Microbiota insights in endometriosis

The study of the gut microbiota has emerged as a pivotal area of research for elucidating the pathophysiology of endometriosis for several compelling reasons. Firstly, the digestive microbiota represents the most abundant microbial community in the human body, and these symbionts express approximately 10 times more unique genes than the host's genome, underscoring their substantial genetic diversity and potential impact [ 29 ]. Secondly, the field of immunology has experienced a paradigm shift with the recognition of the pivotal role played by the microbiota in shaping and modulating the induction, education, and functioning of the immune system [ 30 ]. Thirdly, individuals afflicted with endometriosis exhibit a threefold higher incidence of intestinal functional disorders, which have been linked to intestinal dysbiosis [ 31 , 32 ]. These converging factors highlight the significance of investigation of the interplay between the gut microbiota and endometriosis. Human studies focusing on the gut microbiota in endometriosis-affected patients compared to controls have also revealed a wide discrepancy in results, summarized in Table  1 [ 40 , 41 ]. Patients with stage III or IV (moderate to severe) endometriosis ( n  = 14) were more likely to have abundant Shigella and Escherichia than healthy participants ( n  = 14) [ 33 ]. An observational cross-sectional pilot study to characterize the gut microbiome profiles among endometriosis patients ( n  = 35) and healthy controls ( n  = 24) did not detect significant differences between groups [ 34 ]. The microbiota is often analyzed using α-diversity (representing the bacterial richness of an individual) and β-diversity (bacterial diversity between two groups) or using ratios such as Firmicutes/Bacteroidetes, which has an important influence in maintaining normal intestinal homeostasis [ 42 ]. A large study found that β-diversity was different in healthy controls ( n  = 198) compared to patients with endometriosis ( n  = 66), with a lower α-diversity in endometriosis patients [ 37 ]. Authors reported the abundance of 12 bacteria belonging to the classes Bacilli, Bacteroidia, Clostridia, Coriobacteriia, and Gammaproteobacteria that differed significantly between stool samples from endometriosis patients ( n  = 66) and those from matched healthy controls ( n  = 198).Two bacteria from the class Bacteroidia ( Bacteroides and Parabacteroides ) and two belonging to the class Clostridia ( Oscillospira and Coprococcus ) were present in higher abundances in endometriosis patients, whereas two bacterial species from the classes Bacteroidia ( Paraprevotella and one unidentified) and Clostridia ( Lachnospira and one unidentified) were present at lower abundances compared to in individuals without endometriosis [ 37 ]. On the other hand, another study found that patients with stage III or IV endometriosis ( n  = 12) had a higher Firmicutes/Bacteroidetes ratio than the healthy controls, the two groups nevertheless exhibited differences in β-diversity [ 35 ]. A recent study compared patients with stage I-II versus III-IV, showing a significant difference in the composition of the gut microbiota. In addition, the digestive microbiota of the most severe patients (stage III-IV with dysmenorrhoea) had a significantly different composition [ 39 ]. The largest study to date by Pérez Prieto et al., comparing 136 women with endometriosis to 864 controls, found no significant differences between groups. However, it is important to note that the control group lacked radiological or surgical confirmation to rule out endometriosis [ 38 ]. Variability in the findings across these studies points to the need for larger studies in which potential confounders (e.g., age, race/ethnicity, other health conditions, medication, and diet) are taken into account. In addition, while the changes in microbiota may be due to endometriosis, they may also be due independently to the inflammation it triggers. These different studies allow us to explore a phenomenon of association without being able to distinguish whether it is the endometriosis itself or the inflammation or dysimmunity that it causes. Table 1 Gut microbiota modification in human endometriosis Study Material and Methods Design Results (EMS vs controls) Ata et al. 2019 [ 33 ] Human Feces RNA 16S V3-V4 14 EMS vs. 14 controls ↑ Proteobacteria (BGN): Escherichia, Shigella Perrota et al. 2020 [ 34 ] Human Feces RNA 16S V4 21 EMS stage I-II vs. 14 EMS stage III-IV Similar α and β-diversity Shan et al. 2021 [ 35 ] Human Feces RNA 16S V3-V4 12 EMS stage III-IV vs. 12 controls Different relative abundance (PCoA) ↑ Firmicutes/Bacteroidetes ratio ↑ Actinobacteria, Saccharibacteria, Acidobacteria , Cyanobacteria, Fusobacteria and Prevotella Huang et al. 2021 [ 36 ] Human Feces RNA 16S V4 21 EMS vs. 20 controls Different relative abundance (PCoA) ↓ α-diversity ↑ Eggerthella lenta, Eubacterium dolichum ↓ Clostridia Clostridiales, Lachnospiraceae Ruminococcus, Clostridiales Lachnospiraceae, Ruminococcaceae Ruminococcus Svensson et al. 2021 [ 37 ] Human Feces RNA 16S V1-V3 66 EMS vs. 198 controls Different relative abundance (PCoA) ↓ α-diversity ↑ Clostridia ↓ Bacteroidia, Coriobacteriia ↑ Gammaproteobacter, Bacilli Huang et al. 2021 [ 36 ] Human Feces RNA 16S V3-V4 18 EMS vs. 18 controls Different relative abundance (PCoA) ↓ α-diversity ↑ Proteobacteria ↓ Bacteroidota and Firmicutes Pérez-Prieto et al. 2024 [ 38 ] Human Feces DNA Shotgun 136 EMS vs. 864 controls Similar α and β-diversity Cai et al. 2025 [ 39 ] Human Feces RNA 16S V3-V4 39 EMS stage I-II vs. 36 EMS stage III-IV Different relative abundance (PCoA) Similar α-diversity Different relative abundance in subgroupe analysis of patients with dysmenorrhea Abbreviations: D day, EMS endometriosis, PCoA Principal Coordinates Analysis Gut microbiota modification in human endometriosis Human Feces RNA 16S V3-V4 Human Feces RNA 16S V4 Human Feces RNA 16S V3-V4 Different relative abundance (PCoA) ↑ Firmicutes/Bacteroidetes ratio ↑ Actinobacteria, Saccharibacteria, Acidobacteria , Cyanobacteria, Fusobacteria and Prevotella Human Feces RNA 16S V4 Different relative abundance (PCoA) ↓ α-diversity ↑ Eggerthella lenta, Eubacterium dolichum ↓ Clostridia Clostridiales, Lachnospiraceae Ruminococcus, Clostridiales Lachnospiraceae, Ruminococcaceae Ruminococcus Human Feces RNA 16S V1-V3 Different relative abundance (PCoA) ↓ α-diversity ↑ Clostridia ↓ Bacteroidia, Coriobacteriia ↑ Gammaproteobacter, Bacilli Human Feces RNA 16S V3-V4 Different relative abundance (PCoA) ↓ α-diversity ↑ Proteobacteria ↓ Bacteroidota and Firmicutes Human Feces DNA Shotgun Human Feces RNA 16S V3-V4 Different relative abundance (PCoA) Similar α-diversity Different relative abundance in subgroupe analysis of patients with dysmenorrhea Abbreviations: D day, EMS endometriosis, PCoA Principal Coordinates Analysis Another interesting yet overlooked aspect of clinical studies concerns the digestive symptoms reported by patients. It is estimated that more than a half of individuals with endometriosis experience debilitating digestive symptoms, including bowel movement disorders, digestive pain, and bloating [ 43 ]. The main hypothesis generally put forward is the presence of digestive tract lesions responsible for these symptoms. However, in most cases, these signs occur in the absence of any visible digestive lesions [ 44 ]. This suggests that alterations in the gut microbiota—despite their variability—may be linked to these disorders. To date, no study has specifically explored the relationship between digestive symptoms and microbiota variation. Further studies are therefore needed and should be conducted longitudinally across the menstrual cycle to account for the known exacerbation of digestive symptoms in the premenstrual phase. It has indeed been shown that both gut microbiota composition and intestinal gas production are influenced by hormonal fluctuations—particularly those associated with the menstrual cycle and the use of oral contraceptives, the latter being especially relevant in endometriosis due to their widespread therapeutic use [ 45 ]. The main advantage of studying the intestinal microbiota in animal models lies in the strict control of the various parameters that can modify the microbiota, of which diet is the most important. In mouse models, studies of the potential role played by the intestinal microbiota in endometriosis have yielded heterogenous results, summarized in Table  2 . Feces samples were found with lower gut α and β diversities and abundances in mice with endometriosis compared to those of control mice [ 54 ]. Another study conducted on feces reported no difference in the α- and β-diversities between mice with and without endometriosis [ 48 ]. However, it should be noted that their experiment ended 21 days after the surgical induction of endometriosis and that the feces samples were taken from the cage and not by surgical means from the colon [ 48 ]. In a third study, dysbiosis of the gut microbiota was observed 42 days after endometriosis induction, with an elevated Firmicutes/Bacteroidetes ratio and an elevated abundance of Bifidobacterium [ 47 ]. The same results were observed in another study of mice [ 47 ], while an elevated Firmicutes/Bacteroidetes ratio and decreased abundance of Ruminococcaceae were observed in feces samples from female rats with endometriosis [ 50 ]. Three studies found higher gut microbiota diversity and an increased Firmicutes/Bacteroidetes ratio [ 35 , 46 , 47 , 49 ], which is a marker of dysbiosis [ 55 ]. These data support the presupposed interrelation between endometriosis and the microbiota, but they also raise two critical questions: the impact of inflammation related to endometriosis on the microbiota, and the impact of microbiota on the endometriosis pathogenesis. Table 2 Gut microbiota modification in animal endometriosis Study Material et Methods Design Results (EMS vs controls) Bailey et al. 2002 [ 46 ] Rhesus monkey Selective and differential agars 8 EMS vs. 10 controls ↓ Lactobacillus species ↑ Aerobic and gram negative bacteria Yuan et al. 2018 [ 47 ] Mice Feces D7, D14, D28, D42 ARN 16S V4 22 EMS vs. 20 controls Different relative abundance (PCoA) ↑ Firmicutes/Bacteroidetes ratio Hantschel et al. 2019 [ 48 ] Mice Feces D21 ARN 16S V4-V5 8 EMS vs. 8 controls Similar relative abundance ↑ Bacteroidales, Lactobacillus species, Prevotellaceae and Lachnospiraceae Chadchan et al. 2019 [ 49 ] Mice Feces D21 ARN 16S 5 EMS vs. 5 controls Different relative abundance (PCoA) ↑ α-diversity ↓ Firmicutes/Bacteroidetes ratio Cao et al. 2020 [ 50 ] Mice Feces D56 ARN 16S V3-V4 8 EMS vs. 8 controls Different relative abundance (PCoA) ↓ α-diversity ↓ Ruminococcaceae Ni et al. 2020 [ 51 ] Mice Feces D21 ARN 16S V3-V4 6 EMS vs. 6 controls Different relative abundance (PCoA) ↓ α-diversity ↑ Proteobacteria and Verrucomicrobia ↓ Firmicutes/Bacteroides ratio Parpex et al. 2024 [ 52 , 53 ] Mice Feces W7 ARN 16S V3-V4 30 EMS vs. 28 controls Different relative abundance (PCoA) Similar α-diversity and Firmicutes/Bacteroides ratio ↓ Akkermansia  Abbrevations: D  day, W  week, EMS  endometriosis, PCoA  Principal Coordinates Analysis Gut microbiota modification in animal endometriosis Rhesus monkey Selective and differential agars ↓ Lactobacillus species ↑ Aerobic and gram negative bacteria Mice Feces D7, D14, D28, D42 ARN 16S V4 Different relative abundance (PCoA) ↑ Firmicutes/Bacteroidetes ratio Mice Feces D21 ARN 16S V4-V5 Similar relative abundance ↑ Bacteroidales, Lactobacillus species, Prevotellaceae and Lachnospiraceae Mice Feces D21 ARN 16S Different relative abundance (PCoA) ↑ α-diversity ↓ Firmicutes/Bacteroidetes ratio Mice Feces D56 ARN 16S V3-V4 Different relative abundance (PCoA) ↓ α-diversity ↓ Ruminococcaceae Mice Feces D21 ARN 16S V3-V4 Different relative abundance (PCoA) ↓ α-diversity ↑ Proteobacteria and Verrucomicrobia ↓ Firmicutes/Bacteroides ratio Mice Feces W7 ARN 16S V3-V4 Different relative abundance (PCoA) Similar α-diversity and Firmicutes/Bacteroides ratio ↓ Akkermansia Abbrevations: D  day, W  week, EMS  endometriosis, PCoA  Principal Coordinates Analysis In the study of the microbiota—particularly the gut microbiota—in endometriosis, a recurring question is whether endometriosis alters the microbiota or whether a pre-existing dysbiosis contributes to the development of the disease. Several experimental mouse studies support the latter hypothesis, showing that the microbiota may participate in the initiation or progression of endometriosis. For example, fecal microbiota transfer (FMT) from mice with endometriosis to healthy mice resulted in larger lesions compared to transfers from healthy donors [ 56 ]. Similarly, induction of dysbiosis through a Western diet also worsened lesion severity [ 52 ]. These findings support the emerging view of the microbiota as a functional "last organ" [ 57 ] and suggest that it may contribute to the pathogenesis of endometriosis—an argument that further supports its classification as a systemic disease. It should be noted that endometriosis most commonly originates from retrograde menstruation, and that the microbiota likely serves as a co-factor in lesion development and symptom expression, rather than as a primary cause. On the other hand, endometriosis has been shown to alter the gut microbiota in both human and mouse models—suggesting a potential reverse causality. Dysbiosis may help explain the high prevalence of digestive symptoms in affected patients, as these symptoms could arise secondarily from microbiota alterations triggered by the onset of endometriosis. The use of probiotics, and their effects on gut microbiota composition and clinical symptoms, warrant further investigation in this context. The gut microbiota produces an extremely diverse metabolite repertoire by anaerobic fermentation of dietary components [ 58 , 59 ]. The main metabolic end products are SCFAs, including butyrate, acetic acid, butyric acid, and propionic acid. These SCFAs interact with host cells via epithelial cells to influence the immune response. They induce the overexpression of Treg lymphocytes and myeloid cells [ 60 ]. The immune responses mediated by SCFAs consist of the regulation of Treg inhibition, histone-deacetylases, and B cells by regulation of the mitogen-activated protein kinases (MAPK) and mammalian target of rapamycin (mTOR) pathways [ 61 ]. SCFAs have been shown to be involved in various pathologies, exerting either a deleterious or protective effect [ 5 , 62 , 63 ]. SCFAs have been described as activators of the pro-inflammatory pathway as NOD-like receptor family, pyrin domain containing 3 (NLRP3), and Nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB), and the production of interleukins (IL)−1β and IL-18 [ 64 , 65 ]. Other SCFA with anti-inflammatory properties, Such as n-butyrate, are decreased in endometriosis, as reported by Chadchan et al. In a mouse model of endometriosis the authors demonstrated that 1) endometriosis alters the gut microbiome, resulting in reduced production of the n-butyrate, and 2) n-butyrate, but not acetate or propionate, inhibits endometriotic lesion growth [ 66 ]. Once translocated into the systemic circulation, these molecules can be further recognized by pattern recognition receptors (PRR) including Toll-like receptors (TLR) and NOD-like receptors (NLRs), which are able to recognize certain patterns of microbial molecules and activate innate immune pathways, then modifying the growth of lesions [ 67 ]. Dysbiosis of the gut microbiota leads to reduced butyrate production and impaired metabolism, as seen in inflammatory bowel disease [ 68 ]. While butyrate is present in the diet, its levels may be insufficient to counteract inflammation in the gut. Oral butyrate Supplementation has shown dose-dependent anti-inflammatory effects in both preclinical and clinical studies, with significant benefits observed at doses of 100 mg/kg/day [ 69 ]. Although butyrate is rapidly absorbed in the duodenum, its use as a dietary supplement—assuming it can reach the colon—represents an interesting therapeutic avenue. A study investigating the correlation between fecal metabolomics and gut microbiota in a mouse model of endometriosis revealed significant alterations in bile acid and fatty acid profiles, including an increase alpha-linolenic acid levels [ 51 , 70 ]. In a subsequent study by the same group, alpha-linolenic acid supplementation was shown to attenuate the inflammatory response by reducing nitrite and prostaglandin E2 accumulation, key mediators of inflammation [ 54 ]. Furthermore, alpha-linolenic acid has been shown to suppress the pro-inflammatory activity of M1 macrophages, notably by reducing the secretion of IL-1β and IL-6 [ 71 ]. It also contributes to improving the peritoneal inflammatory environment in endometriosis by decreasing lipopolysaccharide (LPS) levels [ 54 ]. Other gut microbiota–derived SCFAs and related metabolites have also been implicated in endometriosis. Recent research explored the combined gut microbiota and metabolomic profiles in stool samples from women with endometriosis, identifying novel bioactive metabolites with therapeutic potential [ 72 ]. Among them, 4-hydroxyindole, a compound derived from bacterial tryptophan metabolism, emerged as a promising candidate. This metabolite was shown to prevent the formation of endometriotic lesions, promote lesion regression, and significantly reduce pain in experimental models.

The cervix acts as a crucial barrier, exhibiting a distinct microbiota composition compared with the endometrium. These differences suggest that the cervix plays a pivotal role in bacterial clearance within the female reproductive tract. Cervicovaginal mucus, an essential component of this barrier, is a complex secretion composed of water, electrolytes, lipids, and proteins. During the ovulation phase, mucus selectively restricts sperm transport [ 93 ] and prevents the colonization of unwanted vaginal microorganisms [ 94 ]. Cervicovaginal mucus contains many immunomodulatory and antimicrobial molecules, such as IgA, mucins, defensins, and lysozyme [ 95 – 98 ]. Mucus appears to be a relevant area to study alongside microbiota, as it is in direct contact with bacteria that use the sugars attached to mucins as an energy reserve. Functionally, two major types of mucins can be distinguished: transmembrane mucins and gel-forming mucins [ 99 ]. By binding water with their O-linked glycans, mucins impart gel-like properties to mucus. Gel-forming mucins are the backbone of mucus. Glycosylation of mucins has a major impact on their function, affecting their structure [ 100 ]. Interestingly, it has been shown that the hyposialylation pattern of endometriotic cells appears to be associated with the enhancement of their migratory abilities in case of endometriosis, a mechanism that could explain the invasion of lesions [ 101 ]. Despite the lack of systematic mapping of endometrial mucins, their close association with the microbiota, as observed in the intestinal barrier, is evident. In vitro fermentation models have demonstrated that the addition of mucins leads to altered bacterial profiles, favoring mucin-degrading bacteria (Bacteroidetes, Akkermansia, and Lachnospiraceae species) while decreasing the abundance of Lactobacillus species and Bifidobacterium [ 102 ]. Studies have identified the expression of several mucins, including MUC1 (with the known CA19-9 epitope) [ 103 ], MUC2, MUC4, MUC5AC, MUC6 [ 104 ], and MUC16 (with the known CA125 epitope) [ 105 ], in the endometrial epithelium. However, investigations of mucin polymorphisms in endometriosis have primarily focused on MUC2 and MUC4, linking specific single-nucleotide polymorphisms (SNPs) to the development of endometriosis and associated infertility [ 106 , 107 ]. Functional studies regarding the involvement of mucins in endometriosis and the link with the endometrial or vaginal microbiota are lacking, however. A large number of studies have suggested a degree of crosstalk between the gut microbiota and the immune system [ 30 , 108 ]. This dialogue begins at the tissular level, with the combined action of bacteria, epithelial barrier, mucus, and immune cells [ 109 ]. The correct functioning of this relationship results in homeostasis. It is the sensor that enables both tolerance to healthy microbiota and response to diseased microbiota. The immune system plays a vital role in the interaction between the host and the microbiota. Peritoneal macrophages play a central role in the development and maintenance of endometriotic lesions. Macrophages function as innate immune cells by phagocytosing and sterilizing foreign substances such as bacteria and play a central role in defending the host against infection. Increased in numbers in the peritoneal cavity as well as in the lesions, they participate in chronic inflammation and the recruitment and activation of other immune cells through their secretion of cytokines [ 110 , 111 ]. Their role in eliminating endometrial debris by phagocytosis is severely impaired, whereas the number of macrophages and their activation is greatly increased in endometriosis [ 112 , 113 ]. A reduced phagocytotic capacity of peritoneal macrophages in endometriosis has been described in association with decreased levels of protein and matrix metalloproteinase-9 enzyme activity [ 114 ]. Emani et al. found that leakage of bacterial products from the gut results in increased numbers of macrophages in the peritoneal cavity [ 115 ]. According to their activation status, macrophages are divided into two populations: classically activated (M1) and alternatively activated (M2). This activation state is highly dependent on the cellular environment, and macrophages can switch from one type to another. M1 macrophages are capable of pro-inflammatory responses. In contrast, M2 macrophages are involved in angiogenesis, and they are capable of anti-inflammatory responses and repair of damaged tissues [ 116 , 117 ]. In infected tissues, macrophages are first polarized to the pro-inflammatory M1 phenotype to assist the host against pathogens. Subsequently, macrophages are polarized to the M2 phenotype to mount an anti-inflammatory response and repair damaged tissue. A shift from M1 to M2 within eutopic endometrium is observed in association with endometriosis [ 118 ]. In a mouse model of endometriosis, analysis of the immune cell populations in the peritoneal fluid of vehicle versus antibiotic-induced microbiota-depleted mice found fewer total and CD206 + (M2-like) macrophages in the peritoneum [ 56 ]. Finally, at the endometrial level, no study to date has investigated the impact of the endometrial microbiota on macrophage activation or differentiation [ 119 ]. The implication of neutrophils in the modulation of microbiota in endometriosis has not been studied. However, neutrophils cells in endometriosis are known to generate inflammation as it is demonstrated by an increase level of chemotactic factors, such as IL-8 in the plasma or in peritoneal fluid of affected women [ 120 ]. Neutrophil infiltration and migration can be promoted by their own IL-17 production and by estrogen [ 121 , 122 ]. Their modulation by the microbiota in endometriosis has not been studied to date. The involvement of natural killer (NK) cells in the disturbance of microbiota in endometriosis women has not been clearly elucidated. Their low cytotoxic activity against ectopic endometrial cells may be due to the consequences of chronic disease-induced inflammation [ 123 ]. Macrophages-secreted IL-6, IL-10, and transforming growth factor β in the peritoneal environment have been shown to reduce the cytolytic activity of NK cells [ 124 , 125 ]. It has been shown that butyrate (e.g., microbial short-chain fatty acids) has a strong anti-inflammatory effect on NK cells. Co-culture of NK cells with butyrate has been shown to decrease cytokine production (interferon-γ, tumor necrosis factor-α, and IL-22) and to downregulate mTORC1 activity [ 126 ]. The impact of NK and neutrophil cells on variations in the microbiota deserves further investigation. Given the central role of macrophages in endometriosis, the trained immunity associated with macrophages and its role in bacterial dialogue should be discussed. Briefly, macrophages can be reprogrammed to acquire memory-like characteristics after antigenic challenge to reinforce or inhibit a subsequent immune response, a phenomenon called trained immunity [ 127 ]. Jeljeli et al. trained peritoneal macrophages in a mouse model of endometriosis. When exposed to bacillus Calmette–Guérin (BCG), macrophages enhance the growth of lesions. Conversely, with repeated exposure to low doses of LPS, immunotolerance was observed, resulting in smaller lesions and less production of the pro-inflammatory cytokine IL-10 [ 23 ]. This model is fully adaptable to bacterial exposure; LPS is contained in the membrane of gram-negative bacteria. Exposure to higher doses of LPS produces opposite effects, with larger lesions [ 128 ]. The bacterial biomass studied in the microbiota is closer to the doses used in trained immunity. B and T lymphocytes play an essential role in the survival and proliferation of endometrial cells. Indeed, endometriosis is characterized by reduced activity of cytotoxic T cells, modulation of cytokine secretion by T helper cells, and autoantibody production by B lymphocytes [ 129 ]. Le et al. studied a baboon model for endometriosis that presents several advantages such as natural menses, development of spontaneous endometriosis [ 130 ], and adequate animal body size for repeated sampling of sufficient size. They noticed alteration of the gut microbiota in baboons after induction of endometriosis. These changes were accompanied by an increase in the Tregs/Th17 cell ratio in peripheral blood [ 26 ]. Regulatory T cells (Tregs) are potent suppressors of inflammatory immune responses. Th17 cells are a pro-inflammatory effector CD4 T cell population that initiates an inflammatory response mainly through recruitment, activation, and migration of neutrophils [ 131 ]. Interestingly, they found a correlation between certain bacterial species and these T-cell populations. The phyla Bacteroidetes, Firmicutes, and Proteobacteria were negatively correlated with the level of peripheral Treg cells; while Prevotella and Sutterella were positively correlated with the level of Th17 cell populations. No data are available for other T and B cells and their interaction with microbiota in the context of endometriosis. More generally, the most studied form of homeostatic immunity to the gut microbiota is the one associated with IgA specific for commensal-derived antigen responses secreted by T cells in Peyer’s patches [ 132 ]. At steady state, most Th17 and Th1 cells are found at barrier sites, and their frequencies are severely reduced in the context of antibiotic-induced depletion of microbiota [ 133 ]. The endometrial defenses against bacteria involve their recognition by receptors. These receptors are found on the surface of endometrial immune cells and comprise TLRs [ 134 ] and several NLRs [ 135 ], which bind molecules specific to microbial organisms, which are often called pathogen-associated molecular patterns (PAMPs). No relationship between the microbiota and the expression of these receptors in endometrium has been documented to date in endometriosis. While bacteria can colonize the mucosa, the endometrium can also defend itself against these bacteria by producing antimicrobial peptides [ 136 ]. Antimicrobial peptides are effective against bacteria, fungi, enveloped viruses, and protozoa. Transcription of antimicrobial peptides is increased when pattern recognition receptors are activated and in response to cytokines [ 137 ]. In addition to direct inhibition of microorganisms, antimicrobial peptides also help protect epithelia against microbial proteases and help in resolving inflammation. The endometrium has a unique set of antimicrobial peptides that are mainly secreted by epithelial cells and leukocytes: β-defensin 1–4, α-defensin (human defensin 5), elafin, and secretory leukocyte protease inhibitor [ 138 ]. β-Defensins are the most common endometrial defensins. Activated by NF-kB transcription factors, they are known to have an antibacterial effect and the ability to neutralize LPS [ 139 ]. Intriguingly, these antimicrobial peptides were identified in a physiological setting prior to the discovery of an endometrial microbiota. To date, there are no data linking the expression of these antimicrobial peptides to endometrial microbiota or endometriosis.

Endometriosis, a chronic disease predominantly affecting women of childbearing age, has garnered increasing attention in the context of microbiota research. Investigations into the digestive and gynecological microbiota in relation to endometriosis have provided valuable insights. Notably, the interplay between the immune system and bacterial interactions has demonstrated promising outcomes concerning lesion development and the potential for microbiota modulation as a therapeutic approach for patients. These findings highlight the significance of understanding the intricate connections between the microbiota, immune system, and endometriosis pathogenesis, paving the way for innovative treatment strategies. However, great caution must be exercised regarding various potential biases, such as the menstrual cycle, lifestyle habits, sexual habits, and precise selection of control groups. Further exploration of this complex interplay holds immense potential for improving the management and outcomes of individuals affected by endometriosis.

Adenomyosis, characterized by the presence of endometrium in the myometrial tissue, is a significant threat to women’s health due to its high incidence [ 184 ]. Adenomyosis, which is frequently associated with endometriosis (~ 30%), causes pain, infertility, and abnormal uterine bleeding [ 185 ]. Diffuse internal adenomyosis (in relation to the myometrium) is the most common form, compared to focal external adenomyosis, which shares more characteristics with deep infiltrating endometriosis [ 186 ]. Chen et al. recently investigated the impact of adenomyosis on the gut microbiota [ 187 ]. They found an increase in the ratio of Firmicutes/Bacteroidetes and the relative abundance of Lactobacillus species in cases of adenomyosis. In a study of 38 patients with adenomyosis—89% of whom also had deep infiltrating endometriosis—and 46 controls, 16S rRNA sequencing revealed significantly reduced gut microbial α-diversity and distinct gut and vaginal microbiota compositions in patients with adenomyosis. Several bacterial taxa were differentially represented in the gut and endometrial microbiota, with specific profiles associated with internal versus external adenomyosis phenotypes. Data are available regarding the association between adenomyosis and imbalance in the vaginal microbiota. Kunaseth et al. found an increase in microbial richness in patients with adenomyosis, with abundant Alloscardovia , Oscillospirales , Ruminococcaceae , Oscillospiraceae , Enhydrobacter , Megamonas , Selenomonadaceae , and Faecalibacterium [ 188 , 189 ]. Interestingly, Atopobium is a known biomarker of vaginal microbiota in patients with endometriosis combined with adenomyosis [ 190 ]. Lin et al. studied the endometrial microbiota in 38 women with ( n  = 21) and without ( n  = 17) adenomyosis. In contrast with the results for patients with endometriosis, patients with adenomyosis had a significantly lower richness than the control group. Their findings identified Lactobacillus zeae , Burkholderia cepacia , Weissella confusa , Prevotella copr i, and Citrobacter freundii as potential biomarkers for adenomyosis [ 191 ]. A recent systematic review concluded that adenomyosis is associated with significant microbial alterations, including a depletion of protective vaginal Lactobacillus species and an enrichment of vaginal and endometrial opportunistic anaerobic bacteria rich in virulent cell wall components such as LPS [ 192 ]. It has also been shown that these microbial differences are cycle-dependent and more pronounced during the luteal phase [ 193 ]. The proximity of adenomyosis lesions to the uterine cavity and the vagina makes them a potential therapeutic target for vaginal probiotics.

Endometriosis is a chronic, estrogen-dependent, gynecological condition characterized by the presence of endometrial-like tissue outside the uterine cavity [ 1 , 2 ]. Endometriosis likely results from the convergence of multiple biological mechanisms. The leading theory—retrograde menstruation—suggests that endometrial cells flow backward through the fallopian tubes during menstruation and implant in the peritoneal cavity [ 3 ]. Approximately 10% of reproductive-age women are afflicted with endometriosis, but the discrepancy between this prevalence and the quasi-systematic retrograde menstruations in cycling women (80%) suggests other susceptibility factors are implicated at the individual level [ 4 ]. This suggests the involvement of additional mechanisms [ 5 ]. The implantation of ectopic endometrial cells appears to be facilitated by: (i) alterations in local immunity, particularly within the peritoneal cavity, including impaired macrophage clearance and dysregulation of pro-inflammatory cytokines [ 6 ]; (ii) a hormonal environment marked by relative hyperestrogenism and progesterone resistance, which promotes ectopic lesion survival and growth [ 7 ]; and (iii) neovascularization and tissue invasion, driven by growth factors and matrix metalloproteinases, contributing to lesion progression [ 8 ]. Although the retrograde menstruation theory explains the typical pelvic distribution of lesions, it fails to account for extrapelvic forms of endometriosis, such as inguinal or pulmonary involvement [ 9 ]. Alternative hypotheses include Müllerian remnants [ 10 ], coelomic metaplasia, and lymphovascular dissemination [ 3 , 11 ], which account for cases in non-menstruating women and distant lesion sites. More recently, the role of endometrial stem cells has gained attention, suggesting their aberrant migration and differentiation at ectopic locations [ 12 ]. Together, these theories underscore the multifactorial nature of endometriosis pathogenesis. Endometriosis is classically responsible for pain and infertility, thus significantly impacting quality of life [ 13 ]. A correlation between the location of lesions and patients' symptoms was first demonstrated during surgical treatment of endometriosis [ 14 ], then thanks to the considerable development of medical imaging techniques [ 15 ]. Some classifications of the disease have sought to link the intensity of painful symptoms to the anatomical staging of lesions. Severe stages (III and IV) and mild to moderate stages (I and II) have thus been defined by the American Society for Reproductive Medicine [ 16 ]. In some cases, this correlation was not accurate, as shown by patients with a severe stage on imaging but no painful symptoms or infertility [ 17 ]. The concept of disease defined by the location of lesions and their correlation with symptoms has progressively evolved. Surgical removal of endometriosis lesions could result in residual pain, such as neuropathic pain [ 18 ]. The discovery of an association between endometriosis and autoimmune or inflammatory diseases (systemic lupus erythematosus, multiple sclerosis, Sjögren's syndrome and inflammatory bowel disease) [ 19 , 20 ], the presence of migraine [ 21 ] or cardiovascular comorbidities [ 22 ] argues in favour of a systemic feature to endometriosis. Recent studies have revealed a modulatory effect of training innate immunity in endometriosis, suggesting a role for environmental factors, especially the effects of bacterial stimulation [ 23 ]. The influence of the microbiome on the development of various inflammatory diseases is well established, implicating bacteria in enhanced immune responses and the perpetuation of chronic inflammation [ 24 , 25 ]. Unique intestinal microbiota signatures and distinct associated immune cell profiles in peripheral blood and endometrium, characterized by an enhanced Th17/Treg ratio, have been observed in a baboon model of endometriosis, thus highlighting the importance of bacteria–immune system interaction in this chronic disorder [ 26 ]. Focusing on the intestinal mucosa, factors such as lipopolysaccharide (LPS), antimicrobial peptides, mucins, short-chain fatty acids (SCFAs), and immunoglobulins are thought to play a role, but their precise contributions have not yet been clearly defined. Furthermore, the discovery of an endometrial microbiota through next-generation sequencing (NGS), particularly targeting 16S ribosomal RNA, has opened new avenues for investigation in many gynecological conditions such as endometriosis and infertility [ 27 , 28 ]. Focusing on endometriosis, the exact mechanisms of interaction between the endometrial microbiota and the endometrium, as well as the exact impact on the disease process, remain unknown. This review aims to provide a comprehensive understanding of the mechanisms by which the digestive and gynecological microbiota may participate in endometriosis pathogenesis. By elucidating these mechanisms, new diagnostic, and therapeutic strategies for managing this multifactorial disease can be explored.

Neurogenesis is defined as the generation of new neurons, glial cells, and other neural lineages. Endometriosis-associated pain is complex. It is widely accepted that no correlation exists between the extent of endometriosis seen at laparoscopy or radiology and the degree of pain symptoms [ 174 ]. The experience of pain is complex and involves many mechanisms and interactions between the peripheral and the central nervous systems [ 175 ]. This involves the activation of nociceptors [ 176 ] and neurogenesis of lesions activated by neurophils [ 177 ]. No data are available on a possible interaction between the microbiota and neurogenesis at both the digestive and endometrial sites. The release of inflammatory mediators by the gut microbiota can lead to neuroinflammation [ 178 ]. In addition, microbiota-derived short-chain fatty acids can induce the proliferation and differentiation of human neural progenitor cells [ 179 ]. Although the relationship between microbiota and neurogenesis has not been specifically explored in the context of endometriosis, the role of the gut microbiota in modulating visceral pain is well documented [ 180 ], particularly in irritable bowel syndrome. Studies in germ-free mice have shown that the absence of commensal gut microbes leads to enhanced visceral pain sensitivity, a phenotype that can be reversed upon microbial colonization [ 181 ]. This pain-related sensitivity has also been shown to be transmissible via fecal microbiota transplantation (FMT), and attenuated by antibiotic treatment [ 182 ]. Mechanistically, these effects appear to involve inflammasome signaling pathways [ 183 ], suggesting that microbial composition may shape nociceptive pathways through immune–neural interactions.

Inflammation and aberrant immune responses are well-known factors involved in the pathophysiology of endometriosis. Both can be triggered by bacterial endotoxins, which promote the secretion of inflammatory cytokines and chemokines. The bacteria found in the microbiota are sources of pathogen-associated molecular patterns and metabolites. Pathogen-associated molecular patterns are mainly represented by LPS (a component of the outer cell wall of gram-negative bacteria), lipoteichoic acid, and flagellin. These highly conserved structural motifs serve as ligands that are recognized by PRRs to trigger immune responses. Bacterial adaptive changes including modulation of LPS synthesis and structure are conserved processes in infections. Activation of macrophages and dendritic cells was found to occur by the binding of LPS to the membrane surface of certain gram-negative bacteria, to its TLR4 receptor, resulting in nuclear translocation of NF-kB, which induces the expression of inflammatory cytokines (IL-6 and TNFα) [ 194 ]. In patients with endometriosis, Khan et al. found a higher colony formation of Escherichia coli in menstrual blood and endotoxin levels in menstrual fluid [ 195 ]. The hypothesis that E. coli is a key pathogenic bacterium in endometriosis has been reinforced by a recent publication from our team focusing on the only pathognomonic infection of the disease: endometrioma infections. An ecological analysis of aspirated fluid from infected endometriomas revealed that the majority of infections were caused by E. coli [ 53 ]. Several studies have shown that intraperitoneal LPS injected into mice increases endometriosis lesions through activation of the NF-kB pathway [ 128 , 196 ]. Moreover, the use of antibiotics before or after induction of endometriosis in rodent models results in a decrease in lesion size and inflammation. It is not possible to know whether the antibiotic action was induced by modification of the microbiota, by inhibition of certain inflammatory pathways, or by both [ 197 ]. A recent study revealed that infection of endometrial cells with Fusobacterium led to a transition from quiescent fibroblasts to myofibroblasts, with an increase in their proliferation, adhesion, and migration [ 198 ]. These results were confirmed by Fusobacterium inoculation in a syngeneic mouse model of endometriosis, which resulted in an increase in the number and weight of endometriotic lesions. This effect was also reversible with antibiotic treatment. Interestingly, Lin et al. conducted a nationwide prospective cohort study to determine whether lower genital tract infections increase the risk of endometriosis. A total of 79 512 patients were included in the lower genital tract infection group, with an equal number of controls. The incidence of endometriosis (HR = 2.01; p  < 0.001) was higher in patients than in the controls [ 199 ]. Concerning upper genital tract infection, a strong association has been found between endometriosis and chronic endometritis [ 200 ]. Pelvic inflammatory disease was shown to be a major risk factor (three-fold) for developing endometriosis within 10 years in another nationwide retrospective cohort study involving a total of 141 460 patients [ 201 ]. Several questions remain unanswered, however, and are likely to be key to understanding the bacterial hypothesis in the pathophysiology of endometriosis. Do the bacteria responsible for these infections come from the endometrium? Are they responsible for the alteration of the endometrium, from eutopic to ectopic, and do they then promote the implantation of lesions in the peritoneal cavity? Modulating the microbiota could be beneficial for patients (Fig. 1 ). In a mouse model of endometriosis, oral administration of Lactobacillus reduced the size of lesions by activating NK cells, mimicking their stimulation by IL12 [ 202 ]. Furthermore, treatment with one or two probiotics ( Saccharomyces boulardii and/or Lactobacillus acidophilus ) has been reported to exert favorable effects on clinical, immune, and physiologic parameters in a mouse model of endometriosis [ 203 ]. In patients with severe endometriosis (Stages III and IV), a randomized placebo-controlled trial showed some beneficial effects of oral administration of Lactobacillus on endometriosis-related pain [ 204 ]. In addition to modulating the microbiota, the use of broad-spectrum antibiotics has been tested in a mouse model of surgically-induced endometriosis, with the important observation of reduction of the development of lesions [ 49 ]. Finally, as demonstrated in the mouse study by Parpex et al., the marked depletion of Akkermansia in the gut, observed in association with the most severe endometriotic lesions, highlights its potential as a candidate for probiotic-based therapeutic strategies [ 52 ]. Fig. 1 Gut and Gynecological Microbiota in Endometriosis: Current Knowledge and Future Directions Gut and Gynecological Microbiota in Endometriosis: Current Knowledge and Future Directions A potential lead lies in the effect of dietary modifications on these symptoms. The low-FODMAP (Fermentable Oligosaccharides, Disaccharides, Monosaccharides And Polyols) diet was introduced in 2005 to group together certain poorly absorbed fermentable carbohydrates responsible for intestinal symptoms, particularly in patients with irritable bowel syndrome [ 205 ]. Clinical studies confirmed the effectiveness of the low-FODMAP diet in irritable bowel syndrome patients, which reduces pain, bloating, and bowel movement disorders by limiting the osmotic and fermentative effects of these sugars in the gut [ 206 ]. This dietary approach is based on well-defined pathophysiological mechanisms, including incomplete intestinal absorption, rapid fermentation by the microbiota, and visceral hypersensitivity, all of which vary depending on the type of FODMAP and individual sensitivity. Moore et al. evaluated the efficacy of this diet on gastrointestinal symptoms in 160 female patients diagnosed with irritable bowel syndrome. Among them, 59 (37%) had a confirmed diagnosis of endometriosis. After one month on the diet, improvement defined as a > 50% reduction in gastrointestinal symptoms was more pronounced in patients with both endometriosis and irritable bowel syndrome [ 207 ]. A randomized controlled trial conducted by the same team further explored this effect by comparing the low-FODMAP diet to both the patients’ baseline diet and a control diet based on national Australian dietary guidelines [ 208 ]. The study used a crossover design with three 4-week periods (including a 4-week washout). A significant improvement in quality of life was observed after 4 weeks on the low-FODMAP diet, compared to both the baseline diet group ( p  < 0.001) and the control diet group ( p  = 0.004). Additional clinical symptoms such as abdominal pain, stool consistency, and abdominal bloating were also significantly improved. This is the first study providing robust evidence that a dietary intervention can significantly enhance the quality of life of patients with endometriosis. The changes in gut microbiota before, during, and after implementation of the low-FODMAP diet need to be investigated in patients with endometriosis. This promising diet in endometriosis is known to induce long-lasting changes in the gut microbiota of patients with irritable bowel syndrome (lower Bifidobacteria ) [ 209 ]. If a sustained improvement in symptoms associated with microbiota alterations were also observed in endometriosis, this would constitute strong evidence for considering the gut microbiota as an effective therapeutic target. Dysbiosis has been associated with increased intestinal permeability (“leaky gut”), allowing the translocation of antigens and bacteria, which may promote chronic inflammation and contribute to endometriosis progression [ 210 ]. A pilot study by Mohling et al. found that nearly half of patients with laparoscopically confirmed endometriosis had impaired intestinal permeability, compared to none in the control group, suggesting a possible link between barrier dysfunction and the disease [ 211 ]. Zonulin, a key regulator of tight junctions, is overexpressed in states of dysbiosis and may facilitate immune activation, potentially explaining gastrointestinal symptoms commonly reported in endometriosis, even in the absence of bowel involvement [ 212 ]. Future studies should investigate zonulin as a biomarker of gut permeability and explore probiotic interventions to alleviate symptoms and improve quality of life in affected patients. The role of the gynecological microbiota—both vaginal and endometrial—in endometriosis remains less well characterized to date. However, several hypotheses can be proposed, starting from a consistent observation: most studies report alterations in the vaginal and endometrial microbiota. While the composition of a healthy vaginal microbiota is well defined, the endometrial microbiota is less understood and likely originates from the vaginal tract. This endometrial microbiota warrants further investigation, as it is in direct contact with the eutopic endometrium prior to retrograde menstruation and may contribute to lesion establishment in the peritoneal cavity. Consequently, it could be involved in peritoneal inflammation through its anatomical continuity via the fallopian tubes and may also play a role in infertility mechanisms through the presence of pathogenic bacteria (subclinical endometritis) and associated cervical mucus dysregulation.