BRD4 Antagonist and Histone Deacetylase Inhibitor: A Novel Synergistic Combination Against Human Acute Myeloid Leukemia (AML) Cells

The BET (bromodomain and extraterminal) protein family members including BRD4 bind to acetylated lysines on the histone proteins, help assemble transcriptional regulators at the target gene promoters and enhancers, and regulate the expression of important oncogenes, e.g., MYC and BCL-2. Here we determined the effect of histone deacetylase (HDAC) inhibitor panobinostat (PS) on sensitizing human AML blast progenitor cells (BPCs) to the BET protein inhibitor JQ1. Treatment with JQ1, but not its inactive enantiomer (R-JQ1), dose-dependently increased the % of cells in the G1 phase and reduced the % of S phase cells, while concomitantly inducing apoptosis in the AML BPCs expressing mutant NPM1c+ (OCI-AML3 cells) or with the ectopic co-expression of FLT3-ITD (OCI-AML3/FLT3-ITD cells), or in AML cells expressing MLL fusion oncoprotein (MV4-11 and MOLM13 cells). JQ1 was equally effective in inducing apoptosis of OCI-AML3 and OCI-AML3/FLT3-ITD cells. JQ1 treatment inhibited the clonogenic survival of OCI-AML3 more than of MOLM13 cells. Treatment with JQ1 also dose-dependently exerted lethal anti-leukemia effects against 10 primary CD34+ AML cell samples. JQ1 treatment reduced binding of BRD4 and RNA polymerase II to the DNA of MYC and BCL2. The heat map from the gene expression microarray profile in OCI-AML3 cells demonstrated that JQ1 treatment downregulated the mRNA expression of more genes, as compared to the number of genes whose mRNA expression was up regulated by JQ1 treatment. Total RNA from the untreated and JQ1-treated cells used for the microarray analysis was also used for quantitative PCR analysis, which confirmed that JQ1 treatment markedly decreased the mRNA expression of the MYC and BCL2 genes. Depletion of BRD4 by shRNA phenocopied the effects of JQ1 treatment in OCI-AML3 cells. While it had no effect on BRD4 and NPM1, JQ1 treatment dose-dependently depleted the protein levels of MYC, BCL2, CDK6 and pSer2 RNA POL II, as well as induced the levels of p21, p27, BIM and cleaved PARP in OCI-AML3. Similar effects were observed in a representative primary AML BPC sample that expressed mutant NPM1c+ and FLT3-ITD. As compared to each agent alone, co-treatment with JQ1 (but not its inactive enantiomer, R-JQ1) and panobinostat (PS) synergistically induced apoptosis of the AML BPCs, especially the stem cell sub-population of CD34+CD38-Lin- BPCs, but not of normal CD34+ hematopoietic progenitor cells. This was associated with greater attenuation of MYC and BCL2, while increasing the levels of p21, BIM and cleaved PARP levels in the AML BPCs. In the cells treated with BRD4 shRNA, but not with the non-targeted shRNA, PS treatment induced significantly more apoptosis of OCI-AML3 cells (p < 0.05). As compared to treatment with the vehicle alone, daily intra-peritoneal (IP) treatment x 5 days, for three weeks, with either JQ1 (50 mg/kg formulated in 10% 2-hydroxypropyl-β-cyclodextrin) or panobinostat (5 mg/kg by IP injection 3 times per week) significantly improved the survival of the NOD/SCID mice engrafted with OCI-AML3 AML xenografts in the bone marrow, without inducing any toxicity (p < 0.05). Notably, combined treatment with PS and JQ1 further significantly improved survival of the mice, as compared to treatment with JQ1 or PS alone (p < 0.001). In a Kaplan Meier plot depicting survival, this translated into a plateau in the survival curve. Co-treatment with PS and JQ1 was also associated with the most reduction in the levels of MYC, BCL-2 and CDK6 proteins in the AML xenograft. Collectively, these pre-clinical findings highlight a potential, therapeutically efficacious, combination of BRD4 antagonist and HDAC inhibitor for further development as a therapy of AML.