De-Maw Chuang, PhD - US grants

Molecular mechanisms underlying cell death are a major focus of current biomedical research as aberrant cell death is involved in the pathogenesis of a vast number of diseases including the CNS disorders. Apoptosis, also referred to as programmed cell death, is a process in which a cell dies by activation of an intrinsic genetic program. Apoptosis plays an important role in the development and aging processes in the central nervous system (CNS). Abnormality in apoptosis has been linked to pathogenesis of neurodegenerative diseases and implicated in neuropsychiatric disorders. We have used primary cultures of rat CNS neurons and neurally related cell lines as a model to study molecular mechanisms underlying neuronal apoptosis. We found that glyceraldehyde-3-phospahate dehydrogenase (GAPDH), a glycolytic enzyme, is over- expressed during apoptosis induced by aging of rat cerebellar granule cells (CGCs) in culture. Antisense oligonucleotides to GAPDH block GAPDH over-expression and effectively delay age-induced apoptosis of these cerebellar neurons. These results provide the first evidence for a role of GAPDH in neuronal apoptosis. By the same criteria, we found that GAPDH is involved in age-induced apoptosis of rat cerebral cortical neurons and apoptosis of CGC induced by extracellular potassium deprivation and exposure to cytosine arabinoside (AraC). We found that the AraC-induced apoptosis is robustly protected by the neurotrophins brain-derived neurotrophic factor (BDNF) and NT 4/ 5, but not NT-3. The apoptosis is also associated with an increased expression of two death genes, p53 and Bax. An antisense oligonucleotide to p53 reduces GAPDH induction and protects against AraC-induced apoptosis, suggesting that GAPDH over-expression depends on p53 expression. Moreover, tranfection of the p53 gene into PC12 cells induces GAPDH over-expression and causes ultimate cell death, suggesting that GAPDH is a novel target gene regulated by p53. AraC-induced GAPDH over-expression is predominantly accumulated in the nucleus assessed by subcellular fractionation study and electromicroscopic immunohistochemistry. Translocation of GAPDH to the nucleus occurs in parallel with a loss of GAPDH glycolytic and uracil glycosylase activities, suggesting an alteration in the structure and function of nuclear GAPDH. We also demonstrated that at least six GAPDH isoforms can be detected in the nucleus of cerebellar granule cells. These nuclear isoforms differ in their abundance and are translocated to the nucleus according to a distinct time-course following AraC treatment. Evidence is also available that nuclear GAPDH accumulation precedes activation of caspase-3 and cleavage of its nuclear substrate lamin B1. In CGCs, we studied the role of GAPDH in the excitotoxicity induced by SYM 2081 (2S,4R,-4-methylglutamate), an inhibitor of excitatory amino acid transporter and an agonist of low affinity kainate receptors. We found that SYM-induced excitotoxicity is N-methyl-D-aspartate (NMDA) receptor-mediated and is associated with nuclear translocation of GAPDH. Pretreatment of neurons with valproate, a mood stabilizing drug, suppresses GAPDH nuclear accumulation with a concomitant neuroprotective effect. Chromatin immunoprecipitation revealed that GAPDH is co-present with acetylated histone H3 and that valproate treatment caused a time-dependent decrease in levels of nuclear GAPDH with a concomitant increase in acetylated histone in the chromatin immunoprecipitation complex. Our results strongly suggest that valproate protects neurons from excitotoxicity through inhibition of histone deacetylase and that this neuroprotection involves suppression of excitotoxicity-induced accumulation of GAPDH in the nucleus. We have studied GAPDH abnormalities in a transgenic mouse model of Huntington's disease in which the disease protein (huntingtin) gene expresses expanded (89) CAG repeats. We found that GAPDH is overexpressed in neurons of discrete brain areas such as the hippocampal formation, caudate-putamen and globus pallidus, compared with wild type control. Confocal microscopic analysis revealed a predominant increase of GAPDH in the nucleus. Thus, mutation of huntingtin is associated with GAPDH overexpression and nuclear translocation in discrete populations of neurons in selective brain areas. Our results suggest that over-expression of GAPDH in CNS neurons could induce neuronal apoptosis and ultimate neurodegeneration. Moreover, the ability of GAPDH antisense oligonucleotides to rescue neurons from undergoing apoptotic cell death also suggests that such oligonucleotides may be useful for treating neurodegenerative diseases. In an attempt to further elucidate molecular mechanisms underlying apoptosis, we have used CGCs treated with AIDS-related neurotoxins, such as 3-OH- kynurenine (3-HK) and quinolinic acid. We found that 3-HK induces apoptosis in the range of 50-1000 micro-M, while quinolinic acid is ineffective. The 3-HK neurotoxicity is potentiated by superoxide dismutase (SOD) and a cell-permeable SOD mimetic, MnTBAP. Both 3-HK-induced and SOD-potentiated neurotoxicities are blocked by catalase, suggestive of hydrogen peroxide formation. We also found that 3-HK induces apoptotic death of PC12 pheochromocytoma cells and hypothalamic GT1-7 cells. In both cell types, 3-HK-induced apoptosis is robustly protected by dantrolene, a drug that inhibits Ca-2+ efflux from the endoplasmic reticulum to the mitochondria. Moreover, overexpression of Bcl-2, an anti-apoptotic gene product, in GT1-7 cells arrests 3-HK-induced neurotoxicity. Thus, pharmacological manipulation with dantrolene and/or gene therapy with Bcl-2 are potentially useful for the treatment of 3-HK-related neurodegeneration and other forms of neurodegenerative diseases. Glutamate excitotoxicity has been implicated in a variety of neurodegenerative diseases. Using CGCs as a model system, we have studied mechanisms underlying glutamate-induced apoptosis in these cell types. We found that glutamate excitotoxicity is associated with increased expression of apoptotic proteins, p53 and Bax, but with decreased expression of the cytoprotective protein, Bcl-2. The excitotoxicity is also concurrent with inhibition of the cell survival factors, Akt-1 and p-CREB (cyclic AMP-responsive element binding protein) due to activation of protein phosphatase PP2A and PP1, respectively. Additionally, glutamate-induced apoptosis is preceded by activation of p38 kinase and JNK (c-Jun N-terminal kinase), and suppression of these kinase activities results in neuroprotection. Activation of these kinases results in phosphorylation and activation of p53. Moreover, glutamate excitotoxicity mediated through NMDA receptors involves nuclear accumulation of GAPDH. In an attempt to elucidate the role of numerous proteins in apoptosis and neuroprotection, we have succeeded in developing high transfection efficiency technology of siRNA in primary cultures of neurons. Using rat cortical neurons transfected with siRNA for glycogen synthase kinase-3 (GSK-3), we have demonstrated that glutamate-induced apoptotic death involves activation of GSK-3alpha and beta isoforms and that silencing either isoform with its specific siRNA provides full protection against excitotoxicity. This conclusion is corroborated by using inhibitors of GSK-3 such as lithium, SB216763, SB415280, inhibitor I and inhibitor VII. Additionally, our experiments using siRNA silencing of GSK-3alpha and GSK-3beta provide evidence that these two isoforms can have distinct roles in the regulation of transcription factors such as CREB, NF-kB, EGR-1 and Smad3/4.

Lithium was introduced into psychiatry 50 years ago and remains to be the most commonly used drug for the treatment of manic depressive illness. The precise mechanisms underlying its clinical efficacy remain to be defined. We have previously shown that long-term treatment of cerebellar granule cells (CGCs) with relatively low doses of lithium increases the mRNA level of the immediate early gene product, c-Fos, and up-regulates the expression and function of m3-muscarinic receptors in these neurons. Since AP-1 and cAMP-responsive element (CRE) are key elements of transcriptional regulation for a variety of proteins with neurophysiological importance, we have investigated the effects of lithium on gene transcription in cultured CGCs and in rats by measuring transcription factor binding to AP-1 and CRE sites. We found that treatment of CGCs with lithium chloride induces a time- and concentration-dependent increase in AP-1 and CRE binding activities. Chronic dietary treatment of rats with lithium carbonate for 4 weeks also significantly increases AP-1 and CRE binding activity in the frontal cortex, hippocampus, amygdala and cerebellum. p-CREB, Jun D and Fos family proteins are present in the AP-1 binding sites in the rat brain and in CGCs. In a parallel study, we explored the neuroprotective effects of lithium against excitotoxicity elicited by glutamate, a major excitatory amino acid neurotransmitter involved in the synaptic plasticity and pathogenesis of neurodegenerative and neuropsychiatric disorders. We found that long-term exposure to lithium chloride dramatically protects cultured rat CGCs, cerebral cortical and hippocampal neurons against glutamate-induced excitotoxicity which involves apoptosis mediated by N-methyl-D-aspartate (NMDA) receptors. This neuroprotection occurs at therapeutically relevant concentrations (0.5-5.0 mM) and requires treatment for 6-7 days for complete protection to occur, while a 24-hr treatment is ineffective. The protection is specific for glutamate-induced excitotoxicity and can be attributed to inhibition of NMDA receptor-mediated calcium influx. The long-term effects of lithium are not due to down-regulation of NMDA receptor sites, nor are they related to its known ability to block inositol monophosphatase activity. Our results suggest that modulation of glutamate receptor hyperactivity represents, at least in part, the molecular mechanisms by which lithium alters brain function and exerts its clinical efficacy in the treatment for manic depressive illness. These novel actions of lithium also suggest that excessive glutamatergic neurotransmission may be the pathogenic mechanism underlying bipolar illness. In a related study, we found that lithium also robustly protects against carbamazepine and phenytoin-induced apoptosis and modestly inhibits age-induced death of CGCs. However, unlike the protection against glutamate excitotoxicity, this effect of lithium on anticonvulsant-induced neurotoxicity does not require long-term treatment and occurs at relatively high concentrations (less than or equale to 5 mM). This neuroprotective effect of lithium is inhibited by a phosphatidylinositol 3-kinase (PI 3-K) inhibitor, LY294002. The inhibition of carbamazepine neurotoxicity by lithium may have clinical implications, as combined treatment with lithium and carbamazepine is known to potentiate the effect of either drug alone in the treatment of manic depressive illness. We have studied the role of expression of genes involved in pro- apoptosis and cytoprotection. In CGCs, treatment with neuroprotective concentrations of lithium induces a time- dependent increase in the levels of mRNA and protein of the cytoprotective gene Bcl-2. Conversely, the levels of mRNA and protein of the pro-apoptotic genes, Bax and p53 are decreased by lithium. The ratio of Bcl-2/Bax is increased by about 6-fold after treatment with 3 mM LiCl for 7 days. In contrast, glutamate treatment induces a rapid increase in the levels of Bax and p53 mRNA and protein. Pretreatment with lithium suppresses the glutamate-induced increase in Bax and p53 and maintains Bcl-2 at an elevated level. Lithium also blocks glutamate-elicited cytochrome c release from mitochondria and the cleavage of the caspase-3 substrate, lamin B. In a parallel study, we found that glutamate induces a rapid, reversible decrease in the activity of a cell survival factor, Akt, through enhanced dephosphorylation due to activation of protein phosphatase(s). In contrast, lithium activates the PI 3-K/Akt signalling pathway and enhances the phosphorylation of glycogen synthase kinase 3-beta. Pretreatment with lithium facilitates the recovery of glutamate-induced loss of Akt activity. Taken together, our results suggest that lithium-induced up-regulation of cytoprotective gene products and down-regulation of pro- apoptotic gene products play a prominent role in its neuroprotective actions. We have expanded from our in vitro cell culture studies by using animal models of cerebral ischemia and Huntingtons disease. We found that subcutaneous injection of rats with LiCl for 16 days reduces the size of ischemic brain infarct volume by more than 50% in rats subjected to occlusion of the left middle cerebral artery. The focal ischemia-induced neurological deficits such as abnormal posture and hemiplegia, measured at 24 hr after middle artery occlusion, are significantly reduced by chronic lithium pretreatment. In the Huntingtons disease animal model, we injected quinolinic acid, a partial agonist of the NMDA receptor, into the left side of rat striatum. This treatment results in about 70% lesion of the striatum which involves the loss of GABAnergic neurons expressing dopamine D-1 receptors. This quinolinic acid-induced lesion requires activation of the transcription factor NF-kB and induction of p53 and c-Myc. Our results show that chronic lithium treatment or one day pretreatment decreases the size of striatal lesion by 40-50%. Moreover, lithium neuroprotection effects are associated with over-expression of Bcl-2 in affected areas and other brain structures. Thus, our in vitro and in vivo studies raise the possibility that lithium, in addition to its use for bipolar depressive illness, may have expanded use for the treatment of neurodegenerative diseases, particularly those linked to excitotoxicity.