A Study Of Factors Underlying BCG Immunogenicity Differences Across Countries: The Influence Of DNA Methylation Patterns And Antigen Presenting Cells

Introduction: The protective efficacy of the BCG vaccine against pulmonary tuberculosis varies across the globe and is lower in countries near the equator. Previous studies have shown disparate immunogenicity profiles after infant BCG vaccination in The United Kingdom, Malawi, The Gambia and Indonesia. Pre-exposure to environmental mycobacteria and maternal or infant coinfections are one of the mechanisms that could explain these variations. Exact molecular mechanisms that drive sustained differences in BCG immunogenicity are not known. Methods and Hypothesis: In this study we aimed to investigate molecular mechanisms that drive differences in BCG immunogenicity. We hypothesised that differential DNA methylation or dynamic phenotypes of antigen presenting cells contribute to underlying differences in BCG immunogenicity. We used samples from South African, Ugandan and UK BCG vaccinated infants to investigate their BCG-induced immune responses. In South Africa we compared DNA methylation profiles between high and low BCG responders. In the UK and Uganda, we compared BCG specific immune responses and DNA methylation profiles between both countries. Lastly, in Uganda, we investigated longitudinal changes in dendritic cell phenotypes. Results: We found that differential DNA methylation of immune pathways such as T cell activation, or potassium and calcium channel Signalling is associated with variation in BCG immunogenicity in South African infants. The 4 comparison of immune responses between UK and Ugandan infants revealed that Ugandan infants have a strong inflammatory response to BCG, but levels of TH1 and TH2 cytokines are similar. The analysis of DNA methylation between these countries revealed that genes of B cell activation pathway are more methylated in Uganda, whereas genes of TGFb regulatory pathways are more methylated in the UK. The analysis of antigen presenting cells revealed that key dendritic cell populations are absent at birth and develop within the first year of life. We also highlight that cytomegalovirus (CMV), Epstein-Barr virus (EBV) and maternal latent tuberculosis infection (LTBI) have profound effects on dendritic cell phenotypes. Implications: These results point towards a role for DNA methylation in the regulation of BCG immune responses, however, in vitro studies focusing on highlighted gene candidates and pathways should be conducted. Small differences in immunogenicity profiles after BCG vaccination in the UK and Uganda suggest that BCG efficacy in these countries should be similar, an area requiring a further investigation. The absence of dendritic cells at birth should be investigated and the kinetics of their development within the first week of life examined. Special attention should be paid to the development of vaccines or treatments for CMV, EBV and LTBI, as these pathogens modulate DC phenotypes