Targeting Chronic Myeloid Leukemia Stem Cells through Pharmacological Inhibition of HIF-1α

Chronic myeloid leukemia (CML) is a clonal hematopoietic stem cell (HSC) disorder due to the 9;22(q34.1;q11.2) reciprocal chromosomal translocation, which results in the generation of BCR/abl "fusion" oncogene and the expression of the BCR/Abl oncoprotein, a constitutively-active tyrosine kinase (TK). To date, chronic-phase CML patients are treated with TK inhibitors (TKi) such as imatinib- mesylate (IM) and dasatinib. However, to eradicate CML, it is necessary to eliminate Leukemia Stem Cells (LSC), which are largely insensitive to TKi. Thus, TKi, rather than curing CML definitively, induce a state of minimal residual disease (MRD) apparently sustained by LSC persistence. Therefore, the identification of non-TKi drugs capable to target and eradicate LSC responsible for MRD and therefore the risk of relapse of disease is of primary importance.

LSC, as well as normal HSC, in vivo reside in low oxygen tension areas of the bone marrow (BM) called "HSC niche". Several previous studies from our group and others demonstrate that low oxygen tension maintains the survival and stem cell potential of HSC and LSC, favoring their self-renewal. Interestingly, we also found that the expression of BCR/Abl oncoprotein is suppressed in low oxygen, which well explains the refractoriness of LSC to TKi, provided they manage to survive in the absence of BCR/Abl signaling.

Hypoxia-Inducible Factor-1α (HIF-1α) is a key regulator of cell adaptation to low oxygen and of LSC maintenance. This study addresses the in vitro and in vivo effects of the pharmacological inhibition of HIF-1α on the maintenance of CML stem cell potential.

After screening several HIF-1α inhibitors, we found that Acriflavine (ACF) induced apoptosis and DNA damage in CML cells and identified ACF as a very effective inhibitor of CML cell survival and growth in low oxygen, as regards the number of viable cells in culture and colony formation efficiency of CML cell lines or human or murine primary cells. Interestingly, ACF did not affect the colony formation efficiency of normal human hematopoietic cells. We also demonstrated in this study that IM and ACF exhibited an at least additive effect as far as the reduction of the total number of viable cells in low oxygen is concerned. Moreover, we demonstrated that low-oxygen incubation of CML cells increased the expression of genes related to stem cell potential, such as Nanog, Sox2 and Oct4, and that ACF treatment reduced their expression. These results on one hand confirmed that hypoxia selects cells with SC potential, on the other hand they demonstrated that ACF treatment reduces their maintenance.

Moreover, we evaluated the maintenance of CML stem cell potential using a stem cell assay, demonstrating that ACF suppressed the stem cell potential of CML cell lines or primary cells. On the contrary, IM or Dasatinib, were unable to affect the stem cell potential of CML cells also when administered to primary samples neither TKi-treated before nor carrying primary mutations of BCR/Abl kinase domain. This is in agreement with what observed before by us and others and confirms the refractoriness of LSC to TKi.

The effects of ACF in vivo were evaluated using a murine CML model where the disease was induced by a BCR/abl-carrying retrovirus. A 10 day-long ACF treatment was sufficient to attenuate CML development in mice, as evidenced by lower spleen weight, fewer leukemia cells in peripheral blood and BM, and a markedly reduced leukemia cell infiltration into the lungs. Moreover, it was possible to confirm in vivo the effects of ACF on LSC observed in vitro. ACF treatment indeed reduced the maintenance of the Long-Term LSC-containing LSK cell subset (CD34^- LSK) in the BM of CML mice. Interestingly and in keeping with the less-severe addiction of HSC to HIF-1α deficiency, we found that non-CML cells were significantly less sensitive to ACF treatment than CML cells. This points to a good therapeutic index of ACF in discriminating leukemic from normal hematopoietic cells.

On the basis of all above, the ACF/TKi combination may represent a novel therapeutic approach to target at one time CML cell bulk and LSC, in order to induce remission and prevent MRD of CML. Interestingly, ACF is an already FDA-approved drug for non-oncological uses in humans.