Studies of Adenosine and GABAA Receptor Functions in Rat Hippocampal Slices

Recent evidence has indicated that adenosine, in addition to potassium and calcium currents, may also affect chloride movement in hippocampal neurones. This project was undertaken to determine the possible role of adenosine on chloride channels and synaptic plasticity in comparison with a selective GABA[A] agonist, muscimol. Extracellular recordings were made from the CA1 pyramidal cell layer of hippocampal slices in response to stimulation of Schaffer collateral fibres in stratum radiatum (0.01 Hz). Adenosine and muscimol induced concentration dependent reductions in the amplitude of orthodromically induced population potentials. In order to eliminate effects of these agents on potassium channels, experiments were performed in the presence of barium, 1mM (in some experiments in the presence of 1 mM tolbutamide). This concentration increased potential size, and reduced the inhibitory effect of adenosine on population spike size, but not synaptic potential size. This profile is consistent with the blockade of potassium channels associated only with the postsynaptic effects of adenosine. However, muscimol responses were unaffected. Adenosine potentiated the ability of muscimol to inhibit evoked potentials in the absence or presence of barium. The potentiation was prevented by the Al selective antagonist 8-cyclopentyltheophylline. The effects of adenosine, as well as muscimol, were reduced by the chloride channel blocker DIDS, which also prevented the adenosine potentiation of muscimol. The results indicate the ability of adenosine to operate chloride channels in hippocampal neurones, and suggest a potentiative interaction between adenosine and muscimol which also involve chloride channels. The second part of this study was to examine neurosteroids which have been reported to be positive modulators of the GABAA receptors. Alphaxalone and 5alpha-pregnan-3alpha-o1-20-one potentiated the inhibitory effect of muscimol on the population spike size at low concentrations (0.5 and 1muM) that had no significant effect on the spike size by themselves. This profile is in agreement with other reports which have described the effect of these neurosteroids as barbiturate-like. Alphaxalone and 5alpha-pregnan-3alpha-o1-20-one also at low concentrations potentiated the inhibitory effect of adenosine alone and in the presence of barium 1 mM which blocked adenosine activated potassium channels. Alphaxalone failed to potentiate the inhibitory effect of adenosine in the presence of bicuculline 1muM. It is concluded that these neurosteroids enhanced the potentiative interaction between adenosine and muscimol in the presence of barium. The results indicate that adenosine's effects are normally enhanced by virtue of the potentiative interaction occurring with endogenous GABA. In addition to this, the results show the chloride channels activated by adenosine to be different from those operated by the GABA[A] receptor. The third part of this project was to investigate the role of GABA[A] and adenosine receptors on long-term depression (LTD) and synaptic plasticity. Unlike long-term potentiation, LTD in the central nervous system remains poorly understood. Muscimol induced a time and concentration-dependent LTD in the amplitude of orthodromic potentials. Increasing the stimulation frequency from 0.01 Hz to 1 Hz for 10 seconds reversed the LTD induced by muscimol. Although adenosine decreased the spike size in a concentration-dependent manner, it failed to induce LTD. Muscimol also induced LTD in the absence of electrical stimulation. Alphaxalone and 5alpha-pregnan-3alpha-o1-20-one at concentrations that did not have any effect themselves on the population spike (0.5 and 1 muM), potentiated the inhibitory effect of muscimol on the population spike size. These neurosteroids at high concentrations (5 and 10 muM) decreased the spike size by themselves. On the other hand, at the low concentrations both steroids were able to potentiate the ability of muscimol to induce LTD. Moreover, muscimol 1 muM which is not able to induce LTD, alphaxalone and 5alpha-pregnan-3alpha-o1-20-one 1 muM maintained the LTD induced by muscimol 10 muM. Bicuculline 5 muM reversed the LTD induced by muscimol 10 muM. To examine the possible role of glutamate receptors in LTD induced by muscimol a number of NMDA and metabotropic receptors agonists and antagonists were used. The NMDA receptor antagonist 2-AP5, the NMDA/metabotropic antagonist 2-AP3 and selective metabotropic antagonist L(+)-AP3 failed to modify the LTD. Quisqualic acid and (1S,3R)- aminocyclopentane dicarboxylic acid (ACPD), a selective agonist at metabotropic receptors, did not induce LTD or short-term depression, whereas kynurenic acid prevented the reversal of the LTD obtained at 1 Hz. It is concluded that LTD can be induced by the selective activation of GABA[A] receptors and the lack of involvement of glutamate receptors in the protocol which is presented in this study confirms the unique role of classical GABA[A] receptors in the effect of muscimol and may indicate a novel type of long-lasting depression. Furthermore, the failure of adenosine to induce LTD taken together with the earlier results, suggests that the adenosine activated chloride channel differs from the GABA[A] receptor chloride channel