Identification of Serum, Placental and Fetal Serum Metabolite Biomarkers for Preeclampsia Based on LC-MS

Background: Preeclampsia is a common complication of pregnancy seriously affecting the health of mothers and babies. There is no independent methods for diagnosing and monitoring preeclampsia with sufficient accuracy, specificity and sensitivity. Changes in body metabolism usually occur before the appearance of clinical symptoms of the disease. Therefore, predicting the occurrence and development of preeclampsia by detecting changes in metabolites has become a research trend in this field. Methods In this study, a nontargeted metabolomics approach based on liquid chromatography (LC)-mass spectrometry (MS) was used to analyze the metabolites in the serum, placenta and fetal serum samples of 6 cases of late preeclampsia patients and 6 cases of normal pregnant women. Combine multivariate and univariate statistical methods to screen the endogenous differential metabolites in each sample, and use MetaboAnalyst software to perform enrichment analysis on the metabolic pathways involved in these differential metabolites. In addition, based on a binary logistic regression model to screen biomarkers that may be used to predict preeclampsia. Results Compared with normal pregnant women, there were significant differences in serum, placental and fetal serum metabolic profiles of late preeclampsia patients. We found that the late preeclampsia patients mainly had metabolic abnormalities in the metabolic pathways of ALA, taurine, glycine, serine and threonine; in addition, placenta samples showed abnormalities in the metabolic pathways of linolenic acid, taurine, α-linoleic acid and other unsaturated fatty acids, which in turn caused abnormalities in the metabolic pathways of glutamine and glutamate, ALA, alanine, aspartic acid and glutamate, and other polyunsaturated fatty acids in the fetuses of these patients. Moreover, we generated the ROC curves of differential metabolites using a binary logistic regression model and selected four serum metabolites with an AUC value greater than 0.9, most of which were polyunsaturated fatty acids. Conclusions There are significant differences in serum, placental, and fetal serum metabolic profiles between late preeclampsia patients and normal third trimester pregnant women in this study and influence the metabolic phenotypes of offspring. This result not only provides a meaningful biomarker for the diagnosis of preeclampsia, but also provides a basis for understanding the pathogenesis of preeclampsia.