acase@tulane.edu / Major Depressive Disorder is a pervasive and devastating neuropsychiatric disease. One in three patients receive no therapeutic benefit from standard antidepressant treatments, and those who do respond do so only after chronic treatment for weeks to months. A single low dose of ketamine, an NMDA receptor antagonist, results in a rapid and prolonged antidepressant effect in as many as 90 percent of treatment-resistant patients. In this thesis we elucidate some mechanisms through which ketamine functions. Here we demonstrate in mice that ketamine treatment initiates mTOR-dependent protein synthesis, increases expression of synaptic proteins, increases excitatory synaptic drive in mPFC, and elicits an antidepressant-like response. Furthermore, genetic deletion of NMDAR subunit GluN2B from cortical pyramidal neurons is sufficient to mimic and occlude ketamine’s effects on protein synthesis, excitatory synaptic function, and depression-related behaviors. We demonstrate that GluN2B-containing NMDARs are uniquely tonically activated by ambient glutamate in mPFC, and that altering levels of ambient glutamate bidirectional alters excitatory synaptic drive in mPFC and depression-related behaviors. Next we demonstrate that viral deletion of GluN2B from pyramidal neurons in mPFC enhances excitatory synaptic drive onto these cells in an input-specific manner and is sufficient to elicit an antidepressant-like response. Finally, using genome-wide profiling of translating mRNAs, we observe that antidepressant-dose ketamine enhances the expression of gene-sets involved in translational processes, metabolism, GPCR signaling, and angiogenesis, among others. Individual genes were analyzed for a potential role in ketamine’s antidepressant-like response, and we identified one GPCR, VPAC2, as being a candidate target. This thesis contributes important steps towards a mechanistic understanding of ketamine’s effects as an antidepressant, and reveals novel targets for depression therapy. / 1 / Oliver H Miller

Indiana University-Purdue University Indianapolis (IUPUI) / Excitotoxicity, a major hallmark of neurodegeneration associated with cerebral ischemia, is a result of accumulation of extracellular glutamate. This excess glutamate leads to hyperactivation of glutamate receptors such as the N-methyl-D-asparate (NMDA) receptors (NMDARs) following the activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor (AMPARs). Excessive activation of NMDARs causes an influx of calcium, which can eventually activate apoptotic pathways and lead to death of neurons. Regulation of NMDAR subunit composition, localization, surface expression, and activity can balance cell survival via activation of either pro-death or pro-survival pathways after a course of an ischemic insult. Specifically, phosphorylation of different NMDAR subunits defines their activity and downstream signaling pathways. NMDARs are phosphorylated by multiple kinases and dephosphorylated by different phosphatases. Besides phosphatases and kinases, per se, phosphorylation of synaptic proteins that regulate kinase or phosphatase targeting and activity also mediate NMDAR phosphorylation. Spinophilin, a major synaptic scaffolding and protein phosphatase 1 (PP1) targeting protein, mediates substrate phosphorylation via its ability to bind PP1. Our studies focus on delineating the role of spinophilin in the regulation of phosphorylation and function of the GluN2B subunit of the NMDA receptor as well as the role of spinophilin in modulating glutamate-induced neurotoxicity. Interestingly, our data demonstrate that spinophilin sequesters PP1 away from GluN2B thereby enhancing phosphorylation of GluN2B at Ser-1284. These changes impact GluN2B protein interactions, subcellular localization, and surface expression, leading to alterations in the amount of calcium entering the neuron via GluN2B-containing NMDARs. Our data show that spinophilin biphasically regulates GluN2B function. Specifically, Ser-1284 phosphorylation enhances calcium influx through GluN2B containing NMDA receptors, but spinophilin leads to dramatic decreases in the surface expression of the receptor independent of Ser-1284 phosphorylation. Moreover, in spinophilin knockout mice, we observe less PP1 binding to GluN2B and less phosphorylation of Ser-1284, but more surface expression of GluN2B and greater levels of caspase activity. Together, these observations suggest a potential neuroprotective role for spinophilin by decreasing GluN2B-containing NMDA receptor-dependent surface expression and thereby decreasing intracellular calcium and neuronal cell death.

NMDARs are ligand-gated cation channels which are activated by the neurotransmitter glutamate. NMDARs are essential in coupling electrical activity to biochemical signalling as a consequence of their high Ca2+ permeability. This Ca2+ influx acts as a secondary messenger to mediate neurodevelopment, synaptic plasticity, neuroprotection and neurodegeneration. The biological outcome of NMDAR activation is determined by a complicated interrelationship between the concentration of Ca2+ influx, NMDAR location (synaptic vs. extrasynaptic) as well as the subtype of the GluN2 subunit. Despite the recognition that NMDAR mediated physiology is multifaceted, tools used to study subunit and location dependent signalling are poorly characterized and in other cases, non-existent. Therefore, the aim of this thesis is to address this issue. Firstly, I assessed the current pharmacological approach used to selectively activate extrasynaptic NMDARs. Here, synaptic NMDARs are first blocked with MK-801 during phasic activation and then extrasynaptic NMDARs are tonically activated. This approach relies on the continual irreversible blockade of synaptic NMDARs by MK-801 yet contrary to the current dogma, I demonstrate this blockade is unstable during tonic agonist exposure and even more so when physiologically relevant concentrations of Mg2+ are present. This confines a temporal limit in which selective activation of extrasynaptic NMDARs can occur with significant consequences for studying synaptic vs. extrasynaptic NMDAR signalling. Dissecting subunit-dependent signalling mediated by the two major GluN2 subunits in the forebrain, GluN2A and GluN2B, has been advanced significantly by selective GluN2B antagonism yet a reciprocal GluN2A selective antagonist has been lacking. Utilizing novel GluN2A-specific antagonists, I demonstrate a developmental upregulation of GluN2A-mediated NMDA currents which concurrently dilutes the contribution of GluN2B-mediated currents. Moreover, I tested the hypothesis that the Cterminus of GluN2A and GluN2B are essential in controlling the developmental switch of GluN2 subunits utilizing knock-in mice whereby the C-terminus of GluN2A is replaced with that of GluN2B. Surprisingly, the exchange of the C-terminus does not impede the developmental switch in subunits nor the proportion of NMDARs at synaptic vs extrasynaptic sites. However, replacing the C-terminus of GluN2A with that of GluN2B induces a greater neuronal vulnerability to NMDA-dependent excitotoxicity. Collectively, this work enhances our understanding of the complex physiology mediated by the NMDAR by determining how pharmacological tools are best utilized to study the roles of NMDAR location and subunit composition in addition to revealing the importance of the GluN2 C-terminus in development and excitotoxicity.

Excitotoxicity, a major hallmark of neurodegeneration associated with cerebral ischemia, is a result of accumulation of extracellular glutamate. This excess glutamate leads to hyperactivation of glutamate receptors such as the N-methyl-D-asparate (NMDA) receptors (NMDARs) following the activation of α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor (AMPARs). Excessive activation of NMDARs causes an influx of calcium, which can eventually activate apoptotic pathways and lead to death of neurons. Regulation of NMDAR subunit composition, localization, surface expression, and activity can balance cell survival via activation of either pro-death or pro-survival pathways after a course of an ischemic insult. Specifically, phosphorylation of different NMDAR subunits defines their activity and downstream signaling pathways. NMDARs are phosphorylated by multiple kinases and dephosphorylated by different phosphatases. Besides phosphatases and kinases, per se, phosphorylation of synaptic proteins that regulate kinase or phosphatase targeting and activity also mediate NMDAR phosphorylation. Spinophilin, a major synaptic scaffolding and protein phosphatase 1 (PP1) targeting protein, mediates substrate phosphorylation via its ability to bind PP1. Our studies focus on delineating the role of spinophilin in the regulation of phosphorylation and function of the GluN2B subunit of the NMDA receptor as well as the role of spinophilin in modulating glutamate-induced neurotoxicity. Interestingly, our data demonstrate that spinophilin sequesters PP1 away from GluN2B thereby enhancing phosphorylation of GluN2B at Ser-1284. These changes impact GluN2B protein interactions, subcellular localization, and surface expression, leading to alterations in the amount of calcium entering the neuron via GluN2B-containing NMDARs. Our data show that spinophilin biphasically regulates GluN2B function. Specifically, Ser-1284 phosphorylation enhances calcium influx through GluN2B containing NMDA receptors, but spinophilin leads to dramatic decreases in the surface expression of the receptor independent of Ser-1284 phosphorylation. Moreover, in spinophilin knockout mice, we observe less PP1 binding to GluN2B and less phosphorylation of Ser-1284, but more surface expression of GluN2B and greater levels of caspase activity. Together, these observations suggest a potential neuroprotective role for spinophilin by decreasing GluN2B-containing NMDA receptor-dependent surface expression and thereby decreasing intracellular calcium and neuronal cell death.

Dopamine (DA) neurons of the substantia nigra pars compacta (SNc) have a key role in regulation of voluntary movement control. Their death is a hallmark of Parkinson’s disease, characterised by inhibited motor control, including muscle rigidity and tremor. Excitatory input to SNc-DA neurons is primarily from the subthalamic nucleus, and in PD these afferents display a higher frequency firing, as well as increased burst firing, which could cause increased excitatory activity in SNc-DA neurons. NMDA receptors (NMDARs) bind the excitatory neurotransmitter glutamate, and are essential for learning and memory. In SNc-DA neurons, NMDARs have a putative triheteromeric subunit arrangement of GluN1 plus GluN2B and/or GluN2D. Wild type (WT) mice, and those lacking the gene for GluN2D (Grin2D-null), were used to explore its role in various aspects of DA neuronal function and dysfunction using patch-clamp electrophysiology, viability assaying, and immunofluorescence. Pharmacological intervention using subunit-specific inhibitors ifenprodil and DQP-1105 on elicited NMDAR-EPSCs suggested a developmental shift from primarily GluN2B to GluN2B/D. Activity dependent regulation was assessed by high frequency burst stimulation of glutamatergic afferents: in comparison to controls, significant downregulation of NMDARs was observed in SNc-DA neurons, though no differences were observed based on genotype. This regulatory function may be a neuroprotective or homeostatic response. Ambient extracellular glutamate elicits tonic NMDAR activity in SNc-DA neurons, which may be important for maintaining basal levels of excitability: the role of GluN2D was assessed by recording the deflection in baseline current caused by application of competitive NMDAR antagonist D-AP5. There was a significantly larger NMDAR-mediated current in WT vs Grin2D-null mice, indicating that GluN2D has a role in binding ambient glutamate. Dysfunction of glutamate uptake could be a secondary pathophysiological occurrence in the SNc, leading to increased ambient glutamate: the effect of this was explored by application of the competitive glutamate transporter blocker TBOA. Here, the NMDAR-mediated portion of this current was significantly higher in WT mice in comparison to Grin2D-null. Interestingly, dose-response data obtained from bath application of NMDA showed significantly larger currents in Grin2D-null animals vs WT, but only at the top of the response curve (~1-10 mM), which may indicate a capability for larger conductance in Grin2D-null animals at high NMDAR saturation due to replacement of GluN2D with GluN2B. GluN2D may therefore be neuroprotective, by attenuating peak current flow in response to very high agonist concentrations. Lastly, GluN2D has been found to decrease NMDAR open probability under hypoxic conditions, potentially conferring resistance to hypoxia / ischemia related excitotoxicity. Therefore, low (15% O2 / 80% N2 / 5% CO2) vs high (95% O2 / 5% CO2) oxygen conditions were used along with immunofluorescent propidium iodide cell death assaying and immunofluorescent labeling for DA neurons in order to compare levels of DA neuronal death in the SNc based on oxygen status and genotype. Whilst there was a significant submaximal effect based on O2 status, genotype did not confer a practical resistance under these conditions. In summary, NMDARs have diverse roles in SNc-DA neurons which may both serve to maintain normal function and protect the cell against potentially pathological conditions.

The N-methyl-D-aspartate receptor (NMDAR) hypofunctional hypothesis is believed to explain one of the contributing factors to schizophrenia. This hypothesis suggests the dysregulation of NMDAR, a protein responsible for receiving signals from the synapses between neurons, is the cause of some of the symptoms seen in schizophrenia. The post synaptic density protein 95 (PSD95) uses its PDZ-domains to help facilitate the received signal from NMDAR to α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) which in turn transmits the signal through the neuron. One way to increase the function of NMDAR could be to increase its affinity towards PDZ-domains of PSD95 using a small molecule. Fragment based drug design (FBDD) is one way to screen for molecules that modulates the NMDAR-PDZ interaction. This work describes the development of differential scanning fluorimetry (DSF) and surface plasmon resonance (SPR) assays using a fusion protein to screen for molecules that potentiate the interaction between NMDAR and AMPAR as well as methods assisting in the prioritization of hits based on both affinity, selectivity, and mechanism. The developed assays were used to screen a library containing 768 compounds. Screen positives and other compounds of interest were triaged and evaluated based on affinity, selectivity, and ability to modulated peptide binding resulting in eight confirmed hits that interacts with the two PDZ-domains of PSD95 investigated. As part of this work, the dissociation constant (KD) was determined for a panel of peptides representing versions of the truncated NMDAR GluN2b-subunit C-terminal towards PDZ1 and 2 of PSD95.