Principal component analysis based unsupervised feature extraction applied to publicly available gene expression profiles provides new insights into the mechanisms of action of histone deacetylase inhibitors

Abstract

Publicly available gene expression profiles of the hippocampus measured during the successful administration of the histone deacetylase inhibitor, CI-994, to assist the extinction of mice contextual fear conditioning were re-analyzed using the recently proposed principal component analysis based unsupervised feature extraction. We identified 30 genes associated with differential gene expression in the hippocampus of mice treated with the HDAC inhibitor compared to controls; most of these genes code for postsynaptic density proteins. These 30 genes significantly overlapped with those detected by treatment with another HDAC inhibitor, FTY720, during similar contextual fear conditioning. However, because the 30 genes did not strongly overlap with genes associated with histone acetylation during contextual fear conditioning, altered histone modification in response to HDAC inhibitor treatment might not be the primary mechanism of effective extinction of contextual fear conditioning. Based on the results of our analyses we propose that HDAC inhibitors affect the temporal expression of the above genes via direct as well as indirect mechanisms that involve calcium signaling.