Abstract
Schizophrenia is a complex mental disorder in which the aetiology is not well understood, although evidence suggests that maternal infection increases the incidence of schizophrenia in the offspring. In the present thesis, we used an animal model of a neurodevelopmental insult where pregnant rat dams were injected with polyriboinosinic-polyribocytidilic acid (termed MIA animals) and offspring were tested in a number of tasks that model behaviours where patients with schizophrenia often show impairments.
In Chapter 2 we tested the model using behavioural tasks that probe aspects of the negative symptoms of schizophrenia, such as motivation. Contrary to what is observed in patients with schizophrenia, MIA animals exhibited what appeared to be enhanced motivation. Through a series of experiments we found that this was actually due to an inability of MIA animals to detect changes in the contingencies between their responding and the resulting outcomes.
In Chapter 3 we tested a form of memory that has been implicated in the cognitive symptoms of schizophrenia. In line with patients, MIA animals exhibited memory impairment in a displaced object recognition task. We also employed electrophysiology techniques to examine hippocampal local field potentials during exploration. We found between group differences in low beta rhythm that may relate to differences in memory discrimination capabilities between MIA and control animals.
In Chapters 4 and 5 we changed direction to test a novel therapeutic intervention protocol on control animals prior to using it on the MIA animals. Subsequently, in Chapters 4 and 5 we assessed under what conditions fluoxetine, an antidepressant that has been shown to enhance hippocampal neurogenesis; exerts a behavioural effect in hippocampal tasks with and without a navigational component. In the water maze and in the negative patterning procedure, fluoxetine was most beneficial when there was a four week delay between administration and testing and under high loads of interference.
Building on data from the MIA animals in Chapters 2 and 3 and from fluoxetine data in Chapters 4 and 5, in Chapter 6 we combined fluoxetine with MIA to assess fear and extinction memory in a neurogenesis-dependent protocol. We found that there was no difference in fear learning between fluoxetine and saline animals and between MIA and control animals. However, during extinction learning, MIA animals who were administered fluoxetine froze significantly less than control animals who were administered fluoxetine. We attributed these findings to differences in the mechanisms underlying behavioural flexibility and pattern separation and suggest that fluoxetine may affect an aberrant system differently than a healthy system.
Collectively, the data presented here contribute to the extensive evidence showing that maternal infection produces offspring whose behaviour mimics schizophrenia-like symptomology. Furthermore, fluoxetine differentially affected MIA and control animals. Our acute-delayed fluoxetine protocol is the first of its kind and the effects we observed correspond to the time taken for adult born neurons to become functionally integrated into hippocampal networks and the therapeutic delay associated with fluoxetine. We propose that MIA may mediate hippocampal neurogenesis later in life that may manifest as psychosis.