Activation of the Metabotropic Glutamate Receptor Attenuates N
Total Page:16
File Type:pdf, Size:1020Kb
Proc. Natl. Acad. Sci. USA Vol. 88, pp. 9431-9435, November 1991 Neurobiology Activation of the metabotropic glutamate receptor attenuates N-methyl-D-aspartate neurotoxicity in cortical cultures (trans-1-aminocyclopentyl-1,3-dicarboxylic acid/carbachol/Alzheimer disease/development/neurodegeneration) JAE-YOUNG KOH, ELIZABETH PALMER, AND CARL W. COTMAN* Department of Psychobiology, University of California, Irvine, CA 92717 Communicated by James L. McGaugh, July 15, 1991 (receivedfor review April 3, 1991) ABSTRACT Excitatory amino acid receptor-mediated neu- methylisoxazole-4-propionic acid and kainate receptors) are rotoxicity (excitotoxicity) has been proposed to contribute to directly associated with the specific ligand-gated ionophores neuronal loss in a wide variety of neurodegenerative conditions. that permit influx of cations upon activation (ionotropic Although considerable evidence has accumulated implicating receptor). In contrast, the recently discovered metabotropic N-methyl-D-aspartate (NMDA), kainate, and a-amino-3- glutamate, or trans-1-aminocyclopentyl-1,3-dicarboxylic hydroxy-5-methylisoxazole4propionic acid receptors in the acid (ACPD), receptor (10-12) activates a phospholipase, via processes of excitotoxicity, relatively little research has focused a guanine nucleotide-binding regulatory protein, that hydro- on the ability of other neurotransmitter systems to influence lyzes membrane phospholipid to generate the second mes- sengers diacylglycerol (DAG) and inositol trisphosphate excitotoxic neuronal injury. In the present study, we examined (1P3). DAG, in combination with Ca2+, activates protein the effects of rans-l-aminocyclopentyl-1,3-dicarboxylic acid kinase C (PKC) (13), whereas IP3 mobilizes Ca2+ from (ACPD), a selective agonist for the metabotropic glutamate, or intracellular stores in various cell types (14). ACPD, receptor, and carbachol, an agonist at the acetylcholine Considering the evidence that calcium ions may play a receptor, on neuronal degeneration produced by brief exposure critical role in producing excitotoxic neuronal damage (15, to NMDA in murine cortical cultures. Since excitotoxic neuronal 16), each of the three receptors appears capable of contrib- injury is probably caused by increases in intracellular Ca21 uting to neuronal damage by increasing intraneuronal free concentrations, the two transmitter agonists were of particular calcium concentrations. While activation of the ionotropic interest as both have been shown to mobilize intracellular receptors is sufficient to produce most of the EAA neuro- calcium stores. Contrary to what might be expected, ACPD and, toxicity (1, 17, 18), probably through a large influx ofcalcium to a lesser degree, carbachol attenuated NMDA neurotoxicity. (15, 16, 19), evidence for involvement of the metabotropic The neuroprotective effect of ACPD, but not of carbachol, was receptor in excitotoxic neuronal damage was not available dependent upon the developmental state of cultures; in older until recently, largely due to the lack of selective agonists or cultures (.18 days in vitro), the protective effect decreased. The antagonists. In this regard, certain unusual pharmacological neuroprotection by ACPD may be, in part, mediated by protein effects of a nonselective agonist such as quisqualate have kina, since protection is partially reversed by the protein often been difficult to interpret. For example, Garthwaite and kinase antagonists H-7 and HA-1004. These data suggest that Garthwaite (20) proposed that the 6-cyano-2,3-dihydroxy-7- concomitant activation of the ACPD receptor may serve as a nitroquinoxaline (CNQX)-insensitive induction of CNQX- protective mechanism against neurotoxicity that could be pro- sensitive neurotoxicity by a pulse exposure to quisqualate duced by briefintense NMDA receptor activation during normal may be mediated by the metabotropic receptor. However, a similar quisqualate neurotoxicity in cortical cultures has been or abnormal brain function. shown to be largely attributable to the cellular uptake and subsequent release of quisqualate (21). The recent discovery Excitatory amino acid (EAA) neurotoxicity (excitotoxicity) has of ACPD, a selective agonist for the metabotropic receptor been proposed to contribute to neuronal loss in a broad spec- (12), now makes it possible to directly examine the role ofthe trum ofneurodegenerative conditions including ischemia, Hun- ACPD receptor in excitotoxic neuronal damage. We have tington disease, and Alzheimer disease (AD) (1, 2). As neuro- found that ACPD is not neurotoxic itself nor does it poten- degenerative diseases are known to be associated with a variety tiate kainate or a-amino-3-hydroxy-5-methylisoxazole-4- of specific pathologies, understanding the factors that can alter propionic acid neurotoxicity in cortical cultures (22). How- EAA neurotoxicity may prove clinically important. For exam- ever, effects of ACPD on NMDA receptor-mediated neuro- ple, f/A4 protein that accumulates in amyloid plaques of AD toxicity were not examined. patients increases the vulnerability ofcortical neurons to EAA- As a step toward a better understanding of neurotrans- mediated neuronal injury (3). Similarly, other neurotransmitter mitter interactions in neuronal degeneration, the present systems, which are often altered anatomically orfunctionally in study examines the effects of ACPD and carbachol, an various disease states, could affect the susceptibility ofneurons agonist at the acetylcholine receptor, on neurotoxicity pro- to excitotoxic damage. Forinstance, in AD, there is widespread duced by brief pulse exposure to NMDA. reduction in cholinergic (4) and somatostatinergic markers (5, 6). The ability ofthe non-EAA system to affect EAA-mediated MATERIALS AND METHODS injury may reflect transmitter interactions in normal brain, as Cortical Cell Culture. Primary cerebral cortical cultures serotonin has been reported to change electrophysiological were prepared from 14- to 15-day-old fetal mice generally as responses of neurons to N-methyl-D-aspartate (NMDA) (7, 8). It is now well established that there are at least three Abbreviations: NMDA, N-methyl-D-aspartate; EAA, excitatory general classes of glutamate receptors (9). The NMDA and amino acid; ACPD, trans-1-aminocyclopentyl-1,3-dicarboxylic acid; the non-NMDA receptors (i.e., a-amino-3-hydroxy-5- CNQX, 6-cyano-2,3-dihydroxy-7-nitroquinoxaline; DIV, days in vi- tro; LDH, lactate dehydrogenase; PI, phosphatidylinositol; AD, Alzheimer disease; DAG, diacylglycerol; IP3, inositol trisphosphate; The publication costs of this article were defrayed in part by page charge PKC, protein kinase C; CPP, 3-[(+)-2-carboxypiperazin-4-yl]propyl- payment. This article must therefore be hereby marked "advertisement" 1-phosphonic acid. in accordance with 18 U.S.C. §1734 solely to indicate this fact. *To whom reprint requests should be addressed. 9431 Downloaded by guest on September 27, 2021 9432 Neurobiology: Koh et al. Proc. Natl. Acad. Sci. USA 88 (1991) described by Choi et al. (23) with minor modifications (3). with a phase-bright microscope (Olympus model IMT-2); dam- Dissociated cortical cells were plated in 16-mm multiwell aged neurons were easily identifiable as most turned into debris vessels (4 x 105 cells per well) in Eagle's minimal essential (see Fig. lA) and were stainable with trypan blue (0.4%, 5 min)). medium (with Earle's salts) supplemented with 10% heat- In most experiments, neuronal cell injury was further quanti- inactivated horse serum, 10o fetal bovine serum, glutamine tated chemically by measurement ofthe lactate dehydrogenase (2 mM), and glucose (21 mM). After 5-10 days in vitro (DIV), (LDH) released by damaged cells into the extracellular fluid 1 nonneuronal cell division was inhibited by exposure to 10 ,uM day after EAA exposure (25), using an automated microplate cytosine arabinoside for 1-3 days, and the cells were shifted reader (Thermomax; Molecular Devices, Palo Alto, CA) and into a serum-free maintenance medium containing ingredi- kinetics software. Previous control studies have demonstrated ents described by Brewer and Cotman (24). Subsequent that specific LDH release is linearly proportional to the number medium change was carried out twice a week. For neuro- of neurons damaged (23, 25). toxicity experiments, only cultures ofDIV 14 or greater were Phosphatidylinositol (PI) Hydrolysis. Agonist-induced PI used, since previous studies showed that younger cortical hydrolysis was measured as described (26). Briefly, cells neurons are resistant to brief glutamate exposure (23). were labeled for 24 hr at 37°C with 1 ml of serum-free culture Exposure to EAAs. Exposure to EAAs was carried out as medium containing 2.5 gCi (1 Ci = 37 GBq) of myo- described (21). All agonist exposure was performed at room [3H]inositol. The cells were washed three times with 1 ml of temperature in a Hepes-buffered salt solution with the fol- physiological saline buffer with the following composition: lowing composition: 120 mM NaCl, 5.4 mM KCl, 0.8 mM 127 mM NaCl, 2 mM KCI, 1.25 mM KH2PO4, 26 mM MgCI2, 1.8 mM CaC12, 20 mM Hepes (pH 7.4 at 250C), and 15 NaHCO3, 2 mM CaCl2, 2 mM MgSO4, and 10 mM D-glucose, mM glucose. After the exposure, serum-free maintenance equilibrated with 95% 02/5% CO2. The cells were then medium was replaced, and cultures were returned to the 5% preincubated for 15 min at 37°C in 500 ul of physiological C02/95% 02 incubator until the next day, when the evalua- saline buffer containing 10 mM LiCl, 500 nM tetrodotoxin, 50 tion of