Correction of Aberrant E-Cadherin Splicing with HDAC Inhibitors in CLL

Abstract 1776

E-cadherin gene is aberrantly spliced in CLL cells. This results in a transcript that lacks exon 11 and is non-functional. As CLL specimens express higher proportion of these transcripts as compared to normal B cells, this aberrant splicing is a mechanism of E-cadherin inactivation that results in wnt pathway activation. To design ways to inhibit this aberrant splicing, we analyzed epigenetic histone modifications in CLL specimens as the histones are critically involved in gene transcription, exon definition and splicing. Our findings show that increasing histone acetylation by histone deacetylase inhibitors (HDACi) can increase the expression of correctly spliced transcript and at the same time reduce the aberrantly spliced transcript (exon 11 skipped) resulting in re-expression of this tumor suppressor gene. It is known that increase in histone acetylation upon HDACi exposure results in an open chromatin structure and an increase in transcription. Exposure of CLL specimens (n=10) to HDACi, MS275 (0.1-1μM for 24–48 hours) resulted in a 4–16 fold increase in E-cadherin expression with a decrease in the aberrant transcript (undetectable levels in 6 CLL specimens). Similar results were obtained with other HDACi, SAHA and Valproic acid and the increase in expression was also confirmed by western blot analysis. Chromatin immunoprecipitation (ChIP) experiments with a pan-acetyl H3 and H4 antibody confirms that the E-cadherin promoter and exon 11 region of this gene are indeed hypoacetylated as compared to PBMC and the treatment with HDAC inhibitor increases histone acetylation at both the promoter and exon 11 region.

We investigated the wnt pathway as lack of E-cadherin expression is a mechanism of wnt pathway activation in CLL cells. Re-expression of E-cadherin by HDACi led to inhibition of this pathway as two downstream effectors, LEF and cyclinD1 genes were downregulated by real time PCR analysis indicating that this strategy could be used to inhibit this pathway in a clinical setting as well. Apoptosis was also observed in CLL specimens treated with HDACi. To analyze the global changes induced by histone modifications and their reversal by HDACi, we analyzed a defined gene set PCR ChIP promoter array. CLL specimens were compared to PBMC and also analyzed for the change in histone patterns upon HDAC inhibitor treatment. From this array a number of interesting candidate genes that are also affected by histone modifications were identified including E-cadherin, BAX, a bcl2 inhibitor and CDKN2B, a gene that inhibits activation of CDK kinases were hypoacetylated in CLL specimens as compared PBMC, however underwent hyperacetylation with HDACi treatment. BCL2 and NFKB1 promoters were hyperacetylated or transcriptionally active in CLL specimens. Besides the histone modifications on the ChIP array, the change in expression of these genes upon HDACi treatment was also confirmed by real time PCR analysis. With the E-cadherin gene as a model of histone epigenetic changes in CLL specimens, the studies highlight a signature pattern of these modifications in CLL cells. It is expected that these epigenetic histone changes will influence gene expression levels, RNA isoforms and splice variants and will have a role in CLL biology.