Abstract
The association between gut microbiome and chronic metabolic disease including polycystic ovary syndrome (PCOS), is well documented, however, the relationship between the gut microbiota and serum metabolites remains unknown. In this study, untargeted metabolomics together with a 16S rRNA gene sequencing tool was used to detect small molecule serum metabolites and the gut microbiome. We identified 15 differential metabolites between PCOS patients and the healthy control. Lysophosphatidylcholine (LPC) (18:2, 20:3, 18:1, P-16:0, 17:0, 15:0, 18:3, 20:4), phosphatidylcholine(PC), ganglioside GA2 (d18:1/16:0) and 1-linoleoylglycerophosphocholine were increased in the PCOS group, and the concentrations of phosphoniodidous acid, bilirubin, nicotinate beta-d-ribonucleotide and citric acid were decreased in the PCOS group, suggesting a lipid metabolism and energy metabolism disorder in the PCOS patients. The diversity of gut microbiota in PCOS group was lower than that in healthy controls. Escherichia/Shigella, Alistipes and an unnamed strain 0319_6G20 belonging to Proteobacteria were important distinguishing genera (LDA > 3.5) in PCOS. Prevotella_9 was positively correlated with phosphoniodidous acid, nicotinate beta-d-ribonucleotide and citric acid concentrations, and negatively correlated with the concentration of LPC (20:3) and 1-linoleoylglycerophosphocholine; Roseburia was negatively correlated with LPC concentration (20:4), while the characteristic genus 0319_6G20 of PCOS was positively correlated with LPC concentration (20:3) (COR > 0.45). SF-36 in the PCOS group was significantly lower than that in the healthy control (HC) group, which was associated with the presence of Escherichia-Shigella and Alistipes. Our finding demonstrated the correlation between the gut microbiota and serum metabolites in PCOS, and therefore characteristic gut microbiota and metabolites may play an important role in the insulin resistance and the mood changes of PCOS patients.
Similar content being viewed by others
Introduction
Polycystic Ovary Syndrome (PCOS) occurs mostly in women of reproductive age. It is a complex syndrome characterized by excessive androgen, ovulation dysfunction, and organic polycystic ovary. Most patients also suffer from insulin resistance and lipid metabolism disorders1. It has been estimated that the clinical morbidity of PCOS in women of reproductive age worldwide is as high as 10%. According to the diagnostic criteria of the AE-PCOS Society and Rotterdam, the incidence can be as high as 15.3% and 19.9%, respectively2. PCOS can cause irregular menstrual cycles, infertility, and metabolic syndromes such as obesity, lipid metabolism disorders, insulin resistance, etc. It has become an important public health problem affecting women's physical and mental health.
However, increasing evidence illustrates a more multi-factorial and complex nature of the syndrome not previously apparent arising from a combination of genetic, psychological and environmental factors3. The relationship between gut microbiota dysbiosis and PCOS is attracted increasing attention. Kelley et al. found that the intestinal microbe composition changed significantly in a mice model study using letrozole to induce PCOS. The treatment of adolescent female mice with letrozole reduced the intestinal flora diversity and resulted in a species-specific and time-dependent shifts in the relative abundance of in particular, Bacteroides and Firmicutes4. Another study showed Lactobacillus, Ruminococcus, and Clostridium abundance were lower in letrozole-treated PCOS rats5. Microbiota changes in mice and rat model were also demonstrated in human studies. Clinical trials showed a lower diversity and a modified phylogenetic profile in PCOS patient stool microbiomes6. In addition, the abundance of Escherichia-Shigella and Streptococcus in PCOS patients were increased, and the abundance of Akkermansia and Rumenococcus bacteria were decreased in other human studies7.
Metabolomics, using the qualitative and quantitative analysis of blood, urine, feces and other body fluids, improves our understanding of how specific metabolites, diseases and their phenotypic changes correlate. The human body is now considered as a superorganism as trillions of commensal microbes live on and inside the body, interacting through the process of metabolic exchange and "co-metabolism"8. The combination of serum metabolomics and 16S rRNA gene sequencing helps explain the close relationship between the gut microbiota and the host.
Using the non-targeted metabolomics technology to detect the serum and follicular fluid of PCOS patients, Xu et al. found that small molecules such as 1-methylhistidine, threonine and citric acid had underwent significant changes9. Zhao et al. found that discrete metabolites in PCOS patients are closely related to clinical symptoms10. Therefore, we recruited PCOS patients and healthy controls (HC) to study the relationship between the gut microbiota and metabolic changes in PCOS patients. Our research shows that the changes in the gut microbiota are related to host metabolism in PCOS patients, and the microbiota profile associated with the psychological state of PCOS, which provided a new perspective to explain the etiology and pathogenesis of PCOS.
Results
Baseline information
The study subjects were all women from the Pixian area of Chengdu, China. There was no significant difference in age between the PCOS patients and the healthy subjects (p > 0.05). The BMI of PCOS patients was increased compared with health subjects (p < 0.05), but the mean value is lower than 24. In a Chinese study, the authors recruited 999 volunteers with PCOS in Southern China, finding that the proportion of PCOS patients with a BMI above 23 kg/m2 was 34.63%11. It suggests that PCOS patients in China tend to have a lower BMI compared to world average, approximately 50% of PCOS women are overweight or obese12. Compared with healthy controls, the serum testosterone (T), luteinizing hormone (LH), LH/follicle-stimulating hormone (FSH) ratio and fasting insulin levels of PCOS patients were higher than those of the control group (p < 0.05). Further, the quality of life score SF-36 was lower than that of the control group (p < 0.01) (Table 1).
Potential serum metabolomic biomarkers for PCOS
We found 513 discernible features in positive ion mode, and 202 features in negative ion mode (supplementary Tables 1 and 2). In order to determine the difference in the metabolites between the two groups in our experiment design, partial least squares discriminant analysis (PLS-DA) was applied. The PLS-DA Scores plot is shown in Fig. 1, from which it can be seen that HC group and PCOS group are completely separated. The Variable Importance in the Projection (VIP) value of the PLS-DA model (threshold > 1) was adopted, combined with an independent sample T-test (P < 0.05) to search for different metabolites. A total of 15 different metabolites were detected (p < 0.05, VIP > 1, Table 2). The concentration of lysophosphatidylcholine (LPC) (18:2, 20:3, 18): 1, P-16:0, 17:0, 15:0, 18:3, 20:4), Phosphatidylcholine (PC), Ganglioside GA2 (d18:1/16: 0)), 1-Linoleoylglycerophosphocholine were higher in the PCOS group, while phosphoniodidous acid, bilirubin, nicotinate beta-d-ribonucleotide and citric acid concentrations were lower. Using MetaboAnalyst 3.0 to further analyze the differential metabolites and their related metabolic pathways, we found metabolites related to glycerophospholipid metabolism, such as LPC (18:2, 20:3, 18:1, P-16:0, 17:0, 15: 0,18:3,20:4), PC, 1-linoleoylglycerophosphocholine were enriched in the PCOS group; conversely, metabolites that related to energy metabolism, such as citric acid in the tricarboxylic acid cycle, the intermediate product nicotinate beta-d-ribonucleotide in the nicotinate and niacinamide metabolic pathway were significantly reduced in the PCOS group (Table 2).
The altered gut microbiota in women with PCOS
We performed 16S rRNA gene V3-V4 regions sequencing to evaluate the gut microbiota of the PCOS patients. Using QIIME2 (2019,4), a total of 4125 ASVs were identified. The number of overlapped ASVs between the two groups was 1179, and the PCOS group had 1409 unique ASV sequences, while the healthy controls had 1519 unique ASV sequences (Fig. 2A). The rank abundance curve demonstrates the species richness and evenness. At ASV level, we found the rank-abundance curves of PCOS showed downward trends compared with the control group, suggesting the diversity of gut microbiota in the PCOS group was reduced (Fig. 2B). To further verify the difference of diversity and evenness between the two groups, we did alpha diversity analysis. It reveals the Chao1 index and Shannon index in the PCOS group were significantly lower than those of the healthy controls (P < 0.05) (Fig. 2C). Principal coordinates analysis (PCoA) and non-metric multidimensional scaling (NMDS) analysis of all the subjects displayed the dysbiosis caused by PCOS (Fig. 2D,E), in accordance with previous studies13,14.
We then focused on the differences between groups on the taxonomic levels. At the phylum level, compared with healthy controls, the abundance of the Firmicutes, Bacteroidetes, Actinobacteria, Tenericutes, and Gemmatimonadetes in the PCOS group were decreased, and the proportion of Proteobacteria, Verrucomicrobia, Fusobacteria, Acidobacteria, and Cyanobacteria were increased (Fig. 3A supplementary Table 3). Among the top ten most abundant genera, the proportions of Escherichia-Shigella, Megamonas and Parasutterella in the PCOS group were increased, while the levels of the remaining seven genera were decreased (Fig. 3B supplementary Table 4. LefSe analysis showed that Escherichia-Shigella, Alistipes and an unnamed genus 0319_6G20 belonging to Proteobacteria were the most important characteristic genera of the PCOS group (LDA > 3.5) (Fig. 3C,D).
Correlation analysis of serum metabolites and intestinal microbiota in PCOS group
The Pearson correlation analysis between 15 important metabolites in serum and the gut microbiota of PCOS patients was carried out at the genus level. We found that Prevotella_9 was positively correlated with the concentrations of Phosphoniodidous acid, nicotinate beta-d-ribonucleotide, and citric acid, and negatively correlated with the concentrations of LPC (18:2) and 1-Linoleoylglycerophosphocholine; Roseburia was negatively correlated with the concentrations of LPC (20:4), while the characteristic genus 0319_6G20 of PCOS is positively correlated with the concentration of LPC (20:3) (Pearson Correlation Coefficient > 0.45) (Fig. 4).
Discussion
In this study, we found the BMI of PCOS patients was increased compared with health subjects, but the mean is lower than 24. In a Chinese study, the authors recruited 999 volunteers with PCOS in Southern China, finding that the proportion of PCOS patients with a BMI above 23 kg/m2 was 34.63%13. It suggests that PCOS patients in China tend to have a lower BMI compared to world average, approximately 50% of PCOS women are overweight or obese14.
This study revealed a significant correlation between serum small molecular metabolites and the gut microbiota in PCOS patients, based on the analysis of the relative differences in serum metabolites in the PCOS and healthy controls. The important metabolic pathways were screened and 15 characteristic metabolites in serum were identified, which we suggest as potential biomarkers in future PCOS research. The 16S sequencing of fecal samples displayed a compositional dysbiosis of gut microbiota in the PCOS group. Furthermore, correlation analysis showed that the gut microbiota was associated with alterations in specific serum metabolites.
In this study, we found that in contrast to healthy controls, among the 15 different serum metabolites in PCOS samples, 13 are relating to the glycerophospholipid metabolism pathway, and 2 to energy metabolism, suggesting significant abnormalities with fat metabolism in patients with PCOS. Dysfunction of glycerophospholipid in PCOS patients' follicles play a discernible association with declining the 2 pronuclei (PN) fertilization rate during IVF procedure15. The abnormal PC/LPC ratio in PCOS, which may result in changes in serum arachidonic acid concentration, was regulated by circulating insulin and androgens. Further understanding of the molecular mechanisms that lead to the altered lipid profiles identified here, together with genomic and proteomic studies, may provide new insights into the pathogenic mechanisms of PCOS and inform novel therapeutic strategies16.
Following analysis of the alpha and beta diversity of the gut microbiota of the PCOS and control group showed that there is an imbalance in the microbial composition of PCOS patients, which is consistent with previous studies6,7,17. Compared with the control group, the ratio of Firmicutes/Bacteroidetes in PCOS decreased, while the Proteobacteria, Verrucomicrobia and Fusobacteria increased, and the Actinobacteria decreased. Generally, a healthy human gut microbial community is mainly composed of Firmicutes and Bacteroidetes, and the decline in the ratio of Firmicutes/Bacteroidetes is implicated in obesity18. The abundance of Proteobacteria and Verrucomicrobia in the gut of patients with type 2 diabetes is significantly increased19. In the intestinal microbial community, Actinomycetes account for a relatively small proportion, comparing to Firmicutes and Bacteroidetes, but studies have found that it plays an important role in maintaining the homeostasis of the intestinal microbiota. The genus of Actinomycota can produce large amounts of short-chain fatty acids (SCFAs), which provide energy for intestinal epithelial cells and protect the intestinal barrier from bacterial infections20. Fusobacterium is a commensal organism common in the oral cavity and known to be pathogenic under some conditions, associated with several human diseases, especially gastrointestinal disorders21.
Further, findings in this study show that among the top ten genera in relative abundance, the proportion of Escherichia-Shigella, Megamonas, and Lachnospira in the PCOS group were increased, while in the healthy control group and beneficial Roseburia and Bifidobacterium are increased Escherichia-Shigella, Alistipes and an unnamed genus 0319_6G20 belonging to Proteobacteria were identified as the most important characteristic genus in the PCOS group. Studies have shown that Escherichia-Shigella22 and Alistipes23 are significantly increased in the gut microbiota of patients with depression. In our study, based on the SF-36 questionnaire survey, we also found that the quality of life scores of the PCOS patients were significantly lower than those of healthy controls, indicating a correlation between depression and dysbiosis of gut microbiota in this population24. The brain-gut-axis is an interrelated system that affects both neural functions and eating behaviour25. Changes in gut microbiota affect the brain's physiological, behavioral, and cognitive functions through the influence of hormones, immune factors, and metabolites26 Alistipes is an indole-positive organism, and therefore decreases serotonin availability, which is associated with depression23. A previous study has displayed that Escherichia/Shigella is negatively correlated with the concentration of ghrelin7. Serotonin, peptide YY (PYY) and ghrelin are mediators of the brain–gut axis. This may shed new light on why PCOS patients are more prone to depression than healthy controls. Nevertheless, further mechanism studies are needed to prove whether the changes of gut microbiota are associated with the depression tendency.
Correlation analysis shows that some key gut microbial members were associated with the potential serum biomarkers of PCOS. Findings illustrate that among the top ten abundant bacterial genera all subjects in this study, the abundance of Prevotella was significantly reduced in PCOS patients, and negatively correlated with the serum metabolites LPC(18:2) and 1-Linoleoylglycerophosphocholine, and positively correlated with the concentration of β-nicotinic acid nucleotide and citric acid; Roseburia was negatively correlated with the concentration of LPC(20:4), while the 0319_6G20 genus and LPC(20:4) were positively correlated (cor > 0.45).
LPCs are essential substances and they are associated with metabolic disorders, such as inflammatory diseases27. In-vivo studies showed LPC production by hydrolysis of phosphatidylcholine induced by phospholipase A2, the main phospholipid component of oxidized low-density lipoprotein, is related to the occurrence of atherosclerosis. LPC can activate RhoA, a GTPase protein, through the PKC-α pathway and thus cause the dysfunction of the endothelial barrier. It also damage thw endothelial cells by preventing the synthesis of endogenous relaxing factors, as well as increasing the expressions of monocyte chemotactic protein-1 and interleukin-8 in endothelial cells through the activation of the NADH/NADPH oxidase system, which can lead to atherosclerosis and inflammatory diseases28,29. Some studies showed that LPCs can cause insulin resistance in diabetic patients but meanwhile lower the blood sugar30. Knowledge of this apparently opposing effect of LPC on human health is currently limited, and is an area for further research. In this study the LPCs of PCOS patients were significantly increased, supporting a plausible reason for the relatively high morbidity rate and cardiovascular disease in PCOS patients.
1-Linoleoylglycerophosphocholine, the product of PC metabolic pathway, was previously associated with heightened insulin resistance31. Our clinical data showed that PCOS group fasting insulin levels are within the normal range, but higher than the HC group, suggesting 1-Linoleoylglycerophosphocholine concentration can predict the rise of insulin resistance. Several studies have shown that a range of fatty acids represented by LPC appear at elevated levels in various diseases and are closely associated with gut microbiota32,33. Prevotella is considered as a gut microbial commensal in healthy human who consumes plant-rich diet, thus it is cosidered as a beneficial microbe34. However, Prevotella has also been shown to be associated with various diseases such as hypertension33, rheumatoid arthritis, periodontitis, and metabolic disorders35. Interestingly, in our study, Prevotella is associated with bile acid levels, including concentration of chenodeoxycholic acid (CDCA) and ursodeoxycholic acid (UDCA) which were positively correlated. Bile salts possess direct antimicrobial activities, thus being able to shape the structure of gut microbiota36. On the other side, gut microbes are able to produce bile salt hydrolases to deconjugate bile acids37. Bile acids can act as signaling molecules that regulate host metabolism by binding to the nuclear receptor farnesoid X receptor (FXR) and the Takeda G-protein coupled bile acid receptor TGR5, involving in the regulation of lipid metabolism38. The cause might be Prevotella is a large genus that includes more than 50 different species. The limitations of the resolution Prevotella genera which does not take into account species and strain level attributes which most likely account for the good and bad effects.
The concentration of nicotinate beta-d-ribonucleotide and citric acid in PCOS patients were lower than those in healthy controls. Nicotinate beta-d-ribonucleotid is a precursor for the synthesis of nicotinamide adenine dinucleotide (NAD+), which is essential for cell energy metabolism, cell protection and biosynthesis processes39. Studies have shown that NAD+ is negatively correlated with PCOS: The concentration of NAD+ in ovarian granulosa cells (GCs) in PCOS patients is significantly lower than that in healthy controls. Restoring NAD+ levels in PCOS patients can reduce the mitochondrial dysfunction of GCs40. Citric acid is an important intermediate product of the tricarboxylic acid cycle, which can reduce lipid peroxidation and reduce inflammation41. In a PCOS mouse model, the kidney tricarboxylic acid cycle products (citric acid, fumaric acid and succinic acid) and NAD+ levels were significantly altered42. There are consistent with the results of this study suggesting that oxidative stress and energy metabolism disorders in PCOS patients, which could arise through one of the the pathogenesis indicators of PCOS. Previous studies also found that after 6 weeks of ingesting whole grains and high dietary fiber diets, healthy, overweight adults with high abundance of Prevotella lost more body weight than those with low Prevotella abundance43. This may relate to its effective utilization of complex carbohydrates44, improving glucose metabolism, and therefore promoting glycogen storage45. The Pearson correlation analysis showed Prevotella was positively correlated with of citric acid and β-nicotinic acid nucleotides concentration, which is related to energy metabolism, suggesting that Prevotella may be related to energy metabolism and thus influence the metabolism of PCOS patients.
Roseburia is a genus of bacteria that produce acetate, propionate and butyrate. Its abundance in the gut of people with metabolic and inflammatory diseases is reduced46, and its abundance is inversely correlated to the occurrence of coronary atherosclerosis47. Roseburia ferments plant fibers and produces organic acids, one of which is butyrate, which can improve colon movement, immune maintenance and anti-inflammatory effects48. This study found that Roseburia was negatively correlated with the concentration of LPC (20:4). Other studies have also found the elevated concentration of butyrate-producing bacteria, with reduced levels of certain plasma LPC49. Butyric acid, a short-chain fatty acid (SCFAs), acts as a signaling molecule, notably through the G-protein coupled receptors GPR43/FFAR2 and GPR41/FFAR3. Activation of GPR43 on L-cells increases secretion of glucagon-like peptide-1 (GLP-1) and acetate induces anti-lipolytic activity and improves glucose and lipid metabolism50.
The findings are presented here within the limitations of the study: the sample size is relatively small, stated correlations have not yet been further verified, and our research is based on 16S rRNA sequencing results within the limits of resolution. A future study is planned to compare these finding in higher sample sizes and in animal models, which should further extend our understanding through functional analysis of the microbiota. The longer-term aim of this work is to design metatranscriptomic and metametabolomic analysis techniques and possibly selectively managing gut microbiota of PCOS patients. In this study, for the first time, the results of serum metabolomics and gut microbiota were combined to explore the possible mechanism of metabolic disorders in PCOS patients. The findings support and inform a biological treatment of PCOS in patients. In summary, non-targeted metabolomics and 16S rRNA gene sequencing revealed characteristic changes in fecal metabolites and gut microbiota of PCOS patients. Escherichia-Shigella, Alistipes, and an unnamed genus 0319_6G20 in the phylum Proteobacteria were observed in PCOS group patients. Linoleoylglycerophosphocholine, LPC (18:2, 20:3, 18:1, P-16:0, 17:0, 15:0, 18:3, 20:4), phosphoniodidous acid, PC, bilirubin, ganglioside GA2, β-nicotinic acid nucleotide, citric acid are characteristic metabolites in the PCOS group. In these patients, there is a close correlation between the gut microbiota and the serum metabolites. Study of the characteristic gut microbiota and its metabolites should be further extended to, for example, examine relationships to the physiological and emotional changes in patients. In future research, we plan to use the findings here to design a target strain transplantation and when complete, compare with the findings of this study to develop a biogenic and effective treatment for PCOS.
Methods
Participants
From November 2015 to December 2017, PCOS patients (PCOS group, n = 20) attending the outpatient clinic of Pixian Hospital of Traditional Chinese Medicine, Republic of China were recruited. The gynecologists used Rotterdam criteria to diagnose PCOS. Participants who presented with two out of three following criteria were diagnosed as PCOS: (1) clinical and/or biochemical hyperandrogenism (hirsutism and/or an increased testosterone concentration); (2) oligo- and/or anovulation (menstrual cycles > 35 days and/or the absence of menstruation for at least 3 months); (3) polycystic ovaries (assessed through gynecological ultrasound or medical history).
The healthy control group aged 18–40 years old (HC group, n = 20) had no history of diagnosed PCOS and did not meet the Rotterdam criteria. All participants met the exclusion criteria, including no use of oral contraceptives, antiandrogens, or insulin sensitizers in the past three months; pregnancy; other known disorders that can cause hyperandrogen and ovulation, such as 21-hydroxylase deficiency, congenital adrenal hyperplasia, Cushing’s syndrome, androgen secretory tumors, thyroid disease and hyperprolactinemia; any mental or organic disease; the use of corticosteroids or sex steroids; drug and alcohol abuse in the past 2 years; and the use of antibiotics, probiotics or prebiotics in the past 3 months. All subjects were from Pidu district Chengdu city region in Sichuan province, China, to minimize any confounding influences related to geographical location. This study was approved by the Ethics Committee of the Hospital of Traditional Chinese Medicine in Pixian Chengdu, China (2015KL-001). Every subject completed an informed consent form in accordance with the Declaration of Helsinki.
Collection of peripheral venous blood and stool samples
All subjects’ peripheral venous blood was collected on the third day of the menstrual cycle and the levels of sex hormones (luteinizing hormone (LH), follicle stimulating hormone (FSH), and testosterone (T)) were determined to ensure that no abnormalities were present. Stool samples were obtained from the participants 3–5 days after menstruation. Sterile plastic spoons and plastic tubes were used by each subject to collect about 10 g of fresh stool samples. The blood and fecal samples were transported within 2 h and preserved at − 80 °C until further processing.
Non-targeted metabolomics analysis
The serum samples were thawed from − 80 °C to room temperature in the dark. 300 μl of methanol solution (Dannstadt, Gennany) containing internal standard (5ug/mL, l-2-chloro-phenylalanine) was added into 100ul serum, vortexed for 1 min, and centrifuged at 13,000 rpm for 15 min at 4 °C. 200 μl of the supernatant was added into a sample vial for subsequent metabolomics analysis. In addition, an equal volume of each sample solution without blood to be tested was mixed together, serving as the QC (Quality Control).
The LC–MS detection platform was: liquid chromatography Agilent 1290 Infinity UHPLC, mass spectrometry Agilent 6538 UHD and Accurate-Mass Q-TOF /MS, chromatographic column XSelect HSS T3 (2.1 × 100 mm, 2.5 m). The mass spectrum data was collected by Accurate-Mass Q-TOF/MS, and the working mode divided into positive ion and negative ion mode. The gas temperature was 350 °C, the intake air volume 11 L/min, the fragmentation voltage 120 V, and the center of mass data collected in the range of 50–1000 m/z. Agilent Masshunter Qualitative Analysis B.04.00 software (Agilent Technologies, USA) was used to convert the raw data into a common (mz.data) format. The XCMS package51 in R was used to perform downstream analysis, such as peak identification, retention time correction, and automatic integration. We did internal standard normalization, and the number of sample features was screened according to the 80% principle. After that, we obtained a visualization matrix containing sample name, m/z-RT pair, and peak area. Importation of the visualization matrix into the R language platform was done, using the Pareto Scaling method for normalization, and to perform multivariate statistical analyses.
Access to cited metabolomics data: MTBLS3975 (http://www.ebi.ac.uk/metabolights/MTBLS3975)52.
16S rRNA gene sequencing
Stool samples were thawed in 4 °C water, centrifuged for 10 min, and DNA was extracted using Axygen Axy Prep DNA Gel Kit (Axygen, USA). The amplification method was applied, with primers and F:AYTGGGYDTAAAGNG R:TACNVGGGTATCTAATCC. Library prep for V3-V4 region, and the library pool was submitted to the Illumina MiSeq platform with 2 × 250 paired-end sequencing. Data were analyzed using QIIME2 (2019.4)53 software. Taxonomic results were summarized in supplementary Table 5. Accession to cite for these SRA data: PRJNA779930.
Statistical analysis
Statistical analyses were performed using SPSS 22.0 version software (IBM Inc., USA). Continuous variables are described by mean and standard deviation (SD). If the changes in α and β diversity indexes of metabolites and microbial communities were normally distributed, a covariance analysis was performed; otherwise the non-parametric Kruskal Wallis test was used. Differences in metabolites and gut microbiota between the two groups were analyzed using two independent samples t test or Wilcoxon rank sum test. Partial least squares discriminant analysis (PLS-DA) was used to analyze the two groups of samples. VIP (Variable Importance in the Projection) values of PLS-DA model (threshold > 1) were used to search for differential metabolites combined with independent sample T-test (P < 0.05). The Pearson correlation was used to analyze the correlation between each discriminated metabolite and the gut microbiota. A two-tailed p < 0.05 threshold indicated that the differences were statistically significant.
Data availability
The 16S rRNA gene sequencing datasets generated and analyzed during the current study are available in the NCBI (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA779930/), and Non-targeted metabolomics analysis datasets generated and analyzed during the current study are available in the Metbolights (http://www.ebi.ac.uk/metabolights/MTBLS3975).
References
Goodarzi, M. O., Dumesic, D. A., Chazenbalk, G. & Azziz, R. Polycystic ovary syndrome: Etiology, pathogenesis and diagnosis. Nat. Rev. Endocrinol. 7, 219–231 (2011).
Yildiz, B. O., Bozdag, G., Yapici, Z., Esinler, I. & Yarali, H. Prevalence, phenotype and cardiometabolic risk of polycystic ovary syndrome under different diagnostic criteria. Hum. Reprod. 27, 3067–3073 (2012).
Escobar-Morreale, H. F. Polycystic ovary syndrome: Definition, aetiology, diagnosis and treatment. Nat. Rev. Endocrinol. 14, 270–284 (2018).
Kelley, S. T. et al. The gut microbiome is altered in a letrozole-induced mouse model of polycystic ovary syndrome. PLoS ONE 11, e0146509. https://doi.org/10.1371/journal.pone.0146509 (2016).
Guo, Y. et al. Association between polycystic ovary syndrome and gut microbiota. PLoS ONE 11, e153196. https://doi.org/10.1371/journal.pone.0153196 (2016).
Lindheim, L. et al. Alterations in gut microbiome composition and barrier function are associated with reproductive and metabolic defects in women with polycystic ovary syndrome (PCOS): A pilot study. PLoS ONE 12, e0168390. https://doi.org/10.1371/journal.pone.0168390 (2017).
Liu, R. et al. Dysbiosis of gut microbiota associated with clinical parameters in polycystic ovary syndrome. Front. Microbiol. 8, 324 (2017).
Nicholson, J. K., Holmes, E. & Wilson, I. D. Gut microorganisms, mammalian metabolism and personalized health care. Nat. Rev. Microbiol. 3, 431–438 (2005).
Xu, W. L. et al. Untargeted metabolomics analysis of serum and follicular fluid samples from women with polycystic ovary syndrome. Minerva Endocrinol. https://doi.org/10.23736/S0391-1977.20.03199-5 (2020).
Zhao, X. et al. How to screen and prevent metabolic syndrome in patients of PCOS early: Implications from metabolomics. Front. Endocrinol. 12, 659268. https://doi.org/10.3389/fendo.2021.659268 (2021).
Chen, X. et al. Appropriate BMI levels for PCOS patients in Southern China. Hum. Reprod. 25, 1295–1302 (2010).
Gambineri, A. et al. Obesity and the polycystic ovary syndrome. Int. J. Obes. 26, 883–896 (2002).
Zhou, L. et al. Characteristic gut microbiota and predicted metabolic functions in women with PCOS. Endocr. Connect. 9, 63–73 (2020).
Zhang, F. et al. Diversity of the gut microbiota in dihydrotestosterone-induced PCOS rats and the pharmacologic effects of diane-35, probiotics, and berberine. Front. Microbiol. 10, 175. https://doi.org/10.3389/fmicb.2019.00175 (2019).
Liu, L. et al. Follicular dynamics of glycerophospholipid and sphingolipid metabolisms in polycystic ovary syndrome patients. J. Steroid Biochem. Mol. Biol. 185, 142–149 (2019).
Li, S. et al. Discovery of novel lipid profiles in PCOS: Do insulin and androgen oppositely regulate bioactive lipid production?. J. Clin. Endocrinol. Metab. 102, 810–821 (2017).
Torres, P. J. et al. Gut microbial diversity in women with polycystic ovary syndrome correlates with hyperandrogenism. J. Clin. Endocrinol. Metab. 103, 1502–1511 (2018).
Koliada, A. et al. Association between body mass index and Firmicutes/Bacteroidetes ratio in an adult Ukrainian population. BMC Microbiol. 17, 120 (2017).
Zhao, X. et al. The alteration in composition and function of gut microbiome in patients with type 2 diabetes. J. Diabetes Res. 2020, 8842651. https://doi.org/10.1155/2020/8842651 (2020).
Binda, C. et al. Actinobacteria: A relevant minority for the maintenance of gut homeostasis. Dig. Liver Dis. 50, 421–428 (2018).
Han, Y. W. Fusobacterium nucleatum: A commensal-turned pathogen. Curr. Opin. Microbiol. 23, 141–147 (2015).
Rong, H. et al. Similarly in depression, nuances of gut microbiota: Evidences from a shotgun metagenomics sequencing study on major depressive disorder versus bipolar disorder with current major depressive episode patients. J. Psychiatr. Res. 113, 90–99 (2019).
Jiang, H. et al. Altered fecal microbiota composition in patients with major depressive disorder. Brain Behav. Immun. 48, 186–194 (2015).
Valles-Colomer, M. et al. The neuroactive potential of the human gut microbiota in quality of life and depression. Nat. Microbiol. 4, 623–632 (2019).
Zanchi, D. et al. The impact of gut hormones on the neural circuit of appetite and satiety: A systematic review. Neurosci. Biobehav. Rev. 80, 457–475 (2017).
Wang, H. X. & Wang, Y. P. Gut microbiota-brain axis. Chin. Med. J. 129, 2373–2380 (2016).
Matsumoto, T., Kobayashi, T. & Kamata, K. Role of lysophosphatidylcholine (LPC) in atherosclerosis. Curr. Med. Chem. 14, 3209–3220 (2007).
Schmitz, G. & Ruebsaamen, K. Metabolism and atherogenic disease association of lysophosphatidylcholine. Atherosclerosis 208, 10–18 (2010).
Mehta, D. Lysophosphatidylcholine: An enigmatic lysolipid. Am. J. Physiol. Lung Cell Mol. Physiol. 289, L174–L175 (2005).
Liu, P. et al. The mechanisms of lysophosphatidylcholine in the development of diseases. Life Sci. 247, 117443. https://doi.org/10.1016/j.lfs.2020.117443 (2020).
Ruiz-Argüelles, A., Méndez-Huerta, M. A., Lozano, C. D. & Ruiz-Argüelles, G. J. Metabolomic profile of insulin resistance in patients with multiple sclerosis is associated to the severity of the disease. Mult. Scler. Relat. Disord. 25, 316–321 (2018).
Nie, J., Zhang, L., Zhao, G. & Du, X. Quercetin reduces atherosclerotic lesions by altering the gut microbiota and reducing atherogenic lipid metabolites. J. Appl. Microbiol. 127, 1824–1834 (2019).
Li, J. et al. Gut microbiota dysbiosis contributes to the development of hypertension. Microbiome. 5, 14 (2017).
Ley, R. E. Gut microbiota in 2015: Prevotella in the gut: choose carefully. Nat. Rev. Gastroenterol. Hepatol. 13, 69–70 (2016).
Iljazovic, A., Amend, L., Galvez, E. J. C., de Oliveira, R. & Strowig, T. Modulation of inflammatory responses by gastrointestinal Prevotella spp.—From associations to functional studies. Int. J. Med. Microbiol. 311, 151472. https://doi.org/10.1016/j.ijmm (2021).
Tian, Y. et al. The microbiome modulating activity of bile acids. Gut Microbes. 11, 979–996 (2020).
Guzior, D. V. & Quinn, R. A. Review: Microbial transformations of human bile acids. Microbiome. 9, 140 (2021).
Ghaffarzadegan, T. et al. Determination of free and conjugated bile acids in serum of Apoe(−/−) mice fed different lingonberry fractions by UHPLC-MS. Sci. Rep. 9, 3800. https://doi.org/10.1038/s41598-019-40272-8 (2019).
Yang, Y. & Sauve, A. A. NAD(+) metabolism: Bioenergetics, signaling and manipulation for therapy. Biochim. Biophys. Acta. 1864, 1787–1800 (2016).
Wang, Y. et al. NAD+ deficiency and mitochondrial dysfunction in granulosa cells of women with polycystic ovary syndrome. Biol. Reprod. 105, 371–380 (2021).
Abdel-Salam, O. M. et al. Citric acid effects on brain and liver oxidative stress in lipopolysaccharide-treated mice. J. Med. Food. 17, 588–598 (2014).
Selen, E. S. et al. NMR metabolomics show evidence for mitochondrial oxidative stress in a mouse model of polycystic ovary syndrome. J. Proteome Res. 14, 3284–3291 (2015).
Christensen, L. et al. Prevotella abundance predicts weight loss success in healthy, overweight adults consuming a whole-grain diet ad libitum: A post hoc analysis of a 6-wk randomized controlled trial. J. Nutr. 149, 2174–2181 (2019).
Gálvez, E. J. C. et al. Distinct polysaccharide utilization determines interspecies competition between intestinal Prevotella spp. Cell Host Microbe. 28, 838–852 (2020).
Kovatcheva-Datchary, P. et al. Dietary fiber-induced improvement in glucose metabolism is associated with increased abundance of Prevotella. Cell Metab. 22, 971–982 (2015).
Machiels, K. et al. A decrease of the butyrate-producing species Roseburia hominis and Faecalibacterium prausnitzii defines dysbiosis in patients with ulcerative colitis. Gut 63, 1275–1283 (2014).
Kasahara, K. et al. Interactions between Roseburia intestinalis and diet modulate atherogenesis in a murine model. Nat Microbiol. 3, 1461–1471 (2018).
Tamanai-Shacoori, Z. et al. Roseburia spp.: A marker of health?. Future Microbiol. 12, 157–170 (2017).
Ryan, P. M. et al. Metformin and dipeptidyl peptidase-4 inhibitor differentially modulate the intestinal microbiota and plasma metabolome of metabolically dysfunctional mice. Can. J. Diabetes. 44, 146-155.e2 (2020).
Christiansen, C. B. et al. The impact of short-chain fatty acids on GLP-1 and PYY secretion from the isolated perfused rat colon. Am. J. Physiol. Gastrointest. Liver Physiol. 315, G53–G65 (2018).
Benton, H. P., Want, E. J. & Ebbels, T. M. Correction of mass calibration gaps in liquid chromatography-mass spectrometry metabolomics data. Bioinformatics 26, 2488–2489 (2010).
Haug, K. et al. MetaboLights: A resource evolving in response to the needs of its scientific community. Nucleic Acids Res. 48, D440–D444 (2020).
Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37, 852–857 (2019).
Funding
This study was supported by Innovation Team and Talents Cultivation Program of National Administration of Traditional Chinese Medicine (No: ZYYCXTD-D-202003), National Natural Science Foundation of China (82174517), and Science and Technology Department of Sichuan Province, basic research for application (No. 2015JY0214).
Author information
Authors and Affiliations
Contributions
Study conception and design: Z.Y. and E.Q.Q. Clinical subject recruitment: Y.Q.L. and J.Y. Methodology: Z.Y. and R.L. Writing—original draft: Z.Y., S.R.C. and E.Q.Q. Writing—review and editing: Z.Y., H.Y., T.X., R.L., F.R.L., J.Y. and S.G.Y. All authors approved the publication of this study.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Yu, Z., Qin, E., Cheng, S. et al. Gut microbiome in PCOS associates to serum metabolomics: a cross-sectional study. Sci Rep 12, 22184 (2022). https://doi.org/10.1038/s41598-022-25041-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-022-25041-4
- Springer Nature Limited
This article is cited by
-
The Novel Insight of Gut Microbiota from Mouse Model to Clinical Patients and the Role of NF-κB Pathway in Polycystic Ovary Syndrome
Reproductive Sciences (2024)
-
Perturbations in gut microbiota composition in patients with polycystic ovary syndrome: a systematic review and meta-analysis
BMC Medicine (2023)