The BET (bromodomain and extra terminal) protein family, including BRD4 are chromatin reader proteins that bind to acetylated lysines on histone proteins and recruit transcriptional regulatory complexes to gene promoters, thereby coupling histone acetylation to RNA polymerase II (RNAP2)-mediated transcript elongation. Preferential binding of BRD4 to clustered enhancers or ‘super enhancers’ enables BRD4, through the mediator complex, to regulate the transcription of specific oncogenes, including c-MYC, BCL2 and CDK4/6 in AML blast progenitor cells (BPCs). Through its C-terminal domain (CTD), BRD4 recruits pTEFb (positive transcription elongation factor b), the heterodimer composed of cyclin T and CDK9, which phosphorylates the heptad repeats on the CTD of RNAP2 at serine 2. The kinase activity of CDK9 in pTEFb is also required for the phosphorylation of DSIF and NELF, which abrogates their inhibitory hold on the pause-release of RNAP2 and productive transcript elongation of specific oncogenes and their target genes. pTEFb exists in two forms, the catalytically active form, which is associated with BRD4 and RNAP2, and a catalytically inactive form associated with 7SK snRNP (small nuclear ribonucleoprotein) and the inhibitory protein HEXIM1 (hexamethylene bis-acetamide or HMBA-inducible protein 1). The transition of pTEFb between the inactive and active complex is dynamic and regulated by cell growth and stress signaling. HEXIM1 binds to and sequesters pTEFb, thereby inhibiting its kinase activity for RNAP II. We recently reported (Mol Cancer Ther, 2014) that the BRD4 antagonist JQ1 and I-BET151 alone, and synergistically with the pan-histone deacetylase inhibitor panobinostat, inhibits the growth and induces apoptosis in cultured (OCI-AML3 and MOLM13) and patient-derived primary AML (pAML) BPCs. JQ1 treatment also improved the survival of the immune-depleted mice engrafted with MOLM13 or pAML BPCs. Notably, the lethal activity of JQ1 was associated with marked induction of HEXIM1. Therefore, we hypothesized that HEXIM1 induction, by sequestering pTEFb and inhibiting RNAP2, contributes to the growth inhibitory and lethal effects of JQ1. We additionally hypothesized that co-treatment with those agents that further augment HEXIM1 induction would synergize with JQ1 against AML BPCs. To validate these hypotheses, we achieved a stable lentivirus mediated knockdown (KD) of the mRNA and protein expressions of HEXIM1 by approximately 90% in OCI-AML3 and MOLM13 cells, as well as transient KD in pAML BPCs. Compared to the parental control cells, OCI-AML3/HKD and MOLM13/HKD cells showing HEXIM1 knockdown (HKD) exhibited reduced binding of HEXIM1 with cyclin T (as revealed by cyclin T pull-down and immunofluorescence microscopy), but showed augmented suspension culture growth and increased protein levels (by Western blot) of cyclin T, pRNAP2 and cMYC levels. However, in contrast to the parental control cells, JQ1 (1000 nM) treatment-mediated HEXIM1 induction and binding to cyclin T was markedly inhibited in the HKD AML BPCs, which was associated with significant inhibition of JQ1-induced pRNAP2, as well as differentiation and apoptosis of HKD BPCs (p < 0.01). Notably, whereas treatment with JQ1 alone improved the median survival of NOD/SCID mice engrafted with MOLM13 BPCs, this survival improvement was significantly impaired in the mice engrafted with MOLM13/HKD AML BPCs (p < 0.02). This indicated that HEXIM1-induction mechanistically contributes to JQ1-induced growth inhibition, differentiation and apoptosis of AML BPCs. Further, the synergistic apoptosis observed in MOLM13, OCI-AML3 and primary AML cells induced by co-treatment with JQ1 (250 to 1000 nM) and PS (5 to 10 nM) or HMBA (1 to 2 mM) was also associated with greater induction of HEXIM1. However, this synergy was impaired in the cultured and pAML BPCs with HEXIM1 knockdown. These findings confirm that HEXIM1 induction is a biomarker and plays a mechanistic role in mediating growth inhibition, differentiation and apoptosis induced by BRD4 antagonist in human AML BPCs. These findings support, and we have embarked upon, studies involving future pre-clinical development of those agents that in combination significantly augment BRD4 antagonist-induced HEXIM1 in human AML BPCs.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution