Vaginal metatranscriptome meta-analysis reveals functional BV subgroups and novel colonisation strategies

The application of '-omics' technologies to study bacterial vaginosis (BV) has uncovered vast differences in composition and scale between the vaginal microbiomes of healthy and BV patients. Compared to amplicon sequencing and shotgun metagenomic approaches focusing on a single or few species, investigating the transcriptome of the vaginal microbiome at a system-wide level can provide insight into the functions which are actively expressed and differential between states of health and disease. We conducted a meta-analysis of vaginal metatranscriptomes from three studies, split into exploratory (n = 42) and validation (n = 297) datasets, accounting for the compositional nature of sequencing data and differences in scale between healthy and BV microbiomes. Conducting differential expression analyses on the exploratory dataset, we identified a multitude of strategies employed by microbes associated with states of health and BV to evade host cationic antimicrobial peptides (CAMPs); putative mechanisms used by BV-associated species to resist and counteract the low vaginal pH; and potential approaches to disrupt vaginal epithelial integrity so as to establish sites for adherence and biofilm formation. Moreover, we identified several distinct functional subgroups within the BV population, distinguished by genes involved in motility, chemotaxis, biofilm formation and co-factor biosynthesis. After defining molecular states of health and BV in the validation dataset using KEGG orthology terms rather than community state types, differential expression analysis confirmed earlier observations regarding CAMP resistance and compromising epithelial barrier integrity in healthy and BV microbiomes and also supported the existence of motile vs. non-motile subgroups in the BV population. These findings were independent of the enzyme classification system used (KEGG or EggNOG). Our findings highlight a need to focus on functional rather than taxonomic differences when considering the role of microbiomes in disease and identify pathways for further research as potential BV treatment targets.