reanalyzed from our published study (Hurabielle et al., 2020). ^∗∗∗∗ FDR ˂ 0.0001. (J) MLV SU expression detected by flow cytometry on the surface of Sca-1⁺CD49f⁺ keratinocytes from the ear pinnae 5 days after the beginning of IMQ treatment. ^∗P < 0.05, ^∗∗P < 0.01, ^∗∗∗P < 0.001, ^∗∗∗∗P < 0.0001 as calculated with Student’s t test (A), (H) and (J) or two-way ANOVA with Holm-Šidák multiple comparison test (C), (D) and (E). Data are represented as mean ± SEM. Each dot (A), C-E, H-J) represents an individual mouse. Data are representative of two (B), (F), (G), (H), (I) and (J) or three (A), (C), (D) and (E) independent experiments using 4-5 mice per group.

Figure 6

STING signaling and retroelement activity contribute to microbiota-induced inflammatory responses (A–D) WT, cGas^−/−, Sting^−/−, Ifnar1^−/−, Krt14^cre-Sting^flox/flox, and Krt14^cre+Sting^flox/flox mice fed a HF diet were topically associated with S. epi or left unassoc. (A and C) Ear thickness measurement reported as the change relative to baseline at day 0. (B and D) Frequency and absolute number of indicated lymphocyte subsets. (E) WT mice on HF diet were treated daily for 8 weeks with vehicle control (vehicle) or antiretrovirals starting at 2 weeks after the beginning of the HF-diet regimen. At 6 weeks post-HF diet, mice were associated or not with S. epi. (F) Ear thickness measurement. (G) Frequency of indicated lymphocyte subsets at 14 days post-association. Each dot represents an individual mouse. ^∗ in the flow plots indicates significant difference compared to WT (B), Krt14^Cre-Sting^flox/flox (D), and vehicle (G) S. epi associated conditions. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001 as calculated with two-way ANOVA with Holm-Šidák multiple comparison test (A, C, and F) or one-way ANOVA with Holm-Šidák multiple comparison test (B, D, and G). Data are represented as mean ± SEM. Data are representative of two independent experiments using 3–7 mice per group. See also Figure S7.

Figure S7

S. epidermidis induced aberrant skin inflammation in the context of high-fat diet is dependent on ERVs/cGas/Sting/IFN axis, related to Figure 6 (A-B) Absolute number of CD8⁺ T cells, CD4⁺ T cells, γδTCR^lowT cells and MAIT cells and frequency of IL-17A⁺ or IFN-γ⁺ CD4⁺ T cells and IL-17A⁺ γδTCR^low or MAIT cells in the ear pinnae from S. epidermidis-associated WT, cGas^–/–, Sting^–/– and Ifnar1^–/– mice fed a high-fat diet (A) or from unassociated or S. epidermidis-associated Krt14^cre-Sting^flox/flox and Krt14^cre+Sting^flox/flox mice fed a high-fat diet (B). (C) WT mice fed a high-fat diet (HF) were treated daily with either vehicle control (vehicle) or a combination of tenofovir disoproxil fumarate and emtricitabine (antiretroviral), beginning at 2 weeks post HF diet for a total of 8 weeks. At 6 weeks post HF diet, mice were topically associated with S. epidermidis (S. epi) or left unassociated. Absolute number of CD8⁺ T cells, CD4⁺ T cells and γδTCR^low T cells and frequencies of IL-17A⁺ or IFN-γ⁺ CD4⁺ T cells and IL-17A⁺ γδTCR^low cells. All cells were stimulated with PMA and ionomycin. Data are represented as mean ± SEM. Each dot represents an individual mouse. ^∗P ˂ 0.05, ^∗∗P < 0.01, ^∗∗∗P < 0.001, ^∗∗∗∗P < 0.0001 as calculated with one-way ANOVA (A) and (C) or two-way ANOVA with Holm-Šidák multiple comparison test (B). Data are representative of two independent experiments using 3-7 mice per group.