Serum Metabonomic Study Of Patients With Endometriosis

OBJECTIVE: Ultra-high liquid chromatography-mass spectrometry was used to screen serological differential metabolites that can identify EMT and non-EMT.Find potential biomarkers for the diagnosis of endometriosis and explain the metabolic pathways involved in differential metabolites from a metabolomic perspective.METHODS:Ultra-high performance liquid chromatography-mass spectrometry(UHPLC-HRMS)was used to analysis the serum of 22 EMT patients and 22 non-EMT patients by non-targeted metabolomics.The obtained raw data is subjected to peak matching,filtering and normalization.The processed data was analyzed by Metabo Analyst 3.0 software for supervised pattern recognition Principal component analysis(PCA)and unsupervised pattern recognition partial least-squares-discriminant analysis(PLS-DA),based on the variable weight value(VIP)obtained by PLS-DA.small molecular differential metabolites that may be associated with endometriosis were screened by criteria of VIP > 1.8 and P < 0.05.According to the mass-to-charge ratio and abundance ratio of the differential metabolites,the molecular formulas matched with them are obtained,and the structure of the differential metabolites is determined from the metabolite database.These differential metabolites are then analyzed by KEGG clustering and directed to differential metabolic pathways.The pathological mechanism of endometriosis was elucidated from the perspective of metabolomics.The Receiver operating characteristic(ROC)curve was plotted and the area under the curve(AUC)was calculated to determine the combination of differential metabolites with the best specificity and sensitivity.RESULTS: According to single factor analysis and multivariate statistical analysis,the metabolomics characteristics of serum in EMT group were significantly different from those in healthy womens.Eight different metabolites were selected,including Methionine sulfoxide,Sphingosine-1-phosphate,lysophosphatidylcholine 18:0(LPC18:0),LPC 18:0P and LPC 16:0e levels were significantly incresed in EMT patients.In contrast,L-Methionine,Kynurenic acid and Aspartyl-Threonine levels are reduced in EMT.KEGG cluster analysis of these eight differential metabolites revealed that cysteine and methionine metabolism,glycerophospholipid metabolism,and sphingolipid metabolic pathway were different in the EMT group and the healthy control group.After binary logistic regression of differential metabolites,a predictive model containing Kynurenic acid and Sphingosine-1-phosphate was constructed.The ROC curve analysis confirmed the diagnostic value of these two metabolites for EMT.The accuracy of the predicted EMT was 99.8%,the sensitivity was 95.5%,and the specificity was 100%.Conclusion: There is a significant difference in the metabolomics characteristics of serum between EMT group and non-EMT group.It has been found that Kynurenic acid and sphingosine-1-phosphate have good predictive power in the diagnosis of endometriosis and provided a Serum biomarkers with high sensitivity and high specificity to non-invasive diagnosis of endometriosis.The mechanism of the development of endometriosis is revealed from the level of serum metabolomics.