P-333 Combining deep shotgun metagenomic sequencing and culturomics for in-depth characterization of a potential endometrial microbiome

Abstract Study question Does the endometrium host a microbiome, and if so, are these taxa distinct from the vaginal/cervical microbiome? Summary answer Multifaceted analysis of endometrial tissue reveals no distinct microbial taxa compared to the vaginal/cervical samples, questioning whether detected microbiota are endometrium-related or result from cross-contamination. What is known already Recent studies suggest a role for the endometrial microbiota in reproductive health, yet the existence of an endometrial microbiome and its specific bacterial taxa remain enigmatic. The female reproductive tract, with a bacterial abundance gradient, is predominantly Lactobacillus-dominated in the lower region. Inconsistent findings on Lactobacillus dominance in endometrial samples are often ascribed to cross-contamination and methodological constraints, highlighting the need for precise methodologies in studying the low-biomass endometrial environment. Study design, size, duration Fifteen women, undergoing a diagnostic hysteroscopy and endometrial biopsy, as part of the subfertility work-up, were included in this study. Microbiota profiles of endometrial biopsies were compared to paired vaginal and cervical swabs from nine patients using 16S amplicon sequencing. Furthermore, the microbial composition of six additional endometrium biopsies was characterized using both a culturomics-based approach and deep metagenomic sequencing. Participants/materials, setting, methods Endometrial biopsies (n = 15) were taken with a sterile Pipelle de Cornier and snap-frozen in liquid nitrogen, followed by -80 °C storage and 16S amplicon sequencing (n = 9) targeting the V4-region or deep shotgun metagenomic sequencing (n = 6) (Prebiomics, Italy). Simultaneously, in nine women, paired cervical and vaginal swabs (Copan Diagnostics, Italy) were taken and preserved in eNAT medium. In six participants, an additional endometrial biopsy was collected and stored in Amies medium for immediate culturomics. Main results and the role of chance We observed a high similarity in the microbial composition of endometrial biopsies, cervical, and vaginal samples using 16 amplicon sequencing, but also some taxa were only found in the endometrial samples. Notably, the presence of Gemmobacter, a waterborne genus, was only found in endometrial biopsies and could be a source of contamination from water used during hysteroscopy to flush the uterus. Most endometrial microbiome profiles were Lactobacillus dominated, with exceptions observed in three patients. No significant differences for alpha diversity measures (richness and evenness) of the microbiota composition were observed between the different niches. For the first time we applied both culturomics and deep shotgun metagenomic sequencing for the characterization of the microbial composition in the same endometrial biopsies. Commonly identified genera using both techniques include Lactobacillus, Prevotella, Bifidobacterium, Anaerococcus, and Cuticubacterium. Remarkably, at the species level, the two approaches revealed almost no overlapping species within each biopsy, underscoring the importance of combining multiple methodologies in examining low-biomass niches like the endometrium. Limitations, reasons for caution A limitation of this study is the low-biomass nature of endometrial biopsies which are prone to cross-contamination with the lower female reproductive tract. This challenges downstream analysis, as it becomes difficult to distinguish between a relevant microbiological signal and contaminating noise. Wider implications of the findings This study emphasizes the need to consider a multifaceted microbiome approach when exploring low-biomass environments, like the endometrium. The critical consideration of potential cross-contamination by the lower female reproductive tract and its implications for downstream analysis is paramount in clinical microbiology research and demands meticulous handling. Trial registration number not applicable