The impact of maternal infection with Mycobacterium tuberculosis on the infant response to BCG immunisation

Tuberculosis (TB) is still a very important health problem in sub-Saharan Africa. Many people are infected with Mycobacterium tuberculosis (M.tb) (the bacterium that causes tuberculosis) in these countries, but the infection is kept in check by the immune system and the people remain healthy. We call this latent infection. Usually, only about 10% of infected individuals will progress to having the active form of the disease in their lifetime. However, when people do get the active disease, they can infect many other people, and so the cycle continues. TB kills approximately 2 million people every year, and another 10 million people become infected with M.tb every year.

One way to eradicate TB would be to immunise everyone with an effective vaccine. Unfortunately, to date there is only one licensed vaccine, BCG, which was first introduced almost 100 years ago. BCG is given in many countries at, or around the time of birth. In sub-Saharan Africa, this immunisation schedule has been very effective at preventing some of the more severe (and deadly) forms of the disease in children, but the vaccine does not protect against the more common, active pulmonary form of the disease (the disease affecting the lungs) that occurs in adolescents and adults. By contrast, in countries such as the UK, BCG immunisation gives up to 80% protection against active pulmonary disease in adolescents and adults. We do not know why BCG should give poorer protection against pulmonary TB in sub-Saharan African countries compared to European countries, and our study is aimed at understanding this phenomenon.

Since latent M.tb infection is more prevalent in sub-Saharan Africa than in the UK (for example), we think that part of the answer may lie in whether a mother has an underlying, latent M.tb infection. We think that this infection in a pregnant woman may affect the developing baby's own immune system. For example, if the baby is exposed to maternal M.tb infection while in the womb, when its immune system is developing, the immune system may regard the presence of Mtb infection as "normal" and fail to mount a suitable reaction to the infection. In such a case, when the baby is born and is given the BCG vaccine, the baby may not respond in the same way as a baby who was born of a mother without a latent M.tb infection (such as mothers in the UK).

Our study will therefore look at the infant immune response to the BCG vaccine. We will examine immune responses in babies born in sub-Saharan Africa from mothers who do and don't have a latent M.tb infection, and we will compare their results with results from babies born of mothers in the UK, where M.tb infection is much less common. These results will give us a good idea of whether maternal M.tb infection status affects the infant response to BCG. If it does, then we may have to think about testing young women for the infection, and treating them before or during pregnancy.

We propose that the poor effectiveness of BCG in sub-Saharan Africa may be due to maternal Mycobacterium tuberculosis (M.tb) infection, which is endemic in this setting. If maternal M.tb infection influences the infant response to BCG, the timing, magnitude, or quality of the BCG-specific immune response may be altered in infants of M.tb-infected women compared to those without such exposure. Such effects may be manifested as a delayed peak in response to BCG immunisation, or a lower overall recruitment of BCG-specific immune cells in the priming phase, leading to a lower set point of BCG-specific immunological memory, and as differences in the effector response profile.

We will therefore conduct longitudinal studies to examine the timing, magnitude and quality of the initial response to BCG immunisation, and will follow infants to age one year in order to assess the establishment of BCG-specific immunological memory. To investigate the effects of maternal M.tb infection, we will compare these parameters between infants of mothers infected with M.tb, and infants of uninfected mothers, in Uganda. We will also explore mechanisms by which such effects may be mediated by studying differences in dendritic cell profiles, mRNA-Seq profiles and in circulating molecules - such as chemokines and cytokines - that may control the immunological milieu and hence, ultimately, effector cell function. To investigate whether maternal M.tb infection contributes significantly to the association between latitude and response to BCG immunisation we will compare infant responses between Uganda and the UK and assess whether differences are qualitatively similar to those between infants of mothers with and without M.tb infection.

1. The primary, long-term, intended beneficiaries of our research are the populations of TB endemic countries. If our study shows that maternal M.tb infection affects the infant response to BCG the next step will be to investigate the benefits of treatment of latent tuberculosis in young women, before or during pregnancy, for the response to BCG immunisation in their infants. These further studies will be designed to inform local, national and international guidelines on the treatment of M.tb infection (latent or otherwise) in young women before, or during pregnancy, to allow for a better protective vaccine outcome in infants and children in TB endemic settings. This will have a knock-on effect of potentially decreasing the large burden of latent M.tb infection, thereby decreasing the number of latently infected adults and thus driving down the burden of TB. This would have obvious benefits for any endemic country by decreasing morbidity and mortality associated with M.tb infection, and of having healthier adults contributing to the health and wealth of the nation.

2. In addition, there will be particular immediate and long-term beneficiaries in Uganda.
i. The mothers and their infants participating in the study will benefit, during the critical first year of the infant's life, from the excellent, free medical care provided by the research clinic.
ii. The staff of Entebbe Hospital, with whom the research team collaborates, will benefit from on-going training in research methods and good clinical research practice, as well as from salary enhancement in appreciation of their contribution to the work.
iii. The research team. The Co-infection Studies Programme has a strong track record in research capacity building, having developed a research team trained in good clinical research practice now numbering 40 strong, and having contributed to research training for 20 Masters, 12 PhD students and 5 post-doctoral fellows, as well as numerous school-leaver and undergraduate internships. The proposed work will continue to act as a platform for such research capacity development.
iv. Technology and skills transfer. The London team will support the transfer of new techniques to Uganda, particularly the use of Luminex assays, which are just being established in Uganda, thus enhancing capacity in cutting edge immunology research. The analysis of complex immunological data is of increasing importance given availability of techniques, such as Luminex and microarray, which can produce results on hundreds or thousands of outcomes from a small amount of material. Dr Emily Webb will work with statisticians in Uganda on the analysis of Luminex data using principal component analysis. Ugandan trainees will also have the opportunity to learn about the analysis of microarray data through the collaboration with VGTI: this project is one of several currently in progress between VGTI and the Uganda Virus Research Institute and together with other projects, will contribute to the development of bioinformatics expertise at the MRC Uganda Unit.