Neutral and selective drivers of TCR repertoire diversity

Abstract

The vertebrate adaptive immune system relies on the immense diversity of T cell receptors (TCRs) and immunoglobulins (IGs) to recognize a wide array of pathogens. Each T and B cell expresses receptors encoded by recombination of gene segments during lympchocyte development. This somatic generation of diversity depends on the germline genes available for recombination and, for TCRs, on interactions with the major histocompatibility complex (MHC). Adaptive immune repertoire diversity enables immune surveillance and protection. In this dissertation, I investigate the evolutionary and selective forces shaping TCR repertoire diversity across multiple levels, from germline gene evolution to repertoire structure, using whole-genome and immune repertoire sequencing across humans, non-human great apes, mice, and bank voles. First, I examined the evolution of germline TRBV and IGHV gene families in non-human great apes using read-depth-based copy number estimates and McDonald-Kreitman tests. Despite their immunological importance, I found only modest copy number differences across species and limited evidence for lineage-specific expansions or adaptive evolution. V gene allelic variation showed signatures of purifying selection consistent with most protein-coding genes. Together, these results show that V genes in the great apes did not experience diversifying selection despite differing pathogen history and geographic distributions. Second, I tested the hypothesis that greater MHC diversity within individuals expands the breadth of the TCR repertoire by increasing the number of TCRs that pass thymic selection. Using ecological and network-based diversity metrics in a large human cohort, I consistently found no strong association between MHC diversity and TCR repertoire composition. This surprising null result held across multiple MHC diversity measures and covariate adjustments, and was supported by comparative analyses in bank voles. Despite wide variation in MHC genotypes, most individuals appeared to achieve a similar minimum level of repertoire diversity - suggesting that the immune system generates sufficient diversity. Finally, by comparing productive (selected) and nonproductive (unselected) TCR sequences in genetically identical mice, I isolated the effects of thymic and peripheral selection. Public TCRs were consistently enriched and more abundant in the naive productive repertoire, showing that selection shapes TCR usage beyond neutral generation biases. Together, these results suggest that the overall structure of the repertoire is robust to variation in MHC diversity. Moreover, recent evolution of germline V gene content in great apes shows little evidence of adaptive expansion. However, interactions with self-peptides presented by the MHC enhances the abundance of public TCRs. These findings support a model in which the adaptive immune system achieves sufficient diversity through robust, cross-reactive repertoires shaped largely by stochastic processes - downplaying the role of MHC-mediated selection in shaping overall repertoire structure.