Preview

PDF, English Download (6MB) | Terms of use

Abstract

Schizophrenia is a neurodevelopmental disorder affecting about 1 % of the population. This thesis focuses on the development of an animal models of schizophrenia-related pathophysiology based on the so called “two-hit” hypothesis. According to this hypothesis, schizophrenia is the product of a combination of at least two factors, acting during the critical phases of development of the central nervous system. One “hit” reproduces hypofunction of the NMDA receptor signaling via early postnatal treatment with the NMDA receptor antagonist phencyclidine (PCP), a treatment that is toxic to inhibitory networks of brain areas such as the cerebral cortex. Second “hit”, prenatal treatment with polyinosinic : polycytidylic acid (Poly I:C), mimics maternal infection, one of the most commonly reported environmental risk factors of schizophrenia. In a sharp contrast to acute models (e.g. amphetamine-induced locomotor hyperactivity), developmental models are inherently suffering from long periods between the sickness-inducing treatments and the test phase, when various compensatory and other confounding environmental factors can interfere with the model performance. As a result, developmental models often lack robustness that is most commonly expressed as high inter-individual variability. There are several approaches to deal with these problems that are proposed and presented in this thesis. First, current analysis emphasizes the need to develop and validate tasks that can be applied repeatedly to monitor the development of the disease-like state (and potential response to pharmacological and other treatment). One such task is presented in details – a novel reinforcement learning task that is not sensitive to practice effects and robust enough to be administered for many weeks, if not months. Second, an intermediate test phase was introduced to bridge manipulations during the perinatal period of development and assessments made in adult subjects. Based on the juvenile social play behavior of the rat, correlation and cluster analyses were conducted in order to develop predictors of adult behavioral and non-behavioral abnormalities (i.e. to identify adult animals most affected by the Poly I:C/PCP treatment). In schizophrenia research, adolescence is seen as a critical time period and it was hypothesized, that pharmacological intervention during this phase could have a preventive effect on brain pathology and behavioral abnormalities in adults. Two drugs, minocycline and pregnenolone, previously reported to have anti-inflammatory, neuroprotective and disease modifying potential, were shown to prevent brain pathology (alterations in microglia density and GAD67 expression) in rats of the two-hit group when applied during adolescence. These results have two important consequences. On the one hand, they call for more discussion on the need to develop disease-modifying therapy of schizophrenia, a subject that is still not well accepted in the medical community at least in part due to stigmatization of the disease. On the other hand, such studies are the best illustration of the potential value of neurodevelopmental models. Present thesis clearly indicates that, given the limited robustness, the value of these models is not as unequivocal as one may want to have. These issues are currently not discussed in the literature that is dominated by positive evaluations of dozens of various neurodevelopmental models. Despite challenges, the Poly I:C/PCP model can be recommended for the further development and validation. Developmental models not only help to understand the impact of factors implicated in the pathophysiology of schizophrenia but are also the key to study novel therapeutic approaches.