N-methyl-D-aspartate (NMDA) receptors are a subtype of ligand-gated ionotropic glutamate receptors that are involved in most fast, excitatory neuronal transmission in the mammalian central nervous system (CNS). NMDA receptor activity is crucial for normal brain function, and NMDA receptor dysregulation has been linked to a number of diseases of the CNS. There are several NMDA receptor subtypes. Each subtype has a unique temporal and spatial expression pattern, suggesting that different subtypes play different physiological roles in the CNS. Here, we have investigated how changes in membrane voltage impact the activity of various NMDA receptor subtypes in the absence and presence of the highly physiologically relevant channel blocker magnesium (Mg&#xB2;&#x207A;). Mg²⁺ strongly blocks all NMDA receptor subtypes at, and near, typical resting membrane potentials. Only upon depolarization is Mg²⁺ block relieved. We found that, upon depolarization, NMDA receptors containing NR2C or NR2D subunits unblock Mg2+ very rapidly (τ < 1 ms), while Mg²⁺ unblock from NMDA receptors containing the NR2A or NR2B subunit displays a prominent slow component (τ of several ms). We go on to show that the slow component of Mg²⁺ unblock from NR2A and NR2B containing NMDA receptors actually reflects inherent voltage-dependent alterations in NMDA receptor gating. In the absence of Mg²⁺, NR2A and NR2B containing NMDA receptor currents are enhanced upon membrane depolarization. Utilizing data collected in the absence of Mg²⁺, we developed kinetic models of NR2A and NR2B containing NMDA receptors that included inherent voltage sensitivity such that the receptors open more rapidly at positive membrane potentials. The NR2 subunit specific models reproduce experimentally recorded currents during changes in membrane voltage in both the absence and presence of Mg²⁺. The models also reproduce several other previously described voltage-dependent characteristics of the NMDA receptor channel. Inherent voltage dependence further emphasizes the strong link between NMDA receptor activity and neuronal depolarization.

In the mouse HT22 hippocampal cell line and immature primary cortical neurons, excessive glutamate treatment results in intracellular cysteine depletion, subsequent glutathione loss and the steady accumulation of reactive oxygen species (ROS). This form of oxidative stress ultimately leads to cell death. Previous data from our laboratory had shown that delayed and persistent activation of extracellular signal-regulated kinases-1/2 (ERK1/2) is associated with glutamate induced oxidative toxicity in HT22 cells and immature primary neurons. In addition, U0126, a specific inhibitor of the ERK-activating kinase, MEK-1/2, inhibits ERK activation and prevents cells death induced by glutamate. However the mechanisms responsible for this chronic activation of ERK during oxidative stress have not been well characterized. Results from this thesis demonstrated that overexpression of a dominant negative mutant of MEK1 blocked glutamate toxicity in transfected HT22 cells. These data confirmed previous results and illustrated that ERK1/2 activation is necessary for oxidative toxicity. However overexpression of a constitutively active MEK1ERK chimera (LA-MEK1ERK2) that induced robust ERK activation and translocation into nucleus did not trigger toxicity in HT22 cells. Thus, ERK1/2 phosphorylation and activation is not sufficient for glutamate induced cell oxidative toxicity. Activation of ERK1/2 in HT22 cells has a distinct kinetic profile with an initial peak occurring between 30 minutes and 1 hour of glutamate treatment and a second peak typically emerging after 6 hours. I demonstrate here that the initial phase of ERK1/2 induction is due to activation of metabotropic glutamate receptor type I (mGluRI). ERK1/2 activation by mGluRI contributes to an HT22 cell adaptive response to oxidative stress as glutamate induced toxicity is enhanced upon pharmacological inhibition of mGluRI. The protective effect of ERK1/2 activation at early times after glutamate treatment is mediated by a restoration of glutathione (GSH) levels that are reduced due to depletion of intracellular cysteine pools. Additional results suggest that mGluRI may be involved in regulating mRNA and protein levels of glutamate-cysteine ligase (GCL), which would lead to enhanced glutathione synthesis. Thus, ERK1/2 activation by mGluRI protects HT22 cells from oxidative toxicity through upregulation of GCL transcription and translation, and subsequent enhancement of GSH levels.

Verbal working memory (VWM) is the ability to dynamically preserve and manipulate verbal information for brief periods of time. VWM is maintained through a silent inner speech process (Baddeley, 1986; Baddeley & Hitch, 1974). It is well established in the behavioral and neuroimaging literature that VWM can be disrupted by the simultaneous (concurrent) performance of simple speech tasks (e.g. overt concurrent articulation of a word or digit) (Caplan et al., 2000; Larsen & Baddeley, 2003). Our primary goal in these experiments is to test whether VWM and overt concurrent articulation will have one or more overlapping regions of activation in areas commonly associated with speech processing, and to determine whether such regions are active during simple tapping tasks. Due to concerns about overt movement artifacts, we also explore covert version of speech and tapping tasks. Experiment 1 was a behavioral study that examined the effects of overt and covert concurrent articulation and finger tapping on VWM. We found that overtly and covertly concurrently articulating the were the most detrimental to subjects recall ability. These effects could be attributed to dual-task interference effects at the level of inner speech in VWM, thus, indicating a shared set of neural regions for all speech and VWM. At the same time, the effect sizes were different for the overt and covert versions of our tasks, raising questions about the common assumption of shared substrates. Experiment 2 was an imaging study designed to examine whether there were shared neural regions between simple speech tasks and VWM and to further explore differences between overt and covert tasks. The results from this experiment provided only weak evidence implicating two candidate regions as the shared locus of activation: the left cerebellum and left superior temporal gyrus. We also found interesting evidence in support of distinct sets of regions for overt versus covert versions of the tasks.

The behavioral consequences of norepinephrine (NE) in the basolateral amygdala (BLA) have long been well established. NE increases in the amygdala in response to the presentation of aversive stimuli, presumably due to an activation of locus coeruleus (LC) neurons that send NE efferents to the BLA. The following studies examine the electrophysiological consequences of alterations of the NE system on neuronal activity within the BLA. Single unit recordings of neurons of the BLA were performed, and responses to systemic administration of the anxiogenic agent yohimbine were examined. Yohimbine had both excitatory and inhibitory effects on spontaneous and afferent-evoked neuronal activity of BLA neurons. This was accompanied by a yohimbine-induced increase in NE levels within the BLA, confirmed with microdialysis. To more precisely examine the effects of NE within the BLA on neuronal activity, we used iontophoresis combined with single unit recordings of BLA neurons. NE directly applied to BLA neurons causes predominantly inhibitory effects. Spontaneous activity was inhibited, presumably via alpha-2 receptor mechanisms, while a smaller subset of neurons were excited via beta receptor actions. NE also inhibited afferent-evoked activity of BLA neurons. Footshock and LC stimulation each caused both excitatory and inhibitory effects on BLA neuronal activity; those effects could be mimicked by NE iontophoresis. Therefore, NE effects are representative of those caused by aversive stimulus presentation (footshock), or by activation of LC neurons. Chronic stress alters the activity of the NE system, the responsivity of BLA neurons, and behavioral consequences of NE on targets. Our final studies addressed whether chronic cold exposure (7 or 14 days, 5C) alters NE modulation of BLA neuronal activity. After 14 days of v cold exposure, NE caused more excitation of spontaneous and afferent BLA neuron activity, in contrast to the NE-induced inhibition seen in control rats. Seven days of cold stress caused only moderate changes in NE modulation of evoked activity. These data demonstrate that prolonged stress alters the way in which NE affects neuronal activity in target regions. We suggest BLA neurons become hyperexcitable, and this pathology may underlie some of the behavioral deficits and symptoms associated with exposure to chronic stress.

Oxidative stress induced by glutathione depletion in the mouse HT22 neuronal cell line and embryonic rat immature cortical neurons causes a delayed, sustained activation of extracellular signal-regulated kinases-1/2 (ERK1/2), which results in cell death. Previous studies from our lab have shown that this sustained activation of ERK1/2 is mediated primarily by a selective, reversible inhibition of ERK1/2-directed phosphatases. However, the mechanisms underlying the inhibition remain unclear. Results from this thesis demonstrate that the inhibition of ERK1/2 phosphatases in HT22 cells and immature neurons is a consequence of oxidative stress induced by glutathione depletion as phosphatase activity is restored in cells treated with the antioxidant BHA. This agent leads to reduced ERK1/2 activation and neuroprotection. However, we also show that an increase in free intracellular Zn2+ that accompanies glutathione depletion-induced oxidative stress in HT22 cells and immature neurons contributes to selective inhibition of ERK1/2 phosphatase activity and consequently ERK1/2 activation and cell death. ERK1/2 also functions to maintain elevated levels of Zn2+. Thus the elevation of intracellular Zn2+ within neurons subjected to oxidative stress can trigger a robust positive feedback loop operating through activated ERK1/2 that rapidly sets into motion a Zn2+ -dependent pathway of cell death. Previous data from our lab have suggested that PP2A, a Serine/Threonine protein phosphatase, is the predominant ERK1/2 phosphatase in primary neurons. I have confirmed this result by using specific peptide inhibitor of PP2A. Furthermore, I have revealed for the first time the reversible cysteine oxidation in the catalytic subunit of PP2A, suggesting that PP2A may be the phosphatase that is susceptible to oxidation and inactivation following glutamate treatment in HT22 cells and primary neurons. ERK1/2 activation contributes to neuronal death following focal ischemia/reperfusion. In the third part of my thesis, I investigated the mechanisms responsible for ERK1/2 activation following ischemia/reperfusion. I have demonstrated that the selective inhibition of ERK1/2 phosphatases contributes to ERK1/2 activation following ischemia/reperfusion in both focal and global ischemia models. Altogether, these results implicate the inhibition of ERK1/2 phosphatases as an important mechanism for ERK1/2 activation in oxidatively stressed neurons.

Parkinson's disease (PD) neurodegeneration of the dopaminergic cells of the substantia nigra has been linked to various types of cellular injury, including oxidative stress, mitochondrial dysfunction, and dysfunction of the ubiquitin proteasome system. Multiple genetic mutations and high prevalence of idiopathic disease conceals the unifying mechanism for PD. Given the selective vulnerability of dopaminergic cells, dopamine (DA) may play a major role in PD pathogenesis. DA metabolism and oxidation into DA quinone (DAQ) leads to the production of reactive oxygen species. In addition, DAQ can react with reduced sulfhydryls, covalently modifying cysteine residues. DAQ modification of free cysteines, glutathione, and cysteines in proteins could result in decreased antioxidant capacity and inactivation and/or misfolding of proteins. In this study, I measured the effect of DA treatment in differentiated PC12 cells, and found that DA exposure was toxic, lead to increased DAQ modified free cysteines, glutathione, and cysteines in proteins, and decreased ATP levels. I also demonstrated that metabolism of DA by monoamine oxidase did not influence DA-induced toxicity, but that DA uptake by the dopamine transporter was necessary for DA-induced cell death. Further, I demonstrated that activation of endoplasmic stress (ER) also occurred following DA exposure, with increases in ER chaperone proteins calreticulin, ERp29, ERp99, Grp58, Grp78, Grp94, and Orp150. Decreased mitochondrial levels of the glycolytic enzyme aldolase A and increased levels of whole cell aldolase A were also observed following DA exposure, suggesting that DA may affect ATP levels by altering energy-related proteins. Finally, to determine the role of DA oxidation in the rotenone model of PD, I used the mitochondrial complex I inhibitor in DA depleted cells. I found no protection with DA depletion, but significant increases in rotenone toxicity when co-treated with methamphetamine, which leads to the cytoplasmic release of DA. Since I used sub-toxic levels of methamphetamine, this data suggests that the increased levels of DA oxidation lead to potentiation of rotenone-induced toxicity. Therefore, in this thesis I show that DA oxidation is linked to oxidative stress, ER stress activation, and mitochondrial dysfunction, and thus may play a role in the pathogenesis of PD.

Detection of blood volume deficits alters a rats motivational state by stimulating thirst and salt appetite. In consequence, PEG-treated rats with established hypovolemia adaptively drink both water and hypertonic NaCl solution; indeed, they quickly alternate between drinking both fluids and concoct a mixture isotonic to body fluids a concentration that is most effective in repairing plasma volume deficits without perturbing pOsm. However, their plasma volume deficits cannot be restored until ingested fluid is absorbed from the GI tract. The present experiment sought to address the issue of whether ingested water accelerates ingestion, gastric emptying, and small intestinal absorption of 0.30 M NaCl. In fact, ingestion of both water and 0.30 M NaCl did accelerate fluid delivery into the systemic circulation. Moreover, as a consequence of fluid leaving the GI tract more quickly, GI distension signals associated with inhibition of fluid intake are quickly removed, leading to larger fluid intakes. The unique behavior of PEG-treated rats corresponds to restoration of their body fluid deficits, including behavioral, physiological, and hormonal aspects of body fluid homeostasis. Clearly, co-existence of thirst and salt appetite is an adaptive behavioral response to hypovolemia.

Schizophrenia is a complex and devastating psychiatric disorder that creates a substantial emotional and economic burden on individuals with the illness, their families, and society. Understanding the causes and identifying the molecular alterations in the brain that underlie the pathophysiology of core clinical features of schizophrenia are central to the development of new therapeutic interventions. In particular, schizophrenia is characterized by impairments in working memory, which are thought to result from a deficit in GABA neurotransmission in the dorsolateral prefrontal cortex (DLPFC). Interestingly, exposure to cannabis has been associated with an increased risk for developing schizophrenia and cannabis use is associated with DLPFC-related working memory impairments similar to those observed in schizophrenia. The effects of cannabis are mediated by the brain cannabinoid 1 (CB1) receptor, which in the rodent, is heavily localized to certain inhibitory axon terminals and, when activated, inhibits GABA release. Here, we have investigated the anatomical distribution of the CB1 receptor in the primate brain and characterized the cellular localization and synaptic targets of the CB1 receptor in the primate DLPFC. In addition, we explored the potential relationship between CB1 receptor signaling and altered GABA neurotransmission in schizophrenia by evaluating CB1 receptor mRNA and protein expression in the DLPFC of subjects with schizophrenia. We found that CB1 receptors are highly expressed in the primate DLPFC and that CB1 receptors are localized in the terminals of the subtype of perisomatic-targeting GABA interneurons that contain the neuropeptide cholecystokinin (CCK). We found that CB1 mRNA and protein are reduced in schizophrenia, which may represent a compensatory mechanism to increase GABA transmission from perisomatic-targeting CCK neurons with impaired GABA synthesis. We conclude that reductions in the expression of the CB1 receptor mRNA and protein in CCK neurons represent a novel neuropathological entity in the DLPFC of individuals with schizophrenia. These findings suggest a novel drug target for the treatment of cognitive dysfunction in schizophrenia.

Calcium influx through presynaptic calcium channels triggers transmitter release, but many of the details that underlie calcium-triggered secretion are not well understood. In an attempt to increase our understanding of this process, synaptic transmission at the frog neuromuscular junction has been investigated using physiological experiments and computational modeling. Pharmacological manipulations ((R)roscovitine and DAP) were used as tools to modulate presynaptic calcium influx and study effects on transmitter release. I showed that (R)-roscovitine predominately slowed deactivation kinetics of calcium current (by 427%), and as a result, increased the integral of calcium channel current evoked by a physiological action potential waveform (by 44%). (R)-roscovitine also increased the quantal content of acetylcholine released from the motor nerve terminals (by 149%) without changing paired-pulse facilitation under low calcium conditions. In contrast, exposure to 3,4-diaminopyridine (which affects transmitter release evoked by partially blocking potassium channels, altering the amplitude of the presynaptic action potential, and indirectly increasing calcium entry) increased paired-pulse facilitation (by 23%). In normal calcium conditions, both pharmacological treatments showed relatively similar effects on paired-pulse facilitation. I used a computational model, constrained by previous reports in the literature and my physiological measurements, to simulate my experimental data. This model faithfully reproduced calcium current with a single action potential, the average number of released synaptic vesicles, and the effects of (R)-roscovitine and DAP on calcium influx and vesicle release. Using this model, I made several predictions about the mechanisms underlying transmitter release. First, calcium ions originating from one or two voltage-gated calcium channels most often contributed to cause the fusion of each vesicle. Second, the calcium channel closest to a vesicle that fuses, provides 77% of calcium ions. My simulation of paired-pulse facilitation using the present model needed more adjustments, and in the process of adjusting the model parameters, various hypotheses that might explain observed short-term synaptic plasticity, including the effects of changes in buffer conditions, the effects of uneven calcium channel distribution, reducing terminal volume by adding vesicles to a storage pool, changes in the second action potential waveform, and possible persistent changes in vesicle release machinery were explored.

Complex motor behaviors enable mammals to adapt to their internal and external environments. The lateral hypothalamic area (LHA) contributes importantly to autonomic and endocrine regulation, behavioral states, and energy balance. Orexin (Orx) neuropeptides, produced exclusively by LHA cells, are crucial in the integration of sleep and arousal. The LHA projects densely to the ventral tegmental area (VTA), a dopamine (DA) region that is essential for modulating goal-directed behaviors. Extensive investigations of reward function implicate the LHA-VTA connectivity, an arrangement not yet characterized in detail at the ultrastructural level. The present research sought to clarify the precise interactions of LHA axons with VTA cells. Considering reported physiological responses of VTA cells to LHA stimulation and Orx actions, we hypothesized that both projections interact heavily with DA and GABA cell groups in the VTA, and that LHA axons provide a predominant inhibitory innervation. We used immunocytochemistry to visualize DA or GABA neurons in combination with 1) tract tracer identification of LHA axons or 2) immunolabeling for Orx. Electron microscopic analysis of the VTA revealed that, while the bulk of LHA and Orx projections pass through the VTA, their connections with DA and GABA neurons are a complementary mixture of excitatory and inhibitory synapses. The details of morphology herein suggest many different mechanisms of signal transmission by which LHA axons might contribute information concerning interoceptive state to the adaptive performance of complex motor behaviors modulated by the VTA.