Modulation of the brain aversive system by GABAergic and serotonergic mechanisms

Experiments performed in our laboratory, using electrical stimulation combined with microinjection of drugs in the dorsal midbrain central grey (CG) of the rat, evidenced that direct stimulation of GABA receptors with locally administered gamma-aminobutyric acid (GABA) or the GABAA receptor agonists 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol, isoguvacine and muscimol raised the aversive threshold, defined as the lowest electrical current intensity inducing flight or escape behaviour when applied to the dorsal CG. The GABAB receptor agonist baclofen was ineffective. Also, enhancement of endogenous GABA action through local injection of the benzodiazepines chlordiazepoxide and midazolam or of pentobarbital resulted in anti-aversive effects. Ro 15-1788 antagonized both chlordiazepoxide and midazolam, suggesting benzodiazepine receptor mediation. In contrast to pro-GABAergic drugs, microinjection of the GABA antagonists bicuculline and picrotoxin into the CG elicited flight behaviour, like the electrical stimulation. Similar experiments with drugs influencing serotonergic neurotransmission evidenced that intra-CG microinjection of serotonin (5-HT) or of the direct 5-HT receptor agonist 5-methoxy-N,N-dimethyltryptamine increased the aversive threshold. The anti-aversive effect of 5-HT was potentiated by the selective inhibitor of 5-HT neuronal uptake, zimelidine. Also, the latter drug increased the aversive threshold when given alone. The anti-aversive effect of 5-HT was antagonized by local pretreatment with either metergoline or ketanserin, the latter being a selective blocker of 5-HT2 receptors. In contrast to the GABA antagonists mentioned above, the 5-HT receptor blockers did not evoke aversive behaviour per se. Therefore, both GABAergic and serotonergic mechanisms are likely to play an inhibitory role in the dorsal CG integrating aversive behaviour. The former seem to act tonically, whereas 5-HT would act in a phasic way. The implications of these results for the pathophysiology and drug treatment of chronic anxiety, panic states and pain disorders are briefly discussed.