- Top of page
- Materials and methods
Mild metabolic stress may increase resistance of neurons in the brain to subsequent, more severe insults, as demonstrated by the ability of ischemic pre-conditioning and dietary restriction to protect neurons in experimental models of stroke- and age-related neurodegenerative disorders. In the present study we employed iodoacetic acid (IAA), an inhibitor of glyceraldehyde-3-phosphate dehydrogenase, to test the hypothesis that inhibition of glycolysis can protect neurons. Pre-treatment of cultured hippocampal neurons with IAA can protect them against cell death induced by glutamate, iron and trophic factor withdrawal. Surprisingly, protection occurred with concentrations of IAA (2–200 nm) much lower than those required to inhibit glycolysis. Pre-treatment with IAA results in suppression of oxyradical production and stabilization of mitochondrial function in neurons after exposure to oxidative insults. Levels of the stress heat-shock proteins HSP70 and HSP90, and of the anti-apoptotic protein Bcl-2, were increased in neurons exposed to IAA. Our data demonstrate that IAA can stimulate cytoprotective mechanisms within neurons, and suggest the possible use of IAA and related compounds in the prevention and/or treatment of neurodegenerative conditions.
Exposure of neurons to moderate levels of stress can increase their resistance to more severe and otherwise lethal insults in experimental cell culture and animal models relevant to the pathogenesis of ischemic stroke and other neurodegenerative conditions. For example, exposure of cultured neurons to elevated temperatures (heat-shock) increases their resistance to excitotoxicity (Lowenstein et al. 1991), as does prior exposure to a low subtoxic level of N-methyl-d-aspartate (Marini and Paul 1992). In vivo, a mild ‘pre-conditioning’ ischemia can increase resistance of hippocampal and cortical neurons to a subsequent severe ischemic insult (Liu et al. 1993; Amin et al. 1995; Matsushima and Hakim 1995; Kitagawa et al. 1997), and mild impairment of oxidative phosphorylation in cell culture and in vivo can also elicit a cytoprotective response (Riepe et al. 1997). In addition to aerobic metabolism, neurons can also produce energy anaerobically via the process of glycolysis in which glucose is phosphorylated and then metabolized to 3-carbon sugars that undergo oxidation and reactions of phosphate bonds to generate ATP. Under basal conditions, glycolysis is thought to make only a minor contribution to the total ATP production of neurons (Lutz 1992). Under hypoxic conditions, anaerobic glycolysis provides the adult brain with a limited amount of energy and time to maintain ion homoeostasis and other essential processes before several events occur that lead to brain cell damage and death. It is not known whether inhibition of glycolysis can induce a cytoprotective response in neurons. If this were the case, then chemicals that inhibit glycolysis may have potential clinical applications.
Biochemical cascades that result in cell death, particularly those involved in a form of programmed cell death called apoptosis, occur not only in the cytoplasm, but also in the mitochondria and endoplasmic reticulum (Mattson 2000). For example, neuronal apoptosis induced by excitotoxic and oxidative insults involves changes in mitochondrial membrane permeability (Green and Reed 1998), release of calcium from the endoplasmic reticulum (Guo et al. 1998, 1999), and activation of proteolytic cascades involving one or more caspases (Chan and Mattson 1999). Two major families of proteins can prevent apoptosis, the Bcl-2 family (Sadoul 1998) and the inhibitor of apoptosis protein (IAP) family (Robertson et al. 2000). Bcl-2 may prevent cell death by stabilizing mitochondrial membranes and suppressing oxyradical damage, while IAPs can directly inhibit caspases. The mechanisms whereby moderate stress increases neuronal resistance to injury are beginning to be identified. One class of proteins involved in the stress response are molecular chaperones that include heat-shock proteins and glucose-regulated proteins (Koroshetz and Bonventre 1994; Yu et al. 1999). These proteins act by binding to dysfunctional and misfolded proteins, which often targets them for proteolytic degradation. Chaperone proteins are localized in specific subcellular compartments. For example, heat-shock protein 70 (HSP70) is present in the cytosol, HSP75 is primarily in the mitochondria and glucose-regulated protein 78 (GRP78) is located in the endoplasmic reticulum.
Because inhibition of glycolysis might be expected to induce a mild stress response in neurons without compromising their viability or function, we sought to determine whether inhibition of glycolysis might protect neurons against death. To this end, we employed iodoacetic acid (IAA), an inhibitor of glyceraldehyde-3-phosphate dehydrogenase (Sabri and Ochs 1971). Our data demonstrate strong neuroprotective effects of IAA against excitotoxic and oxidative insults, and trophic factor withdrawal. The protective effect occurs with concentrations of IAA lower than those which inhibit ATP production, and involves increased production of HSP70, HSP90 and Bcl-2.
- Top of page
- Materials and methods
The results of the present study demonstrate that IAA can exert a strong cytoprotective action in cultured hippocampal neurons under conditions relevant to the pathogenesis of several different neurodegenerative disorders. Overactivation of glutamate receptors and oxidative stress are believed to contribute to the death of neurons in stroke (Dirnagl et al. 1999), traumatic brain injuries (Obrenovitch and Urenjak 1997), Alzheimer's disease (Mattson 1998), Parkinson's disease (Beal 1998) and amyotrophic lateral sclerosis (Rothstein 1996). IAA pre-treatment protected hippocampal neurons against glutamate toxicity and death induced by iron, a metal that induces hydroxyl radical production and membrane lipid peroxidation (Mark et al. 1997). IAA was also effective in protecting hippocampal neurons against death induced by trophic factor deprivation, a model of developmental programmed cell death (Chan et al. 1999). Previous studies using an identical hippocampal cell culture system showed that the predominant mode of neuronal death after trophic factor withdrawal, or exposure to glutamate or iron, is apoptosis (Mattson et al. 1997; Chan et al. 1999; Glazner et al. 2000). The ability of IAA to protect neurons against these different insults therefore suggests that iodoacetate can prevent neuronal apoptosis. Treatment with IAA resulted in a stabilization of mitochondrial function and suppression of oxyradical production in neurons exposed to the oxidative and apoptotic insults. Previous studies have shown that neuronal deaths induced by glutamate, trophic factor deprivation and iron each involve oxyradical production and mitochondrial dysfunction (Mattson et al. 1993, 1995, 1997; Greenlund et al. 1995; Schinder et al. 1996; White and Reynolds 1996; Keller et al. 1998; Chan et al. 1999; Putcha et al. 1999). It therefore seems likely that IAA acts on one or more cellular pathways that stabilize mitochondrial function and suppress oxyradical production. In contrast to its ability to protect neurons against death induced by glutamate and iron, IAA was ineffective in protecting neurons against death induced by 4-hydroxynonenal, a toxic aldehyde produced when membrane lipids are peroxidized (Kruman et al. 1997; Mark et al. 1997). This suggests that the neuroprotective effect of IAA may act at a step(s) prior to membrane lipid peroxidation. Consistent with this possibility, we found that insult-induced increases in dihydrorhodamine fluorescence, a measure of hydroxyl radical and peroxynitrite levels, were suppressed in neurons treated with IAA.
Our data suggest that the mechanism whereby IAA protects neurons involves up-regulation of certain protein chaperones and the anti-apoptotic protein Bcl-2. Levels of HSP70, HSP90 and Bcl-2 were significantly increased within 6–12 h of exposure to IAA, and levels of these proteins remained elevated through at least 24 h of exposure to IAA. HSP70 and HSP90 are protein chaperones that have been shown to prevent apoptotic death in a variety of cell types (Beere and Green 2001). These proteins may bind to proteins either upstream (e.g. Akt kinase and Bcl-2 family members) or downstream (e.g. caspases and Apaf-1) of mitochondrial alterations involved in cell death cascades (Sato et al. 2000). Previous studies have linked production of stress proteins including HSP70 and GRP78 to preconditioning neuroprotective effects of heat-shock and mild ischemia (Lowenstein et al. 1991; Liu et al. 1993; Yu and Mattson 1999). Although the role of HSP90 in neuroprotection has not been established, studies of ischemic preconditioning in the heart suggest a cytoprotective role for HSP90 (Nayeem et al. 1997).
Bcl-2 family members are thought to exert their anti- or pro-apoptotic actions at the level of the mitochondrial membrane (Kroemer 1999). Bcl-2 can stabilize mitochondrial membrane potential and suppress oxyradical production in several different paradigms of neuronal apoptosis (Borner et al. 1994; Ellerby et al. 1996; Guo et al. 1997; Kruman et al. 1997). Several different environmental signals have been shown to induce production of Bcl-2 in neurons, including neurotrophic factors (Allsopp et al. 1995), cytokines (Tamatani et al. 1999), cell adhesion (Gary and Mattson 2001) and exposure to insults such as ischemia (Zhu et al. 1999b) and seizures (Henshall et al. 2000). In addition, it was recently reported that subtoxic levels of the succinate dehydrogenase inhibitor 3-nitropropionic acid can induce an increase in the levels of Bcl-2 in neurons in the rat brain (Brambrink et al. 2000). The increased level of Bcl-2 in hippocampal neurons exposed to IAA suggests that an adaptive anti-apoptotic response occurs in neurons. Previous studies have shown that Bcl-2 can stabilize mitochondrial function in neurons subjected to excitotoxic and oxidative insults (Bruce-Keller et al. 1998), suggesting an important role for Bcl-2 in the neuroprotective effects of IAA demonstrated in the present study. In the hippocampal culture system employed in the present study, the mode of cell death induced by glutamate, trophic factor withdrawal and iron is predominantely apoptosis (Kruman et al. 1997; Chan et al. 1999; Glazner et al. 2000). It therefore appears to be the case that IAA pre-treatment protects neurons against apoptosis. The possible effects of IAA in modulating necrotic neuronal death remains to be determined.
Interestingly, we found that IAA increases HSP70 levels, but does not increase GRP78 levels. This contrasts with results obtained in studies in which cultured hippocampal neurons were exposed to the non-metabolizable glucose analog 2-deoxy-d-glucose, which induced an increase in levels of both HSP70 and GRP78 (Lee et al. 1999). Therefore, the cellular response to limiting glucose availability by inhibition of glycolysis with 2-deoxy-d-glucose can be distinguished from that induced by IAA based upon the profile of stress proteins produced. GRP78 is localized in endoplasmic reticulum and is responsive to stress in that organelle. The lack of effect of IAA on GRP78 levels therefore suggests that IAA may not induce an endoplasmic reticulum stress response. Another distinction between the neuroprotective mechanisms of IAA and 2-deoxy-d-glucose is that, whereas IAA induces an increase in Bcl-2 levels (present study), 2-deoxy-d-glucose may not (Lee et al. 1999). It will be of considerable interest to identify the specific signaling cascades and transcription factors involved in neuroprotective responses to different types of metabolic inhibitors.
Finally, the implications of the present findings for physiological regulation of neuronal survival by changes in energy metabolism, and for neuroprotective strategies for neurodegenerative disorders, should be considered. In addition to the neuroprotective effect of preconditioning ischemia, recent studies have demonstrated quite striking neuroprotective effects of dietary restriction in rodents. Dietary restriction induces cytoprotective mechanisms in neurons in several different brain regions in experimental models relevant to the pathogenesis of Alzheimer's disease (Bruce-Keller et al. 1999; Zhu et al. 1999a), Parkinson's disease (Duan and Mattson 1999) and stroke (Yu and Mattson 1999). The latter studies, and more recent findings (Lee et al. 2000; Duan et al. 2001), suggest that dietary restriction protects neurons by inducing the up-regulation of HSP70 and brain-derived neurotrophic factor. By modulating the expression of heat-shock proteins, Bcl-2 family members and neurotrophic factors changes in brain energy metabolism may affect neuronal plasticity and survival. Interestingly, administration of 2-deoxy-d-glucose to adult rats and mice can protect neurons against excitotoxic, oxidative and ischemic injury (Duan and Mattson 1999; Lee et al. 1999; Yu and Mattson 1999). In light of the present findings, future studies should therefore be peformed to determine whether IAA and related compounds can exert neuroprotective effects in vivo. Agents that modify cellular energy metabolism may prove valuable in the prevention and treatment of various neurodegenerative conditions.