Bacterial biomarkers for the diagnosis of polycystic ovary syndrome in obese individuals

Abstract

The present invention relates to the diagnosis, risk prediction, or prognosis monitoring of polycystic ovary syndrome (PCOS) in obese people. Based on sequencing and bioinformatics analysis, biomarkers for PCOS in obese people are provided, along with a PCOS screening kit, providing a highly sensitive and specific detection tool for clinical diagnosis and treatment.

Description

Technical Field

[0001] The present invention belongs to the field of biomedicine technology, and specifically relates to an intestinal flora marker and its application in the diagnosis, disease risk prediction or prognosis monitoring of polycystic ovary syndrome in obese people. Background Art

[0002] Polycystic ovary syndrome (PCOS) is a common female reproductive endocrine and metabolic disorder, with clinical manifestations of hyperandrogenism, chronic anovulation, and polycystic ovaries. The prevalence in women of childbearing age can reach 6% to 20%. PCOS is one of the most common endocrine diseases and can cause infertility in women of childbearing age. Many studies have shown that PCOS patients have a higher risk of metabolic-related diseases, such as insulin resistance (IR), type 2 diabetes mellitus (T2DM), dyslipidemia, and cardiovascular disease. To date, the cause of PCOS is still unclear. Researchers speculate that multiple factors, such as congenital heredity, intrauterine environment, lifestyle, and potential changes in intestinal flora, may promote the occurrence and development of the disease.

[0003] There is growing evidence that intestinal bacteria play an important role in obesity, obesity-related inflammation, and insulin resistance. For example, Professor Zhao Liping of Shanghai Jiao Tong University found that short-chain fatty acid (SCFA)-producing bacteria may improve glycated hemoglobin (HbAlc) levels in patients with T2DM by increasing the production of glucagon-like peptide-1 (GLP-1). The commensal bacterium Bacteroides thetaiotaomicron can reduce plasma glutamate concentrations and alleviate diet-induced weight gain and obesity in mice, and its abundance is associated with the efficiency of bariatric surgery. Recent studies on PCOS and intestinal flora have shown that PCOS patients have reduced bacterial alpha diversity and the abundance of Bacteroidaceae, Clostridiaceae, Erysipelotrichidae, Lachnospiraceae, Lactobacillaceae, Porphyromonadaceae, Prevotellaceae, Ruminococcaceae, and Actinobacteria. Qi et al. found that the abundance of Bacteroides vulgatus in the intestines of PCOS patients was significantly increased, while the levels of glycodeoxycholic acid and tauroursodeoxycholic acid were decreased. Hyperandrogenism is associated with intestinal microbial dysbiosis, suggesting that androgens may regulate the intestinal microbiome, suggesting that manipulating the intestinal microbiome may be a potential treatment for PCOS. For example, transplanting the microbiota of women with PCOS into mice induced a PCOS-like phenotype, while transplanting the intestinal microbiota of healthy women alleviated PCOS symptoms in mice.

[0004] Body mass index (BMI) is a commonly used international indicator for measuring obesity and health. A normal BMI ranges from 20 to 25; a value above 25 is considered overweight, and a value above 30 is considered obese. PCOS patients experience IR, gut microbiome imbalance, and chronic inflammation, which are further exacerbated in obese patients. Traditional clinical diagnosis of PCOS relies primarily on menstrual patterns, androgen levels, and ovarian ultrasound. According to the Rotterdam ESHRE / ASRM (2003) criteria, two of the following three criteria must be met: oligomenorrhea; clinical or biochemical androgen excess; and bilateral or unilateral polycystic ovarian changes and / or enlargement. However, PCOS diagnosed using these methods is highly heterogeneous and fails to incorporate the degree of metabolic abnormalities, a key clinical characteristic, into subtypes of PCOS patients. Recent studies suggest that gut microbiome disturbances play a significant role in the development and progression of PCOS. However, relatively little research has examined the relationship between PCOS and obesity. Further research is needed to understand the differences in gut microbiota between healthy women with a high BMI and those with PCOS, and whether PCOS risk assessment can be performed based on gut microbial abundance in healthy women with a high BMI. Variations in gut bacterial abundance may provide new avenues for characterizing or diagnosing PCOS.

[0005] Therefore, providing a group of bacterial markers that can be used to detect polycystic ovary syndrome can be used for rapid and efficient diagnosis of polycystic ovary syndrome patients in obese populations. Summary of the Invention

[0006] The purpose of the present invention is to solve the problems existing in the prior art and propose a genetic detection kit for polycystic ovary syndrome in obese people and its preparation method and application.

[0007] In one aspect, the present invention provides a biomarker that can be used for diagnosing polycystic ovary syndrome, predicting disease risk, or monitoring prognosis in obese people. The biomarker is the bacterium Erysipelotrichaceae UCG-003. The biomarkers also include Collinsella, Barnesiella, Blautia, Ruminococcaceae_UCG.005, Escherichia.Shigella, Butyricicoccus, Agathobacter, Dorea, Lachnoclostridium, Lachnospira, Coprococcus_3, Faecalibacterium, Dialister, Anaerostipes, Paraprevotella, Ruminoc occaceae_UCG.002, Sutterella, Romboutsia, Ruminococcus_1, Ruminococcus_torques_group, Christensenellaceae_R.7_group, Odoribacter, Subdoligranulum, Phascolarctobacterium, Eggerthella, Bilophila, Ruminococcus_gnavus_group and Parabacteroides bacteria.

[0008] Another aspect of the present invention provides a kit for diagnosing, predicting the risk of polycystic ovary syndrome, or monitoring prognosis in obese people, the kit comprising a reagent for detecting the bacterial content of the biomarker, wherein the reagent is selected from a PCR method, a microarray method, a sequencing method, or an immunoassay reagent.

[0009] Another aspect of the present invention provides the use of a reagent for detecting biomarkers in preparing a kit for diagnosing polycystic ovary syndrome in obese people, predicting disease risk, or monitoring prognosis.

[0010] Compared with the existing technology, the patent of this invention has the following beneficial technical effects: by studying feces to reveal the differences in intestinal bacteria between PCOS patients and healthy people, the risk assessment and early diagnosis of PCOS can be accurately performed. Based on the comparison and analysis of the intestinal flora of PCOS patients and healthy people, in particular, the difference in flora between healthy women with high BMI and PCOS patients with high BMI, combined with the characteristic bacterial data of PCOS people with high BMI and healthy women with high BMI as a training set, it can accurately assess the risk of PCOS patients and make early diagnosis. Compared with currently commonly used diagnostic methods, this method is convenient and fast.

[0011] The PCOS-related biomarkers proposed in this paper are valuable for early diagnosis. First, the markers have high specificity and sensitivity, effectively distinguishing PCOS patients with high BMI from healthy controls. Second, stool analysis is noninvasive, which can increase patient compliance. Characteristic bacterial groups identified by high-throughput sequencing can be identified using high-throughput sequencing or qPCR. Third, the indicator bacteria of this invention can also be used as a tool for monitoring treatment responses in PCOS patients, allowing the abundance of characteristic bacteria to guide treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The above and other objects, features and advantages will become apparent from the following description of specific embodiments of the present invention, as illustrated in the accompanying drawings, which are not necessarily to scale, emphasis instead being placed on illustrating the principles of various embodiments of the present invention.

[0013] Figure 1 Comparison of microbial diversity among the LC, OC, LP, and OP groups. (A) α diversity; (B) β diversity.

[0014] Figure 2 Comparison of bacterial community composition and structure among the LC, OC, LP, and OP groups. (A) Phylum level; (B) Genus level.

[0015] Figure 3 This is the LEfSe analysis among the four groups: LC, OC, LP and OP.

[0016] Figure 4 (A) Importance ranking of feature species for the disease model constructed for the OC and OP groups based on the random forest algorithm; (B) ROC curve constructed based on feature values.

[0017] Figure 5 Ranking of the importance of characteristic species groups for the disease models constructed for the LC and LP groups based on the random forest algorithm. DETAILED DESCRIPTION

[0018] The present invention will be further described below with reference to specific examples so that those skilled in the art can better understand the present invention and implement it, but the examples are not intended to limit the present invention.

[0019] Example 1

[0020] 1.1 Sample Collection

[0021] According to the 2003 Rotterdam criteria, women diagnosed with PCOS should meet the following criteria: (1) oligoovulation and / or anovulation; (2) clinical and / or biochemical signs of androgen excess; and (3) ultrasound findings of polycystic ovaries in one or both ovaries, ≥12 follicles with a diameter of 2-9 mm, and / or an ovarian volume ≥10 mL. All recruited PCOS patients were newly diagnosed patients and had not received PCOS-related treatment. A total of 41 healthy women and 47 women with PCOS were included. The population was divided into normal BMI (BMI < 24) and high BMI (BMI ≥ 24) groups based on BMI = 24. Therefore, the number of normal BMI healthy women (LC), high BMI healthy women (OC), normal BMI PCOS patients (LP), and high BMI PCOS patients (OP) were 21, 20, 22, and 25, respectively.

[0022] 1.2 DNA extraction and 16S rRNA gene sequencing

[0023] On the day of the physical examination, stool samples were collected from healthy individuals and PCOS patients and immediately frozen at -80°C. DNA was extracted from approximately 200 mg of stool according to the kit instructions. 16S rRNA gene amplification was performed using the forward primer: 5'-CCTACGGRRBGCASCAGKVRVGAAT-3'; reverse primer: 5'-GGACTACNVGGGTWTCTAATCC-3'. DNA libraries were quality-checked using Qubit 3.0 and sequenced on an Illumina instrument (Illumina, San Diego, CA, USA).

[0024] 1.3 Sequencing Data Analysis

[0025] Fastp software (version 0.19.6) was used to trim the raw sequencing data for quality and length, and to remove Illumina adapters. 16S sequencing data were processed and analyzed using QIIME2 (version 2019.4). Sequencing primers were removed using cutadapt, and noise reduction was performed using Dada2. Representative sequences and species abundance tables based on these sequences were generated using Dada2. Sequences were annotated to species using the q2-feature-classifier classify-sklearn.

[0026] 1.4 Statistical analysis

[0027] The Shannon index of bacterial alpha diversity was calculated based on the OTUs table. For the OTUs table, the distance matrix was calculated based on the Bray Curtis distance and the PCoA diagram was drawn. The Kruskal-Wallis method was used for statistical testing, and the Benjamini-Hochberg (BH) FDR method was used for p-value correction. The LDA cutoff value of the linear discriminant analysis effect size (LEfSe) was 2, which was used to identify differential bacterial groups that distinguished between groups, thereby finding statistically significant biomarkers. The RandomForest algorithm in the randomForest R package was used for training based on bacterial classification or characteristic OTU features. The entire dataset was divided into training (80%) and testing (20%) datasets. In the testing phase, the predictive performance of each machine learning model was evaluated by performance parameters such as the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity.

[0028] Example 2

[0029] 2.1 Analysis of bacterial α and β diversity

[0030] like Figure 1 As shown in A, there was no significant difference in the Shannon index of bacterial α diversity among the four groups of LC, OC, LP and OP (p>0.05). However, the OC group showed higher bacterial diversity and the OP group showed lower bacterial diversity. PCoA analysis based on Bray-Curtis distance found that the bacterial community structures of the four groups were relatively similar, and PERMANNOVA analysis showed that the differences were not significant (p>0.05) ( Figure 1 B).

[0031] 2.2 Analysis of bacterial flora structure

[0032] like Figure 2 As shown in A, from the bacterial phylum level, the abundance of Firmicutes, Bacteroidetes, Proteobacteria and Actinobacteria dominated in healthy people and PCOS patients. Among them, the average abundance of Firmicutes in the OC group was the lowest at about 53%, while the average abundance of Bacteroidetes and Proteobacteria was the highest among the four groups, while the abundance of these three bacterial phyla in the LP group was exactly the opposite of that in the OC group. From the bacterial genus level, the abundance of Bacteroides, Prevotella_9, Faecalibacterium and Roseburia dominated in healthy people and PCOS patients ( Figure 2 B).

[0033] 2.3 Bacterial marker analysis

[0034] LEfSe analysis of characteristic markers across the LC, OC, LP, and OP groups revealed that each group had its own distinct markers. For example, Erysipelotrichaceae UCG-003 was a characteristic marker for the OP group; Agathobacter, Lactobacillaceae, and Lactobacillus were characteristic markers for the LP group; Blautia, Dorea, the Eubacterium eligens group, and Coprococcus_1 were characteristic markers for the LC group; and Ruminococcus_1, Parabacteroides merdae, and Collinsella were characteristic bacterial markers for the OC group, representing 22 groups. Even when an LDA value of 3.5 was used as the cutoff, characteristic bacterial groups such as Agathobacter, Blautia, Dorea, and Ruminococcus_1 were still observed. These results suggest that these bacterial types may serve as important disease indicator microorganisms.

[0035] 2.4 Analysis of bacterial community structure

[0036] In order to screen out characteristic microbial groups that can distinguish OC and OP groups and thus facilitate disease diagnosis, we used the random forest algorithm to rank the feature importance of the entire OTU group of OC and OP groups. Based on the average reduced Gini index, it can be found that Collinsella, Erysipelotrichaceae UCG_003, Bamesiella, Blautia, Ruminococcaceae UCG_005, Escherichia / Shigella and Butyricicoccus, Agathobacter and Dorea are the most important bacterial groups in the model ( Figure 4 A). The training set model constructed with the top 30 important features has an AUC of 0.70 for distinguishing the OC group from the OP group ( Figure 4 B). Similarly, we also attempted to construct a random forest model for disease diagnosis in the LC and LP groups from the perspective of bacteria. Based on the average reduction of the Gini index, we screened out important characteristic bacterial groups such as Coprococcus_1, Lachnospiraceae UCG_010, Bifidobacterium, Alistipes and Streptococcus ( Figure 5 ). Therefore, the characteristic bacteria have a good discriminative effect on OC and OP groups.

[0037] In conclusion, based on random forest and LEfSe analyses, we found that the biomarker combination of Collinsella, Erysipelotrichaceae UCG-003, Barnesiellaceae, Blautia, Ruminococcaceae UCG_005, and Butyricicoccus bacteria can be used as a marker for whether women with high BMI have PCOS or to predict the risk of PCOS in subjects.

[0038] It should be understood that while the invention has been particularly shown and described with reference to exemplary embodiments thereof, it should be understood by those skilled in the art that various changes in form and details may be made therein and any combination of various embodiments may be made without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A biomarker for diagnosing polycystic ovary syndrome in obese people, characterized by: The biomarker is composed of Erysipelotrichaceae UCG-003, Collinsella, Barnesiella, Blautia, Ruminococcaceae_UCG.005, Escherichia.Shigella, Butyricicoccus, Agathobacter, Dorea, Lachnoclostridium, Lachnospira, Coprococcus_3, Faecalibacterium, Dialister, Anaerostipes, Paraprevotella, Ruminococcace ae_UCG.002, Sutterella, Romboutsia, Ruminococcus_1, Ruminococcus_torques_group, Christensenellaceae_R.7_group, Odoribacter, Subdoligranulum, Phascolarctobacterium, Eggerthella, Bilophila, Ruminococcus._gnavus_group and Parabacteroides bacterial composition.

2. The biomarker according to claim 1, characterized in that The biomarker is gut bacteria.

3. A kit for diagnosing polycystic ovary syndrome in obese people, characterized in that: Contains a reagent for detecting the bacterial content of the biomarker according to any one of claims 1-2.

4. The kit according to claim 3, wherein The detection reagent is selected from PCR method, microarray method or immunoassay reagent.

5. Use of a reagent for detecting the biomarker according to any one of claims 1 to 2 in preparing a diagnostic kit for polycystic ovary syndrome in obese people.

Images