Recent findings allow us to hypothesize an additional mechanism through which deregulated PAR formation exerts a detrimental role in the CNS. Several lines of evidence support the so-called "signalling hypothesis" (Chiarugi 2002b). For instance, PARP-1 hyperactivation is a powerful trigger of mitochondrial release of apoptosis inducing factor (AIF) in cultured neurons undergoing excitotoxicity (Hong et al. 2004; Yu et al. 2002) as well as in astrocytes (Alano et al. 2004) and ischemic brain (Komjati et al. 2004). These findings, together with knowledge that AIF may contribute to post-ischemic neurodegeneration (Cao et al. 2003; Culmsee et al. 2005; Plesnila et al. 2004; Zhu et al. 2003), suggest that PAR formation is a pivotal event in neuronal cell death that follows the ischemic insult. Recently, Andrabi et al. reported that PAR polymer is endowed with intrinsic cytotoxicity and that its toxic effect is abolished by pre-treatment with the PAR-degrading enzymes phosphodiesterase (PD1) or poly (ADP-ribose) glycohydrolase (PARG). Moreover, interfering with PAR polymer signalling by means of neutralizing PAR antibodies or PARG overexpression reduces PARP-1-dependent NMDA excitotoxicity and N-methyl-N'-nitro-N-nitrosoguanidine (MNNG)-induced cell death (Andrabi et al. 2006; Wang et al. 2003) In keeping with this, transgenic mice overexpressing PARG have markedly reduced infarct volumes after 2 h of MCAo. Consistent with the notion that PAR chains are toxic, experimental evidence indicates that (1) siRNA knockdown of PARG leads to enhanced MNNG toxicity, (2) PARG +/- cortical cultures are more sensitive to NMDA excitotoxicity and (3) PARG +/- mice have larger infarct volumes 24 h after 1.5 h of MCAo. These results suggest that PAR is a key signal in the nucleus-mitochondria cross-talk during PARP-1-dependent neuronal death, substantiating the relevance of the signaling hypothesis to PAR neurotoxicity (Andrabi et al. 2006a; Wang et al. 2003). To date, molecular mechanisms responsible for PAR-dependent AIF release in neurons are unknown. Of note, recent findings support an important role of PARP-1 in the release of AIF from mitochondria after ischemic brain injury (Culmsee et al. 2005). Inhibition of PARP-1 prevented AIF translocation from mitochondria to the nucleus after oxygen-glucose deprivation in vitro and in focal cerebral ischemia in vivo. It has been suggested that mitochondrial NAD depletion may represent a link between PARP-1 overactivation and the release of AIF from mitochondria (Yu et al. 2002). Based on these data, these authors suggest that DNA damage and PARP-1 activation subsequent to cerebral ischemia, lead to a decrease of mitochondrial NAD+ contents that induces Bid-mediated mitochondrial membrane pore formation and hence the AIF release from mitochondria to nucleus in which it initiates nuclear condensation (Susin et al. 1999). More recently, a report provides novel insights into the molecular mechanism involved in mitochondrial AIF release upon MNNG-dependent PARP-1 activation (Moubarak et al. 2007). In light of these new findings, calpain appears to link PARP-1 activation to Bax activation and AIF release during MNNG-induced necrosis. Recently, both Bax and AIF have been identified as targets of calpains. Indeed, generation of a Bax fragment is an early event in the induction of apoptosis via calpains (Altznauer et al. 2004; Gao and Dou 2000; Polster et al. 2005) and AIF becomes a soluble and death-promoting protein after its cleavage by calpains (Polster et al. 2005). These results indicate that calpains control mitochondrial AIF release during PARP-1 activation.
Finally, ADP-ribose, the monomeric constituent of PAR and product of PARG, could also be involved in mediating mitochondrial disfunction consequent to PARP-1 hyperactivation (Dumitriu et al. 2004). Following PARP-1 activation, the increasing levels of poly(ADP-ribose) recruit PARG to the nucleus. PARG hydrolyses the protein-bound poly(ADP-ribose), thereby generating free oligo- and monomers of ADP-ribose. ADP-ribose potently inhibits the activity of the ATP-binding cassette (ABC) transporters, involved in the transport of multiple substrates across cellular membranes, among them mitochondrial membranes. Therefore, considering the homology between ADP-ribose, ADP and ATP, it is tempting to speculate that ADP-ribose might compete with these metabolites for its binding site on ATP-dependent membrane transporters (Dumitriu et al. 2004).
Was this article helpful?