Influence of circadian phase and test illumination on pre-clinical models of anxiety

Abstract

Pre-clinical models of anxiety, particularly the elevated plus-maze (EPM), have been shown to be sensitive to a variety of methodological variations. Recent research has implicated circadian phase of testing in influencing the behavioural profile of 5-HT_1A ligands on the EPM. The present study investigated the effects of testing animals during the dark and light phases and in light and subjective dark test conditions on baseline behaviour in animal models of anxiety. Eighty singly housed male Sprague–Dawley rats were exposed to a battery of unconditioned, exploratory tests (EPM, open field arena, holeboard) and a new model of extreme anxiety, the unstable elevated exposed plus-maze (UEEPM). Circadian phase of testing failed to consistently alter behaviour on any model. Level of test illumination had no effect on subjects' response to the open field arena, holeboard or UEEPM. Dark testing increased locomotor activity on the EPM (total arm entries, closed arm entries and distance moved) without decreasing open-arm avoidance. The construct of anxiety as measured by a number of different paradigms withstood major intra-laboratory manipulation of circadian phase of testing and illumination of apparatus. It is suggested that the effects of circadian rhythmicity may be confined to the behavioural profiles of serotonergic, particularly 5-HT_1A, ligands on the EPM.

Introduction

Pre-clinical models of anxiety have failed to provide consistent profiles for established and novel anxiolytic agents. For example, the widely used elevated plus-maze test (EPM) has provided contrasting profiles for serotonergic (5-HT), adrenoceptor and cholecystokinin (CCK) receptor ligands (for review, see Ref. [26]). Methodological variables such as gender, age and strain of subjects have been shown to influence behaviour on animal models and may go some way to explaining inconsistencies. For instance, sex differences have been reported in the social interaction test and the Vogel conflict test [17], while female subjects have been found less anxious than males [17], [25] on the EPM. Trullas and Skolnick [28] found that 70% of variability in plus-maze behaviours between mouse strains could be attributed to genetic factors. Strain differences have been reported for locomotor and thigmotactic behaviour in the open field arena [29] and for escape-related responses in a new model of extreme anxiety, the unstable elevated exposed plus-maze (UEEPM) [18]. Young rats have also shown less propensity to escape the UEEPM than adult rats [19]. In addition, procedural variables such as prior handling [1], individual housing, prior exposure to a novel arena [25] and illumination levels have been shown to affect baseline behaviour in pre-clinical models, hence creating further potentially confounding methodological variables.

High lighting levels have been shown to suppress locomotion and rearing and increase thigmotaxis in the open field arena [29] and decrease exploration in the black/white test box [4]. In contrast, Griebel et al. [10] reported that rats exposed to the EPM under low illumination (centre square 100 lux, open arms 80 lux, closed arms 90 lux) entered open arms more, spent a greater proportion of the trial on the open arms and were generally more active than those tested under high illumination (centre square 230 lux, open arms 220 lux, closed arms 100 lux). Thus, bright light appeared to increase the aversiveness of the test situation. However, different research groups have found both pigmented [22] and albino [2], [7] rats' behaviour on the EPM is independent of light levels ranging from 9.5 lux to 900 lux. Interestingly, level of test illumination has been found to dramatically affect the profiles of serotonergic ligands on the EPM. An identical dose of 8-OH-DPAT has produced anxiolysis at high levels of illumination and anxiogenesis at low levels [12].

Recently, rodent circadian rhythmicity has been suggested as a possible source of variability in the EPM. Rodents display circadian rhythms of approximately 24 h in a range of physiological and behavioural measures including body temperature [30], corticosterone release [31], locomotor activity [9] and response to pain [20]. Cao and Rodgers [3] propose that circadian phase of testing may be responsible for the reported contrasting effects of the selective 5-HT_1A antagonist LY 297996. The compound has produced anxiolysis in mice tested on the EPM during the mid-dark phase, but has been ineffective in mice tested during the mid-light phase. Results reported in abstract form show WAY 100635, a selective 5-HT_1A antagonist, produces significantly different effects on extracellular levels of 5-HT depending on circadian phase of administration [11]. Microdialysis of the rat hippocampus found the compound had no effect on 5-HT levels during the light phase, yet levels rose to 350% of control values during the dark phase.

Although rats' behaviour on the EPM has been shown to vary at different times during the light phase [10], to date no research has compared baseline behaviour in each of the circadian phases on a variety of animal models of anxiety. The present study aimed to assess the validity of circadian factors as potential methodological confounds. Behavioural home cage monitoring of three subjects over a 24-h period provided an index of any diurnal/nocturnal variations of activity in a familiar environment. A further group of experimentally naive, untreated rats were then exposed to a battery of unconditioned behavioural tests (UEEPM, EPM, open field arena, holeboard) during their subjective light and dark phases. To determine whether any observed differences were a function of lighting conditions and thus independent of circadian phase, two levels of test illumination, subjective light and subjective dark, were employed.