The dissertation is divided into two separate experiments that explore the effects of visual-spatial learning on PSD-95 dorsal hippocampal expression. Specifically, the aim of these studies was to explore the effect of learning an assay, the Hebb-Williams mazes, on the protein expression of PSD-95 in Fmr1 KO mice. PSD-95 is an important scaffolding protein hypothesized to be involved in learning and memory. In cellular models of Fragile X Syndrome it has been shown to be dysregulated but it has never been measured following behavioural learning. Establishment of a deficit using an ecologically valid behavioural assay could lead to the development of novel interventions. Study one employed a subset of the Hebb-Williams mazes of various levels of difficulty to evaluate PSD-95 protein expression in Fmrp intact and Fmr1 KO mice following learning. The results revealed significant increases in PSD-95 protein expression in control runners when compared to Fmr1 KO mice. There was a negative correlation between PSD-95 protein levels and mean total errors on the mazes meaning that as expression was increased, errors were decreased. The goals of study two were to reverse the molecular and behavioural deficits using pharmacological antagonist treatment shown to be effective in cellular models of Fragile X Syndrome. Fmr1 KO mice were treated with either saline or 20 mg/kg of a metabotropic glutamate receptor antagonist, 2-Methyl-6-(phenylethynyl) pyridine (MPEP). Relative to saline treated controls, drug treated Fmr1 KO mice made fewer errors on the same subset of Hebb-Williams mazes used in study one. Latency to complete these mazes did not differ between groups, indicating that MPEP treatment does not adversely affect motor functioning. Protein assessment revealed that PSD-95 was selectively rescued in MPEP treated mice and not saline controls. Similar to study one, a negative correlation between PSD-95 protein levels and mean total errors was observed. When taken together, these studies indicate that protein deficits are associated with a deficit of learning that can be reversed with a selective glutamate receptor antagonist. One of the strengths of the Hebb-Williams mazes is that performance is measurable without floor or ceiling effects, which plague other common behavioural assays. These data further suggest that pharmacological antagonist treatments may be promising in correcting the learning deficits in human Fragile X Syndrome patients.

This dissertation describes separate but related studies that explore visual spatial learning and memory in Fragile X Syndrome. Across all studies, either the performance of individuals affected by FXS and/or fmr1 KO mice was compared to comparison controls on seven H-W mazes of increasing difficulty levels. Study one employed the traditional configuration of the H-W mazes to evaluate performance variables that include latency to complete the maze and number of the errors. The results of study 1 revealed significant differences in performance for both FXS groups as compared to mental age-matched comparison individuals and wild type mice, respectively. In contrast to the FXS group, performance of the comparison group improved as indicated by significantly fewer errors across trials. A similar pattern of results was observed when latency across trials was analyzed. Taken together, the results of study one support the hypothesis that a selective deficit in spatial learning and memory characteristic of the FXS phenotype can be observed in the murine model of FXS, if equivalent tasks are employed in testing humans and mice. Study two expanded on these findings by adding landmarks to the maze environment to evaluate how these may impact spatial learning and memory in fmr1 KO mice. Contrary to our hypotheses, landmarks significantly impaired wild type control performance. In addition, results revealed that the performance of the fmr1 KO mice generally did not differ between landmark and non-landmark tasks, indicating that the presence of landmarks neither enhanced nor hindered mouse performance. Lastly, study three entailed a more in-depth behavior analysis of maze navigation performance for FXS individuals from study 1. Consistent with the hypotheses and findings from study 1, results revealed significant differences in performance variables between individuals, with FXS participants generally performing worse than the comparison group participants. Taken together, the results of study 3 generally supported the hypothesis that there was greater impairment in performance for individuals affected by FXS as compared to controls. This impairment was evident in the pattern of pathways taken to solve H-W mazes, consistent with the notion that affected individuals employed different behavioral strategies.