Deciphering the Mechanisms of Osteoblast-Induced Resistance of Leukemic Stem Cell (LSC) in Ph+ CML: Role of PI3-Kinase, BRD4 and MYC and Development of Strategies to Overcome Osteoblast-Induced Resistance

Chronic myeloid leukemia (CML) is a stem cell neoplasms characterized by the chromosome translocation t(9;22) and the related BCR-ABL1 fusion gene. Therapy with BCR-ABL1 kinase inhibitors is highly effective in the treatment of CML and deep molecular responses are achieved in most patients. However, not all patients respond to these drugs due to resistance of leukemic stem cells (LSC). Recent data suggest that the disease-related microenvironment, known as the stem cell niche contributes to drug resistance and relapse in CML. So far, little is known about the resistance mechanisms protecting niche cells in the bone marrow of patients with CML. We have recently shown that osteoblasts are a major site of LSC-mediated resistance against BCR-ABL1-targeting drugs in CML. In the current study, we screened for drugs that are able to suppress growth and viability of osteoblasts and/or other niche-related cells and can thereby overcome drug resistance of CML LSC. Proliferation was analyzed by determining ³H-thymidine uptake in niche-related cells and apoptosis was measured by Annexin-V/DAPI-staining and flow cytometry. We found that the dual PI3 kinase (PI3K) and mTOR inhibitor BEZ235 and the selective pan-PI3K inhibitor copanlisib suppress proliferation of primary osteoblasts (BEZ235 IC ₅₀: 0.05 µM; copanlisib IC ₅₀: 0.05 µM), the osteoblastic cell line CAL-72 (BEZ235 IC ₅₀: 0.5 µM; copanlisib IC ₅₀: 1 µM), primary human umbilical vein endothelial cells (BEZ235 IC ₅₀: 0.5 µM; copanlisib IC ₅₀: 0.5 µM) and the endothelial cell line HMEC-1 (BEZ235 IC ₅₀: 1 µM; copanlisib IC ₅₀: 1 µM), whereas no comparable effects were seen with the mTOR inhibitor rapamycin. As determined by flow cytometry, BEZ235 and copanlisib suppressed the expression of phosphorylated (p) pAKT and pS6 in endothelial cells and osteoblasts whereas rapamycin downregulated pS6 expression but did not decrease expression of pAKT. Moreover, we found that BEZ235 and copanlisib cooperate with nilotinib and ponatinib in suppressing growth and viability of osteoblasts and endothelial cells. Furthermore, BEZ235 and copanlisib were found to overcome osteoblast-induced resistance of K562, KU812 cells, and primary CD34 ⁺/CD38 ⁻ CML LSC against nilotinib and ponatinib. This effect was also seen when CAL-72 cells were first exposed to BEZ235 or copanlisib and washed before co-cultures with CML cells and BCR-ABL1 inhibitors were prepared, suggesting that osteoblast inhibition was a crucial event capable of disrupting LSC resistance in these co-cultures. Of all other drugs tested, only the BRD4-targeting drug JQ1 was found to suppress CAL72-induced resistance in the CML cell lines KU812 and K562, suggesting that osteoblast-induced resistance of CML cells is also mediated by a BRD4-MYC pathway. In a next step, we examined the expression of resistance-mediating immune checkpoint molecules on CML cells (KU812, K562, LSC) and on osteoblasts by flow cytometry. We found that CML cells and CAL72 cells constitutively express PD-L1 and that interferon-gamma (IFN-G) promotes the expression of PD-L1 in all cell types tested. Moreover, we found that the BRD4 blocker JQ1 and the BRD4-degrader dBET6 suppress the IFN-G-induced upregulation of PD-L1 in CML LSC and osteoblasts. In conclusion, our data show that osteoblast-induced resistance of CML stem cells is mediated by a PI3K-dependent pathway and BRD4/MYC, and that BRD4-inhibition or BRD4-degradation counteracts osteoblast-induced resistance of CML (stem) cells against BCR-ABL1 inhibitors and PD-L1 expression on CML LSC and osteoblasts. We hypothesize that checkpoint inhibition may assist in drug-induced eradication of CML LSC and thus in the development of curative drug therapies in Ph ⁺ CML.