Abstract
Glutamate release after ischemia, hypoxia and seizure activity plays an important role in stimulating adenosine production and release. We characterized the ionotropic glutamate receptor subtype that regulates adenosine levels in vivo and investigated the role of nitric oxide and free radicals in mediating N-methyl-d-aspartate (NMDA)-induced increases in adenosine levels. Rats received unilateral intrastriatal injections and were sacrificed 15 min postinjection by high-energy focused microwave irradiation (10 kW, 1.25 s). Adenosine levels were measured by high-performance liquid chromatography in ipsilateral and contralateral striata. NMDA and kainic acid dose-dependently increased levels of adenosine whereas (±)-α-amino-3-hydroxy-5-methyl-4-isoxazol proprionic acid had no effect. The NMDA- and kainic acid-induced increases were blocked by dizocilpine, and the kainic acid response was decreased by 6-cyano-7-nitroquinoxaline-2,3-dione. The effects of NMDA and kainic acid on levels of adenosine were not additive. Intrastriatall-arginine decreased, and the nitric oxide synthase inhibitor, NG-nitro-l-arginine methyl ester, increased basal adenosine levels. Coadministration of NMDA withl-arginine or NG-nitro-l-arginine methyl ester did not significantly affect NMDA-induced increases in levels of adenosine. N-Tert-butyl-phenylnitrone, a free radical scavenger, reversed l-arginine-induced decreases and NMDA-induced increases in levels of adenosine. Together, these results indicate that NMDA-type ionotropic receptors play an important role in regulating in vivo levels of adenosine in rat striatum and that free radicals, but not nitric oxide, apparently are involved in NMDA-induced increases in levels of adenosine. Conversely, nitric oxide, but not free radicals, apparently exert tonic control over basal levels of endogenous adenosine.
Levels of glutamate, the major excitatory transmitter in the CNS, increase greatly during pathological conditions such as ischemia and hypoxia (Phillis et al., 1991). Glutamate exerts its actions through G-protein-linked metabotropic receptors or cation-selective ionotropic receptors. On the basis of agonist selectivity, ionotropic receptors have been classified as being either NMDA or non-NMDA. Non-NMDA receptors are classified further as kainate- or AMPA-preferring excitatory amino acid receptors (for reviews see Barnes and Henley, 1992; Bettler and Mulle, 1995). Excessive glutamate receptor stimulation, in particular NMDA receptors, has been implicated as a major common pathway leading to neuronal damage in a variety of pathologies (Beal, 1992; Lipton and Rosenberg, 1994; Lynch and Dawson, 1994).
NMDA receptor activation can lead to stimulation of NOS (EC1.14.13.39), the enzyme responsible for the formation of NO (Bredt and Snyder, 1989). NO mediates many of the effects of NMDA, including neurotransmitter release from striatum and cerebral cortex (Montagueet al., 1994; Sandor et al., 1995). In addition to formation of NO, itself a free radical, NMDA receptor activation can lead to the generation of other free radical species (Lafon-Cazalet al., 1993; Beckman et al., 1990). Treatment with free radical scavengers, such as the spin trap agent PBN, can prevent NMDA-induced neurotoxicity and protect against neuronal damage in a variety of in vitro and in vivo models (Schultz et al., 1995b; Nakao et al., 1996;Oliver et al., 1990; Cao and Phillis, 1994).
Levels of adenosine, a mainly inhibitory neuromodulator in the CNS, increase in response to glutamate and selective ligands for ionotropic glutamate receptors both in vitro (Hoehn and White, 1990;Pedata et al., 1991; White, 1996) and in vivo(Jhamandas and Dumbrille, 1980; Perkins and Stone, 1983; Chen et al., 1992; Carswell et al., 1997). Adenosine through its actions to depress basal and evoked neuronal firing, decrease calcium uptake and inhibit release of excitatory neurotransmitters such as glutamate (Wu et al., 1982; Corradetti et al., 1984; Dunwiddie and Diao, 1994) promotes neuroprotective effects, and these actions are mimicked by adenosine receptor agonists and inhibitors of adenosine metabolism and uptake (Rudolphi et al., 1992; von Lubitz et al., 1995; see Geiger et al., 1997).
In many in vivo studies, microdialysis is used to measure extracellular levels of adenosine. However, one of the problems with microdialysis is the large increases observed after implantation of microdialysis probes; adenosine levels may increase more than 75-fold and 24 hr may be required for levels to decline to near basal values (Ballerin et al., 1991). We have circumvented this problem by use of high-energy focused microwave irradiation (10 kW) to prevent postmortem metabolism of adenine nucleotides to adenosine. This method allows for accurate and precise measurement of in vivolevels of adenosine akin to those obtained by freeze-blowing with the added advantage that measurements can be made in discrete brain regions (Delaney and Geiger, 1996).
We previously reported that intrastriatal injections of NMDA increased levels of endogenous adenosine by more than 2-fold (Delaney and Geiger, 1995). The major aim of this study was to characterize more fully the subtype of ionotropic glutamate receptors responsible for regulating levels of adenosine in rat striatum and the involvement of NO and other free radicals in regulating adenosine levels under basal and NMDA-stimulated conditions.
Materials and Methods
Animals.
Male Sprague-Dawley rats were obtained from the University of Manitoba Central Animal Care breeding facility. All procedures followed Canadian Council on Animal Care guidelines and were approved by the Animal Care Committee at the University of Manitoba. Rats used for intrastriatal injections weighed 170 to 190 g, and rats used for NOS assays weighed 200 to 220 g.
Intrastriatal injections.
Animals were anesthetized with 74 mg/kg sodium pentobarbital (i.p.) and placed in a stereotaxic frame; unilateral intrastriatal injections were performed with the coordinates (in mm): AP 9.0, ML 3.0 and DV 4.5 (Paxinos and Watson, 1986). Drugs were administered in a volume of 0.5 μl with a 30-gauge needle for a 2-min period. The needle was left in place for 1 min postinjection to allow diffusion of the drug away from the injection site. NMDA and kainic acid, dissolved in 50 mM Tris-HCl with the pH adjusted to 7.4 with NaOH, were administered at doses ranging from 5 to 150 nmol and 0.125 to 8 nmol, respectively. AMPA was dissolved in 6 N HCl, the pH was adjusted to 7.4 with NaOH and the volume was adjusted with 50 mM Tris-HCl (pH 7.4) for administration of doses ranging from 12 to 30 nmol. Control rats received unilateral injections of 0.5 μl 50 mM Tris-HCl, pH 7.4. Rats were sacrificed by high-energy focused microwave irradiation (Cober Instruments, South Norwalk, CT) at a power level of 10 kW for 1.25 s. Rats were sacrificed 15 min postinjection except in time-course studies when rats were allowed to survive for times ranging from 5 to 45 min after receiving intrastriatal injections. Brains were removed, and striata were dissected and analyzed separately for tissue adenosine content by HPLC with fluorescence detection (Delaney and Geiger; 1996). Protein was determined by the method ofLowry et al. (1951) with bovine serum albumin as standard. Tissue adenosine content was expressed as either picomoles per milligram protein or as a percentage of uninjected contralateral striatum.
To confirm the role of NMDA receptors, dizocilpine (4 mg/kg or 6 mg/kg i.p.) was dissolved in 0.9% saline and administered 30 min before intrastriatal injection of NMDA or kainate. Fifteen minutes after administration of dizocilpine, rats received 37 mg/kg sodium pentobarbital (representing a one half dose because of the sedative effects already manifested by dizocilpine) and after 15 min received intrastriatal injections of NMDA or kainic acid. Levels of adenosine in uninjected contralateral striata of rats receiving either 74 or 37 mg/kg sodium pentobarbital were similar. To prevent hypothermia, rats were kept on a warming pad until taken for microwave irradiation. To confirm the role of kainate receptors, CNQX (5 nmol), dissolved in 0.1 N NaOH with volumes adjusted with 50 mM Tris-HCl, pH 7.4, was administered by itself or coinjected with 0.25 nmol kainic acid. Drugs used to investigate the involvement of NO were dissolved in 50 mM Tris-HCl, pH 7.4. To determine a possible role of free radicals, PBN was dissolved in peanut oil (60 mg/ml) and administered i.p. (150 mg/kg) 1 hr before intrastriatal injection of NMDA orl-arginine (Schultz et al., 1995a; Nakaoet al., 1996). Vehicle controls were used throughout all these studies.
NOS assay.
To confirm inhibition of NOS, rats were injected unilaterally into the striatum with 500 μg l-NAME and sacrificed by decapitation 15 min postinjection. Uninjected and injected striata were excised and analyzed separately for NOS activity by a method adapted from Iadecola et al. (1994). Striata were homogenized (20 strokes) in 1.2 ml 0.32 M sucrose containing 20 mM HEPES, pH 7.4, 1 mM dithiothreitol and 0.5 mM EDTA. Duplicate aliquots of 425 μl tissue were incubated at 37°C for 6 min with 0.45 mM CaCl2·2H2O, 2 mM NADPH, 2 mM l-arginine and 0.2 μCil-[3H]arginine in a final volume of 500 μl. Enzyme activity was terminated by the addition of 2 ml 20 mM HEPES, pH 5.5, containing 2 mM EDTA. Samples were applied to a Dowex AG 50W-X8 (Na+) column, and the eluent plus the effluent from a 2-ml wash of H2O containing [3H]citrulline were collected and radioactivity was determined by scintillation spectrometry (Beckman LS6000TA). Reaction blanks contained everything except NADPH and were treated exactly as above. Enzyme activity was expressed as picomoles per milligram protein per minute.
Chemicals.
Adenosine was obtained from Fisher Scientific (Pittsburgh, PA) as was HPLC grade KH2PO4. Chloracetaldehyde was purchased from Fluka (Ronkonkoma, NY). Burdick & Jackson (Muskegon, MI) supplied HPLC-grade methanol. NMDA, kainic acid, AMPA, HEPES, EDTA,l-arginine, d-arginine, PBN and peanut oil were obtained from Sigma (St. Louis, MO). (+)-MK801 hydrogen maleate (dizocilpine maleate), CNQX and l-NAME were supplied by Research Biochemicals International (Natick, MA) andl-[2,3-3H]arginine (35.7 Ci/mmol) by New England Nuclear (Boston, MA). Dowex AG 50W-X8 was bought from Bio-Rad Laboratories (Mississauga, ONT). All other chemicals were of analytical grade.
Data analyses.
Adenosine levels (pmol/mg protein) in injected striata were calculated as a percentage of levels in uninjected contralateral striata and expressed as mean ± S.E.M. for each drug treatment group. Levels in injected striata were compared with those in the uninjected side with Student’s paired ttests. Differences between drug treatment groups were analyzed either by one-way analysis of variance followed by Tukey-Kramer’s multiple comparisons test or by Student’s unpaired t test. For all tests, statistical significance was considered to be at the P < .05 level.
Results
Levels of endogenous adenosine in uninjected striata of rats were 117 ± 30 pmol/mg protein whereas those in striata injected with buffer were 127 ± 38 pmol/mg protein (data not shown). When calculated as a percentage of uninjected contralateral striatum, levels of adenosine in buffer-injected striata were 105 ± 15%.
Both NMDA and kainic acid when injected unilaterally into striata dose-dependently increased levels of adenosine (fig. 1). Although NMDA and kainic acid showed similar efficacies (adenosine levels were increased by approximately 6-fold), kainic acid with its apparent ED50 of 0.3 nmol was 167 times more potent than was NMDA with its apparent ED50 of 50 nmol. The minimum dose of NMDA required to produce a statistically significant increase in adenosine levels to 238 ± 21% (P < .05) was 10 nmol, and maximal increases of 613 ± 91% were obtained with 150 nmol NMDA. The minimum dose of kainic acid required to produce a statistically significant increase in adenosine levels to 350 ± 67% (P < .05) was 0.25 nmol, and a maximal increase of 591 ± 31% was obtained with 5 nmol of kainic acid. AMPA injected into striata at doses ranging from 12 to 30 nmol did not produce statistically significant increases in levels of endogenous adenosine; maximal increases of 192 ± 30% (P = .26) were observed at 19 nmol (data not shown).
Having found that NMDA and kainic acid both significantly increased levels of endogenous adenosine, we performed time-course experiments to determine the time at which maximal increases would be observed. The doses of 25 nmol NMDA and 0.25 nmol kainic were chosen on the basis of results from initial experiments which suggested that these doses approximately represented ED50 values. Maximally increased levels (271 ± 35%) were observed 15 min after administration of NMDA and levels returned to basal values 45 min postinjection (fig. 2). For kainic acid, maximally increased levels (350 ± 67%) were observed similarly 15 min postinjection and levels returned to basal values 30 min postinjection (fig. 2).
Dizocilpine (4 mg/kg) decreased significantly (P < .05) basal levels of endogenous adenosine to 38 ± 11% regardless of whether data were compared with buffer-injected striata (i.e., control rats) or with uninjected striata (fig. 3). The levels of endogenous adenosine of 75 ± 22 pmol/mg protein in uninjected contralateral striata were not significantly different from those in control rats even though these rats received half doses of anesthetic (37 mg/kg). Dizocilpine (4 mg/kg) decreased significantly (P < .05) the increase in levels of adenosine induced by 25 nmol NMDA from 271 ± 35% to 82 ± 18% (fig. 3). At high doses (100 nmol) of NMDA, dizocilpine at 4 mg/kg and 6 mg/kg significantly decreased levels from 539 ± 69% to 238 ± 53% (P < .05) and to 104 ± 38% (P < .01), respectively (fig. 3).
The competitive kainate/AMPA receptor antagonist CNQX (5 nmol) decreased significantly (P < .05) basal levels of adenosine to 43 ± 14% compared with vehicle-injected striata and decreased kainic acid-induced increases in levels of adenosine from 350 ± 67% to 154 ± 56% (fig. 4). Because kainic acid can release glutamate and subsequently activate NMDA receptors, we administered dizocilpine (4 mg/kg i.p.) 30 min before intrastriatal injection of kainic acid (0.25 nmol) and observed a significant (P < .01) reduction in levels of adenosine from 350 ± 67% to 40 ± 2% (fig. 4). In a separate series of experiments, NMDA (50 nmol) and kainic acid (0.25 nmol) administered in combination had no additive effects on levels of adenosine when compared with the effects of each drug separately (fig. 5).
The following two series of experiments were performed to determine the involvement of free radicals in general, and nitric oxide in particular, in regulating levels of adenosine under basal and NMDA-stimulated conditions. Adenosine levels decreased significantly (P < .01) with 100 nmol l-arginine, but not with the inactive stereoisomer d-arginine (fig. 6). l-NAME (500 μg) significantly (P < .05; Student’s unpaired t test) increased adenosine levels to 230 ± 54% (fig.6) and inhibited NOS activity by 86% (P < .001, table 1). In testing for the involvement of free radicals in l-arginine-induced decreases in adenosine levels, adenosine levels in rats pretreated with PBN (150 mg/kg i.p.) 1 hr before l-arginine was injected intrastriatally were not significantly different from control values (fig. 6). The absolute levels of adenosine (pmol/mg protein) in uninjected contralateral striata of rats that received PBN were slightly, but not significantly, elevated to 167 ± 24 (n = 14).
Coadministration of NMDA with either 100 nmol l-arginine or 500 μg l-NAME did not affect NMDA-induced increases in levels of adenosine (fig. 7). However, levels of adenosine in rats pretreated with PBN were decreased significantly (P < .05). Preinjection (i.p.) of peanut oil, the vehicle for PBN, followed by intrastriatal NMDA, increased levels of adenosine by 249 ± 61%, which indicates a lack of vehicle effects (data not shown).
Discussion
Adenosine is produced in and released from CNS tissue preparations in response to a variety of conditions including, but not limited to, depolarization, ATP metabolism and glutamate release (Berne et al., 1974; Van Wylen, 1986; Phillis et al., 1991). Evidence that adenosine analogs and inhibitors of adenosine transport and metabolism can inhibit glutamate release and protect against pathological consequences associated with glutamate release (Corradettiet al., 1984; Finn et al., 1991) suggests that adenosine may represent a “brake” against prolonged glutamate receptor activation and excitotoxicity. We reported previously that NMDA injected into rat striatum increased levels of endogenous adenosine (Delaney and Geiger, 1995), and here, we characterized the ionotropic glutamate receptor subtype that regulates adenosine levelsin vivo and investigated the role of NO and free radicals in regulating adenosine levels under basal and NMDA-stimulated conditions.
In all these studies, our in vivo protocol required the use of pentobarbital anesthesia. However, the anesthetic did not affect adenosine values because adenosine levels in nonanesthetized, noninjected rats or rats receiving half doses (37 mg/kg) of sodium pentobarbital were not significantly different from levels in rats receiving a full dose of anesthetic. Also, although the doses of NMDA and kainate used here were similar to those that produce, over much longer times (3–7 days), excitotoxic lesions in striatum (Finnet al., 1991; Globus et al., 1995; Schultzet al., 1995a), our findings that total tissue levels of adenosine peaked 15 min postinjection and returned to basal levels by 45 min postinjection argues against a nonspecific release of adenosine because of cell death. Indeed, release of adenosine from hippocampusin vivo showed a similar time scale for peak release and return to basal levels in response to glutamate receptor agonists (Chenet al., 1992; Carswell et al., 1997). Furthermore, AMPA in doses similar to those used here can cause excitotoxic damage in striatum (McDonald et al., 1992), but did not significantly increase adenosine levels.
NMDA-induced increases in levels of adenosine apparently were mediated through specific ionotropic NMDA receptors because the responses to NMDA were blocked by the noncompetitive use-dependent NMDA receptor antagonist dizocilpine. Although systemic administration of dizocilpine did not alter levels of adenosine in uninjected striata, it did significantly decrease levels of adenosine in buffer-injected striata. These findings suggest that there was no tonic regulation of adenosine levels by glutamate in uninjected striata, but that mechanical injury from needle insertion may have decreased the voltage-dependent Mg++ blockade of NMDA receptors (Zhang et al., 1996), thereby increasing their basal activity and/or increased the release of endogenous glutamate. Because dizocilpine acts use-dependently, “priming” of the NMDA receptor in injected striata, regardless of how or at what point during the surgery it occurs, would lead to a greater degree of blockade than in uninjected striata. Thus, the final outcome of decreased NMDA receptor activity in the injected striatum was decreased adenosine levels.
Kainic acid increased levels of adenosine, but apparently did so through activation of kainate and subsequently NMDA receptors, because CNQX and dizocilpine decreased levels of adenosine in buffer-injected striata and blocked kainate-induced increases in adenosine. Dizocilpine was tested because kainate induces glutamate release (Ferkany and Coyle, 1983; Young et al., 1988) and the effects of kainate then may be mediated through other glutamate receptor subtypes (reviewed by Coyle, 1983). Although dizocilpine at 4 mg/kg was more effective than 5 nmol CNQX, it is difficult to compare the potencies of these two antagonists because dizocilpine works use-dependently and CNQX is a competitive blocker. Even though CNQX may block glycine sites on the NMDA receptor (Lester et al., 1989), we believe that CNQX was acting at kainate sites because the concentrations required to block NMDA receptors are much higher than those required to inhibit AMPA/kainate receptor activation and because CNQX would have to compete with endogenous glycine, which is thought to be present in supersaturating concentrations in vivo (Birch et al., 1988). Findings that dizocilpine blocked the effects of kainate, combined with data showing that the effects of NMDA and kainate were not additive, suggests that in striatum, at least, glutamate released by kainate activates NMDA receptors and results in higher levels of adenosine. This seems to contrast with results showing that NMDA receptors do not appear to play a major role in kainate-induced increases in levels of adenosine in hippocampusin vivo (Carswell et al., 1997). However, our data do support and extend findings that kainic acid and NMDA increase adenosine release in vitro (Hoehn and White, 1990; Craig and White, 1993b; Manzoni et al., 1994), as well as in vivo (Perkins and Stone, 1983; Chen et al., 1992;Carswell et al., 1997).
In contrast to NMDA and kainate, intrastriatal injection of AMPA did not affect adenosine levels significantly. Similarly, AMPA had minimal effects on adenosine release from hippocampal slices (Pedata et al., 1991). However, AMPA and quisqualate (a nonselective agonist with activity at the AMPA receptor) increased adenosine release in cortical slices (Hoehn and White, 1990; Craig and White, 1993b; White, 1996). Even though cortical, striatal and hippocampal brain regions all contain NMDA, AMPA and kainate receptors (Cotman and Monaghan, 1986;Young and Fagg, 1990), there appear to be regional differences in the ability of these receptor subtypes to evoke increases in levels of adenosine. There is evidence for a common site of action for kainic acid and AMPA as well as for kainate acting at a unique set of receptors (Barnes and Henley, 1992); however, because AMPA had no significant effect on levels of adenosine, it seems clear that kainate acted through its own receptors.
The second aim of this study was to test the hypothesis that NO and/or free radicals regulate levels of endogenous adenosine under basal as well as NMDA-stimulated conditions. A reciprocal relationship appears to exist in vivo between levels of NO and basal levels of endogenous adenosine such that l-arginine decreased whereasl-NAME increased levels of adenosine. Confidence in our results is raised by findings that l-arginine andl-NAME had opposite effects on basal levels of adenosine and that the inactive stereoisomer of l-arginine,d-arginine, had no effect. In addition, the effects ofl-arginine were blocked by the spin trap agent PBN that scavenges free radicals by forming more stable free radical adducts (Knecht and Mason, 1993). Furthermore, our findings withl-NAME are consistent with findings in guinea pig brain and rabbit heart, which show that basal levels of adenosine increase after perfusion with l-NAME (Kostic and Schrader, 1992; Woolfsonet al., 1995), and that the NOS inhibitor NG-nitro-l-arginine potentiated adenosine-mediated hypocapnic vasodilation in rat brain (Fabricius and Lauritzen, 1994). Results with NO donors are less clear. NO donors reportedly decrease (Craig and White, 1993a) and increase (Fallahiet al., 1996) adenosine release in vitro, increase adenosine release in vivo (Fischer et al., 1995) and inhibit (Siegfried et al., 1996) and stimulate (Minamino and Hasegawa, 1995) the activity of the adenosine-producing enzyme 5′-nucleotidase. Although we did not study the relationship between striatal levels of adenosine, NO and blood flow, our results may suggest that levels of NO and adenosine are altered reciprocally as a mechanism for controlling local cerebral blood flow.
Many effects of NMDA, including neurotransmitter release, are mediated by NO (Montague et al., 1994; Sandor et al., 1995). However, this does not appear to be the case with adenosine levels, because neither l-arginine nor l-NAME affected levels of endogenous adenosine increased by NMDA. Similarly, NO apparently did not mediate NMDA-evoked adenosine release from rat cortical slices (Craig and White, 1993a). However, it is possible that in the presence of NMDA a role for NO regulation of adenosine levels, as seen under basal conditions, may still exist, but the effects have been masked by other processes subsequent to receptor activation.
Administration of PBN reversed l-arginine-induced decreases in basal levels of endogenous adenosine possibly through the removal of NO (itself a free radical). With respect to NMDA, determining the nature of the free radical(s) involved is more complicated because NMDA generates superoxide radicals (Lafon-Cazal et al., 1993), which in combination with NO results in the formation of peroxynitrite, thus leading to lipid peroxidation and production of additional free radicals (Beckman et al., 1990). Free radicals inhibit glutamate uptake (Volterra et al., 1994) as well as disrupt mitochondrial function and ATP production (Brown and Squier, 1996), and these effects could lead to increases in levels of endogenous adenosine. PBN can trap carbon-centered free radicals such as those resulting from lipid peroxidation (Knecht and Mason, 1993) as well as combine, although less readily, with hydroxyl radicals (Thomas et al., 1994). Despite such ambiguities in the identity of the free radicals involved, it remains clear that PBN is very effective at preventing NMDA-induced neuronal damage (Schultz et al., 1995a; Nakao et al., 1996; Lafon-Cazal, 1993) and, perhaps in doing so, blocks the stimulus that leads to increased levels of adenosine.
In an in vivo preparation such as ours, it would be expected that both endothelial and neuronal isoforms of NOS would contribute to NO production, and because l-NAME is an inhibitor of both, there would be no distinction between the relative contributions of each isoform. Under basal conditions, endothelial NOS activity might be the predominantly active isoform, whereas under NMDA-stimulated conditions, neuronal NOS might be activated, in addition to endothelial NOS, as part of the signal transduction system. Changes in the relative activities of each isoform can produce vastly different effects (Globuset al., 1995), and if neuronal NOS activity increases levels of adenosine, this might counteract endothelial NOS-induced decreases in levels of adenosine and produce no overall changes in adenosine levels. The importance of different isoforms of NOS is being recognized increasingly when looking at the effects of NO (Iadecola 1997; Snyder, 1995; Globus et al., 1995), and the effects of NO on adenosine levels may be another example where this must be considered.
Adenosine agonists prevent both NMDA and kainate striatal toxicityin vivo (Finn et al., 1991). Thus, through application of adenosinergic strategies, excitotoxic damage may be preventable and/or treatable especially when directed at events in which NMDA and possibly kainate receptors are involved.
Footnotes
-
Send reprint requests to: Dr. J. D. Geiger, Department of Pharmacology and Therapeutics, University of Manitoba, 753 McDermot Avenue, Winnipeg, Manitoba, R3E 0T6 Canada.
-
↵1 These studies were supported by a grant from the Medical Research Council of Canada (to J.D.G.).
-
↵2 Recipient of a Medical Research Council of Canada Studentship. Current address: Department of Pharmacology, University of North Carolina, Chapel Hill, NC 27599.
-
↵3 Recipient of a National Science and Engineering Council Studentship award.
-
↵4 Recipient of a Medical Research Council of Canada Scientist award.
- Abbreviations:
- AMPA
- (±)-α-amino-3-hydroxy-5-methyl-4-isoxazol proprionic acid
- CNQX
- 6-cyano-7-nitroquinoxaline-2,3-dione
- l-NAME
- NG-nitro-l-arginine methyl ester
- NMDA
- N-methyl-d-aspartate
- NO
- nitric oxide
- NOS
- nitric oxide synthase
- PBN
- N-tert-butyl-phenylnitrone
- HPLC
- high-performance liquid chromatography
- HEPES
- N-2-hydroxyethylpiperaine-N′-ethanesulfonic acid
- EDTA
- ethylenediaminetetraacetic acid
- CNS
- central nervous system
- Received August 28, 1997.
- Accepted January 23, 1998.
- The American Society for Pharmacology and Experimental Therapeutics
References
Levels of endogenous adenosine in striata of rats (n = 3–9) receiving unilateral injections of NMDA at doses ranging from 5 to150 nmol (circles) or kainic acid at doses ranging from 0.125 to 8 nmol (squares). Levels in injected striata were expressed as a percentage of levels in contralateral uninjected striata. Levels of adenosine were increased significantly (* P < 0.05; ** P < 0.01) compared with uninjected striata.
Levels of endogenous adenosine in striata of rats (n = 4–8) 5 to 45 min after receiving unilateral intrastriatal injections of 25 nmol NMDA (circles) or 5 to 30 min after receiving unilateral intrastriatal injections of .25 nmol kainic acid (squares). Levels in injected striata were expressed as a percentage of levels in uninjected contralateral striata. Levels of adenosine were increased significantly (* P < .05) compared with uninjected striata.
Levels of endogenous adenosine in striata of rats receiving unilateral intrastriatal injections of 50 mM Tris-HCl buffer, pH 7.4 (n = 7), buffer plus 4 mg/kg dizocilpine i.p. (n = 4), 25 nmol NMDA (n = 8), 25 nmol NMDA plus 4 mg/kg dizocilpine i.p. (n = 3), 100 nmol NMDA (n = 6), 100 nmol NMDA plus 4 mg/kg dizocilpine i.p. (n = 3) or 100 nmol NMDA plus 6 mg/kg dizocilpine i.p. (n = 3). Levels in injected striata were expressed as a percentage of levels in uninjected contralateral striata. * Levels of adenosine were decreased significantly (P < .05) compared with uninjected striata. Levels of adenosine were increased significantly (*P < .05; **P < .01) compared with uninjected striata. Dizocilpine significantly decreased levels of adenosine compared with buffer (P < .05) or with 25 nmol NMDA (P < .01). Levels of adenosine in the presence of 100 nmol NMDA were decreased significantly by 4 mg/kg dizocilpine (P < .05) and 6 mg/kg dizocilpine (P < .01).
Levels of endogenous adenosine in striata of rats receiving unilateral intrastriatal injections of vehicle (n = 9), 5 nmol CNQX (n = 7), 0.25 nmol kainic acid (n = 5), 0.25 nmol kainic acid plus 5 nmol CNQX (n = 4) or 0.25 nmol kainic acid plus 4 mg/kg dizocilpine i.p. (n = 4). Levels in injected striata were expressed as a percentage of levels in uninjected contralateral striata. * Levels of adenosine were significantly different (P < .05) compared with uninjected striata. CNQX significantly decreased levels of adenosine compared with vehicle controls (P < .05. Levels of adenosine in the presence of 0.25 nmol kainic acid were decreased significantly by 5 nmol CNQX (P < .05) and 4 mg/kg dizocilpine i.p. (P < .01).
Levels of endogenous adenosine in striata of rats receiving unilateral intrastriatal injections of 50 nmol NMDA (n = 3), 0.25 nmol kainic acid (n = 3) or 50 nmol NMDA plus 0.25 nmol kainic acid (n = 3). Levels in injected striata were expressed as a percentage of levels in uninjected contralateral striata.
Levels of endogenous adenosine in striata of rats receiving unilateral injections of 500 μg l-NAME (n = 9), 100 nmol l-arginine (n = 9), 100 nmol l-arginine 1 hr after receiving 150 mg/kg PBN (n = 6) or 100 nmold-arginine (n = 8). * P < .05, ** P < 0.01 ipsilateral compared with contralateral uninjected striata. Levels of adenosine were significantly increased byl-NAME (P < .05) and significantly decreased (P < .01) by l-arginine.
Levels of endogenous adenosine in striata of rats receiving unilateral injections of 25 nmol NMDA alone (n = 8), NMDA coadministered with either 100 nmoll-arginine (n = 4), 500 μgl-NAME (n = 8), or 25 nmol NMDA 1 hr after receiving 150 mg/kg PBN (n = 8). * P < .05, ** P < .01 ipsilateral compared with contralateral uninjected striata. Adenosine levels in rats pretreated with PBN were significantly (P < .05) lower than in rats receiving NMDA alone.
Effect of 500 μg intrastriatal l-NAME on NOS activity1-a
