Medullary thymic epithelial cells and CD8α+ dendritic cells coordinately regulate central tolerance but CD8α+ cells are dispensable for thymic regulatory T cell production

Abstract

In the thymus, antigen presenting cells (APCs) namely, medullary thymic epithelial cells (mTECs) and thymic dendritic cells (tDCs) regulate T cell tolerance through elimination of autoreactive T cells and production of thymic T regulatory (tTreg) cells. How the different APCs in the thymus share the burden of tolerazing the emerging T cell repertoire remains unclear. For example, while mutations that inhibit mTEC development or function associate with peripheral autoimmunity, the role of tDCs in organ-specific autoimmunity and tTreg cell production remains controversial. In this report we used mice depleted of mTECs and/or CD8α⁺ DCs, to examine the contributions of these cell populations in thymic tolerance. We found that while mice depleted of CD8α⁺ DCs or mTECs were normal or developed liver inflammation respectively, combined depletion of mTECs and CD8α⁺ DCs resulted in overt peripheral autoimmunity. The autoimmune manifestations in mice depleted of both mTECs and CD8α⁺ cDCs associated with increased percentages of CD4⁺ and CD8⁺ T cells in the thymus. In contrast, while mTEC depletion resulted in reduced percentages of tTreg cells, no additional effect was observed when CD8α⁺ DCs were also depleted. These results reveal that: 1) mTECs and CD8α⁺ DCs cooperatively safeguard against peripheral autoimmunity through thymic T cell deletion; 2) CD8α⁺ DCs are dispensable for tTreg cell production, whereas mTECs play a non-redundant role in this process; 3) mTECs and CD8α⁺ DCs make unique contributions to tolerance induction that cannot be compensated for by other thymic APCs such as migratory SIRPα⁺ or plasmacytoid DCs.

Keywords: Autoimmunity; Dendritic cells; Medullary thymic epithelial cells; Regulatory T cells.

Figure 1

Conventional CD8α⁺ DCs are depleted in the thymus of Irf8^−/−/Traf6ΔTEC and Batf3^−/−/Traf6ΔTEC dKO mice. (A–C) Representative examples of thymic cDCs (CD11c^highPDCA1⁻) stained with anti-CD8α and -SIRPα mAb using total thymocyte suspensions from mice with indicated genotype. (D) Quantification of SIRPα⁺ and CD8α⁺ cDCs, expressed as percentages of thymic cDCs, in Irf8^−/−/Traf6ΔTEC mice (left bar graph), Batf3^−/−/Traf6ΔTEC (middle bar graph) and chimeric Batf3^−/−→Traf6ΔTEC mice. Results are expressed as mean + s.e.m. (n>3 mice per group). Depletion of CD8α⁺ cDCs in Irf8^−/−, Irf8^−/−/Traf6ΔTEC, Batf3^−/−, Batf3^−/−/Traf6ΔTEC and Batf3^−/−→Traf6ΔTEC chimeras was statistically significant compared to WT and Traf6ΔTEC mice using 1-way anova with Tukey's Multiple Comparison Test.

Figure 2. Lower weight in mice depleted of both mTECs and CD8α⁺ cDCs

Photographs and weight quantification of (A) 5 to 6-week-old WT, Traf6ΔTEC, Irf8^−/− and Irf8^−/−/Traf6ΔTEC dKO mice and of (B) 4 to 5-week-old WT, Traf6ΔTEC, Batf3^−/− and Batf3^−/−/Traf6ΔTEC dKO mice. Scale bar 1cm. (C) Graph representing the percentage of initial weight over time starting four weeks after irradiation-transplantation of Batf3^−/−/Traf6ΔTEC chimeras and (D) area under the curve of percentage of initial weight shown in (C). Results are shown as mean and individual mice (A–B), and mean + s.e.m (C–D). (n=3 Batf3^−/− and >5 mice for all other groups of mice). Statistical significance was analyzed by 1-way anova with Tukey's Multiple Comparison Test.

Figure 3. Combined depletion of mTECs and CD8α⁺ cDCs leads to multi-organ autoimmunity

Photographs of H&E-stained paraffin-embedded sections of indicated organs from (A) WT, Traf6ΔTEC, Irf8^−/− and Irf8^−/−/Traf6ΔTEC dKO and (B) from WT→WT, WT→Traf6ΔTEC, Batf3^−/−→WT, Batf3^−/−→Traf6ΔTEC chimeras. (C, D) Sera of WT, Traf6ΔTEC, Irf8^−/− and Irf8^−/−/Traf6ΔTEC dKO mice (C) and WT→WT, WT→Traf6ΔTEC, Batf3^−/−→WT, Batf3^−/−→Traf6ΔTEC chimeras (D) were used to stain HEp-2 cells and the presence of auto-antibodies (AutoAb) was revealed by anti–mouse IgG–FITC and fluorescence microscopy Quantification of AutoAb was performed on an arbitrary scale of 0–5, based on positive staining and fluorescence intensity. Scale bars 100µm. Results are shown as mean and individual mice . n>5 mice per group. Statistical significance was analyzed by 1-way anova with Tukey's Multiple Comparison Test.

Figure 4. Increased frequency of SP Thymocytes in mice lacking mTECs and CD8α⁺ cDCs

(A) Thymic cellularity and (B) percentage of CD4⁺CD8α⁻ (CD4 SP) and CD4⁻CD8α⁺ (CD8 SP) thymocytes in WT, Traf6ΔTEC, Irf8^−/− and Irf8^−/−/Traf6ΔTEC dKO mice (left bar graph), WT, Traf6ΔTEC, Batf3^−/− and Batf3^−/−/Traf6ΔTEC dKO mice (middle bar graph) and in WT→WT, WT→Traf6ΔTEC, Batf3^−/−→WT, Batf3^−/−→Traf6ΔTEC chimeras (right bar graph). Results are expressed as mean + s.e.m of individual mice. n=3 Batf3^−/− and >5 mice for all other groups of mice. Statistical significance was analyzed with a 1-way anova with Tukey's Multiple Comparison Test.

Figure 5. CD8α⁺ cells are dispensable for thymic regulatory T cell production

(A) Percentage of Foxp3⁺CD25⁺ tTreg among CD4 SP in WT, Traf6ΔTEC, Irf8^−/− and Irf8^−/−/Traf6ΔTEC dKO mice (left bar graph), WT, Traf6ΔTEC, Batf3^−/− and Batf3^−/−/Traf6ΔTEC dKO mice (middle bar graph) and in WT→WT, WT→Traf6ΔTEC, Batf3^−/−→WT, Batf3^−/− →Traf6ΔTEC chimeric mice (right bar graph). (B) Quantification of Treg (CD25⁺GFP^Foxp3+) among CD4⁺ T cells five days after in vitro incubation of CD4⁺CD8⁻GFP^Foxp3−CD25⁻ -or CD4⁺CD8⁻GFP^Foxp3+CD25⁻ thymocytes with thymic DC subset. Quantification of viable cells by flow cytometry (AnnexinV⁻PI⁻) after 24 hours of in vitro incubation of total thymocytes from OTII mice with indicated thymic DC subsets with or without Ova_323–339. Results are expressed as mean + s.e.m of individual mice (A) or of culture triplicates (B–C). n=3 Batf3^−/− and >5 mice for all other groups of mice for (A), (B–C) are representative of two independent experiments with similar results. Statistical significance was analyzed with a 1-way anova with (A) Tukey's Multiple Comparison Test and (B–C) Dunnett's Multiple Comparison Test.