Over the past decade, a combination of genetic, biological and imaging approaches have identified Disrupted-in-Schizophrenia-1 (DISC1) as a strong susceptibility gene for psychiatric illnesses including schizophrenia, bipolar disorder and major depression. DISC1 regulates various neurodevelopmental processes including neurogenesis, neuronal migration, neurite outgrowth, spine development and neurotransmitter signaling. Human post-mortem brain studies in schizophrenia have revealed some consistent neuropathological findings such as abnormal neuron morphology, reduced spine density, aberrant cytoarchitecture and interneuron deficits in the dorsolateral prefrontal cortex and hippocampus. However, the etiology and development of these histological abnormalities remain unclear. Therefore, we investigated brain histology of two independently-derived Disc1 mutant mice with point mutations (Disc1-Q31L and -L100P). Both mutants displayed reductions in cortical neuron density, decreased neurogenesis and altered cortical neuron distribution when compared to wild-type controls. Frontal cortical pyramidal neurons had shorter dendrites and reduced dendritic surface area. Spine density was also reduced on apical dendrites of both frontal cortical and hippocampal pyramidal neurons. In addition, we observed a pronounced defect in tangential migration of interneurons in the embryonic brains of Disc1-L100P mutants when compared to wild-type mice. Adult Disc1-L100P mutants also have selective alterations of calbindin- and parvalbumin-expressing interneurons in the cortex and hippocampus, decreased glutamate decarboxylase 67/parvalbumin co-localization and mis-positioned interneurons across the neocortex. Finally, we investigated the effects of GSK3α inactivation on frontal cortical neuron morphology in Disc1-L100P mutants. Disc1-L100P, GSK3α -/- and 100P/100P;GSK3α double mutants all exhibited significant reductions in dendritic length and surface area while spine density was significantly reduced only in Disc1-L100P and 100P/100P;GSK3α +/- mutants when compared to wild-type mice. Interestingly, spine density was rescued in 100P/100P;GSK3α -/- double mutants and comparable to wild-type mice. Overall, these findings are consistent with the anomalies seen in post-mortem schizophrenia studies and other Disc1 mutant mouse models, providing further evidence that DISC1 participates in neurodevelopment. GSK3 is only one of many pathways modulated by DISC1, so more research is required to fully understand how the network of DISC1-interacting proteins is involved in the pathophysiology of psychiatric disorders. Better understanding of these mechanisms could lead to the development of new treatments.

Over the past decade, a combination of genetic, biological and imaging approaches have identified Disrupted-in-Schizophrenia-1 (DISC1) as a strong susceptibility gene for psychiatric illnesses including schizophrenia, bipolar disorder and major depression. DISC1 regulates various neurodevelopmental processes including neurogenesis, neuronal migration, neurite outgrowth, spine development and neurotransmitter signaling. Human post-mortem brain studies in schizophrenia have revealed some consistent neuropathological findings such as abnormal neuron morphology, reduced spine density, aberrant cytoarchitecture and interneuron deficits in the dorsolateral prefrontal cortex and hippocampus. However, the etiology and development of these histological abnormalities remain unclear. Therefore, we investigated brain histology of two independently-derived Disc1 mutant mice with point mutations (Disc1-Q31L and -L100P). Both mutants displayed reductions in cortical neuron density, decreased neurogenesis and altered cortical neuron distribution when compared to wild-type controls. Frontal cortical pyramidal neurons had shorter dendrites and reduced dendritic surface area. Spine density was also reduced on apical dendrites of both frontal cortical and hippocampal pyramidal neurons. In addition, we observed a pronounced defect in tangential migration of interneurons in the embryonic brains of Disc1-L100P mutants when compared to wild-type mice. Adult Disc1-L100P mutants also have selective alterations of calbindin- and parvalbumin-expressing interneurons in the cortex and hippocampus, decreased glutamate decarboxylase 67/parvalbumin co-localization and mis-positioned interneurons across the neocortex. Finally, we investigated the effects of GSK3α inactivation on frontal cortical neuron morphology in Disc1-L100P mutants. Disc1-L100P, GSK3α -/- and 100P/100P;GSK3α double mutants all exhibited significant reductions in dendritic length and surface area while spine density was significantly reduced only in Disc1-L100P and 100P/100P;GSK3α +/- mutants when compared to wild-type mice. Interestingly, spine density was rescued in 100P/100P;GSK3α -/- double mutants and comparable to wild-type mice. Overall, these findings are consistent with the anomalies seen in post-mortem schizophrenia studies and other Disc1 mutant mouse models, providing further evidence that DISC1 participates in neurodevelopment. GSK3 is only one of many pathways modulated by DISC1, so more research is required to fully understand how the network of DISC1-interacting proteins is involved in the pathophysiology of psychiatric disorders. Better understanding of these mechanisms could lead to the development of new treatments.

The genetic contribution to the aetiology of psychiatric illness is well-established; however, few variants that alter the encoded protein have been irrefutably identified as causative, leading to the hypothesis that variants affecting gene regulation may play a pathogenic role. This thesis focuses on two genes, Neuregulin 1 (NRG1) and Disrupted in Schizophrenia 1 (DISC1), for which there is strong genetic evidence for involvement in psychiatric illness, as well as evidence for altered expression in patients. Association analysis was carried out to assess the involvement of six intronic NRG1 single nucleotide polymorphisms (SNPs) in schizophrenia and bipolar disorder in two independent samples from the Scottish (Scottish 2; n = 307 control subjects, 303 schizophrenic patients, and 239 bipolar disorder patients and German populations (n = 397 control subjects, 396 schizophrenic patients, and 400 bipolar disorder patients). These SNPs form two haplotypes, one encompassing the 5’ and promoter region of the gene and the other located at the 3’ end of the gene, that were previously associated with schizophrenia and bipolar disorder in a Scottish sample (Scottish 1). The location of these haplotypes, together with the prior evidence for altered NRG1 expression in schizophrenia, suggested the potential involvement of regulatory variants. On combining the Scottish 1 and Scottish 2 samples (combined n = 765 control subjects, 682 schizophrenic patients and 601 bipolar disorder patients), a two- SNP haplotype spanning both coding and non-coding regions in the 3’ region was associated with schizophrenia (p = 0.0037, OR=1.3, 95% CI: 1.1-1.6) and the combined schizophrenia and bipolar disorder case group (p = 0.0080, OR=1.2, 95% CI: 1.1-1.5), with both these associations remaining significant after permutation analysis (p = 0.022 and p = 0.044, respectively). To further understanding of how DISC1, a leading candidate gene for schizophrenia that has also been implicated in other psychiatric disorders, is regulated the previously uncharacterised promoter region was assessed both bioinformatically and in vitro using the dual luciferase reporter assay. The region was found to lack canonical promoter motifs but to contain a CpG island, consistent with DISC1’s ubiquitous pattern of expression. A region located 300bp to -177bp relative to the transcription start site (TSS) was identified as contributing positively to DISC1 promoter activity, whilst a region -982bp to -301bp relative to the TSS was found to confer a repressive effect. FOXP2, a transcription factor which is mutated in a rare speech and language disorder and implicated in autism pathogenesis, was found to repress transcription from the DISC1 promoter. Two pathogenic FOXP2 point mutations reduced this transcriptional repression. Preliminary evidence for a bi-directional regulatory relationship between DISC1 and FOXP2 was observed: a mouse model of schizophrenia that carries a Disc1 L100P amino acid substitution and shows altered developmental Disc1 expression was also found to show altered developmental expression of Foxp2. These results further understanding of two genes whose altered expression might contribute to the pathogenesis of psychiatric illness.

Disrupted in Schizophrenia 1 (DISC1) is a well supported risk factor for schizophrenia, bipolar disorder and major recurrent depression. DISC1 is a multifunctional multicompartmentalised scaffold protein with essential roles in neuronal proliferation, differentiation, migration and integration. DISC1 also modulates pathways of vital importance for neuronal signalling and plasticity. One of the major hypotheses for the cause of psychiatric illness is N-methyl-D-aspartate (NMDA) receptor hypofunction. It was observed that NMDA receptor antagonists can induce symptoms of schizophrenia in unaffected individuals, and exacerbate symptoms in patients with schizophrenia. Recent work in our laboratory showed that DISC1 complexes with NMDA receptors within the cell body and at synapse of neurons. Here I studied whether DISC1, or DISC1 missense variants, affect the trafficking of NMDA receptors. This was done by quantifying surface NMDA receptor expression in the presence of DISC1 or variant DISC1. I found that one common variant, 607F, causes a significant reduction in surface expressed NMDA receptors. I went on to show that DISC1 reduces the number of internalised receptors associating with early RAB5-containing endosomes. This indicates that DISC1 may be involved in the trafficking and recycling of NMDA receptors, a process that may be affected by the missense DISC1 variant 607F. Further to this I studied the effects on NMDA receptor trafficking of DISC1 pathway partners Nuclear Distribution Element 1 (NDE1) and Trafficking-protein kinesin binding 1 (TRAK1), both regulators of neuronal intracellular trafficking. Phosphorylation of NDE1 at T131 has been shown to be modulated by DISC1. Using phospho-mimic and phospho-dead NDE1 expression constructs I observed a significant reduction in the surface-expressed NMDA receptors in cells expressing the phospho-mimic form of NDE1. NDE1 may therefore be involved in the trafficking of NMDA receptors, and this role may be modulated by phosphorylation of NDE1. Finally, TRAK1 was shown to associate robustly with the GluN2B subunit, and to decrease the surface expression of NMDA receptors, most likely by sequestering them. The TRAK1-induced GluN2B sequestration may be an artefact, but the association of the trafficking molecule TRAK1 with this subunit may point towards a role in NMDA receptor trafficking. These proteins have been shown to associate with each other and may form a complex in order to traffic NMDA receptors. Disruption of this complex by defective DISC1 expression may affect NMDA receptor trafficking. In the brain this could conceivably contribute to NMDA receptor hypofunction and the development of psychiatric illness.

Schizophrenia is a relatively poorly understood, debilitating psychiatric disorder affecting around 0.5% of the population worldwide. The main characteristics of the disease are hallucinations, delusions and cognitive impairment such as difficulty in learning. It has been recently suggested that Disrupted-in-Schizophrenia-1 (DISC1) might be one of the main genetic risk factors for this disease. Mouse Disc1 has been implicated in brain development, mainly in neurite outgrowth, integration of newborn neurons, neuronal precursor proliferation/differentiation and neuronal migration. Disc1 function in the cortical excitatory cells was studied in fair detail but there is little data on Disc1 role in cortical interneuron development. In this study I have investigated development of the cortical interneurons in 21 days old mice with ENU-induced point mutations in the mouse Disc1 sequence - L100P and Q31L; previously characterized as ‘schizophrenic-like’ and ‘depressive-like’ respectively. Bin analysis was performed on five brain regions: frontal and central primary somatosensory (fSSp and SSp respectively) cortices, ventral auditory (vAud) cortex, visual (Vis) cortex and medial prefrontal cortex (MPFC); for four major interneuronal markers: parvalbumin (PV), somatostatin (STT), calretitnin (CLR) and glutamate decarboxylase 67 (GAD67). A significant decrease in PV (protein and mRNA) expression was observed in a subclass of the cortical interneurons in the fSSp, SSp, vAud and Vis cortices of L100P homozyogous (L100P) and heterozygous (L100P +/-) mouse brains when compared to their wild-type (WT) littermates. No such difference in the PV positive cells was found in the MPFC in the L100P mouse brain. Other interneuronal markers expression was not different in the L100P and L100P +/- brain from that in the WT littermate controls. Furthermore, there was no significant difference in any of the interneuronal markers expression in the Q31L mouse brain cortex. A minor change in the relative distribution of the interneurons (GAD67 positive cells) was found in the L100P but not Q31L brain. With no difference in the number of the interneurons and the nature of PV expression regulation, the cell non-autonomous effect of L100P Disc1 on this subpopulation of intereneurons was investigated. Overexpression of the mouse Disc1-100P in utero in the radial glia cells born at E14.5 (future layer II/III and IV excitatory cells) resulted in a significant decrease in the PV positive cells in all of the electroporated regions (fSSp, SSp, vAud and Vis cortices) when compared to mouse WT Disc1 overexpression. Furthermore, a decrease in the PV cells on the contralateral side was observed in the SSp and Vis cortices. This study demonstrates that mouse Disc1 is involved in the generation of parvalbumin expressing interneurons within the cortex in a cell non-autonomous way. The L100P point mutation in Disc1 led to downregulation of parvalbumin, which in turn would result in abnormal inhibitory properties of this interneuron subtype.

DISC1, a strong genetic candidate for psychiatric illness, is a molecular scaffold residing in multiple subcellular compartments, where it regulates the function of interacting proteins with key roles in neurodevelopment and plasticity. Both common and rare DISC1 missense variants are associated with risk of mental illness and/or brain abnormalities in healthy carriers, but the underlying mechanisms are unclear. In this thesis, I initially examine the effect of a panel of common and rare single amino acid substitutions on DISC1 subcellular targeting, establishing that the rare mutation R37W and the common variant L607F disrupt DISC1 nuclear targeting in a dominant-negative fashion. This finding predicts that DISC1 nuclear expression is severely impaired in 37W and 607F carriers. In addition, I show that the L607F substitution results in aberrant cytoplasmic and cytoskeletal distribution of DISC1. In the nucleus, DISC1 interacts with the transcription factor ATF4, which is involved in the regulation of cellular stress responses and memory consolidation. Here I show that at basal cAMP levels, wild-type DISC1 strongly inhibits the transcriptional activity of ATF4, and this effect is ablated by 37W and 607F, most likely as a consequence of their defective nuclear targeting. 607F additionally reduces DISC1/ATF4 interaction, which likely contributes to its weakened inhibitory effect. I also demonstrate that DISC1 modulates transcriptional responses to endoplasmic reticulum stress, and that this modulatory effect is also ablated by 37W and 607F. By providing evidence that single amino acid substitutions of DISC1 associated with psychiatric illness impair its regulatory function on ATF4-dependent transcription, I highlight a potential mechanism by which these protein variants may impact on molecular pathways underlying cognition and stress responses, two processes of direct relevance to psychiatric disease.

Major mental illness is now recognised as one of the leading causes of adult morbidity. Of the adult onset psychiatric disorders, the functional psychoses (schizophrenia, bipolar disorder and recurrent major depression) are the most severe and most common in the general population. Evidence suggests that certain genetic factors influence an individual’s susceptibility to developing these disorders when combined with appropriate social and environmental conditions. Several good candidate genes have been identified. Of relevance to this study is Disrupted in Schizophrenia 1 (DISC1) which was identified in a large Scottish family that carried a balanced translocation (t1:11) and had a history of major mental illness. In 2008, two ENU mutant mouse models with missense mutations in exon 2 of Disc1 were characterised and found to have behavioural and neuroanatomical phenotypes consistent with schizophrenia and major depression. The primary aim of this thesis is to further analyse these mouse models by performing whole genome gene expression studies and secondary protein analysis to identify genes involved in the aetiology of schizophrenia and major depression. My initial analysis used Illumina BeadChip microarray technology to identify 368 genes that were differentially expressed in ENU mutant animals under different biological conditions, compared to appropriate control animals. Nine biological groups were compared including one embryonic group at E13, and three groups treated with appropriate anti-psychotic or anti-depressant drugs. Of the 368 genes identified as differentially expressed, 46 were chosen for validation by qRT-PCR based on fold-change, p-value, functional significance, overenrichment of GO terms, pathway analysis and previous implications in major mental illness. NRXN1, NRXN3 and CDH11 were found to be significantly up-regulated in the schizophrenia mouse model with EGR4 significantly down-regulated compared to C57BL/6J wild-type controls. These findings were also replicated in an independent sample using wildtype littermates. The mental retardation gene PAK3 was up-regulated in the schizophrenia mouse model and expression levels were corrected to a level not significantly different to wild-type, when treated with the PDE4 inhibitor Rolipram. Semi-quantitative western blotting also confirmed the disregulation of EGR4 and PAK3 at the protein level in these animals. RNA expression profiles were also characterised for each of the genes above, and DISC1, through development. In summary this thesis describes the striking disregulation of four prominent genetic candidates of major mental illness in an independent animal model. A first functional link between DISC1 and NRXN1 is described suggesting, for the first time, a DISC1- dependant mechanism for regulating neurexin gene expression.

BACKGROUND: Schizophrenia is a debilitating disorder that has a profound impact on quality of life due to the presence of both cognitive deficits and psychotic symptoms. Despite having significant global economic and social costs and a worldwide prevalence of 1%, schizophrenia is still not well understood. Research has been making strides in uncovering the pathophysiology and the etiology that drive this disease, ranging from genetic abnormalities, disrupted circuitry, changes in microarchitecture, to impaired synaptic connectivity. Evidence suggests that disrupted-in-schizophrenia-1 (DISC1) driven genetic disturbances in fast-spiking parvalbumin (PV) neurons and their surrounding perineuronal nets (PNNs) likely contribute to schizophrenia etiology as they are part of the microcircuits required for working memory, a cognitive function that has been consistently impaired in schizophrenic patients. OBJECTIVE: To identify the neuroanatomical changes in PV neurons and surrounding PNNs in the superficial and deep layers of the prelimbic and infralimbic prefrontal cortex of a rat DISC-1 knockout model. METHODS: 19 DISC1-KO male rats and 15 wildtype rats were treated with saline or MK-801. They were sacrificed between P268-269 and brains were extracted and separated at the corpus callosum. After fixing and preserving, the brains were sliced then stained to visualize parvalbumin and perineuronal nets with immunohistochemistry. Slices were imaged and analyzed for PV, PNN, and PV+PNN counts in the superficial and deep regions of the prelimbic and infralimbic cortices. Averages counts within each group were taken and analyzed via 2-way ANOVAs for each brain region and dependent variable. RESULTS: DISC1-KO rats displayed the following trending changes: decreased PV cells in deep layers of infralimbic and decreased PNNs throughout the prelimbic cortex. MK-801 appears to increase the number of unsheathed PV cells in the superficial layers of prelimbic and infralimbic cortex. It decreased the number of PNNs in the prelimbic of wildtype animals but not in the DISC1-KO cohort. MK-801 moderately increased PV counts in DISC1-KO. CONCLUSIONS: This DISC1-KO model is a promising model of schizophrenia as we see the same directionality of decreases in PV and PNN as post mortem human studies. Furthermore, MK-801 is seen to have an increasing trend effect on PV cells, which should be considered when interpreting findings in future studies that look at these markers.

Amygdala kindling is commonly used to study the neural mechanisms of temporal lobe epilepsy and its behavioral consequences. The repetitive seizure activity that occurs during kindling is thought to induce an extensive array of structural and functional modifications within the brain, particularly in the hippocampus and dentate gyrus regions. Some of these changes include the growth or sprouting of new axonal connections as well as the birth and integration of new neurons into hippocampal circuits. Previous work has shown that these changes in structural and functional plasticity are not necessarily beneficial events. For instance, the growth and reorganization of synaptic terminals in the hippocampus and other brain regions might serve as a substrate that enhances hyperexcitability and seizure generation. In addition, although seizures induce the birth of new neurons, many of these newly generated cells migrate and function improperly within the hippocampal networks. Considering the prominent role of the hippocampus in a variety of behaviours, including learning, memory, and mood regulation, it would appear that alterations involving the structural and functional properties of both mature and newly born neurons in this region could impact these hippocampal-dependent functions. However, to date, the role of kindling-induced changes in hippocampal structural plasticity and neurogenesis on behaviour is incomplete, and the molecular mechanisms that govern these pathological events are poorly understood.<p/> The aim of this dissertation is to gain a better understanding of the changes in synaptic plasticity and neurogenesis within the hippocampus that occur after amygdala kindling. In chapter 2, we will examine if kindling alters the expression of synapsin I, a molecular marker of synaptic growth and activity, in both the hippocampus and other brain regions. In addition, we will also set out to determine if changes in synapsin I are related to the development of behavioural impairments associated with kindling. In chapter 3, the effect of kindling on hippocampal neurogenesis will be examined. In addition, we will also evaluate the effect of kindling on the expression of Reelin and Disrupted-in-Schizophrenia 1 (DISC1), two proteins instrumental for mediating proper neuronal migrational and maturation during development. In chapter 4, the effect of altered DISC1 expression in the dentate gyrus after kindling will be examined more extensively. We will examine whether altered DISC1 expression in the dentate contributes to some of the pathological features associated with seizure-induced hippocampal neurogenesis, such as ectopic cell migration and dentate granule cell layer dispersion. Finally, in chapter 5, the impact of aberrant seizure-induced neurogenesis on behaviour will be examined by determining if seizure-generated neurons functionally integrate and participate in hippocampal circuits related to memory processing. The results of this dissertation enhances our understanding of the functional consequences that altered hippocampal synaptic plasticity and neurogenesis may have on the development of epilepsy and emergence of cognitive impairments associated with chronic seizures.<p/>

Amygdala kindling is commonly used to study the neural mechanisms of temporal lobe epilepsy and its behavioral consequences. The repetitive seizure activity that occurs during kindling is thought to induce an extensive array of structural and functional modifications within the brain, particularly in the hippocampus and dentate gyrus regions. Some of these changes include the growth or sprouting of new axonal connections as well as the birth and integration of new neurons into hippocampal circuits. Previous work has shown that these changes in structural and functional plasticity are not necessarily beneficial events. For instance, the growth and reorganization of synaptic terminals in the hippocampus and other brain regions might serve as a substrate that enhances hyperexcitability and seizure generation. In addition, although seizures induce the birth of new neurons, many of these newly generated cells migrate and function improperly within the hippocampal networks. Considering the prominent role of the hippocampus in a variety of behaviours, including learning, memory, and mood regulation, it would appear that alterations involving the structural and functional properties of both mature and newly born neurons in this region could impact these hippocampal-dependent functions. However, to date, the role of kindling-induced changes in hippocampal structural plasticity and neurogenesis on behaviour is incomplete, and the molecular mechanisms that govern these pathological events are poorly understood.<p/> The aim of this dissertation is to gain a better understanding of the changes in synaptic plasticity and neurogenesis within the hippocampus that occur after amygdala kindling. In chapter 2, we will examine if kindling alters the expression of synapsin I, a molecular marker of synaptic growth and activity, in both the hippocampus and other brain regions. In addition, we will also set out to determine if changes in synapsin I are related to the development of behavioural impairments associated with kindling. In chapter 3, the effect of kindling on hippocampal neurogenesis will be examined. In addition, we will also evaluate the effect of kindling on the expression of Reelin and Disrupted-in-Schizophrenia 1 (DISC1), two proteins instrumental for mediating proper neuronal migrational and maturation during development. In chapter 4, the effect of altered DISC1 expression in the dentate gyrus after kindling will be examined more extensively. We will examine whether altered DISC1 expression in the dentate contributes to some of the pathological features associated with seizure-induced hippocampal neurogenesis, such as ectopic cell migration and dentate granule cell layer dispersion. Finally, in chapter 5, the impact of aberrant seizure-induced neurogenesis on behaviour will be examined by determining if seizure-generated neurons functionally integrate and participate in hippocampal circuits related to memory processing. The results of this dissertation enhances our understanding of the functional consequences that altered hippocampal synaptic plasticity and neurogenesis may have on the development of epilepsy and emergence of cognitive impairments associated with chronic seizures.<p/>