Recurrent implantation failure (RIF) is usually defined as a woman under 40 who has received at least 4 high-quality embryos in at least three fresh or frozen cycles or has transferred more than 10 high-quality embryos and still cannot obtain a clinical pregnancy (1–3). Recent studies had confirmed that chronic endometritis (CE) was an important cause of recurrent reproductive failure (included recurrent miscarriage and recurrent implantation failure) (4). Chronic endometritis (CE) was a persistent inflammation of the endometrium, and CD138 immunohistochemical (IHC) staining of plasma cells was a more accurate and sensitive diagnostic method for the diagnosis of CE (5–9).

Multiple studies had shown that a good balance of innate and adaptive local and peripheral immune systems was the basis for a successful pregnancy (10, 11). The study by Yuye Li et al. showed that the abundance of endometrial immune cells in CE patients was increased, and the high abundance of immune cells might be related to decreased endometrial receptivity and repeated pregnancy failures (4).

It was now clear that microorganisms affect immunity (12, 13). It also knew that disturbances in the balance between the microbiota and the immune system might lead to inappropriate immune responses or excessive inflammation, or down-regulation of the immune response and pathogenic bacteria’s dominance of normal symbionts “ecological disorders” (14). However, the mechanism of action of endometrial microbiota in RIF patients with CE remains unclear. The purpose of this study was to explore the regulation of endometrial microbes on the endometrial immune cells of RIF patients with CE.

Chronic endometritis (CE) is an important factor for recurrent implantation failure (31). The endometrium of CE patients showed a high abundance of immune cell infiltration, which would cause a decrease in the receptivity of the endometrium, an important feature of recurrent implantation failure (4). It has been proven that immune adaptation, one of the key mechanisms for the establishment and maintenance of pregnancy, was directly affected by local microorganisms (32). In this study, we combined the endometrial transcriptome and microbial diversity analysis to illustrate that endometrial microbiota regulated the abundance of endometrial immune cells through metabolic activities.

The functional enrichment analysis of differential genes in the endometrial transcriptome showed that the functions of DEGs were mainly enriched in immune and inflammation-related items. In the network diagram of enriched pathways, cytokine-cytokine receptor interaction and Toll-like receptor signaling pathway were the central pathways in the network. Toll-like receptors (TLRs) are the first line of defense against pathogen invasion and play a key role in inflammation, immune cell regulation, survival, and proliferation (33). This is the result of the body’s defense during endometritis. At the same time, this pathway and cytokine-cytokine receptor interaction played an important role in the regulation of T cell plasticity and differentiation (33, 34). After calculating the infiltration of endometrial immune cells by CIBERSORT, we found that resting CD4 memory T cells was high infiltration in the non-CE group. This might indicate that in CE, CD4 T cells are significantly activated due to inflammation. This was consistent with the conclusion of another study (4).

It has been proven that bacteria can affect immunity (12, 13). We introduced 16s rRNA sequencing technology to analyze the impact of endometrial microbiota on immune cells. The results of LEfSe showed that the abundance of Phyllobacterium and Sphingomonas in the CE group was significantly higher than that in the non-CE group. In the network diagram of the relationship between different genus and immune cell marker genes, we found that Phyllobacterium and Sphingomonas had a significant positive correlation with B cells. Since the infiltration of plasma cells in the endometrial stroma was considered a characteristic of endometritis (35), it confirmed that these two different genera played an important role in endometritis.

At the same time, we noticed a significant positive correlation between Sphingomonas and uterus NK (uNK) cells. In addition, in the differential analysis of the transcriptome, the expression of CD16 (FcγRIII) in the CE group was significantly higher than that in the non-CE group. CD16+ is a characteristic of the uNK cell population that specifically kills infected cells (36). The abnormal differentiation of uNK cells might lead to embryo implantation failure through the development of endometrial and/or decidual blood vessels and trophoblasts (37). The study of May-Tal Sauerbrun-Cutler et al. showed that a higher level of CD16 + uNK cells was a risk factor for embryo transfer failure in infertile women before frozen embryo transfer (38). Macrophage was known as a common anti-epidemic presenting cell (APC). It was the main cytokine producer in the human endometrium (39) and played an important role in the establishment of endometrial receptivity (40). The results of this study suggested that Phyllobacterium and Sphingomonas were significantly negatively correlated with macrophages. This might be one of the important mechanisms of endometrial microbiota affecting endometrial receptivity.

It was known that metabolic activity had an important regulatory effect on immune cells (41, 42). Endometrial microbiota was metabolically active, and the crosstalk of metabolic activity might be an important way for endometrial microbiota to interact with endometrial immune cells. In the study, two metabolic pathways were highly active in the CE group, PWY-7347, and SUCSYN-PWY. SUCSYN-PWY described the process of generating sucrose by phosphoglycerol and 3-phospho-D-glyceric acid. The intermediate products included UDP-α-D-glucose. PWY-7347 described the process of glucose-6-phosphate and UDP-α-D-glucose to produce sucrose, and phosphate and UDP were produced in the middle. Obviously, these two metabolic pathways were interaction with glucose and lipid metabolism, which had been proved to be closely related to the regulation of T cells and macrophages (41, 42). These active microbiotas competed for energy materials of the host (including endometrial cells and immune cells) by consuming intermediate products of glycolysis.

Th1 is a subtype of CD4 T cells, and its increased abundance was considered to be unfavorable for embryo implantation (43). In this study, we found that endometrial microorganisms with high Th1 abundance exhibited significant-high activity in many important metabolic pathways, including L-aspartic acid and L-asparagine biosynthesis, reducing TCA cycle, degradation of purine nucleobases, and biosynthesis of thiamine diphosphate. Obviously, this was related to multiple links in carbohydrate metabolism (including glycolysis). Therefore, the abnormal activities of the microbiota in the endometrium might regulate the conversion of Th1/Th2 by interfering with the process of carbohydrate metabolism and/or fat metabolism. In addition, Th17 and Treg also played an important role in the regulation of immune adaptation during embryo implantation, and the imbalance of Th17/Treg ratio would have an adverse effect on embryo implantation. In the study of intestinal microbes, the researchers confirmed that lipopolysaccharide (LPS) could induce a decrease in the number of Tregs and an increase in the number of Th17 and Th1 (44). In our study, the endometrial microbiota of patients with high Th17 abundance was significantly active in the relevant pathways of LPS synthesis. It meant that the imbalance in the number and proportion of immune cells in the endometrium may be caused by a similar mechanism.

In this study, we combined the transcriptome and 16s rRNA sequencing technology to analyze the interaction between the endometrium microbial disorder caused by chronic endometritis and the immune cells in the endometrium of patients with recurrent implantation failure. We found that the composition of endometrial microorganisms in CE and non-CE patients were significantly different. Phyllobacterium and Sphingomonas mainly regulated immune cells by interfering with the process of carbohydrate metabolism and/or fat metabolism in the endometrium. The results of this study supplemented part of the mechanism of microbial regulation of the immune environment of the endometrium.

The transcriptome sequencing for 112 endometrium samples have been deposited with the National Center for Biotechnology Information (NCBI) under reference number PRJNA747622. The 16s rRNA gene sequencing for 43 endometrial microbiota samples have been deposited with the NCBI under reference number PRJNA732058.

The studies involving human participants were reviewed and approved by Sixth Affiliated Hospital of Sun Yat-Sen University. The patients/participants provided their written informed consent to participate in this study.

CF, TL, and PeC carried out the study. PeC and TL analyzed and interpreted the data and drafted the manuscript. YG and PaC collected and followed up the samples. TL and CF revised the manuscript. CF, TL, and PeC coordinated the study, participated in the design, and reviewed the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the National Natural Science Foundation of China (grant no. 81871214), the National Key R&D Program of China (grant no. 2017YFC1001603), and the National Natural Science Foundation of China (grant no. 81070495).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.