Multiple serotonin mechanisms in animal models of anxiety: environmental, emotional and cognitive factors.

Responses to serotonergic drugs in animal models of 'anxiety' are reviewed with emphasis on the elevated X-maze. Evidence for the 'classic' hypothesis, that decreasing serotonergic function is anxiolytic and increasing it anxiogenic, is most consistent in models of behavioural inhibition where the stimulus inhibits an approach response (conflict models). However, paradoxical drug effects are also frequent, especially where the aversive stimulus evokes an active response. Both types of drug effect are equally frequent in the elevated X-maze. 'Anxiety' models may detect multiple sites and mechanisms of action of the same drug; this may indicate multiple anxiety-related neurological mechanisms in the brain. However, not all drug effects in 'anxiety' models are necessarily related to anxiety itself. It is possible that cognitive factors may affect stimulus evaluation, and response inhibition by an aversive stimulus may be a special case of a wider role for serotonin in behavioural control. Clinical implications of these observations are considered.

Effects of beta-adrenoceptor ligands in the elevated X-maze 'anxiety' model and antagonism of the 'anxiogenic' response to 8-OH-DPAT.

Effects of ß-adrenoceptor agonists and antagonists have been examined on open/total arm entry ratio (OTR) over a wide range of doses in the rat elevated X-maze 'anxiety' model. For clenbuterol, terbutaline and adrenaline, OTR was reduced only at doses at or above those reducing total entries; dobutamine was inactive. Neither the ß(1)-adrenoceptor antagonist metoprolol nor the ß( 2)-adrenoceptor antagonist ICI 118,551 affected OTR or total entries. The peripherally acting ß(1)-antagonist practolol was also inactive. The elevated X-maze therefore does not appear to detect ß-adrenoceptor-mediated changes in 'anxiety'. Among the ß-adrenoceptor antagonists which also bind to 5-HT(1) receptors, sotalol and timolol were inactive but restricted doses of alprenolol (0.1 mg/kg) and pindolol (0.1-0.25 mg/kg) caused an anxiolytic-like increase in OTR while propranolol (5-10 mg/kg) and pindolol (1.0 mg/kg) reduced OTR without affecting total entries. These effects may be attributable to the activity of these agents at 5-HT( 1) receptors. Since metoprolol (3.0 mg/kg) and ICI 118,551 (1.0 mg/kg) did not alter the fall in OTR caused by the selective 5-HT(1A) agonist 8-OH-DPAT, the antagonism of this fall by alprenolol (0.5 mg/kg), pindolol (0.5 mg/kg), propranolol (3.0 mg/kg) and timolol (3.0 mg/kg) is most likely to represent a 5-HT(1) receptor antagonist effect of these agents.

Actions and some interactions of 5-HT1A ligands in the elevated X-maze and effects of dorsal raphe lesions.

Effects of 5-HT1A agonists and partial agonists on open/total arm entry ratio (OTR) have been examined in the elevated X-maze anxiety model. 8-OH-DPAT (0.05-0.2 mg/kg), RU 24969 (0.5-2.0 mg/kg) and BAY R 1521 (0.1-1.2 mg/kg) produced dose-dependent reductions in OTR, signifying anxiogenic effects. Buspirone reduced OTR only at doses (0.25-5.0 mg/kg) decreasing total entries; gepirone (0.1-5.0 mg/kg) was inactive. Ipsapirone (0.25-5.0 mg/kg) increased OTR and at 1.0 mg/kg antagonised the anxiogenic action of 8-OH-D-PAT, RU 24969 and BAY R 1531. Gepirone (2.5 mg/kg) failed to antagonise 8-OH-DPAT, but the dose was limited by its effect on total entries. The anxiogenic effect of a low dose of 8-OH-DPAT was also prevented by p-chlorophenylalanine (p-CPA) pretreatment and reversed to anxiolytic by 5,7-dihydroxytryptamine lesions of dorsal raphe, which spared median raphe. These lesions also abolished the anxiolytic effect of ipsapirone without affecting the anxiogenic response to yohimbine. This study provides preliminary evidence that 8-OH-DPAT may be capable of acting as an agonist and ipsapirone as an antagonist at a presynaptic site related to dorsal raphe which is separate from the site of action of yohimbine. 5-HT1A agonists and partial agonists may have multiple sites and/or mechanisms of action in the elevated X-maze.

Neurochemical and behavioral aspects of lamotrigine.

Lamotrigine (LTG), a new anticonvulsant, chemically unrelated to current antiepileptic drugs (AEDs), resembles phenytoin (PHT) and carbamazepine (CBZ) in ability to block hindlimb extension in both the maximal electroshock test and leptazol-induced seizures. Results indicate that LTG may be of value in both partial and generalized seizures. In in vitro studies, LTG has been shown to inhibit veratrine-evoked release of glutamate when a threshold depolarizing concentration (4 micrograms/ml) is used, and also inhibits aspartate release when a larger stimulus is given (10 micrograms/ml). However, LTG does not block potassium-evoked transmitter release. LTG is a less potent inhibitor of the release of gamma-aminobutyric acid (GABA), acetylcholine, noradrenaline, and dopamine. LTG blocks the neurotoxicity of kainic acid in vivo, supporting the in vitro findings, and suggests that the anticonvulsant effect of LTG may be due to inhibition of glutamate release. In a test of working memory and phencyclidine (PCP) discrimination studies, LTG had no effect, indicating no sharing of the same PCP-like side effects associated with NMDA receptor blockade. In the gerbil model of global ischemia, high doses of LTG provided protection against damage to the CA1 region of the hippocampus. Analogues of LTG of higher potency to block the release of glutamate may be necessary to ensure protection against ischemic brain damage.

Effects in the X-maze anxiety model of agents acting at 5-HT1 and 5-HT2 receptors.

Three 5-HT agonists produced a dose-related fall in open/total arm entry ratio in the elevated X-maze model of anxiety at doses which did not affect total entries. The relative potency, 8-hydroxy-2-(di-n-propylamino)tetra lin (8-OH-DPAT) much greater than 5-methoxy-N,N-dimethyltryptamine (5-MeODMT) greater than or equal to 5-methoxy-3(tetrahydropyridin-4-yl)1H-indole (RU 24969), was unrelated to the occurrence of wet dog shakes and suggests that 5-HT1 rather than 5-HT2 receptors may be involved. However, the 5-HT2 receptor antagonists ritanserin, ketanserin and seganserin caused an anxiolytic-like increase in entry ratio, although only ritanserin produced this effect across the dose range tested. +/- Pindolol, an antagonist at 5-HT1 receptors, showed a biphasic dose-response curve with a fall in entry ratio at one high dose. The effect of a submaximal dose of 8-OH-DPAT was prevented by pindolol but not by a similarly anxiolytic dose of ritanserin or diazepam. A higher dose of diazepam caused intense muscle hypotonia in combination with 8-OH-DPAT. Since open/total entry ratio appears to represent choice, rather than suppression or delay, of a response, the effects seen may indicate involvement of 5-HT receptors in anxiety separately from any change in the ability to withhold a response. The precise role of each receptor subtype, however, remains to be determined.

Cardiac reactivity in rats with acute renal failure.

Cardiac reactivity has been determined in rats with acute renal failure (ARF) induced by either bilateral nephrectomy or intramuscular glycerol injection. Rats with bilateral nephrectomy showed reduced chronotropic responses to cervical sympathetic stimulation but no appreciable alteration in the chronotropic responses to vagal stimulation. By contrast, we have previously noted that rats with glycerol-induced ARF show diminished chronotropic responses to stimulation of both nerves. The negative chronotropic and inotropic responses to carbachol in isolated atria from both nephrectomized and glycerol-injected rats were not significantly different from their respective controls. In both models of ARF the atrial positive chronotropic responses to isoprenaline were significantly decreased whilst positive inotropic responses were not significantly different from controls. The results indicate that the cause of the reduced chronotropic response to vagal stimulation observed in glycerol-injected rats is of presynaptic origin whilst the reduced chronotropic response to cervical sympathetic stimulation noted in both models of ARF may be due to an impaired postsynaptic response.