Autism spectrum disorder (ASD) imposes large lifetime social and economic costs on families and communities. Causes, differentially affecting boys, likely are multifactorial. In search of modifiable risk factors, several studies have found associations of ASD risk with prenatal ambient air pollution exposure; evidence from human epidemiological and animal studies is converging on the neurotoxic effects of fine particulate matter less than 2.5 m in diameter (PM2.5). Recently, early life exposure to currently unregulated ultrafine PM0.1 was shown to cause autism-like behavioral traits specific to males, but methods have not been available to examine effects of PM0.1 or of components of the complex PM mixture likely to be causal. We hypothesize that ASD will be associated with novel PM2.5 and PM0.1 exposure estimates with high temporal and spatial resolution at maternal residences and workplaces during pregnancy (Aim 1a), and that associations will be driven by specific PM components (Aim 1b). We will assess this hypothesis in a pregnancy-birth cohort of 400,000 mother-offspring pairs followed through Kaiser Permanente Southern California (KPSC), a population resource with standardized algorithms for systematic screening and diagnosis of ASD and gestational risk factors, available through an exceptionally high quality electronic medical record. Maternal immune activation (MIA) has been proposed as a common mechanism linking prenatal risk factors, including diverse viral and bacterial infections, asthma, pre-pregnancy obesity, diabetes, to subsequent ASD risk. Because prenatal PM exposure, and ASD phenotype, have in common pro-inflammatory and oxidative stress pathways, we hypothesize that MIA-related maternal comorbidities will increase fetal susceptibility to neurodevelopmental effects of PM exposure leading to ASD (Aim 2). This novel hypothesis can only be studied in large population studies such as the KPSC cohort with sufficient power for assessing interactions that together could explain a much larger proportion of ASD than single risk factor epidemiology. Finally, ASD epidemiology has been limited by the lack of biological markers both of environmental exposures and of pathways of effects. Using a novel assay for electrophilic adducts to human serum albumin in neonatal archived dried blood spots from 420 children selected from the cohort based on exposure, we will examine biological markers for PM exposure and oxidative stress. We hypothesize that PM2.5 will be associated with a targeted panel of adducts reflecting exposure and with adducts reflecting oxidative stress (Aim 3a). We hypothesize that these targeted biological markers of exposures and pathways, and additional ones identified in an untargeted HSA ?adductome?, will be associated with an abnormal Checklist for Autism in Toddlers (CHAT)1 administered at ages 18 and 24 months to all participants in the KPSC cohort (Aim 3b). The study will address critical gaps in knowledge of effects of PM and PM composition, and ASD susceptibility, and will provide clues to biological pathways underlying PM effects on ASD.
The aims of this study are to assess effects of prenatal and early life exposure to ambient particulate matter mass and composition on autism spectrum disorder, and susceptibility to exposure in a large pregnancy cohort. In an informatively selected sample, PM effects on novel early life biological markers of exposure and oxidative stress and their association with early life neurodevelopmental phenotype will also be investigated.
