Inhibition of BRD4 Bromodomains Is a Potent Novel Strategy to Target Apoptosis Resistance in Paediatric ALL

Abstract 77

Despite significant improvements in survival rates for children with B-cell acute lymphoblastic leukaemia (ALL), relapsed ALL still remains an incurable disease warranting more targeted therapies. Furthermore, there is a requirement for therapies with less toxic side-effects in order to improve the health of ALL survivors. We have previously reported that a subset of paediatric ALL tumours exhibit a defect in the apoptotic response to DNA damage in vitro, that these responses are associated with a poor clinical response in vivo, and that deregulation of multiple prosurvival pathways underlies this resistance. We also observed that pharmacological inhibition of individual prosurvival pathways leads to widely heterogeneous responses in vitro and does not appear to be the optimal strategy to overcome ALL apoptotic resistance.

Here, we present a novel approach in the treatment of resistant ALL that targets the histone acetyl-lysine binding bromodomains of a protein regulator of epigenetic memory and transcriptional activity, BRD4. Recent research has revealed that BRD4 remains bound to transcriptional start sites of genes expressed during the M/G1 transition, influencing mitotic progression, and mediating transcriptional elongation. We used the potent and highly selective synthetic inhibitor of BRD4 domains, JQ1, which recently demonstrated striking activity in a pre-clinical mouse model of the aggressive NUT-midline carcinoma (Filippakopoulos P et al. 2010; Nature; 468; 1067–1073).

Using a luminescent ATP-based cytotoxicity assay we measured cell viability following 72 hr exposure to JQ1 in a panel of ALL cell lines, including Ph⁺ ALL (SUP-B15, SD1, TOM-1); drug-resistant Ph⁻ ALL (REH, NALM-6); and T-ALL (Jurkat); as well as primary ALL tumours. JQ1 led to an impressive decrease in cell viability in all cell lines tested, with IC50 values below 850 nM. We subsequently observed that JQ1 also caused significant cytotoxicity in primary ALL tumours (n=13) within nanomolar concentrations compared to control PBMCs. Importantly, primary tumours derived from patients with high-risk, apoptosis-resistant ALL were found to be sensitive to the inhibitor. Finally, combined Dexamethasone/JQ1 treatment of ALL cell lines and primary ALL tumours revealed a high level of synergism.

We next addressed the mechanism of JQ1-induced killing. We observed that treatment of pre-B ALL cell lines with 1 μM JQ1 over 96 hr induced caspase-dependant apoptosis. Consistent with this, rapid dephosphorylation of pro-apoptotic BAD at Ser112 was noted 24 hr after incubation. Furthermore, JQ1 induced dramatic G1-arrest in all treated cells and c-Myc protein, constitutively expressed in all ALL cells, was found to be completely downregulated at the same time point.

To identify the full effect of BRD4 inhibition on transcription, we analysed global gene expression in 8 primary ALL tumours before and after exposure to 1 μM JQ1 using Affymetrix GeneChip Human Gene 1.0 ST Arrays. We identified 96 differentially downregulated genes including those that promote angiogenesis (ANGPT2, HGF, ELK-3) and inhibit apoptosis (BIRC3), whereas 127 genes were upregulated including those with a role in anti-angiogenesis (TIMP3), translational repression (EIF4EBP2, PAIP2B), and p53 stabilisation (PPP1R13B, DHRS2).

BRD4 has previously been reported to play a role in DNA replication (Maruyama T et al. 2002; Mol. Cell. Biol; 22; 18; 6509–6520), thus we reasoned that BRD4 inhibition might compromise this highly regulated process. We used the DNA fibre labelling technique to visualise the effect of JQ1 on replicating cells and showed that 24 hr exposure of NALM-6 cells to 1 μM JQ1 led to a decreased rate of replication fork progression as well as a substantial increase in stalled replication forks.

In conclusion, our results suggest that BRD4 inhibition instigates potent sensitisation of resistant ALL cells by multiple mechanisms: induction of apoptosis, repression of translation, cell cycle inhibition and inhibition of replication. This data strongly supports the rationale for the targeting of BRD4, and potentially other family members within this new class of chromatin-binding protein, as a new therapeutic strategy for the treatment of resistant ALL tumours.