Neurochemical Research, Vol. 24, No. 7, 1999, pp. 843-848

NcG-Nitro-L-Arginine, a Nitric Oxide Synthase Inhibitor,Antagonizes Quinolinic Acid-Induced Neurotoxicityand Oxidative Stress in Rat Striatal Slices

Daniel Santamaria,1 Velia Espinoza-Gonzalez,2 Camilo Rios,2 and Abel Santamaria2,3

(Accepted February 25, 1999)

Nitric oxide (NO) is a potential contributor to neurotoxicity following overactivation of N-methyl-D-aspartate (NMDA) receptors. In this work we investigated the effect of Nco-nitro-L-arginine(L-NARG 25, 50, or 100 uM), a selective inhibitor of nitric oxide synthase (NOS) -the syntheticenzyme of NO- on quinolinic acid (QUIN 100 uM)-induced neurotoxicity (measured as lactatedehydrogenase (LDH) leakage) in rat Striatal slices. Oxidative stress was also measured both aslipid peroxidation and as the levels of reduced (GSH) and oxidized (GSSG) glutathione, in an ef-fort to elucidate a possible participation of NO in the toxic mechanisms involved in NMDA re-ceptor-mediated neuronal injury. The action of L-arginine (L-ARG 100 or 200 uM), a well-knownNO precursor, was also tested on QUIN-induced neurotoxicity and oxidative stress. Resultsshowed that QUIN produced significant changes in both cell damage (177%) and oxidative in-jury (203% in lipid peroxidation, 68% in GSH, and 123% in GSSG) as compared to control val-ues. All these effects were antagonized by adding L-NARG to the incubation media, whereasL-ARG alone, or in combination with QUIN, significantly enhanced both lipid peroxidation andLDH leakage. Moreover, the protective effects of L-NARG on QUIN-induced lipid peroxidationwere reversed by addition of an excess of L-ARG to the media. These findings indicate thatNO is probably mediating the mechanism of neurotoxicity produced by QUIN, which may be ofpotential value to explain the molecular basis of neurodegenerative processes linked to QUIN-mediated NMDA receptor overactivation.

KEY WORDS: Quinolinic acid; nitric oxide; neurotoxicity; lipid peroxidation; oxidative stress.

INTRODUCTION

Quinolinic acid (QUIN), an N-methyl-D-aspartate(NMDA) subtype of glutamate receptor agonist exhibit-ing excitatory activity in the brain (1), has been in-

1 Department of Neurochemistry, National Institute of Pediatrics,S.S.A., Mexico.

2 Department of Neurochemistry, National Institute of Neurologyand Neurosurgery Manuel Velasco Sudrez, S.S.A., Mexico.

3 To whom to address reprint requests. Departamento de Neuro-qufmica, Instituto Nacional de Neurologia y Neurocirugia ManuelVelasco Sudrez, S.S.A. Ave. Insurgentes Sur 3877, Tlalpan D.F.,Mexico 14269, MEXICO. Tel.: (+525)606-3822 (ext. 2005). Fax:(+525)528-0095. E-mail: abels@fournier.facmed.unam.mx.

843

volved in inflammatory and infectious disorders of thenervous system (2). As an heterocyclic amino acid de-rived from L-tryptophan, QUIN is a precursor of NAD+at the kynurenine pathway (3). QUIN is also able to pro-duce excessive membrane depolarizations and calciuminflux to the neurons, which in turn, may result in sev-eral toxic consequences, such as activation of proteasesand lipases, generation of free radicals, activation of theconstitutive nitric oxide synthase (NOS), disruption ofmitochondrial oxidative phosphorylation, enhancedbrain lipid peroxidation and increased levels of carbonmonoxide (1,4-6). Recently, the neurotoxic action ofQUIN has been associated to the production of reactiveoxygen species (7).

0364-3190/99/0700-0843$16.00/0 © 1999 Plenum Publishing Corporation

844 Santamaria, Espinoza-Gonzalez, Rios, and Santamaria

On the other hand, as a highly-diffusible moleculeexhibiting physiological activity in the brain, NO hasbeen commonly proposed as a putative cellular messen-ger (8-10), also showing neurotoxic effects dependingeither on its redox status (11,12) or its concentrations(13). Therefore, NO has been currently related to patho-logical events such as apoptosis, oxidative stress and celldamage (14), involving the Ca2+-calmodulin-dependentactivation of the constitutive nitric oxide synthase(cNOS) after NMDA receptor stimulation (15,16). Fol-lowing this line of research, we have recently shown thatNco-nitro-L-arginine (L-NARG), a well-known NOS in-hibitor, successfully prevent the QUIN-induced lipidperoxidation in synaptosomes from whole rat brain,whereas a precursor of nitric oxide (NO), L-arginine(L-ARG), was effective to increase lipid peroxidationunder the same experimental conditions (17). Moreover,NO has been directly implicated in the pattern of toxic-ity produced by QUIN (18), since Perez-Severiano andcoworkers have shown that NO-nitro-L-arginine methylester, a selective NOS inhibitor, is able to preventQUIN-induced circling behavior, lipid peroxidation andNOS activation in a Ca2+-dependent manner.

Taken all these evidences together, it is likely thatelevated concentrations of QUIN and the consequentoveractivation of NMDA receptors might result in toxiclevels of NO as part of a complex pattern of neuro-toxicity. In order to elucidate whether the addition of aNOS inhibitor, L-NARG, to the striatal slices may re-duce the neurotoxic and oxidative actions of QUIN, weconducted the following experiments. We also evalu-ated the effect of a well-known NO precursor, L-argi-nine (L-ARG), on some of the toxic events elicited byQUIN. Striatal slices were selected as suitable biologi-cal preparations in light of both the susceptibility ofsuch a brain region to the toxic action of QUIN as wellas the integrative character of the slices for in vitrostudies.

EXPERIMENTAL PROCEDURE

Reagents. Most of reagents were obtained from Sigma Chemi-cal Co. (St. Louis, MO). All other reagents were obtained from E.Merck (Mexico) and Mallinkrodt/Baker (Mexico).

Dissection of Striatal Slices. Male Wistar rats (250-300 g) werekilled by decapitation. Their brains were removed and briefly storedin cold saline solution and the corpora striata were then dissected out.Groups of 5 striatal slices (200-300 nm of thickness) were sectionedby hand. After a recovery period of 45 min, they were incubated in 2ml of Ringer solution (containing 118 mM NaCl, 1.2 mM KH2PO4,4.7 mM KC1, 2.5 mM CaCl2, 1.17 mM MgSO4, 5.6 mM D(+)-Glu-cose and 2.5 mM TRIS) at 37°C in an oxygenated shaking-water bathfor 60 minutes in the presence of either QUIN (100 uM as final con-

centration), L-ARG (100 or 200 uM), L-NARG (25, 50, or 100 uM)or some combinations of them.

The Assay of Lactate Dehydrogenase (LDH) Leakage. The ex-tent of quinolinate-induced cell injury was estimated according toWroblewski and LaDue (19), by measurement of LDH (EC 1.1.1.27)leakage from damaged cells of the striatal slices. After the treatmentof the slices, 500 ul-aliquots were obtained from the incubation media.LDH activity was measured employing the LDH diagnostic kit fromHuman (containing 50 mmol/1 of TRIS buffer pH 7.4 + 1.2 mmol/1 ofpyruvate + 5 mmol/1 of EDTA + 0.15 mmol/1 of NADH), by follow-ing the absorbance at 340 nm of wavelength. Results were expressedas International Units (IU) per mg of protein, according to the calcu-lation procedure of the LDH diagnostic kit supplier.

The Assay of Thiobarbituric Acid-Reactive Substances (TBARS).Lipid peroxidation was measured in the striatal slices by detectionof TBA-reactive substances, according to previous reports (5,17).Briefly, after exposure to the incubation conditions, the slices were re-moved from the media and placed into tubes containing 1.5 ml ofdeionized water. The tissue was then sonicated and the homogenateswere used for measurement of peroxidation by addition of 2 ml of theTBA reagent (0.375 g of TBA + 15 g of trichloroacetic acid + 2.5 mlof HC1 in 100 ml of water). The final solution was heated in a boilingwater bath for 30 min. After cooling the samples on ice, they werecentrifuged at 3,000 g for 15 min and the absorbance of the respectivesupernatants was measured at 532 nm in a Beckman DU-6 spec-trophotometer. Concentrations of TBARS were calculated by interpo-lation in a standard curve of periodic oxidation of 2-deoxy-D-ribose.Final results were expressed as nmoles of TBARS per mg of protein.

The Fluorometric Assay of Reduced (GSH) and Oxidized (GSSG)Glutathione. Reduced (GSH) and oxidized (GSSG) glutathione levelswere both measured according to the method previously described byHissin and Hilf (20). After incubating in the presence of differenttreatments, the striatal slices (five per probe) were homogenized in3.75 ml EDTA-phosphate buffer plus 1 ml HPO3 (25%). Homogenateswere then centrifuged at 3,000 g for 15 min and the resultant super-natants were separated in two equal volumes of 0.5 ml.

For measurement of GSH, the first 0.5 ml aliquots were addedto 4.5 ml of phosphate buffer (pH 8.0) and mixed. One-hundred ulof the mixtures were added to 1.8 ml phosphate buffer (pH 8.0) plus100 ul o-phthalaldehyde (OPA). The new mixtures were incubatedat room temperature for 15 min and their fluorescent signals wererecorded in a Perkin-Elmer LS50B Luminescence Spectrometer at420 nm of emission and 350 nm of excitation wavelengths.

For measurement of GSSG, the second 0.5 ml fractions of super-natants were added to 200 ul of 0.04 M N-ethylmaleimide (NEM)0.04 M, in order to prevent oxidation of GSH to GSSG. Mixtureswere incubated at room temperature for 30 min and then, 4.3 ml of0.1 N NaOH was added. One-hundred ul of the resultant mixtureswere added to 1.8 ml of 0.1 N NaOH plus 100 ul of OPA. The sam-ples were then incubated at room temperature for 15 min and theirfluorescent signals were recorded at the same conditions that we em-ployed for GSH.

Final results were expressed as nmoles of GSH or GSSG per mgof protein or as the GSH/GSSG ratio, as previously reported (21).

Protein Measurement. Protein contents in the homogenates weremeasured by the Folin & Ciocalteu's phenol reagent (22) in a Beck-man DU-6 Spectrophotometer at 550 nm of wavelength. All resultswere finally corrected by protein content.

Statistical Analysis. Two-way or three-way analysis of variance(ANOVA) and Tukey's tests for multiple comparisons (23) were ap-plied for the statistical analysis of all results. Values of P < 0.05 andP < 0.01 were considered of statistical significance.

Fig. 1. The action of increasing concentrations of Nw-nitro-L-arginine(L-NARG) on quinolinic acid (QUIN)-induced cell damage in rat stri-atal slices. Cell injury was measured as lactate dehydrogenase (LDH)leakage (international units per mg of protein) in probes containingfive slices each. Tissue was added with 2 ml of Ringer solution andincubation was done in an oxygenated shaking water bath at 37°C for60 minutes in the presence of QUIN, L-NARG or some combinationsof them. Error bars represent SEM of 5 to 8 experiments. aP < 0.01compared to control and bP < 0.01 versus QUIN.

Fig. 2. The action of NO-nitro-L-arginine (L-NARG) on L-arginine(L-ARG)- or L-ARG plus quinolinic acid (QUIN)-induced celldamage in rat striatal slices. Cell injury was measured as lactate de-hydrogenase (LDH) leakage (international units per mg of protein)in probes containing five slices each. Tissue was added with 2 mlof Ringer solution and incubation was done in an oxygenated shak-ing water bath at 37°C for 60 minutes in the presence of QUIN,L-NARG, L-ARG or some combinations of them. Error bars repre-sent SEM of 5 to 8 experiments. aP < 0.01 compared to control andbP < 0.01 versus QUIN.

RESULTS

Effects of L-NARG on QUIN- and L-ARG-InducedLDH Leakage. Two independent series of experimentswere performed and presented respectively as Fig. 1 andFig. 2. Significant increases in LDH leakage were foundin the media from the QUIN-treated striatal slices(177% and 181%) as compared to control values (100%)(Fig. 1 and 2). The incubation of the slices in the pres-ence of 100 uM L-NARG produced values of QUIN-and L-ARG-induced LDH leakage similar to controls(109% and 113%, respectively) (Fig. 1 and 2). No pro-tection against QUIN-induced damage was found in thepresence of 25 uM L-NARG (166% vs control) whereas50 uM L-NARG partially prevented QUIN-inducedLDH leakage (132% vs control) (Fig. 1). The additionof L-ARG alone or L-ARG + QUIN to the incubationmedia resulted in enhanced leakage of LDH (193% and207% vs control, respectively), exhibiting values simi-lar to those produced by QUIN alone (Fig. 2). When co-incubated with QUIN + L-ARG, L-NARG producedvalues of LDH leakage not different from those in con-trol (Fig. 2). Finally, L-NARG alone had no effect onthe leakage of LDH (112 % vs control).

Effects of L-NARG on QUIN- and L-ARG-InducedLipid Peroxidation. Table I summarizes the actions ofL-NARG on QUIN- and L-ARG-induced lipid peroxi-

dation. Quinolinate alone significantly enhanced per-oxidation as compared to control. Also, L-ARG alone,at 200 uM, but not at 100 uM, significantly increasedlipid peroxidation as compared to controls; however,in combination with QUIN, the effects were not addi-tive. Alone, L-NARG had no effect on lipid peroxida-tion. When co-administered with QUIN, L-ARG orQUIN + L-ARG, L-NARG produced values of lipidperoxidation similar to those of controls. The additionto the media of L-ARG (100 uM) 30 min after thetreatment of the slices with QUIN + L-NARG hadno effect on TBARS as compared to controls, whereasthe addition of an excess of L-ARG (200 uM) underthe same experimental conditions completely reversedthe protective action of L-NARG on QUIN-inducedlipid peroxidation.

Effect of L-NARG on QUIN-induced Changes inGSH and GSSG. Table II shows the effect of L-NARGon QUIN-induced changes in the basal levels of GSHand GSSG. Significant changes in the levels of bothGSH and GSSG were observed after incubation of thestriatal slices with QUIN, as compared to controls.Consequently, the GSH/GSSG ratio was also affectedby the toxic action of QUIN when compared to controlvalues. All these changes were prevented by addition tothe incubation media of L-NARG. No effects of thisdrug (as compared to controls) either on GSH or GSSG

Nco-Nitro-L-Arginine Antagonizes Quinolinic Acid-Induced Neurotoxicity 845

Table II. The Action of Nco-Nitro-L-Arginine (L-NARG) on Quinolinic Acid(QUIN)-Induced Changes in Glutathione Levels of Rat Striatal Slices

ControlQUIN (100 uM)L-NARG (100 uM)QUIN + L-NARG

nmol/mg prot (mean ± SEM)

GSH

172 ±6116±4*168 ±7179 ±5

GSSG

8± 110± 1*7 ± 38 ± 1

GSH/GSSG ratio

21.50 ± 1.0212.04+0.65*20.14+1.2622.11 ±0.86

Both reduced (GSH) and oxidized (GSSG) forms of glutathione were measuredin probes containing five striatal slices each. Tissue was added with EDTA-phosphate buffer plus HPO3 and incubation was done in a shaking water bath at37°C for 60 minutes in the presence of QUIN, L-NARG or their combination.O-phthalaldehyde (OPA) was used for the fluorometric assay of both GSH andGSSG. For GSSG measurement, an additional incubation was done in the pres-ence of N-ethylmaleimide (NEM). Error bars represent SEM of 6 experiments.*P < 0.05 compared to control.

Table I. The Action of Nco-Nitro-L-Arginine (L-NARG) on Quinolinic Acid (QUIN)-and L-Arginine (L-ARG)-Induced Lipid Peroxidation in Rat Striatal Slices

ControlQUIN (100 uM)L-ARG (100 uM)L-ARG (200 uM)L-NARG (100 uM)QUIN + L-ARG (100)QUIN + L-NARGL-ARG (100) + L-NARGQUIN + L-NARG + L-ARG (100)

(All simultaneously added to the medium)QUIN + L-NARG + L-ARG (100)

(L-ARG added 30 min after QUIN and L-NARG)QUIN + L-NARG + L-ARG (200)

(L-ARG added 30 min after QUIN and L-NARG)

nmol TBARS/mg prot

(mean ± S.E.M.)

0.062 ± 0.0010.1 26 ±0.025a

0.093 ± 0.0050.15210.016b

0.059 ± 0.0040.12710.007a

0.067 ± 0.0020.064 ± 0.0070.071+0.004

0.076 ± 0.006

0.146+0.013b

% vs Control

10020315024595

205108103115

123

235

Lipid peroxidation was measured by the assay of thiobarbituric acid (nmoles of TBA-reactivesubstances per mg of protein) in probes containing five striatal slices each. Tissue was added with2 ml of Ringer solution and incubation was done in a shaking water bath at 37°C for 60 minutesin the presence of QUIN, L-ARG or their combination. Error bars represent SEM of 5 to 8 ex-periments. aP < 0.05 and bP < 0.01, compared to control.

846 Santamaria, Espinoza-Gonzalez, Rios, and Santamaria

levels were observed when administered alone to themedia.

DISCUSSION

We have previously shown that L-NARG, a se-lective inhibitor of NOS, can reduce the oxidative in-jury produced by QUIN in vitro in synaptosomal frac-tions prepared from whole rat brain (17). In order toprovide further evidence of a possible participation ofNO in the pattern of toxicity elicited by QUIN in vitro,

in this study we tested the effect of L-NARG on dif-ferent markers of QUIN-induced toxicity in rat striatalslices. We found that QUIN and L-ARG are both ableto induce significant leakage of LDH. Interestingly,the coadministration of QUIN plus L-ARG had no ad-ditive effect, suggesting that both drugs could be shar-ing the same mechanisms to exert their toxicity. Thecell damage induced by QUIN was completely pre-vented by the addition of L-NARG.

On the other hand, QUIN treatment alone, ac-cording to previous reports (5,6,7,17), resulted an im-portant lipid peroxidant in our experimental system.

Nw-Nitro-L-Arginine Antagonizes Quinolinic Acid-Induced Neurotoxicity 847

When treated with L-ARG alone and QUIN plusL-ARG, the slices exhibited enhanced levels of lipidperoxidation as compared to control slices. The mostremarkable finding of this series corresponded to thereversed effect: an excess of L-ARG antagonized theprotective action of L-NARG against QUIN-inducedlipid peroxidation when L-ARG was added to the in-cubation media 30 min after the addition of other treat-ments. These data suggest that both L-ARG andL-NARG are acting in opposite directions, but in thesame metabolic system, in a concentration-dependentmanner.

As another index of oxidative stress, we also mea-sured the levels of both reduced (GSH) and oxidized(GSSG) glutathione, as well as the GSH/GSG ratio.Changes on these markers were observed after treat-ment with QUIN alone whereas L-NARG preventedthese alterations mediated by QUIN. It is well knownthat the tripeptide glutathione fulfills a wide variety ofimportant physiological functions. One of them is thatof the reduction of various peroxides and free radicals(24), and this protection against oxygen toxicity re-sults in the conversion of GSH to GSSG. Therefore,the thiol/disulfide (GSH/GSSG) ratio has an importanteffect on the redox state of the protein thiols with mod-ulation of protein conformation and enzyme activity,and its alteration constitutes a suitable marker of ox-idative stress. Moreover, oxygen-derived free radicalsmay overwhelm the radical scavenging potential ofcells leading to lipid peroxidation and destruction ofcell membranes, which in turn can result in cellular de-pletion of GSH with subsequent increases of GSSG(24). The altered GSH/GSSG ratio by QUIN reflectsthe capability of this neurotoxin to exert its toxic ac-tion on the redox systems, affecting the primary an-tioxidant pathways by an NMDA receptor mediatedmechanism and supporting our previous findings onlipid peroxidation (5,6,17,18). In fact, while in our ex-perimental system it was expected that LDH might beleaking mainly from the striatal neurons (since theirwell-known susceptibility to the toxic and destructiveactions of QUIN over other cell types), GSH was ex-pected to be affected preferentially from astrocytes,since it has been shown that the concentration ofglutathione (both forms) is much higher in culturedastrocytes than in neurons (25). This finding also sug-gest a preferential susceptibility of neurons to be af-fected by oxidative stress by their lack of antioxidantmechanisms.

It has been suggested that QUIN-induced lipidperoxidation in rat brain synaptosomes is mediatedby NO, since some NOS inhibitors, such as L-NARG

and L-NAME may both singificantly reduce the ox-idative action of QUIN in vitro and in vivo (17,18).Our present results provide additional evidence to theproposal that the generation of oxidative stress andcellular damage after exposure to QUIN are related toeach other, and these toxic events are thought to bemediated by NO. Therefore, although NO has beenshown to exert either modulatory or neurotoxic ac-tions on the CNS, the toxic action of QUIN seems tobe strongly related to the production of NO. More-over, in vitro studies have demostrated that cells ex-posed to oxidative stress respond to peroxynitrite (atoxic molecule derived from NO) by releasing GSH,suggesting that NO-derived oxidants might involveseveral toxic mechanisms and the action of other mol-ecules such as O2, to produce oxidative stress. Fur-thermore, it is likely that NO, in a similar way to car-bon monoxide, eventually may exert its toxic actionby generation of free radicals, involving the disrup-tion of the mitochondrial electron transport, inhibitionof Na+/K+-ATPase, or inhibition of cytochrome A3

(4,11,12,27). Consequently, from our findings, NOcould be suggested to be responsible for the neuro-toxic effects induced by QUIN after the NMDA re-ceptor overactivation, acting itself as an oxidative andneurotoxic intermediate, or via other derivatives. In-deed, the NO-derived molecule, peroxynitrite, hasbeen reported to be responsible of a series of redoxchanges in cells, such as irreversible inhibition of mi-tochondrial respiration, decreased mitochondrial po-tential and proton leak at the respiratory chain, alsoinvolving membrane lipid peroxidation (26).

Results of this study lead us to propose that NOis an important mediator of cell injury and oxidativedamage associated with the complex pattern of toxi-city elicited by QUIN. Cell damage and oxidativestress produced by QUIN seem both to be closely re-lated during the first stages of acute neurotoxicity,involving the action of oxidative mediators. How-ever, further experiments must be carried out in orderto determine whether the cellular redox status can bealso affected by derivatives of NO, such as peroxyni-trite. Since QUIN has been successfully employed asan experimental model of Huntington's disease and isalso involved in several degenerative events (1,2),our fiindings assume pathophysiological relevance tothe proposal that the cell damage associated to thegeneration of free radicals or other reactive oxygenspecies observed in neurodegenerative disorders (i.e.,Huntington's and Parkinson's diseases) might even-tually involve the production of NO as an oxidativemediator.

848 Santamaria, Espinoza-Gonzalez, Rios, and Santamaria

ACKNOWLEDGMENTS

This work was supported by CONACyT Grant J28612-M andCONACyT's fellowship Grant 96114 (Mexico). The authors expresstheir gratitude to Leticia Andres-Martinez for her technical assistance.

REFERENCES

1. Stone, T. W. 1993. Neuropharmacology of quinolinic andkynurenic acids. Pharmacol. Rev. 45: 309-379.

2. Heyes, M. P., Saito K., Crowley, J. S., Davies, L. E., Demitrack,M. A., Der, M., Dilling, M. A., Elia, J., Kruesi, M. J. P., Lackner,A., Larsen, S. A., Lee, K., Leonard, H. L., Markey, S. P., Martin,A., Milstein, S., Mouradian, M. M., Pranzatelli, M. R., Quearry,B. J., Salazar, A., Smith, M., Strauss, S. E., Sunderland, T.,Swedo, S. W., and Tourtellotte, W. W. 1992. Quinolinic acid andkynurenine pathway metabolism in inflammatory and non-inflammatory neurological diseases. Brain 115: 1249-1273.

3. Schwarcz, R., Foster, A. C., French, E. D., Whetsell Jr., W. O.,and Kohler, C. 1984. Excitotoxic models for neurodegenerativedisorders. Life Sci. 35: 19-32.

4. Nathanson, J. A., Scavone, C., Scalon, C., and McKee, M. 1995.The cellular Na+ pump as a site of action for carbon monoxideand glutamate: a mechanism for long-term modulation of cellu-lar activity. Neuron 14: 781-794.

5. Rios, C., and Santamaria, A. 1991. Quinolinic acid is a potentlipid peroxidant in rat brain homogenates. Neurochem. Res. 16:1139-1143.

6. Santamaria, A., and Rios, C. 1993. MK-801, an N-methyl-D-aspartate receptor antagonist, blocks quinolinic acid-inducedlipid peroxidation in rat corpus striatum. Neurosci. Lett. 159:151-154.

7. Sti'pek, S., Stastny, F., Platenik, J., Crkovska, J., and Zima, T.1997. The effect of quinolinate on rat brain lipid peroxidation isdependent on iron. Neurochem. Int. 30: 233-237.

8. Bredt, D. S., and Snyder, S. H. 1990. Isolation of nitric oxidesynthase, a calmodulin-requiring enzime. Proc. Natl. Acad. Sci.U.S.A. 87: 682-685.

9. Garthwaite, J. 1991. Glutamate, nitric oxide and cell-cell sig-naling in the nervous system. Trends Neurosci. 14: 60-67.

10. Moncada, S., Palmer, R.M.J., and Higgs, E.A. 1989. Biosynthe-sis on nitric oxide from L-arginine: a pathway for the regulationof cell function and communication. Biochem. Pharmacol. 38:1709-1715.

11. Lipton, S. A., Choi, Y. B., Pan, Z. H., Lei, S. Z., Chen, H. S.V., Sucher, N. J., Loscalzo, J., Singel, D. J., and Stamler, J. S.1993. A redox-based mechanism for the neurodestructive ef-

fects of nitric oxide and related nitroso-compounds. Nature364: 626-632.

12. Snyder, S. H. 1993. Janus faces of nitric oxide. Nature 364: 577.13. Kashii, S., Mandai, M., Kikuchi, M., Honda, Y., Tamura, Y.,

Kaneda, K., and Akaike, A. 1996. Dual actions of nitric oxidein N-methyl-D-aspartate receptor-mediated neurotoxicity incultured retinal neurons. Brain Res. 711: 93-101.

14. Rubbo, H., Radi, R., Trujillo, M., Telleri, R., Kalyanaraman, B.,Barnes, S., Kirk, M., and Freeman, B. A. 1994. Nitric oxide reg-ulation of superoxide and peroxynitrite-dependent lipid peroxi-dation. J. Biol. Chem. 269: 26066-26075.

15. Garthwaite, J., Garthwaite, G., Palmer, R. M. J., and Moncada, S.1989. NMDA receptor activation induces nitric oxide synthesisfrom arginine in rat brain slices. Eur. J. Pharmacol. 172: 413-416.

16. Kiedrowski, L., Costa, E., and Wroblewski, J. T. 1992. Glutamatereceptor agonists stimulate nitric oxide synthase in primary cul-tures of cerebellar granule cells. J. Neurochem. 58: 335-341.

17. Santamaria, A., Santamaria, D., Diaz-Munoz, M., Espinoza-Gonzalez, V., and Rios, C. 1997. Effects of NO-nitro-L-arginineand L-arginine on quinolinic acid-induced lipid peroxidation.Toxicol. Lett. 93: 117-124.

18. Perez-Severiano, F., Escalante, B., and Rios, C. 1998. Nitricoxide synthase inhibition prevents acute quinolinate-inducedstriatal neurotoxicity. Neurochem. Res. 23: 1297-1302.

19. Wroblewski, F., and LaDue, J. S. 1995. Lactic dehydrogenaseactivity in blood. Proc. Soc. Exp. Biol. Med. 90: 210-213.

20. Hissin, P. J., and Hilf, R. 1976. A fluorometric method for de-termination of oxidized and reduced glutathione in tissues.Anal. Biochem. 74: 214-226.

21. Shukla, G. S., Srivastava, R. S., and Chandra, S. V. 1988. Pre-vention of cadmium-induced effects on regional glutathione sta-tus of rat brain by vitamin E. J. Appl. Toxicol. 8: 355-358.

22. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.1951. Protein measurement with the Folin-phenol reagent. J. Biol.Chem. 193: 265-275.

23. Steel, R. G. D., and Torrie, J. H. 1980. Principles and proce-dures of statistics. McGraw-Hill, New York, NY.

24. Harvey, P.R.C., Ilson, R.G., and Strasberg, S.M. 1989. The si-multaneous determination of oxidized and reduced glutathionesin liver tissue by ion pairing reverse phase high performance liq-uid chromatography with a coulometric electrochemical detec-tor. Clin. Chim. Acta 180: 203-212.

25. Raps, S. P., Lai, J. C. K., Hertz, L., and Cooper, A. J. L. 1989.Glutathione is present in high concentrations in cultured astro-cytes but not in cultured neurons. Brain Res. 493:398-401.

26. Brookes, P. S., Land, J. M., Clark, J. B., and Heales, S. J. R.1998. Peroxynitrite and brain mitochondria: evidence for in-creased proton leak. J. Neurochem. 70: 2195-2202.

27. Zhang, J., and Piantadosi, C. A. 1992. Mitochondrial oxidativestress after carbon monoxide hypoxia in the rat brain. J. Clin. In-vest. 90:1193-1199.