CCK receptor antagonists in animal models of anxiety: comparison between exploration tests, conflict procedures and a model based on defensive behaviours

The present experiments compared the behavioural effects of one cholecystokinin(A) (CCK(A); lorglumide) and two CCK(B) (PD 135,158 and LY 288513) receptor antagonists in classical animal models of anxiety, including conflict tests (punished lever pressing and Vogel drinking tests in rats) and exploratory models (elevated plus-maze test in rats and light/dark choice test in mice), and a recently developed mouse defence test battery (MDTB) which has been validated for the screening of both anti-panic and classical anxiolytic (i.e. benzodiazepines) drugs. Diazepam was used as a positive control. Results showed that all three CCK receptor antagonists were inactive in both conflict tests. Furthermore, despite the incorporation of more ethologically-derived measures (i.e. risk assessment activities or directed exploration, or both) no effects were observed in the elevated plus-maze and in the light/dark tests. These profiles contrast with that of diazepam which displayed clear anxiolytic-like effects in these models. In the MDTB, the CCK receptor antagonists failed to modify parameters (i.e. risk assessment, defensive threat/attack and escape attempts), which have been shown to be particularly sensitive to drugs effective in the treatment of generalized anxiety. By contrast, the CCK(B) receptor antagonists PD 135,158 (0.001-0.01, 1 mg/kg, i.p.) and LY 288513 (1 and 3 mg/kg, i.p.) significantly decreased avoidance distance when the rat was first placed in the test apparatus, an effect which is consistent with an anti-panic-like action. Overall, these findings support the idea that classical animal models of anxiety may not be suitable for evaluation of the behavioural effects of CCK receptor antagonists, whereas tests which may model certain aspects of human panic such as the MDTB appear to be more reliable tools when screening such compounds.