Triggering mouth-resident antiviral CD8+ T cells potentiates experimental periodontitis

Abstract

Emerging evidence indicates that gingival-resident helper CD4⁺ T cells are major drivers of periodontal inflammation in response to commensal and pathogenic oral microorganisms. Whether tissue-resident memory CD8⁺ T cells (T_RM), which principally safeguard against viruses and cancer but also drive certain autoimmune and inflammatory conditions, impact periodontitis progression and severity remain unknown. We asked whether local reactivation of oral CD8⁺ T_RM of a defined antigen specificity could exacerbate ligature-induced periodontitis (LIP), a well-established model of periodontal disease in mice. Topical application of virus-mimicking peptides to the oral mucosa concurrent with LIP 1) intensified alveolar bone loss, 2) amplified gingival and cervical lymph node inflammation, and 3) stimulated gingival transcriptional changes in genes related to innate immune sensing and cell-mediated cytotoxicity. Therapeutic depletion of CD103-expressing oral CD8⁺ T_RM in advance of LIP prevented exacerbation of disease. These observations provide evidence that oral CD103⁺ CD8⁺ T_RM have the potential to participate in gingival inflammation, alveolar bone loss, and periodontitis.

Keywords: Bone loss; Inflammation; Periodontitis; T cell; Virus.

Fig. 1.. Oral CD8⁺ T_RM reactivation aggravates experimental periodontitis.

(A) VPEP model and experimental design. (B) Enumeration of gingival OT-I cells, CD4⁺ T cells, and PMNs seven days post-LIP in mice orally swabbed with gp33 or SIIN peptide. Data representative of two independent experiments with four-six mice per group per experiment. (C) Quantification of total bone within predefined coordinates surrounding the non-ligated second molar (Ref.; Left) and ligated second molar (Target; Middle), and their difference (Change; Right). (D) Representative maxilla iso-surfaces from LIP mice exposed to gp33 or SIIN peptide. Shaded regions highlight superficial differences between groups. (E) Alveolar bone loss, normalized to gp33/LIP mice from each of two independent experiments with two-four mice per group per experiment. Normal. = Normalized. (F) Experimental strategy to access the impact of oral CD8⁺ T_RM reactivation on cLN cellularity and gingival inflammatory infiltrate. (G) Abundance of Ki67⁺ OT-I T cells within cLNs of day three LIP mice exposed to either gp33 or SIIN peptide. White circles represent % Ki67⁺ OT-I T cells within cLNs of non-ligated VPEP mice (No LIP). (H) Enumeration of total LN cellularity and individual DC subsets within cLNs of day three LIP mice exposed to gp33 (grey) or SIIN (blue) peptide. White bars represent DCs of the indicated subset within cLNs of non-ligated VPEP mice. Data in G and H representative of two independent experiments with four-five mice per group per experiment. (I) Ce3D imaging of day three LIP gingiva from mice of the indicated treatment group, stained with antibodies against E-Cadherin (teal) or pan-CD45 (red). ‘1′ indicates location of first molar. ‘B’ and ‘P’ denote buccal and palatal orientation, respectively. Insets magnify regions of interest. Data representative of three-four mice per group. Scale bar represents 1 mm. (J) Immunofluorescence (IF) microscopy of interstitial gingiva separating the second and third molar of non-ligated VPEP mice (Left), gp33/LIP mice (Middle), and SIIN/LIP mice (Right) seven days post ligation. White and yellow arrowheads indicate OT-I T cells in gingiva and adjacent molar dental pulp, respectively. Representative images are from at least three decalcified maxillae sections, per mouse, of three or more individual mice per group. I.P. = interproximal; 2nd = second molar; 3rd = third molar; NS = non-specific. Scale bar represents 50 μm. Error bars in all graphs represent mean ± SEM. Dots in C, E, and G represent individual mice. *, P < 0.05; **, P < 0.01; ****, P < 0.0001 as determined by an unpaired Student’s t test between the relevant comparisons.

Fig. 2.. CD8⁺ T_RM driven exacerbation of LIP is CD4⁺ T cell independent.

(A) CD4⁺ T cell frequencies (Left) and absolute numbers (Middle) within spleen and gingiva of naïve C57BL/6J mice treated two days earlier with 200 μg of GK1.5 antibody (green circles), and representative flow cytometry (Right). Red numbers indicate average fold decrease comparing No Tx (no treatment) and GK1.5 treated mice. N = Four mice per group. (B) Experimental design. Representative flow cytometry plots (Left) and enumeration of CD4⁺ T cells (Right) in (C) spleens and (D) tongues of gp33/LIP or SIIN/LIP mice ± GK1.5 antibody treatment. (E) Quantification of OT-I T cells within tongues of mice from the indicated treatment groups seven days post-LIP. (F) Representative flow cytometry plots showing DC subsets within cLNs from mice treated as indicated. (G) Quantification of DC subsets within day seven LIP cLNs from mice of the indicated treatment groups, as gated in E. Data in D-G represents three–five mice per group. (H) Quantification of total bone within predefined coordinates surrounding the non-ligated second molar (Ref.; Left) and ligated second molar (Target; Middle), and their difference (Change; Right). (I) Representative maxilla iso-surfaces from gp33/LIP (Left) and SIIN/LIP (Right) mice treated with GK1.5 antibody. Shaded regions highlight superficial differences between groups. (J) Alveolar bone loss, normalized to gp33/LIP mice treated with isotype-control antibody from two independent experiments with two-five mice per group per experiment. Scale bars in all graphs represent mean ± SEM. Dots in A, C, D, E, H, and J represent individual mice. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 as determined by an unpaired Student’s t test between the relevant comparisons.

Fig. 3.. Antigen restimulation of oral CD8⁺ T_RM amplifies gingival transcriptional responses in LIP.

(A) Experimental design to assess transcriptional changes in LIP gingiva driven by oral CD8⁺ T_RM reactivation. (B) Venn-diagram summarizing the number of DUGs identified in gp33/LIP vs. SIIN/LIP mice three- and seven-days post-LIP, compared to non-ligated VPEP mice. Red arrows highlight shared genes. (C) Log2 mean fold change in genes shared between gp33/LIP and SIIN/LIP mice at days three and seven (Left). Yellow shade highlights significance cutoff. (Right) Average % increase in shared genes between gp33/LIP and SIIN/LIP mice at the indicated timepoint. Expression values for shared genes in gp33/LIP mice were set at 100 %. Percent increase in shared SIIN/LIP genes calculated as 100 % + [(SIIN/LIP value − gp33/LIP value) ÷ gp33/LIP value × 100]. (D) Average expression of the top 10 DUGs in SIIN/LIP mice (blue circles) at day three and day seven, and their values in gp33/LIP mice (grey circles). Red line highlights Il6 expression. (E) Fold change in the indicated T cell effector genes comparing gp33/LIP vs. SIIN/LIP mice three (open circles) and seven days (closed circles) post LIP, normalized to No Tx controls. (F) Fold change in Cd69, Fasl, and Ptprc comparing gp33/LIP vs. SIIN/LIP mice three (open circles) and seven days (closed circles) post LIP, normalized to No Tx controls. (G) Heatmaps showing Log2 fold expression changes for the indicated genes in gp33/LIP vs. SIIN/LIP mice at three- and seven-days post-LIP, normalized to No Tx controls. (H) Fold change in the indicated periodontitis-associated genes comparing gp33/LIP vs. SIIN/LIP mice three (open circles) and seven (closed circles) days post LIP, normalized to No Tx controls. Group numbers are as follows: No Tx; N = 5, gp33/LIP; N = 4 per timepoint, SIIN/LIP; N = 4 per timepoint. Error bars in all graphs represent mean ± SEM. Dots in E, F, and H represent individual mice. Hashed line in E, F, and H denotes a value of 1. *, P < 0.05; **, P < 0.01; ****, P < 0.0001 as determined by a paired (C, D) or unpaired (E, F, H) Student’s t test between the relevant comparisons.

Fig. 4.. Unique genes and transcriptional pathways driven by oral CD8⁺ T_RM reactivation and LIP.

(A) Venn-diagram highlighting the number of unique DUGs in gingiva of gp33/LIP and SIIN/LIP mice at days three and seven, compared to No Tx. Red arrows highlight unique genes. Gene ontogeny enrichment analysis of genes uniquely expressed in gp33/LIP or SIIN/LIP gingiva at day 3 (B) and day 7 (C) post-ligature placement. GO terms were categorized using the ‘Biological Process’ ontology. (D) Heatmap showing Log2 fold expression changes in the top 15/31 DUGs shared at both day three and seven in gingiva of SIIN/LIP mice. Group numbers are as follows: No Tx; N = 5, gp33/LIP; N = 4 per timepoint, SIIN/LIP; N = 4 per timepoint.

Fig. 5.. CD103⁺ CD8⁺ T_RM catalyze LIP-associated alveolar bone loss in the presence of restimulating antigen.

(A) Experimental design to assess the contribution of oral CD103⁺ CD8⁺ T_RM from CD103^neg recirculating CD8⁺ T cells on aggravated bone loss following LIP. Experiments were performed in CD103^KO host mice. (B) Fold change in the number of CD103⁺ OT-I T cells, CD4⁺ T cells, and PMNs isolated from LIP gingiva from mice of the indicated treatment group, normalized to their abundance in IgG/gp33 treated mice. Data representative of two independent experiments with three-four mice per group per experiment. (C) Quantification of total bone within predefined coordinates surrounding the non-ligated second molar (Ref.; Left) and ligated second molar (Target; Middle), and their differences (Change; Right). (D) Representative maxilla iso-surfaces from LIP mice exposed to SIIN peptide with an intact (Left; blue) or depleted (Right; green) oral CD103⁺ CD8⁺ T_RM compartment. Shaded regions highlight superficial differences between groups. (E) Alveolar bone loss, normalized to IgG/gp33 mice, from each of two independent experiments with three-five mice per group per experiment. Error bars in all graphs represent mean ± SEM. Dots in C and E represent individual mice. **, P < 0.01; ***, P < 0.001 as determined by an unpaired Student’s t test between the relevant comparisons.