AbstractSpontaneous preterm birth (sPTB) is a significant cause of childhood mortality worldwide. Preterm birth is a syndrome with many risk factors including dysbiosis of the vaginal microbiota- bacterial vaginosis (BV). BV is the replacement of beneficial Lactobacillus species with a diverse range of anaerobic bacteria. The change in microbiota profile leads to a shift in the metabolic profile and an increase of vaginal pH, with anti-inflammatory lactate replaced with pro-inflammatory acetate. Integration of microbiota and metabolome data improves prediction of sPTB. Although the same species which are associated with BV are often found in women who deliver preterm, relating the influence of specific species to sPTB is complicated. Bacteria associated with BV (BVAB) can also be found in a eubiotic microbiota. Notably Gardnerella vaginalis, which is found in nearly all BV cases, is also present in many otherwise eubiotic microbiotas. Patterns of co-occurrence have been identified for BV associated bacteria in sPTB but the impact of strain level differences on these interactions is unclear. Further research is therefore required to understand how the different species contribute to the dysbiotic environment either alone or in combinations.
To first understand the impact of BV on the vaginal environment, NMR metabolomics was used to generate profiles of important vaginal bacteria in brain heart infusion (BHI)media. Seven strains of G. vaginalis were separated based on fermentation strategy which affected the ratio of acetate and lactate produced. Inter- and intra-strain variation in metabolic profile was also identified for Lactobacillus species relating to acetate and succinate production as well as arginine metabolism.
To confirm whether these metabolic profiles are physiologically relevant, the metabolism of both BVAB and Lactobacillus species was then characterised in a chemically defined medium to mimic the vaginal environment. This media contained both glycogen and glucose as fermentable substrates but, inter alia, lacks pyruvate, uracil and arginine which were important in supporting the growth of both BVAB and key Lactobacillus strains. Glycogen is a major carbon source in vaginal fluid but is absent from conventional culture media. In this media, glycogen breakdown was observed for three representative isolates of Lactobacillus in addition to all the anaerobic pathogens, this was detected through an increase in maltose and glucose in the spent culture media. However, the overall growth of pathogenic bacteria was poor compared to BHI mediumand, while future research can be envisaged to develop physiologically relevant media that universally supports growth of BVAB and Lactobacillus species, co-culture experiments should be conducted in BHI for now.
To determine how species co-occurrence can affect the vaginal metabolome coculture studies were therefore carried out in BHI. A mutualistic relationship was identified between G. vaginalis and P. bivia. The fermentation strategy did not affect whether there was a positive interaction between the two species, however only G. vaginalis strains belonging to a subgroup of Gardnerella showed a positive interaction. There were also strain-level differences in the metabolic output of coculture which wasdependent on fermentation strategy, leading to an increase in either acetate or aspartate, both of which are important predictors of sPTB. A commensal relationship was also identified between P. bivia and Pe. anaerobius which led to a significant increase in acetate and a decrease in lactate.
Taken together, this thesis established the importance of studying intra- and inter-strain variation both in relation to metabolism and interaction with other species in the vaginal microbiota and may help a better understanding of the association between BV and sPTB to develop.
|Date of Award
|1 Aug 2022
|James Mason (Supervisor), Rachel Tribe (Supervisor) & John Sutton (Supervisor)