Kainic acid

Objectives: The formation and release of adenosine following graded excitatory stimulation of the brain may serve important physiological functions such as sleep regulation, as well as an early resistance mechanism against excitotoxicity. However, adenosine at high levels may reflect merely the results of obstructed energy metabolism.

Methods: We examined the extent to which levels of adenosine and adenylate energy charge are affected in vivo by graded excitatory stimulations of brain using unilateral intrastriatal injections of glutamatergic agents and head-focused high energy microwaving for accurate and precise measures of purines.

Results: Our results confirmed that adenosine levels rise when adenylate energy charge decreases and showed that these increases occurred in three distinct phases with the rate of adenosine formation in each phase increasing as tissue adenylate energy charge was further depleted. In addition, we observed that, in most cases, the effects of focal excitatory stimulation on changes in tissue purine levels were restricted spatially within the immediate vicinity of the injection site; however, when strongly depolarizing stimuli were used, changes in purine levels could be observed in adjacent and, occasionally, even in contralateral brain regions.

Discussion: These results provide new insight into purine regulation that occurs under physiologically relevant conditions, such as sleep and during the early stages of brain insults that induce excitotoxicity. [Neurol Res 2005; 27: 139-148]

Keywords: Adenosine; adenylate energy charge; ATP; glutamate; microwave irradiation

INTRODUCTION

When ATP levels are compromised, as might occur during sleep deprivation and brain ischemia, a cascade of harmful excitotoxic events may be initiated1,2. During excitoxicity, released glutamate perpetuates neuronal depolarization through activation of AMPA and kainate receptors, alters intracellular Ca^sup 2+^ signaling through activation of NMDA receptors and limits its removal by reversing glutamate uptake5. In these early stages of excitotoxicity adenosine levels rise rapidly as demonstrated with in vivo whole brain and in vitro tissue slice and cell culture models4-10. Adenosine accumulating from ATP metabolism associated with elevated cellular activity can stimulate compensatory responses, such as vasodilation and attenuation of neurotransmission giving this endogenous compound the distinction of retaliatory metabolite11-16.

The actions of adenosine are mediated through high affinity A^sub 1^, slightly lower affinity A^sub 2A^, and very low affinity A^sub 2B^ and A^sub 3^ cell-surface G-protein coupled receptors17. Adenosine levels are highest at the core of an insult and activated receptors initiate compensatory responses that couple physiological responses to energy status18. Under basal conditions when adenosine levels are lowest, activation of A^sub 1^ receptors may contain the spread of excitatory signaling by limiting glutamate release and improve cell viability in models of excitotoxicity13,19-25. Where adenosine levels are sufficiently elevated to activate A^sub 2A^ receptors, both protective actions such as vasodilation, as well as harmful effects including desensitization of protective A^sub 1^ receptor-mediated actions, potentiation of glutamate release and activation of inflammatory responses have been reported26. Under conditions where tissues are damaged, adenosine levels are high enough to stimulate A^sub 30^ adenosine receptors and cause apoptosis and astrocyte proliferation18,27.

Because of the extremely labile nature of ATP, ADP, AMP and adenosine, extreme measures including in situ freezing with liquid nitrogen, freeze blowing and head-focused high energy microwave irradiation have been used to prevent post-mortem changes in the levels of these compounds28-30. Of these methods, only high-energy microwave irradiation prevents postmortem purine metabolism while preserving the structural integrity of the brain. Using this technique with our focal excitotoxicity model we found that adenosine levels not only increased with adenylate energy charge depletion, but did so in distinct phases at increasing rates suggesting that different adenosineaccumulating mechanisms operate within specific adenylate energy charge ranges.

METHODS

Animals

Adult male Sprague-Dawley rats obtained from the University of Manitoba Central Animal Care breeding facility weighing 200 ± 20 g were used in these studies. Rats were allowed at least 5 days to acclimatize before procedures were carried out. All protocols were performed in accordance with University of Manitoba Animal Care Ethics Committee guidelines.

Chemicals

L-glutamate (monosodium salt), kainic acid, N-methylD-aspartate (NMDA) and tri-n-octylamine (TOA) were purchased from Sigma Chemical Company (St Louis, MO). Chloroacetaldehyde was obtained from Fluka (Ronkonkoma, NY). 5-Methyl-10,11-dihydro-5Hdibenzocyclohepten-5,10-imine maleate (MK-801), 6cyano-7-nitroquinoxaline-2,3-dione (CNQX), 2,3-dihydroxy6-nitro-7-sulphamoylbenzo(F)-quinoxaline (NBQX), L-(-)threo-3-hydroxyaspartic acid (HAsp) and (1S,3R)-1aminocyclopentane-1,3-dicarboxylic acid (ACPD) were purchased from Tocris Cookson (Ballwin, MO). All HPLC reagents were ordered from Fisher Scientific (Nepean, ON).

Intrastriatal injections

Drug doses used in each experiment are described in the figure legends. Tris injection vehicle was composed of 50 mM Tris. Artificial CSF (aCSF) vehicle contained 124 mM NaCI, 5 mM KCl, 0.1 mM CaCl^sub 2^, 1.2 mM MgCl^sub 2^, 26 mM NaHCO^sub 3^, 10 mM glucose. Salt solution was composed of 140 mM NaCl, 4.7 mM KCl, 1.2 mM MgSO^sub 4^, 2.5 mM CaCl^sub 2^, 10 mM glucose, 1.4 mM KH^sub 2^PO^sub 4^ and 4.3 mM Na^sub 2^HPO^sub 4^. All vehicle and drug solutions were prepared on the day of use and pH levels were always adjusted to 7.4 using either HCl or NaOH. MK-801 was administered i.p. (4 mg/kg) in a 0.9% NaCl solution 30 minutes prior to intrastriatal injection of drugs.

Surgical procedures

Rats were anaesthetized with sodium pentobarbital (74 mg/kg i.p.) and placed in a stereotaxic frame. All drug injections were directed to the middle of the right striatum. From the intersection of the midline and bregma on the surface of the skull, the injection needle was moved 3.0 mm laterally and then lowered 4.5 mm ventrally through a small burr hole31. Drug injections were delivered through a 30gauge needle in either 0.5 or 1.0 μl vols. The 1.0 μl injections were administered over 2 minutes and an additional 2 minutes was allowed for drug diffusion before needle removal. For 0.5 μl injections, injection and diffusion times were 1 minute each. Following drug injection, animals were placed on a heating pad to maintain normal body temperature and were killed by microwave irradiation 15 minutes after drug injection. For cortex stimulation experiments, we mechanically stimulated the cortex with forceps for 3-5 seconds within the area of the burr hole to test whether cortical stimulation had any effect on purine levels in the striatum. For no-volume experiments, the injection needle was inserted and left in place for 4 minutes; however, no fluid volume was delivered.

Microwave irradiation and tissue dissection

Following intrastriatal injection, animals were killed by focused microwave irradiation (Cober Electronics) at a power setting of 10 kW for 1.5 seconds as described previously4-6. Immediately following microwave irradiation, brains were removed, and whole striata were dissected and then cut coronally into three portions of equal rostrocaudal length designated anterior (mainly caudate-putamen), middle (mainly striatum) and posterior (striatum plus globus pallidus). Tissue samples were stored at -80°C until analysis by high performance liquid chromatography (HPLC).

Tissue preparation

Frozen tissue samples were weighed in plastic centrifuge tubes (80Ti-type, Beckman) containing 600 μl of ice-cold 2% trichloroacetic acid (TCA) and volumes were then adjusted to yield concentrations between 10 and 20 mg tissue/ml of TCA. Tissues were homogenized using a Polytron homogenizer (Brinkmann) at setting #6 and 50 μl aliquots of each homogenate were removed and stored on ice for subsequent protein analysis. Remaining homogenates were centrifuged at 4°C for 15 minutes at 25,000 g and 500 μl of each resulting supernatant were neutralized with an equal volume of tri-n-octylamine/freon (45:155) in 1.5ml microcentrifuge tubes. Samples were mixed by vortex for 7 seconds, centrifuged for 2 minutes at 13,000 g to separate organic and aqueous phases, and then placed on ice. The top aqueous layer of each sample was collected for analysis by HPLC.

Analyses of purines, protein and data

The analysis of adenosine, ATP, ADP and AMP by HPLC was conducted in duplicate as described previously30. For adenosine and adenine nucleotide analyses, compounds were identified by comparison to standards and peak areas were quantified using Millennium v.3.05 software (Waters). Protein content of samples was determined as described using a commercial kit (BioRad)32. All data were reported as mean ± SEM. Statistical comparisons between two groups (contralateral versus ipsilateral) were carried out using unpaired two-tailed f-tests and statistically significant differences were annotated by asterisk (*). For comparisons between more than two data groups, ANOVA was carried out followed by a Student-Neuman-Keuls multiple comparisons test and statistically significant differences were indicated by a dagger ([dagger]). Multiple regression analyses were carried out using GraphPad Prism software. For assessing the energy state of each tissue sample, we used the adenylate energy charge parameter:

EC = ([ATP] + 1/2 [ADP])/([ATP] + [ADP] + [AMP])

where the maximum EC value approaches 1.033. This equation expresses the ratio of hydrolysable high-energy phosphate bond-containing adenine nucleotides to the entire adenine nucleotide pool, giving an indication of the short-term metabolic potential energy of a tissue.

RESULTS

Focal excitatory stimulation in whole striatum

Excitatory amino acid (EAA)-induced increases in brain adenosine levels have been characterized using the glutamate receptor agonists AMPA, kainate, and NMDA4-6. Here, using L-glutamate, we found that an injection of 1000 nmol produced large and statistically significant (p

Concurrent analysis of adenosine and nucleotide levels in striatal subregions

Using the excitatory amino acid analog kainic acid, alone or in combination with the glutamate uptake blocker HAsp34, we investigated the extent to which increasing excitatory stimulation affected tissue adenine nucleotide and adenosine levels. Comparing contralateral and ipsilateral middle subregions, intrastriatal injection of kainate produced a significant decrease (6%, p

Adenosine level responses to stimulation intensity

To determine how adenosine levels changed relative to the degree of excitatory stimulation, we tested the effects of different stimuli having increasing depolarizing strength and ability to activate glutamate receptors (Figure 4). With mechanical stimulation of the cortex we observed only sporadic increases in contralateral middle striatal subregion adenosine levels (152 ± 59 pmol/mg protein) that on average were not statistically different from corresponding ipsilateral subregions (61 ± 14 pmol/mg protein). Although intrastriatal needle insertion produced adenosine level increases in ipsilateral subregions (134 ± 38 pmol/mg protein) compared with corresponding contralateral regions (48 ± 9 pmol/mg protein), these changes were not statistically significant (p=0.052).

Mg^sup 2+^-free Tris vehicle by itself induced statistically significant adenosine level increases (Figure 1) possibly because voltage-dependent Mg^sup 2+^ block of NMDA receptors was relieved. In contrast, intrastriatal injection of aCSF or salt solution, both containing physiological levels of Mg^sup 2+^, produced adenosine levels of 133 ± 56 and 151 ± 56 pmol/mg protein that were not statistically different from levels in corresponding contralateral subregions, but were significantly lower (p

The excitatory potential of Tris was further evident in injection experiments where NMDA was delivered in either aCSF or Tris; adenosine levels were 309 ± 64 with aCSF and were significantly higher (p

Relationship between adenylate energy charge, adenine nucleotides and adenosine levels

Pooling data from the experiments described in Figures 3 and 4 we used the adenosine and adenine nucleotide levels from the resulting 363 striatal subregions to correlate adenosine levels with adenylate energy charge depletion. Plotting adenosine levels as a function of adenylate energy charge, we observed that adenosine levels increased with declining adenylate energy charge in three distinct stages that we have termed phase 1, 2, and 3 (Figure 5A). Phase 1 (Figure 5B), encompassing the adenylate energy charge range between 0.90 and 0.75 where most of our data was concentrated, had a slope value of 1170 ± 100 (r^sup 2^ = 0.90). In phase 2 (Figure 5C), adenosine levels in the 0.74-0.60 adenylate energy charge range increased more rapidly with adenylate energy charge depletion yielding a slope value of 4080 ± 730 (r^sup 2^=0.71). Although we had limited data from the phase 3 region (Figure 5D), the adenylate energy charge in the range of 0.59-0.45 was highly correlated (r^sup 2^ = 0.92) with adenosine levels with a slope of 15,850 ± 1860. The three slopes were significantly different from each other (p

Using the same data pool described in Figure 5, we averaged the ATP, ADP and AMP values that gave rise to each adenylate energy charge level and plotted nucleotide levels as a function of their corresponding adenylate energy charge (Figure 6). At an adenylate energy charge of 0.90, ATP, ADP and AMP levels were 30.2 ± 4.2, 6.0 ± 1.0, and 0.7 ± 0.1 nmol/mg protein, respectively (Figure 6). As adenylate energy charge levels decreased from 0.90, we observed a gradual decline in ATP levels and a corresponding increase in AMP levels to a point where ATP, ADP and AMP levels merged at an adenylate energy charge value of approximately 0.55. This pattern is very similar to that proposed in the original EC equation model33.

Subregion data distribution

From the pooled data (Table 7) we observed that 81% of contralateral subregions tell within the phase 1 range. Compared with contralateral data, ipsilateral data were shifted toward the lower boundary of phase 1 and into phase 2, a trend particularly evident with the middle (injected) ipsilateral subregion data. Minimally invasive treatments (group 1) and those involving innocuous drug vehicles containing physiological levels of Mg^sup 2+^ (group 2) yielded subregions that fell predominantly in the phase 1 range. As treatments began to increase in excitatory potential (group 3) there was a shift of subregion adenylate energy charge values from phase 1 to phase 2, and this trend was most evident with the most strongly depolarizing treatments that would best activate NMDA receptors (group 4). Although we rarely observed tissue adenylate energy charge levels in the phase 3 range, these values were most frequent with the strongest excitatory treatments.

DISCUSSION

Under basal conditions, brain adenosine levels are nearly 1000-fold lower than ATP levels and therefore from a stoichiometric standpoint, even a very small drop in ATP levels has the potential to produce large adenosine level increases. Our data showed that adenosine levels increased as tissue energy levels were depleted following excitatory stimulation. However, this relationship was not stoichiometric, but appeared to occur in three distinct phases each defined by a linear relationship with a unique slope. Although the three phases of adenosine level regulation were arbitrarily chosen, experimental evidence exists for differential activation of glutamate and adenosine receptors under conditions of varying excitatory stimulation or depolarization. It follows, therefore, that with a decline in tissue energy state, different glutamate and adenosine-related mechanisms can be selectively activated, resulting in different rates of adenosine formation. In addition, our data indicate that different types of adenosine regulation occur between adjacent tissue regions and, in some cases, adenosine levels can be affected in brain regions distant from the site of stimulus.

Phase 1 regulation

Assuming that contralateral striatal subregions were largely unaffected by injections into the opposite side of the brain, we interpreted the large distribution of these data in the phase 1 range to be indicative of energy levels occurring in the absence of significant external excitatory stimulation. This interpretation is supported by studies showing that basal adenylate energy charge levels in whole brain range between 0.96 and 0.80, and that measurable changes in brain adenosine levels can be observed only when basal neuronal signaling is inhibited28,30,35. In addition, the injection treatments expected to have minimal excitatory effects, including those carried out in the presence of CNQX and NBQX, as well as group 1 treatments, produced adenosine levels in the phase 1 range. Within phase 1, adenylate energy charge levels appear to be representative of quiescent (0.90-0.85), medium (0.84-0.80) and elevated (0.79-0.75) levels of neuronal activity given the changes in phase 1 data distribution following treatments with increasing excitatory potential.

AMPA and kainate receptors mediate fast excitatory signaling, whereas NMDA receptors require prior depolarization to alleviate voltage-dependent Mg^sup 2+^ block36,37. Adenosine level increases in brain tissue have been shown following activation of AMPA, kainate and NMDA-type glutamate receptors and NMDA receptor antagonists only affected large adenosine level increases induced by strongly depolarizing electrical or K+ stimuli6,8,35,38,39. Therefore, it appears that moderate adenosine level increases can occur with non-NMDA glutamate receptor activation; however, when the degree of depolarization reaches the threshold to allow significant activation of NMDA receptors, additional adenosine forming mechanisms are activated to further boost adenosine levels. Illustrating this point, we saw considerable differences in adenosine level increases when comparing the effects of either vehicle or NMDA injections in the absence or presence of physiological levels of Mg^sup 2+^ suggesting that NMDA receptor activation played a significant additive role. Large-scale activation of NMDA receptors, we propose, is the defining mechanism that separates phase 1 and phase 2 adenosine level regulation. In phase 1, we would expect that the degree of depolarization, as reflected by tissue adenylate energy charge levels, is insufficient to cause widespread activation of NMDA receptors or release of endogenous glutamate. A distinct adenylate energy charge or adenosine level threshold for widespread NMDA receptor activation appears to exist because released adenosine can inhibit excitatory conduction through activation of high affinity A^sub 1^ receptors and adenosine formed with low level NMDA receptor activation provides an inhibitory threshold against additional NMDA receptor activation15,40,41. Together, these mechanisms would resist excitatory depolarization and the transition from phase 1 to phase 2 adenosine level regulation.

Phase 2 regulation

With injection treatments expected to have moderate depolarizing effects (group 3) we observed a large shift in adenosine and adenylate energy charge values to the phase 2 range. The group 3 treatments involving the strong excitatory agents NMDA and glutamate were counterbalanced by the inclusion of NMDA receptor inhibitors Mg^sup 2+^ and Mg^sup 2+^/MK-801. Kainate, although a potent excitotoxin, was used at a very low dose (0.25 nmol) compared with doses used in other studies (10 nmol), showing significant kainate toxicity42. These treatments were therefore expected to induce moderate depolarization and NMDA receptor activation, and indeed yielded adenylate energy charge levels in the upper phase 2 range. The majority of data in phase 2, however, came from strongly depolarizing group 4 treatments.

Mechanical manipulation of brain tissue, which can reduce voltage-dependent block of NMDA receptors, together with the administration of Mg^sup 2+^ free agents probably had synergistic effects by dis-inhibiting NMDA receptors resulting in the large adenosine level increases caused by Tris vehicle alone43. By adding NMDA to the Tris vehicle, adenosine level increases were boosted by almost the same magnitude as when HAsp was added to kainate injections. The additive effect of kainate/HAsp injections likely resulted from significant NMDA receptor activation induced by extracellular glutamate given that kainate causes presynaptic glutamate release in the striatum and HAsp further increases extracellular glutamate levels through hetero-exchange mechanisms involving glutamate transporters34,44,45. Therefore, phase 2 adenosine level regulation begins when moderate depolarization results in significant NMDA receptor activation. As the depolarizing stimulus increases in intensity we would anticipate more endogenous glutamate release, further depolarization and additional NMDA receptor activation resulting in a continuous drop in tissue adenylate energy charge and a consequent rise in adenosine levels throughout the phase 2 region46,47.

The relative affinities of adenosine A^sub 1^ and A^sub 2A^ receptors for adenosine are approximately 70 and 150 nM, respectively and the adenosine concentrations midway through phase 1 and phase 2 ranges are approximately 100 and 600 pmol/mg protein, respectively17. Therefore, it is tempting to speculate that if phase 1 is indicative of adenosine levels under basal conditions where A^sub 1^ receptors are preferentially activated, then in phase 2, where tissue adenosine levels are approximately 6 times higher, adenosine A^sub 2A^ receptors may mediate the actions of adenosine. Whereas adenosine A^sub 1^ receptor mediated effects are neuroprotective under excitatory conditions, the role of A^sub 2A^ receptor activation is less clearly defined. It has been suggested that activation of A^sub 2A^ receptors on non-neural cells can induce protective anti-inflammatory and vasodilation effects, but with respect to neuronal signaling mechanisms, it is A^sub 2A^ antagonists that are found to be neuroprotective in models of brain ischemia26,48.

Phase 3 regulation

The high rate of adenosine formation observed in phase 3 may involve mechanisms different from those in phases 1 and 2. With the high levels of tissue depolarization, as suggested by the low adenylate energy charge values in phase 3, glutamate release may be occurring which could be further amplified by depolarization-induced inhibition and reversal of glutamate uptake3. Under these conditions, widespread activation of NMDA receptors could facilitate a cascade of excitotoxic events including intracellular Ca^sup 2+^ dysregulation, free radical production and inhibition of mitochondrial function which would inhibit energy production resulting in rapid adenosine formation49-52. Clearly, the highest percentage of tissue samples in the phase 3 region came from the strongly depolarizing group 4 treatments.

The activation of low affinity adenosine A^sub 3^ receptors is thought to occur under conditions of extreme metabolic stress where adenosine levels are extremely high resulting in the mobilization of tissue repair mechanisms including apoptosis and astrocyte proliferation18,27. Drastic insults to the brain, such as global ischemia, which reduces brain adenylate energy charge levels to the 0.20-0.30 range or the ischemia associated with decapitation, which increases striatal adenosine levels to approximately 2500 pmol/mg protein, outline the physiological extremes for adenosine and adenylate energy charge levels30,53,54. It is likely, therefore, that phase 3 represents a range of tissue energy levels over which significant adenosine A^sub 3^ receptor activation occurs.

Contralateral effects

We demonstrated that both adenine nucleotide and adenosine level changes were largely limited to the region immediately surrounding the stimulus site. Nevertheless, we observed occasional adenosine level increases in contralateral striata following unilateral drug injection and although inconsistent, appeared mainly when strong depolarizing stimuli were administered, primarily 1000 nmol glutamate. In contrast, however, when glutamate receptor antagonists CNQX and NBQX were injected, we not only observed consistently low adenosine levels in ipsilateral striata, but contralateral adenosine levels were also significantly lowered. These findings suggested that intense depolarization of the ipsilateral striatum can alter neuronal activity in the contralateral striatum, a possibility considering that both striata are indirectly connected through bilaterally projecting neurons . Contralateral responses following stimulation of ipsilateral striatum have been reported elsewhere, as well as in other brain regions receiving afferent input from areas directly affected by excitotoxins56-59. These data suggest that using contralateral regions as internal controls can lead to underestimation of effects.

Our model is sensitive enough to detect changes in levels of purines across a wide range of stimulus intensities. With this method hypotheses about the role of purines in sleep regulation, and cell life and death are being tested.

REFERENCES

1 Benington JH, Heller HC. Restoration of brain energy metabolism as the function of sleep. Prog Neumbiol 1995; 45: 347-360

2 Basheer R, Strecker RE, Thakkar MM, et al. Adenosine and sleepwake regulation. Prog Neumbiol 2004; 73: 379-396

3 Danbolt NC. Glutamate uptake. Prog Neumbiol 2001; 1: 1-105

4 Delaney SM, Geiger JD. Enhancement of NMDA-induced increases in levels of endogenous adenosine by adenosine deaminase and adenosine transport inhibition in rat striatum. Brain Res 1995; 1-2: 72-76

5 Delaney SM, Geiger JD. Levels of endogenous adenosine in rat striatum. II. Regulation of basal and N-methyl-D-aspartate-induced levels by inhibitors of adenosine transport and metabolism. J Pharmacol Exp Ther 1998; 2: 568-572

6 Delaney SM, Shepel PN, Geiger JD. Levels of endogenous adenosine in rat striatum. I. Regulation by ionotropic glutamate receptors, nitric oxide and free radicals. J Pharmacol Exp Ther 1998; 2: 561-567

7 Hoehn K, White TD. Role of excitatory amino acid receptors in K+ - and glutamate-evoked release of endogenous adenosine from rat cortical slices. I Neurochem 1990; 1: 256-265

8 Pedata F, Pazzagli M, Pepeu G. Endogenous adenosine release from hippocampal slices: excitatory amino acid agonists stimulate release, antagonists reduce the electrically- evoked release. Naunyn-Schmied Arch Pharmacol 1991; 5: 538-543.

9 Craig CG, White TD. N-methyl-D-aspartate- and non-N-methyl-D-aspartate-evoked adenosine release from rat cortical slices: distinct purinergic sources and mechanisms of release. J Neurochem 1 993; 3: 1073-1080

10 Schousboe A, Frandsen A, Drejer J. Evidence for evoked release of adenosine and glutamate from cultured cerebellar granule cells. Neurochem Res 1989; 9: 871-875

11 Li J, ladecola C. Nitric oxide and adenosine mediate vasodilation during functional activation in cerebellar cortex. Neumpharmacol 1994; 11: 1453-1461

12 Pelligrino DA, Wang Q, Koenig HM, et al. Role of nitric oxide, adenosine, N-methyl-D-aspartate receptors, and neuronal activation in hypoxia-induced pial arteriolar dilation in rats. Brain Res 1995; 1: 61-70

13 Evans MC, Swan JH, Meldrum BS. An adenosine analogue, 2-chloroadenosine, protects against long term development of ischaemic cell loss in the rat hippocampus. Neurosci Lett 1987; 3: 287-292

14 Yoon KW, Rothman SM. Adenosine inhibits excitatory but not inhibitory synaptic transmission in the hippocampus. J Neurosci 1991; 5: 1375-1380

15 Flagmeyer I, Haas HL, Stevens DR. Adenosine Al receptor-mediated depression of corticostriatal and thalamostriatal glutamatergic synaptic potentials in vitro. Brain Res 1997; 1: 178-185

16 Newby AC. Adenosine and the concept of 'retaliatory metabolites'. Trends Biochem Sci 1984; 9: 42-44

17 Dunwiddie TV, Masino SA. The role and regulation of adenosine in the central nervous system. Ann Rev Neurosci 2001; 24: 31-55

18 von Lubitz DK, Ye W, McClellan J, et al. Stimulation of adenosine A3 receptors in cerebral ischemia. Neuronal death, recovery, or both? Ann NY Acad Sci 1999; 890: 93-106

19 Corradetti R, Lo Conte C, Moroni F, et al. Adenosine decreases aspartate and glutamate release from rat hippocampal slices. Eur J Pharmacol 1984; 1-2: 19-26

20 Simpson RE, O'Regan MH, Perkins LM, et al. Excitatory transmitter amino acid release from the ischemic rat cerebral cortex: effects of adenosine receptor agonists and antagonists. J Neurochem 1992; 5: 1683-1690

21 Heron A, Lasbennes F, Seylaz J. Adenosine modulation of amino acid release in rat hippocampus during ischemia and veratridine depolarization. Brain Res 1993; 1: 27-32

22 Bennett GC, Boarder MR. The effect of nucleotides and adenosine on stimulus-evoked glutamate release from rat brain cortical slices. Br J Pharmacol 2000; 3: 617-623

23 Finn SF, Swartz KJ, Beal MF. 2-Chloroadenosine attenuates NMDA, kainate, and quisqualate toxicity. Neurose; Lett 1991; 2: 191-194

24 Roucher P, Meric P, Correze JL, et al. Metabolic effects of R-phenylisopropyladenosine during reversible forebrain ischemia studied by in vivo 31 P nuclear magnetic resonance spectroscopy. J Cereb Blood Flow Metab 1991; 3: 453-458

25 Perez-Pinzon MA, Mumford PL, Rosenthal M, et al. Anoxie preconditioning in hippocampal slices: role of adenosine. Neuroscience 1996; 3: 687-694

26 Pedata F, Corsi C, Melani A, et al. Adenosine extracellular brain concentrations and role of A2A receptors in ischemia. Ann NY Acad Sci 2001; 74-84

27 Abbracchio MP, Ceruti S, Brambilla R, et al. Modulation of apoptosis by adenosine in the central nervous system: a possible role for the A3 receptor. Pathophysiological significance and therapeutic implications for neurodegenerative disorders. Ann NY Acad Sci 1997; 825: 11-22

28 Ponten U, Ratcheson RA, Salford LC, et al. Optimal freezing conditions for cerebral metabolites in rats. J Neurochem 1973; 21: 1127-1138

29 Veech RL, Harris RL, Veloso D, et al. Freeze-blowing: A technique for the study of brain in vivo. J Neurochem 1973; 20: 183-188

30 Delaney SM, Geiger JD. Brain regional levels of adenosine and adenosine nucleotides in rats killed by high-energy focused microwave irradiation. J Neurosci Methods 1996; 2: 151-156

31 Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. New York: Academic Press, 1986

32 Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal Biochem 1976; 72: 248-254

33 Atkinson DE. The energy charge of the adenylate pool as a regulatory parameter. Interaction with feedback modifiers. Biochem 1968; 11: 4030-4034

34 Bridges RJ, Kavanaugh MP, Chamberlin AR. A pharmacological review of competitive inhibitors and substrates of high-affinity, sodium-dependent glutamate transport in the central nervous system. Curr Pharm Des 1999; 5: 363-379

35 Pazzagli M, Pedata F, Pepeu G. Effect of K+ depolarization, tetrodotoxin, and NMDA receptor inhibition on extracellular adenosine levels in rat striatum. Eur J Pharmacol 1993; 1: 61-65

36 Bettler B, Mulle C. Review: neurotransmitter receptors. II. AMPA and kainate receptors. Neuropharmacology 1995; 2: 123-139

37 Nowak L, Bregestovski P, Ascher P, et al. Magnesium gates glutamate-activated channels in mouse central neurones. Nature 1984; 5950: 462-465

38 Perkins MN, Stone TW. In vivo release of [3H]-purines by quinolinic acid and related compounds. Br J Pharmacol 1983; 2: 263-267

39 Hoehn K, White TD. N-methyl-D-aspartate, kainate and quisqualate release endogenous adenosine from rat cortical slices. Neuroscience 1990; 2: 441-450

40 Calabresi P, Centonze D, Gubellini P, et al. Synaptic transmission in the striatum: from plasticity to neurodegeneration. Prog Neurobiol 2000; 3: 231-265

41 White TD, Craig CG, Hoehn K. Extracellular adenosine, formed during low level NMDA receptor activation, provides an inhibitory threshold against further NMDA receptor-mediated neurotransmission in the cortex. Drug Dev Res 1993; 406-409

42 McBean GJ, Roberts PJ. Neurotoxicity of L-glutamate and DL-threo-3-hydroxyaspartate in the rat striatum. J Neurochem 1985; 1: 247-254

43 Zhang L, Rzigalinski BA, Ellis EF, et al. Reduction of voltage-dependent Mg^sup 2+^ blockade of NMDA current in mechanically injured neurons. Science 1996; 5294: 1921-1923

44 Ferkany JW, Zaczek R, Coyle JT. Kainic acid stimulates excitatory amino acid neurotransmitter release at presynaptic receptors. Nature 1982; 5876: 757-759

45 Young AM, Crowder JM, Bradford HF. Potentiation by kainate of excitatory amino acid release in striatum: complementary in vivo and in vitro experiments. J Neurochem 1988; 2: 337-345

46 Novelli A, Reilly JA, Lysko PG, et al. Glutamate becomes neurotoxic via the N-methyl-D-aspartate receptor when intracellular energy levels are reduced. Brain Res 1988; 1-2: 205-212

47 Zeevalk GD, Nicklas WJ. Evidence that the loss of the voltagedependent Mg^sup 2+^ block at the N-methyl-D-aspartate receptor underlies receptor activation during inhibition of neuronal metabolism. J Neurochem 1992; 4: 1211-1220

48 Sitkovsky MV. Use of the A(2A) adenosine receptor as a physiological immunosuppressor and to engineer inflammation in vivo. Biochem Pharmacol 2003; 4: 493-501

49 Mody I, MacDonald JF. NMDA receptor-dependent excitotoxicity: the role of intracellular Ca^sup 2+^ release. Trends Pharmacol Sci 1995; 10: 356-359

50 Schulz JB, Henshaw DR, Siwek D, et al. Involvement of free radicals in excitotoxicity in vivo. J Neurochem 1995; 5: 2239-2247

51 Nakao N, Crasbon-Frodl EM, Widner H, et al. Antioxidant treatment protects striatal neurons against excitotoxic insults. Neuroscience 1996; 1: 185-200

52 Nicholls DG, Budd SL, Castilho RF, et al. Glutamate excitotoxicity and neuronal energy metabolism. Ann NY Acad Sci 1999; 1-12

53 Onodera H, Iijima K, Kogure K. Mononucleotide metabolism in the rat brain after transient ischemia. J Neurochem 1986; 6: 1704-1710

54 Verhaegen M, Iaizzo PA, Todd MM. A comparison of the effects of hypothermia, pentobarbital, and isoflurane on cerebral energy stores at the time of ischemic depolarization. Anesthesiology 1995; 5: 1209-1215

55 Wilson CJ. Morphology and synaptic connections of crossed corticostriatal neurons in the rat. J Comp Neurol 1987; 4: 567-580

56 Wilson CJ. Postsynaptic potentials evoked in spiny neostriatal projection neurons by stimulation of ipsilateral and contralateral neocortex. Brain Res 1986; 1-2: 201-213

57 Ruzicka BB, Jhamandas K. Elevation of Met-enkephalin-like immunoreactivity in the rat striatum and globus pallidus following the focal injection of excitotoxins. Brain Res 1990; 1-2: 227-239

58 Lerner-Natoli M, Rondouin G, Belaidi M, et al. N-[1-(2-thienyl)-cyclohexyl]-piperidine (TCP) does not block kainic acid- induced status epilepticus but reduces secondary hippocampal damage. Neurosci Lett 1991; 2: 174-178

59 van Lookeren Campagne M, Lucassen PJ, Vermeulen JP, et al. NMDA and kainate induce internucleosomal DNA cleavage associated with both apoptotic and necrotic cell death in the neonatal rat brain, Eur J Neurosci 1995; 7: 1627-1640

P. Nickolas Shepel*,[dagger], David Ramonet[double dagger], Patrick Stevens[double dagger] and Jonathan D. Geiger[dagger],[double dagger]/p>

[dagger] Department of Pharmacology and Therapeutics, University of Manitoba Faculty of Medicine, Winnipeg, Manitoba, R3E 0W3, Canada

[double dagger] Department of Pharmacology, Physiology and Therapeutics, University of North Dakota School of Medicine and Health Sciences, Grand Forks, North Dakota 58203, USA

Correspondence and reprint requests to: Dr Jonathan D. Geiger, Department of Pharmacology, Physiology and Therapeutics, University of North Dakota, School of Medicine and Health Sciences, 501 North Columbia Road, Grand Forks, North Dakota 58203, USA. [jgeiger@medicine.nodak.edu] Accepted for publication December 2004.

* Deceased 17 February 2004.

ACKNOWLEDGMENTS

Dr P. Nickolas Shepel (Nick) was instrumental to the success of our work described in part here. His excellent scientific work and joie de vive will be missed greatly. Our work was supported by operating grants received from the National Institute on Aging (AG17628), the National Centre for Research Resources (P20RR017699) and the Canadian Institutes of Health Research (HOP8901, MOP53329). P.N.S. was supported by a Doctoral research award from the Canadian Institutes of Health Research.

Copyright Maney Publishing Mar 2005
Provided by ProQuest Information and Learning Company. All rights Reserved