The BH3-Mimetic Obatoclax Restores the Response to Dexamethasone in Glucocorticoid-Resistant ALL through Induction of Autophagy.

Incorporation of cell death-inducing agents into current therapeutic regimens is an attractive strategy to improve treatment for drug resistant leukemia. We tested the potential of the pan-Bcl-2 family antagonist BH3 mimetic obatoclax (GX15-070) to restore the glucocorticoid (GC) response in GC-resistant childhood acute lymphoblastic leukemia (ALL). One tenth of the IC50 dose of obatoclax was sufficient to restore GC-sensitivity in vitro of primary ALL cells from T-cell and precursor B-cell ALL patients with poor in vivo response to prednisone. In GC-resistant cell lines, this effect was associated with strong caspase activation, externalization of annexin V and resulted in caspase-dependent cell death. Overexpression of either Mcl-1 or Bcl-X_L interfered with the cytotoxic effect of obatoclax as single agent, but not with the GC-sensitizing effect of low-dose obatoclax. We hypothesized that an alternative mechanism could be involved, which may be independent of the regulation of the intrinsic pathway of apoptosis by members of the Bcl-2 family. Consistent with this hypothesis, we observed that obatoclax was also cytotoxic for mouse embryonal fibroblasts with homozygous deletions of Bax and Bak (Bax/Bak DKO), in which the intrinsic pathway is blocked completely. The IC50 of obatoclax on DKO cells was 200 nM. Given that induction of autophagy can lead to an alternative mode of programmed cell death, and because autophagy induction is also controlled by anti-apoptotic Bcl2-family members via an interaction with the essential autophagy BH3-only protein Beclin-1, we asked whether autophagy involved in mediating the therapeutic effects of obatoclax in our model. Using the LC3-GFP reporter system, we detected strong induction of autophagic vacuoles after treatment of Bax/Bak DKO and ALL cells with IC50 concentrations of 200 and 800 nM obatoclax, respectively. Autophagy induction was confirmed by detection of the expected cleavage products of the endogenous autophagy protein LC-3 by Western blotting. Low dose obatoclax induced autophagy only in combination with dexamethasone. Inhibition of autophagy using 3-methyl-adenosine (3-MA) or downregulation of Beclin-1 by RNA interference abrogated the GC-sensitizing effect of obatoclax completely in ALL cells, and strongly interfered with the cytotoxic effect of obatoclax on DKO cells. The restoration of GC-resistance by Beclin-1 knock-down was associated with a marked decrease of caspase-3 activation. These data indicate that induction of autophagy by obatoclax is required to initiate a caspase-dependent death pathway. Furthermore, low-dose obatoclax did not impair the response to daunorubicine, vincristine and asparaginase in GC-resistant ALL cells, suggesting that induction of autophagy will not lead to drug resistance in this context. Ongoing experiments will validate the effect of obatoclax in a leukemia xenograft model of GC-resistant ALL. Collectively, our results support the notion that the incorporation of obatoclax into GC-based regimens is a promising strategy to improve the treatment of GC-resistant ALL and provide new angles for the investigation of mechanisms that are implicated in altered drug responses in resistant disease.