Genetic analysis of DNA methylation and gene expression levels in whole blood of healthy human subjects

BMC Genomics Genetic analysis of DNA methylation and gene expression levels in whole blood of healthy human subjects Kristel R van Eijk 1 Simone de Jong 2 Marco PM Boks 1 Terry Langeveld Fabrice Colas Jan H Veldink Carolien GF de Kovel Esther Janson Eric Strengman 2 Peter Langfelder Ren S Kahn 1 Leonard H van den Berg Steve Horvath Roel A Ophoff 1 2 0 4357C , 695 Charles E. Young Drive, South Los Angeles, CA 90095-1761 , USA 1 Department of Psychiatry, Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht , Utrecht 3508, GA , The Netherlands 2 Center for Neurobehavioral Genetics, University of California Los Angeles , Box 951761 Gonda Background: The predominant model for regulation of gene expression through DNA methylation is an inverse association in which increased methylation results in decreased gene expression levels. However, recent studies suggest that the relationship between genetic variation, DNA methylation and expression is more complex. Results: Systems genetic approaches for examining relationships between gene expression and methylation array data were used to find both negative and positive associations between these levels. A weighted correlation network analysis revealed that i) both transcriptome and methylome are organized in modules, ii) co-expression modules are generally not preserved in the methylation data and vice-versa, and iii) highly significant correlations exist between co-expression and co-methylation modules, suggesting the existence of factors that affect expression and methylation of different modules (i.e., trans effects at the level of modules). We observed that methylation probes associated with expression in cis were more likely to be located outside CpG islands, whereas specificity for CpG island shores was present when methylation, associated with expression, was under local genetic control. A structural equation model based analysis found strong support in particular for a traditional causal model in which gene expression is regulated by genetic variation via DNA methylation instead of gene expression affecting DNA methylation levels. Conclusions: Our results provide new insights into the complex mechanisms between genetic markers, epigenetic mechanisms and gene expression. We find strong support for the classical model of genetic variants regulating methylation, which in turn regulates gene expression. Moreover we show that, although the methylation and expression modules differ, they are highly correlated. DNA methylation; Gene expression; Association; Epigenetics; WGCNA - Background Epigenetics has been described as the structural adaptation of chromosomal regions so as to register, signal or perpetuate altered activity states [1]. DNA methylation is one of several forms of epigenetic modifications and involves the covalent binding of a methyl group to a Cytosine-5 at a C-phosphate-G (CpG) site. These sites are relatively rare in the genome but more common at promoter regions of genes, also called CpG islands (CGIs). CpGs in these islands are less likely to be methylated than CpGs outside these islands. Recent studies have shown that specifically the CpGs in the shore of CGIs are most frequently involved in differential methylation between tissues or experimental groups [2,3]. Increased methylation of CpG islands at 5 end of a gene is associated with gene repression. Possible mechanisms for repression include interference with transcription factor binding or through the recruitment of repressors such as histone deacetylases [4]. Although one would expect DNA methylation at CGIs and expression of the nearby gene to be inversely correlated, this is not necessarily the case. Recent reports also identified positive associations between expression and methylation levels [5-7]. However, negative associations between methylation and expression were found to be enriched particularly in CGIs [6] and promoter regions [5]. Around 30% of gene expression levels in cell lines [8] and 23% of DNA methylation levels in blood are heritable [9] and genetic variation associated with expression and methylation levels has been identified in several organisms [6,10-12], tissues [13] and populations [14]. Local (cis) and distal (trans) associations of genetic variation with gene expression levels have been observed. With the arrival of high-throughput DNA methylation assays, methylation quantitative trait loci (mQTLs) can now be studied genome-wide in any tissue or cell type of interest. Similar to expression (eQTLs), more cis than trans regulation has been identified [5-7] but peak enrichment for mQTLs is located in much closer proximity to transcription start sites than that of eQTLs [6]. Attempts to identify three-way associations between genetic variants, expression and methylation on a genome-wide scale in four different brain regions did not identify co-regulation of methylation and expression by the same genetic variants [6], while a study of cerebellar samples did identify three-way associations for a number of genes [7]. In lymphoblastoid cell lines of 77 individuals of the Yoruba Hapmap population, coregulation of expression and methylation levels by the same genetic variants was also found, suggesting a shared mechanism, whereby a genetic variant influences methylation, which in turn influences expression levels [5]. Strong evidence exists that both patterns of CpG methylation [15,16] and gene expression [13,17,18] differ between tissues. The aims of the current study are i) to relate expression levels to methylation levels, ii) to relate coexpression modules (clusters of expression probes) to co-methylation modules, iii) and to study the relationship between genetic markers, methylation and expression in whole blood of a relatively large (n=148) set of healthy human subjects. For the genetic analysis, we examined the associations of methylation and expression levels and identified genetic markers associated with these levels. To infer directionality in the relationships between genetic variants, methylation and expression, we calculated local edge orienting (LEO) scores based on structural equation models [19]. This method has been applied successfully before and will aid in elucidating the nature of relationship between genetic variation, methylation and expression [20-23]. Results Associations between methylation and expression levels A multivariate linear model analysis for regressing a gene expression level on a methylation level and age and gender resulted in the identification of 522 negative and 276 positive cis associations between methylation and expression levels (False Discovery Rate (FDR) 5% corrected). A negative association between methylation and transcript level means that increased methylation levels correlate with decreased expression levels, whereas a positive correlation includes levels that both increase or decrease. These associations involved 517 different cis-acting CpG loci (...truncated)