On the diversity of T cell receptors in the genus Mus

Abstract

The diversity of T cell receptors (TCRs) is one of the backbones of an effective adaptive immune system. This diversity is generated by somatic rearrangements of gene segments in two separate peptide chains that dimerize to form a unique receptor that can specifically recognize antigens. The generative process of TCR repertoire formation is largely defined by stochastic events that can theoretically give rise to more than 10^15 unique receptors. Strikingly, immune responses to common pathogens are frequently driven by identical or very similar TCRs. Consequently, there is significant non-random sharing of such “public” receptors between individuals. This has invoked the idea that genetically encoded factors contribute to the shaping of an individual’s TCR repertoire, but experimental validation of such factors has been lacking due to the technical challenge of capturing the sheer size of diverse TCR repertoires. I have developed a single-cell and high-throughput TCR sequencing protocol capable of generating paired TCR sequencing data from millions of CD8+ T cells. To reveal the contribution of genetic factors in the generation of TCRs, I generated TCR repertoires from 32 mice representing the reference lab mouse and three sister species, as well as F1 hybrids between them. Collectively, these mice span an evolutionary divergence time of approximately three million years and represent an exceptional model to study germline determinants of TCR repertoire formation. By conducting a comprehensive comparison of the variable, diversity and joining gene segments across the different species, I showed that the TCR alpha variable gene segment locus has undergone a major locus expansion. Following this observation, I was able to show that the usage frequencies of gene segments of TCRs varied significantly across species but were remarkably conserved in intra-species repertoires. Using F1 hybrids, I demonstrated genetic control in usage for specific gene segments, because individual parental alleles retain differential usage frequencies despite a shared heterozygous genetic background. Further, I have evaluated the impact of thymic selection on the shaping of an individual’s repertoire. TCR repertoire diversity reduction caused by thymic selection is mostly defined by rejection of variable gene segments in TCR beta chains and occurs strictly through direct protein-protein interaction with antigen-presenting major histocompatibility complexes. This has significant consequences for the sharing of identical and similar TCRs across several individuals. Further, by comparing the frequencies of short amino acid motifs of TCRs, I showed that even in those regions, arising from seemingly random fusion of gene segments, abundances of particular amino acids motifs are remarkably dependent of the respective genotype of an individual. This work not only provides an approach to analyze TCR repertoires at unprecedented scale but also reveals a surprising extent of genetic contribution to the shaping of an individual’s TCR repertoire.