Investigating Lipid-Responsive T Cells in Tuberculosis : Paving the Way for New Lipid-Based Vaccines

Mycobacterium tuberculosis (M.tb), which causes Tuberculosis (TB), remains one of the leading causes of death worldwide. The lack of an effective vaccine and the persistence of multi-drug resistant strains highlights the urgent need for improved therapeutic interventions for TB. Unconventional T cells, including CD1-restricted T cells, represent potential targets for future therapeutics due to the non-polymorphic nature of CD1 and the ability to respond to lipid antigens such as those found in the lipid-rich cell wall of M.tb. Although subsets of CD1c-restricted T cells specifically recognise mycobacterial lipids, the majority of T cells restricted by CD1c exhibit autoreactivity as they recognise CD1c bound to self-lipids. Importantly, these autoreactive T cells exhibit dual recognition of both foreign, pathogen derived lipid antigens and self-derived lipids when bound to CD1c. This may indicate a role for CD1c-restricted T cells in TB infection, but their exact function remains unknown. To investigate the hypothesis that "CD1c autoreactive T cells modulate the host-pathogen interaction in human TB infection", I first investigate molecular mechanisms that underpin CD1c recognition by a cognate TCR. I employed methods such as site directed mutagenesis and lipid pulsing to investigate TCR binding footprint and lipid antigen reactivity, respectively. I demonstrate that binding of a CD1c autoreactive αβTCR is likely focused away from the F' roof of CD1c, but further molecular studies are required to unravel exact binding footprint. Furthermore, I identified that this TCR also exhibits promiscuous recognition of various self-derived lipid cargo when bound to CD1c, but recognition is augmented by adjustments in the lipid alkyl chains suggesting a novel mechanism of recognition driven by a degree of fine specificity for lipid alkyl chains. Using CD1c tetramer guided cell sorting, I demonstrate the isolation and cloning of bona fide CD1c autoreactive γδTCRs. New TCRs can be employed in future studies to unravel the molecular mechanisms underpinning their binding to CD1c. Next, I optimised a short-term culture assay to investigate CD1c autoreactive T cells in a small cohort of healthy donors. My results showed that these responses are present in the circulation and they readily expand in vitro in response to CD1c+ APC with the majority of expanded cells exhibiting a CD4-CD8- phenotype. I then go on to investigate CD1c immunity in the context of M.tb infection. I show for the first time using immunohistochemistry that both γδ T cells and CD1c+ cells are present in the lungs of TB patients, but are largely found away from areas of caseous necrosis and reside in inflammatory tissue distal to the infection focus. Moreover, in a cohort of South African TB patients, I demonstrate an increased frequency of CD1c autoreactive αβ T cells in the peripheral blood and CD1c autoreactive Vδ1+ T cells in the lungs when compared to healthy controls. My results also reveal that lung resident Vδ1+ T cells of TB patients express elevated levels of PD1 compared to uninfected controls, suggesting that these cells had become activated and then exhausted in response to antigen. Taken together, my results suggest that M.tb drives CD1c mediated responses in vivo, indicating an important role for CD1c mediated immune responses in the host pathogen interaction in TB. I also investigated the hypothesis that "CD1d-restricted iNKTs are associated with disease severity" in the macaque model of infection. I carried out longitudinal studies in the peripheral blood of a small cohort of macaques, pre- and post-M.tb challenge, to monitor iNKT numerical and functional changes. My results demonstrate an increase in iNKT frequency at 8 weeks post challenge, however proliferative responses following stimulation with the strong iNKT agonist α-Galactosylceramide (α-GalCer) were impaired. Recovery of the proliferative response at the 4 week time point was observed in some animals, but due to the small cohort size no correlations could be made with lung iNKT frequency, bacteriology and pathology scores at the time of necropsy. Finally, I demonstrate the generation of Cynomolgus macaque CD1c tetramers (mCD1c). In comparison to human CD1c tetramers, positive staining of macaque T cells was observed with the mCD1c tetramer. Further validation of this tool is required in order to utilise these tetramers to study CD1c-restricted T cells in future M.tb in vivo challenge studies.