Valproic acid disrupts the oscillatory expression of core circadian rhythm transcription factors.

Valproic acid (VPA) is a well-established therapeutic used in treatment of seizure and mood disorders as well as migraines and a known hepatotoxicant. About 50% of VPA users experience metabolic disruptions, including weight gain, hyperlipidemia, and hyperinsulinemia, among others. Several of these metabolic abnormalities are similar to the effects of circadian rhythm disruption. In the current study, we examine the effect of VPA exposure on the expression of core circadian transcription factors that drive the circadian clock via a transcription-translation feedback loop. In cells with an unsynchronized clock, VPA simultaneously upregulated the expression of genes encoding core circadian transcription factors that regulate the positive and negative limbs of the feedback loop. Using low dose glucocorticoid, we synchronized cultured fibroblast cells to a circadian oscillatory pattern. Whether VPA was added at the time of synchronization or 12h later at CT12, we found that VPA disrupted the oscillatory expression of multiple genes encoding essential transcription factors that regulate circadian rhythm. Therefore, we conclude that VPA has a potent effect on the circadian rhythm transcription-translation feedback loop that may be linked to negative VPA side effects in humans. Furthermore, our study suggests potential chronopharmacology implications of VPA usage.

Class I lysine deacetylases promote glucocorticoid-induced transcriptional repression through functional interaction with LSD1.

Small molecule inhibitors of lysine deacetylases (KDACs) are approved for clinical use in treatment of several diseases. Nuclear receptors, such as the glucocorticoid receptor (GR) use lysine acetyltransferases (KATs or HATs) and KDACs to regulate transcription through acetylation and deacetylation of protein targets such as histones. Previously we have shown that KDAC1 activity facilitates GR-activated transcription at about half of all cellular target genes. In the current study we examine the role of Class I KDACs in glucocorticoid-mediated repression of gene expression. Inhibition of KDACs through two structurally distinct Class I-selective inhibitors prevented dexamethasone (Dex)-mediated transcriptional repression in a gene-selective fashion. In addition, KDAC activity is also necessary to maintain repression. Steroid receptor coactivator 2 (SRC2), which is known to play a vital role in GR-mediated repression of pro-inflammatory genes, was found to be dispensable for repression of glucocorticoid target genes sensitive to KDAC inhibition. At the promoters of these genes, KDAC inhibition did not result in altered nucleosome occupancy or histone H3 acetylation. Surprisingly, KDAC inhibition rapidly induced a significant decrease in H3K4Me2 at promoter nucleosomes with no corresponding change in H3K4Me3, suggesting the activation of the lysine demethylase, LSD1/KDM1A. Depletion of LSD1 expression via siRNA restored Dex-mediated repression in the presence of KDAC inhibitors, suggesting that LSD1 activation at these gene promoters is incompatible with transcriptional repression. Treatment with KDAC inhibitors does not alter cellular levels of LSD1 or its association with Dex-repressed gene promoters. Therefore, we conclude that Class I KDACs facilitate Dex-induced transcriptional repression by suppressing LSD1 complex activity at selected target gene promoters. Rather than facilitating repression of transcription, LSD1 opposes it in these gene contexts.