Comparative analysis of vaginal microbiota sampling using menstrual cups and high vaginal swabs pregnant women living with HIV-1 infection

Background: Menstrual cups are increasingly used to collect cervicovaginal secretions to characterise vaginal mucosal immunology, in conjunction with high vaginal swabs (HVS) for metataxonomics, particularly in HIV transmission studies. We hypothesised that both methods of collecting bacterial biomass are equivalent for 16S rRNA gene amplicon sequencing. Material and Methods: Cervicovaginal fluid (CVF) samples from 16 HIV-1 infected pregnant women were included to represent the major vaginal bacterial community state types (CST I-V). Women underwent sampling during the second trimester by liquid amies HVS followed by a menstrual cup (Soft disc) for a minimum of five minutes. Samples were immediately transferred to the laboratory and frozen at -80°C. Bacterial cell pellets obtained from swab elution and cup were resuspended for DNA extraction. Bacterial 16S RNA amplicon sequencing was performed using V1-V2 primers. 16S rRNA gene sequences were analysed using MOTHUR. Paired total DNA, bacterial load, amplicon read counts, diversity matrices and bacterial taxa were compared by sampling method using MicrobiomeAnalyst, SPSS and R. Results: The total DNA eluted from one aliquot of diluted CVF from a menstrual cup was similar to that of a HVS swab (993ng and 609ng, p=0.18); the mean bacterial loads were also comparable for both methods (cup: 6.3 log10 16S rRNA gene copies versus HVS: 6.2 log10 16S rRNA gene copies, p=0.27). The mean number of sequence reads generated from cup samples was lower than from HVS (cup: 12730; HVS:14830, p=0.05). The alpha diversity metrices were similar for both techniques; cup Species Observed: 41 (range 12-96) versus HVS:47 (range 16-96), p=0.15; cup Inverse Simpson Index:1.98 (range 1.0-4.0) versus HVS: 0.48 (range 1.0-4.4), p= 0.22). The three most abundance species observed in the dataset were: Lactobacillus iners, Lactobacillus crispatus and Gardnerella vaginalis. Hierarchical clustering of relative abundance data showed that samples obtained from the same individual via different methods clustered together in the same CST group. Conclusions: These data demonstrate that despite sampling slightly different areas of the lower genital tract, menstrual cup and HVS samples are both suitable for assessing vaginal microbiota composition and diversity in HIV-1 pregnant individuals. The menstrual cup offers some advantages, including a higher total volume of sample available for DNA extraction and complimentary assays.