Abstract

Abstract 2439

Tumor metabolism has emerged as a hallmark of cancer by which the oncogenic profile of cancer cells pairs energy availability with cell growth and survival. BCR-ABL positive acute lymphoblastic leukemia (BCR-ABL+ ALL) is a highly resistant phenotype, and the BCR-ABL fusion protein has been correlated with alterations in glucose metabolism. The glucose analogue 2-deoxy-D-glucose (2DG) has been found to be an effective antitumor agent in both animal models and human clinical trials for solid tumors based on the dependency of hypoxic tumor cells on anaerobic glycolysis to generate ATP. 2DG inhibits two key glycolytic enzymes, hexokinase (HK) and phosphoglucose isomerase (PGI), and interfers with N-linked glycosylation by its incorporation in place of mannose into lipid-linked oligosaccharide (LLO) chains which leads to premature termination of LLO synthesis. Unlike hypoxic solid tumors, we recently demonstrated that 2DG induces cell death in ALL under normoxia preferentially by inhibiting N-linked glycosylation resulting in ER stress/UPR-mediated apoptosis, although inhibition of glycolysis also contributed to cell death in a phenotype specific-manner (Mol Cancer Res 10:969, 2012). Among ALL subtypes, BCR-ABL+ ALL t(9;22) cell lines SupB15 and TOM1 exhibited the highest sensitivity to 2-DG suggesting that BCR-ABL expression may be linked to sensitivity to 2DG. To investigate the role of BCR-ABL expression in this process, we constructed NALM6 (Bp-ALL) stable cell lines expressing BCR-ABL p190 fusion, and assessed their level of sensitivity to 2DG under normoxia. Expression of the BCR-ABL p190 fusion in NALM6 cells significantly increased 2DG-induced apoptosis compared to mock transfected NALM6 cells, indicating that BCR-ABL expression results in higher sensitivity to 2DG in this experimental system. Western blot analysis of BCR-ABL+ ALL (SUPB15 and TOM1) and Bp-ALL (NALM6) cell lines demonstrated that BCR-ABL expression induced alterations in the PI3K/Akt/p70S6K, GSK-3β/Mcl-1, p-STAT3 (Ser727), p-PKCδ (Thr505), and GRP78/p-eIF2α (Ser51) signaling pathways. Treatment with 2DG resulted in upregulation of p-AMPK (Thr172) and p-GSK-3β (Ser9), and downregulation of p-mTOR (Ser2448), p-p70S6K (Thr389), p-STAT3 (Ser727) and Mcl-1 in both BCR-ABL+ ALL (SUPB15 and TOM1) and Bp-ALL cells (NALM6). As we have previously described, 2DG induces ER stress and the expression of the unfolded protein response (UPR) markers (Mol Cancer Res 10:969, 2012). In BCR-ABL+ ALL cells, 2DG led to upregulation of IRE1α, GRP78 and CHOP (a marker of UPR induced cell death) and resulted in apoptotic cell death (>60%), which correlated with lower level of Mcl-1 expression. To further investigate the role of Mcl-1 on 2DG's sensitivity we used pharmacological inhibitors and shRNA. We found that Mcl-1 down-regulation sensitized Bp-ALL cells to 2DG (36% ± 4% cell death; p<0.001 for NALM6/shMcl-1 vs. NALM6/shScramble treated with 2DG 0.5mM). We then examined the role of GSK-3β/Mcl-1 signaling in K562, a highly 2DG resistant CML cell line expressing BCR-ABL p210 fusion. We found that K562 cells expressed high levels of Mcl-1. Inhibition of PI3K/Akt/GSK-3β pathway using LY294002 sensitized K562 cells to 2DG, suggesting that downregulation of Mcl-1 by GSK-3β activation leads to 2DG sensitivity. Finally, we co-targeted BCR-ABL+ ALL cells with TKIs (Imatinib) plus the glycolytic/glycosylation inhibitor 2DG and found significant increase in cell death for the combination compared to single drug (62.6% ± 0.8% cell death; p<0.001 for 2DG 4mM + Imatinib 0.5μM vs. single drug). Further, no cross-resistance was detected between 2DG and Imatinib in TKI-resistant BCR-ABL+ ALL cells. Taken together, our data demonstrate that BCR-ABL p190 expression sensitizes Bp-ALL cells to 2DG by altering the GSK-3β/Mcl-1 signaling pathway and inducing ER stress/UPR mediated cell death. This study supports further consideration of strategies combining inhibitors of glucose metabolism such as 2DG with BCR-ABL TKIs for future clinical translation in BCR-ABL+ ALL.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.