Swimming behaviour and post-swimming activity in Vitamin D receptor knockout mice

Abstract

Animal experiments have shown that Vitamin D plays a role in both brain development and adult brain function. The adult Vitamin D receptor null mutant mouse (VDR −/−) is reported to be less active and more anxious than wild-type litter mate controls and to have poor swimming ability. However, an “anxious” behavioural phenotype is inferred from differences in locomotor behaviour. This is a general problem in behavioural phenotyping where a neurological phenotype is inferred from changes in locomotion which will be affected by non-neurological factors, such as muscle fatigue. In this study of VDR −/−, we conducted a detailed examination of one form of motor behaviour, swimming, compared to wildtype littermate controls. Swimming was assessed using a forced swim test, a laneway swimming test and a watermaze test using a visible platform. Post-swimming activity was assessed by comparing grooming and rearing behaviour before, and 5 min after, the forced swimming test. We replicated previous findings in which VDR −/− mice demonstrate more sinking episodes than wildtype controls in the forced swim test but they were similar to controls in the time taken to swim a 1 m laneway, and in the time taken to reach a visible platform in the watermaze. Thus, the VDR −/− mice were able to swim but were not able to float. Grooming and rearing behaviour of the VDR −/− mice was similar to wildtype controls before the forced swim but the VDR −/− were much less active after the swim compared with wildtype mice which displayed high levels of grooming and rearing. We conclude that VDR −/− mice have muscular and motor impairments that do not affect their ability to swim but significantly alters the ability to float as well as their post-swimming activity. Differences in muscle strength may confound tests of activity that are used to infer an “anxious” phenotype.

Introduction

Vitamin D has a well established role in the regulation of calcium and phosphate homeostasis and in bone formation [8]. However, there is also a growing body of evidence linking Vitamin D with various aspects of adult brain function [13]. The Vitamin D receptor (VDR) is present in the brain throughout development and into adulthood [11], [25], [28]. A range of animal experiments have also demonstrated that low prenatal Vitamin D alters brain development [10], [21], and that transient prenatal hypovitaminosis D is associated with persistent changes in brain structure and neurotrophin level [12]. Of particular interest, as adults, these animals have subtle behavioural changes, characterized by spontaneous hyperlocomotion [4], [5] and altered latent inhibition [1].

In contrast to the phenotype of the rat exposed to developmental Vitamin D deficiency, VDR null mutant mice are characterized by alopecia, reductions in both body size and weight and impaired motor coordination [3]. A recent study showed that VDR knockout mice have reduced exploration in a range of tests including the holeboard, open field and elevated plus maze [16]. Scores from these tests are frequently used in animal models of anxiety, however the valid interpretation of these measures rests on normal muscle function. Another feature of the VDR knockout behavioural phenotype is poor swimming ability (as assessed by the forced swimming test) [17]. Their swimming behaviour was characterized by frequent sinking episodes and a vertical posture with a reduction in the time spent immobile (compared to both wildtype and heterozygous controls). While the forced swim test is frequently used as a model of behavioural ‘despair’, or learned helplessness, this measure can also be influenced by more ‘peripheral’ features of the musculo-skeletal system. These issues have added salience in light of the growing body of evidencing linking low Vitamin D levels and altered muscle function [22].

The aim of this study was to refine the behavioural phenotype of the VDR knockout mouse with particular reference to the possible confounding effects of muscle weakness on performance on swimming and grooming tasks. In the present study we examined the swimming behaviour of VDR knockout mice in more detail using a forced swim test, a laneway swim test and a visible platform water maze test. The latter two tests provide ways to stop swimming other than floating (i.e. escape or find a platform). As a further indication of motor behaviour we measured grooming and rearing after swimming, which has been used as a model of post-exercise fatigue in mice [6]. This has been used to show, for example, that mice inoculated with an avirulent strain of Toxoplasma gondii display exercise induced muscle fatigue, suggesting that immune stimulation can induce muscle fatigue [6]. Using this paradigm, a post-swim activity assessment was made in mice by assessing both grooming and rearing behaviour before and after a 5 min forced swim test.