The induction of proline oxidase accompanying p53-dependent induction of apoptosis suggests that the proline metabolic pathway plays a role in programmed cell death. We are studying this metabolic pathway at the level of : a) proline oxidase - its induction and its role in apoptosis; b) pyrroline 5-carboxylate (P5C) and P5C reductase in cell signaling; and c) the function of prolidase in cell physiology. a) Proline oxidase - its induction and its role in apoptosis. We have focused on proline oxidase, the enzyme which catalyzes the conversion of proline to P5C. The enzyme is bound to mitochondrial inner membranes and donates electrons to complex II. Recently others have shown that proline oxidase (POX) is one of only 14 genes (from 7202 genes monitored by SAGE) highly induced accompanying p53-dependent induction of apoptosis in colorectal tumor cells. To investigate the role of POX, we first used RT-PCR to examine several cell types and the stimuli which will result in POX induction. In human colorectal cancer (LoVo) cells, mouse colonic epithelial cells transformed by temperature sensitive SV40 Young Adult Mouse Colon epithelia (YAMC), as well as a variety of carcinoma cell lines, we found that serum starvation as well as cytotoxic drugs induced proline oxidase in a p53-dependent fashion. We proposed that proline oxidase may contribute to the metabolic events accompanying apoptosis by generating reactive oxygen species (ROS). This possibility was examined in LoVo and YAMC cells by using a fluorescent probe and laser cytometry to monitor ROS. Importantly, we included DLD1-POX, a p53-negative colon tumor cell line stably transfected with POX. Interestingly, the addition of proline markedly increased ROS generation in a concentration-dependent manner but only when POX was expressed. Glutamate, on the other hand, was without effect. Importantly, the proline-POX effect is sufficient to induce mitochondria-mediated apoptosis. Using DLD-tet-POX cells which are p53 nulls, we showed that induction of POX results in decreased cell growth and increase in SubG1 fractions on cell cytometry, nuclear fragmentation on TUNEL assay and DNA laddering and release of cytochrome c from mitochondrial and caspase 9 activation. We are currently using antisense POX in DLD-p53 cells to ascertain whether POX is necessary for apoptosis. b) Pyrroline 5-carboxylate (P5C) as a signaling and regulatory molecule. P5C, the product of proline oxidase, ornithine aminotransferase and glutamate synthase, has been shown to have regulatory activities. Recent work suggests that P5C reductase (P5CR) may play a role in receptor-mediated regulation. The two known isozymes of P5CR may serve distinct functions. P5CR1 is a NADH-dependent enzyme regulated to produce proline, whereas P5CR2 is a NADPH-dependent enzyme coupled with redox-related regulation. In lymphocytes activated by lectins, it appears that P5CR2 is induced. We are defining the interaction of P5CR2 with membrane receptors. c) Inhibition of prolidase by nickel. Prolidase hydrolyzes imidodipeptides with carboxyl terminus proline and is a metalloenzyme. Nickel which can cause pulmonary fibrosis and cancer in humans, is a potent inhibitor of prolidase activity. Kinetic studies show that nickel is a competitive inhibitor of the interaction with manganese, the most potent activator of catalytic activity. The inhibitory effect can be demonstrated not only in cell-free systems, but also in intact CHO-K1 cells which are proline auxotrophs but which also can grow on GLY-PRO. Nickel markedly inhibits growth of these cells on GLY-PRO but not on PRO. We are using these cells to perturb the processing of imidodipeptides and other proline containing peptides to investigate their potential role in cell signaling.
