Introduction: histone acetylation is a post-translational modification that activates gene expression, and can be targeted by a class of compounds that inhibit HDAC enzymes. Pharmacological HDAC inhibitors such as SAHA are currently in clinical use with more than 20 compounds in phase I, II and III trials. In addition, they are known to suppress inflammatory responses and recent evidence suggests that HDAC inhibitors could be used for the treatment of cardiovascular disease and diabetes. Whereas HDAC inhibitors cause histone hyperacetylation which has long been thought to be the paradigmatic mechanism of action, the extent of the genome-wide increases and decreases in acetylation in healthy and diseased tissues remain poorly defined. Hence, the aim of this study is to investigate the regulation of gene expression in endothelial cells derived from healthy and diabetic individuals.

Methods: primary human aortic endothelial cells (HAECs) derived from healthy and diseased (diabetic) individuals were stimulated with SAHA (12hs, 2 μM, in triplicate). We profiled gene expression using RNA-seq and histone H3 acetylation of lysine 9 and 14 (H3K9/14ac) using chromatin immunoprecipitation sequencing (ChIP-seq).

Results: in addition to histone acetylation associated with gene activation, we observed broad histone deacetylation correlated with gene suppression following HDAC inhibition in healthy HAECs. Indeed, 7775 promoters were reduced by up to eight-fold the starting level. Integration with ENCODE transcription factor binding sites data identified deacetylation predominantly at target genes of the histone acetyltransferase p300-CBP coactivator family member, p300. RNA-seq of HAECs stimulated with C646, an inhibitor of HATs p300 and CBP, and p300 siRNA suggest p300/CBP-dependent gene suppression of some deacetylated genes. We also observe reduced CBP binding at the promoters of IL6 and BMX determined by ChIP-PCR. In contrast to the strong deacetylation response in healthy HAECs, 4868 promoters were deacetylated in diabetic HAECs stimulated with SAHA, with acetylation reduced by up to two-fold. Further investigation identified that initial levels of acetylation were significantly lower in the diabetic cells, including at genes regulated by p300. Indeed, SAHA stimulation in diabetic HAECs resulted in a shift of the acetylation signature to one that more closely resembles healthy cells.

Conclusion: we show that HDAC inhibitor mediated histone deacetylation differs in healthy and diseased diabetic HAECs, and may be p300-dependent.