Abstract
T cell receptors (TCRs) recognise peptides bound to the major histocompatibility complex(MHC) as part of cellular immunity, leading to death of the antigen-presenting cell. Although
normally targeting pathogenic cells, TCRs can be used for immunotherapy, whereby they are
engineered to target a specific peptide antigen on cancer cells. One approach for immunotherapy
are soluble TCRs (e.g. ImmTACs, which target cancer). This requires TCRs binding peptideMHC (pMHC) complexes with high affinity. The development of high affinity TCRs for
therapeutic applications has been successful, but the efficiency of the process could be improved
by incorporating computational approaches. So far, such approaches have been limited, due to
not reaching sufficiently high affinity or specificity (i.e. leading to cross-reactivity). In this thesis,
various TCR-pMHC systems are explored to improve understanding of the system, with the
aim of developing workflows for simulation-led therapeutic TCR design. Initially, evaluations
of the dynamics of the non-polymorphic MHC class 1 like molecule, MR1, were performed to
improve understanding of the impact of different ligands and variants on the dynamics. Then,
computational workflows for efficient and reliable prediction of the TCR-pMHC binding affinity
were developed, using several different TCR-pMHC systems to evaluate the generality of the
approach. Further, comparisons of cross-reactive and specific TCRs using molecular dynamics
simulation demonstrated that increases in the interactions between TCR and MHC could drive
cross-reactivity, allowing binding of the TCR to peptides that are not the intended target.
This understanding allowed the development of a workflow for predicting specificity, which,
together with the affinity prediction workflow, was applied to a computational TCR redesign
workflow to generate TCRs of high affinity and specificity for therapeutic applications. Overall,
this thesis increases our understanding of TCR affinity and specificity and demonstrates how
this understanding can be used to develop simulation-led approach for engineering therapeutic
TCRs.
| Date of Award | 17 Sept 2024 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Marc W Van der Kamp (Supervisor) & Linda Wooldridge (Supervisor) |