Bortezomib Has Anti-Proliferative and Apoptotic Effects Against CML Stem Cells, Including the Quiescent Population.

CML is a disease of the haemopoietic stem cell (HSC) caused by the constitutively active tyrosine kinase Bcr-Abl. The current first-line therapy for CML is imatinib (IM), a tyrosine kinase inhibitor (TKI) effective in establishing durable disease control in the majority of patients in chronic phase. However, the quiescent leukaemic HSC (Ph+HSC) population is insensitive to the apoptotic effects of this and more potent TKIs - such as dasatinib or nilotinib. These TKIs exert potent anti-proliferative effects against CML stem and progenitor cells, resulting in accumulation of quiescent cells in vitro. This Ph+HSC reservoir of disease in treated patients is now recognised to be a critical target if we wish to achieve complete eradication of residual disease. Bcr-Abl+ cells have abnormally high levels of proteasome activity and expression and in common with other cancer cells have been shown to be relatively more sensitive to the effects of proteasome inhibitors (PI) than their non-malignant counterparts. We have assessed the potential role for PI in CML using Bcr-Abl+ cell lines (K562 and Ba/F3 with mutant/wild type Bcr-Abl) and Ph+HSC obtained from leukapheresis products of patients, taken with consent, at time of diagnosis with CML. We have demonstrated apoptotic and anti-proliferative effects of PI (MG132 and bortezomib) in wild type and mutated Bcr-Abl+ cell lines (including T315I) and in Ph+HSC (n=7) by viable cell counts, thymidine incorporation and flow cytometry for annexin V and 7AAD. The apoptotic effects also occurred in the quiescent Ph+HSC fraction with an IC50 of 5–7.5nM (bortezomib)- similar to that seen in the total CD34+ cell population and lower than that previously reported for cells from multiple myeloma patients. These results were compared to parallel FACS-sorted cell populations from non-CML Ph- leukapheresis products and a therapeutic window was seen. CFSE tracking of cell division in CML samples in vitro indicated depletion of the quiescent fraction, in contrast to the accumulation of these cells demonstrated with IM, nilotinib or dasatinib, either alone or in combination. The intracellular effects of exposure of these cell populations to bortezomib were assessed by quantitatively measuring the intracellular accumulation of polyubiquitinated proteins seen following PI treatment. Interestingly no clear effect on Bcr-Abl activity was demonstrated with PI treatment of Ph+HSC (measured by immunoblotting with antibody to phosphorylated Crkl), though in Bcr-Abl+ cell lines detectable activity was reduced. Drug combination experiments (Calcusyn) were performed with bortezomib and IM or dasatinib to assess the effect of simultaneous inhibition of Bcr-Abl and the proteasome - the combination of a TKI with a PI resulted in additive effects, however no clear synergistic effects of significance were seen. Our results indicate that PI are capable of inducing CML stem cell specific apoptosis, suggesting the proteasome may be a relevant target for eradication of residual disease in CML following TKI therapy.