Regulatory T cells enforce self-nonself discrimination by constraining conventional T cells of matched self-specificity during infection

During infections, Foxp3+ regulatory T (Treg) cells must control autoreactive conventional T (Tconv) cell responses against self-peptide antigens while permitting those against pathogen-derived “nonself” peptides. We sought to define the basis of this selectivity using mice in which Tregs reactive to a single prostate-specific self-peptide/MHC (self-pMHC) antigen, termed "C4", were selectively depleted ("C4ΔTEC" mice). We demonstrated that Tregs reactive to the C4 peptide were dispensable for the control of Tconv cells of matched antigen specificity at homeostasis but were required to control such Tconv cells and prevent fulminant prostatic autoimmunity following infection with the pathogen L monocytogenes engineered to express the C4 peptide ("Lm[C4]"). We performed the experiment outlined here to understand how the absence of C4-specific Treg cells in C4ΔTEC mice changes the phenotypic state of the C4-specific Tconv cells following Lm[C4] infection to drive autoimmunity. We found that C4-specific Tconv cells adopted preferential and differential cell states in both C4ΔTEC mice and wild-type mice that have C4-specific Treg cells. Specifically, gene expesion data revealed that C4-specific Tconv cells in wild-type mice preferentially adopted a non-proliferating effector state, while C4-specific Tconv in C4ΔTEC mice appeared as proliferating effector cells and proliferating central-memory cells. Wild-type and C4ΔTEC male mice were infected intraveneously with 10^7 CFU of attenuated (ΔActA) Lm[C4]. 4 days post-infection spleens were harvested, CD4+ T cells were enriched using MACS CD4+ negative selection, cells were stained with C4/I-Ab APC- and PE-tetramers, fluorophores were cross-linked with anti-fluor MACS microbeads and tetramer-binding cells were positively selected over an AutoMACS column. Eniched samples from individual mice were stained with surface antibodies and Biolegend hashtag antibodies, then pooled into a sample containing enriched cells from 9 or 10 mice, and FACS purified. This entire process was completed twice, first with 4 wild-type and 5 C4ΔTEC mice, then with 5 wild-type and 5 C4ΔTEC mice, staggering the harvest by ~4 hours. After sorting, there were 2 samples total (Samples A and B), where each sample contained hashtagged C4-tetramer+ cells from 9 or 10 mice, respectively. These samples were loaded into one of two lanes of a sequencing chip and then subjected to Drop-seq (10X Genomics) to coencapsulate individual cells in reverse emulsion droplets in oil together with one uniquely barcoded mRNA-capture bead. Libraries were derived from single cells, then subjected to next-generation sequencing.