T cell receptors (TCR) bind to peptides from various sources on MHC (Major Histocompatibility Complex) molecules. A long-standing goal in the field is to understand the mechanisms of MHC-peptide exchange and MHC-TCR interactions. Here, I present work from three uniquely different systems that address the following: HLA-DR1 conformational stability, self-tolerant mechanisms of TCRs isolated from self-reactive TCR transgenic mice, and TCR cross-reactivity mechanisms between LCMV and VV.
First, I present a crystal structure of HLA-DR1 in complex with A1L9 peptide, a peptide with two amino acid substitutions from the parental peptide. The singly substituted A1 peptide, which has a pocket 1 alanine substitution, decreases intrinsic half-life between MHC-peptide and increases susceptibility to HLA-DM mediated peptide exchange. This data agrees with previous models of HLA-DM-mediated peptide exchange in which the major determinant is located at the HLA-DR1 pocket 1. However, the L9 substituted peptide, which has a pocket 9 leucine substitution, displays the opposite phenotype: increased intrinsic half-life and decreased HLA-DM susceptibility. The crystal structure presented here shows that HLA-DR1 in complex with a doubly substituted peptide, A1L9, is in the same conformation as HLA-DR1 with the wild-type peptide, demonstrating that pocket 9 residues can rescue pocket 1 residue binding deficiencies and that HLA-DR1 stability is determined by amino acids along the peptide, not only at pocket 1.
Next, I present crystal structures of two self-tolerant TCRs in complex with IAb-3K pMHC. To elucidate molecular mechanism for self-reactivity and self-tolerance, the TCRs J809.B5 and 14.C6 are compared to each other and its parental self-reactive TCR, YAe-62.8. In comparison to YAe-62.8, J809.B5 interacts with the same pMHC, but utilizes more peptide specific interactions, a mechanism that may distinguish self-reactive receptors from self-tolerant receptors. Additionally, the crystal structure of 14.C6 TCR, which bears a different CDR3α sequence from J809.B5, demonstrates that CDR3 sequences can modulate interactions of germline encoded CDR1 and CDR2 loops. Together, these results highlight that in addition to CDR3 VDJ recombination, diversity is generated in the mature TCR repertoire by differential chain pairing, either of which can affect the interactions of germline encoded CDR loops.
Next, I present a detailed analysis of cross-reactive TCRs between Kb-GP34 and Kb-A11R. The mature LCMV-immune repertoire was analyzed by DNA deep sequencing of TCRβ CDR3 sequences, which led to the identification of new cross-reactive sequence motifs. Cross-reactive sequence motifs varied by each Vβ gene, suggesting a role of CDR1, CDR2, and CDR3 loop interplay in cross-reactivity.
Lastly, I present the crystal structures of a GP34/A11R cross-reactive TCR in complex with both Kb-GP34 and Kb-A11R. Analysis of the crystal structures revealed that the two complexes are largely the same, despite differences in peptide sequences. Surprisingly, the TCR to peptide interactions were dominated by three out of eight peptide side-chains. Cross-reactivity between these two complexes is likely due to a large amount of interactions from TCR to MHC compared to interactions of TCR to peptide. We note two unique MHC-peptide interactions that may allow Kb to be an allele prone to cross-reactivity. The first is an interaction at the C-terminus of the A11R peptide which pulls A11R P7 asparagine away from TCR interactions. The second interaction is from an arginine at position 155, which sits at the interface between TCRα and TCRβ , and contributes the most buried surface area in the interaction interface. Because Kb’s arginine 155 is a long side chain that hydrogen bonds with the peptide backbone, and is also at the center of the TCR-peptide interface, GP34 and A11R peptide sequence differences may be occluded from TCR discrimination by Kb presentation.
The data presented in this dissertation demonstrate that interactions between MHC-peptide and MHC-TCR act harmoniously and coopertively, whereby proximal interactions are affected by interactions elsewhere. While previous models of HLA-DR/HLA-DM interactions demonstrate the importance of interactions at HLA-DR1 pocket 1, I showed that pocket 9 also contributes to HLA-DR stability and therefore, HLA-DM susceptibility. I also showed that TCR CDR3 loop sequences affect germline CDR1/CDR2 loop interactions and vice versa. Lastly, I showed that allele specific MHC side chain interactions with the bound peptide influence TCR ligand binding and hence, TCR cross-reactivity.
| Identifer | oai:union.ndltd.org:umassmed.edu/oai:escholarship.umassmed.edu:gsbs_diss-1995 |
| Date | 15 May 2018 |
| Creators | Trenh, Peter |
| Publisher | eScholarship@UMMS |
| Source Sets | University of Massachusetts Medical School |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | GSBS Dissertations and Theses |
| Rights | Licensed under a Creative Commons license, http://creativecommons.org/licenses/by/4.0/ |
Page generated in 0.019 seconds