Phenformin Suppresses Calcium Responses to Glutamate and Protects Hippocampal Neurons against Excitotoxicity

Experimental Neurology 175, 161–167 (2002) doi:10.1006/exnr.2002.7864, available online at http://www.idealibrary.com on

Phenformin Suppresses Calcium Responses to Glutamate and Protects Hippocampal Neurons against Excitotoxicity Jaewon Lee,* ,† Sic L. Chan,* Chengbiao Lu,* Mark A. Lane,* and Mark P. Mattson* ,‡ *Laboratory of Neurosciences, National Institute on Aging Gerontology Research Center, 5600 Nathan Shock Drive, Baltimore, Maryland 21224; †Department of Anatomy and Neurobiology, University of Kentucky, Lexington, Kentucky 40536; and ‡Department of Neuroscience, Johns Hopkins University School of Medicine, 725 North Wolfe Street, Baltimore, Maryland 21205 Received July 20, 2001; accepted December 14, 2001

Phenformin is a biguanide compound that can modulate glucose metabolism and promote weight loss and is therefore used to treat patients with type-2 diabetes. While phenformin may indirectly affect neurons by changing peripheral energy metabolism, the possibility that it directly affects neurons has not been examined. We now report that phenformin suppresses responses of hippocampal neurons to glutamate and decreases their vulnerability to excitotoxicity. Pretreatment of embryonic rat hippocampal cell cultures with phenformin protected neurons against glutamate-induced death, which was correlated with reduced calcium responses to glutamate. Immunoblot analyses showed that levels of the N-methyl-D-aspartate (NMDA) subunits NR1 and NR2A were significantly decreased in neurons exposed to phenformin, whereas levels of the AMPA receptor subunit GluR1 were unchanged. Whole-cell patch clamp analyses revealed that NMDA-induced currents were decreased, and AMPA-induced currents were unchanged in neurons pretreated with phenformin. Our data demonstrate that phenformin can protect neurons against excitotoxicity by differentially modulating levels of NMDA receptor subunits in a manner that decreases glutamate-induced calcium influx. These findings show that phenformin can modulate neuronal responses to glutamate, and suggest possible use of phenformin and related compounds in the prevention and/or treatment of neurodegenerative conditions. ©

2002 Elsevier Science (USA)

Key Words: AMPA; biguanide; diabetes; glutamate; ischemic stroke; NMDA; oxidative stress; patch clamp.

INTRODUCTION

Glutamate is the major excitatory neurotransmitter in the mammalian central nervous system and, as such, plays a fundamental role in most functions of the nervous system including the control of body movements, learning and memory, emotions, and sensory perception (18). While glutamate normally serves

adaptive functions, overactivation of glutamate receptors can result in neuronal degeneration and death, and such excitotoxic cell death is thought to contribute to the pathogenesis of many different disorders, including ischemic stroke, Alzheimer’s and Parkinson’s diseases, and amyotrophic lateral sclerosis (29). Excitotoxicity is mediated by calcium influx which, if not rapidly removed or buffered, initiates a cascade of events involving oxyradical production, mitochondrial dysfunction, and protease activation that ultimately kill the neuron. The vulnerability of neurons to excitotoxicity is determined by a number of factors including the complement of glutamate receptors expressed by the cell, and the abilities of the cell to extrude and buffer calcium, and to suppress oxidative stress and stabilize mitochondrial function. There are two different types of glutamate receptors, ligand-gated ion channels (ionotropic receptors) and GTP-binding protein-coupled receptors (metabotropic receptors). Ionotropic glutamate receptors can be subdivided into those selectively activated by NMDA, those activated by AMPA, and those activated by kainate. NMDA receptors are heteromeric complexes formed by an obligatory NR1 subunit, and one or more NR2 subunits designated NR2A-D (25). NMDA channel opening is dependent upon glutamate binding and prior membrane depolarization, which results in Ca 2⫹ influx. AMPA receptors are heterodimers composed of subunits GluR1–GluR4; these receptors flux both Na ⫹ and Ca 2⫹, and play a major role in glutamate-induced membrane depolarization and Ca 2⫹ influx (10, 25). Phenformin, and related biguanide compounds such as metformin, can normalize blood glucose levels, and have therefore been used in the treatment of patients with diabetes (1, 27). Phenformin and metformin intefere with cellular glucose utilization by diverting glucose metabolism through nonoxidative pathways resulting in increased lactate production, and they can also increase mitochondrial and peroxisomal ␤-oxidation (20). These biguanides have also been used to promote weight loss in obese nondiabetics. Although

161 ©

0014-4886/02 $35.00 2002 Elsevier Science (USA) All rights reserved.

162

LEE ET AL.

studied from the perspective of obesity, biguanides might also be expected to affect energy metabolism in neurons in potentially beneficial ways. In the present study we found that phenformin can protect neurons against excitotoxicity by a mechanism involving stabilization of cellular calcium homeostasis. METHODS

Hippocampal Cell Cultures, Experimental Treatments, and Quantification of Neuron Survival Hippocampi were removed from embryonic day 18 Sprague–Dawley rats (Harlan, Inc.), and cells were dissociated by mild trypsination and trituration. Cells were plated in polyethyleneimine-coated plastic 35- or 60-mm-diameter dishes or 22-mm 2 glass coverslips at a density of approximately 100 cells/mm 2 of culture surface. Cultures were maintained in Neurobasal medium containing B-27 supplements (Gibco BRL), 2 mM Lglutamine, 25 mg/ml gentamycin, 1 mM Hepes, and 0.001% gentamicin sulfate (Sigma). All experiments were performed using 8- to 10-day-old cultures. Phenformin, glutamate and N-methyl-D-aspartate (NMDA) were prepared as 200 –500X stocks in Locke’s buffer, phenformin was present during the glutamate/NMDA exposure. Neuron survival was quantified by counting viable neurons in premarked fields (10X objective; 4 fields/culture; and a minimum of 100 neurons/dish) before experimental treatment and at specified time points thereafter, as described previously (24). A minimum of four cultures were assessed for each treatment condition. Neurons with intact neurites of uniform diameter and soma with a smooth round appearance were considered viable, whereas neurons with fragmented neurites and vacuolated soma were considered nonviable. Measurement of Cellular ATP Levels A luciferin/luciferase-based assay was used to quantify ATP levels. After experimental treatment the cells were rinsed with PBS and lysed with 0.2 ml of cell lysis reagent (Roche, Mannheim, Germany), and10 ␮l of the lysate was taken for protein determination. ATP concentrations in lysates were quantified using an ATP Bioluminescence Assay Kit CLS II (Roche, Mannheim, Germany) and a luminometer (Optocomp II; MGM Instruments, Hamden, CT), according to manufacturers’ protocols. Solutions of known ATP concentrations were used to generate a standard curve, and cell lysates were diluted so that readings fell within the linear range. ATP levels were calculated as nanomole ATP per milligram protein. Measurement of Intracellular Ca 2⫹ Levels Intracellular free Ca 2⫹ levels ([Ca 2⫹] i) were quantified by fluorescence imaging of the calcium indicator

dye fura-2 as described previously (24). Briefly, cells were incubated for 30 min in the presence of 2 ␮M acetoxymethylester form of fura-2 (Molecular Probes, Eugene, OR), washed with Locke’s buffer, and incubated for 40 min prior to imaging. Cells were imaged on a Zeiss Axiovert microscope (40X oil immersion objective) coupled to an Attofluor imaging system. Phenformin was not present in the medium during the calcium imaging; before the imaging, dishes were washed three times with Locke’s buffer (without Mg2⫹ in NMDA response experiments). The [Ca 2⫹] i in 12–20 neuronal cell bodies per microscope field was monitored prior to and after exposure of cells to glutamate receptor agonists which were added to the bathing medium by dilution from 5–10X stocks. Immunoblots The immunoblot methods were similar to those described previously (16). Briefly, 50 ␮g of solubilized proteins were separated by electrophoresis in a polyacrylamide gel, transferred to a nitrocellulose sheet, and immunoreacted with primary antibody overnight at 4°C. The nitrocellulose sheet was further processed using HRP-conjugated anti-mouse secondary antibody and a chemiluminescence detection method (Amersham). The primary antibodies included: a mouse monoclonal antibody against HSP70 (Sigma, St. Louis, MO; 1:4000 dilution); a rabbit polyclonal antibody against GRP78 (StressGen, Vancouver, Canada; 1:1000 dilution); a mouse monoclonal antibody against tubulin (Sigma; 1:1000 dilution); and rabbit polyclonal antibodies against NR1, NR2A, and NR2B (1:400 dilution) or GluR1 (1:1000) (Chemicon, Inc., Temecula, CA). Whole-Cell Recordings of Excitatory Amino AcidInduced Currents These methods were similar to those used in our previous studies (13). Responses were recorded using a whole-cell recording configuration with a patch-clamp amplifier (Axopatch-1D), and data acquisition and analysis was performed using pCLAMP-8 software with filtering at 1 kHz. AMPA currents were recorded using an external solution containing 150 mM NaCl, 5 mM KCl, 2 mM CaCl 2, 1 mM MgCl 2, 10 mM glucose, 10 mM Hepes, 0.3 ␮M tetrodotoxin, and 3 ␮M cyclothiazide (pH 7.4). NMDA currents were recorded using an external solution containing 150 mM NaCl, 5 mM KCl, 2 mM CaCl 2, 10 mM glucose, 10 mM Hepes, 0.3 ␮M tetrodotoxin, and 10 ␮M glycine (pH 7.4). The external solution for AMPA application was the same as that for NMDA except that it lacked glycine and contained MgCl 2, and also contained 30 ␮M cyclothiazide to block fast desensitization of the AMPA channels. The internal solution contained 90 mM N-methyl-D-glucamine, 30 mM CsCl, 20 mM tetraethylammonium chloride, 4

PHENFORMIN PROTECTS NEURONS

163

and then returned to baseline levels by 24 h (Fig. 1a). Cultures were therefore pretreated for 48 h with phenformin or vehicle and then exposed to the excitotoxic neurotransmitter glutamate at a concentration of 100 ␮M. Measurements of neuron survival were made at 24 and 48 h after exposure to glutamate. In cultures not pretreated with phenformin glutamate killed 40 –50% of the neurons (Fig. 1b). Significantly more neurons survived exposure to glutamate in cultures that had been pretreated with phenformin. The survival of neurons pretreated with phenformin was maintained through 48 h of exposure to glutamate (Fig. 1b), suggesting a long-term protective effect of phenformin pretreatment, although a delay of the cell death process cannot be ruled out. Phenformin did not affect the pH of the culture medium when added to cultures at concentrations of 1 ␮M or less, but did decrease pH at

FIG. 1. Phenformin decreases cellular ATP levels in cultured hippocampal neurons and protects against excitotoxicity. (a) Cultures were exposed to vehicle or 1 ␮M phenformin for the indicated time periods, and levels of ATP in cell lysates were then quantified. Values are the mean and SE of determinations made in 5 separate cultures. *P ⬍ 0.05 compared to corresponding control value. (b) Cultures were pretreated for 48 h with 1 ␮M phenformin (Phen) or vehicle, and were then exposed to glutamate, and neuron survival was quantified at the indicated time points. Values are the mean and SE of determinations made in 10 separate cultures. **P ⬍ 0.01 compared to the value for cultures pretreated with phenformin and then exposed to glutamate (ANOVA with Scheffe post-hoc test).

mM Mg-ATP, 10 mM EGTA and 10 mM Hepes (pH 7.2). Cells were continuously perfused with external solution using a glass pipette positioned approximately 100 ␮m from the cell. Test agents were applied to neurons by rapidly switching from control perfusion solution to drug-containing perfusion solution. RESULTS

Phenformin Protects Hippocampal Neurons against Excitotoxic Injury In a preliminary experiment we found that concentrations of phenformin greater than 20 ␮M were toxic to cultured hippocampal neurons, and we therefore determined whether lower concentrations might exert a neuroprotective effect. Levels of ATP were decreased by approximately 20% in neurons treated with 1 ␮M phenformin; ATP levels were decreased within 12 h

FIG. 2. Calcium responses to glutamate are attenuated in hippocampal neurons pretreated with phenformin. (a) Cultures were pretreated for 48 h with a low concentration of phenformin (PL; 0.1 ␮M), a higher concentration of phenformin (PH; 1.0 ␮M), or vehicle (CON). Levels of intracellular free Ca 2⫹ were then measured prior to and during exposure to 100 ␮M glutamate. Values are the mean and SE in measurements made in 20 –30 neurons). (b) Values for basal intracellular free Ca 2⫹ concentrations, and the peak Ca 2⫹ concentration after exposure of neurons to 100 ␮M glutamate, in neurons that had been pretreated for 48 h with a low (PL; 0.1 ␮M) or higher (PH; 1.0 ␮M) concentration of phenformin. Values are the mean and SE of determinations made in at least four separate cultures. *P ⬍ 0.05 compared to control value (ANOVA with Scheffe post-hoc test).

164

LEE ET AL.

FIG. 3. Calcium responses to NMDA are attenuated in hippocampal neurons pretreated with phenformin. (a) Cultures were pretreated for 48 h with 1.0 ␮M phenformin or vehicle (CON). Levels of intracellular free Ca 2⫹ were then measured prior to and during exposure to 100 ␮M NMDA (in Mg 2⫹-free buffer). Values are the mean and SE. (b) Values for basal intracellular free Ca 2⫹ concentrations, and the peak and plateau Ca 2⫹ concentrations, after exposure of neurons to 100 ␮M NMDA, in neurons that had been pretreated for 48 h with 1 ␮M phenformin. Values are the mean and SE of determinations made in five separate cultures. *P ⬍ 0.05 compared to control value (ANOVA with Scheffe post-hoc test).

higher concentrations (data not shown), indicating that the neuroprotective effect of phenformin was not secondary to a change in pH.

FIG. 4. Levels of the stress proteins HSP70 and GRP78 are not changed in hippocampal neurons treated with phenformin. Hippocampal cultures were exposed to either 0.1 ␮M phenformin (PL) 1 ␮M phenformin (PH) or vehicle (CON) for 48 h and proteins in cell lysates were subjected to immunoblot analysis (50 ␮g protein/lane) using antibodies against GRP78 or HSP70.

FIG. 5. Phenformin induces a selective decrease in the amount of NR1 and NR2A in hippocampal neurons. Cultures were exposed to 1 ␮M phenformin (PH) or vehicle (CON) for 48 h and proteins in cell lysates were subjected to immunoblot analysis (50 ␮g protein/lane) using antibodies against the indicated NMDA (NR1, NR2A, and NR2B) or AMPA (GluR1) subunits, or an antibody against ␤-tubulin. (a) Representative immunoblot analysis. (b) Values obtained from densitometric analysis (mean and SE; n ⫽ 6).

Calcium Responses to Glutamate Are Decreased in Neurons Pretreated with Phenformin Because glutamate-induced cell death involves Ca 2⫹ influx through NMDA receptor channels (8, 13), we performed calcium imaging analyses of intracellular Ca 2⫹ levels in control and phenformin-treated neurons using the calcium indicator dye fura-2 as a probe. The basal concentration of intracellular free Ca 2⫹ was similar in neurons in control cultures and cultures pretreated for 48 h with 0.1–1.0 ␮M phenformin, ranging between 60 and 75 nM (Fig. 2). In neurons in control cultures, glutamate induced a rapid increase in the calcium concentration to a peak of approximately 375 nM, which was sustained for several minutes. Both the peak and sustained Ca 2⫹ response to glutamate were significantly attenuated, in a concentration-dependent manner, in neurons pretreated with phenformin (Fig. 2). The peak calcium response to NMDA was significantly attenuated, and the sustained calcium concentration reduced by approximately 30% in neurons pretreated with phenformin (Fig. 3).

165

PHENFORMIN PROTECTS NEURONS

FIG. 6. NMDA-induced currents are decreased in hippocampal neurons pretreated with phenformin. Cultures were treated for 48 h with either vehicle (Control) or 1 ␮M phenformin. Whole-cell currents induced by NMDA or AMPA were then recorded (see Methods). (a, b, d, e) Representative recordings of NMDA- and AMPA-induced currents. (c, f) Summary data for peak and steady-state currents induced by NMDA and AMPA. Values are the mean and SE of recordings made in at least five neurons per condition. *P ⬍ 0.05, **P ⬍ 0.01 compared to corresponding Control value.

Phenformin Selectively Alters Levels of NMDA Receptor Subunits and Decreases NMDA-Induced Currents Because phenformin transiently decreased ATP levels and because previous studies linked increases in stress proteins to neuroprotective effects of agents that induce energetic stress (11, 19, 34), we determined the effects of phenformin on two such stress proteins, heatshock protein-70 (HSP70), and glucose-regulated protein-78 (GRP78). There was no increase in levels of either HSP70 or GRP78 in neurons exposed to phenformin for 48 h (Fig. 4), suggesting that the excitoprotective mechanism of action of phenformin did not involve a stress response. In order to determine whether the decreased sensitivity of phenformin-treated neurons to glutamate might result from changes in the expression of glutamate receptor channels, we performed immunoblot analyses to assess levels of NMDA and AMPA receptor subunits in neurons that had been exposed to phenformin. After 48 h of exposure to phenformin, levels of NR1 and NR2A were decreased in neurons exposed to 1 ␮M phenformin; levels of NR2B were decreased, but not to a level that reached statistical significance (Fig. 5). Levels of the AMPA receptor subunit GluR1 were unchanged in neurons treated with phenformin (Fig. 5). In order to determine the electrophysiological consequences of changes in NMDA receptor subunit levels in neurons treated with phenformin, we recorded whole-

cell currents induced by either NMDA or AMPA. Cultures were pretreated for 48 h with phenformin or vehicle and NMDA- and AMPA-induced currents were recorded. The amplitudes of both the peak and steadystate NMDA-induced current were decreased in neurons in phenformin-treated cultures compared to control cultures (Fig. 6). In contrast, neither the peak or steady-state currents induced by AMPA were affected by phenformin treatment. Phenformin was not a direct blocker of NMDA receptor channels because acute phenformin treatment did not affect NMDA-induced whole-cell current, and did not alter calcium responses to NMDA (data not shown). DISCUSSION

Phenformin has been used in humans to treat diabetes, and studies of its anti-diabetic mechanisms of action suggest that it induces a mild energetic stress (decreased aerobic metabolism), enhances lactate production, and increases insulin sensitivity (20). We found that treatment of cultured hippocampal neurons with phenformin, at concentrations that cause a small and transient decrease in ATP levels, results in an increased resistance of the neurons to excitotoxicity. Previous studies have shown that exposure of animals to mild energetic stressors, such as a brief ischemia (2, 21), dietary restriction (3, 33), or 2-deoxy-D-glucose (11, 34) can increase resistance of neurons to excitotoxic and ischemic cell death. Similar “preconditioning” ef-

166

LEE ET AL.

fects of mild stress have been reported in studies of cultured neurons (19, 22, 35). Although a similar preconditioning mechanism could account for the neuroprotective effect of phenformin we observed, this appears not to be the case because phenformin did not cause an increase in levels of the stress proteins HSP70 and GRP78. The modest reduction in ATP levels induced by phenformin may not be sufficient to induce the same type of stress response previously documented in cultured hippocampal neurons exposed to 2-deoxy-D-glucose (19). Phenformin can increase lactate production, and lactate can serve as an energy source for neurons, and may protect them against ischemic and excitotoxic injury (30). It is therefore possible that increased lactate production contributes to the neuroprotective effects of phenformin, although we did not detect a change in pH which usually accompanies lactate formation. Instead, our data suggest a novel mechanism whereby phenformin protects neurons against excitotoxicity is by suppressing calcium influx through NMDA receptor channels. The decreased sensitivity to glutamate toxicity conferred by phenformin was associated with a decreased calcium response to glutamate and NMDA. The decreases in levels of the NR1 and NR2A subunits are consistent with the observed decrease in NMDA sensitivity. NMDA receptors are heterooligomeric complexes consisting of NR1 plus one of four NR2 subunits (NR2A, NR2B, NR2C, and NR2D). Electrophysiological analysis of recombinant NMDA receptors revealed that the type of NR2 subunit present in NMDA receptor complex can determine the functional properties of the NMDA channel (5, 17, 27). The peak channel open probability for NR1/NR2A channels is considerably greater than for NR1/NR2B channels. Thus, the decreased amount of NR2A caused by phenformin would be expected to decrease the calcium responses to glutamate, which is precisely what we observed. Although the present findings are the first to document an effect of a biguanide compound on responses of neurons to glutamate, one previous study showed that metformin can inhibit ganglionic transmission in the peripheral nervous system (26). It is not known how phenformin affects levels of the NMDA receptor subunits, but one possibility is suggested by the ability of phenformin to enhance insulin signaling (33). Previous studies have shown that neurotrophic factors that activate signal transduction pathways similar to that of insulin, including insulin-like growth factors, can protect neurons against excitotoxic insults (6, 23). One such excitoprotective signaling cascade involves the Akt kinase (4, 14). Precedence for modulation of glutamate receptor subunit expression by neuroprotective signals comes from studies showing that excitoprotective neurotrophic factors and cytokines can alter expression of NMDA and/or AMPA receptor subunits in a subunit selective manner (7).

From the perspective of potential therapeutic use of phenformin, it will now be important to establish its efficacy in animal models of neurodegenerative conditions, such as stroke and traumatic brain injury, that involve overactivation of glutamate receptors. Previous studies have shown that long-term administration of biguanides such as phenformin and metformin is welltolerated in humans (30), although phenformin may have deleterious effects on mitochondria (15). Moreover, it has been reported that lifelong administration of phenformin can reduce tumor incidence and extend lifespan in mice (9). Although the extent to which phenformin penetrates the blood-brain barrier is not known, the related biguanide compounds meta-chlorophenylbiguanide (mCPBG) and 1-(2-naphthyl) biguanide (2-NBG) have been shown to exert effects in the brain when given systemically (12). The only well-established means of extending lifespan is by caloric restriction (32). Phenformin exerts several physiological effects in vivo similar to caloric restriction including decreased plasma glucose levels and increased insulin sensitivity. Manipulating cellular energy metabolism through caloric restriction and pharmacological approaches is proving effective in preclinical tests of neuroprotective compounds (11, 34). The present demonstration of a neuroprotective action of an antidiabetic compound therefore suggests the potential efficacy of such compounds in the neurology clinic. REFERENCES 1. 2.

3.

4.

5.

6.

7.

8.

Bailey, C. J. 1992. Biguanides and NIDDM. Diabetes Care 15: 755–772. Barone, F. C., R. F. White, P. A. Spera, J. Ellison, R. W. Currie, X. Wang, and G. Z. Feurstein. 1998. Ischemic preconditioning and brain tolerance: Temporal histological and functional outcomes, protein synthesis requirement, and interleukin-1 receptor antagonist and early gene expression. Stroke 29: 1937–1950. Bruce-Keller, A. J., G. Umberger, R. McFall, and M. P. Mattson. 1999. Food restriction reduces brain damage and improves behavioral outcome following excitotoxic and metabolic insults. Ann. Neurol. 45: 8 –15. Chalecka-Franaszek, E., and D. M. Chuang. 1999. Lithium activates the serine/threonine kinase Akt-1 and suppresses glutamate-induced inhibition of Akt-1 activity in neurons. Proc. Natl. Acad. Sci. USA 96: 8745– 8750. Chen, N., T. Luo, and L. A. Raymond. 1999. Subtype-dependence of NMDA receptor channel open probability. J. Neurosci. 19: 6844 – 6854. Cheng, B., and M. P. Mattson. 1992. IGF-I and IGF-II protect cultured hippocampal and septal neurons against calcium-mediated hypoglycemic damage. J. Neurosci. 12: 1558 –1566. Cheng, B., K. Furukawa, J. A. O’Keefe, Y. Goodman, M. Kihiko, T. Fabian, and M. P. Mattson. 1995. Basic fibroblast growth factor selectively increases AMPA-receptor subunit GluR1 protein level and differentially modulates Ca2⫹ responses to AMPA and NMDA in hippocampal neurons. J. Neurochem. 65: 2525–2536. Choi, D. W. 1994. Calcium and excitotoxic neuronal injury. Ann. N. Y. Acad. Sci. 747: 162–171.

PHENFORMIN PROTECTS NEURONS 9.

10. 11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Dilman, V. M. and V. N. Anisimov. 1980. Effect of treatment with phenformin, diphenylhydantoin or L-dopa on the ife span and tumour incidence in C3H/Sn mice. Gerontology 26: 241– 245. Doble, A. 1995. Excitatory amino acid receptors and neurodegeneration. Therapie 50: 319 –337. Duan, W., and M. P. Mattson. 1999. Dietary restriction and 2-deoxyglucose administration improve behavioral outcome and reduce degeneration of dopaminergic neurons in models of Parkinson’s disease. J. Neurosci. Res. 57: 195–206. Dukat, M., R. Young, N. N. Barmani, B. Ahmed and R. A. Glennon. 2000. The 5-HT3 agent N-(3-chlorophenyl)guanidine (MD-354) serves as a discriminative stimulus in rats and displays partial agonist character in a shrew emesis assay. Psychopharmacology (Berlin) 150: 200 –207. Furukawa, K., W. Fu., W. Witke, D. J. Kwiatkowski, and M. P. Mattson. 1997. The actin-severing protein gelsolin modulates calcium channel and NMDA receptor activities and vulnerability to excitotoxicity in hippocampal neurons. J. Neurosci. 17: 8178 – 8186. Gary, D. S. and M. P. Mattson. 2001. Integrin signaling via the PI3 kinase–Akt pathway increases neuronal resistance to glutamate-induced apoptosis. J. Neurochem 76: 1485–1496. Gettings, S. D., J. E. Reeve, and L. J. King. 1988. Possible role of intracellular Ca2⫹ in the toxicity of phenformin. Biochem. Pharmacol. 37: 281–289. Glazner, G. W., S. L. Chan, C. Lu and M. P. Mattson. 2000. Caspase-mediated degradation of AMPA receptor subunits: A mechanism for preventing excitotoxic necrosis and ensuring apoptosis. J. Neurosci. 20: 3641–3649. Hoffmann, H., T. Gremme, H. Hatt, and K. Gottmann. 2000. Synaptic activity-dependent developmental regulation of NMDA receptor subunit expression in cultured neocortical neurons. J. Neurochem. 75: 1590 –1599. Hudspith, M. J. 1997. Glutamate: A role in normal brain function, anaesthesia, analgesia and CNS injury. Br. J. Anaesth. 78: 731–747. Lee, J., A. J. Bruce-Keller, I. Kruman, S. L. Chan, and M. P. Mattson. 1999. 2-deoxy-D-glucose protects hippocampal neurons against excitotoxic and oxidative injury: Evidence for the involvement of stress proteins. J. Neurosci. Res. 57: 48 – 61. Lenhard, J. M., S. A. Kliewer, M. A. Paulik, K. D. Plunket, J. M. Lehmann, and J. E. Weiel. 1997. Effects of troglitazone and metformin on glucose and lipid metabolism. Biochem. Pharmacol. 54: 801– 808. Liu, Y., H. Kato, N. Nakata, and K. Kogure. 1993. Temporal profile of heat shock protein 70 synthesis in ischemic tolerance induced by preconditioning ischemia in rat hippocampus. Neuroscience 56: 921–927. Lowenstein, D. H., P. Chan, and M. Miles. (1991) The stress protein response in cultured neurons: Characterization and

23.

24.

25.

26.

27.

28.

29. 30.

31.

32. 33.

34.

35.

167

evidence for a protective role in excitotoxicity. Neuron 7: 1053– 1060. Mattson, M. P., Y. Zhang, and S. Bose. 1993. Growth factors prevent mitochondrial dysfunction, loss of calcium homeostasis, and cell injury, but not ATP depletion in hippocampal neurons deprived of glucose. Exp. Neurol. 121, 1–13. Mattson, M. P., M. A. Lovell, K. Furukawa, and W. R. Markesbery. 1995. Neurotrophic factors attenuate glutamate-induced accumulation of peroxides, elevation of intracellular Ca 2⫹ concentration, and neurotoxicity and increase antioxidant enzyme activities in hippocampal neurons. J. Neurochem. 65: 1740 – 1751. Michaelis, E. K. 1998. Molecular biology of glutamate receptors in the central nervous system and their role in excitotoxicity, oxidative stress and aging. Prog. Neurobiol. 54: 369 – 415. Petersen, J. S., W. Liu, D. R. Kapusta, and K. J. Varner. 1997. Metformin inhibits ganglionic neurotransmission in renal nerves. Hypertension 29: 1173–1117. Rendell, M. S., and W. R. Kirchain. 2000. Pharmacotherapy of type 2 diabetes mellitus. Ann. Pharmacother. 34(7– 8):878 – 895. Roberts, E. B., and A. S. Ramoa. 1999. Enhanced NR2A subunit expression and decreased NMDA receptor decay time at the onset of ocular dominance plasticity in the ferret. J. Neurophysiol. 81, 2587–2591. Sattler, R., and M. Tymianski. 2000. Molecular mechanisms of calcium-dependent excitotoxicity. J. Mol. Med. 78: 3–13. Schurr, A., R. S. Payne, J. Miller, and B. M. Rigor. 1997. Brain lactate is an obligatory aerobic energy substrate for functional recovery after hypoxia: Further in vitro validation. J. Neurochem. 69: 423– 426. Sirtori, C. R., and C. Pasik. 1994. Re-evaluation of a biguanide, metformin: Mechanism of action and tolerability. Pharmacol. Res. 30: 187–228. Sohal, R. S., and R. Weindruch. 1996. Oxidative stress, caloric restriction, and aging. Science 273: 59 – 63. Vigneri, R., B. Maddux, and I. D. Goldfine. 1984. The effect of phenformin and other adenosine triphosphate-lowering agents on insulin binding to IM-9 human cultured lymphocytes. J. Cell. Biochem. 24: 177–186. Yu, Z. F., and M. P. Mattson. 1999. Dietary restriction and 2-deoxyglucose administration reduce focal ischemic brain damage and improve behavioral outcome: Evidence for a preconditioning mechanism. J. Neurosci. Res. 57: 830 – 839. Yu, Z. F., H. Luo, W. Fu,, and M. P. Mattson. 1999. The endoplasmic reticulum stress-responsive protein GRP78 protects neurons against excitotoxicity and apoptosis: Suppression of oxidative stress and stabilization of calcium homeostasis. Exp. Neurol. 155: 302–314.