Broad Epigenetic Signature of Maternal Care in the Brain of Adult Rats

Background Maternal care is associated with long-term effects on behavior and epigenetic programming of the NR3C1 (GLUCOCORTICOID RECEPTOR) gene in the hippocampus of both rats and humans. In the rat, these effects are reversed by cross-fostering, demonstrating that they are defined by epigenetic rather than genetic processes. However, epigenetic changes at a single gene promoter are unlikely to account for the range of outcomes and the persistent change in expression of hundreds of additional genes in adult rats in response to differences in maternal care. Methodology/Principal Findings We examine here using high-density oligonucleotide array the state of DNA methylation, histone acetylation and gene expression in a 7 million base pair region of chromosome 18 containing the NR3C1 gene in the hippocampus of adult rats. Natural variations in maternal care are associated with coordinate epigenetic changes spanning over a hundred kilobase pairs. The adult offspring of high compared to low maternal care mothers show epigenetic changes in promoters, exons, and gene ends associated with higher transcriptional activity across many genes within the locus examined. Other genes in this region remain unchanged, indicating a clustered yet specific and patterned response. Interestingly, the chromosomal region containing the protocadherin-α, -β, and -γ (Pcdh) gene families implicated in synaptogenesis show the highest differential response to maternal care. Conclusions/Significance The results suggest for the first time that the epigenetic response to maternal care is coordinated in clusters across broad genomic areas. The data indicate that the epigenetic response to maternal care involves not only single candidate gene promoters but includes transcriptional and intragenic sequences, as well as those residing distantly from transcription start sites. These epigenetic and transcriptional profiles constitute the first tiling microarray data set exploring the relationship between epigenetic modifications and RNA expression in both protein coding and non-coding regions across a chromosomal locus in the mammalian brain.

were observed for exactly the same number periods, there was some variation across days. The data were analyzed as the percentage of observations in which animals engaged in the target behavior. The following behaviors were scored: mother off pups, mother licking/grooming any pup, or mother nursing pups in an arched-back posture, a "blanket" posture in which the mother lies over the pups, or a passive posture in which the mother is lying on her back or side while the pups nurse. A detailed description of these behaviors is provided elsewhere [2].
Note that behavioral categories are not mutually exclusive. For, example, licking/grooming most often occurred while the mother was nursing the pups. The frequency of maternal licking/grooming and arched-back nursing across a large number of mothers is normally and not bimodally distributed [3]. Hence, the high and low LG mothers represent two ends of a continuum, rather than two distinct populations. To define these populations for the current study we observed the maternal behavior in a cohort of 32 mothers and devised the group mean and standard deviation for each behavior over the first 7 days of life. High LG mothers were defined as females whose frequency scores for both licking/grooming and arched-back nursing were >1 SD above the mean. Low LG mothers were defined as females whose frequency scores for both licking/grooming and arched-back nursing were >1 SD below the mean. Previous reports [4,5] of licking/grooming behavior suggest that the frequency of licking/grooming and arched-back nursing are highly correlated (r > +0.90).
Microarray analysis. Differential methylation, acetylation and expression between maternal care groups was determined in two stages to ensure both statistical significance and biological relevance. In the first stage, linear models implemented in the 'limma' package [6] of Bioconductor [7] were used to compute a modified t statistic at the individual probe level. An individual probe was called differentially methylated/acetylated/expressed if the significance of its t statistic was at most 0.05 (uncorrected for multiple testing) and the associated difference of means between the groups was at least 0.5. Given that the DNA samples were sonicated into 200-700bp fragments prior to hybridization, we assumed that probes within 500bp should have approximately similar probe scores. Therefore, in the second stage, we computed differential statistics for 1000bp intervals from the differential statistics of the probes that they contained. The intervals tiled the entire 7Mb region under investigation at 500bp spacing. Differential significance of these intervals was determined using the Wilcoxon rank-sum test comparing t statistics of the probes within the interval against those of all the probes on the microarray. Significance levels were then adjusted to obtain false discovery rates. An interval was called differentially methylated/acetylated/expressed if it satisfied each of the following: 1. its false discovery rate was at most 0.2, and 2. the 1000 bp interval contained at least one probe called differentially methylated/acetylated/expressed.
The first requirement ensured that several probes in the interval have similar group differences, and the second requirement ensured that the difference was not simply weakly distributed across the entire interval and consequently difficult to validate. For methylation and acetylation, intervals satisfying these tests were called an RDme or RDac (differentially methylated/acetylated region, respectively). Consecutive RDme/RDac were combined into a single RDme/RDac.   For data shown in Figure 3c, modified t-statistics (described above) are used for each probe rather than for estimated levels in order to illustrate the distribution of DNA methylation, H3K9 acetylation and transcriptional differences across the gene.
Figures 4a, and 5a-b were created by viewing UCSC browser tracks containing average microarray probe intensity differences between LG offspring groups.