Under homeostatic conditions, the main endoergonic processes of neurons are ion-pumping (responsible for about 50% of ATP hydrolysis), biosynthesis of macro-molecules (proteins, lipids, carbohydrates and nucleic acids) and neurotransmitters, intracellular molecular transport and phosphorylation. All these ATP-consuming processes are almost totally impaired in neurons of the ischemic core. Conversely, in the ischemic penumbra, physiologic cellular functions are partially maintained. At this level, numerous strategies capable of reducing energy consumption by neurons and glia have profound neuroprotective effects (Beal 2000). In this regard, several lines of evidence demonstrate that inhibition of poly(ADP-ribosyl)ation in
Fig. 7.1 Proposed role of PARP-1 in ischemic neuronal death. In the ischemic brain, overactivation of neuronal glutamate receptors occurs because of glutamate release triggered by energy failure and peri-infarct depolarization. The ensuing excitotoxic cascade prompts intracellular accumulation of Ca2+ (T[Ca2+]i), oxidative and nitrosative stress eventually leading to massive DNA damage. Recruitment of immune cells in the infarcted area also causes genotoxic stress and activation of transcription factors involved in cell death signaling such as NF-kB and p53. Accumulation of intracellular Ca2+ can also directly activate nuclear PARP-1 activity (Homburg et al. 2000). Overall, these events cause hyperactivation of PARP-1, which consumes excessive NAD and ATP, leading to glycolytic block and mitochondrial dysfunction, leading to cellular necrosis. Impairment of mitochondrial functioning also prompts AIF release and excitotoxic cell death. Hyperactivation of PARP-1 also alters the enzyme-dependent fine-tuning of transcription factor activation, resulting in transcriptional derangements, and abnormalities in gene expression that eventually contribute to neuronal death the ischemic brain tissue preserves energy dynamics. It is well appreciated that massive genotoxic stress occurs in tissues subjected to ischemia/reperfusion or mild ischemia because of formation of reactive oxygen and nitrogen radicals such as superoxide ion (O2-), hydroxyl radical (OH), NO and peroxynitrite (ONOO-) (Lee et al. 2000; Lipton 1999). Profuse DNA damage in turn prompts PARP-1 hyperactivation, ATP consumption and NAD depletion, thereby worsening energy dynamics already compromised by the ischemic insult. Under these stressful conditions, the NAD salvage pathway is activated with re-synthesis of NAD, thanks to the concerted actions of nicotinamide phosphoribosyl transferase (NaPRT) and nicoti-namide mononucleotide adenylyl transferase (NaMNAT). Notably, because both enzymes utilize ATP, DNA damage-dependent PARP-1 hyperactivation ultimately depletes cellular ATP pools and triggers cell death. This pathogenetic interpretation, the so-called "suicide hypothesis", is the oldest interpretation of PAR-dependent neurotoxicity. It originally stems from the pioneering studies of Berger and colleagues on the role of PARP-1 in radiation-induced cell death (Berger 1985).
In light of the remarkable therapeutic efficacy of PARP-1 inhibitors in models of brain ischemia, several researchers consider energy utilization by hyper-poly(ADP-ribosyl)ation as causative in post-ischemic brain damage (Szabo and Dawson 1998). Several reports are in keeping with this interpretation. For instance, in line with the "suicide hypothesis" the PARP-1 inhibitor 3-aminobenzamide (3-ABA) prevents NAD depletion in the ischemic tissue of mice subjected to 2 h MCAo (Endres et al. 1997). Also, NAD and ATP shortage in rat brains subjected to transient focal ischemia is significantly reduced by inhibiting PARP-1 with nicotinamide (Yang et al. 2002). Similarly, the PARP-1 inhibitor FR247304 prompts NAD rescue in a transient ischemia model in the rat (Iwashita et al. 2004). As further evidence that hyperactivation of PARP-1 triggers death of neural cells because of energy failure, energetic substrates such as tricarboxylic acid cycle intermediates prevent death of cultured neurons undergoing massive poly(ADP-ribosyl)ation (Ying et al. 2002), and NAD repletion rescues astrocytes exposed to alkylating agents (Ying et al. 2002). However, as for its relevance to the pathogenesis of ischemic neuronal death, conflicting results have been reported. Plaschke and colleagues (Plaschke et al. 2000) show that the neuroprotective effect of PARP-1 inhibition in a rat model of global ischemia (15 min) is associated with early rescue of NAD contents without a parallel increase in those of ATP. Similarly, NAD concentrations are neither reduced in the hippocampus of rats subjected to sub-lethal transient global ischemia and reperfusion (Nagayama et al. 2000), nor in the brain of mice exposed to 1 h MCAo and different times of reperfusion (Paschen et al. 2000). Finally, the study by Goto and associates (Goto et al. 2002) shows that the brains of PARP KO mice are more resistant to ischemic stroke despite undergoing an energy depletion similar to those of wild type (WT) animals. This report provides a significant challenge to the relevance of the suicide hypothesis to ischemic brain injury and suggest that mechanisms in addition to energy derangement underlie the detrimental role of PARP-1 in ischemic brain injury. The "suicide hypothesis", therefore, might explain the neurotoxic effects of PAR only in conditions of massive DNA rupture and intense PARP-1 activation within the CNS.
Was this article helpful?