Final Report Summary - E-DOHAD (Environmentally-induced Developmental Origins of Health and Disease)
The application of DOHaD concept to short half-lived endocrine disruptors is well supported by toxicological studies. Most human studies on the topic assessed exposures from spot biospecimens, inducing large exposure misclassification. In contrast, human studies reported effects of early-life exposure to atmospheric pollutants, an issue relatively less tackled in toxicology. Generally, human observations and toxicological experiments can in principle be complementary, but are most often conducted totally independently.
E-DOHaD project aimed at characterizing the impact of environmental exposures during intra-uterine life on childhood health. Our focus is on two families of pollutants with highly prevalent and controllable exposure in humans: atmospheric pollutants and specific high-volume non-persistent chemicals with endocrine-disrupting properties.
Regarding early-life exposure to non-persistent chemicals of the phenols and phthalates families, we have documented detectable levels in urine of French pregnant women. In children from Eden mother-child cohort, respiratory health was assessed around 5 years of age from validated questionnaires and lung function (spirometry) examinations. Among the compounds considered, ethyl-paraben tended to be assessed with a decreased lung function in boys. Parabens are used in antimicrobial agents in cosmetic products and food, but no study had considered their possible effects on respiratory health in humans so far, to our knowledge.
We identified possible increases in troubles of male behavior at 3-5 years of age in relation to maternal triclosan levels during pregnancy. Triclosan is a still widely-used (and authorized in the EU) organochlorine pesticide found in soaps or tooth paste, known to disrupt the thyroid axis, which controls brain development. We had previously reported a possible effect of triclosan on head circumference at birth in boys. There was no evidence of an effect of these compounds on the cognitive function (intelligence) of boys.
We have documented that the levels of these compounds in the body of the mother have very strong variations during pregnancy, even within the course of the day, as well as between the days of the week and across several weeks of the pregnancy. We have shown that this variability strongly impacts the validity of studies aiming at characterizing the health effects of non-persistent chemicals. For the compounds with the strongest temporal variations, such as bisphenol A, the amplitude of the bias is expected to be around 80% and to correspond to attenuation, that is, a strong under-estimation of any real effect of the exposure if its assessment is based on a spot urine sample, as in most previous studies. As a possible cure, we have developed a within-subject biospecimens pooling approach, consisting in collecting several biological samples per subject during a toxicologically-relevant exposure window and pooling them before assaying the chemical compound(s) of interest. We have shown the validity of this approach both theoretically (through a simulation study) and empirically, and have implemented it at a large scale in the context of SEPAGES couple-child cohort (Grenoble area, France), in which 471 women collected over 40 urine samples each during pregnancy.
In addition, in this cohort, prenatal growth (through ultrasound measurements), postnatal growth, neurodevelopment (through validated questionnaires and clinical tests by trained neuropsychologists) and respiratory health in mother and children have been assessed. The respiratory health assessment of the child has been done through a particularly innovative approach allowing objective measurements of lung function at much earlier ages than is done in most previous cohorts, in which lung function assessment is usually done at age 5-6 years or later. In SEPAGES cohort, lung function was assessed with a clinical device used in the child’s natural sleep allowing quantification of the lung clearance index at 6 weeks of age, as well as through oscillometry at 3 years of age.
When it comes to atmospheric pollutants, maternal benzene personal exposure during pregnancy tended to be associated with decreases in offspring birth weight and head circumference in the newborns of SEPAGES cohort. Nitrogen dioxide, a marker of traffic-related air pollution, also tended to be associated with decreased head circumference. In an animal (rabbit) experiment focused on the impact of exposure to diesel engine exhaust during pregnancy, diesel engine exhaust altered placental function, with a decreased placental efficiency and placental blood flow. In addition, in the second generation (i.e. the offspring of the animals exposed during intra-uterine life) metabolic function was altered, with increased cholesterol levels and triglyceride levels. The rabbit model had been chosen because of the similarity of its placental function compared to other animal models such as rodents.
In conclusion, from a methodological point of view, E-DOHaD project allowed to develop and validate approaches to assess with much more accuracy than previously possible the potential health effects of early-life exposure to time-varying exposures present at low levels in the general population, which is a critical step to better understand the effect of environmental factors on health and ultimately improve the health of populations. E-DOHaD also documented effects of chemical factors at levels encountered in the general population of the EU on children health. The project also illustrated the feasibility of relevance of having toxicological and epidemiological studies being conducted in parallel on similar factors. E-DOHaD project allowed to set up a unique cohort of a new type, with very intense and accurate assessment of environmental exposures and health, which constitutes a great resource to document the effect of early life exposure to non-persistent exposures, and of the Exposome in general, on children health.