Abstract 4448

Epigenetic therapies are being increasingly used to treat hematological disorders, such as myelodysplasia (MDS), with good responses in a significant proportion of patients. Histone deacetylase (HDAC) inhibitors are one of the classes of agents available for this purpose. Their therapeutic function is due to the ability to modify gene expression profile through acetylation of histones and non-histone proteins but little is currently known regarding the molecular mechanism by which HDAC inhibitors modulate gene expression.

The aim of this study is to dissect the promoter sequence characteristics required for transcriptional regulation of house-keeping genes by HDAC inhibitors and to identify genes which could be potential targets for this epigenetic therapy.

To identify promoter regions that predict response to HDAC inhibitors we first assessed the effect of sodium butyrate, and vorinostat on the expression levels of several house-keeping genes in hematopoietic cell lines K562 (erythroid) and HL60 (myeloid) by quantitative RT-PCR (qRT-PCR). Amongst the analyzed genes, the expression levels of IER3, RAI3, cFOS, Gravin, and COX2 were highly increased by sodium butyrate and vorinostat in both cells lines. Analysis of the expression levels of these genes in response to sodium butyrate and vorinostat in primary pro-myeloid (CD33+) and pro-erythroid (CD235A+) cells, isolated from normal bone marrows from patients with non-myeloid pathologies, revealed that, similarly to HL60 and K562, the expression levels of IER3, cFOS, and COX2 are also significantly increased in cells treated with these epigenetic agents when compared to control cells. Inhibition of SP1 transcriptional activity by mithramycin A led to a partial reduction of sodium butyrate-induced expression of these genes in HL60 and, by contrast, to an enhancement of both sodium butyrate- and vorinostat-increased expression of these genes in K562. These observations suggest that the action of these HDAC inhibitors in these hematopoietic cell lines are in part mediated by SP1. However the effect defends on the lineage: while in HL60 the sodium butyrate-induced gene expression requires SP1 binding to the promoter regions of these genes, in K562 both sodium butyrate and vorinostat-increased gene expression involve the disruption of SP1 binding from the promoter regions of these genes. To identify the promoter elements of IER3 that mediate their expression in response to sodium butyrate and vorinostat, a series of progressive deletion mutants of IER3 promoter-luciferase constructs were generated. The reporter assays performed with IER3 deletion constructs in both HL60 and K562 reveals that deletion of the -91 to -61 sequence leads to a significant decrease of IER3 basal transcriptional activity and additionally abolishes vorinostat-increased transcriptional activity of this gene, indicating that this region contains the minimal vorinostat-responsive element and also the minimal cis-acting sequences required for basal promoter activity in these cells. Bioinformatic analysis of the IER3 -91 to -61 sequence shows that this region contains canonical binding sites for several transcription factors (TFs), some of them shown to be critical for IER3 transcriptional activity in other cells. To precisely map the vorinostat-responsive element within -91 to -61 IER3 promoter region, several mutant reporter plasmids for the different TF binding sites present in this region were obtained by site-directed mutagenesis. Transcriptional activity of these mutant reporters in response to vorinostat in both K562 and HL60 was assessed by reporter assays. We found that mutation of a specific binding site for two zinc-finger TFs within -91 to -61 IER3 promoter significantly reduces IER3 basal transcriptional activity and fully abrogates its vorinostat-induced expression in both K562 and HL60. These results indicate that vorinostat-mediated IER3 expression in hematopoietic cells requires this canonical binding site. Further studies need to be performed to confirm the TF involved.

These results will be used to identify target genes amongst those known to be suppressed in MDS. The expression levels of these responsive genes to vorinostat in bone marrow cells of MDS patients could then be used as predictive marker to distinguish patients that will respond to this epigenetic therapy and thus allowing an effective targeting personalized therapy.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution