Viral-mimetic treatment during pregnancy, and its effects on postnatal brain and behaviour

Epidemiological studies have indicated an association between infection during pregnancy and an increased risk of neurodevelopmental disorders such as schizophrenia and autism in offspring. Animal models to study the prenatal origin of such outcomes of pregnancy have largely used conventional rodents which are immature at birth compared to the human neonate. The main aim of this thesis is to develop and characterise a model of the neurodevelopment basis of mental illness in the spiny mouse (Acomys cahirinus), whose offspring have completed organogenesis at birth. However the behavioural characterisation of the ‘naïve’ spiny mouse is essential to develop and optimise a set of behavioural paradigms, from which the behavioural and cognitive outcome of experimentally altered spiny mice can be assessed. The aim of studies outlined in Chapter 2 is to identify the behaviour of spiny mice in a set of commonly used behavioural tests from birth to adulthood. Behaviour tests that assessed exploratory activity and social interaction could be used from at 1d of postnatal age in the spiny mice. Tests used to assess fear and behaviour, were found not to produce reliable results before 10 days of age. The novel object recognition test is not appropriate in young spiny mice (1-20 days of age), as they do not explore the ‘novel’ objects in the open field adequately to make it a measure of learning and memory. In addition, prepulse inhibition does not occur in spiny mice until after 40 days of age. Where as a test of motor coordination, the rotarod test, does not pose a challenge to spiny mice older than 40d of age, and may not be an appropriate test for spiny mice of that age. The only difference found between female and male spiny mice across all tests was in their acoustic startle response, with females having a far greater acoustic startle response than males to the highest acoustic startle pulse although when weight was taken into account this difference did not occur. This study has optimised a set of behavioural tests that can be used to assess postnatal behaviours in the spiny mouse and provided important information about critical periods of behavioural development throughout postnatal life. The studies outlined in Chapters 3 and 4 aimed to develop and characterise an animal model of the neurodevelopmental basis of mental illness disorders such as schizophrenia and autism. A low dose of the viral mimetic, polyriboinosinic polyribocytidylic acid (Poly I:C), recognised by toll-like receptor (TLR)-3, was administered at mid-gestation to pregnant spiny mice to determine the behavioural and neurological outcome in the offspring. The treatment had no effect on maternal, fetal or neonatal survival, or postnatal growth of the offspring to 100d of age. However, offspring prenatally exposed to a viral mimeitc showed significant impairments in non-spatial memory and learning tasks, and motor activity. Examination of brain sections at 1 and 100 days of age revealed significant decreases in Reelin, increased glial fibrillary acidic protein (GFAP) expression, and increased numbers of activated microglia, specifically in the hippocampus. These findings have also been reported in the brains of autistic and schizophrenia patients in post-mortem studies. Further assessment on a more comprehensive behavioural battery is required before we are able to conclude that our animal model, involving low dose viral-mimetic treatment during pregnancy in the spiny mouse, is an appropriate in vivo model of the neurodevelopmental basis of mental illness. Nevertheless, results in Chapters 3 and 4 indicate long term debilitating effects in offspring exposed to a viral infection during pregnancy. As infections arising from various causes have a similar debilitating effect, it is thought that the maternal response, common to most infections, may be the critical factor altering fetal brain development. The aim of the study described in Chapter 5 was to determine if the immunological factors, common to most infection when recognised by TLRs, are altered within 24hr after varying doses of Poly I:C administration at mid-gestation in placental and fetal tissue. Poly I:C administration at mid-gestation decreased mRNA levels of the transcription factors NF-ĸB and IRF-3 in the placenta and fetal brain. In addition, Poly I:C administration caused a significant decrease in the mRNA levels of pro-inflammatory cytokines, IL-6 and TNF-α, in the fetal brain. These decreases in transcription factors and cytokines were not evident, in the most part, after treatment with 5mg/kg Poly I:C. Changes in gene expression in the placenta and fetal brain after Poly I:C administration, may be the mechanism underlying the disruption in fetal brain development, which subsequently leads to postnatal abnormalities, although further investigation is required to provide greater understanding. In Chapter 3 and 4 it was shown that prenatal exposure to a low dose (0.5mg/kg) of Poly I:C causes behaviour abnormalities in spiny mouse offspring. However in Chapter 5 it was shown that a moderate dose (5mg/kg) of Poly I:C causes different cytokine expression levels in the fetal brain and placenta immediately following administration in a pregnant spiny mice, compared to the lower dose. Therefore the aim of Chapter 6 was to investigate if prenatal exposure to a moderate dose (5mg/kg) of Poly I:C causes behavioural abnormalities in offspring, assessed using a more comprehensive behaviour battery. We found that prenatal exposure to a moderate dose of the viral mimetic, Poly I:C causes abnormalities in sensorimotor gating, social interaction and memory and learning. This study demonstrates that varying the severity of an infection has differential effects on the outcome of the offspring. This study provides further evidence that this is an appropriate animal model of the neurodevelopmental basis of mental illness disorders, such as schizophrenia and autism, that can be used to test potential therapeutic interventions. We have demonstrated for the first time in a precocial animal model, that prenatal exposure to a viral-like infection causes behavioral and neurological abnormalities in offspring from a young age. The results presented here lay the basis of testing treatments that might prevent or reverse these effects on brain development, and might therefore lead to therapies that help prevent the development of postnatal conditions such as autism and schizophrenia. The findings is this thesis provide compelling evidence that even a very low grade maternal virus-like infection may alter brain development, causing behavioural and neurological abnormalities to emerge in postnatal life. Thus, we believe that maternal viral infection during pregnancy is likely to be a contributory factor to the development of certain mental illness disorders in the human population.