Placental DNA methylation signatures of prenatal air pollution exposure and potential effects on birth outcomes: an analysis of three prospective cohorts.

BACKGROUND: Pregnancy air pollution exposure (PAPE) has been linked to a wide range of adverse birth and childhood outcomes, but there is a paucity of data on its influence on the placental epigenome, which can regulate the programming of physiological functions and affect child development. This study aimed to investigate the association between prenatal air pollutant exposure concentrations and changes in placental DNA methylation patterns, and to explore the potential windows of susceptibility and sex-specific alterations. METHODS: This multi-site study used three prospective population-based mother-child cohorts: EDEN, PELAGIE, and SEPAGES, originating from four French geographical regions (Nancy, Poitiers, Brittany, and Grenoble). Pregnant women were included between 2003 and 2006 for EDEN and PELAGIE, and between 2014 and 2017 for SEPAGES. The main eligibility criteria were: being older than 18 years, having a singleton pregnancy, and living and planning to deliver in one of the maternity clinics in one of the study areas. A total of 1539 mother-child pairs were analysed, measuring placental DNA methylation using Illumina BeadChips. We used validated spatiotemporally resolved models to estimate PM2·5, PM10, and NO2 exposure over each trimester of pregnancy at the maternal residential address. We conducted a pooled adjusted epigenome-wide association study to identify differentially methylated 5'-C-phosphate-G-3' (CpG) sites and regions (assessed using the Infinium HumanMethylationEPIC BeadChip array, n=871), including sex-specific and sex-linked alterations, and independently validated our results (assessed using the Infinium HumanMethylation450 BeadChip array, n=668). FINDINGS: We identified four CpGs and 28 regions associated with PAPE in the total population, 469 CpGs and 87 regions in male infants, and 150 CpGs and 66 regions in female infants. We validated 35% of the CpGs available. More than 30% of the identified CpGs were related to one (or more) birth outcome and most significant alterations were enriched for neural development, immunity, and metabolism related genes. The 28 regions identified for both sexes overlapped with imprinted genes (four genes), and were associated with neurodevelopment (nine genes), immune system (seven genes), and metabolism (five genes). Most associations were observed for the third trimester for female infants (134 of 150 CpGs), and throughout pregnancy (281 of 469 CpGs) and the first trimester (237 of 469 CpGs) for male infants. INTERPRETATION: These findings highlight the molecular pathways through which PAPE might affect child health in a widespread and sex-specific manner, identifying the genes involved in the major physiological functions of a developing child. Further studies are needed to elucidate whether these epigenetic changes persist and affect health later in life. FUNDING: French Agency for National Research, Fondation pour la Recherche Médicale, Fondation de France, and the Plan Cancer.

An integrated multi-tissue approach for endometriosis candidate biomarkers: a systematic review.

Biomarker identification could help in deciphering endometriosis pathophysiology in addition to their use in the development of non invasive diagnostic and prognostic approaches, that are essential to greatly improve patient care. Despite extensive efforts, no single potential biomarker or combination has been clinically validated for endometriosis.Many studies have investigated endometriosis-associated biological markers in specific tissues, but an integrative approach across tissues is lacking. The aim of this review is to propose a comprehensive overview of identified biomarkers based on tissue or biological compartment, while taking into account endometriosis phenotypes (superficial, ovarian or deep, or rASRM stages), menstrual cycle phases, treatments and symptoms.We searched PubMed and Embase databases for articles matching the following criteria: 'endometriosis' present in the title and the associated term 'biomarkers' found as Medical Subject Headings (MeSH) terms or in all fields. We restricted to publications in English and on human populations. Relevant articles published between 01 January 2005 (when endometriosis phenotypes start to be described in papers) and 01 September 2022 were critically analysed and discussed.Four hundred forty seven articles on endometriosis biomarkers that included a control group without endometriosis and provided specific information on endometriosis phenotypes are included in this review. Presence of information or adjustment controlling for menstrual cycle phase, symptoms and treatments is highlighted, and the results are further summarized by biological compartment. The 9 biological compartments studied for endometriosis biomarker research are in order of frequency: peripheral blood, eutopic endometrium, peritoneal fluid, ovaries, urine, menstrual blood, saliva, feces and cervical mucus. Adjustments of results on disease phenotypes, cycle phases, treatments and symptoms are present in 70%, 29%, 3% and 6% of selected articles, respectively. A total of 1107 biomarkers were identified in these biological compartments. Of these, 74 were found in several biological compartments by at least two independent research teams and only 4 (TNF-a, MMP-9, TIMP-1 and miR-451) are detected in at least 3 tissues with cohorts of 30 women or more.Integrative analysis is a crucial step to highlight potential pitfalls behind the lack of success in the search for clinically relevant endometriosis biomarkers, and to illuminate the physiopathology of this disease.