Amniotic-fluid metabolomics identifies phospholipid remodeling as a metabolic signature of intrauterine exposure in pregnancies with polycystic ovary syndrome

Abstract

Background Polycystic ovary syndrome is associated with metabolic and hormonal disturbances during pregnancy, but whether these alterations are reflected in the fetal intrauterine exposure environment remains incompletely understood. This study aimed to identify polycystic ovary syndrome-related intrauterine metabolic signatures using late-gestation amniotic fluid. Methods Untargeted metabolomic profiling was performed on late-gestation amniotic fluid samples from women with polycystic ovary syndrome and controls. Differential metabolite analysis, pathway-level analysis, and multilevel sensitivity analyses were conducted to identify robust metabolic alterations associated with maternal polycystic ovary syndrome. Exploratory placental transcriptomic analysis and targeted RT-qPCR assessment in an independent sample set were further used to examine tissue-level molecular changes related to the lipidomic findings. Results Amniotic fluid from pregnancies with polycystic ovary syndrome showed a distinct metabolic profile dominated by lipid remodeling. Differential metabolites were mainly enriched in membrane phospholipids, sphingolipid-related metabolites, polyunsaturated fatty acid-related pathways, and selected steroid hormone-related metabolites. Phospholipid remodeling was characterized by decreased phosphatidylcholine species, increased phosphatidylethanolamine species, a lower phosphatidylcholine/phosphatidylethanolamine ratio, and redistribution of arachidonic acid-containing phospholipids. Alpha-linolenic acid metabolism provided an additional polyunsaturated fatty acid-related signal. Sphingolipid enrichment suggested that lipid alterations extended from membrane structural remodeling to lipid-mediated signaling. Selected steroid hormone-related metabolites were also elevated, consistent with an altered hormonal milieu in pregnancies with polycystic ovary syndrome. These signatures remained largely stable across sensitivity analyses, as did the multivariable-adjusted inverse association between PE(16:0/20:4) and birth weight. Exploratory placental transcriptomic analysis and targeted RT-qPCR assessment in a small independent cohort provided preliminary tissue-level support for phospholipid-related molecular alterations, with the clearest changes involving PLA2-family genes and PCYT1A, suggesting cross-cohort convergence at the pathway level. Conclusions These findings identify phospholipid remodeling as a major amniotic-fluid metabolic signature of polycystic ovary syndrome-related intrauterine exposure. PUFA-related metabolism, sphingolipid enrichment, and steroid hormone-related alterations provide additional metabolic context. Complementary placental molecular findings and the inverse association between PE(16:0/20:4) and birth weight further suggest potential links with the maternal–fetal interface and fetal growth. Further studies are needed to clarify the biological relevance of these changes and their potential role in offspring metabolic susceptibility.