FGF-21, a novel metabolic regulator, has a robust neuroprotective role and is markedly elevated in neurons by mood stabilizers.

Fibroblast growth factor-21 (FGF-21) is a new member of the FGF super-family and an important endogenous regulator of glucose and lipid metabolism. It has been proposed as a therapeutic target for diabetes and obesity. Its function in the central nervous system (CNS) remains unknown. Previous studies from our laboratory demonstrated that aging primary neurons are more vulnerable to glutamate-induced excitotoxicity, and that co-treatment with the mood stabilizers lithium and valproic acid (VPA) induces synergistic neuroprotective effects. This study sought to identify molecule(s) involved in these synergistic effects. We found that FGF-21 mRNA was selectively and markedly elevated by co-treatment with lithium and VPA in primary rat brain neurons. FGF-21 protein levels were also robustly increased in neuronal lysates and culture medium following lithium-VPA co-treatment. Combining glycogen synthase kinase-3 (GSK-3) inhibitors with VPA or histone deacetylase (HDAC) inhibitors with lithium synergistically increased FGF-21 mRNA levels, supporting that synergistic effects of lithium and VPA are mediated via GSK-3 and HDAC inhibition, respectively. Exogenous FGF-21 protein completely protected aging neurons from glutamate challenge. This neuroprotection was associated with enhanced Akt-1 activation and GSK-3 inhibition. Lithium-VPA co-treatment markedly prolonged lithium-induced Akt-1 activation and augmented GSK-3 inhibition. Akt-1 knockdown markedly decreased FGF-21 mRNA levels and reduced the neuroprotection induced by FGF-21 or lithium-VPA co-treatment. In addition, FGF-21 knockdown reduced lithium-VPA co-treatment-induced Akt-1 activation and neuroprotection against excitotoxicity. Together, our novel results suggest that FGF-21 is a key mediator of the effects of these mood stabilizers and a potential new therapeutic target for CNS disorders.

Protracted lithium treatment protects against the ER stress elicited by thapsigargin in rat PC12 cells: roles of intracellular calcium, GRP78 and Bcl-2.

We investigated the cytoprotective effects of lithium, the mood-stabilizer, on thapsigargin-induced stress on the endoplasmic reticulum (ER) in rat PC12 cells. Protracted lithium pretreatment of PC12 cells elicited cytoprotection against thapsigargin-induced cytotoxicity. Lithium protection was concurrent with inhibition of thapsigargin-induced intracellular calcium increase and with elevated expression of the molecular chaperone GRP78. Moreover, lithium pretreatment upregulated the antiapoptotic protein Bcl-2, and blocked Bcl-2 downregulation elicited by thapsigargin. Prior to the induction of GRP78, lithium treatment alone increased the expression of c-Fos whose induction by ER stress is necessary for GRP78 induction. Curcumin, an inhibitor of transcription factor AP-1, blocked lithium cytoprotection against thapsigargin cytotoxicity. Thus, the induction of GRP78 and Bcl-2, and activation of AP-1 likely contribute to lithium-induced protection against cytotoxicity resulting from ER stress. Additionally, thapsigargin-induced cytotoxicity was suppressed by pretreatment with another mood-stabilizer, valproate, indicating that cytoprotection against ER stress is a common action of mood-stabilizing drugs.

Neurotrophins protect against cytosine arabinoside-induced apoptosis of immature rat cerebellar neurons.

Neurotrophin-induced neuroprotection against apoptosis was investigated using immature cultured cerebellar granule cells (CGC) from newborn rat pups. Apoptotic cell death induced by treatment with cytosine arabinoside (AraC) was confirmed by DNA fragmentation and quantified by cell survival assays. AraC was most effective in inducing apoptosis when added to CGC on the day of culture preparation, while less or no effect was observed when added at 24 or 48h after plating, respectively. Pretreatment of CGC cultures for 24h with brain-derived neurotrophic factor (BDNF) or neurotrophin-4 (NT-4), but not neurotrophin-3 (NT-3), robustly protected against AraC neurotoxicity. K252a, an inhibitor of the tropomyosin-related kinase (Trk) tyrosine kinase receptor family which showed no toxicity by itself, blocked BDNF protection of AraC-induced apoptosis in a concentration-dependent manner. Neither protein kinase C activation nor inhibition mimicked or affected BDNF protection against AraC neurotoxicity. BDNF, but not NT-3, treatment of immature CGC caused a marked, but transient activation of Akt through phosphatidylinositol (PI) 3-kinase. The neuroprotective effects of BDNF were suppressed by pretreatment with LY 294002 (a PI 3-kinase inhibitor). BDNF neuroprotection was also preceded by activation of mitogen activated protein kinase (MAPK) and suppressed by two MAPK/ERK (MEK)-selective inhibitors, PD 98059 and U-0126. Moreover, inhibitors of PI 3-kinase and MEK potentiated AraC-induced neurotoxicity. These results show that neurotrophins protect against AraC-induced apoptosis, at least in part, through TrkB-mediated activation of the PI 3-kinase/Akt and MEK signaling pathways.

Valproic acid inhibits histone deacetylase activity and suppresses excitotoxicity-induced GAPDH nuclear accumulation and apoptotic death in neurons.

Valproic acid (VPA), used to treat bipolar mood disorder and seizures, also inhibits histone deacetylase (HDAC). Here, we found that VPA and other HDAC inhibitors, butyrate and trichostatin A, robustly protected mature cerebellar granule cell cultures from excitotoxicity induced by SYM 2081 ((2S, 4R)-4-methylglutamate), an inhibitor of excitatory amino-acid transporters and an agonist of low-affinity kainate receptors. These neuroprotective effects required protracted treatment and were correlated with enhanced acetylated histone levels, indicating HDAC inhibition. SYM-induced excitotoxicity was blocked by MK-801 ((5R,10S)-(+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-imine hydrogen maleate), supporting that the toxicity was largely N-methyl-D-aspartate receptor dependent. SYM excitotoxicity had apoptotic characteristics and was prevented by a caspase inhibitor. SYM-induced apoptosis was associated with a rapid and robust nuclear accumulation of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a housekeeping gene previously shown to be proapoptotic. VPA pretreatment suppressed SYM 2081-induced GAPDH nuclear accumulation, concurrent with its neuroprotective effects. Chromatin immunoprecipitation (ChIP) revealed that GAPDH is copresent with acetylated histone H3, including Lys9-acetylated histone, and that VPA treatment caused a time-dependent decrease in the levels of nuclear GAPDH with a concomitant increase in acetylated histones in the ChIP complex. Our results strongly suggest that VPA protects neurons from excitotoxicity through inhibition of HDAC activity and that this protective effect may involve suppression of excitotoxicity-induced accumulation of GAPDH protein in the nucleus.

Lithium stimulates progenitor proliferation in cultured brain neurons.

The number of neurons in the brain is controlled by production of new neurons and neuronal death. Neural progenitor proliferation in the developing and adult brain plays a prominent role in the production of new neurons. Here, we examined the effects of lithium, a mood-stabilizing drug, on neuronal proliferation in rat primary neuronal cultures. The incorporation of 5-bromo-2'-deoxyuridine (BrdU) into replicating DNA was used to label proliferating cells. BrdU incorporation was detected by immunocytochemistry in cerebellar granule cells prepared from postnatal rats and cerebral cortical cultures prepared from embryonic rats. Quantification of BrdU incorporation into cultures was performed by counting BrdU-positive cells and BrdU-coupled enzyme-linked immunosorbent assay. Both methods revealed that lithium increased BrdU incorporation in cerebellar granule cells and cerebral cortical cultures. Most BrdU-positive cells colocalized with nestin, a neuroblast cell marker, in cerebral cortical cultures. Blockade of DNA replication by cytosine arabinoside almost completely abolished BrdU incorporation, suggesting that lithium-induced BrdU incorporation was mainly due to enhanced DNA replication. Glutamate, glucocorticoids and haloperidol were found to markedly reduce neural progenitor proliferation in cerebellar granule cells. The presence of lithium prevented the loss of proliferation induced by these agents. Lithium-induced neural progenitor proliferation in vitro suggests that similar effects might occur in vivo and this action could also be related to its clinical efficacy. Cultured brain neurons may provide a valuable model for studying the molecular mechanisms underlying lithium-induced up-regulation of neural proliferation.

Neuronal apoptosis induced by pharmacological concentrations of 3-hydroxykynurenine: characterization and protection by dantrolene and Bcl-2 overexpression.

We have studied neurotoxicity induced by pharmacological concentrations of 3-hydroxykynurenine (3-HK), an endogenous toxin implicated in certain neurodegenerative diseases, in cerebellar granule cells, PC12 pheochromocytoma cells, and GT1-7 hypothalamic neurosecretory cells. In all three cell types, the toxicity was induced in a dose-dependent manner by 3-HK at high micromolar concentrations and had features characteristic of apoptosis, including chromatin condensation and internucleosomal DNA cleavage. In cerebellar granule cells, the 3-HK neurotoxicity was unaffected by xanthine oxidase inhibitors but markedly potentiated by superoxide dismutase and its hemelike mimetic, MnTBAP [manganese(III) tetrakis(benzoic acid)porphyrin chloride]. Catalase blocked 3-HK neurotoxicity in the absence and presence of superoxide dismutase or MnTBAP. The formation of H(2)O(2) was demonstrated in PC12 and GT1-7 cells treated with 3-HK, by measuring the increase in the fluorescent product, 2',7'-dichlorofluorescein. In both PC12 and cerebellar granule cells, inhibitors of the neutral amino acid transporter that mediates the uptake of 3-HK failed to block 3-HK toxicity. However, their toxicity was slightly potentiated by the iron chelator, deferoxamine. Taken together, our results suggest that neurotoxicity induced by pharmacological concentrations of 3-HK in these cell types is mediated primarily by H(2)O(2), which is formed most likely by auto-oxidation of 3-HK in extracellular compartments. 3-HK-induced death of PC12 and GT1-7 cells was protected by dantrolene, an inhibitor of calcium release from the endoplasmic reticulum. The protection by dantrolene was associated with a marked increase in the protein level of Bcl-2, a prominent antiapoptotic gene product. Moreover, overexpression of Bcl-2 in GT1-7 cells elicited by gene transfection suppressed 3-HK toxicity. Thus, dantrolene may elicit its neuroprotective effects by mechanisms involving up-regulation of the level and function of Bcl-2 protein.

beta-amyloid peptide-induced death of PC 12 cells and cerebellar granule cell neurons is inhibited by long-term lithium treatment.

Treatment of rat pheochromocytoma cells (PC 12) cells with beta-amyloid peptide-(1-42) for 24 h induced a concentration-dependent decrease in cellular redox activity in the dose range of 1 to 20 microM. These effects were markedly attenuated by pretreatment with 2 mM LiCl for 7 days, whereas 1-day pretreatment was ineffective. Measurements of live and dead cells by double-staining with fluorescein diacetate and propidium iodide, respectively revealed that protracted lithium pretreatment attenuated PC 12 cell death induced by beta-amyloid-(1-42) and cerebellar granule cell death induced by beta-amyloid-(25-35). Preceding PC 12 cell death, beta-amyloid peptide elicited a slight decrease in protein levels of Bcl-2. Conversely, 7-day pretreatment with lithium resulted in an approximate doubling of Bcl-2 protein levels in cells treated with or without beta-amyloid peptide-(1-42). Lithium-induced Bcl-2 upregulation was temporally associated with the cytoprotective effects of this drug. Thus, lithium protection against beta-amyloid peptide neurotoxicity might involve Bcl-2 overexpression, and lithium treatment for Alzheimer's disease should be reexamined.

Inhibition of excessive neuronal apoptosis by the calcium antagonist amlodipine and antioxidants in cerebellar granule cells.

Neuronal cell death as a result of apoptosis is associated with cerebrovascular stroke and various neurodegenerative disorders. Pharmacological agents that maintain normal intracellular Ca2+ levels and inhibit cellular oxidative stress may be effective in blocking abnormal neuronal apoptosis. In this study, a spontaneous (also referred to as age-induced) model of apoptosis consisting of rat cerebellar granule cells was used to evaluate the antiapoptotic activities of voltage-sensitive Ca2+ channel blockers and various antioxidants. The results of these experiments demonstrated that the charged, dihydropyridine Ca2+ channel blocker amlodipine had very potent neuroprotective activity in this system, compared with antioxidants and neutral Ca2+ channel blockers (nifedipine and nimodipine). Within its effective pharmacological range (10-100 nM), amlodipine attenuated intracellular neuronal Ca2+ increases elicited by KCl depolarization but did not affect Ca2+ changes triggered by N-methyl-D-aspartate receptor activation. Amlodipine also inhibited free radical-induced damage to lipid constituents of the membrane in a dose-dependent manner, independent of Ca2+ channel modulation. In parallel experiments, spontaneous neuronal apoptosis was inhibited in dose- and time-dependent manners by antioxidants (U-78439G, alpha-tocopherol, and melatonin), nitric oxide synthase inhibitors (N-nitro-L-arginine and N-nitro-D-arginine), and a nitric oxide chelator (hemoglobin) in the micromolar range. These results suggest that spontaneous neuronal apoptosis is associated with excessive Ca2+ influx, leading to further intracellular Ca2+ increases and the generation of reactive oxygen species. Agents such as amlodipine that block voltage-sensitive Ca2+ channels and inhibit cellular oxidative stress may be effective in the treatment of cerebrovascular stroke and neurodegenerative diseases associated with excessive apoptosis.

The c-myc coding region determinant-binding protein: a member of a family of KH domain RNA-binding proteins.

The half-life of c- myc mRNA is regulated when cells change their growth rates or differentiate. Two regions within c- myc mRNA determine its short half-life. One is in the 3'-untranslated region, the other is in the coding region. A cytoplasmic protein, the coding region determinant-binding protein (CRD-BP), binds in vitro to the c- myc coding region instability determinant. We have proposed that the CRD-BP, when bound to the mRNA, shields the mRNA from endonucleolytic attack and thereby prolongs the mRNA half-life. Here we report the cloning and further characterization of the mouse CRD-BP, a 577 amino acid protein containing four hnRNP K-homology domains, two RNP domains, an RGG RNA-binding domain and nuclear import and export signals. The CRD-BP is closely related to the chicken beta-actin zipcode-binding protein and is similar to three other proteins, one of which is overexpressed in some human cancers. Recombinant mouse CRD-BP binds specifically to c- myc CRD RNA in vitro and reacts with antibody against human CRD-BP. Most of the CRD-BP in the cell is cytoplasmic and co-sediments with ribosomal subunits.

Purification and characterization of a polysome-associated endoribonuclease that degrades c-myc mRNA in vitro.

The regulation of mRNA half-lives is determined by multiple factors, including the activity of the messenger RNases (mRNases) responsible for destroying mRNA molecules. Previously, we used cell-free mRNA decay assays to identify a polysome-associated endonuclease that cleaves c-myc mRNA within the coding region. A similar activity has been solubilized and partially purified from a high salt extract of adult rat liver polysomes. Based on a correlation between protein and enzyme activity, the endonuclease is tentatively identified as a approximately 39-kDa protein. It cleaves the coding region stability determinant of c-myc mRNA with considerable specificity. Cleavages occur predominantly in an A-rich segment of the RNA. The endonuclease is resistant to RNase A inhibitors, sensitive to vanadyl ribonucleoside complex, and dependent on magnesium. In these and other respects, the soluble enzyme we have purified resembles the polysome-associated c-myc mRNase.

Developmental regulation of CRD-BP, an RNA-binding protein that stabilizes c-myc mRNA in vitro.

We previously isolated and characterized a coding region determinant-binding protein (CRD-BP) that might regulate c-myc mRNA post-transcriptionally. CRD-BP binds specifically to the coding region of c-myc mRNA and might stabilize c-myc mRNA in vitro by protecting it from endonucleolytic cleavage. Since c-myc abundance is regulated during embryonic development and cell replication, we investigated whether CRD-BP is also regulated in animal tissues. We focused on CRD-BP expression during rat liver development and liver regeneration, because c-myc mRNA is regulated post-transcriptionally in both cases. CRD-BP expression parallels c-myc expression during liver development; the protein is present in fetal and neonatal liver but is absent or in low abundance in adult liver. In contrast, the up-regulation of c-myc mRNA following partial hepatectomy is not accompanied by up-regulation of CRD-BP. To our knowledge, CRD-BP is the first example of a putative mammalian mRNA-binding protein that is abundant in a fetal tissue but either absent from or scarce in adult tissues. Its expression in fetal liver and in transformed cell lines suggests CRD-BP is an oncofetal protein.

Carbamazepine induction of apoptosis in cultured cerebellar neurons: effects of N-methyl-D-aspartate, aurintricarboxylic acid and cycloheximide.

We have previously demonstrated that carbamazepine (CBZ) at concentrations above the therapeutic range is toxic to cultured cerebellar granule cells. Here, we ask whether the effect of CBZ involves neuronal apoptosis or necrosis. Treatment of cultured cerebellar granule cells with CBZ for 3 days resulted in a concentration-dependent fragmentation of DNA revealed as a laddered pattern in agarose gel electrophoresis, a phenomenon characteristic of apoptosis. Pretreatment of cells with N-methyl-D-aspartate (NMDA) blocked CBZ-induced DNA fragmentation and neurotoxicity as assayed by loss of mitochondrial activity with MTT or by [3H]ouabain binding to Na+/K(+)-ATPase. Aurintricarboxylic acid (ATA), a polyanionic dye, also markedly suppressed DNA fragmentation and cell death detected by morphological examination. A considerable level of DNA ladder formation was detected in untreated cells and this basal DNA fragmentation was also blocked by NMDA and ATA. Moreover, NMDA and ATA prevented CBZ-induced chromatin condensation as revealed by DNA binding with the fluorescent dye Hoechst 33258. Pretreatment of cells with cycloheximide, a protein synthesis inhibitor, prevented CBZ-induced cell death detected morphologically and attenuated CBZ-induced neurotoxicity assessed by mitochondrial activity and [3H]ouabain binding assays. Taken together, our results suggest that CBZ-induces death of cerebellar granule cells by an apoptotic process that is sensitive to NMDA, ATA and cycloheximide.

The majority of yeast UPF1 co-localizes with polyribosomes in the cytoplasm.

In Saccharomyces cerevisiae the UPF1 protein is required for nonsense-mediated mRNA decay, the accelerated turnover of mRNAs containing a nonsense mutation. Several lines of evidence suggest that translation plays an important role in the mechanism of nonsense mRNA decay, including a previous report that nonsense mRNAs assemble in polyribosomes. In this study we show that UPF1 and ribosomal protein L1 co-localize in the cytoplasm and that UPF1 co-sediments with polyribosomes. To detect UPF1, three copies of the influenza hemagglutinin epitope were placed at the C-terminus. The tagged protein, UPF1-3EP, retains 86% (+/- 5%) of function. Using immunological detection, we found that UPF1-3EP is primarily cytoplasmic and was not detected either in the nucleus or in the mitochondrion. UPF1-3EP and L1 co-distributed with polyribosomes fractionated in a 7-47% sucrose gradient. The sucrose sedimentation profiles for UPF1-3EP and L1 exhibited similar changes using three different sets of conditions that altered the polyribosome profile. When polyribosomes were disaggregated, UPF1-3EP and L1 accumulated in fractions coincident with 80S ribosomal particles. These results suggest that UPF1-3EP associates with polyribosomes. L3 and S3 mRNAs, which code for ribosomal proteins of the 60S and 40S ribosomal subunits, respectively, were on average about 100-fold more abundant than UPF1 mRNA. Assuming that translation rates for L3, S3, and UPF1 mRNA are similar, this result suggests that there are far fewer UPF1 molecules than ribosomes per cell. Constraints imposed by the low UPF1 abundance on the functional relationships between UPF1, polyribosomes, and nonsense mRNA turnover are discussed.

Gene products that promote mRNA turnover in Saccharomyces cerevisiae.

We showed previously that the increased rate of mRNA turnover associated with premature translational termination in the yeast Saccharomyces cerevisiae requires a functional UPF1 gene product. In this study, we show that the UPF1 gene codes for a 109-kDa primary translation product whose function is not essential for growth. The protein contains a potential zinc-dependent nucleic acid-binding domain and a nucleoside triphosphate-binding domain. A 300-amino-acid segment of the UPF1 protein is 36% identical to a segment of the yeast SEN1 protein, which is required for endonucleolytic processing of intron-containing pre-tRNAs. The same region is 32% identical to a segment of Mov-10, a mouse protein of unknown function. Dominant-negative upf1 mutations were isolated following in vitro mutagenesis of a plasmid containing the UPF1 gene. They mapped exclusively at conserved positions within the sequence element common to all three proteins, whereas the recessive upf1-2 mutation maps outside this region. The clustering of dominant-negative mutations suggests the presence of a functional domain in UPF1 that may be shared by all three proteins. We also identified upf mutations in three other genes designated UPF2, UPF3, and UPF4. When alleles of each gene were screened for effects on mRNA accumulation, we found that the recessive mutation upf3-1 causes increased accumulation of mRNA containing a premature stop codon. When mRNA half-lives were measured, we found that excess mRNA accumulation was due to mRNA stabilization. On the basis of these results, we suggest that the products of at least two genes, UPF1 and UPF3, are responsible for the accelerated rate of mRNA decay associated with premature translational termination.

The product of the yeast UPF1 gene is required for rapid turnover of mRNAs containing a premature translational termination codon.

mRNA decay rates often increase when translation is terminated prematurely due to a frameshift or nonsense mutation. We have identified a yeast gene, UPF1, that codes for a trans-acting factor whose function is necessary for enhanced turnover of mRNAs containing a premature stop codon. In the absence of UPF1 function, frameshift or nonsense mutations in the HIS4 or LEU2 genes that normally cause rapid mRNA decay fail to have this effect. Instead, the mRNAs decay at rates similar to the corresponding wild-type mRNAs. The stabilization of frameshift or nonsense mRNAs observed in upf1- strains does not appear to result from enhanced readthrough of the termination signal. Loss of UPF1 function has no effect on the accumulation or stability of HIS4+ or LEU2+ mRNA, suggesting that the UPF1 product functions only in response to a premature termination signal. When we examined the accumulation and stability of other wild-type mRNAs in the presence or absence of UPF1, including MAT alpha 1, STE3, ACT1, PGK1, PAB1, and URA3 mRNAs, only the URA3 transcript was affected. On the basis of these and other results, the UPF1 product appears to participate in a previously uncharacterized pathway leading to the degradation of a limited class of yeast transcripts.

Frameshift suppressor mutations affecting the major glycine transfer RNAs of Saccharomyces cerevisiae.

Mutations have been identified in Saccharomyces cerevisiae glycine tRNA genes that result in suppression of +1 frameshift mutations in glycine codons. Wild-type and suppressor alleles of genes encoding the two major glycine tRNAs, tRNA(GCC) and tRNA(UCC), were examined in this study. The genes were identified by genetic complementation and by hybridization to a yeast genomic library using purified tRNA probes. tRNA(UCC) is encoded by three genes, whereas approximately 15 genes encode tRNA(GCC). The frameshift suppressor genes suf1+, suf4+ and suf6+ were shown to encode the wild-type tRNA(UCC) tRNA. The suf1+ and suf4+ genes were identical in DNA sequence, whereas the suf6+ gene, whose DNA sequence was not determined, was shown by a hybridization experiment to encode tRNA(UCC). The ultraviolet light-induced SU F1-1 and spontaneous SU F4-1 suppressor mutations were each shown to differ from wild-type at two positions in the anticodon, including a +1 base-pair insertion and a base-pair substitution. These changes resulted in a CCCC four-base anticodon rather than the CCU three-base anticodon found in wild-type. The RNA sequence of tRNA(UCC) was shown to contain a modified uridine in the wobble position. Mutant tRNA(CCCC) isolated from a SU F1-1 strain lacked this modification. Three unlinked genes that encode wild-type tRNA(GCC), suf20+, trn2, and suf17+, were identical in DNA sequence to the previously described suf16+ frameshift suppressor gene. Spontaneous suppressor mutations at the SU F20 and SU F17 loci were analyzed. The SU F20-2 suppressor allele contained a CCCC anticodon. This allele was derived in two serial selections through two independent mutational events, a +1 base insertion and a base substitution in the anticodon. Presumably, the original suppressor allele, SU F20-1, contained the single base insertion. The SU F17-1 suppressor allele also contained a CCCC anticodon resulting from two mutations, a +1 insertion and a base substitution. However, this allele contained an additional base substitution at position 33 adjacent to the 5' side of the four-base anticodon. The possible origin and significance of multiple mutations leading to frameshift suppression is discussed.