The molecular mechanisms that couple glycolysis to cancer drug resistance remain unclear. Here we identify an ATP-binding motif within the NADPH oxidase isoform, NOX4, and show that ATP directly binds and negatively regulates NOX4 activity. We find that NOX4 localizes to the inner mitochondria membrane and that subcellular redistribution of ATP levels from the mitochondria act as an allosteric switch to activate NOX4. We provide evidence that NOX4-derived reactive oxygen species (ROS) inhibits P300/CBP-associated factor (PCAF)-dependent acetylation and lysosomal degradation of the pyruvate kinase-M2 isoform (PKM2). Finally, we show that NOX4 silencing, through PKM2, sensitizes cultured and ex vivo freshly isolated human-renal carcinoma cells to drug-induced cell death in xenograft models and ex vivo cultures. These findings highlight yet unidentified insights into the molecular events driving cancer evasive resistance and suggest modulation of ATP levels together with cytotoxic drugs could overcome drug-resistance in glycolytic cancers.

We evaluated the long-term natural history of renal function after radical cystectomy with urinary diversion and determined factors associated with decreased renal function. We reviewed the records of 1,631 patients who underwent radical cystectomy between 1980 and 2006. The estimated glomerular filtration rate was calculated preoperatively and at various intervals after surgery. A renal function decrease was defined as a greater than 10 ml per minute/1.73 m(2) reduction in the estimated glomerular filtration rate. Multivariate analysis was done to evaluate the association of clinicopathological features, incontinent vs continent diversion type and postoperative complications with decreased renal function. A total of 1,241 patients (76%) underwent incontinent diversion and 390 (24%) underwent continent diversion. Median followup after radical cystectomy in patients alive at last followup was 10.5 years (IQR 7.1, 15.3). The median preoperative estimated glomerular filtration rate was higher in the continent diversion cohort (67 vs 59 ml per minute/1.73 m(2), p <0.0001). This difference was maintained until 7 years postoperatively, after which no difference was noted in renal function by diversion type. By 10 years after radical cystectomy the risk of a renal function decrease was similar for incontinent and continent diversion (71% and 74%, respectively, p = 0.13). On multivariate analysis risk factors associated with decreased renal function included age (HR 1.03, p <0.0001), preoperative estimated glomerular filtration rate (HR 1.05, p <0.0001), chronic hypertension (HR 1.2, p = 0.01), postoperative hydronephrosis (HR 1.2, p = 0.03), pyelonephritis (HR 1.3, p = 0.01) and ureteroenteric stricture (HR 1.6, p <0.0001). Decreased renal function is noted in most patients during long-term followup after radical cystectomy. Postoperative hydronephrosis, pyelonephritis and ureteroenteric stricture represent potentially modifiable factors associated with a decrease. Choice of urinary diversion was not independently associated with decreased renal function.