[Back to Number 7 ToC] [Back to Journal Contents] [Back to Biokhimiya Home page]

REVIEW: Nitric Oxide in Mechanisms of Brain Damage Induced by Neurotoxic Effect of Glutamate

V. G. Bashkatova* and K. S. Rayevsky

Institute of Pharmacology, Russian Academy of Medical Sciences, Baltiiskaya ul. 8, Moscow, 125315 Russia; fax: (095) 151-1261; E-mail: ksrayif@glasnet.ru

* To whom correspondence should be addressed.

Received February 26, 1998
Nitric oxide (NO) is a gaseous chemical messenger which plays the role of a universal modulator of various physiological functions of animals including the nervous system, i.e., interneuronal communications, synaptic plasticity, memory formation, receptor functions, intracellular signal transmission, release of neurotransmitters. The possible role of NO is considered in some basic diseases of the central nervous system which are associated with the neurotoxic effect of glutamate (ischemia, stroke, convulsive disorders, etc.). NO is believed to be a key pathophysiological factor of these diseases. The production of NO in the brain cortex of rats was found to significantly increase during convulsions of various origin.
KEY WORDS: nitric oxide, NO-synthase, glutamate receptors, brain ischemia, convulsions, lipid peroxidation, EPR spectroscopy

Abbreviations: NO) nitric oxide; NMDA) N-methyl-D-aspartate; NMDLA) N-methyl-DL-aspartate; EPR) electron paramagnetic resonance; MNFe-DETC) iron mononitrosyl--diethyldithiocarbamate complex; MES) maximal electric shock; NOS) NO-synthase; L-NNA) N-nitro-L-arginine; TSC) thiosemicarbazide; GABA) gamma-aminobutyric acid.

The finding of the polyfunctional physiological regulator nitric oxide (NO) is one of the most significant achievements in biology during the last decade [1, 2]. According to present conceptions, NO is a gaseous chemical free radical messenger which plays the role of a universal modulator of various physiological functions of animals including the central nervous system. Interest in the probable role of NO as a neuronal messenger appeared after the finding of NO production by the granular cerebellum cells in response to exposure to agonists of glutamate receptors in vitro [3, 4]. It is also known that the production of nitroxyl radical along with other reactive oxygen forms (superoxide-anion, hydroxyl radical) is the main link in the destruction of biological membranes of the nervous cells under conditions of ischemia, trauma, convulsions, etc. [5].

NO has been recently established to play a pathogenic role in neurogenerative diseases of the central nervous system such as ischemia, brain stroke, epilepsy, and other convulsive disorders, Parkinson's disease, etc. [6-9]. It was shown that NO was involved in the formation of neuronal memory, in particular, in the initiation of long-term potentiation in the hippocampus and modulated the synaptic neurotransmission [2] including the functions of NMDA-subtype glutamate receptors [10].


The development of various trends in science, including immunology, physiology, and pharmacology of the cardiovascular system, toxicology, neurobiology, etc. revealed a universal regulatory molecule, NO, which possessed features of a biological messenger. The role of NO as biological messenger is first of all determined by its physicochemical properties. NO is a short-lived reactive free radical of high lability [11, 12]. It was long thought that the brain functions depended on two neurotransmitter types: exciting and inhibiting; these include acetylcholine, biogenic monoamines, amino acids, and neuropeptides [10].

NO is considered to be the first of a new family of signal molecules which possess neurotransmitter properties [2, 13-16]. Unlike the traditional neurotransmitters, NO is not reserved in synaptic vesicles of nerve endings but is released into the synaptic cleft by free diffusion, but not via exocytosis. The NO molecule is synthesized in response to physiological demand via the enzyme NO-synthase (NOS) from its biological precursor L-arginine. Therefore, precisely the synthesis of NO seems to be the main key in the regulation of the activity of this messenger. The biological efficiency of NO significantly depends on the small size of the NO molecule and on its high reactivity and diffusibility in tissues including the central nervous system. Based on this, NO was called a retrograde messenger [17].

Earlier the short lifetime of NO of some seconds made difficult its quantitative determination in tissues and, particularly, studies on localization of NO-producing neurons in the mammalian brain. Therefore, a neuronal isoform of NO-synthase (nNOS) was given special attention. The localization of nNOS in the brain was studied in detail by immunohistochemical methods and by in situ hybridization [1]. First of all, nNOS was found to be colocalized with somatostatin, neuropeptide Y, or gamma-aminobutyric acid (GABA) in the glutamatergic granular cells, GABA-ergic basket cerebellum cells, and cortex neurons of the brain. A similar pattern was found in the striatum. It should be noted that NADPH-diaphorase was also found in the neurons along with NO-synthase [18, 19]. A special treatment of tissue permitted us to selectively label the nNOS-containing neurons. Such neurons are about 1-2% of the total cell population in the brain cortex and striatum. Pyramidal cells in the CA1 hippocampus area virtually do not contain the neuronal form of NO-synthase; however, a significant content of the endothelial enzyme form (eNOS) was found in this area and also in the CA3 area and in the dentate gyrus granular cells.


Mechanisms of NO biosynthesis have been studied in detail during recent years [15, 20]. Although NO can be produced by nearly all tissues of the body, under normal conditions the content of NOS is the highest in the brain. Processes of NOS-containing neurons are so widely ramified that virtually all neurons of the central nervous system are located not farther than in some micrometers from a source of NO [21]. The olfactory bulb and cerebellum were found to be the most productive sources of NO [22].

In humans and other animals, NO is produced from L-arginine by the two-step enzymatic oxidation of its guanidine group with an intermediate production of NG-hydroxy-L-arginine. The hydroxylation takes place with the involvement of the coenzyme tetrahydrobiopterin, depends on calcium-calmodulin, and is inhibited by carbon monoxide. During the next step of the reaction, a free radical NO and L-citrulline are produced stoichiometrically. A number of NOS isoforms has been described: a constitutive one which is permanently present in tissues (cNOS) and an inducible one (iNOS); moreover, neuronal (nNOS), endothelial (eNOS), and macrophagal (macNOS) forms of NO-synthase are classified according to their localization. The first two forms are mainly constitutive ones, whereas the latter is an inducible form of NOS [15].

The dependence of cNOS on calcium-calmodulin suggests that the enzyme can produce NO in response to stimulation of receptors, such as the activation of the NMDA-receptors in brain ischemia or exposure to agents which increase the level of intracellular Ca2+. cNOS has been found mainly in nerve and endothelial cells [23]. The inducible enzyme form (iNOS) is produced, as a rule, in response to activation of the cells by cytokines. This form was found in macrophages, endothelial cell, hepatocytes, and smooth muscles [15].

The first attempts to obtain purified NOS were unsuccessful because the enzyme rapidly lost its activity during the purification. The purified enzyme was obtained from the brain (nNOS) after the finding of its dependence on calmodulin [1, 15]. A short time later NOS was obtained not only from the brain but also from the endothelium (eNOS) and macrophages (macNOS) [22]. The structure of NOS indicates that a set of regulatory mechanisms exists. A large number of NOS inhibitors is known, and, first of all, these are N-substituted derivatives of L-arginine, i.e., structural analogs of the natural substrate of the enzyme. Nitro- and alkylderivatives of L-arginine seem to be the best known analogs. In the presence of excess L-arginine the inhibitory effect is decreased, and this is evidence of the competitive mechanism of inhibition.

Individual NOS isoforms differ in their sensitivity to inhibitors: thus, NG-hydroxy-L-arginine is a powerful inhibitor of neuronal and endothelial enzymes with Ki of 200-500 nM, whereas NG-amino-L-arginine is known to be one of the most efficient inhibitors of the macrophagal enzyme (Ki of 1-5 nM). Precisely the highly specific inhibitors to individual NOS isoforms are suggested to be of interest in medicine. Since NO is likely to be involved in the pathogenesis of various diseases of the central nervous system, intensive studies on some of the enzyme inhibitors were started, e.g., probable neuroprotectors in ischemia, anticonvulsive agents in epilepsy models, and under some other conditions. Recently a new group of selective inhibitors of nNOS was described--derivatives of 7-nitroindazole [24]. 7-Nitroindazole was reported to selectively inhibit the neuronal NOS isoform (IC50 = 0.71 ± 0.01 µM for rat cerebellum homogenate) along with antinociceptive and anticonvulsive effects [8]. No hypertensive effect (which is associated with the effect of other inhibitors of NOS) was observed.

Nitric oxide is inactivated during the interaction with hemoglobin and other iron-containing compounds. On the other hand, vasodilators such as sodium nitroprusside and nitroglycerol are donors of NO, and their effect is suggested to be associated with the production of NO.

Ca2+ and calmodulin play an important role in the regulation of neuronal and endothelial NOS, whereas macrophagal NOS is insensitive to them. Thus, the stimulation of brain glutamate receptors by NMDA initiates the entrance into the cell of Ca2+ which binds to calmodulin and activates NOS. Consequently, the production of NO is an intracellular step in the transmission of glutamatergic signal. The interaction of NO with guanylate cyclase is the next step [17]. The primary effect of NO manifests itself in complexation with iron of guanylate cyclase with the resulting activation of the enzyme and increased production of cGMP. The increased content of the latter activates cGMP-dependent protein kinases which catalyze the phosphorylation of various protein substrates. A local injection of NMDA into the hippocampus was shown to induce the death of neurons, i.e., a neurotoxic effect took place which could be prevented by inhibitors of NOS [9]. The pathogenic role of NO under conditions of this model was indirectly confirmed by the neurotoxic effect which was also found after the injection of metabolic donors of NO, such as sodium nitroprusside. The neuronal and endothelial forms of NOS are sensitive to inhibitors of calmodulin, in particular, to phenothiazine neuroleptics aminazine and trifluoperazine (IC50 = 10 µM).

On the contrary, inducible NOS of both macrophagal and other origin was neither stimulated by Ca2+ nor inhibited by inhibitors of calmodulin despite the presence of the calmodulin-recognizing site.

Phosphorylation under the influence of some protein kinases is important in the regulation of NOS. Thus, nNOS is phosphorylated by cAMP-dependent protein kinase, protein kinase C, cGMP-dependent protein kinase, and Ca2+-calmodulin-dependent protein kinase. The phosphorylation decreases the enzyme activity [25, 26]. On the other hand, a protein phosphatase calcineurin dephosphorylated NOS with a resulting increase in its catalytic activity [27]. Thus, the phosphorylation of NOS regulates its activity at many levels. Moreover, various coenzymes, in particular, tetrahydrobiopterin, are also involved in the regulation of the enzyme activity [20].

As mentioned, the neuronal and endothelial isoforms of NO-synthase (nNOS and eNOS) are in most cases of the constitutive type and their activation does not require the synthesis of new enzyme protein. However, protein synthesis can occur under some pathologic conditions, e.g., trauma or stroke [28]. Human genes for nNOS and eNOS are located in the 12th and 7th chromosomes, respectively [29].


An exciting neurotransmitter glutamate and its structural analogs which specifically bind to various subtypes of glutamate receptors are known to be in some cases neurotoxic due to hyperactivation of receptor--channel complexes and resulting damage, degeneration, and death of neurons [30, 31]. The stimulation of NMDA-subtype receptor by an excess of glutamate initiated a reaction cascade including an increase in the intracellular calcium content, generation of NO, activation of guanylate cyclase, increased synthesis of cGMP and lipid peroxidation (LPO), and, finally, the death of neurons and development of stable disorders in brain activity. This so-called glutamate toxicity [32] provides the basis for frequent neurodegenerative diseases of the central nervous system including ischemic stroke, epilepsy, Parkinson's and Alzheimer's diseases, etc.

Interest in studies on the role of NO in the neurotoxic effect of glutamate and its analogs especially increased during recent years [2, 33-36]. The interaction of NO with reactive oxygen forms produces compounds of high toxicity, in particular, peroxynitrite, and this results in the damage and death of neurons.


The activation of NMDA-subtype glutamate receptors in cerebellum granular cell culture is known to cause the release of NO from these cells [3]. A local injection of NMDA into the rat hippocampus is also known to induce the death of neurons, i.e., glutamate toxicity was observed [9]. Donors of NO--sodium nitroprusside, (+)-S-nitroso-N-acetyl-penicillamine (SNAP), S-nitrosoglutathione (SNOG)--produced a similar effect.

The association of NO with the neurotransmitter function of glutamate was the reason for studies in detail on the possible involvement of NO in pathophysiological mechanisms of disorders in the brain blood flow and, first of all, of brain ischemia with subsequent development of stroke [28, 36]. The endothelial NOS which is also present in the brain was shown to participate in these mechanisms along with the neuronal NOS. Both enzymes are similar in properties but differ in their localization: nNOS is a cytosolic and eNOS a membrane-bound protein [37]. With deficiency of L-arginine, nNOS produced superoxide-anion and hydrogen peroxide which are neurotoxic in ischemia [38]. The inducible form of NOS is likely to be still more important in pathophysiological mechanisms of ischemia and then in the development of stroke because this form is activated by cytokines and produce an excess of NO which is toxic for neurons [2, 20]. The cytotoxic effect of NO seems not to depend on guanylate cyclase and the generation of cGMP [39]. Inhibitors of NOS were found to manifest a protective effect in ischemia. The vasodilating activity of NO is also known to decrease in the presence of superoxide-anion [40], whereas superoxide dismutase (SOD) increased the cytotoxicity of NO in microglia cell culture. The vascular endothelium was also shown to regulate the effects of NO through the generation of superoxide-radical [41]. This phenomenon has not been explained.

The vasodilating activity of NO, nitro compounds, and NO donors has been studied in detail including an effect on the brain blood flow. The results of these studies are reviewed elsewhere [28]. Inhibitors of NOS induced the contraction of brain vessels, and in some cases their injection was accompanied with a decrease in the basal brain blood flow [42]. The problems related to the involvement of NO in the regulation of the blood flow are beyond the scope of this paper and are the subject of special studies. Let us consider briefly only studies on the use of NO donors and NO-synthase inhibitors as possible modulators of brain ischemic lesions under model conditions of disorders in the brain blood flow which result in ischemia with the subsequent death of neurons.

The increased production of NO during ischemia associated with hyperactivation of NMDA-subtype glutamate receptors can influence the ischemic situation either negatively (due to increased production of free radicals of peroxynitrite type) or positively. The latter is explained not only by the direct vasodilation of brain blood vessels but also by a possible retrograde blockade of NMDA-receptors by NO [43].

Direct measurements of NO level in brain tissue during ischemia caused by a bilateral occlusion of the carotid arteries followed by reperfusion were carried out using an original method of EPR-spectroscopy which was based on the use of diethyldithiocarbamate as a scavenger of the produced NO [44]. These measurements showed a significantly increased content of NO in the brain cortex under conditions of ischemia--reperfusion [45]. The use of NO donors, such as sodium nitroprusside or 3-morpholinosydnonimine increased the local brain flow and decreased the volume of the resulting infarction zone [46]. On the contrary, the injection of a natural precursor of NO, L-arginine, failed to produce such an effect [47].

Contradictory results were obtained with the use of NOS inhibitors, in particular, of N-nitro- and N-methylated analogs of L-arginine as neuroprotectors. A single use of a NOS inhibitor, L-NAME, in the model of focal brain ischemia in rats decreased the infarction zone [47]. However, the opposite results were obtained in other studies although they were performed under similar conditions [9, 48]. It was suggested that the effect of NO was favorable during the first period of ischemia (from some data, up to 8 min) but became neurotoxic later [47].

Two phases of NO generation are suggested to exist under ischemia--reperfusion conditions. The increased production of NO during the ischemia seems to be a physiological response which is associated with an increase in the blood flow. On the other hand, the production of NO during reperfusion can be accompanied by tissue damage [49].

A fundamentally new concept in studies on the molecular mechanism of brain ischemia was developed due to use of model mutant animals which lack one or the other NOS isoform. Thus, mice with inherited deficiency of nNOS were found to be more resistant to the brain ischemia than the control mice. This finding has confirmed the suggestion on the determining role of NO which was produced under the influence of the neuronal NO-synthase in the pathogenesis of ischemic brain infarction.


The participation of NO in mediating various functions of the central nervous system under both normal and pathological conditions including the glutamate toxicity [2, 49] and the related convulsive states is now doubtless [13, 35, 50]. It should be noted that data on the role of NO in pathophysiological mechanisms of convulsive disorders are inconsistent: some authors believe NO to manifest an anticonvulsive activity [51, 52], while others think it to be a proconvulsant [53, 54]. This difference in the results seems to be caused by different conditions during the experiments and the diversity of approaches, doses, animal species, etc. [8, 51, 52]. Thus, a precursor of NO, L-arginine, was used in various studies at the dose range from 10 [51] to 750 mg/kg [55]. Doses of L-NNA, the competitive but non-selective inhibitor of nNOS were varied from 1 to 250 mg/kg [56, 57]. In some cases other inhibitors of NOS were used, in particular, 7-nitroindazole which is a selective inhibitor of the neuronal isoform. Thus, there is no unified concept on the role of NO in the pathogenesis of convulsive disorders. Some authors consider NO to be an endogenous neuroprotector [58, 59] or anticonvulsant [56]. This opinion is confirmed by the finding that repeated injections of L-NNA into rats during four days inhibited NOS and this inhibition was accompanied by potentiating of convulsive attacks which were induced by kainic acid or pilocarpine, i.e., by convulsants of different mechanism [8, 52]. The authors concluded that NO regulated the initiation of convulsive activity expansion. As a retrograde messenger, NO is believed to initiate a cascade of reactions which prevent the expansion of convulsive activity, that is, NO is considered to be an endogenous anticonvulsant [60]. According to other data, NO is a proconvulsant that may exert a neurotoxic effect [53]. The generation of NO was also reported to significantly increase the production of free radicals which dramatically increase the neurotoxic effect [11, 61].

There are some data on the involvement of NO in the development of epileptiform convulsions [19, 62], and, particularly, the proconvulsive effect of kainic acid which is an agonist of glutamate receptors was accompanied by an increased NO content in the brain [8]. It should be noted that, as a rule, indirect approaches were used in the works under consideration, such as the use of metabolic precursors (L-arginine), donors of NO (sodium nitroprusside, etc.), or inhibitors of NOS. Until recently, the lack of an adequate method of quantitative determination of NO in tissues because of its short life-time (5-6 sec) was the fundamental difficulty.

The purpose of our work was to study the possible involvement of NO into pathophysiological mechanisms of appearance and development of convulsions of various origin. The direct quantitative method developed by Vanin et al. [44] was used to evaluate the rate of NO production in rat brain by its incorporation in complexes with bivalent iron and diethyldithiocarbamate (Fe2+--DETC) with the resulting paramagnetic iron mononitrosyl complexes with DETC (MNFe--DETC). These complexes are seen as an EPR-signal of g-factor 2.035 and 2.0 and a triplet hyperfine structure at g. To obtain MNFe--DETC in the organism, rats were simultaneously injected with an aqueous solution of Na-DETC (Sigma, USA) at the dose 500 mg/kg and a solution of Fe2+-citrate. Thirty minutes after the injection of the trap, the rats were decapitated, the brain was isolated immediately, the brain cortex was placed into a glass tube and frozen in liquid nitrogen to be analyzed afterwards by EPR-spectroscopy. EPR signals of the samples were recorded using a Radiopan radiospectrometer (Poland). Under these conditions, in the brain tissue a distinct signal was recorded which represented the content of MNFe--DETC complexes, i.e., of nitric oxide [8, 25, 63-65].

Various models of convulsive states were used as follows: maximal electroshock (MES), convulsions induced by thiosemicarbazide (TSC) which is an inhibitor of glutamate decarboxylase and causes a gradual decrease in the brain content of the inhibitory neurotransmitter GABA, pentylentetrazole-induced attacks, and also convulsions due to activation of NMDA-subtype of glutamate receptors after the direct injection of NMDA into the brain ventricles.

In all cases specific convulsive manifestations were observed in animals, mainly clonic-tonic type of attacks. In the control group of animals a weak signal was recorded which corresponded to the NO content of 1.65 nmoles/g wet tissue for 30 min. On the height of convulsive attack induced by MES (tonic phase of convulsions) or by injection of pentylentetrazole, TSC, or NMDLA, the content of NO in the rat brain cortex was 3-5 times higher than in the control [63-65]. On average, under conditions of MES, the content of NO produced in the brain cortex for 30 min was 5 ± 1 as compared to 1.5 ± 0.5 nmoles/g wet tissue in the brain cortex of the control rats. During pentylentetrazole-induced clonic-tonic convulsions (120 mg/kg) the content of NO in the rat brain cortex increased ~5 times versus the control [64]. On the height of clonic convulsions induced by the intraventricular injection of NMDLA (28.8 µg in 5 µl), the content of NO in the brain cortex increased more than fourfold [66, 67]. During TSC-induced convulsions, the content of NO (6.0 ± 1.0 nmoles/g) was also significantly higher than in the control. These results are in agreement with findings of other authors who used similar approaches to determine the content of NO in the brain during convulsions which were induced by kainic acid, an agonist of non-NMDA type glutamate receptors [8].

As a whole, our results indicate that convulsive attacks of various origin were accompanied by a significantly increased content of MNFe--DETC complexes, i.e., by the increased production of NO independently of the convulsion-inducing agent. Our results do not allow us to conclude whether the increased generation of NO from L-arginine in the rat brain cortex was associated with the activation of constitutive NOS of neuronal or endothelial origin. It should be noted that the injection to animals of bivalent iron ions was required to produce MNFe--DETC complexes in the brain cortex. These complexes were not recorded in both the control and experimental animals after the injection of only DETC, and this is consistent with the previous findings [44, 64].

The mean content of NO during clonic convulsions was not significantly higher than during the tonic attacks, thus, there was no direct relation between the convulsion type (tonic or clonic attack) and the level of increase in the content of NO in the rat brain cortex.

To analyze the observed increase in the brain content of NO, a competitive inhibitor of NO-synthase, L-NNA, was used at the dose of 250 mg/kg. Under these conditions, no increase in the brain content of NO was found although clonic convulsions took place in these animals. The injection of L-NNA completely prevented an increase in NO content after MES or injections of pentylentetrazole, TSC, or NMDLA [63, 64, 67].

The pentylentetrazole model was studied in detail because it represented the convulsive attack of different molecular mechanism which was associated with the blockade of chlorine channel of the GABA-A--receptor complex. To study the possible involvement of NO in the pathophysiological mechanism of the convulsive attack, this model was used to study effects of the NO precursor L-arginine and of a NOS inhibitor L-NNA. The appearance of distinct signs of the convulsive attack was recorded. L-Arginine (300 mg/kg) failed to significantly influence the convulsive attacks. L-NNA (10 mg/kg) significantly decreased the latency of the first twich onset which corresponded to the moment of convulsive electric discharge in the motor cortex in response to pentylentetrazole. And the appearance of clonic convulsions was significantly (twofold) delayed. The inhibitor at the dose of 250 mg/kg significantly increased the latent period of the true clonus. The same dose of L-NNA postponed the moment of onset of tonic convulsions [63].

But the interpretation of these results is rather difficult because the inhibition of NO-synthase was shown to increase convulsions of various origin [11], and this is in agreement with our finding of the decreased latency of the first twich while inhibiting NO-synthase. It is likely that the protective, especially anticonvulsive effect of NO took place only at the initial trigger moment of the convulsive attack and failed to influence the subsequent steps of the generalization of the attack. It also seems that L-NNA somehow suppressed the excitability of neurons and thus inhibited the propagation of convulsive activity. There is a report in the literature on the neuroprotective properties of L-NNA [46].

Experiments were carried out to reveal any correlation between the drug-induced suppression of various convulsive signs, on one hand, and the inhibition of NO production and LPO activation, on the other hand. The following drugs were studied: chemically different anticonvulsants, a noncompetitive antagonist of NMDA-receptors disocilpin (MK-801) and L-NNA. The drug efficiency was determined in the preventing successive clonic and tonic components of a convulsive attack and also in increasing the latency of convulsion onset. The effect of lamotrigine is associated with the inhibition of the presynaptic release of glutamate, and this drug at the dose of 20 mg/kg completely prevented tonic convulsions and decreased the intensity of clonic convulsions. An injection of phenobarbital completely prevented all manifestations of convulsive attacks. The other anticonvulsant, carbamazepine (15 mg/kg), significantly decreased the intensity of all the main components of the convulsive attack. The inhibitor of NO-synthase, L-NNA, at the dose of 250 mg/kg significantly weakened convulsions and completely abolished the tonic phase of the attack [63, 64]. To analyze the involvement of NMDA-subtype glutamate receptors into the genesis of convulsive states and the associated neurotoxicity due to direct activation of these receptors, NMDLA was injected into the brain because such injection is known to be accompanied by the development of repeated convulsive clonic attacks. The noncompetitive antagonist of NMDA-receptors disocilpin at dose 1 mg/kg completely inhibited these convulsions. An injection of L-NNA (250 mg/kg) significantly weakened the convulsions but failed to completely arrest them [67].

Along with suppression of convulsions, the anticonvulsants more or less prevented the concomitant increase in the production of NO but their effects were different: carbamazepine and lamotrigine were rather efficient although they failed to completely inhibit the production of NO specific for convulsive attacks. Thus, lamotrigine decreased the production of NO by more than 50%, whereas the more potent anticonvulsant carbamazepine decreased the production of NO still less. Phenobarbital completely prevented an increase in the level of NO in rat brain. Disocilpin (1 mg/kg) also completely inhibited the increase in NO production after the injection of NMDLA in the rat brain ventricles, and this indicates that receptors were involved into the increase in NO content during the convulsive attack [67].

Thus, our findings in total are consistent with the modern concept of the trigger role of NO generation at convulsive states [8]. As a retrograde messenger, NO is suggested to start the reaction chain which limits the expansion of the convulsive activity and thus be an endogenous anticonvulsant [60]. In fact, the observed decrease in the latent period of the first convulsive component of the corazole-induced attack under the inhibition of NO-synthase and the absence of NO production can be explained by the prevention of retrograde blockade by the last of the NMDA-receptors which participated in the initiation of convulsions [43]. Molecular mechanisms underlying the generation of NO are not quite elucidated. It is likely that the initial steps of the "trigger" reaction are common and, in particular, include the activation of the NMDA-subtype glutamate receptors which are abundant in various brain structures and participate in the initiation of reactions which promote the increased entrance of Ca2+ into the cell, the activation of guanylate cyclase, and the generation of free radicals with the subsequent neurotoxic effect [16, 61].

This work was supported by the Russian Foundation for Basic Research (Grant No. 95-04-1861) and INTAS (Grant No. 94-500).

The authors are grateful to Dr. V. B. Narkevich for help in the preparation of this paper.


1.Bredt, D. S., Glatt, C. E., Hwang, P. M., Fotichi, M., Dawson, T. M., and Snyder, S. H. (1991) Neuron, 7, 615-624.
2.Moncada, S., Palmer, R. M., and Higgs, E. A. (1991) Pharmacol. Rev., 43, 109-142.
3.Garthwaite, J., Charles, S. L., and Chess-Williams, R. (1988) Nature, 336, 385-388.
4.Garthwaite, J., Garthwaite, G., Palmer, R. M., and Moncada, S. (1989) Eur. J. Pharmacol., 172, 413-416.
5.Hall, E. D., and Braughler, J. M. (1989) Free Rad. Biol. Med., 6, 303-313.
6.Youdim, M. B. I., Lavie, L., and Riederer, P. (1994) in The Neurobiology of NO and OH (Chiueh, C. C., Gilbert, D. A., and Colton, C. A., eds.) The New York Academy of Sciences, New York, pp. 64-68.
7.Mitrovic, B., Pierre, B. A. S., Mackenzie-Graham, A. J., and Merilli, J. E. (1994) in The Neurobiology of NO and OH (Chiueh, C. C., Gilbert, D. A., and Colton, C. A., eds.) The New York Academy of Sciences, New York, pp. 64-68.
8.Mulsch, A., Busse, R., Mordvintsev, P., Vanin, A. F., Nielsen, E. O., Scheel-Kruger, J., and Olesen, S. P. (1994) Neuroreport, 5, 2325-2328.
9.Moncada, S., Lekieffre, D., Arvin, B., and Meldrum, B. (1992) Neuroreport, 3, 530-532.
10.Cooper, J. R., Bloom, F. E., and Roth, R. H. (1996) The Biochemical Basis of Neuropharmacology, Oxford University Press, New York, pp. 126-458.
11.Boldyrev, A. A. (1995) Neirokhimiya, 12, 3-13.
12.Reutov, V. P. (1995) Usp. Biol. Khim., 35, 189-228.
13.Rayevsky, K. S. (1997) Byull. Eksp. Biol. Med., 123, 484-490.
14.Schuman, E. M. (1995) in Nitric Oxide in the Nervous System (Vincent, S., ed.) Academic Press, London, pp. 125-150.
15.Bredt, D. S., and Snyder, S. H. (1994) Ann. Rev. Biochem., 63, 175-195.
16.Garthwaite, J. (1991) Trends Pharm. Sci., 14, 60-67.
17.Dawson, T. M., and Dawson, V. L. (1994) The Neuroscientist, 9, 9-20.
18.Dawson, T. M., Bredt, D. S., Fotuhi, M., Hwang, P. M., and Snyder, S. H. (1991) Proc. Natl. Acad. Sci. USA, 88, 7797-7801.
19.Hope, B. T., Michael, G. J., Knigge, K. M., and Vincent, S. R. (1991) Proc. Natl. Acad. Sci. USA, 88, 2811-2814.
20.Marletta, M. A. (1993) J. Biol. Chem., 268, 12231-12234.
21.Bredt, D. S., Hwang, P. M., and Snyder, S. H. (1990) Nature (London), 347, 768-770.
22.Varner, P. D., and Beckman, J. S. (1995) in Nitric Oxide in the Nervous System (Vincent, S., ed.) Academic Press, London, pp. 191-206.
23.Schuman, E. M., and Madison, D. V. (1994) Ann. Rev. Neurosci., 17, 153-183.
24.Moore, P. K., Wallace, P., Gaffen, Z., Hart, S. L., and Babbedge, R. C. (1993) Br. J. Pharmacol., 110, 219-224.
25.Mikoyan, V. D., Kubrina, L. N., and Vanin, A. F. (1994) Biofizika, 39, 915-918.
26.Bredt, D. S., Ferris, C. D., and Snyder, S. H. (1992) J. Biol. Chem., 267, 10976-10981.
27.Dawson, T. M., Steiner, J. P., Dawson, V. L., Dinerman, J. L., Uhl, G. R., and Snyder, S. H. (1993) Proc. Natl. Acad. Sci. USA, 90, 9808-9812.
28.Faraci, F. M., and Brian, J. E. (1994) Stroke, 25, 692-703.
29.Nathan, C., and Xie, Q.-W. (1994) J. Biol. Chem., 19, 13725-13728.
30.Rayevsky, K. S. (1990) Patol. Fiziol. Eksp. Terap., 6, 3-9.
31.Meldrum, B. S. (1994) Neurology, 44, 14-23.
32.Choi, D. W. (1988) Neuron, 1, 623-634.
33.Gulyaeva, N. V. (1995) Neirokhimiya, 12, 63-66.
34.Markov, Kh. M. (1996) Usp. Fiziol. Nauk, 27, 30-57.
35.Gurin, A. V. (1997) Usp. Fiziol. Nauk, 28, 53-60.
36.Myers, P. R., Minor, R. L., Guerra, R., Bates, J. N., and Harrison, D. G. (1990) Nature, 345, 161-164.
37.Forstermann, U., Schmidt, H. H. H. W., Pollack, J. S., Sheng, H., Mitchell, J. A., Warner, T. D., Nakane, M., and Murad, F. (1991) Biochem. Pharmacol., 42, 1849-1857.
38.Pou, S., Pou, W. S., Bredt, D. S., Snyder, S. H., and Rosen, G. M. (1992) J. Biol. Chem., 267, 24173-24176.
39.Schmidt, H. W., Lohmann, S. M., and Walter, U. (1993) Biochim. Biophys. Acta, 1178, 153-175.
40.Palmer, A. M., Ferrigi, A. G., and Moncada, S. (1987) Nature, 327, 524-526.
41.Halliwell, B. (1984) Med. Biol., 62, 71-77.
42.Kovach, A. G. B., Szabo, C., Benyo, Z., Csaki, C., Greenberg, J. H., and Reivich, M. (1992) J. Physiol., 449, 183-196.
43.Manzoni, O., Prezeau, L., Marin, P., Deshager, S., and Bockaert, J. (1992) Neuron, 8, 653-662.
44.Vanin, A. F., Mordvintsev, P. I., and Kleshchev, A. L. (1984) Stud. Biophys., 102, 135-137.
45.Sato, S., Tominaga, T., Ohnishi, T., and Ohnishi, S. (1993) Biochim. Biophys. Acta, 181, 195-197.
46.Zhang, F., and Iadecola, C. (1993) Neuroreport, 4, 559-562.
47.Pelligrino, D. A. (1993) Neurosurg. Anesthesiol., 5, 221-231.
48.Dawson, D. A., Kusumoto, K., Graham, D. I., McCulloch, J., and Macrae, I. M. (1992) Neurosci. Lett., 142, 151-154.
49.Sato, S., Tominaga, T., Ohnishi, T., and Ohnishi, S. (1994) Brain Res., 647, 91-96.
50.Rundfeldt, C., Koch, R., Richter, A., Mevissen, M., Gerecke, U., and Loescher, W. (1995) Eur. J. Pharmacol., 274, 73-81.
51.Przegalinski, E., Baran, L., and Siwanowicz, J. (1996) Neurosci. Lett., 217, 145-148.
52.Maggio, R., Fumagalli, F., Donati, E., Barbier, P., Racagni, G., Corsini, G. U., and Riva, M. (1995) Brain Res., 679, 184-187.
53.Mollace, V., Bagetta, G., and Nistico, G. (1991) Neuroreport, 2, 269-272.
54.De Sarro, G., and de Sarro, A. (1993) Neuropharmacology, 32, 51-58.
55.Nowicki, J. P., Duval, D., Poignet, H., and Scatton, B. (1991) Eur. J. Pharmacol., 204, 339-340.
56.Buisson, A., Lakhmeche, N., Verrecchia, C., Plotkine, M., and Boulu, R. G. (1993) Neuroreport, 4, 444-446.
57.Starr, M. S., and Starr, B. S. (1993) Pharmacol. Biochem. Behav., 45, 321-325.
58.Przegalinski, E., Baran, L., and Siwanowicz, J. (1994) Neurosci. Lett., 170, 74-76.
59.Wang, Q., Theard, M. A., Pelligrino, D. A., Baughman, V. L., Hoffman, W. E., Albrecht, R. F., Cwick, M., Paulson, O. B., and Lassen, N. A. (1994) Brain Res., 658, 192-198.
60.Theard, M. A., Baughman, V. L., Wang, Q., Pelligrino, D. A., and Albrecht, R. F. (1995) Neuroreport, 6, 921-924.
61.Moncada, S., and Higgs, A. (1993) New Engl. J. Med., 329, 2003-2011.
62.Dawson, V. L., Dawson, T. M., London, E. D., Bredt, D. S., and Snyder, S. H. (1991) Proc. Natl. Acad. Sci. USA, 88, 6368-6371.
63.Bashkatova, V. G., Vitskova, G. Yu., Narkevich, V. B., Vanin, A. F., Mikoyan, V. D., Kubrina, L. N., and Rayevsky, K. S. (1996) Neurokhimiya, 13, 110-115.
64.Bashkatova, V. G., Mikoyan, V. D., Kosacheva, E. S., Kubrina, L. N., Vanin, A. F., Rayevsky, K. S., et al. (1996) Dokl. Ros. Akad. Nauk, 348, 119-121.
65.Bashkatova, V. G., Vitskova, G. Yu., Narkevich, V. B., Mikoyan, V. D., Vanin, A. F., and Rayevsky, K. S. (1998) Byull. Eksp. Biol. Med., 125, 26-29.
66.Bashkatova, V. G., Vitskova, G. Yu., Vanin, A. F., Mikoyan, V. D., Narkevich, V. B., and Rayevsky, K. S. (1997) Soc. Neurosci. Abst., 23, 2427.
67.Rayevsky, K. S., Bashkatova, V. G., Vitskova, G. Yu., Narkevich, V. B., Mikoyan, V. D., and Vanin, A. F. (1998) Eksp. Klin. Farmakol., 61, 13-16.