How is it possible that one twin of a monozygotic twin pair develops cancer or depression and the other twin remains free of physical or mental illness?  The answer is simple: if identical twins have spent large part of their adult lives exposed to different environments they are not identical anymore. Their genome undergoes epigenetic (Greek epi=’above’ the genome) modifications that impact gene expression without actually changing the sequence of the DNA letters. ‘Epigenetics’ can be defined as the study of biochemical modifications of the DNA influencing gene expression without altering the structural base-pair sequence itself. The epigenome is not a stable entity but is designed to dynamically interact with the environment. Although during the life course changes in structural DNA sequences (mutations) occur rarely, epigenetic changes resulting in permanent alterations in gene expression, including silencing of genes, occur more frequently than we ever had imagined. Child development might be conceptualized as experiences becoming sculpted in the organism’s DNA through methylation, one of the major epigenetic mechanisms of change.
The application of epigenetics to the study of child development is a fascinating next step in unraveling the intricate interplay between rearing environment and the child’s genome shaping child development. New questions about intergenerational transmission of epigenetic changes, and reversibility of DNA methylation in children through psychosocial intervention  or pharmacological treatment emerge. From an epigenetic perspective divisions between genes, brain and behavior are artificial as the environment becomes embodied in the epigenome. It is the epigenomic modified DNA sequence that results in protein synthesis which in turn canalizes development. In fact, to a large extent nature is nurture. Methylation matters if one wants to understand how the early environment leaves its lasting imprint on the child.

Although the number of studies on methylation in humans have increased quite dramatically in recent years, little is known about its stability across time and similarity across tissues, or its validity as marker of human behavior. Some recent studies of our team are the following.

In a pilot study on correspondence of DNA methylation levels from the Illumina 450K array across batches, tissues and time we found better agreement between technical replicates versus repeated samplings, and better for buccal cells than for blood or blood spots. Gender, tissue and batch effects were identified. An epigenome-wide association study of maternal prenatal stress and neonatal methylation patterns in cord blood revealed some suggestive hits in the Generation R sample (N = 912) that did not replicate in the Avon Longitudinal Study of Parents and Children (ALSPAC, N = 828). Combining data from the two population-based samples in a meta-analysis, we found no large effects of prenatal maternal stress exposure on neonatal DNA methylation. Epigenetic results might be difficult to replicate because of various confounding influences that differ between studies.

In two other studies we examined the role of methylation in modulating the correlates of OXTR genotype. The OXTR rs53576 (G/A) variation was targeted to explain variation in children’s autistic symptoms and adult depressive disorder. In Generation R main and interactive effects of OXTR rs53576 genotype, prenatal maternal stress exposure, and OXTR methylation on child autistic symptoms were studied in N = 680 6-years old children whose cord blood was used for genotyping and assessment of methylation. Prenatal maternal stress exposure, but not OXTR rs53576 genotype or OXTR methylation, was related to child autistic symptoms. In addition, we found a significant OXTR rs53576 genotype by OXTR methylation interaction effect: OXTR methylation levels were positively associated with autistic symptoms for OXTR rs53576 GG children but not for A-allele carriers.

In a sample of 43 clinically depressed women recruited from a psychosomatic inpatient unit and 42 healthy female control subjects, DNA methylation profiles of the OXTR gene were assessed from leukocyte DNA. Depressed female patients had decreased OXTR exon 1 DNA methylation compared to non-depressed women. The association between depression and methylation level was moderated by OXTR rs53576 genotype. Exon 2 methylation was associated with OXTR rs53576 genotype but not with depression.

Our results highlight the promises of incorporating epi-allelic information in study designs, but they also show the slow progress and various pitfalls and problems facing researchers in this field.


Rijlaarsdam, Pappa, Walton et al (2016).Epigen. An epigenome-wide ass Meta-Analysis prenatal maternal stress in neonates
Rijlaarsdam, Van IJzendoorn, Verhulst et al (2016). Autism_Research prenatal stress OXTR methylation autism Proofs
Reiner et al (2015). Methylation of the Oxytocin receptor gene
VanIJzendoorn, Bakermans-Kranenburg, Ebstein 2011 Methylation Matters in Child Development, pdf file
VanIJzendoorn, Caspers, Bakermans-Kranenburg, Beach, Philibert 2010 Methylation Matters: Interaction Between Methylation Density and Serotonin Transporter Genotype, pdf file