Research progress of gut microbiota and its metabolites in polycystic ovary syndrome

Robust microbial diversity is fundamental to the gut microbiota’s resilience against stressors and serves as a primary indicator of overall health. Conversely, a reduction in this diversity, often driven by the proliferation of pathogenic microorganisms, signals a compromised physiological state (14). In PCOS patients, the gut microbiota often exhibits altered diversity. Specifically, α-diversity reflects the richness and variety of species within a specific community, while β-diversity measures the similarity or difference between distinct samples or groups (12).

Numerous clinical studies have indicated that the α-diversity of gut microbiota is significantly reduced in PCOS patients. Commonly used α-diversity indices in current research, such as Shannon index, Chao1 index, Simpson index, and observed operational taxonomic units (OTUs), are generally lower in the PCOS group compared to healthy controls (15–21). However, some clinical studies failed to observe significant differences in α-diversity between PCOS patients and healthy controls (22–25). Notably, Dong et al. (26) conducted a subgroup analysis of PCOS patients, revealing that significant differences in the Shannon index persisted between specific PCOS subgroups and the control group, when stratified by homeostasis model assessment of insulin resistance (HOMA-IR) or body mass index (BMI). This suggests that an overall analysis may obscure α-diversity alterations within distinct subgroups. In animal model studies, both Torres’s (27) and Han’s (28) studies observed a decrease in the Faith’s phylogenetic diversity (PD) index in the PCOS model group. Conversely, studies by Zheng (29), Zheng (30), Rodriguez Paris (31), Yu (32), and Esfandiarinezhad (33) showed that α-diversity indices such as Chao1, Shannon, Simpson, abundance-based coverage estimator(ACE), and species evenness in the PCOS animal model group were not significantly different from the control group. However, Li et al. (34) observed in their animal study that the amplicon sequence variants (ASVs) count, Shannon, and Simpson indices in the PCOS model group were, on the contrary, increased.

Regarding β-diversity, most current studies are analyzed using methods such as principal coordinate analysis (PCoA) and permutational multivariate analysis of variance (PERMANOVA), primarily based on distances like UniFrac and Bray-Curtis. In clinical studies, some have detected significant differences in gut microbiota structure between PCOS patients and healthy controls (15, 18–22, 25, 35). However, several other studies failed to observe significant differences in β-diversity (16, 17, 23, 24, 26, 36). Among these studies, Chu (36), Dong (26), and He (23) further performed subgroup analyses of PCOS patients stratified by BMI or IR. Even after such stratification, they still failed to detect significant differences in gut microbiota structure between these PCOS subgroups and the control group. In animal model studies, most investigations identified significant differences in gut microbiota β-diversity between PCOS model groups and control groups (21, 27, 29, 31–34), with only a few studies reporting no significant differences (28, 30, 37). Among these, Zheng (30) and Pieczyńska-Zając (37) noted that significant changes were observed in the high-fat PCOS subgroup.

It can be seen that the current research results on gut microbiota diversity in PCOS exhibit a certain degree of heterogeneity, both in clinical cohorts and animal models. However, the overall trend indicates that PCOS may be associated with changes in gut microbiota diversity. Han et al. (38) analyzed and found that the α-diversity of animal models showed higher stability. They proposed that the discrepancy in results between PCOS clinical studies and animal studies might be attributed to the strict control of environmental factors (such as diet and genetics) in animal experiments, which minimizes external confounding variables. In contrast, most human studies do not control or standardize patients’ diets, leading to fluctuations in the detected results of gut microbiota composition and diversity. (Table 1).

The dysbiosis of the gut microbiome in patients with PCOS may not be characterized solely by a decrease in overall diversity, but rather by a more pronounced manifestation of specific compositional changes in certain taxa at the phylum, genus, or species level (36, 39).

In PCOS, a significant increase in the abundance of certain potentially pathogenic and pro-inflammatory bacterial groups is observed. Specifically, the phylum Proteobacteria and its constituent family Enterobacteriaceae, particularly the genera Escherichia/Shigella, exhibit elevated abundance. These opportunistic pathogens are known to secrete virulence factors, thereby exacerbating systemic inflammation and metabolic dysfunction (18, 35, 36). Furthermore, within the phylum Bacteroidota, the family Bacteroidaceae, especially species like Bacteroides vulgatus and Bacteroides fragilis, show increased abundance. These specific bacteria contribute to the exacerbation of IR, ovarian dysfunction, and systemic inflammation through mechanisms such as interfering with bile acids (BAs) metabolism, compromising the intestinal mucosal barrier, and degrading mucin (22, 36). Similarly, species within the genus Prevotella, including Prevotella stercorea and Prevotella copri, also demonstrate increased abundance. They are implicated in promoting intestinal inflammation, leading to systemic metabolic abnormalities, and influencing branched-chain amino acids (BCAAs) synthesis and host metabolism (26, 35, 40). Concurrently, there is a general reduction in the abundance of beneficial bacteria crucial for gut health and short chain fatty acids (SCFAs)-producing bacteria. The overall abundance of the phylum Bacteroidota is generally diminished, which may impair its role in regulating intestinal barrier function and maintaining metabolic health (20, 29). Within the phylum Firmicutes, important SCFAs-producing bacterial families, such as Lachnospiraceae and Ruminococcaceae, show significantly decreased abundance. This reduction leads to diminished production of SCFAs like butyrate, consequently weakening intestinal barrier function, anti-inflammatory effects, and metabolic regulatory capacity (20, 21, 35). Firmicutes and Bacteroidota are the two most dominant phyla in the human gut microbiota, and their relative abundance ratio (F/B ratio) is considered a critical indicator for assessing gut dysbiosis. An elevated F/B ratio is typically closely associated with host metabolic dysregulation, impaired intestinal barrier function, immune dysregulation, and the development and progression of various metabolic diseases, including obesity and IR (41–43) (Table 2).

SCFAs, metabolites produced by gut microbiota through fermenting dietary fiber, include acetate, propionate, butyrate, and valerate (52). A study by Zhang et al. (35) found that the intestinal content of acetate, propionate, and butyrate was reduced by approximately 30% to 66% in PCOS patients compared to healthy controls. Following 10 weeks of probiotic treatment in PCOS patients, the abundance of Lactobacillus in the gut significantly increased, accompanied by a consequent rise in SCFAs levels. Multiple studies have demonstrated a significant reduction in beneficial SCFAs-producing bacteria in PCOS patients, such as Butyricicoccus, Blautia, Coprococcus, Faecalibacterium prausnitzii, the Clostridium innocuum group, and Prevotella, leading to decreased intestinal SCFAs levels (53, 54). This dysbiosis disrupts the metabolic equilibrium between the host and the microbiota, thereby promoting the onset and progression of PCOS. (Figure 1).

BAs play a crucial role in the body through enterohepatic circulation. In the liver, cholesterol is utilized and catalyzed by the enzyme cholesterol 7α-hydroxylase (CYP7A1) to synthesize primary unconjugated BAs, which then conjugate with glycine or taurine to form primary conjugated BAs. These conjugated BAs are concentrated in the gallbladder and released into the intestine. The gut microbiota, including Bacteroides vulgatus, Ruminococcus, Lachnospiraceae, and Prevotella, employs BSH and 7α-dehydroxylase to deconjugate and dehydroxylate them, yielding secondary unconjugated BAs. These secondary unconjugated BAs are reabsorbed back into the liver, re-conjugated, and re-enter the intestine with bile, completing the circulation (12, 59). A study by Yu et al. (60) revealed alterations in the BAs profile of PCOS patients. Specifically, levels of primary bile acids, especially chenodeoxycholic acid (CDCA) and taurochenodeoxycholic acid (TCDCA), and of secondary unconjugated bile acids such as lithocholic acid (LCA) and deoxycholic acid (DCA) were significantly elevated, while the level of the secondary conjugated bile acid, glycodeoxycholic acid (GDCA), was decreased. (Figure 1).

LPS, known as bacterial endotoxin, is a unique component of the cell wall of gram-negative bacteria (GNB) (63). Zheng et al. (29) showed in a study on female mice that serum LPS levels were significantly negatively correlated with the gut microbiota Chao1 index (r=-0.584, p = 0.007), suggesting that elevated LPS levels may be related to reduced gut microbiota richness. Notably, Guan et al. (64) found that the abundance of Akkermansia muciniphila in the intestines of PCOS model mice was significantly increased by 31.99% (p < 0.01) compared to the normal group. The overgrowth of Akkermansia muciniphila extensively consumes mucin in the intestinal mucus layer, leading to thinning of the mucus layer and a reduction in goblet cell numbers, thereby weakening the intestinal physical barrier. This impaired barrier not only induces local inflammation, but the metabolites produced by its mucin degradation also provide nutrients for other GNB, promoting their proliferation and increasing the total LPS synthesis (65). Dubey et al. (66) observed an elevated relative abundance of potential pathogens, including the phylum Proteobacteria and the family Enterobacteriaceae, in the gut microbiota of PCOS patients. These bacteria are capable of secreting toxic substances such as LPS and adhesins. Additionally, Clostridium perfringens can generate enterotoxins that compromise the tight junctions of the intestinal epithelium, thereby increasing intestinal permeability. This enhanced permeability facilitates the translocation of endotoxins into the systemic circulation, which activates a generalized immune response and triggers chronic low-grade inflammation (54). (Figure 1).

The production of TMAO is dependent on the gut microbiota’s metabolism of dietary metabolites. The gut microbiota metabolizes substances like choline, phosphatidylcholine, and L-carnitine to produce trimethylamine (TMA). After being absorbed into the bloodstream, TMA is further oxidized to TMAO by hepatic flavin-containing monooxygenases (FMOs) (71, 72). Specifically, mice studies have confirmed that Lachnoclostridium saccharolyticum and Clostridium sporogenes can produce TMA through choline metabolism. Human metagenomic studies have shown that γ-proteobacteria, especially Escherichia coli, are the main TMA-producing bacteria (73, 74). It is known that TMAO is closely associated with various metabolic and cardiovascular diseases, including atherosclerosis, heart failure, IR, hepatic steatosis, and non-alcoholic fatty liver disease (NAFLD) (75–77). Due to gut microbiota dysbiosis being a key feature of PCOS, women with it are also at a higher risk of developing these related diseases.

BCAAs, including valine, leucine, and isoleucine, are essential amino acids that must be obtained through diet, as they cannot be synthesized endogenously by humans (83). They also serve as crucial substrates metabolized by the gut microbiota. Several studies indicate that BCAAs are significantly elevated in obese patients with PCOS, showing a positive correlation with both IR and HA (84). Paczkowska et al. (85) confirmed this observation, reporting significantly higher BCAAs in PCOS patients (540.59 ± 97.23 nmol/mL vs 501.09 ± 85.33 nmol/mL), particularly in those exhibiting HA. They further noted that BCAAs dysregulation was more pronounced in the subgroup of PCOS patients presenting with abdominal obesity. Multiple studies have identified Prevotella copri and Bacteroides vulgatus as key bacterial species involved in BCAAs synthesis within the human gut (40, 86). Thus, gut microbiota dysbiosis may contribute to PCOS pathology via the abnormal metabolism of BCAAs. Supporting this mechanism, animal experiments have demonstrated that colonization with Prevotella copri can elevate circulating BCAAs in mice, leading to the induction of IR and the exacerbation of glucose intolerance (40). A prospective cohort study conducted by Jing et al. (87) found that the abundance of Parabacteroides merdae was reduced in the gut of PCOS patients, and this reduction was associated with elevated serum levels of BCAAs, particularly isoleucine. (Figure 1).

The state of IR leads to hyperinsulinemia, where excessive insulin binds to its receptors on ovarian theca cells. This action upregulates the activity of key steroidogenic enzymes, specifically cytochrome P450 family 11 subfamily A member 1 (CYP11A1) and cytochrome P450 family 17 subfamily A member 1 (CYP17A1). CYP11A1 catalyzes the conversion of cholesterol to pregnenolone, which serves as an androgen precursor, while CYP17A1 further processes this precursor into biologically active androgens. This mechanism directly results in elevated ovarian androgen synthesis. Concurrently, insulin promotes the expression of LH receptors on the surface of theca cells, thereby amplifying the effects of androgen synthesis and exacerbating androgen oversecretion, culminating in HA (69, 102). The accumulation of excessive androgens within the follicles directly inhibits GCs proliferation and antrum formation, causing follicular arrest at the small antral follicle stage, thereby preventing their development into dominant follicles. Furthermore, patients with IR exhibit significantly elevated levels of AMH secreted by ovarian GCs. AMH inhibits follicular sensitivity to FSH, impeding the transition from primordial to preantral follicles. It also reduces aromatase activity, further suppressing estrogen synthesis, thereby aggravating follicular maturation defects and leading to the characteristic PCOM observed via ultrasound (48).

The endometrium is a target organ for insulin and its receptors, and its cells regulate glucose metabolism through surface glucose transporters (103, 104). Local IR in the endometrium can lead to insufficient glucose supply, interfering with endometrial growth and activity, reducing endometrial receptivity (ER), and thereby affecting embryo implantation and development (10, 105). This process is also closely related to inflammatory responses. PCOS patients have low-grade inflammation, which can interact with IR to promote each other (106). Inflammatory factors can activate the NF-κB signaling pathway, further producing more inflammatory factors such as TNF-α, IL-6, and IL-18 (107, 108). Among them, TNF-α can inhibit insulin receptor tyrosine phosphorylation, block insulin signaling pathways, and reduce the expression of endometrial glucose transporter 4 (GLUT-4), exacerbating local IR in the endometrium (109, 110).

HA directly acts upon the cellular signaling pathways within insulin target organs, such as the liver, muscle, and adipose tissue, thereby inhibiting the normal physiological function of insulin. Within the insulin signaling cascade, androgens can suppress the phosphorylation and activation of key molecules, notably phosphatidylinositol 3-kinase (PI3K) and PKB. These molecules are central to insulin’s role in promoting glucose transport, glycogen synthesis, and lipogenesis. This blockage of signal transduction directly impedes the cell’s ability to effectively uptake and utilize glucose, consequently leading to IR (111, 112). In addition, HA can lower the levels of adiponectin, an insulin sensitivity-related adipokine and an important insulin sensitizer, whose reduction directly weakens the body’s responsiveness to insulin. Conversely, HA increases the release of pro-inflammatory adipokines, such as Leptin and Resistin. These factors not only exacerbate local chronic inflammation within adipose tissue but also act via systemic circulation on peripheral tissues, including the liver and skeletal muscle, further inhibiting the insulin signaling pathway and aggravating IR (113, 114). Adipocytes in PCOS patients are significantly larger than those in BMI-matched healthy women, and adipocyte size exhibits a positive correlation with the level of HA (115). These hypertrophied adipocytes demonstrate markedly reduced insulin sensitivity, coupled with decreased expression and content of GLUT-4. This results in impaired insulin-mediated glucose uptake capacity, directly inducing local IR within the adipose tissue (116).

FMT aims to restore the healthy balance of the gut microbiota or treat gut-related diseases by transplanting fecal microbiota from a healthy donor into the patient’s intestine (117). Restoring a healthy gut microbiota through FMT can simultaneously improve both the metabolic and reproductive phenotypes of PCOS, positioning it as a potentially effective therapeutic strategy. Studies by Huang et al. (118) demonstrated that when fecal microbiota from PCOS patients were transplanted into germ-free mice, the recipient mice developed symptoms such as weight gain, IR, elevated T, abnormal estrous cycles, reduced corpus luteum count, and intestinal barrier damage. Concurrently, the abundance of specific gut bacteria, including Faecalibaculum and Mediterraneibacter genera, was found to be elevated in the gut. Martinelli et al. (119) observed that following the transplantation of fecal microbiota from healthy rats into PCOS model rats, all recipients showed restoration of normal estrous cycles, reduced androgen synthesis, normalization of ovarian morphology, and improved gut microbiota composition. Specifically, the abundance of Lactobacillus and Clostridium genera increased, while the abundance of the Prevotella genus decreased. Thus, FMT can be utilized both to establish PCOS animal models and potentially to modify PCOS-related phenotypes. Among current potential therapeutic strategies, FMT is considered the most holistic approach, as it encompasses the entire microbial ecosystem, including the bacteriome, virome, fungome, archaeome, and even the parasitome. However, research on FMT is predominantly based on animal models. Its efficacy and safety in humans require further validation and clarification, along with the need to explore optimal clinical implementation protocols.

Probiotics are active microorganisms in the gut that exert beneficial effects by maintaining microbial balance and regulating host immunity. Common genera include Lactobacillus, Bacillus, Bifidobacterium, Streptococcus, and Enterococcus (120). Prebiotics are organic substances that selectively promote the growth and metabolism of beneficial gut bacteria, such as galactooligosaccharides, fructooligosaccharides, inulin, and lactulose (121), and can improve the structure and function of the gut microbiota. Probiotics and prebiotics have a positive impact on the gut microecology, demonstrating significant efficacy in improving gut health, regulating immune responses, and assisting in the treatment of various diseases, thus offering new strategies and directions for biological therapy (122–124). Studies by Zhang et al. (35) indicate that dietary supplementation with Bifidobacterium lactis V9 exerts regulatory effects on the gut microbiota and disease indices in PCOS patients. The probiotic, through effective colonization, significantly promotes the abundance of SCFAs-producing bacteria, such as Faecalibacterium, Akkermansia, and Butyricimonas, thereby directly elevating intestinal SCFAs levels. The increase in SCFAs levels, on one hand, lowers intestinal pH and enhances gut barrier integrity, and on the other hand, competitively inhibits the proliferation of LPS-producing pro-inflammatory bacteria, including Parabacteroides, Clostridium, and Escherichia/Shigella, consequently alleviating chronic low-grade inflammation mediated by LPS. Ultimately, this mechanism effectively reduces the LH/FSH ratio and T levels, and improves IR and dyslipidemia. Furthermore, prebiotics, such as inulin, have also shown positive effects. In a PCOS mouse model, inulin decreased the Firmicutes/Bacteroidetes ratio, increased the abundance of SCFAs-producing bacteria, and upregulated tight junction protein expression, enhancing intestinal barrier integrity. Simultaneously, inulin attenuated inflammation by inhibiting the TLR4/MyD88/NF-κB pathway and NLRP3 inflammasome in the ovaries. Based on the mechanism described above, inulin can improve glucose tolerance and insulin sensitivity, reduce the number of ovarian cystic follicles, increase the number of corpora lutea, and restore sex hormone levels and the estrous cycle (125).

TCM, encompassing Chinese herbal medicine, acupuncture, and dietary therapy, has a history of several centuries in China and shows significant potential in the treatment of PCOS, with its role in regulating gut microbiota gaining increasing attention. Recent studies indicate that certain TCM formulas may exert therapeutic effects by altering the composition of the gut microbiota. Therefore, identifying active compounds or formulas from Chinese herbs that target the gut microbiota for PCOS treatment has become a promising area of research (126, 127). Bailin (BL), a proprietary Chinese medicine composed of Cordyceps sinensis, has the effect of warming yang and tonifying qi. Research indicates that BL improves PCOS by modulating gut microbiota, specifically by reducing pro-inflammatory bacteria such as Akkermansia and increasing beneficial bacteria such as Lactobacillus. Furthermore, BL protects intestinal barrier integrity, reduces gut-derived LPS release, and activates the IRS1-PI3K-PKB insulin pathway to improve IR. Concurrently, it inhibits the LPS-TLR4 inflammatory pathway, thereby alleviating chronic inflammation (64). Zishen Qingre Lishi Huayu recipe (ZQLHR), a patented Chinese herbal formula, has also shown significant efficacy in PCOS treatment, effectively improving patients’ spontaneous ovulation rate and ovulatory menstruation rate. Its mechanism of action includes promoting the enrichment of butyrate-producing bacteria like Lachnospira and Faecalibacterium, and reducing pro-inflammatory bacteria such as Escherichia-Shigella, which collectively contribute to significant improvement in host metabolic function. FMT experiments have confirmed its ability to improve ovarian structure in PCOS mice and regulate serum concentrations of acetate, butyrate, and succinate (128). Additionally, acupuncture, as an ancient TCM therapy, can significantly regulate intestinal function and microbial composition via the gut-brain axis, thereby improving PCOS (129). Studies have found that electroacupuncture can increase the abundance of Tenericutes and Prevotella_9, improve the estrous cycle, sex hormone levels, and glucose tolerance in rats, reduce body weight and visceral and subcutaneous fat content, and increase BAT weight (130).