Serum Metabolomics Study Of Polycystic Ovary Syndrome Based On UPLC/Q-TOF-MS Coupled With A Pattern Recongnition Approach

Polycystic ovary syndrome (PCOS) is a common endocrine and metabolic disorder in reproductive-age women. The etiology and pathogenesis are still uncertain now. Since PCOS has obviously clinical heterogeneity, serious psychological and physiological effects on patients, and higher risk for developing metabolic disorders and cardiovascular diseases, it is imperative to further study its pathogenesis and explore biomarkers for early diagnosis. Metabolomics (or metabonomics) is a new branch of science under the development of system biology recently. Based on the high throughput and highly sensitive analytical technology, metabolomics studies all metabolites (endogenous and exogenous metabolites) in biological systems, reflects the final results of perturbations upstream or directly from environment and reveals the physiology and pathology of research subjects. Metabolomics has become an important tool in distinguishing metabolic pathways changes and the diagnosis of human diseases, especially in the study of metabolic diseases.Objective:In this study, based on the platform of ultra performance liquid chromatography tandem quadrupole time-of-flight mass spectrometry (UPLC/Q-TOF-MS), a comprehensive metabolomics approach is applied to study the changes of metabolic profiling in serum between PCOS patients and controls and to select characteristic metabolites for PCOS diagnosis as well as to explore the pathogenesis and therapeutic target of PCOS. At the same time, the effects of IR on PCOS is explored by studying the metabolic changes between IR PCOS patients and non-IR PCOS subjects. Further, we study the pathogenesis of PCOS and IR development by doing in-depth analysis and research of these biomarkers.Methods:Based on the platform of UPLC/Q-TOF-MS and the method of pattern recognition, a comprehensive metabolomics approach has been applied to explore the changes of metabolic profiling between PCOS patients (n=20) and controls (n=15) as well as IR PCOS patients (n=11) and non-IR PCOS subjects (n=9) in serum. The raw data are pretreated and normalized by Marker View v1.2.1. SMCA-P13.0 is used for principal component analysis (PCA), partial least-squares discriminant analysis (PLS-DA), and orthogonal partial least-squares discriminant analysis (OPLS-DA). We further study the changes of metabolic profile data, and fabricate the pattern recognition models. Then the feature ions which can separate PCOS patients from healthy controls and separate IR PCOS subjects from non-IR PCOS subjects are selected based on the variable importance in projection (VIP) values of OPLS-DA model and p values of univariate statistic analysis. All the feature ions are identified and assigned to corresponding metabolites. Meanwhile, the ROC analysis is adopted to evaluate the clinical diagnostic performance of these selected different metabolites.Results:This study successfully fabricated the PLS-DA models of "non-IR PCOS patients-IR PCOS patients-controls" in ESI+and ESI-, and OPLS-DA models of differentiating PCOS patients from controls, as well as separating non-IR PCOS subjects from IR PCOS subjects. In total,36 metabolites were found significantly different between PCOS and controls, and 9 metabolites were discovered significantly different between IR and non-IR PCOS patients. Significant increase in the levels of saturated and unsaturated fatty acids (myristic acid, linoleic acid,9-/13-HODE, etc), fatty amides (palmitic amide, oleamide, palmitoleoyl ethanolamide), dehydroepiandrosterone sulfate (DHEAS), L-glutamic acid, azelaic acid, L-glyceric acid, pyroglutamic acid, and decrease in the levels of lysophosphatidylethanolamines, lysophosphatidylcholines, uridine and L-carnitine were found in PCOS patients compared with controls. In IR PCOS patients, linoleic acid, myristic acid, palmitoleic acid and vaccenic acid also increased significantly accompanied with a significant decrease in creatine, argininosuccinic acid, dodecanedioic acid, indoxyl sulfate and a increase in 5-methoxysalicylic acid compared with non-IR PCOS patients. With the ROC analysis, nine metabolites with good diagnostic value were found, mainly including LPC(18:2), LPC(20:2), ketoleucine, uridine, fructose 6-phosphate, phytosphingosine, etc.Conclusions:Metabonomics is a useful tool to study the differences of metabolites and changes in metabolic pathways of diseases. With metabolomics, we find there exists abnormalities of steroid hormone biosynthesis, amino acids and nucleosides metabolism, glutathione metabolism, lipids and carbohydrates metabolism in PCOS patients. The subgroup IR PCOS patients exhibit greater metabolic deviations than non-IR PCOS patients. These findings may be promising to yield a valuable insight into the pathogenesis and advance the diagnosis and prevention of PCOS.