Secretor status is a modifier of vaginal microbiota-associated preterm birth risk

Abstract

Mutations in the FUT2 gene that result in a lack of expression of histo-blood group antigens on secreted glycoproteins may shape the vaginal microbiota with consequences for birth outcome. To test this, we analysed the relationship between secretor status, vaginal microbiota and gestational length in an ethnically diverse cohort of 302 pregnant women, including 82 who delivered preterm. Lactobacillus gasseri and L. jensenii were found to have distinct co-occurrence patterns with other microbial taxa in non-secretors. Moreover, non-secretors with Lactobacillus spp. depleted high diversity vaginal microbiota in early pregnancy had significantly shorter gestational length than Lactobacillus spp. dominated non-secretors (mean of 241.54 days (sd=47.14) versus 266.21 (23.61); P-value=0.0251). Similar gestational length differences were observed between non-secretors with high vaginal diversity and secretors with Lactobacillus spp. dominance (mean of 262.52 days (SD=27.73); p-value=0.0439) or depletion (mean of 266.05 days (SD=20.81); p-value=0.0312). Our data highlight secretor status and blood-group antigen expression as being important mediators of vaginal microbiota-host interactions in the context of preterm birth risk.

Keywords: Lactobacillus; blood-type antigens; preterm birth; secretor status; vaginal microbiome.

Fig. 1.. Vaginal microbiota composition in secretors and non-secretors.(a) Distribution of the top 10 most abundant bacterial genera (in descending alphabetical order) in the vaginal microbiota of secretors and non-secretors in early, mid and late pregnancy. (b) The distribution of the proportion of Lactobacillus in secretors and non-secretors in early, mid and late pregnancy was similar. (c). Frequencies of each CST through pregnancy in secretors and non-secretors.

Fig. 2.. Co-occurrence of vaginal microbiota in early pregnancy in the context of secretor status. (a) Co-occurrence networks of early pregnancy vaginal microbiota in secretors and non-secretors. Nodes represent bacterial taxa detected at >0.5% relative abundance in two or more samples and have been grouped into L. crispatus, L. gasseri, L. iners, L. jensenii, other Lactobacillus, BV-associated bacteria, pathobionts and others, as defined previously. Statistically significant edges, indicating correlations, are coloured and weighted according to their r value: red (negative correlation); green (positive correlation). In both secretors and non-secretors, many positive associations were observed between BV-associated microbes. (b) Star diagrams (derived from the network analyses) centred on the L. crispatus, L. iners and G. vaginalis nodes indicate Lactobacillus species, especially L. crispatus, tend to be more negatively correlated with other bacteria in the vaginal microbiota of both secretors and non-secretors, compared with G. vaginalis and other BV-associated bacteria.

Fig. 3.. The association between secretor status, vaginal microbiota and gestational length (days). A red dashed line highlights the 37 weeks (259 days) threshold for prematurity. (a) The distribution of gestational length by secretor status and vaginal microbiome shows that non-secretors with Lactobacillus-depleted microbiomes in early pregnancy tend to have shorter gestation, but this relationship disappears by mid-pregnancy. Sample sizes are presented at the top of plot. (b) Classifying the microbiome by CST rather than by Lactobacillus status indicates that reduced gestation in Lactobacillus-depleted non-secretors is mainly driven by women with CST 3 and high diversity CST 4 microbiomes. Sample sizes are presented in Table S3.