Age and longitudinal NR3C1 methylation

Epigenomics of common diseases, such as depression, are prone to a multitude of influences, including age. A sensitive period for the impact of a certain environmental parameter, has been suggested in the literature. In particular in adolescence methylation differences can be expected by age, and, potentially, by perceived stress. Longitudinal research in adolescence, however, is scarce. NR3C1 is a gene often studied in multiple psychiatric phenotypes, we investigate how methylation in NR3C1 in a longitudinal adolescent sample can be affected by age and perceived parenting. From two most extreme perceived parenting groups identified by cluster analysis on a sample of 1,103 Belgian adolescents (MageW1=13.79(SD=0.94);51%Boys), W1: 100 adolescents were randomly selected: N=49 (supportive-guiding parenting), and N=51 (punishing-neglecting parenting). From these adolescents, 67 adolescents consented to a second saliva sample (W2), resulting in a total of 167 samples for bisulfite treatment and pyrosequencing analysis (PyromarkQ24). 8CpGs were genotyped in the 1F promotor of NR3C1. Samples were randomized prioritizing wave, parenting cluster, and gender. Analyses were performed in R using multivariate anova and linear regression, controlling for gender, plate, and, if not stratified by age, age. Following analyses per wave, a significant difference per parenting cluster was found for CpG1 in W2, showing hypomethylation for the bad-parenting cluster (F(1,48)=10.84, p=0.0019), which was not there at W1. Further exploration in W1 revealed a switch at the age of 13.23yrs: individuals in the bad-parenting cluster younger than this age show hypermethylation (6/8 CpGs: p<0.05, min-p=0.0051), individuals from the same cluster older than 13.23yrs, show hypomethylation over the 8 CpGs (8/8: p