Symposium 3: What Genetically-Defined Animals can Teach us About What is Being Learned in Common Laboratory Tasks

From a methodological viewpoint, the use of genetic models permits either bottom up or top down analyses of brain and behaviour relationships. In the first case, the starting point is a gene manipulation (null mutation or gene modification) with subsequent identification of the consequences at the behavioural level. In the second case, the starting point is an observed behavioural difference between selected lines or mutant vs wild type individuals that will be used to reveal correlated variations in brain properties and architecture. The data that will be presented refer to the second type of analysis. Inbred strains of mice like C57BL/6 (C57) and DBA/2 (DBA) differently perform in hippocampal-mediated tasks. That is, C57 show higher scores in spatial learning as well as in contextual fear conditioning relative to DBA. These mice also show differences in hippocampal morphology and functionality. Starting from these observations, we decided to investigated possible strain differences in the functionality of other cortical and subcortical areas involved in such tasks. Ibotenic or electrolytic lesions were performed in frontal and posterior parietal cortex as well as in the hippocampus, the amygdala and the nucleus accumbens ofC57 and DBA mice. We initially found strong differences in the lesions effects according to the genotype. In particular, although hippocampal lesions were found to disrupt spatial performance in both strains (1) frontal cortex and amygdala had a deleterious effect only in high learner C57 mice whereas (2) nucleus accumbens lesions did not produce any effect in C57 but paradoxically increased spatial performance in low learner DBA mice. However, running the same experiments in a different context (heavy cueing vs poor cueing) completely modified the strain specific effect of lesions. In particular, under heavy cueing conditions, spatial performance increased in non lesioned mice but nucleus accumbens lesions were now found to produce strongly disruptive effects in both strains. Finally, placing mice in conditions of acoustic pollution before training also modified performance and brain areas functionality in a strain-specific manner. Taking together, these results indicate that strain-specific patterns of brain and behaviour relationships can be strongly modulated by context manipulations.