Autophagy is dispensable for Kmt2a/Mll-Mllt3/Af9 AML maintenance and anti-leukemic effect of chloroquine.

Recently, macroautophagy/autophagy has emerged as a promising target in various types of solid tumor treatment. However, the impact of autophagy on acute myeloid leukemia (AML) maintenance and the validity of autophagy as a viable target in AML therapy remain unclear. Here we show that Kmt2a/Mll-Mllt3/Af9 AML (MA9-AML) cells have high autophagy flux compared with normal bone marrow cells, but autophagy-specific targeting, either through Rb1cc1-disruption to abolish autophagy initiation, or via Atg5-disruption to prevent phagophore (the autophagosome precursor) membrane elongation, does not affect the growth or survival of MA9-AML cells, either in vitro or in vivo. Mechanistically, neither Atg5 nor Rb1cc1 disruption impairs endolysosome formation or survival signaling pathways. The autophagy inhibitor chloroquine shows autophagy-independent anti-leukemic effects in vitro but has no efficacy in vivo likely due to limited achievable drug efficacy in blood. Further, vesicular exocytosis appears to mediate chloroquine resistance in AML cells, and exocytotic inhibition significantly enhances the anti-leukemic effect of chloroquine. Thus, chloroquine can induce leukemia cell death in vitro in an autophagy-independent manner but with inadequate efficacy in vivo, and vesicular exocytosis is a possible mechanism of chloroquine resistance in MA9-AML. This study also reveals that autophagy-specific targeting is unlikely to benefit MA9-AML therapy.