Using cyclosporin A (CsA) to inhibit P-glycoprotein (P-gp) funcoxidative phosphorylation, such as cyanide and azide, thereby tion we showed previously that there was a discordance depleting cellular ATP and inhibiting all energy-dependent between the ability of acute myeloid leukemic (AML) blast cells processes. This strategy was used originally to demonstrate the to accumulate daunorubicin and P-gp antigen expression (Xie ATP dependence of P-gp, 7 and more recently has been shown et al, Leukemia 1995; 9: 1882-1887). This discordance suggests to inhibit drug efflux via MRP. 8,9 We now report that it also that a CsA-sensitive drug efflux mechanism distinct from P-gp is expressed in many clinical samples. In the present study appears to show the presence of high levels of non-P-gp The following human cell lines were used to study energy- AML lines OCI/AML 1-5, established in this Institute. In

2581 Oxidative metabolism generates intracellular energy and metabolic intermediates necessary to promote the growth of AML cells. Recently, we demonstrated that AML cells are uniquely sensitive to inhibition of mitochondrial translation. Therefore, we characterized the structure and metabolic function of the mitochondria in AML and normal hematopoietic cells. Compared to normal cells (n = 10), 1° AML cells (n = 12) had increased mitochondrial mass and increased levels of the NRF-1, TFAM, EF-Tu and Myc, genes that positively regulate mitochondrial biogenesis. By transmission electron microscopy, we demonstrated that the mitochondria in 1°AML cells were larger in area, but fewer in number compared to normal CD34+ cells. Given the dysregulated mitochondrial biogenesis in 1° AML cells, we examined the activity and reserve capacity of the respiratory complexes in 1° AML and normal cells. When normalized for mitochondrial mass, 1°AML cells (n = 12) had reduced activity of respiratory complexes III, IV and V compared to normal cells (n = 10). Thus, despite the increased mitochondrial mass in AML, respiratory chain complex activity did not increase proportionately. Next, we evaluated the spare reserve capacity in AML cell lines, 1° AML samples, and normal cells. Spare reserve capacity reflects the difference between basal and maximal respiratory rate and was determined by measuring oxygen consumption after treatment with oligomycin to block ATP synthesis and FCCP to uncouple ATP synthesis from the electron transport chain. The spare reserve capacity in AML cells and 1o samples was lower than normal hematopoietic cells. In order to determine the reserve capacity in individual respiratory complexes, we evaluated the rate of oxygen consumption in 1°AML and normal cells by treating the cells with increasing concentrations of the complex I, III IV, and V inhibitors, rotenone, antimycin, sodium azide, and oligomycin, respectively, and measuring changes in oxygen consumption. AML cells displayed less reserve capacity in the individual complexes compared to normal hematopoietic cells, and the differences were most striking for complexes III and IV. Consistent with the reduced reserve capacity, AML cells were more sensitive to respiratory chain inhibitors. We then employed a genetic approach to investigate the relationship between mitochondrial mass and spare reserve capacity using P493 Burkitt's cells with inducible myc as we and others have previously shown that myc regulates mitochondrial mass. Compared to myc knockdown cells, myc +P493 cells had increased mitochondrial mass, larger mitochondria, increased basal oxygen consumption, but reduced activity of respiratory complexes III, IV and V when normalized for mitochondrial mass, compared to myc - cells. In addition, myc expressing cells had less spare reserve capacity in their respiratory chain. Thus, in this isogenic cell line, increased mitochondrial mass was not accompanied by a proportionate increase in respiratory chain activity resulting in decreased spare reserve capacity. Given the reduced reserve capacity in AML cells, we evaluated the effects of increasing electron flux through respiratory chain. We speculated that the low spare reserve capacity would render AML cells more vulnerable to oxidative stress. To test this strategy AML cells and 1° samples as well as normal cells were treated increasing concentrations of the fatty acid substrate palmitate or the TCA cycle substrate dimethyl succinate. Consistent with our hypothesis, treatment with palmitate or dimethyl succinate transiently increased oxygen consumption and decreased spare reserve capacity in AML but not normal cells. Subsequently these treatments, increased reactive oxygen and induced cell death preferentially in AML cells and 1° samples compare to normal hematopoietic cells. Moreover, this treatment preferentially reduced the clonogenic growth of 1° AML cells over normal cells and reduced the engraftment of 1°AML but not normal cells into immune deficient mice. In summary, compared to normal hematopoietic cells, AML cells have greater mitochondrial mass but respiratory chain activity does not increase proportionately. The lack of proportionate rise in respiratory complex activity results in reduced spare reserve capacity in the respiratory complexes and greater sensitivity to oxidative stress. These data highlight a unique metabolic vulnerability in AML. Disclosures: No relevant conflicts of interest to declare.