Disease relapse in B-cell acute lymphoblastic leukemia (ALL), either due to development of acquired resistance after therapy or because of de novo resistance, remains a therapeutic challenge. In the present study, we have developed a cytarabine (Ara-C)resistant REH cell line (REH/Ara-C) as a chemoresistance model. REH/Ara-C 1) was not crossresistant to vincristine or methotrexate; 2) showed a similar proliferation rate and cell surface marker expression as parental REH; 3) demonstrated decreased chemotaxis toward bone marrow stromal cells; and 4) expressed higher transcript levels of cytidine deaminase (CDA) and mito-NEET (CISD1) than the parental REH cell line. Based on these findings, we tested NL-1, a mitoNEET inhibitor, which induced a concentration-dependent decrease in cell viability with a comparable IC 50 value in REH and REH/Ara-C. Furthermore, NL-1 decreased cell viability in six different ALL cell lines and showed inhibitory activity in a hemosphere assay. NL-1 also impaired the migratory ability of leukemic cells, irrespective of the chemoattractant used, in a chemotaxis assay. More importantly, NL-1 showed specific activity in inducing death in a drug-resistant population of leukemic cells within a coculture model that mimicked the acquired resistance and de novo resistance observed in the bone marrow of relapsed patients. Subsequent studies indicated that NL-1 mediates autophagy, and inhibition of autophagy partially decreased NL-1-induced tumor cell death. Finally, NL-1 showed antileukemic activity in an in vivo mouse ALL model. Taken together, our study demonstrates that mitoNEET has potential as a novel antileukemic drug target in treatment refractory or relapsed ALL.