Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Complex Role of Nitric Oxide in the Pathophysiology of Focal Cerebral Ischemia

The Complex Role of Nitric Oxide in the Pathophysiology of Focal Cerebral Ischemia Stroke Research Laboratory, Department of Neurosurgery and Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, U.S.A. Nitrogen monoxide (NO) has recently emerged as an important mediator of cellular and molecular events which impacts the pathophysiology of cerebral ischemia. Although tempting t o ask whether NO is "good or bad" for cerebral ischemia, the question underestimates t h e complexities of NO chemistry and physiology as well as oversimplifies the pathophysiology of focal cerebral ischemia. Important vascular and neuronal actions of NO have been defined which both enhance tissue survival and mediate cellular injury and death, and these will be reviewed. Strategies which modify NO synthesis and / or metabolism may someday assume therapeutic importance, but not until the tissue compartments generating NO, the activities of the enzymes that are inducibly and constitutively expressed, and t h e redox state of NO during the stages of ischemic injury, are defined with greater precision. Our knowledge of these processes is rudimentary. This review will summarize t h e evidence from animal models which supports an emerging role for NO in ischemic pathophysiology. Important aspects of NO synthesis and inhibitors of this process will also be discussed. Introduction phage-mediated cytotoxicity (6,7) and neurotransmission (8,9). Nitric oxide synthase (NOS), a marker of NO biosynthetic activity, was detected within brain parenchyma and vessels (10-12). NO has been proposed as a mediator of cerebrovascular tone (see 13-16), hypercapnic hyperemia (1 7-18), metabolicflow coupling (19-20) and in various pathological processes (8,21) including cerebral ischemia (22). The reader is referred to several excellent reviews (3,4,7-9,12,32,47,79) for more in-depth treatment of this subject. S y n t h e s i s a n d M e t a b o l i s m of N i t r o g e n Monoxide (NO) Nitrogen monoxide (NO), a ubiquitous molecule in mammalian tissues, was recently proposed as a mediator of endothelium-dependent vascular relaxation (1,Z). Its role in ischemic pathophysiology has generated considerable controversy because NO is involved in such diverse physiological processes as regulation of vascular tone (3,4) platelet inhibition (S), macroCorresponding author: Dr. M.A. Moskowitz, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, U.S.A. Tel. +1 (617) 726 8442; Fax +1 (617)726 2547 NO is synthesized from the guanidino nitrogen of L-arginine and molecular oxygen (3,5,23). The reaction requires nicotinamide adenine dinucleotide phosphate (NADPH) and produces citrulline and NO following a five electron-oxidation of L-arginine. Flavins and tetrahydrobiopterin are additional cofactors. At least three isoforms of NOS have been identified: (i) constitutive neuronal and (ii) endothelial isoforms and (iii) an inducible isoform originally isolated from macrophages (24-27). The constitutive isoforms (cNOS) are calcium / calmodulin dependent and activated by small intracellular calcium transients. An increase in intracellular calcium concentration from 100 nM to 500 nM changes the rate of NO synthesis from <5% to >95% of maximum (4,23). The inducible isoform (iNOS) is calciuminsensitive and is stimulated by endotoxins and cytokines, a process that leads to a slower but longer lasting NO increase compared to calcium-dependent NO synthesis (6). The endothelial enzyme and the inducible isoform have a molecular weight around 135 kiloDalton (kDa) whereas the neuronal form is larger (160 kDa) (24-28). All three isoforms also possess NADPH-diaphorase activity, the presence of which has been used to infer the localization of NOS enzyme (29). In the brain, cNOS exists in neurons (ll),astrocytes (very low levels of expression) (12), perivascular nerves and cerebrovascular endothelium (lo), and the inducible form in astrocytes and microglia (12), vascular smooth muscle (30) and endothelial cells (31). Synthesis from the constitu- T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia tively expressed protein is enhanced by either raising intracellular calcium or by increasing the amount of cNOS enzyme (32). Providing more cofactor does not affect the rate of synthesis nor does providing additional substrate. For example, free intracellular Larginine (300-800 pM) (33,34) is greater than the Michaelis-Menton constatnt (Km) derived for bovine endothelial NOS (2.9 pM) (26) or rat cerebellar NOS (2 pM) (24). Despite these considerations, we observed that intravenous infusion of L-arginine leads to pial vessel dilation and blood flow increases within no ma1 rat brain (35-37). The evidence using NOS inhibitors supports (but does not prove) a role for L-arginine-induced enhanced NO synthesis. Once it is formed, NO readily diffuses out of cells. Rapid removal by oxygen radicals and metalloproteins like hemoglobin limits its spread to a few hundred microns and shortens its half-life to the order of seconds (4). However, formation of S-nitrosothiol adducts may stabilize the labile NO radical and prolong its biological half-life. Sulfhydryl groups in proteins (e.g., albumin) represent a rich source of reduced thiol, and S-nitroso-proteins form readily under physiological conditions (38). NO'S short half-life makes detection difficult in brain tissue. Current methods of detection include NOselective electrochemical microsensors (39,40), electron paramagnetic resonance coupled with an in vivo spin-trapping technique (4 l), spectroscopy combined with microdialysis (after trapping NO with hemoglobin) (42), and fluorometric measurements of its stable metabolite nitrite ex vivo (43). Citrulline and guanosine 3',5'-cyclic monophosphate (cGMP) levels are also an indirect index, but their specificity is lower. None of these methods can identify the origin or source of detectable NO. Assays for NOS activity in vitro may also not reflect conditions in vivo inasmuch as co-factors and ions are added to optimize the incubation conditions, and cellular compartmentation is disrupted by tissue homogenization. Under physiological conditions, there are no known mechanisms regulating tissue NO levels other than those regulating biosynthesis (3,9,27). However, degradation of superoxide by superoxide dismutase or removal of hemoglobin prolongs its half-life and increases tissue levels of NO (4). Recent calculations by Wink et al. (44), however, estimate that more than 30 minutes are required for removing NO generated by tissue oxygen, leading to the speculation that some other mechanisms must exist to breakdown NO. Targets of Nitrogen Monoxide (NO) Action lar calcium levels sufficient to activate NO synthase inhibit guanylate cyclase (23,45). Blood vessels may provide such an example. Vasodilatory neuromediators like acetylcholine or bradykinin raise intracellular calcium and stimulate endothelial NOS activity to release NO or a related nitrosothiol (EDRF) from endothelial cells. NO, in turn, enhances cGMP synthesis in smooth muscle cells (1,2,4,46). Nitrovasodilators such as nitroglycerin, nitroprusside or 3-morpholinosydnonimine (SIN-1) raise cGMP in vascular smooth muscle by directly liberating or donating NO (3,47). Platelet inhibition, neurotransmission and penile erection are also mediated by cGMP-dependent mechanisms (3,9). NO may also react reversibly with thiol and metal groups to modulate the activity of certain proteins (32,48). Some antimicrobial effects of NO, inhibition of N-methyl-D-aspartate (NMDA) receptor activity, regulation of adenosine diphosphate (ADP)-ribosylation are examples of the latter mechanism. Interestingly, NO can be interconverted among three redox forms: (i) nitrosonium (NO+), (ii) nitric oxide (NO') and (iii) nitroxyl anion (NO-), under physiological conditions (48). These redox forms favor different effector interactions. For example, NO' is required for activation of guanylate cyclase and this form is also suggested to account for the neurotoxic action of NO whereas NO+ down regulates NMDA receptor activity, therefore, may be neuroprotective (48,49). NO donors may also exert different actions depending upon the redox form of NO generated. Lipton et al. proposed that sodium nitroprusside protected neurons because it generates NO but can be converted to the NO' toxic pathway after reduction with cysteine (49). L-Arginine Analogs and Nitric Oxide Synthase (NOS) Inhibition The actions of NO usually involve activation of the heme-containing enzyme guanylate cyclase following formation of an NO-heme complex (3,4,23). Elevations in cGMP result. However, cGMP dependent mechanisms may only be operational in cells adjacent to those generating N O because intracellu- Most researchers use NOS inhibitors to investigate the role of NO in various physiological and pathological conditions (SO). The most widely studied inhibitors are the substrate analogs of L-arginine such as nitro-L-arginine methyl ester (L-NAME), nitro-L-arginine (L-NA) and monomethyl-L-arginine (L-NMMA). In most instances, the conversion of Larginine and molecular oxygen to citrulline and NO is blocked competitively, although non-competitive inhibition may occur depending upon dosage and treatment duration (3,7,51-53). Like NOS activity itself, enzyme inhibition is enantiomerically-specific. The inhibitors L-NAME, L-NA and L-NMMA restrict both the constitutively expressed and inducible enzymes and do not discriminate between the neuronal and endothelial forms (32). It is notable, however, that L-NA displays a marked preference for the constitutively expressed protein whereas L-NMMA is reportedly more potent and effective against the inducible protein (54). Recently, 7-nitro indazole was reported to inhibit rat brain NOS potently without T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia increasing the blood pressure, suggesting that it may be a selective inhibitor of the neuronal form (55). NOS inhibitors have been administered systemically (intraperitoneal, intravenous) directly into the brain (microdialysis), topically by superfusion (cranial window preparation) or into the cerebrospinal fluid (intracerebroventricular). An advantage of intracranial administration is that it does not increase arterial blood pressure or peripheral vascular resistance and avoids potentially confounding drug actions on large proximal arteries supplying the brain. Hence, the role of NOS can be studied selectively within the cerebral microcirculation. However, local application of high concentrations of these agents for prolonged periods of time increases the likelihood of non-specific effects. Measuring NOS activity after topical or intravenous L-arginine analogs provides useful information which can reduce the possibility of invalid conclusions. This is illustrated by recent studies in which the regional cerebral blood flow (rCBF) response was correlated with the degree of NOS inhibition within underlying cortex following topical L-NAME or LNA (19). Significant inhibition of the physiological response was achieved only at 30 minutes or longer, or until at least 50% of enzyme activity was inhibited after topical L-NA application. Had these experiments ended at 15 minutes rather than at 30 or 60 minutes, incorrect conclusions would have been reached. In all likelihood, the level of enzyme inhibition depends upon the route, dosage and time of administration and differs for each cell and tissue source. More detailed studies will be needed to clarify these issues. NOS inhibitors exhibit other effects which raise concerns about specificity. For example, L-NAME exhibits muscarinic antagonist actions (56). Effects of L-arginine analogs on iron containing reactions like cytochrome C reduction (57) and blockade of endothelium-independent vasodilatation have been reported. In addition, L-NMMA competes with Larginine uptake (58), can be metabolized by endothelial cells to L-citrulline which in turn serve as a precursor for NO synthesis by conversion to Larginine (59). Nitrogen Monoxide (NO) and Focal Ischemia NO has been implicated in the pathophysiology of 30- Slice Number Slice Number Figure 1 L-arginine but not D-arginine infusion reduces infarct size in two different models of cerebral ischemia. Infarction areas on seven coronal slices. In model one (top), neocortical infarct after common carotid / distal middle cerebral arterial (MCA) occlusion; In model two (bottom), neocortical striatal infarct after proximal MCA occlusion. Open circles, filled circles and triangles denote saline-, L-arginine- and D-arginine-treated animals, respectively. Slice one is the most rostral. Data are means 5 S.E. 'P < 0.01 compared with D-arginine-treated group. Reproduced with permission from Am J Physiol 253: H1632-Hl635, 1992, reference 36. focal cerebral ischemia based on its actions as a mediator of tissue injury. Within 3 to 24 minutes after middle cerebral artery (MCA) occlusion, NO increases dramatically from approximately 10 nM to 2.2 pM within cortex as detected by a porphyrinic microsensor (40). Brain nitrite and cGMP levels also rise, and these increases are effectively blocked by prior L-NA administration (43). Brain NOS activity, as assessed ex vivo, increases as well. Nitrite, NO and NOS activity return to baseline within an hour. One speculation holds that cNOS activity increases due to a rise in intracellular Ca ++ / calmodulin complex (9,60). cNOS is activated by intracellular calcium concentrations slightly above the resting level and is maximally stimulated at levels (23) well below those reached in neurons during focal ischemia (10-100 pM) (61). A Ca++/calmodulin-dependent enhancement would not account, however, for increased enzyme activity detected by the ex vivo assay without some other (possibly co-valent) modification such as a change in phosphorylation state (43,62). Neurons, astrocytes, perivascular nerves and cerebrovascular endothelium may form NO during cerebral ischemia. A late but sustained increase in NO levels T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia 1 - MCAO L-ARG Figure 2 L arginine infusion increases ischemic regional ceiebral blood flow (rCBF) and leads to functional recovery The computer print-out of a one-hour recording period illustrates electrocorticogram (ECoG) and rCBF recorded from the same cortical gray matter sample in the penumbral region of dorso-lateral cortex distal to an occluded middle cerebral artery (MCA) End-tidal CO2 and arterial blood pressure are also displayed rCBF fell to 20% of the pre-ischemic level upon clipping the MCA Intravenous L-arginine (300 mg / kg, over ten minutes) starting five minutes after the occlusion increased rCBF An improvement in ECoG began 5 5 minutes after rCBF exceeded 304b of baseline and sustained along with the increase in rCBF No rCBF response to hypercapnia was observed during ischemia although a hyperemic response had been recorded before ischemia (not shown) due to expression of iNOS may also occur but currently there is no data available to support this contention. The fact that microglia comprise 5-12% of brain cells (63) and proliferate in the peri-infarct zone merits attention in this regard. However, there are data t o suggest that during the immediate period following ischemia, an increased NO production in vascular endothelium or perivascular nerves may improve blood flow and be neuroprotective. Indeed, infusion of L-arginine, which dilates pial vessels (blocked by topical L-NAME) and increases rCBF in normal as well as in ischemic brain (35,37), reduces infarct size (36,37) (Fig 1). In fact, L-arginine infusion leads to electrical recovery if blood flow enhancement exceeds t h e functional flow threshold of approximately 30% of pre-ischemic flow (64) (Fig 2). The functional recovery, as detected by a glass microelectrode placed within cortex immediately subjacent to t h e laser-Doppler flow probe, followed the increase in flow within minutes but never vice versa in seven rats studied. The blood flow increase which was first detected by laser-Doppler flowmetry (35,37) and then verified by dynamic magnetic resonance image (MRI) scanning (65) was observed when Larginine was administered shortly after the induc- tion of ischemia and was greatly diminished when L-arginine was administered 30 minutes after MCA occlusion (64). The attenuation of L-arginine's effect on both blood flow and tissue protection roughly parallels the decrease in tissue N O production after ischemia (40,43). Inactivation of NOS (whole brain) following its early activation was reported by Kader et al. (43) and may reflect proteolysis or accumulation of an enzyme inhibitor within ischemic tissue. In all likelihood, endothelial NOS share a similar fate. The NO binding protein, guanylate cyclase appears more robust and resistant t o the effects of ischemia than NOS (66). Hence, administering NO donors leads to blood flow increases within the ischemic tissue for at least one hour after arterial occlusion and a decrease in infarct size in models of focal ischemia (66,67). One may speculate that NO production by perivascular nerves improve blood flow in a zone of hypoperfusion. Parasympathetic fibers appear particularly relevant because they contain NOS and the vasodilators VIP and possibly acetylcholine within the same neurons (10). Not surprisingly, infarction volume caused by MCA occlusion increases .after chronic parasympathectomy. However, infarct size does not increase after acute parasympathetic sectioning, which suggests the possibility that in this instance, t h e dilator response does not require brain stem activation but rather involves the direct release of transmitters from perivascular nerves by chemicals within the perivascular.space (68,69). Although t h e ameliorative effects of raising blood flow within ischemic tissue was shown more than a decade ago (70), the merits of this treatment strategy have not yet been accepted unanimously (71,72). Opponents argue that increasing rCBF promotes deleterious effects on edema formation (73) and free radical generation during reperfusion (74) and that normal tissue may "steal" blood from the ischemic zone (71). However, recent findings with NO donors (66,67) and L-arginine (36,37) and preliminary data following thrombolytic therapy in stroke patients (75) strongly suggest that early restoration of rCBF can increase tissue survival and restore function. Promoting such changes appears especially critical within t h e ischemic penumbra. However, not all measures which elevate brain blood flow d o so within this region (e.g., hypercapnia) (Fig 2). The potential therapeutic role for drugs which specifically enhance brain blood flow within t h e peri-infarct zone merits further consideration. Studies Using Nitric Oxide Synthase (NOS) Inhibitors Protective effects notwithstanding, it is difficult t o reconcile t h e conflicting literature which report either increases or reductions in infarct volume after permanent or transient MCA occlusion and NOS T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia Table 1 Effect of nitric oxide synthase inhibitors on focal cerebral ischemia in permanent and transient middle cerebral artery occlusion (MCAO) models Species Reference Model Duration of Occlusion Survival Time Inhibitor Infarct Volume Decrease in Infarct Volume Mouse pMCAO 7 days 7 days L-NA 1 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NA 1 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NAME 3 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NAME 10 mg/kg intravenous pre 30 minutes L-NAME 0.7 mg/kg intravenous infusion pre 1 hour L-NA 100 mg/kg intraperitoneal pre 15 hours 72% 1 Rat SD pMCAO 3 days 3 days 44% Rat SHR 84-85 pMCAO 2 days 2 days 35% 1 Cat pMCAO 4 hours 3 hours 2.5 hours 4 hours 5 hours 10 days 35% 1 55% 1 100% 1 Rat SHR tMCAO Rat pup tUCAO Hypoxia Increase in Infarct Volume Rat SHR pMCAO 1 day 1 day L-NA 2.4 mg/kg intravenous infusion post 0-1 hour L-NAME 10 mg/kg intracarotid infusion post 0-1 hour L-NA 1 mg/kg intraperitoneal post 5 minutes, 3 and 6 hours L-NAME 3 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NA 1 mg/kg intraperitoneal post 5 minutes, 3 and 6 hours L-NAME 3 mg/kg intraperitoneal post 5 minutes, 3,6,24 and 36 hours L-NA 2 x 30 mg/kg subcutaneous pre and post at 30 minutes L-NAME 5 mg/kg intravenous pre 15 minutes 32% t 60% Rat SD pMCAO 1 day 1 day Rat SD pMCAO 1 day 1 day 67% T Rat SD pMCAO 2 days 2 days 21% Rat LE pMCAO + UCAO 1 day 1 day 68% T Rat LE pMCAO + UCAO 2 days 2 days NC Rat SD pMCAO 4 hours 4 hours 10% Rat W tMCAO 2 hours 3 days >loo% Abbreviations: p, Permanent; t, Transient; MCAO, Middle cerebral artery occlusion; UCAO, Unilateral common carotid artery occlusion; SD, Sprague-Dawley rat; SHR, Spontaneously hypertensive rat; LE, Long Evans rat; W, Wistar rat; L-NA, Nitro-L-arginine; L-NAME, Nitro-L-arginine methyl ester; pre, Pre-ischemia; post, Post-ischemia; NC, No change. T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia Table 2 Nitrogen monoxtde effects which mpact cerebral ischemia Pos/tive Enhancement of regional cerebral blood flow Inhibition of platelet aggregation Inhibition of platelet and neutrophil adhesion Inhibition of Nmethyl-D-aspartate current Negative Direct cytotoxicity Binding to iron-sulfur complexes (inhib tion of mitochondria1 respiratory enzymes, ribonucleotide reductase, aconitase) Nitrosylation of thiols (inhibition of GAPDH) One simple explanation (with complex ramifications) suggests that N O has both neuroprotective and neurotoxic tissue effects (Table 2). There is already evidence t o support this contention. While current hypotheses and available data suggest that endothelial and perivascular NO have a neuroprotective role, parenchymal NO may exert this dual effect. One may envisage that the predominant action of NO may vary with the tissue compartment (e.g., vascular versus parenchymal), or with the time after the onset o ischemia. Accordingly, the protocol f chosen may favor one or the other of the opposing effects, giving rise t o contradictory results. Some answers may be forthcoming with the development of selective inhibitors for the neuronal or vascular isoforms, or transgenic mice whose neuronal or endothelial NOS are selectively knocked-out, and with a wider application of methods for both directly measuring NO and assessing NOS activity in vivo. Nitrogen Monoxide (NO)-Mediated Neurotoxicity DNA nitration Indirect cytotoxicity via formation of Peroxynitrite Hydroxyl radical Nitrogen dioxide Abbreviations: GAPDH, Glyceraldehyde-3-phosphate inhibition (Table 1). Of course, some of the problems inherent with the use of non-selective NOS inhibitors may underlie some of the existing controversies. However, the observed discrepancies cannot be accounted for by the choice of species, anesthetic, or ischemic model in any simple way although each of these can exert significant effects o n NO synthesis and metabolism. For example, anesthetics like halothane or isoflurane are reported to increase pial arterial diameter with an NO-mediated, yet unknown, mechanism (50). The choice of model may also be important. For example, injury in t h e hypoxiaischemic neonatal rat may be more dependent upon excitotoxic mechanisms and less o n reductions in blood flow than models in which tissue injury is initiated by vascular occlusion only (76). Various species and strains (e.g., normotensive versus hypertensive ratsj possess different susceptibilities t o stroke which may reflect the level of blood pressure elevation, differences in endothelial NOS activity (77), or a greater and lesser extent of collateral blood flow. Severe hypertension caused by NOS inhibitors may also potentiate edema formation and cause bleeding into the core territory (73). Many of these factors are difficult to control and may go unnoticed. In fact, most studies only monitor arterial blood pressure briefly and rarely beyond the anesthetic period; NOS activity within brain is measured only infrequently. NO and its degradation products cause cytotoxicity through formation of iron-NO complexes with several enzymes including complex I and I1 of mitochondrial electron transport, oxidation of protein sulfhydryls and DNA nitration (9,48). The inhibition of glyceraldehyde-3-phosphate dehydrogenase by NO-mediated ADP-ribosylation (78) may have detrimental consequences. This enzyme catalyzes the first step in both glycolysis and the hexose monophosphate shunt and is important for NADP+ synthesis a n d for maintaining intracellular reduced gluthatione levels. NO may mediate cell death also through formation of t h e potent oxidant peroxynitrite (ONOO-) (9,79,80). Furthermore, ONOO- decomposes t o another reactive oxygen species closely related to the hydroxyl free radical ('OH) and to the radical nitrogen dioxide (N02), which is a potent activator of lipid peroxidation (79). Nitrogen Monoxide (NO) and the Wmethyl-DAspartate (NMDA) Receptor NO was proposed as the neurotoxic agent mediating NMDA toxicity. Studies in dissociated cell cultures showed that NOS inhibitors effectively blocked NMDA-induced cell death and that L-arginine depletion attenuated NMDA receptor-mediated toxicity (60). Based o n these studies, it was hypothesized that ischemia-induced NMDA receptor over-activation leads to an increase in intracellular calcium which activates calmodulin and increases NO production. As noted above, it was recently shown that NO also inhibits NMDA currents depending o n its redox state (49). The neurotoxic actions of NO derive from NO' form of t h e molecule whereas N O + form reacts with t h e thiol group of NMDA receptor and blocks its function (49). However, whether NO' or N O + predominates during t h e T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia one-hour rise in tissue, NO still remains to be determined. Moreover, formation of peroxynitrite requires a strict co-localization of NO and superoxide productions (80). It appears so that increased NO production in the brain parenchyma may play either a destructive or protective role depending upon the redox state of the tissue, coexisting free radicals and their spatial and temporal relationships. The outcome even becomes less predictable if one considers the possibility that phosphorylation of NOS induced by glutamate activation of protein kinase C may inhibit its activity (62,81). Conclusion 9. Dawson TM, Dawson VL, Snyder SH (1992) A novel neuronal messenger in brain: The free radical, nitric oxide. Ann Neurol 32: 297-31 1 10. Nozaki K, Moskowitz MA, Maynard KI, Koketsu N, Dawson TM, Bredt DS, Snyder SH (1993) Possible origin and distribution of immunoreactive nitric oxide synthase-containing nerve fibers in cerebral arteries. J Cereb Blood Flow Metab 13: 70-79 11. Bredt DS, Hwang PM, Snyder SH (1990) Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347: 768-770 12.Murphy S, Simmons ML, Agullo L, Garcia A, Feinstein DL, Galea E, Reis DJ, Minc-Golomb D, Schwartz JP (1993) Synthesis of nitric oxide in CNS glial cells. Trends Neurosci 16: 323-328 13. Faraci FM (1990) Role of nitric oxide in regulation of basilar artery tone. Am J Physiol 259: HI 216-H 1221 14. Busija DW, Leffler CW, Wagerle LC (1990) Mono-L-arginine-containing compounds dilate piglet arterioles via an endothelium-derived relaxing factor-like substance. Circ Res 67: 1374-1380 15.Rosenblum WI, Nishimura H, Nelson GH (1990) Endothelium-dependent L-arginine and L-NMMA-sensitive mechanisms regulate tone of brain microvessels. A m J PhySiOl 259: H1396-H1401 16.Tanaka K, Gotoh F, Gomi S, Takashima S, Mihara B, Shirai T, Nogaw S, Nagata E (1991) Inhibition of nitric oxide synthesis induces a significant reduction in local cerebral blood flow in the rat. Neurosci Lett 127: 129-132 17. ladecola C (1992) Does nitric oxide mediate the increases in cerebral blood flow elicited by hypercapnia? Proc Natl Acad Sci USA 89: 3913-3916 18.Wang Q, Paulson 0 8 , Lassen NA (1992) Effect of nitric oxide blockade by NG-nitro-L-arginine on cerebral blood flow response to changes in carbon dioxide tension. J Cereb Blood Flow Metab 12: 935-946 19. lrikura K, Maynard KI, Moskowitz M A (1993) The importance of nitric oxide synthase inhibition to the attenuated vascular response induced by topical L-nitro-arginine during vibrissae stimulation. J Cereb Blood Flow Metab (in press) 20. Dirnagl U, Lindauer U, Villringer A (1993) Role of nitric oxide the coupling of cerebral blood flow to neuronal activation in rats. Neurosci Lett 149: 43-36 21. Snyder SH (1993) Janus faces of nitric oxide. Nature 364: 577 22. Pelligrino DA (1993) Saying NO to cerebral ischemia. J Neurosurg Anesthesiol 5: 221-231 23. Knowles RG. Palacios M, Palmer RMJ, Moncada S (1989) Formation of nitric oxide from L-arginine in the central nervous system: A transduction mechanism for stimulation of the soluble guanylate cyclase. Proc Natl Acad Sci USA 86: 5159-5162 24. Bredt DS, Snyder SH (1990) Isolation of nitric oxide synthase, a calmodulin-requiring enzyme. Proc Natl Acad S C USA 87: 682-685 ~ 25. Hevel JM, White KA. Marletta M A (1991) Purification of the inducible murine macrophage nitric oxide synthase: Identification as a flavoprotein. J Biol Chem 266: 2278922791 26.Pollock JS. Forstermann U, Mitchell JA, Warner TD. Schmidt HHHW, Nakane M. Murad F (1991) Purification and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelial cells. Proc Natl Acad Sci USA 88: 1048010484 In conclusion, the discovery of nitric oxide provides a powerful opportunity to explore the pathophysiology and treatment of cerebral ischemia. Controversy abounds, which may reflect the importance of nitric oxide to the diversity of factors (both known as well as unknown) which impact cerebral ischemia. Will drugs that alter NO synthesis and metabolism ever become useful for the treatment of ischemia? Is NO relevant to the pathophysiology of global ischemia? Only time will tell. Nevertheless, answers should be forthcoming as we: (i) develop selective inhibitors for the neuronal or vascular isoforms of NOS, (ii) develop transgenic mice whose neuronal or endothelial NOS are selectively knocked-out, and (iii) discover methods for both directly measuring NO and assessing NOS activity routinely. Acknowledgments Some of the studies described herein were supported by grants from the National Institutes of Health and the Massachusetts General Hospital Health Interdepartmental Stroke Program Project. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Brain Pathology Wiley

The Complex Role of Nitric Oxide in the Pathophysiology of Focal Cerebral Ischemia

Brain Pathology , Volume 4 (1) – Jan 1, 1994

Loading next page...
 
/lp/wiley/the-complex-role-of-nitric-oxide-in-the-pathophysiology-of-focal-KPts0po017

References (104)

Publisher
Wiley
Copyright
Copyright © 1994 Wiley Subscription Services, Inc., A Wiley Company
ISSN
1015-6305
eISSN
1750-3639
DOI
10.1111/j.1750-3639.1994.tb00810.x
Publisher site
See Article on Publisher Site

Abstract

Stroke Research Laboratory, Department of Neurosurgery and Neurology, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, U.S.A. Nitrogen monoxide (NO) has recently emerged as an important mediator of cellular and molecular events which impacts the pathophysiology of cerebral ischemia. Although tempting t o ask whether NO is "good or bad" for cerebral ischemia, the question underestimates t h e complexities of NO chemistry and physiology as well as oversimplifies the pathophysiology of focal cerebral ischemia. Important vascular and neuronal actions of NO have been defined which both enhance tissue survival and mediate cellular injury and death, and these will be reviewed. Strategies which modify NO synthesis and / or metabolism may someday assume therapeutic importance, but not until the tissue compartments generating NO, the activities of the enzymes that are inducibly and constitutively expressed, and t h e redox state of NO during the stages of ischemic injury, are defined with greater precision. Our knowledge of these processes is rudimentary. This review will summarize t h e evidence from animal models which supports an emerging role for NO in ischemic pathophysiology. Important aspects of NO synthesis and inhibitors of this process will also be discussed. Introduction phage-mediated cytotoxicity (6,7) and neurotransmission (8,9). Nitric oxide synthase (NOS), a marker of NO biosynthetic activity, was detected within brain parenchyma and vessels (10-12). NO has been proposed as a mediator of cerebrovascular tone (see 13-16), hypercapnic hyperemia (1 7-18), metabolicflow coupling (19-20) and in various pathological processes (8,21) including cerebral ischemia (22). The reader is referred to several excellent reviews (3,4,7-9,12,32,47,79) for more in-depth treatment of this subject. S y n t h e s i s a n d M e t a b o l i s m of N i t r o g e n Monoxide (NO) Nitrogen monoxide (NO), a ubiquitous molecule in mammalian tissues, was recently proposed as a mediator of endothelium-dependent vascular relaxation (1,Z). Its role in ischemic pathophysiology has generated considerable controversy because NO is involved in such diverse physiological processes as regulation of vascular tone (3,4) platelet inhibition (S), macroCorresponding author: Dr. M.A. Moskowitz, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114, U.S.A. Tel. +1 (617) 726 8442; Fax +1 (617)726 2547 NO is synthesized from the guanidino nitrogen of L-arginine and molecular oxygen (3,5,23). The reaction requires nicotinamide adenine dinucleotide phosphate (NADPH) and produces citrulline and NO following a five electron-oxidation of L-arginine. Flavins and tetrahydrobiopterin are additional cofactors. At least three isoforms of NOS have been identified: (i) constitutive neuronal and (ii) endothelial isoforms and (iii) an inducible isoform originally isolated from macrophages (24-27). The constitutive isoforms (cNOS) are calcium / calmodulin dependent and activated by small intracellular calcium transients. An increase in intracellular calcium concentration from 100 nM to 500 nM changes the rate of NO synthesis from <5% to >95% of maximum (4,23). The inducible isoform (iNOS) is calciuminsensitive and is stimulated by endotoxins and cytokines, a process that leads to a slower but longer lasting NO increase compared to calcium-dependent NO synthesis (6). The endothelial enzyme and the inducible isoform have a molecular weight around 135 kiloDalton (kDa) whereas the neuronal form is larger (160 kDa) (24-28). All three isoforms also possess NADPH-diaphorase activity, the presence of which has been used to infer the localization of NOS enzyme (29). In the brain, cNOS exists in neurons (ll),astrocytes (very low levels of expression) (12), perivascular nerves and cerebrovascular endothelium (lo), and the inducible form in astrocytes and microglia (12), vascular smooth muscle (30) and endothelial cells (31). Synthesis from the constitu- T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia tively expressed protein is enhanced by either raising intracellular calcium or by increasing the amount of cNOS enzyme (32). Providing more cofactor does not affect the rate of synthesis nor does providing additional substrate. For example, free intracellular Larginine (300-800 pM) (33,34) is greater than the Michaelis-Menton constatnt (Km) derived for bovine endothelial NOS (2.9 pM) (26) or rat cerebellar NOS (2 pM) (24). Despite these considerations, we observed that intravenous infusion of L-arginine leads to pial vessel dilation and blood flow increases within no ma1 rat brain (35-37). The evidence using NOS inhibitors supports (but does not prove) a role for L-arginine-induced enhanced NO synthesis. Once it is formed, NO readily diffuses out of cells. Rapid removal by oxygen radicals and metalloproteins like hemoglobin limits its spread to a few hundred microns and shortens its half-life to the order of seconds (4). However, formation of S-nitrosothiol adducts may stabilize the labile NO radical and prolong its biological half-life. Sulfhydryl groups in proteins (e.g., albumin) represent a rich source of reduced thiol, and S-nitroso-proteins form readily under physiological conditions (38). NO'S short half-life makes detection difficult in brain tissue. Current methods of detection include NOselective electrochemical microsensors (39,40), electron paramagnetic resonance coupled with an in vivo spin-trapping technique (4 l), spectroscopy combined with microdialysis (after trapping NO with hemoglobin) (42), and fluorometric measurements of its stable metabolite nitrite ex vivo (43). Citrulline and guanosine 3',5'-cyclic monophosphate (cGMP) levels are also an indirect index, but their specificity is lower. None of these methods can identify the origin or source of detectable NO. Assays for NOS activity in vitro may also not reflect conditions in vivo inasmuch as co-factors and ions are added to optimize the incubation conditions, and cellular compartmentation is disrupted by tissue homogenization. Under physiological conditions, there are no known mechanisms regulating tissue NO levels other than those regulating biosynthesis (3,9,27). However, degradation of superoxide by superoxide dismutase or removal of hemoglobin prolongs its half-life and increases tissue levels of NO (4). Recent calculations by Wink et al. (44), however, estimate that more than 30 minutes are required for removing NO generated by tissue oxygen, leading to the speculation that some other mechanisms must exist to breakdown NO. Targets of Nitrogen Monoxide (NO) Action lar calcium levels sufficient to activate NO synthase inhibit guanylate cyclase (23,45). Blood vessels may provide such an example. Vasodilatory neuromediators like acetylcholine or bradykinin raise intracellular calcium and stimulate endothelial NOS activity to release NO or a related nitrosothiol (EDRF) from endothelial cells. NO, in turn, enhances cGMP synthesis in smooth muscle cells (1,2,4,46). Nitrovasodilators such as nitroglycerin, nitroprusside or 3-morpholinosydnonimine (SIN-1) raise cGMP in vascular smooth muscle by directly liberating or donating NO (3,47). Platelet inhibition, neurotransmission and penile erection are also mediated by cGMP-dependent mechanisms (3,9). NO may also react reversibly with thiol and metal groups to modulate the activity of certain proteins (32,48). Some antimicrobial effects of NO, inhibition of N-methyl-D-aspartate (NMDA) receptor activity, regulation of adenosine diphosphate (ADP)-ribosylation are examples of the latter mechanism. Interestingly, NO can be interconverted among three redox forms: (i) nitrosonium (NO+), (ii) nitric oxide (NO') and (iii) nitroxyl anion (NO-), under physiological conditions (48). These redox forms favor different effector interactions. For example, NO' is required for activation of guanylate cyclase and this form is also suggested to account for the neurotoxic action of NO whereas NO+ down regulates NMDA receptor activity, therefore, may be neuroprotective (48,49). NO donors may also exert different actions depending upon the redox form of NO generated. Lipton et al. proposed that sodium nitroprusside protected neurons because it generates NO but can be converted to the NO' toxic pathway after reduction with cysteine (49). L-Arginine Analogs and Nitric Oxide Synthase (NOS) Inhibition The actions of NO usually involve activation of the heme-containing enzyme guanylate cyclase following formation of an NO-heme complex (3,4,23). Elevations in cGMP result. However, cGMP dependent mechanisms may only be operational in cells adjacent to those generating N O because intracellu- Most researchers use NOS inhibitors to investigate the role of NO in various physiological and pathological conditions (SO). The most widely studied inhibitors are the substrate analogs of L-arginine such as nitro-L-arginine methyl ester (L-NAME), nitro-L-arginine (L-NA) and monomethyl-L-arginine (L-NMMA). In most instances, the conversion of Larginine and molecular oxygen to citrulline and NO is blocked competitively, although non-competitive inhibition may occur depending upon dosage and treatment duration (3,7,51-53). Like NOS activity itself, enzyme inhibition is enantiomerically-specific. The inhibitors L-NAME, L-NA and L-NMMA restrict both the constitutively expressed and inducible enzymes and do not discriminate between the neuronal and endothelial forms (32). It is notable, however, that L-NA displays a marked preference for the constitutively expressed protein whereas L-NMMA is reportedly more potent and effective against the inducible protein (54). Recently, 7-nitro indazole was reported to inhibit rat brain NOS potently without T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia increasing the blood pressure, suggesting that it may be a selective inhibitor of the neuronal form (55). NOS inhibitors have been administered systemically (intraperitoneal, intravenous) directly into the brain (microdialysis), topically by superfusion (cranial window preparation) or into the cerebrospinal fluid (intracerebroventricular). An advantage of intracranial administration is that it does not increase arterial blood pressure or peripheral vascular resistance and avoids potentially confounding drug actions on large proximal arteries supplying the brain. Hence, the role of NOS can be studied selectively within the cerebral microcirculation. However, local application of high concentrations of these agents for prolonged periods of time increases the likelihood of non-specific effects. Measuring NOS activity after topical or intravenous L-arginine analogs provides useful information which can reduce the possibility of invalid conclusions. This is illustrated by recent studies in which the regional cerebral blood flow (rCBF) response was correlated with the degree of NOS inhibition within underlying cortex following topical L-NAME or LNA (19). Significant inhibition of the physiological response was achieved only at 30 minutes or longer, or until at least 50% of enzyme activity was inhibited after topical L-NA application. Had these experiments ended at 15 minutes rather than at 30 or 60 minutes, incorrect conclusions would have been reached. In all likelihood, the level of enzyme inhibition depends upon the route, dosage and time of administration and differs for each cell and tissue source. More detailed studies will be needed to clarify these issues. NOS inhibitors exhibit other effects which raise concerns about specificity. For example, L-NAME exhibits muscarinic antagonist actions (56). Effects of L-arginine analogs on iron containing reactions like cytochrome C reduction (57) and blockade of endothelium-independent vasodilatation have been reported. In addition, L-NMMA competes with Larginine uptake (58), can be metabolized by endothelial cells to L-citrulline which in turn serve as a precursor for NO synthesis by conversion to Larginine (59). Nitrogen Monoxide (NO) and Focal Ischemia NO has been implicated in the pathophysiology of 30- Slice Number Slice Number Figure 1 L-arginine but not D-arginine infusion reduces infarct size in two different models of cerebral ischemia. Infarction areas on seven coronal slices. In model one (top), neocortical infarct after common carotid / distal middle cerebral arterial (MCA) occlusion; In model two (bottom), neocortical striatal infarct after proximal MCA occlusion. Open circles, filled circles and triangles denote saline-, L-arginine- and D-arginine-treated animals, respectively. Slice one is the most rostral. Data are means 5 S.E. 'P < 0.01 compared with D-arginine-treated group. Reproduced with permission from Am J Physiol 253: H1632-Hl635, 1992, reference 36. focal cerebral ischemia based on its actions as a mediator of tissue injury. Within 3 to 24 minutes after middle cerebral artery (MCA) occlusion, NO increases dramatically from approximately 10 nM to 2.2 pM within cortex as detected by a porphyrinic microsensor (40). Brain nitrite and cGMP levels also rise, and these increases are effectively blocked by prior L-NA administration (43). Brain NOS activity, as assessed ex vivo, increases as well. Nitrite, NO and NOS activity return to baseline within an hour. One speculation holds that cNOS activity increases due to a rise in intracellular Ca ++ / calmodulin complex (9,60). cNOS is activated by intracellular calcium concentrations slightly above the resting level and is maximally stimulated at levels (23) well below those reached in neurons during focal ischemia (10-100 pM) (61). A Ca++/calmodulin-dependent enhancement would not account, however, for increased enzyme activity detected by the ex vivo assay without some other (possibly co-valent) modification such as a change in phosphorylation state (43,62). Neurons, astrocytes, perivascular nerves and cerebrovascular endothelium may form NO during cerebral ischemia. A late but sustained increase in NO levels T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia 1 - MCAO L-ARG Figure 2 L arginine infusion increases ischemic regional ceiebral blood flow (rCBF) and leads to functional recovery The computer print-out of a one-hour recording period illustrates electrocorticogram (ECoG) and rCBF recorded from the same cortical gray matter sample in the penumbral region of dorso-lateral cortex distal to an occluded middle cerebral artery (MCA) End-tidal CO2 and arterial blood pressure are also displayed rCBF fell to 20% of the pre-ischemic level upon clipping the MCA Intravenous L-arginine (300 mg / kg, over ten minutes) starting five minutes after the occlusion increased rCBF An improvement in ECoG began 5 5 minutes after rCBF exceeded 304b of baseline and sustained along with the increase in rCBF No rCBF response to hypercapnia was observed during ischemia although a hyperemic response had been recorded before ischemia (not shown) due to expression of iNOS may also occur but currently there is no data available to support this contention. The fact that microglia comprise 5-12% of brain cells (63) and proliferate in the peri-infarct zone merits attention in this regard. However, there are data t o suggest that during the immediate period following ischemia, an increased NO production in vascular endothelium or perivascular nerves may improve blood flow and be neuroprotective. Indeed, infusion of L-arginine, which dilates pial vessels (blocked by topical L-NAME) and increases rCBF in normal as well as in ischemic brain (35,37), reduces infarct size (36,37) (Fig 1). In fact, L-arginine infusion leads to electrical recovery if blood flow enhancement exceeds t h e functional flow threshold of approximately 30% of pre-ischemic flow (64) (Fig 2). The functional recovery, as detected by a glass microelectrode placed within cortex immediately subjacent to t h e laser-Doppler flow probe, followed the increase in flow within minutes but never vice versa in seven rats studied. The blood flow increase which was first detected by laser-Doppler flowmetry (35,37) and then verified by dynamic magnetic resonance image (MRI) scanning (65) was observed when Larginine was administered shortly after the induc- tion of ischemia and was greatly diminished when L-arginine was administered 30 minutes after MCA occlusion (64). The attenuation of L-arginine's effect on both blood flow and tissue protection roughly parallels the decrease in tissue N O production after ischemia (40,43). Inactivation of NOS (whole brain) following its early activation was reported by Kader et al. (43) and may reflect proteolysis or accumulation of an enzyme inhibitor within ischemic tissue. In all likelihood, endothelial NOS share a similar fate. The NO binding protein, guanylate cyclase appears more robust and resistant t o the effects of ischemia than NOS (66). Hence, administering NO donors leads to blood flow increases within the ischemic tissue for at least one hour after arterial occlusion and a decrease in infarct size in models of focal ischemia (66,67). One may speculate that NO production by perivascular nerves improve blood flow in a zone of hypoperfusion. Parasympathetic fibers appear particularly relevant because they contain NOS and the vasodilators VIP and possibly acetylcholine within the same neurons (10). Not surprisingly, infarction volume caused by MCA occlusion increases .after chronic parasympathectomy. However, infarct size does not increase after acute parasympathetic sectioning, which suggests the possibility that in this instance, t h e dilator response does not require brain stem activation but rather involves the direct release of transmitters from perivascular nerves by chemicals within the perivascular.space (68,69). Although t h e ameliorative effects of raising blood flow within ischemic tissue was shown more than a decade ago (70), the merits of this treatment strategy have not yet been accepted unanimously (71,72). Opponents argue that increasing rCBF promotes deleterious effects on edema formation (73) and free radical generation during reperfusion (74) and that normal tissue may "steal" blood from the ischemic zone (71). However, recent findings with NO donors (66,67) and L-arginine (36,37) and preliminary data following thrombolytic therapy in stroke patients (75) strongly suggest that early restoration of rCBF can increase tissue survival and restore function. Promoting such changes appears especially critical within t h e ischemic penumbra. However, not all measures which elevate brain blood flow d o so within this region (e.g., hypercapnia) (Fig 2). The potential therapeutic role for drugs which specifically enhance brain blood flow within t h e peri-infarct zone merits further consideration. Studies Using Nitric Oxide Synthase (NOS) Inhibitors Protective effects notwithstanding, it is difficult t o reconcile t h e conflicting literature which report either increases or reductions in infarct volume after permanent or transient MCA occlusion and NOS T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia Table 1 Effect of nitric oxide synthase inhibitors on focal cerebral ischemia in permanent and transient middle cerebral artery occlusion (MCAO) models Species Reference Model Duration of Occlusion Survival Time Inhibitor Infarct Volume Decrease in Infarct Volume Mouse pMCAO 7 days 7 days L-NA 1 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NA 1 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NAME 3 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NAME 10 mg/kg intravenous pre 30 minutes L-NAME 0.7 mg/kg intravenous infusion pre 1 hour L-NA 100 mg/kg intraperitoneal pre 15 hours 72% 1 Rat SD pMCAO 3 days 3 days 44% Rat SHR 84-85 pMCAO 2 days 2 days 35% 1 Cat pMCAO 4 hours 3 hours 2.5 hours 4 hours 5 hours 10 days 35% 1 55% 1 100% 1 Rat SHR tMCAO Rat pup tUCAO Hypoxia Increase in Infarct Volume Rat SHR pMCAO 1 day 1 day L-NA 2.4 mg/kg intravenous infusion post 0-1 hour L-NAME 10 mg/kg intracarotid infusion post 0-1 hour L-NA 1 mg/kg intraperitoneal post 5 minutes, 3 and 6 hours L-NAME 3 mg/kg intraperitoneal post 5 minutes, 3, 6, 24 and 36 hours L-NA 1 mg/kg intraperitoneal post 5 minutes, 3 and 6 hours L-NAME 3 mg/kg intraperitoneal post 5 minutes, 3,6,24 and 36 hours L-NA 2 x 30 mg/kg subcutaneous pre and post at 30 minutes L-NAME 5 mg/kg intravenous pre 15 minutes 32% t 60% Rat SD pMCAO 1 day 1 day Rat SD pMCAO 1 day 1 day 67% T Rat SD pMCAO 2 days 2 days 21% Rat LE pMCAO + UCAO 1 day 1 day 68% T Rat LE pMCAO + UCAO 2 days 2 days NC Rat SD pMCAO 4 hours 4 hours 10% Rat W tMCAO 2 hours 3 days >loo% Abbreviations: p, Permanent; t, Transient; MCAO, Middle cerebral artery occlusion; UCAO, Unilateral common carotid artery occlusion; SD, Sprague-Dawley rat; SHR, Spontaneously hypertensive rat; LE, Long Evans rat; W, Wistar rat; L-NA, Nitro-L-arginine; L-NAME, Nitro-L-arginine methyl ester; pre, Pre-ischemia; post, Post-ischemia; NC, No change. T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia Table 2 Nitrogen monoxtde effects which mpact cerebral ischemia Pos/tive Enhancement of regional cerebral blood flow Inhibition of platelet aggregation Inhibition of platelet and neutrophil adhesion Inhibition of Nmethyl-D-aspartate current Negative Direct cytotoxicity Binding to iron-sulfur complexes (inhib tion of mitochondria1 respiratory enzymes, ribonucleotide reductase, aconitase) Nitrosylation of thiols (inhibition of GAPDH) One simple explanation (with complex ramifications) suggests that N O has both neuroprotective and neurotoxic tissue effects (Table 2). There is already evidence t o support this contention. While current hypotheses and available data suggest that endothelial and perivascular NO have a neuroprotective role, parenchymal NO may exert this dual effect. One may envisage that the predominant action of NO may vary with the tissue compartment (e.g., vascular versus parenchymal), or with the time after the onset o ischemia. Accordingly, the protocol f chosen may favor one or the other of the opposing effects, giving rise t o contradictory results. Some answers may be forthcoming with the development of selective inhibitors for the neuronal or vascular isoforms, or transgenic mice whose neuronal or endothelial NOS are selectively knocked-out, and with a wider application of methods for both directly measuring NO and assessing NOS activity in vivo. Nitrogen Monoxide (NO)-Mediated Neurotoxicity DNA nitration Indirect cytotoxicity via formation of Peroxynitrite Hydroxyl radical Nitrogen dioxide Abbreviations: GAPDH, Glyceraldehyde-3-phosphate inhibition (Table 1). Of course, some of the problems inherent with the use of non-selective NOS inhibitors may underlie some of the existing controversies. However, the observed discrepancies cannot be accounted for by the choice of species, anesthetic, or ischemic model in any simple way although each of these can exert significant effects o n NO synthesis and metabolism. For example, anesthetics like halothane or isoflurane are reported to increase pial arterial diameter with an NO-mediated, yet unknown, mechanism (50). The choice of model may also be important. For example, injury in t h e hypoxiaischemic neonatal rat may be more dependent upon excitotoxic mechanisms and less o n reductions in blood flow than models in which tissue injury is initiated by vascular occlusion only (76). Various species and strains (e.g., normotensive versus hypertensive ratsj possess different susceptibilities t o stroke which may reflect the level of blood pressure elevation, differences in endothelial NOS activity (77), or a greater and lesser extent of collateral blood flow. Severe hypertension caused by NOS inhibitors may also potentiate edema formation and cause bleeding into the core territory (73). Many of these factors are difficult to control and may go unnoticed. In fact, most studies only monitor arterial blood pressure briefly and rarely beyond the anesthetic period; NOS activity within brain is measured only infrequently. NO and its degradation products cause cytotoxicity through formation of iron-NO complexes with several enzymes including complex I and I1 of mitochondrial electron transport, oxidation of protein sulfhydryls and DNA nitration (9,48). The inhibition of glyceraldehyde-3-phosphate dehydrogenase by NO-mediated ADP-ribosylation (78) may have detrimental consequences. This enzyme catalyzes the first step in both glycolysis and the hexose monophosphate shunt and is important for NADP+ synthesis a n d for maintaining intracellular reduced gluthatione levels. NO may mediate cell death also through formation of t h e potent oxidant peroxynitrite (ONOO-) (9,79,80). Furthermore, ONOO- decomposes t o another reactive oxygen species closely related to the hydroxyl free radical ('OH) and to the radical nitrogen dioxide (N02), which is a potent activator of lipid peroxidation (79). Nitrogen Monoxide (NO) and the Wmethyl-DAspartate (NMDA) Receptor NO was proposed as the neurotoxic agent mediating NMDA toxicity. Studies in dissociated cell cultures showed that NOS inhibitors effectively blocked NMDA-induced cell death and that L-arginine depletion attenuated NMDA receptor-mediated toxicity (60). Based o n these studies, it was hypothesized that ischemia-induced NMDA receptor over-activation leads to an increase in intracellular calcium which activates calmodulin and increases NO production. As noted above, it was recently shown that NO also inhibits NMDA currents depending o n its redox state (49). The neurotoxic actions of NO derive from NO' form of t h e molecule whereas N O + form reacts with t h e thiol group of NMDA receptor and blocks its function (49). However, whether NO' or N O + predominates during t h e T. Dalkara and M.A. Moskowitz: Nitric oxide and cerebral ischemia one-hour rise in tissue, NO still remains to be determined. Moreover, formation of peroxynitrite requires a strict co-localization of NO and superoxide productions (80). It appears so that increased NO production in the brain parenchyma may play either a destructive or protective role depending upon the redox state of the tissue, coexisting free radicals and their spatial and temporal relationships. The outcome even becomes less predictable if one considers the possibility that phosphorylation of NOS induced by glutamate activation of protein kinase C may inhibit its activity (62,81). Conclusion 9. Dawson TM, Dawson VL, Snyder SH (1992) A novel neuronal messenger in brain: The free radical, nitric oxide. Ann Neurol 32: 297-31 1 10. Nozaki K, Moskowitz MA, Maynard KI, Koketsu N, Dawson TM, Bredt DS, Snyder SH (1993) Possible origin and distribution of immunoreactive nitric oxide synthase-containing nerve fibers in cerebral arteries. J Cereb Blood Flow Metab 13: 70-79 11. Bredt DS, Hwang PM, Snyder SH (1990) Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347: 768-770 12.Murphy S, Simmons ML, Agullo L, Garcia A, Feinstein DL, Galea E, Reis DJ, Minc-Golomb D, Schwartz JP (1993) Synthesis of nitric oxide in CNS glial cells. Trends Neurosci 16: 323-328 13. Faraci FM (1990) Role of nitric oxide in regulation of basilar artery tone. Am J Physiol 259: HI 216-H 1221 14. Busija DW, Leffler CW, Wagerle LC (1990) Mono-L-arginine-containing compounds dilate piglet arterioles via an endothelium-derived relaxing factor-like substance. Circ Res 67: 1374-1380 15.Rosenblum WI, Nishimura H, Nelson GH (1990) Endothelium-dependent L-arginine and L-NMMA-sensitive mechanisms regulate tone of brain microvessels. A m J PhySiOl 259: H1396-H1401 16.Tanaka K, Gotoh F, Gomi S, Takashima S, Mihara B, Shirai T, Nogaw S, Nagata E (1991) Inhibition of nitric oxide synthesis induces a significant reduction in local cerebral blood flow in the rat. Neurosci Lett 127: 129-132 17. ladecola C (1992) Does nitric oxide mediate the increases in cerebral blood flow elicited by hypercapnia? Proc Natl Acad Sci USA 89: 3913-3916 18.Wang Q, Paulson 0 8 , Lassen NA (1992) Effect of nitric oxide blockade by NG-nitro-L-arginine on cerebral blood flow response to changes in carbon dioxide tension. J Cereb Blood Flow Metab 12: 935-946 19. lrikura K, Maynard KI, Moskowitz M A (1993) The importance of nitric oxide synthase inhibition to the attenuated vascular response induced by topical L-nitro-arginine during vibrissae stimulation. J Cereb Blood Flow Metab (in press) 20. Dirnagl U, Lindauer U, Villringer A (1993) Role of nitric oxide the coupling of cerebral blood flow to neuronal activation in rats. Neurosci Lett 149: 43-36 21. Snyder SH (1993) Janus faces of nitric oxide. Nature 364: 577 22. Pelligrino DA (1993) Saying NO to cerebral ischemia. J Neurosurg Anesthesiol 5: 221-231 23. Knowles RG. Palacios M, Palmer RMJ, Moncada S (1989) Formation of nitric oxide from L-arginine in the central nervous system: A transduction mechanism for stimulation of the soluble guanylate cyclase. Proc Natl Acad Sci USA 86: 5159-5162 24. Bredt DS, Snyder SH (1990) Isolation of nitric oxide synthase, a calmodulin-requiring enzyme. Proc Natl Acad S C USA 87: 682-685 ~ 25. Hevel JM, White KA. Marletta M A (1991) Purification of the inducible murine macrophage nitric oxide synthase: Identification as a flavoprotein. J Biol Chem 266: 2278922791 26.Pollock JS. Forstermann U, Mitchell JA, Warner TD. Schmidt HHHW, Nakane M. Murad F (1991) Purification and characterization of particulate endothelium-derived relaxing factor synthase from cultured and native bovine aortic endothelial cells. Proc Natl Acad Sci USA 88: 1048010484 In conclusion, the discovery of nitric oxide provides a powerful opportunity to explore the pathophysiology and treatment of cerebral ischemia. Controversy abounds, which may reflect the importance of nitric oxide to the diversity of factors (both known as well as unknown) which impact cerebral ischemia. Will drugs that alter NO synthesis and metabolism ever become useful for the treatment of ischemia? Is NO relevant to the pathophysiology of global ischemia? Only time will tell. Nevertheless, answers should be forthcoming as we: (i) develop selective inhibitors for the neuronal or vascular isoforms of NOS, (ii) develop transgenic mice whose neuronal or endothelial NOS are selectively knocked-out, and (iii) discover methods for both directly measuring NO and assessing NOS activity routinely. Acknowledgments Some of the studies described herein were supported by grants from the National Institutes of Health and the Massachusetts General Hospital Health Interdepartmental Stroke Program Project.

Journal

Brain PathologyWiley

Published: Jan 1, 1994

There are no references for this article.