Unconditioned and conditioned anxiogenic effects of the cannabinoid receptor agonist CP 55,940 in the social interaction test

Abstract

In spite of the addictive properties of cannabinoids, under certain circumstances, they can evoke strong anxiogenic and aversive responses in humans and in animal tests of anxiety. Effects of different doses of CP 55,940 (10, 20, and 40 μg/kg) were tested in the low-light, familiar (LF) apparatus test condition of the social interaction test. The 40-μg/kg dose of CP 55,940 significantly decreased the time spent in social interaction, indicating an anxiogenic effect. This dose also had an independent effect of reducing locomotor activity. In rats tested undrugged 24 h after testing with 40 μg/kg, there was a significant anxiogenic effect, indicating conditioned anxiety. The group of rats injected with 40 μg/kg immediately after the social interaction test showed an unexpected significant anxiolytic effect when tested undrugged 24 h later. In an additional experiment, rats were tested in the high-light, familiar (HF) apparatus test condition after 10 or 40 μg/kg, and only those that were tested after 40 μg/kg showed an anxiogenic effect on the test day and a conditioned anxiogenic effect when tested undrugged 24 h later. Once again, those injected with 40 μg/kg after the social interaction test displayed an anxiolytic effect when tested undrugged 24 h later. We provide the first evidence for unconditioned and conditioned anxiogenic-like responses to a cannabinoid agonist in the social interaction test.

Introduction

The motivation to use marijuana is thought to lie in part on its ability to assuage symptoms of anxiety (Porter and Felder, 2001), but cannabis users report feelings of anxiety and panic reactions Hall et al., 1994, Hall and Solowij, 1998 in equal measure to those of relaxation and euphoria (Hall et al., 1994). It is possible that the reasons for this lie in bidirectional effects of cannabinoids on anxiety, with low doses having anxiolytic effects and high doses having anxiogenic ones. The data from animal tests provide further evidence of bidirectional modulation of anxiety by the cannabinoid system. Low doses of the cannabinoid receptor agonists, nabilone (Onaivi et al., 1990), CP 55,940 Genn et al., 2003, Marco et al., 2004, and Δ⁹-tetrahydrocannabinol (THC) (Berrendero and Maldonado, 2002), induced anxiolytic effects in the elevated plus maze and light–dark crossing tests. In contrast, high doses of the cannabinoid agonist HU-210 produced anxiogenic responses in the defensive withdrawal test (Rodriguez de Fonseca et al., 1996) and enhanced emotional responding to tactile stimulation (Giuliani et al., 2000), whereas mid-high doses of CP 55,940 had anxiogenic effects in the plus maze Arévalo et al., 2001, Marı́n et al., 2003, Marco et al., 2004.

The cannabinoid receptor agonist CP 55,940 binds to the brain cannabinoid CB₁ receptor with high affinity (Herkenham et al., 1991) and has been shown to be approximately 30 times more potent than THC (Little et al., 1988). However, not all the behavioural effects of CP 55,940 seem to be mediated by the CB₁ receptor inasmuch as the reduction in directed exploration in the holeboard and the anxiogenic effect in the plus maze were not reversed by the CB₁ receptor antagonist SR 141716A Arévalo et al., 2001, Romero et al., 2002. It has recently been suggested that there is a role for a novel cannabinoid receptor in the mediation of anxiety (Haller et al., 2002) inasmuch as both wild type and CB₁ knockout mice show changes in anxiety in the plus maze in response to the functional CB₁ receptor antagonist SR141716A. This antagonist binds to the putative novel cannabinoid-sensitive receptor (Haller et al., 2002), which constitutes a candidate for the mediation of preserved effects of cannabinoids in CB₁ knockout mice Jarai et al., 1999, Zimmer et al., 1999, Di Marzo et al., 2000, Breivogel et al., 2001, Hajos et al., 2001. It is therefore possible that different cannabinoid receptors may mediate anxiety in different ways. Evidence for an endogenous anxiolytic cannabinoid tone comes from the intrinsic effects of the receptor antagonist SR141716A. This drug has anxiogenic effects in the defensive withdrawal and elevated plus maze tests in rats Navarro et al., 1997, Arévalo et al., 2001.

The anxiogenic effects of abused drugs may constitute salient stimuli to which drug-associated negative symptoms are classically conditioned. For example, the phenomenon of test-specific conditioned anxiety to an anxiogenic dose of nicotine has been demonstrated in both the social interaction and the elevated plus maze tests File et al., 2002, Tucci et al., 2002. Other drugs of abuse, such as cocaine (DeVries and Pert, 1998), produce a similar pattern of conditioned anxiety, and this phenomenon has been closely linked to cocaine's abuse potential Goeders, 1997, Goeders, 2002.

The purpose of the present experiment was to explore the possible bidirectional effects of CP 55,940 on anxiety in the social interaction test File, 1992, File and Seth, 2003, as well as whether anxiety could be conditioned to a mid-high dose of the drug. In the social interaction test, which models generalised anxiety disorder, the dependent variable is the time spent in social interaction by pairs of rats, and the anxiety generated by this test can be manipulated by changing the light level and/or the rats' familiarity within the test arena. The greatest level of anxiety is generated when rats are tested in an unfamiliar test arena that is lit by high light [high-light unfamiliar (HU) test condition]. A moderate level of anxiety is generated by testing rats in an arena with which they have been familiarised, but which is lit by high light [high-light familiar (HF) test condition]. The lowest level of anxiety is generated by testing rats in a familiar arena, lit by low light [low-light, familiar (LF) test condition]. Increases in social interaction are indicative of an anxiolytic effect, and decreases indicate an anxiogenic response. To test for possible anxiolytic effects, we used the HU and HF test conditions inasmuch as these conditions generate low levels of social interaction in the control group and are most sensitive to the effects of anxiolytics File and Hyde, 1978, File, 1980, File and Seth, 2003. To test for possible anxiogenic effects, we used the LF test condition, which generates the highest level of social interaction and is most sensitive to anxiogenic drug effects (e.g., File et al., 1982, File et al., 1988). Conditioned anxiogenic effects were explored in both the LF and HF test conditions.