Role of poly(ADP-ribosyl)ation in the killing of chronic lymphocytic leukemia cells by purine analogues.

Although the nucleoside analogues fludarabine and chlorodeoxyadenosine have become important therapeutic agents in chronic lymphocytic leukemia (CLL), their effectiveness is limited by drug resistance. Because such resistance is likely to result from impaired drug-induced apoptosis, it is clearly important to understand the mechanisms involved in this process. Whereas p53 can contribute to the nucleoside-induced killing of CLL cells, recent work from this laboratory and elsewhere has shown that such killing can also occur by p53-independent mechanisms. Because poly(ADP-ribose) polymerase (PARP)-mediated NAD+/ATP depletion has been implicated in the nucleoside-induced killing of normal resting lymphocytes, we postulated that this mechanism might account for the p53-independent component of nucleoside cytotoxicity in CLL. To address this question, we used 3-aminobenzamide (3AB) at a concentration (200 microM) known to produce selective inhibition of poly(ADP-ribosyl)ation in intact cells and examined nucleoside-induced killing using a number of different end points (cell membrane disruption, cell shrinkage, mitochondrial depolarization, exposure of phosphatidyl serine, morphological changes, DNA fragmentation, and PARP-1 cleavage). In 27 of the 30 cases of CLL examined, 3AB delayed nucleoside-induced cell membrane disruption without inhibiting other manifestations of cytotoxicity. This indicates that PARP activity, rather than contributing to the induction of cell killing, was accelerating cell membrane disruption during the late stages of apoptosis. This novel observation has important implications for previous studies of PARP-mediated cytotoxicity. However, in cells from one CLL patient, 3AB inhibited all manifestations of nucleoside cytotoxicity; this was the only case in the study known to have a p53 gene defect affecting both alleles. This indicates that PARP activity can occasionally be central to nucleoside-induced killing and that such PARP-mediated killing is p53 independent.