Periconception air pollution, metabolomic biomarkers, and fertility among women undergoing assisted reproduction

Abstract

BackgroundAir pollution exposure has been linked with diminished fertility. Identifying the metabolic changes induced by periconception air pollution exposure among women could enhance our understanding of the potential biological pathways underlying air pollution’s reproductive toxicity. ObjectiveTo identify serum metabolites associated with periconception air pollution exposure and evaluate the extent to which these metabolites mediate the association between air pollution and live birth. MethodsWe included 200 women undergoing a fresh assisted reproductive technology (ART) cycle at Massachusetts General Hospital Fertility Center (2005–2015). A serum sample was collected during stimulation, and untargeted metabolic profiling was conducted using liquid chromatography with ultra-high-resolution mass spectrometry. Exposure to nitrogen dioxide (NO2), ozone (O3), fine particulate matter <2.5 µm (PM2.5), and black carbon (BC) was estimated using validated spatiotemporal models. Multivariable linear regression models were used to evaluate the associations between the air pollutants, live birth, and metabolic feature intensities. A meet in the middle approach was used to identify overlapping features and metabolic pathways. ResultsFrom the C18 and HILIC chromatography columns, 10,803 and 12,968 metabolic features were extracted. There were 190 metabolic features and 18 pathways that were significantly associated with both air pollution and live birth (P < 0.05) across chromatography columns. Eight features were confirmed metabolites implicated in amino acid and nutrient metabolism with downstream effects on oxidative stress and inflammation. Six confirmed metabolites fell into two intuitive clusters – “antioxidants” and “oxidants”- which could potentially mediate some of the association between air pollution and lower odds of live birth. Tryptophan and vitamin B3 metabolism were common pathways linking air pollution exposure to decreased probability of live birth. ConclusionHigher periconception air pollution exposure was associated with metabolites and biologic pathways involved in inflammation and oxidative stress that may mediate the observed associations with lower probability of live birth following ART.