BRD4 Degradation Is a Potent Approach to Block MYC Expression and to Overcome Multiple Forms of Stem Cell Resistance in Ph⁺ CML

Chronic myeloid leukemia (CML) is a hematopoietic stem cell neoplasm in which BCR-ABL1 acts as a major driver of proliferation, differentiation and survival of leukemic cells. In a majority of all patients, leukemic cells can be kept under control by BCR-ABL1 tyrosine kinase inhibitors (TKI). Nevertheless, resistance against one or more TKI may occur. Therefore, research is focusing on novel potential drug targets in CML. We have recently identified the epigenetic reader bromodomain-containing protein 4 (BRD4) as a new therapeutic target in leukemic stem cells (LSC) in acute myeloid leukemia. In the present study, we examine the expression of BRD4 and its downstream effector MYC in CML cells and asked whether BRD4 serves as a drug target in CML cells and whether BRD4-targeting drugs, including JQ1 and newly developed BRD4 degraders (dBET1 and dBET6) are able to overcome LSC resistance in CML. Primary CML cells were obtained from 22 patients with chronic phase (CP) CML and 3 with blast phase (BP) CML. As determined by qPCR and/or immunocytochemistry, the CML cell lines KU812 and K562 as well as primary CML cells expressed BRD4 and MYC. All three BRD4-targeting drugs (JQ1, dBET1 and dBET6) were found to decrease MYC expression in KU812 and K562 cells as assessed by Western blotting. In ³H-thymidine uptake experiments, JQ1 and dBET6 were found to inhibit the proliferation of KU812 in a dose-dependent manner (IC_50, JQ1: 100-500 nM; dBET6: 50-100 nM) whereas dBET1 showed only little if any effects on growth of KU812 cells (IC₅₀: 1-5 µM), and in K562 cells, only dBET6 was found to inhibit growth with a reasonable IC₅₀ value (250-500 nM). Corresponding results were obtained when examining drug effects on survival of CML cell lines by Annexin-V/PI staining. All three BRD4-targeting drugs were found to inhibit proliferation of primary CP CML cells with varying IC₅₀ values. As expected, growth-inhibitory effects of dBET6 were more pronounced (IC₅₀: <100 nM) compared to effects seen with JQ1 and dBET1. dBET1 and dBET6 were also found to inhibit growth of primary CML cells obtained from patients with BP CML, whereas JQ1 was not effective. JQ1 also failed to suppress survival on CML CD34⁺/CD38⁻ LSC. By contrast, dBET1 induced apoptosis in CML LSC at 1 µM and dBET6 induced apoptosis in CML LSC at 0.1 µM. dBET6 induced apoptosis in CML LSC obtained from patients with imatinib-sensitive CML as well as patients with imatinib-resistant CML harboring BCR-ABL1 T315I or BCR-ABL1 F317L. Finally, pre-incubation of CD34⁺ CP CML cells with dBET6 resulted in reduced leukemic engraftment in NSG mice exhibiting human membrane-bound stem cell factor, SCF [NSG-Tg(hu-mSCF)] 6 months after transplantation (engraftment with CD45⁺/CD33⁺/CD19⁻cells in control mice receiving DMSO-treated cells: 8.1±6.6% vs mice receiving dBET6-treated cells: 1.1±0.6%). To further explore the ability of dBET6 to interfere with LSC resistance in CML, we established a co-culture system mimicking LSC-niche interactions in the osteoblastic niche. In this model, co-culturing K562 cells, KU812 cells or primary CML LSC with the osteoblast-like osteosarcoma cell line CAL-72 resulted in resistance against nilotinib and ponatinib. In this culture system, JQ1 was found to partially restore TKI effects in K562 cells and completely restored TKI effects in KU812 cells. Interestingly, JQ1 was not able to restore TKI effects in primary CML LSC in these co-cultures. However, dBET6 was found to overcome niche cell-induced TKI-resistance of primary CML LSC. Finally, we were able to demonstrate that JQ1, dBET1 and dBET6 inhibit interferon-gamma-induced upregulation of PD-L1 expression in CML LSC. Together we show that BRD4 and MYC are potential new therapeutic drug targets in CML and that the BET-degrader dBET6 overcomes multiple forms of LSC resistance, including i) intrinsic resistance, ii) mutation-induced resistance, iii) niche induced resistance and iv) checkpoint-mediated resistance. Whether BRD4 degradation is also able to overcome TKI-resistance of BCR-ABL1+ LSC in vivo in patients with CML remains to be determined in clinical trials.