Overview of results:
- Identify a panel of Mtb epitopes recognized via HLA-E.
- Investigate differences in peptide binding to HLA-E*01:01 and HLA-E*01:03, as well as peptide presentation by these two molecules and T cell recognition, with the objective of selecting peptides presented by both alleles.
To achieve these objectives we used a novel HLA-E-peptide-binding assay which was adapted to test peptide binding to HLA-E*01:01 and HLA-E*01:03 separately. We additionally expanded this assay to test peptide binding to Mamu-E and Qa-1b which are the HLA-E counterparts in monkeys and mice. We were interested in expanding this comparison to all four molecules because of the importance of these animals for pre-clinical trials. Given the differences already seen between the two human molecules, we decided to use this comparison to select peptides which behave similarly in the context of the four molecules, and therefore would translate better from pre-clinical to clinical trials. We found that despite all four molecules sharing main anchor positions, there were differences in the amino acids preferred and therefore each molecule presented its own peptide binding motif.
Based on the results we obtained on the peptide binding motifs, we next developed a prediction algorithm to identify new peptides which HLA-E-binding capacity. We could identify new sequences derived from Mtb that have the capacity to bind HLA-E. We are now currently testing whether these peptides have the capacity to induce protective T cell responses.
- Determine the molecular pre-requisites for HLA-E-peptide-TCR interactions.
To determine the molecular pre-requisites of peptide binding to HLA-E we followed different strategies. First, we performed alanine substitutions on each position of known high and moderate peptide binders to HLA-E to investigate how this substitution could impede or promote peptide binding to HLA-E. Alanine is the smallest amino acid and substitution of any amino acid with alanine is expected to have consequences to the behavior and interactions of the peptide. Then, we tested binding of combinatorial peptide libraries (CPL) which is a collection of 180 peptide pools, each of them containing a single amino acid fixed at one position while the rest of the sequence is random, and therefore this approach can reveal the specific contribution of each residue at each position. Finally, based on our results on peptide binding to HLA-E we did additional amino acid substitutions at selected main anchor positions to investigate the degree of residue flexibility these positions have for peptide binding to HLA-E. For these experiments we also performed a comparison of HLA-E*01:01, HLA-E*01:03, Mamu-E and Qa-1b in order to establish the particular requirements for each molecule. The information gathered with these experiments was in turn used to further inform the HLA-E/peptide binding prediction algorithm described above.
To investigate the molecular requirements for TCR recognition of HLA-E-presented peptides, we performed a TCR repertoire analysis of single cell sorted Mtb-specific HLA-E-restricted T cells from BCG-vaccinated (LTBI) adolescents from South Africa. We obtained GLIPH (Grouping of Lymphocyte Interactions by Paratope Hotspots) clusters, a classification of sequences with similar patterns, in order to establish if there is an enrichment of a certain TCR sequence in the context of HLA-E-restricted T cells. Based on these findings, we made a selection of most prevalent TCR sequences and express them into cell lines to further characterize them in the lab.
- Test capacity of these peptides to induce CD8+ T cell activation and Mtb killing.
To test the capacity of our newly identified peptide to induce protective immune cells we quantify the amount of cells that replicate when cultured with the peptide, as activated T cells are known to replicate. We are currently working on expanding these results and to including the measurement of additional markers for specificity such as tetramer staining and proliferation markers such as CD137.
