The effect of maternal prenatal emotional wellbeing and maternal cortisol on fetal and child devlopment - an epigenetic study

The prenatal period, and therefore the uterine environment, is crucial in the development of the fetal organs and organ systems and therefore in child development. Fetal programming describes the process through which the development of a fetus is altered due to changes in its immediate environment. The effects of this environment can vary at different sensitive periods and can shape the structure and function of the brain and peripheral organ systems, with long-term or even permanent effects on subsequent child and adult physiology, behavior and health (Glover et al.2010;Gluckman and Hanson2004;Gluckman et al.2008;Van den Bergh2011;Reynolds2013b). This research paradigm is commonly referred to as Developmental Origins of Health and Disease (DOHaD) - paradigm". We formulated 4 research aims. First of all we investigated the link between maternal emotional state (more specifically depression, anxiety, pregnancy related anxiety as well as attachment) and maternal cortisol secretion on the one hand, and child development [(fetal and child growth, infant HPA axis function and infant neurodevelopment) on the other hand. Furthermore we investigated whether emotional state and/or maternal cortisol during pregnancy has an association with the methylation pattern of the NR3C1 1B, 1D, and 1F promoter regions in infant DNA of cord blood mononuclear cells at birth. Additionally the relation between the methylation state of the NR3C1 promoter regions in infant DNA at birth and child development was investigated. And finally the relation between infant HPA axis and infant neurodevelopment was explored.We conducted 2 prospective longitudinal clinical studies: a pilot project (chapter 5) and a large comprehensive study, the Prenatal Early life Stress (PELS) study (chapter 4 and chapter 6-9). In the introduction part of this thesis we first give an overview of the literature in the first chapter, this to provide the reader with a broad background. In the second chapter we formulated 4 research aims as previously mentioned. In chapter 3 the epidemiology of both studies is highlighted.In part 1 we explored whether emotional state and/or maternal cortisol during pregnancy has a relation with the methylation pattern of the NR3C1 1B, 1D, and 1F promoter regions in infant DNA of cord blood mononuclear cells at birth. These data indicate that prenatal maternal emotional state, particularly pregnancy related anxiety, are associated with the methylation state of the NR3C1 gene in the child. The results found in the PELS study are relevant since differences in methylation of CpGs in the promoter region can alter gene transcription by affecting transcription factor binding (chapter 2). In part 2, first of all we investigated the association between maternal emotional state (more specifically depression, anxiety, pregnancy related anxiety as well as attachment) and maternal cortisol secretion on the one hand, and child development [fetal and child growth, infant HPA axis function and infant neurodevelopment) on the other hand. In chapter 5 we present results of the pilot study focusing on growth during the prenatal period. This study focused on the prenatal period. Some distinct patterns were observed. First of all cortisol exerted an influence mainly in mid pregnancy and on the growth trajectory between mid and late pregnancy. Secondly in late pregnancy it was observed how emotional state, particularly depressive symptoms and attachment had an influence on growth and analysis of growth between mid and late pregnancy showed that attachment and cortisol in late pregnancy had an influence. In contrast to our pilot study the focus of the PELS study, presented in chapter 5, was not only on the prenatal IUG but also on the longer lasting effects of emotional wellbeing and maternal cortisol during the first year after birth. Growth was approached in a longitudinal way across birth. Only between general anxiety and weight a weak relation has been found. Both studies resemble the normal population, but in the pilot study, depressive symptoms were more pronounced. This might be a possible explanation for these differences in results (chapter 6). In chapter 7 it was concluded that early infant HPA axis function is influenced by both maternal cortisol secretion and depressive symptoms in late pregnancy, though very limited. In chapter 8 it was observed how depressive symptoms during early pregnancy as well as pregnancy related anxiety during pregnancy predicted mental development at 12 months, though again limited. Additionally in part 2, the association between the methylation state of the NR3C1 promoter regions in infant DNA at birth and child development was studied. As concluded in chapter 6 there was only limited evidence for a link between the methylation state of NR3C1 promoter region and growth in a population comparable to general pregnant population (Chapter 6). Early infant HPA axis function at 2 months was limited related to the degree of methylation at CpG sites: D22.24, D25.26, F19 and F20.21 in the NR3C1 promoter region (Chapter 7). And finally in chapter 8 an association between mental development and the methylation state of CpG unit F35 of the NR3C1 promoter region was found as described in chapter 8. Furthermore some borderline significant relations were found between mental neurodevelopment and CpG unit F1.5 and F10.11 as well as between psychomotor development and CpG unit D12.13 and F12.13. Again several of these sites are located in transcription factor binding sites.In the third part of this thesis we investigated the relation between infant HPA axis function and neurodevelopment. It was concluded that early postnatal HPA reactivity during the first year predicts neurodevelopment at 12 months.We must underline that the relations observed were rather weak. The results on growth were inconclusive, though the results on infant HPA axis and neurodevelopment suggest that both are at least partly caused by processes occurring in fetal life.We must underline that our sample resembles the general population. It is reassuring that in a group of mild to moderately emotionally stressed mothers the effects of maternal cortisol and maternal emotional state on infant development seem to be limited. This points out that humans have solid mechanisms to protect their unborn child when exposed to mild to moderate emotionally stressful circumstances. As a consequence we expect the weak effects seen here to enlarge in a more stressed population. We believe that our data further support the fetal programming hypothesis. Of course replication of these data is needed and cautious interpretation is important.status: publishe