Protection against kainate-induced excitotoxicity by adenosine A2A receptor agonists and antagonists

Protection against kainate-induced excitotoxicity by adenosine A2A receptor agonists and antagonists

Pergamon PII: Neuroscience Vol. 85, No. 1, pp. 229–237, 1998 Copyright  1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All ...

244KB Sizes 0 Downloads 60 Views

Pergamon

PII:

Neuroscience Vol. 85, No. 1, pp. 229–237, 1998 Copyright  1998 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(97)00613-1

PROTECTION AGAINST KAINATE-INDUCED EXCITOTOXICITY BY ADENOSINE A2A RECEPTOR AGONISTS AND ANTAGONISTS P. A. JONES,* R. A. SMITH† and T. W. STONE*‡ *Division of Neuroscience and Biomedical Systems, Institute of Biomedical and Life Sciences, West Medical Bldg, University of Glasgow, Glasgow G12 8QQ, U.K. †Laboratory of Human Anatomy, University of Glasgow, Glasgow G12 8QQ, U.K. Abstract––The neuroprotective role of adenosine receptor agonists in various models of ischaemia and neuronal excitotoxicity has been attributed to adenosine A1 receptor activation. In this study we examine the role of the A2A receptor in the kainate model of excitotoxicity. Kainate (10 mg/kg) was administered systemically 10 min after the intraperitoneal injection of adenosine analogues. The A2A agonist 2-p-(2carboxyethyl)phenethylamino-5 -N-ethylcarboxamidoadenosine hydrochloride (CGS21680) protected the hippocampus at concentrations of 0.1 and 0.01 mg/kg, but not at 2 µg/kg. The addition of the centrally acting adenosine A1 receptor antagonist 8-cyclopentyl-1,3-dipropylxanthine partially reduced protection only in the CA3a region, suggesting that only a small proportion of the protection was attributable to the A1 receptor. A less potent A2A agonist, N6-[2-(3,5-dimethyoxyphenyl)-2-(2-methylphenyl)-ethyl]adenosine (1 mg/kg), provided only partial protection against kainate. 4-(2-[7-Amino-2-[2-furyl][1,2,4]triazolo[2,3a][1,3,5]triazin-5-yl-amino]ethyl)phenol, a selective A2A antagonist, also showed protection against kainate-induced neuronal death, when administered alone or in combination with CGS21680. These results show that adenosine A2A receptor activation is protective against excitotoxicity. The protection is largely independent of A1 receptor activation or blockade.  1998 IBRO. Published by Elsevier Science Ltd. Key words: kainate, excitotoxicity, purines, CGS21680, ZM241385, adenosine.

Cerebral ischaemia results in the release of large amounts of excitatory amino acid transmitters, glutamate and aspartate, into the extracellular space. This release is mirrored by a dramatic increase in the extracellular concentration of adenosine.16,17 Adenosine inhibits both the presynaptic release of many neurotransmitters, including glutamate,6,20,41 and hyperpolarizes postsynaptic neurons,15 promoting the view that adenosine is an endogenous protective agent against cerebral ischaemia and excitotoxic neuronal damage. While much work has concentrated on the ischaemic protection displayed by selective adenosine A1 receptor agonists,13,19,30,33,51,52 and the converse potentiation of neuronal damage with A1 antagonists,40,51 a lack of suitable A2 agonists and antagonists has restricted studies investigating the effects of A2 receptors within ischaemic models. Binding studies classified the G-protein-coupled adenosine A2 receptors into two distinct subtypes, ‡To whom correspondence should be addressed. Abbreviations: APEC, 2-[(2-aminoethylamino)-carbonylethylphenylethyl amino]-5 -N-ethylcarboxamidoadenosine; CGS21680, 2-p-(2-carboxyethyl)phenethylamino-5 -Nethylcarboxamidoadenosine hydrochloride; CPX, 8-cyclopentyl-1,3-dipropylxanthine; DPMA, N6-[2-(3,5dimethoxyphenyl)-2-(2-methylphenyl)-ethyl]adenosine; 8-PST, 8-(p-sulphophenyltheophylline); ZM241385, 4-(2[7-amino-2-[2-furyl][1,2,4]triazolo[2,3-a] [1,3,5]triazin-5yl-amino]ethyl)phenol.

A2A and A2B.3 While the low-affinity A2B subtype is distributed throughout the brain, the high-affinity A2A receptors are more localized. The majority of A2A receptors are within the neostriatum and, to a lesser extent, the globus pallidus.22 More recent developments have characterized A2A receptors within the hippocampus and cerebral cortex,12 although these receptors appear to differ pharmacologically from the typical striatal population.11 A2A receptor stimulation increases the release of the excitotoxic amino acids aspartate and glutamate,38 to the detriment of ischaemic neurons. A2A antagonists, which inhibit the further release of glutamate, have repeatedly been shown to increase neuronal survival within ischaemic models.40,49 Activation of the adenosine A2A receptor also increases cerebral blood flow47,48 and inhibits platelet aggregation.14 Both effects increase blood and nutrient supply to any ischaemically compromised area of the brain. 2-p-(2-Carboxyethyl)phenethylamino5 -N-ethylcarboxamidoadenosine hydrochloride (CGS21680), an A2A selective agonist, depresses cerebral glucose utilization in a large number of brain regions, including the hippocampus,35 an effect which may be advantageous to neurons by decreasing the requirement for depleted nutrients during ischaemia. 2-[(2-Aminoethylamino)-carbonylethylphenylethylamino]-5 -N-ethylcarboxamidoadenosine

229

230

P. A. Jones et al.

(APEC), a centrally acting A2A agonist, increases cerebral blood flow, and with chronic exposure, increases neuron preservation in the hippocampus during forebrain ischaemia.49 More recently, CGS21680 has been shown to protect neurons against forebrain ischaemia in gerbils.44 A2 receptors decrease superoxide anion production in neutrophils,9 reducing the possibility of free radicalassociated neuronal death, which has been implicated in ischaemic damage.5,26,31,37 A2A receptor stimulation, therefore, provides a mixture of possible beneficial and detrimental effects, presenting a more confusing picture than that of the A1 receptor. The present study reports the effects in rats of the A2A agonists CGS21680 and N6-[2-(3,5-dimethoxyphenyl)2-(2-methylphenyl)-ethyl]adenosine (DPMA) on excitotoxic neuronal damage induced by kainate, and the effects of antagonists selective for A1 and A2 receptors. EXPERIMENTAL PROCEDURES

Male Wistar rats (Harlan Olac) of weight 240–310 g, housed under standard conditions, were used in all experiments. All injections were made by the intraperitoneal route in a volume no greater than 3 ml/kg. Kainate and CGS21680 were dissolved in saline, DPMA in methanol, 8-(p-sulphophenyltheophylline) (8-PST) in distilled water, 8-cyclopentyl-1,3-dipropylxanthine (CPX) in ethanol and 4-(2-[7-amino-2-[2-furyl][1,2,4]triazolo[2,3-a][1,3,5]triazin-5yl-amino]ethyl)phenol (ZM241385) in polyethylene glycol 300/0.1 M NaOH (1:1, v/v) to provide solutions of the correct concentration for injection. The drugs were all injected 10 min prior to the injection of kainate, which was administered at a standard dose of 10 mg/kg. The animals were left for seven days under standard conditions. Tissue fixing and slicing Rats were anaesthetized with sodium pentobarbitone (60 mg/kg) after seven days. The method of perfusion first described by Brown and Brierley2 was used, in which 50 ml of physiological medium with 12.5% xylocaine was perfused through a cannula placed into the left ventricle before fixation by 250 ml of 40% formaldehyde/glacial acetic acid/ methanol (1:1:8, v/v). The animals were decapitated and the heads stored at 4C in fixative for up to one week. The brain was subsequently removed and the left hemisphere marked with Indian ink. The cerebral hemispheres were cut into 2-mm-thick slices, dehydrated and infused with paraffin wax throughout before embedding in wax. Sections of 6 mm were cut, mounted on slides and stained with haematoxylin and eosin. Sections were examined under a light microscope by observers who were unaware of the drug treatment received. The left (marked) hippocampus was examined and areas CA1, CA2, CA3a, CA3b and CA4 scored for damage. The scale used was a percentage scale, where 0% equated to no damage, while 100% indicated a total loss of neurons. The

damage was estimated by a cell count of viable (bluestained) cells versus pink-stained compromised neurons. Account was also taken of decreased neuronal number and general appearance of the hippocampal subareas. ANOVA followed by post hoc tests were used to determine any statistical significance. Significance refers to results where P<0.05 was obtained. RESULTS

Kainate (10 mg/kg) caused damage in the CA1, CA2 and CA3a areas of the hippocampus (Figs 1, 2), with no evidence of neuron mortality in the CA3b and CA4 regions. The damage associated with the excitotoxin was similar to other reports, with the largest extent of neuronal death observed in the CA1 (48.47.4% damage) and CA3a (38.17.0%) regions, while the CA2 region suffered more moderate (22.26.4%) damage. Saline controls induced no significant hippocampal neuronal death in any region (Fig. 1). CGS21680 was administered at three doses (0.1 mg/kg, 0.01 mg/kg and 2 µg/kg) 10 min prior to the injection of 10 mg/kg kainate. In the CA1 area, both 0.1 and 0.01 mg/kg induced near maximal protection against neuronal cell death (6.94.4% and 7.04.4% damage, respectively), while the lower concentration of 2 µg/kg displayed an intermediate effect (27.09.8% damage), with limited protection against kainate damage, which was significantly greater than in saline controls (Fig. 2). CGS21680 at 0.1 mg/kg protected the CA3a neurons almost entirely against kainate excitotoxicity (5.02.4% damage; Figs 1, 2), while the 0.01 mg/kg dose protected to a notably lesser degree (16.57.7%). The low concentration of CGS21680 showed no protection against kainate. All three doses of CGS21680 displayed significant, and similar, preservation of the CA2 area of the hippocampus (Fig. 2). In contrast, DPMA, a less potent A2A agonist, conferred neuroprotection in the CA1 and CA2 regions only at the higher i.p. concentration of 1.0 mg/kg, remaining largely ineffective at 0.1 mg/kg. No significant protection was observed with either concentration in the CA3a region (Fig. 3). Co-administration of the centrally effective A1 receptor antagonist CPX at a dose of 50 µg/kg with 0.1 mg/kg CGS21680 did not inhibit the protection against kainate damage in either the CA1 or CA2 regions (Fig. 4), although this protection was nonsignificant in the CA2 region. In the CA3a area, CPX increased cell mortality moderately, such that damage was neither significantly different from that obtained with kainate alone nor with kainate plus

Fig. 1. Haemotoxylin and eosin-stained hippocampal sections showing damage. CA1 (A) and CA3a (B) regions of the saline control contrast markedly with hippocampi of animals systemically injected with 10 mg/kg kainate. After kainate administration, only a few normal cells are seen with rounded cell bodies and clear nuclei and nucleoli. Damaged cells are shrunken and distorted, with small dense nuclear remnants. Arrows indicate examples of damaged cells in both the CA1 (C) and CA3a (D) regions of the hippocampus. Administration of 0.1 mg/kg CGS21680 prior to kainate prevented the excitotoxic effects of the glutamate analogue in both CA1 (E) and CA3a (F) areas. Scale bar=50 µm.

A2A protection in excitotoxicity

231

232

P. A. Jones et al.

Fig. 2. Protection by the A2A receptor agonist CGS21680 against kainate-induced excitotoxic hippocampal cell death. Protection was statistically significant at 0.1 and 0.01 mg/kg in the CA1, CA2 and CA3a regions, and at 2 µg/kg in the CA1 and CA3a regions. No damage was produced by kainate in the CA3b or CA4 areas. ***P<0.001, **P<0.01, *P<0.05 versus kainate alone; +++P<0.001, ++P<0.01, + P<0.05 versus saline control.

Fig. 4. Effects of the addition of the centrally acting A1 adenosine receptor antagonist CPX and the peripherally acting adenosine antagonist 8-PST on the protection observed with CGS21680 against kainate excitotoxicity in the CA1, CA2 and CA3a regions. ***P<0.001, **P<0.01, *P<0.05 versus kainate alone.

tion was prevented in the presence of 50 µg/kg CPX (Figs 5, 6). When ZM241385 was combined with 0.1 mg/kg CGS21680, total neuronal preservation was always observed (Fig. 5). DISCUSSION

Fig. 3. Protection observed with the A2A adenosine receptor agonist DPMA in the CA1 and CA2 regions of the hippocampus against kainate-induced neuronal damage at a dose of 1.0 mg/kg, but not at 0.1 mg/kg. **P<0.01, *P<0.05 versus kainate alone.

CGS21680. The polar and non-selective adenosine antagonist 8-PST does not easily cross the blood– brain barrier. When injected at a concentration of 20 mg/kg in conjunction with CGS21680, 8-PST showed inconsistent effects, but prevented any significant degree of protection in the CA1 and CA3a regions (Fig. 4). The specific A2A antagonist ZM241385 was given i.p. at a dose of 10 mg/kg. When given alone, ZM241385 protected against damage induced by 10 mg/kg kainate, to a level similar to CGS21680 alone in the CA3a region (Figs 5, 6). In the CA2 area, ZM241385 reduced neuron damage to 0.71%. Protec-

Kainate has been used extensively in neurobiological research as it preferentially damages neurons in the limbic system, particularly the hippocampus.28 It is considered a good experimental model for temporal lobe epilepsy and global ischaemia.18,46 Adenosine or its analogues have been shown previously to mitigate against kainate-induced excitotoxicity and ischaemic damage, via interaction with the A1 receptor,13,19,30,33,40,51,52 while little attention has been given to the A2 receptor. A2A receptors labelled by autoradiography are located in the neostriatum, globus pallidus, cerebral cortex and hippocampus of the rat brain.12,22 Within the hippocampus, functional A2A receptors are localized to the CA1 and CA3 regions.12 As shown in the present study and several previous reports, kainate caused damage in these areas, as well as the intervening CA2 region, but had no effect in either the CA3b or the CA4 region. The present work indicates that both the adenosine A2A agonists tested, CGS21680 and DMPA, were able to reduce the kainate-induced damage. DPMA (13-fold selective for the rat brain A2 receptor versus A1,34 Ki=4.4 nM)1 significantly decreased neuron degeneration in the CA1 and CA2 areas of the hippocampus at a dose of 1.0 mg/kg. CGS21680 is 140-fold selective for the A2A versus the A1 receptor, with a slightly lower affinity than DPMA for the A2A subtype of the receptor

A2A protection in excitotoxicity

233

Fig. 5. Photomicrographs of the CA1 (A, C) and CA3a (B, D) regions of the rat hippocampus, showing the undamaged regions in animals treated with 10 mg/kg kainate together with 10 mg/kg ZM241385 (A, B). The protective effects of ZM241385 are prevented by the co-administration of 50 µg/kg CPX (C, D). Arrows indicate examples of damaged cells. Scale bar=50 µm.

(Ki=15 nM),21 and has no discernible activity at the low-affinity A2B receptor.22,29 CGS21680 protected the hippocampus at the two higher doses examined, with the lower 2 µg/kg concentration showing only moderate protection in the CA1 and CA2 regions. While similar results were obtained by Sheardown and Knutsen,44 the group used far higher doses of 10 mg/kg CGS21680 (1000 times the concentration required in this study). Since A1 receptor antagonists were not co-administered, the results observed at these higher doses may have been entirely due to A1 receptor activation. There were also variations in the dosing regimen, which may contribute to the differences between their study and the present work.

The limited protection observed with another A2A agonist (APEC)49 may be due more to an affinity for adenosine A3 receptors,25 as both APEC administration and A3 receptor stimulation result in a very similar pattern of protection against ischaemia (in both cases chronic, but note that acute treatment increases neuronal survival).49,50 The selective A1 receptor antagonist CPX (50 µg/ kg), in combination with CGS21680 (0.1 mg/kg), showed no significant blockade of the protection afforded by CGS21680 alone. This suggests that A1 receptor activation by CGS21680 (responsible for the anticonvulsant effects of CGS21680)53 may account for only a small proportion of the protection afforded by this compound, at the doses used here.

234

P. A. Jones et al.

Fig. 6. Significant protection in the kainate model of excitotoxicity with the A2A receptor antagonist ZM241385 (10 mg/kg) alone or in the presence of 0.1 mg/kg CGS21680, which is reversed with the addition of 50 µg/kg of the A1 receptor antagonist CPX. **P<0.01, *P<0.05 versus kainate alone.

The results achieved with the peripherally acting non-selective adenosine antagonist, 8-PST, in combination with CGS21680, were inconsistent and varied from 0% to 60% damage between individual experiments. 8-PST alone has been shown previously to increase the damage associated with kainate,32 possibly by blocking the peripheral effects of endogenously released adenosine. The fluctuation in damage seen in this study may therefore represent a combination of the specific beneficial effects of CGS21680 and the more general detrimental effects observed previously with 8-PST. The experiments with 8-PST, therefore, may imply a peripheral component to the mechanism of action of CGS21680. As A1 receptor activation by the A2A agonists accounts for only a small proportion of the protection, other A2A receptor-mediated mechanisms may be involved. For example, A2A receptor activation has been shown to result in vasodilatation of cerebral blood vessels,48 increasing the supply of nutrients to potentially damaged areas. Increasing blood flow may also increase the rate at which kainate is washed out of the brain. This is less likely to be an important mechanism by which A2A receptor stimulation mediates protection when considering the Sheardown and Knutsen study,44 in which CGS21680 protection was observed within an ischaemia model when given 30 min postoperatively. A2A receptor activation also suppresses leukocyte adhesion to the vascular endothelium and the secretion of injurious reactive oxygen species by neutrophils.8,9 During kainate-induced excitotoxicity, increased energy consumption leads to a decrease in cerebral levels of ATP.36,43 CGS21680 may therefore protect ATP-depleted neurons by reducing neuronal activity,

as CGS21680 has been shown to depress glucose utilization in a number of regions of the brain, including the hippocampus.35 It is noteworthy that, during reperfusion after ischaemic insults, a general depression of glucose metabolism is observed in the brain except for selectively vulnerable regions such as the CA1.24 The selective adenosine antagonist ZM241385, which is 80-fold selective for A2A versus A2B receptors, and 500- to 1000-fold selective for A2A versus A1 receptors,39 protected the hippocampus to a similar degree as CGS21680 in all areas sensitive to kainate damage. This is consistent with other reports of A2A antagonists, where protection has been seen against cerebral ischaemia.40,49 When given in conjunction with CGS21680, total neuronal conservation of all hippocampal neurons was seen. This may suggest that while both the A2A agonist and antagonist are neuroprotective on their own, the mechanism by which this is achieved is different and not mutually exclusive. Due to the low selectivity and potency of ZM241385 for A2B and A3 receptors,42 it is unlikely that the antagonist exerts its neuroprotective effects through these adenosine receptor subtypes; it is possible that ZM241385 may have some degree of agonist activity at A2A receptors in the hippocampus. While no such activity has been observed in the cerebral cortex,42 A2A receptors of the hippocampus differ from those elsewhere in the CNS (including different binding characteristics).11 The role, if any, of the CGS21680 binding sites, distinct from classical A1, A2A, A2B or A3 receptors in rat cortex,23 has yet to be determined. An alternative explanation of protection by ZM241385 may be necessary to account for the blockade of ZM241385 protection by CPX. Recent evidence suggests that A2A receptor activation inhibits A1 receptors,12 an action that could be damaging since A1 receptor stimulation is known to be neuroprotective.33,51 Kainate has been shown to induce the release of adenosine in the CNS.4 This endogenously released adenosine may be enough to activate A2A receptors and thereby inhibit a large proportion of A1 receptors. The addition of the A2A agonist CGS21680 could not then inhibit A1 receptors further, so that largely beneficial effects (such as those on blood vessels) would be seen. The subsequent addition of ZM241385 would block A2A receptors and release A1 receptors, which could be acted upon by endogenously released adenosine to produce protection. Evidence for this is given by the negation of protection by ZM241385 in the presence of CPX. As ZM241385 has no known agonist activity at A1 receptors, it seems likely that the protection is due to the negative association between A1 and A2A receptors. Further evidence for this speculative relationship between A2A and A1 receptors is provided by studies where both A1 and A2A receptors co-exist.7,10

A2A protection in excitotoxicity

Correia-de-Sa and Ribeiro7 first showed that, at the neuromuscular junction, the action of A2A receptorfacilitated neurotransmitter release predominated over A1-induced inhibition. Similarly, Cunha et al.10 observed that endogenously formed adenosine preferentially activated A2A receptors in the hippocampus, enhancing synaptic transmission. CPX has previously been shown to inhibit the cerebral protection exerted by R-phenyl-isopropyladenosine (R-PIA) in the kainate-treated brain.32 However, when given at the same concentration without the addition of R-PIA, there was no increase in damage,33 even though adenosine released endogenously by kainate4 should be sufficient to activate A1 receptors. At a concentration at which it has been speculated that CPX would inhibit both A2A and A1 receptors,33,45 there is an increase in toxicity. Similar results were obtained by Phillis,40 who showed that CPX-enhanced neurotoxicity only occurred at 1.0 mg/kg, but not at the lower dose of 0.1 mg/kg. Although this effect was explained as A1 receptor mediated, this concentration was greater than that shown previously to inhibit the depressant action of A2A receptors27 and much higher than the 50 µg/kg required to prevent R-PIA protection.33 Importantly, the above hypothesis questions the role of the A1 receptor in the neuroprotection afforded by endogenously released adenosine. If inhibition of the A1 receptor does not affect the speculated protection mediated by kainate-released adenosine, there may be no A1 receptor-mediated element in the endogenous situation. Alleviation of damage by A1 agonists may therefore occur only

235

after the exogenous administration of an A1 agonist overcomes this inhibition of the A1 receptor by endogenously activated A2A receptors. This does not discount the use of adenosine agonists and antagonists in the prevention of kainate-induced excitotoxic damage, but does challenge the belief that, in the normal physiology of kainate-induced insult, A1 receptors mediate endogenous protection.

CONCLUSION

These results show the benefits of A2A adenosine receptor agonists in an excitotoxic model. The results also provide evidence for a neuroprotective action of an A2A antagonist, which may be due to interaction between A1 and A2A receptors. While suggesting that the effects of CGS21680 are specific and do not substantially involve A1 receptors, this study suggests that the action is at least partly peripherally mediated. Further work to distinguish between centrally and peripherally mediated effects (e.g., by utilizing the intrahippocampal as opposed to intraperitoneal route of injection) will be required to determine the exact mechanism of action of CGS21680. This would also help to clarify the possibility that peripheral A2A receptor activation mediates protection, while that of the hippocampus (by virtue of inhibiting A1 receptors) increases damage. Acknowledgements—P.A.J. was funded by a University of Glasgow Postgraduate Scholarship. The authors would like to thank Zeneca for the gift of ZM241385.

REFERENCES

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

Bridges A. J., Bruns R. F., Ortwine D. F., Priebe S. R., Szotek D. L. and Trivedi B. K. (1988) N6-[2-(3,5Dimethoxyphenyl)-2-(2-methylphenyl)-ethyl]adenosine and its uronamide derivatives. Novel adenosine agonists and antagonists with both high affinity and high selectivity for the adenosine A2 receptor. J. med. Chem. 31, 1282–1285. Brown A. W. and Brierley J. B. (1972) Anoxic ischaemic cell change in the rat brain. Light microscope and fine structural observations. J. neurol. Sci. 16, 15–84. Bruns R. F., Lu G. H. and Pugsey T. A. (1986) Characterisation of the A2 adenosine receptor labelled by [3H]NECA in rat striatal membranes. Molec. Pharmac. 29, 331–346. Carswell H. V., Graham D. I. and Stone T. W. (1997) Kainate-evoked release of adenosine from the hippocampus of the anaesthetised rat: possible involvement of free radicals. J. Neurochem. 68, 240–247. Cheng Y. and Sun A. Y. (1994) Oxidative mechanisms involved in kainate-induced cytotoxicity in cortical neurones. Neurochem. Res. 19, 1557–1564. Corradetti R., Lo Conte G. and Moroni F. (1984) Adenosine decreases aspartate and glutamate release from rat hippocampal slices. Eur. J. Pharmac. 104, 19–26. Correia-de-Sa P. and Ribeiro J. A. (1994) Tonic adenosine A2A receptor activation modulates nicotinic autoreceptor function at the rat neuromuscular junction. Eur. J. Pharmac. 271, 349–355. Cronstein B. N., Levin R. I., Phillips M. R., Hirschhorn R., Abramson S. B. and Weissmann G. (1992) Neutrophil adherence to endothelium is enhanced via adenosine A1 receptors and inhibited via adenosine A2 receptors. J. Immun. 148, 2201–2206. Cronstein B. N., Rosenstein E. D., Kramer S. B., Weissmann G. and Hirschhorn R. (1985) Adenosine: a physiologic modulator of superoxide anion generation by human neutrophils. Adenosine acts via an A2 receptor on human neutrophils. J. Immun. 135, 1366–1371. Cunha R. A., Correia-de-Sa P., Sebastia˜o A. M. and Ribeiro A. (1996) Preferential activation of excitatory adenosine receptors at rat hippocampal and neuromuscular synapses by adenosine formed from released adenine nucleotides. Br. J. Pharmac. 119, 253–260. Cunha R. A., Johansson B., Constantino M. D., Sebastia˜o A. M. and Fredholm B. B. (1996) Evidence for high affinity binding sites for the adenosine A(2A) receptor agonist [3H]CGS21680 in the rat hippocampus and cerebral cortex that are different from striatal A(2A) receptors. Naunyn-Schmiedeberg’s Arch. Pharmac. 353, 261–271.

236 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

P. A. Jones et al. Cunha R. A., Johansson B., Van der Ploeg I., Sebastia˜o A. M., Ribeiro J. A. and Fredholm B. B. (1994) Evidence for functionally important adenosine A2a receptors in the rat hippocampus. Brain Res. 649, 208–216. Daval J.-L., Von Lubitz D. J. K. E., Deckert J., Redmond D. J. and Marangos P. J. (1989) Protective effect of cyclohexyladenosine A1-receptors, guanine nucleotide and forskolin binding sites following transient brain ischaemia: a quantitative autoradiographic study. Brain Res. 491, 212–226. Dionisotti S., Zocchi C., Varani K., Borea P. A. and Ognini E. (1992) Effects of adenosine derivatives on human and rabbit platelet aggregation. Correlation of adenosine receptor affinities and antiaggregatory activity. NaunynSchmiedeberg’s Arch. Pharmac. 346, 673–676. Gerber U., Greene R. W., Haas H. L. and Stevens D. R. (1989) Characterisation of inhibition mediated by adenosine in the hippocampus of the rat in vitro. J. Physiol. 417, 567–578. Globus M. Y. T., Busto R., Dietrich W. D., Martinez E., Valdes I. and Ginsberg M. D. (1988) Effect of ischaemia on the in vivo release of striatal dopamine, glutamate, and gamma-aminobutyric acid studied by intracerebral microdialysis. J. Neurochem. 51, 1455–1464. Hagberg H., Andersson P., Lacarewicz J., Jacobson I., Butcher S. and Sandberg M. (1987) Extracellular adenosine, inosine, hypoxanthine, and xanthine in relation to tissue nucleotides and purines in rat striatum during transient ischaemia. J. Neurochem. 49, 227–231. Heggli D. A., Aamodt A. and Malthe-Srenssen D. (1981) Kainic acid neurotoxicity, effects of systemic injection of neurotransmission markers in different brain regions. Brain Res. 230, 253–262. Heron A., Lekiettre D., Le Peillet E., Lasbennes F., Seylez J., Plotkine M. and Boulu R. G. (1994) Effects of an A1 adenosine receptor agonist on the neurochemical, behavioural and histological consequences of ischaemia. Brain Res. 641, 217–224. Heron A., Lasbennes F. and Seylez J. (1993) Adenosine modulation of amino acid release in rat hippocampus during ischaemia and veratridine depolarisation. Brain Res. 608, 27–32. Hutchison A. J., Webb R. L., Oei H. H., Ghai G. R., Zimmerman B. and Williams M. (1989) CGS21680, an A2 selective adenosine receptor agonist with preferential hypotensive activity. J. Pharmac. exp. Ther. 251, 47–55. Jarvis M. F. and Williams M. (1989) Direct autoradiographic localisation of adenosine A2 receptors in the brain using the A2-selective agonist, [3H]CGS21680. Eur. J. Pharmac. 168, 243–246. Johannson B. and Fredhom B. B. (1995) Further characterisation of the binding of the adenosine receptor agonist [3H]CGS21680 to rat brain using autoradiography. Neuropharmacology 34, 393–403. Jorgensen M. B. (1993) The role of signal transduction in the delayed necrosis of the hippocampal CA1 pyramidal cells following transient ischaemia. Acta neurol. scand., Suppl. 87, 1–20. Kim H. O., Ji X.-D., Sidiqqi S. M., Olah M. E., Stiles G. L. and Jacobson K. A. (1994) 2-Substitution of N6-benzyladenosine-5 -uronamides enhances selectivity for A3 adenosine receptors. J. med. Chem. 21, 3614–3621. Kitawaga K., Matsumoto M., Oda T., Ninoed M., Hata R., Handa N., Fukunaga R., Isaka Y., Kinmura K., Maeda H., Mikoshiba K. and Kamada T. (1990) Free radical generation during brief period of cerebral ischaemia may trigger delayed neuronal death. Neuroscience 35, 551–558. Lin Y. and Phillis J. W. (1991) Characterisation of the depression of rat cerebral cortical neurons by selective adenosine agonists. Brain Res. 540, 307–310. Lothman E. W. and Collins R. C. (1981) Kainic acid induced limbic seizures: metabolic, behavioural, electroencephalographic and neuropathological correlates. Brain Res. 218, 299–318. Lupica C. R., Cass W. A., Zahniser N. R. and Dunwiddie T. V. (1990) Effects of the selective adenosine agonist CGS21680 on in vitro electrophysiology, cAMP formation and dopamine release in rat hippocampus and striatum. J. Pharmac. exp. Ther. 252, 1134–1141. MacGregor D. G., Jones P. A., Maxwell W. L., Graham D. I. and Stone T. W. (1996) Prevention by a purine analogue of kainate-induced neuropathology in rat hippocampus. Brain Res. 725, 115–120. MacGregor D. G., Higgins M. J., Jones P. A., Maxwell W. L., Watson M. W., Graham D. I. and Stone T. W. (1996) Ascorbate attenuates the systemic kainate-induced neurotoxicity in the rat hippocampus. Brain Res. 727, 133–144. MacGregor D. G. and Stone T. W. (1994) Blockade by 1,3-dipropyl-8-cyclopentylxanthine (CPX) of purine protection against kainate neurotoxicity. Brain Res. 644, 339–342. MacGregor D. G. and Stone T. W. (1993) Inhibition by the adenosine analogue, (R)-N6-phenylisopropyladenosine, of kainic acid neurotoxicity in rat hippocampus after systemic administration. Br. J. Pharmac. 109, 316–321. Merkel L. A., Lappe R. W., Rivera L. M., Cox B. F. and Perrone M. H. (1992) Demonstration of vasorelaxant activity with an A1-selective adenosine agonist in porcine coronary artery: involvement of potassium channels. J. Pharmac. exp. Ther. 260, 437–443. Nehlig A., Daval J.-L. and Boyet S. (1994) Effects of selective adenosine A1 and A2 receptor agonists and antagonists on local rates of energy metabolism in the rat brain. Eur. J. Pharmac. 258, 57–66. Nicklas W. J., Krespan B. and Beri S. (1980) Effect of kainate on ATP levels and glutamate metabolism in cerebellar slices. Eur. J. Pharmac. 62, 209–213. O’Neill M., Canney M., Feeney C., Earley B. and Leonard B. E. (1994) Protective effects of exifone, a free radical scavenger, against ischaemia-induced hippocampal neurodegeneration in the gerbil. Med. Sci. Res. 22, 589–590. O’Regan M. H., Simpson R. E., Perkins L. M. and Phillis J. W. (1992) The selective A2 adenosine receptor agonist CGS21680 enhances excitatory amino acid release from the ischaemic rat cerebral cortex. Neurosci. Lett. 138, 169–172. Palmer T. M., Poucher S. M., Jacobson K. A. and Stiles G. L. (1995) 125I-4-(2-[7-Amino-2-triazolo-triazin-5-ylamino]ethyl)phenol, a high affinity antagonist radioligand selective for the A2a adenosine receptor. Molec. Pharmac. 48, 970–974. Phillis J. W. (1995) The effects of selective A1 and A2a adenosine receptor antagonists of cerebral ischaemic injury in the gerbil. Brain Res. 705, 79–84. Poli A., Lucchi R., Vibio M. and Barnabei O. (1991) Adenosine and glutamate modulate each other’s release from rat hippocampal synaptosomes. J. Neurochem. 57, 298–306. Poucher S. M., Keddie J. R., Singh P., Stoggall S. M., Caulkett P. W. R., Jones G. and Collis M. G. (1995) The in vitro pharmacology of ZM241385, a potent non-xanthine, A2a selective adenosine receptor antagonist. Br. J. Pharmac. 115, 1096–1102.

A2A protection in excitotoxicity 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53.

237

Retz K. C. and Coyle J. T. (1980) Kainic acid lesion of mouse striatum: effects on energy metabolites. Life Sci. 27, 2495–2500. Sheardown M. J. and Knutsen L. J. S. (1996) Unexpected neuroprotection observed with the adenosine A2a receptor agonist CGS21680. Drug Dev. Res. 39, 108–114. Simpson R. E., O’Regan M. H., Perkins L. M. and Phillis J. W. (1992) Excitatory transmitter amino acid release from the ischaemic rat cerebral cortex: effects of adenosine receptor agonists and antagonists. J. Neurochem. 58, 1683–1690. Sperk G. (1994) Kainic acid seizures in the rat. Prog. Neurobiol. 42, 1–32. Stella L., De Novellis V., Berrino L., D’Amico M. and Rossi F. (1996) Evidence that A(2a) and not A(2b) purinoceptors are coupled to production of nitric oxide in the regulation of blood pressure. Environ. Toxic. Pharmac. 2, 327–329. Torregrossa G., Salom J. B., Miranda F. J., Alabadi J. A., Alvarez C. and Alborch E. (1990) Adenosine A2 receptors mediate cerebral vasodilation in the conscious goat. Blood Vessels 27, 24–27. Von Lubitz D. K. J. E., Lin R. C.-S. and Jacobson K. A. (1995) Cerebral ischaemia in gerbils: effects of acute and chronic treatment with adenosine A2a receptor agonist and antagonist. Eur. J. Pharmac. 287, 295–302. Von Lubitz D. J. K. E., Lin R. C.-S. and Popik P. (1994) Adenosine A3 receptor stimulation and cerebral ischaemia. Eur. J. Pharmac. 263, 59–67. Von Lubitz D. J. K. E., Lin R. C.-S., Melman N., Ji X.-D., Carter M. F. and Jacobson K. A. (1994) Chronic administration of selective adenosine A1 receptor agonist or antagonist in cerebral ischaemia. Eur. J. Pharmac. 256, 161–167. Von Lubitz D. J. K. E., Dambrosia J. M., Kempski O. and Redmond D. J. (1988) Cyclohexyl adenosine protects against neuronal death following ischaemia in the CA1 region of gerbil hippocampus. Stroke 19, 1133–1139. Zhang G., Franklin P. H. and Murray T. F. (1994) Activation of adenosine A1 receptors underlies anticonvulsant effect of CGS21680. Eur. J. Pharmac. 255, 239–243. (Accepted 6 November 1997)