Exploratory, anxiety and spatial memory impairments are dissociated in mice lacking the LPA₁ receptor

Abstract

Lysophosphatidic acid (LPA) is a new, intercellular signalling molecule in the brain that has an important role in adult hippocampal plasticity. Mice lacking the LPA₁ receptor exhibit motor, emotional and cognitive alterations. However, the potential relationship among these concomitant impairments was unclear. Wild-type and maLPA₁-null mice were tested on the hole-board for habituation and spatial learning. MaLPA₁-null mice exhibited reduced exploration in a novel context and a defective intersession habituation that also revealed increased anxiety-like behaviour throughout the hole-board testing. In regard to spatial memory, maLPA₁ nulls failed to reach the controls’ performance at the end of the reference memory task. Moreover, their defective working memory on the first training day suggested a delayed acquisition of the task’s working memory rule, which is also a long term memory component. The temporal interval between trials and the task’s difficulty may explain some of the deficits found in these mice. Principal components analysis revealed that alterations found in each behavioural dimension were independent. Therefore, exploratory and emotional impairments did not account for the cognitive deficits that may be attributed to maLPA₁ nulls’ hippocampal malfunction.

Introduction

Lysophosphatidic acid (LPA, 1-acyl-2-sn-glycerol-3-phosphate), acting through 6 G protein-coupled receptors (LPA_1–6), has gained increasing attention over the last few years as an intercellular messenger with several effects on different target tissues (Anliker and Chun, 2004, Birgbauer and Chun, 2006, Choi et al., 2010, Chun, 2005, Chun, 2007, Fukushima et al., 2001, Ishii et al., 2004, Moolenaar et al., 2004, Noguchi et al., 2009, Rivera and Chun, 2008). A growing body of evidence indicates that the LPA pathway is involved in normal and abnormal brain development and function (Anliker and Chun, 2004, Choi et al., 2008, Chun, 2005, Estivill-Torrus et al., 2008). The most extensively studied of these receptors is LPA₁ (Chun, 2005, Contos et al., 2000, Estivill-Torrus et al., 2008, Fukushima et al., 2002, Herr and Chun, 2007, Kingsbury et al., 2003, Matas-Rico et al., 2008).

Recently, Santin et al. (2009) described the behavioural phenotype of the maLPA₁-null mouse, a stable variant of the LPA₁-null mutant strain formerly characterised by Contos et al. (2000) and described in Estivill-Torrús et al. (2008). Impaired spatial memory retention, abnormal use of searching strategies, altered exploration in the open field and increased anxiety-like responses in the elevated plus maze have been reported in the absence of retinal and auditory malfunctions. However, concomitant neurological deficits were observed in olfaction and somesthesis, limb reflexes, co-ordinated limb use and neuromuscular strength (Santin et al., 2009). Interestingly, these behavioural alterations are accompanied by impairments in both hippocampus and cerebral cortex that may be partially responsible for the phenotype (Estivill-Torrus et al., 2008, Matas-Rico et al., 2008).

The complexity of the behavioural phenotype exhibited by the maLPA₁-null mice with impairments in several behavioural domains is frequently observed when transgenic mice are used in research (e.g. Acevedo et al., 2006, Kalueff et al., 2007, Santin et al., 2009). However, the potential relationship among sensorimotor, emotional and cognitive variables is not generally well-addressed and may lead to inaccurate interpretations. To date, it is known that anxiety-related behaviours, exploration and cognition may reflect dissociated or common processes in animal testing (Matzel et al., 2008, Miyagawa et al., 1998, Ohl et al., 2003, Ohl et al., 2002). In this regard, it has been reported that memory could be influenced by the rodent’s inborn anxiety or by its reactivity to a stressor (Herrero et al., 2006, Ribeiro et al., 1999, Wright et al., 2006). The relevance of this point is emphasized in reports that suggest that the performance of some mouse strains in certain tasks may reflect the strain’s anxiety-related behaviour, rather than cognitive functions (Dockstader and van der Kooy, 2001, Ohl et al., 2002). It is important to note that stressors, such as a novel environment or forced swimming, are usually an unavoidable part of the experimental setting even when studying non-emotional cognitive processes. Furthermore, the degree of aversion varies from one task to another, and that may explain disparate memory results between procedures (Hodges, 1996). On the other hand, anxiety levels could be related to increased or reduced locomotion (Kameda et al., 2007, Ramos and Mormede, 1998), and motor activity could influence anxiety and memory when their assessment involves spatial–temporal parameters (Brody and Holtzman, 2006, Kalueff et al., 2007, Strekalova et al., 2005).

The main purpose of this work is to study exploration, anxiety and spatial memory in maLPA₁-null mice, with a focus on the interrelationship among these characteristics, to determine whether motor activity or anxiety impairments might account for cognitive performance. To address this issue, we used the hole-board test and the principal components analysis (PCA) multivariate approach. The hole-board is a frequently used hippocampal-dependent task for measuring spatial learning that is similar to the water maze in that extra-maze cues are used to solve the task (Oades, 1981). Moreover, the hole-board, as well as its modified version, allows the simultaneous evaluation of various potentially interrelated emotional, exploratory and spatial memory measures (Ohl et al., 2002, Ohl et al., 2003, Takeda et al., 1998). PCA is useful to resolve variables into the independent dimensions (factors) that underlie behaviour (Ohl et al., 2002, Ohl et al., 2003, Ramos and Mormede, 1998). Although PCA has successfully been applied to assess behavioural paradigms and inbred strains, it has been less frequently used in studies using transgenic animals (Carola et al., 2002, Fernandes et al., 1999, Gross et al., 2000, Ohl et al., 2003). In this study, we further show the utility of PCA in analysing behavioural research using mutant mice.