CD8+ T cell specificity in thymic selection and in the recognition of antigen

Abstract: Specificity mediated by the T cell receptor (TCR) is of fundamental importance for the ability of T cells to function, and regulate humoral immunity and effector mechanisms of the innate immune system. T cells recognize composite molecules consisting of products of the major histocompatibility complex (MHC) class I or class 11 loci, in complex with a short peptide fragment derived from endogenous or antigenic proteins. TCR-mediated detection of such peptide-MHC complexes is the basis of T cell specificity. This thesis deals with specificity in diverse aspects of T cell biology, such as thymocyte positive and negative selection, and mature T cell recognition of the peptide and MHC components of the peptide-MHC complex. Further, the issue of strict specificity versus degeneracy in T cell recognition is addressed. Finally, specificity in terms of epitope selection during an ongoing T cell response, i.e. the phenomenon of immunodominance, is investigated. Mice with a targeted mutation in the transporter associated with antigen processing I (TAP1) gene do not translocate peptides from the cytosol to the ER for binding to MHC class I molecules. These mice cae therefore serve as a model for CD8+ T cell selection in the absence of the great diversity of TAP1 -dependent self-peptides. Diversity of the CD8+ T cells was investigated functionally by immunization with a set of well-defined CD8+ T cell epitopes, and structurally by analysis of TCR Vß gene usage. The results indicate that a diverse and functional CD8+ T cell repertoire is positively selected in the absence of TAP1-dependent self-peptides, although minor limitations in diversity were detected. These data together with other recent data in the literature suggests that positive thymic selection is semispecific event where peptide dependency varies between T cells. Mice lacking a functional ß2-microglobulin (ß2m) gene express very low levels of properly conformed MHC class I molecules, and cae therefore serve as a model system for CD8+ T cell selection where the selecting MHC ligand is scarcely expressed. Accordingly, these mice have few CD8+ T cells. Nevertheless, the small thymus-dependent CD8+ T cell population was able to generate a peptide specific CD8+ cytotoxic T lymphocyte (CTL) response to peptide immunization. These CTL displayed a dual specificity; peptide specificity against target cells with low MHC expression, and peptide- independent specificity for the MHC restriction element as such against targets with high MHC expression. Thus, the thymic selection window is shifted at low levels of MHC expression such that T cells with high avidity for self-MHC are positively selected. It was further demonstrated that the high avidity for the MHC restriction element was due to high TCR affinity for the MHC component of the peptide-MHC complex. This phenotype was employed to investigate the role TCR recognition of the MHC restriction element in terms of T cell specificity for antigenic peptide. It was found that increased recognition of MHC could compensate for lack of TCR contact residues in the peptide. Thus, variable TCR interaction with MHC modulates T cell peptide-specificity. These data are interpreted in the context of recent structural and functional data of TCR-MHC interactions in thymic selection and recognition of peptide-MHC complexes during the life of mature T cells. The role of self-peptides in shaping the mature T cell repertoire to individual antigens is almost completely undefined. Given the flexible nature of T cell receptor specificity, permanent deletion or disabling of all CD8+ T cells with the capacity to recognize MHC class I-presented self-peptides would severely limit the diversity of the repertoire. To address this problem we have studied T cell reactivity and tolerance to a naturally MHC class I presented selfpeptide derived from the ubiquitously expressed elongation factor 1[alpha] (EF1[alpha]). This peptide was isolated from immunoprecipitated H-2Kb molecules and sequenced by mass spectrometry. Surprisingly, the CD8' T cell repertoire was not completely tolerant to the EF1[alpha] peptide, nor to an isoform of EF1[alpha] differing in on TCR contact residue in the peptide, despite that expression of both isoforms was detected in dendritic cells believed to be responsible for negative selection in the thymus. Low avidity CTL were generated against both isoforms upon peptide immunization, which were able to recognize peptide-pulsed target cells but did not kill cells expressing normal levels of the peptide. The low avidity in CTL recognition of this peptide was due to low TCR affinity. Thus, T cells which are capable of recognizing the self-peptide with low avidity are spared from deletion by negative selection, CTL generated against altered selfpeptide ligands were of intermediate avidity, indicating that epitopes with similarity to self are represented in the corresponding repertoire by T cells with intermediate avidity. A model for maintenance of both maximal diversity and optimal self-tolerance in the CD8+ T cell compartment, based on the presence of "avidity-pits" rather than "holes" in the repertoire is discussed. The propensity of the CD8+ T cell repertoire to focus on a few of many possible epitopes in a complex antigen is defined as "immunodominance". To investigate the hierarchical pattern of CD8+ T cell recognition of multiple epitopes, mice were immunized with mixtures of synthetic peptides representing processed antigenic epitopes, each of which had previously been defined as irnmunodominant in its antigenic system. Novel "superdominance" hierarchies were detected upon co-immunization with these epitopes in adjuvant. Interestingly, dominance was reversed when the epitopes were injected loaded on live dendritic cells (DC). These data are discussed in relation to the current literature on mechanisms for immunodominance, and potential use of the "DC effect" in vaccine development.