# Contribution of endometrial microbiome to inflammation-mediated infertility in women undergoing ART

## TL;DR

This study shows that the endometrial microbiome in women who fail to get pregnant after ART is more diverse and less healthy, and that the metabolite butyrate can harm the endometrial lining and increase inflammation.

## Contribution

The study is the first to show that microbial metabolites like butyrate can directly affect endometrial cell function and receptivity in the context of ART outcomes.

## Key findings

- Women who failed ART had a more diverse endometrial microbiome with fewer lactobacilli and more pathogenic species like Prevotella and Corynebacterium.
- Butyrate, a microbial metabolite, increased inflammation and reduced endometrial epithelial barrier function in vitro.
- Negative ART outcomes were linked to higher microbial diversity and increased expression of receptivity and decidualization markers in endometrial tissue.

## Abstract

Is the endometrial microbiome altered in women who fail to get pregnant after ART and do microbial-derived metabolites influence endometrial cellular mechanisms important for embryo implantation?

The endometrial microbiome in women who fail to get pregnant after ART is more diverse and has fewer lactobacilli species than the endometrial microbiomes of women who become pregnant; the short-chain fatty acid butyrate, a common metabolite found in the presence of increased microbial diversity, diminishes endometrial epithelial barrier function and increases the expression of inflammatory markers.

Shifts in the endometrial microbial community structure have been linked to fertility and pregnancy complications although the underlying mechanisms are poorly understood. Microbial metabolites at other mucosal surfaces, such as the gut, act as important modulators of immune and barrier function, particularly in epithelial cells. Effects of changes in local bacterial microbial populations on fertility, and how their metabolites might influence endometrial cell function have not been explored.

In this prospective longitudinal study of ART outcomes, 29 nulliparous women with unexplained infertility were recruited between October 2016 and February 2018. Endometrial tissue samples were taken for microbiome analysis and endometrial transcriptomics prior to ART. For primary cell culture studies, endometrial biopsies were obtained from fertile women of reproductive age undergoing laparoscopic surgical investigation between February 2021 and September 2023. In vitro models of implantation were established using endometrial cell lines and primary endometrial stromal cells.

Microbiome 16S sequencing analysis was performed on bacterial DNA isolated from endometrial biopsies and correlated with receptivity markers. Endometrial RNA sequencing data from women undergoing ART were used to analyse differential gene expression of receptivity and decidualization markers in women who had a positive or negative ART cycle outcome. In vitro models, using both established endometrial cell lines and primary human endometrial epithelial cells and stromal cells, were developed to investigate the effects of microbial-derived metabolites. An in vitro model of peri-implantation was used to test the effect of butyrate on endometrial epithelial receptivity and stromal cell decidualization.

Endometrial microbiome 16S sequencing revealed a lower abundance of Lactobacillus spp. and significantly higher abundance of pathogenic species such as Prevotella spp. and Corynebacterium spp. in women who did not become pregnant after ART. Endometrial microbiota from women who had positive ART outcomes showed significantly lower diversity indices. Intriguingly, analysis of endometrial RNA sequencing data from women with unexplained infertility undergoing ART showed that negative ART outcomes were associated with higher levels of some receptivity and decidualization markers in their endometrial tissue. Butyrate, but not lactate or acetate, also increased some markers of epithelial receptivity and stromal decidualization. Butyrate exposure also activated defence mechanisms in cultured endometrial epithelial cells by inducing expression of antimicrobial peptide(s) and inflammation markers, as well as impairing the barrier integrity of endometrial epithelial cell monolayers.

The RNA-seq data used for the study can be found in GEO database, GEO ID GSE144895. The data for the 16S sequencing can be accessed in SRA BioProject number PRJNA1338067.

Limitations of our study include the cohort size and technical challenges that precluded absolute butyrate measurement in endometrial tissue biopsies. Biopsy collection from women undergoing gynaecological investigation varied in menstrual cycle staging and fertility diagnoses, which may contribute to the variability between responses obtained from in vitro stimulations. The transferred embryos were not genetically tested, but were all of good or top quality.

Our findings indicate that the endometrial microbiome is altered in women who fail to become pregnant after ART, and that the microbial-derived metabolite butyrate can induce inflammation and impair endometrial epithelial barrier function and drive increased gene expression levels of markers for epithelial receptivity and stromal decidualization in in vitro models of peri-implantation. Endometrial microbial dysbiosis and higher expression of receptivity markers were found in women who failed to establish pregnancy post-ART. Negative ART outcomes in this cohort were found to correlate with the presence of a wider, more diverse microbial community that includes Prevotella spp., which is among the butyrate-producing bacteria. Further investigation of the microbial metabolome in healthy endometrium would help clarify the physiological role of butyrate and other bacterial metabolites in endometrial function.

This research was supported by the Grant for Fertility Innovation from Merck KGaA, grant award number 15692. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

N/A.