Attention-deficit/hyperactivity disorder (ADHD) is the most commonly diagnosed childhood neuropsychiatric disorder. More than twenty years of genetic, behavioral, and pharmacological research support the hypothesis that compromised brain dopamine (DA) signaling impacts risk for the disorder. To date, however, evidence for this idea is indirect, and we therefore lack a construct-valid animal model for ADHD. To gain direct evidence for changes in DA-regulatory molecules in ADHD, we screened subjects with a DSM-IV diagnosis of the disorder and identified multiple, rare DA transporter (DAT, SLC6A3) coding variants. One of these variants, DAT Val559, induces anomalous, non-vesicular DA release in transfected cells, and efflux that can be blocked by the most common ADHD medications, D-amphetamine (AMPH) and methylphenidate. To pursue the significance of these findings in vivo, we engineered DAT Val559 knock-in mice and, as with heterologous expression studies, found the variant to support expression of wild-type (WT) levels of striatal DAT and tyrosine hydroxylase (TH) protein, as well as DA and DA metabolite levels. However, DAT Val559 mice exhibited a novel, conditional hyperactivity phenotype, darting, as well as an impaired sensitivity to the locomotor-activating properties of AMPH. In vivo microdialysis studies in the striatum of DAT Val559 animals demonstrated a pronounced elevation of basal, extracellular DA levels along with a significantly blunted efflux of DA release evoked by locally infused AMPH. Together, our studies confirm a functional impact on extracellular DA homeostasis of the DAT Val559 variant in vivo and establish the first construct-valid animal model of ADHD.

Disruption in insulin signaling is a mechanism hypothesized to underlie many neurologic and psychiatric disorders. Insulin signaling is accomplished by a complex array of intracellular effectors, including the protein kinase Akt. Akt, in turn, is a protein kinase regulated by growth factor, neurocrine, and hormonal factors, which is independently linked to psychiatric and neurologic disease, particularly schizophrenia. Recent data from our lab indicate it is also a common mechanism implicated in control of both the DAT and NET function in brain. The activity of Akt is driven by phosphorylation of this protein at two key residues: Ser473 by mTOR/RICTOR (mammalian target of rapamycin complex 2) and Thr308 by PDK1 (phosphoinositide-dependant kinase 1). Utilizing Cre-loxP technology to delete RICTOR gene in Nestin-expressing cell population, we created a transgenic mouse with disrupted neuronal Akt activity, due to severe impairment in Ser-473 phosphorylation of Akt. At a behavior level, the neuronal Rictor knockout (nKO) mouse exhibits many phenotypes that characterize animal models of schizophrenia, including PPI (pre-pulse inhibition) deficits, open field hyperlocomotion, and hypersensitivity to the psychomotor activating effects of amphetamine. These behavioral phenotypes were previously explained by dysfunctions of the dopaminergic system. However, emerging evidence suggests participation of both DA and NE circuits. At a neurochemical level, our data signifies disturbances in total monoamine homeostasis, with significant decreases in tissue DA content and elevations in NE tissue content, in both prefrontal cortex and dorsal striatum. Intriguingly, marked upregulation of the NET in cortical regions is a potential mechanism underlying the neurochemical and behavioral disruptions in the Rictor-nKO, as the Rictor-nKO demonstrates increased NET expression, and inhibition of the NET restores PPI and cortical DA deficits toward wild type levels. Here, we use the RICTOR KO mouse to analyze how the newly discovered molecular phenotype of disturbed interconnectivity between dopaminergic and noradrenergic systems may produce well-known behavioral phenotypes modeling psychiatric disorders, including schizophrenia, ADHD, major depression, and drug addiction.

The pulvinar is a dorsal thalamic nucleus strongly associated with the visual system, but of uncertain function. One of the central mysteries of the pulvinar is the direction and quality of the information sent from the cortex to the pulvinar and back along a set of projections called cortico-thalamo-cortical loops. Sherman and Guillery (1998) have proposed a methodology for classifying projections from one brain area to another. They suggest that glutamatergic projections may be classified as either drivers (fast-acting projections conveying the main message) or modulators (slow acting projections that only modify the message). In this thesis, the direction, content and organization of information transmitted through the primate pulvinar was investigated using the bush baby as a model species. Single and multiunit electrophysiological recordings were used to define areas in the pulvinar that contain visuotopic maps. Injections in these electrophysiologically defined areas were used to manipulate the activity of pulvinar cells or deliver neural tract tracers. Single and multiunit recordings were performed in the primary visual cortex (V1) while activity in connected neurons in pulvinar was silenced or increased. Immunohistochemical stains were performed on brain sections and used to in conjunction with light microscopy to analyze morphology in the thalamus and cortex. Our findings indicate that the bush baby visual pulvinar is organized into two complete retinotopic maps. Cells in these mapped areas project to the visual cortex including V1 and the secondary visual cortex (V2). Pulvinar projections to V1 have the anatomical features of modulators while pulvinar projections to V2 have the anatomical features of drivers. The pulvinar can exert a type of strong modulatory control over V1 layer II/III cells that suggests that the pulvinar may gate V1 output. Taken together, these data show that the pulvinar plays an active role in the flow of visual information between cortical areas.

Approximately 25% of individuals with autism spectrum disorder (ASD) possess elevated whole blood serotonin (5-HT) levels, termed hyperserotonemia. However, the connection between this biomarker and the pathophysiology of ASD remains unclear. Several rare, hyperfunctional variants of the serotonin transporter (SERT), a key regulator of 5-HT homeostasis in the periphery and the brain, have been identified in children with ASD. To evaluate the developmental and behavioral consequences of impaired 5-HT signaling, a knock-in mouse model was created that expressed the most frequent of these rare, ASD-associated SERT variants, Gly56Ala. In addition to exhibiting hyperserotonemia and global changes in 5-HT homeostasis in the brain, SERT Ala56 mice display alterations in behavior relevant to the core diagnostic features of ASD. While no changes in the developmental trajectory of the 5-HT system were observed in SERT Ala56 and wildtype littermates, offspring of maternal carriers of the Ala56 variant exhibit a unique pattern of developmental perturbations indicative of impaired placental function. Finally, supporting the original genetic association of the Ala56 variant with sensory aversion, we establish a connection between SERT genetic variation and patterns of sensory behavior in children with ASD. Collectively, these studies provide a foundation for future mechanistic work in the SERT Ala56 mouse model that will elucidate the underlying neurobiological causes of ASD.

Parkinsons disease (PD) is a neurodegenerative, motor disorder that is characterized by selective dopaminergic cell loss in the substantia nigra pars compacta. About 10-20% of PD cases have genetic causes; nonetheless, the idiopathic nature of the majority of PD cases calls for the contribution of environmental factors in its etiology. One such factor is the essential trace element manganese (Mn); excessive exposure can result in manganism, which shares similarities with both PD symptomatology and molecular signatures. This overlap warrants investigation into whether a particular genetic risk factor increases susceptibility of DAergic neurons to environmental risk factors. Using the Caenorhabditis elegans (C. elegans) model system, loss of two genes associated with an early-onset, autosomal recessive form of PD, pdr-1/parkin and djr-1/dj-1, results in enhanced Mn accumulation and oxidative stress that can be rescued by the expression of another PD-associated protein known as α-Syn (alpha-synuclein), a protein found aggregated in some PD cases. Moreover, the loss of pdr-1/parkin results in increased mRNA expression of a Mn export gene known as ferroportin (fpn-1). As overexpression of this exporter can rescue pdr-1/parkin mutant phenotypes, these studies support the role of abnormal metal homeostasis as a consequence of genetic mutations associated with early-onset PD.

The dysfunction of dopamine signaling can contribute to mood disorders and drug abuse. Dopamine neurons do not belong to one homogenous group and differ in their electrophysiological properties and molecular expression profiles. The midline dopamine neurons of the RLi and A10dc, which consists of the periaqueductal gray (PAG) and dorsal raphe, are particularly interesting because they project to brain regions that regulate anxiety and stress responsivity. We used a transgenic mouse line that expresses eGFP under control of the tyrosine hydroxylase (TH) promoter to explore the anatomy of dopamine neurons within the VTA, RLi, ventral PAG, and dorsal raphe. Retrograde tracer was injected into the dorsal bed nucleus of the stria terminals (BNST), a region important for stress-induced reinstatement of drug-seeking. As shown previously in rats, both A10dc and VTA dopamine neurons project to the dorsal BNST with the largest numbers of eGFP neurons labeled with tracer in the VTA, RLi, dorsal raphe and adjacent ventral PAG. The RLi neurons receive norepinephrine input, which may prime them for involvement in stress responses. Using the TH-eGFP mouse, we explored the physiology and noradrenergic modulation of these neurons. We find that RLi dopamine neurons differ from VTA dopamine neurons with respect to membrane resistance, capacitance and the hyperpolarization-activated current, Ih. Further, we found that norepinephrine increased the frequency of spontaneous excitatory postsynaptic currents (sEPSCs) on RLi dopamine neurons. This effect was mediated through the α1 adrenergic receptor (AR), as the actions of norepinephrine were mimicked by the α1-AR agonist methoxamine and blocked by the α1-AR antagonist prazosin. This action of norepinephrine did not persist following the subsequent application of prazosin and was therefore not a form of synaptic plasticity. Methoxamine increased the frequency of miniature EPSCs, indicating that the α1-AR action on glutamatergic transmission likely has a presynaptic mechanism. There was a modest decrease in sEPSC frequency with the α2-AR agonist UK-14,304. These studies illustrate a potential mechanism through which norepinephrine could recruit the activity of this population of dopaminergic neurons.

The p75 Neurotrophin Receptor (p75NTR) is a critical regulator of axon pruning and apoptosis during neurodevelopment. Because the receptor has also been associated with many injurious or pathological conditions involving oxidative stress, this dissertation project aimed to investigate a role for p75NTR in neurodegeneration induced by oxidative injury. Our work revealed that receptor is activated by 4-hydroxy-2-nonenal (HNE), a lipid peroxidation product naturally generated during oxidative stress. Treatment of sympathetic neurons with HNE caused axonal degeneration and programmed cell death; however, neurons lacking p75NTR were significantly protected from these effects. HNE exposure was not associated with production of neurotrophins, and a ligand-blocking antibody failed to prevent HNE-induced apoptosis, thus suggesting that oxidative stress activates the receptor through a neurotrophin-independent mechanism. HNE exposure resulted in metalloprotease- and γ-secretase-dependent cleavage of p75NTR, and pharmacological inhibition of these proteolytic events protected neurons from HNE-induced apoptosis. Lastly, p75NTR-/- mice were resistant to oxidative injury caused by administration of 6-hydroxydopamine in vivo. Altogether, these findings indicate that in response to oxidative stress p75NTR is activated through a ligand-independent event which triggers cleavage of the receptor by a metalloprotease and γ-secretase, thereby leading to axonal degeneration and apoptosis.

Epilepsy is a common neurological disease characterized by an enduring predisposition to generate seizures. Although multiple factors contribute to epilepsy, the majority of cases are genetic in origin. Variable expressivity is commonly observed in families with inherited mutations in epilepsy-associated genes, suggesting that variation in genetic modifiers may contribute to epilepsy phenotypes. We previously identified the modulatory voltage-gated potassium channel subunit, Kcnv2, as a candidate modifier gene in a transgenic mouse model of epilepsy. This dissertation outlines: the validation of Kcnv2 as a quantitative modifier of epilepsy in mice; the identification of KCNV2 variants in pediatric epilepsy patients; the determination of Kcnv2 regulatory regions; and the identification of mutations in a delayed-rectifier potassium channel gene, KCNB1, in individuals with epileptic encephalopathy. These studies highlight the importance of delayed-rectifier potassium current in governing neuronal excitability and demonstrate the utility of identifying and characterizing genetic modifiers to elucidate mechanisms of pathogenesis.

Epilepsy is a neurological disorder affecting almost one percent of the population, and genetic epilepsy are those caused by a presumed or unknown genetic factor(s). Mutations in GABAA receptors, pentameric chloride ion channels mediating fast inhibitory neurotransmission, have been identified in patients and families with epilepsy and found to cause epilepsy in animal models. The majority of synaptic GABAARs are αβγ type receptors composed of two α, two β and one γ2 subunits, and half of these epilepsy-associated GABAAR mutations are located in γ2 subunits encoded by the GABRG2 gene. A better understanding of how different types of epilepsy-associated GABRG2 mutations affect receptor trafficking and channel function, and how these mutations cause epilepsy in mouse models, will facilitate future epilepsy diagnosis as well as treatments. Here we have studied three different types of mutations represented by GABRG2(N79S, R82Q, and P83S), GABRG2(Q40X), and GABRG2(Q390X), in cultured HEK cells or animal models. We found that missense mutations located in receptor interface will disrupt receptor assembly and trafficking, which may be improved by slowing receptor biogenesis. We found that nonsense mutations showing loss of function could be partially rescued using gentamicin-induced stop codon read-through. Finally we showed that gene-target therapy could reverse the seizure phenotype in a mouse model carrying a detrimental mutation with dominant negative effects. To conclude, we have shown different molecular mechanisms are associated with these mutations, and distinct mutation-specific therapy may be potentially developed for future treatments.

Left-Right asymmetry of the CNS is a highly conserved feature across vertebrate classes. Asymmetry is manifest at the levels of function and connectivity. But the neural correlates of these asymmetries and their developmental underpinnings are just beginning to be explored. The zebrafish habenular nuclei have functional as well as neuronal asymmetries and offer a molecularly manipulable and highly visualizable model to studying the cellular and developmental origins of CNS left-right asymmetry. Here I report an evolving developmental network that regulates the timing of habenular neurogenesis. This network involved environmental and endocrine cues (light and melatonin), morphogenetic pathways (FGF and Nodal) and several neurogenic genes (dbx1b, lhx9, kip2 and her6). These components act as a temporal neurogenic gate, dictating the onset of neurogenesis. This neurogenic gate is regulated asymmetrically acting first in the left habenula driving earlier neurogenesis there and impacts neuronal cell type specification. Together these findings deepen our understanding of the molecular mechanisms that drive neural fate specification and how these components can give rise to left-right asymmetry in the CNS.