Solid epidemiologic data support that an unfavorable intrauterine environment is associated with adult-onset disorders such as hypertension, coronary heart disease, type 2 diabetes and neuropsychiatric disease. Prenatal nutrition, as assessed by infant size and weight at birth, is the most widely examined variable. Animal studies suggest that poor nutrition might be a marker for other adverse environmental events, such as prenatal stress, that result in permanent changes in metabolism. The molecular mechanism that confers a permanent memory on cells of (adverse) prenatal environmental influences is still unknown. We hypothesize that DNA methylation at CpG dinucleotides fits all requirements to be an important contributor: it can directly affect the expression profile of genes, it is mitotically inheritable and it can be influenced by the environment. As our initial approach to address this hypothesis we propose to optimize animal models to evaluate how supplementation with various methyl donors affects DNA methylation in development and disease in later life. We will initially use existing technologies to screen for genes with CpG methylation changes, but concurrently will develop a """"""""methylation microarray"""""""" for a more sensitive and faster evaluation. We will also measure gene expression using existing cDNA microarrays and establish analysis techniques to correlate CpG methylation profiles with gene expression profiles. Of the candidate genes thus discovered, we will prioritize for further detailed investigation those with a role in neurological function. Finally, to prove the principle that DNA methylation can interact with noxious environmental factors, we will expose mice treated with the methyl donors to low doses of valproic acid in development and study the long-term effects. Valproic acid was chosen because it has been demonstrated that its teratogenic effect on neurons, resulting in axon branching defects, is by means of its potent action as a histone deacetylase (HDAC) inhibitor. HDAC function is required for methylation-dependent transcriptional repression and DNA methylation and VPA could thus interact. These experiments should provide insight into the role of DNA methylation in fetal programming and open many avenues for future studies in this area. ? ?
