DNA methylation and histone post-translational modification stability in post-mortem brain tissue

BackgroundEpigenetic (including DNA and histone) modifications occur in a variety of neurological disorders. If epigenetic features of brain autopsy material are to be studied, it is critical to understand the post-mortem stability of the modifications.MethodsPig and mouse brain tissue were formalin-fixed and paraffin-embedded, or frozen after post-mortem delays of 0, 24, 48, and 72h. Epigenetic modifications frequently reported in the literature were studied by DNA agarose gel electrophoresis, DNA methylation enzyme-linked immunosorbent assays, Western blotting, and immunohistochemistry. We constructed a tissue microarray of human neocortex samples with devitalization or death to fixation times ranging from <60min to 5days.ResultsIn pig and mouse brain tissue, we found that DNA cytosine modifications (5mC, 5hmC, 5fC, and 5caC) were stable for 72h post-mortem. Histone methylation was generally stable for 48h (H3K9me2/K9me3, H3K27me2, H3K36me3) or 72h post-mortem (H3K4me3, H3K27me3). Histone acetylation was generally less stable. The levels of H3K9ac, H3K27ac, H4K5ac, H4K12ac, and H4K16ac declined as early as 24h post-mortem, while the levels of H3K14ac did not change at 48h. Immunohistochemistry showed that histone acetylation loss occurred primarily in the nuclei of large neurons, while immunoreactivity in glial cell nuclei was relatively unchanged. In the human brain tissue array, immunoreactivity for DNA cytosine modifications and histone methylation was stable, while subtle changes were apparent in histone acetylation at 4 to 5days post-mortem.ConclusionWe conclude that global epigenetic studies on human post-mortem brain tissue are feasible, but great caution is needed for selection of post-mortem delay matched controls if histone acetylation is of interest.