Abstract
Although treatment of childhood acute lymphoblastic leukemia (ALL) has greatly improved over the past decades, resistance to the glucocorticoids prednisolone and dexamethasone still causes a serious drawback in the success rate of conventional chemotherapy. Resistance to prednisolone at initial diagnosis in particular is related to an unfavorable event free survival. In addition, in vitro prednisolone resistance is recognized as an important negative parameter for long-term clinical outcome, even in patients who initially have a good in vivo response to one week with glucocorticoids. Therefore, alternative therapies that reverse resistance towards prednisolone and dexamethasone are indispensable to increase treatment efficacy. However, the exact mechanisms that cause cellular resistance to glucocorticoids are currently unknown. A better understanding of the pathways that lead to glucocorticoid resistance will thus be essential to be able to improve treatment results. We previously identified new genes that could be associated with prednisolone resistance in pediatric ALL by gene expression profiling [Holleman et al., N Engl J Med 2004]. Many of these genes were linked to glucose metabolism and were found to be significantly upregulated in prednisolone-resistant precursor B-ALL cells (p≤0.001). Expression levels of hypoxia-inducible factor-1 alpha (HIF-1a), glyceraldehyde-3-phosphatedehydrogenase (GAPDH), carbonic anhydrase 4 (CA4) and glucose transporter 3 (GLUT3/SLC2A3) were higher in prednisolone resistant cells when compared to prednisolone sensitive leukemia cells (fold change = 1.43, 1.26, 1.56 and 2.08 respectively). To investigate the causal relationship between the glycolysis pathway and prednisolone resistance, the glycolytic rate of various prednisolone-resistant and –sensitive leukemic cell lines was determined. The prednisolone-resistant cell lines Jurkat and Molt4 (LC50≥150) showed a higher glucose consumption compared to the prednisolone-sensitive cell lines Tom-1 and RS4;11 (LC50≤0.1), concordant with the microarray data. Several drugs that have been reported to affect glycolysis were subsequently tested for their ability to reverse resistance to glucocorticoids, using the in vitro MTT cytotoxicity assay. Treatment of the prednisolone-resistant cell lines Jurkat and Molt4 with 2-deoxyglucose (2-DG), Lonidamine or 3-Bromopyruvate increased the sensitivity to prednisolone and dexamethasone about ten-fold, while treatment of the prednisolone sensitive cell lines Tom-1 and RS4-11 had little or no effect. Importantly, the sensitizing effect of 2-DG on glucocorticoid resistance was not limited to cell lines, but was also observed in isolated primary leukemia cells being resistant to prednisolone whereas the effect was neglectable in sensitive cases. Moreover, downregulation of the expression of GAPDH by RNA interference also sensitized to prednisolone, comparable to treatment with glycolytic inhibitors. RNA interference for the transcription factor HIF-1a, that regulates the expression of genes involved in glycolysis, did not alter the sensitivity for prednisolone, indicating that the upregulation of HIF-1a is not causally related to glucocorticoid resistance in ALL. Together, these findings indicate the importance of the glycolysis pathway in glucocorticoid resistance in leukemia, and suggest that targeting this pathway is a viable strategy for modulating prednisolone resistance in ALL.
Low dose of 2-deoxy-D-glucose kills acute lymphoblastic leukemia cells and reverses glucocorticoid resistance via N-linked glycosylation inhibition under normoxia