Antonello Bonci, M.D. - US grants

DESCRIPTION (provided by applicant): The goal of the present proposal is to understand the cellular and synaptic events through which cocaine induces long-term plasticity at excitatory synapses in the ventral tegmental area (VTA) after a single in vivo exposure. The VTA considered to play a central role in the initiation of drug-related behaviors such as behavioral sensitization, but the cellular mechanisms that underlie the initiation of such behavioral phenomena are still obscure. Our preliminary data suggest that a single injection of cocaine produces a long-term potentiation (LTP) of AMPA-receptor-mediated post-synaptic currents in the VTA. Furthermore, when mice were challenged with a second injection of cocaine, they displayed context-dependent behavioral sensitization. Further, both cocaine-induced LTP and context-dependent behavioral sensitization are blocked by the NMDA receptor antagonist MK-801. Our first goal is to determine whether the expression of cocaine-induced LTP involves a change in AMPA receptor number, function, or both. Second, we will perform intra-VTA in vivo injections to understand whether intra-VTA injections of cocaine are sufficient for cocaine to induce LTP, and to define the receptors and second messengers involved in producing the cocaine-induced LTP. Third, we will use an in vitro model to study which receptors in the VTA are involved in producing the cocaine-mediated long-term potentiation. Fourth, we will determine whether such cocaine dependent LTP also occurs in tertiary cells, a group of VTA neurons whose function is still unknown. Our long-term goal is to understand the complete sequence of cellular and molecular events through which cocaine produces LTP at AMPA receptors in the VTA and the subsequent initiation of behavioral sensitization.

DESCRIPTION (provided by applicant): Stress increases addictive behaviors such as behavioral sensitization. Although it has been well documented that the ventral tegmental area (VTA) is the brain region whose activation is essential to induce behavioral sensitization, the mechanism through which stress increases behavioral sensitization is unknown. The main goal of this proposal is to elucidate the mechanism through which the corticotropin releasing factor (CRF) mediates NMDA receptor function and behavioral sensitization. Our preliminary data provide the first evidence for a direct synaptic modulation by CRF of synaptic transmission on a subclass of VTA dopamine (DA) neurons. Specifically, our data show that CRF activates CRF 2 receptors (CRF2R), which in combination with the CRF binding-protein (CRF-BP) increases NMDA currents in the VTA. Further, our data show that the CRF-dependent increase of NMDA currents in the VTA depends on the activation of phospholipase C (PLC) and PKC. We hypothesize that the potentiation of NMDA currents by CRF and the CRF-BP represents a key cellular phenomenon underlying stress-induced sensitization to cocaine. In specific aim 1, we want to characterize in detail the anatomical projections of the subset of DA neurons that are sensitive to CRF. Specific aim 2 will define the intracellular pathway activated by PLC that is responsible for the CRF-dependent increase of NMDA currents. Finally, specific aim 3 will study the role of CRF and of CRF2R agonists and antagonists in modulating context-dependent behavioral sensitization and stress-dependent behavioral sensitization to cocaine.

Dopamine neurons in the VTA play a very important role in a variety of physiological as well as addictive behaviors. The main goal of the present proposal is to elucidate the relationship between plasticity at excitatory synapses in the ventral tegmental area (VTA)and addictive behaviors such as behavioral sensitization and self-administration of cocaine. During the current funding period (April1st, 2002- February 2006), we have collected evidence that might explain the sequence of events leading from NMDAR activation in the VTA produced by acute cocaine application, to long-term potentiation of VTA neurons that results as a consequence of in vivo cocaine exposure. Further, our preliminary data suggest that long-term changes of excitatory synaptic transmission in the VTA are not only produced by passive cocaine administration (e.g.in vivo cocaine injections), but operant behaviors such as cocaine self-administration. Specific aim 1 will test the hypothesis that in vivo cocaine-induced potentiation involves a direct action of cocaine in the VTA mediated by NMDARs, D5 receptors and the cAMP/PKA-dependent pathway. Specific aim 2 will test the role and time-course of protein synthesis in mediating long-term plasticity at VTA synapses and behavioral sensitization. Finally, specific aim 3 will characterize whether long-term synaptic changes at glutamatergic synapses in the VTA are produced during forced abstinence or extinction of operant responding from either food or cocaine. Taken together, the results from these experiments will likely help us understand the role of plasticity at glutamatergic synapses in the VTA in mediating cocaine-dependent behaviors.

DESCRIPTION (provided by applicant): Stress increases addictive behaviors. Corticotropin-releasing factor (CRF) is released in the ventral tegmental area (VTA) during stressful events, and produces relapse to cocaine seeking. However, the mechanism by which CRF produces stress-dependent relapse to cocaine seeking is poorly understood. The main goal of this proposal is to understand the role of CRF-R1 and CRF-R2, and the CRF-BP in the VTA in modulating dopamine release and stress-induced relapse to cocaine seeking. Over the past four years, my laboratory has collected evidence showing that CRF activates CRF-R2 to increase NMDAR-mediated currents in VTA DA neurons. Furthermore, we have evidence that CRF-R1 activation in VTA DA neurons increases firing activity via activation of Ih. By performing patch-clamp recordings in the VTA, specific aim 1 will elucidate in detail: a) the intracellular pathway responsible for the CRFR1-dependent increase in firing rate, and b) the intracellular pathways responsible for the CRF-R2-dependent increase in NMDAR currents in the VTA. Specific aim 2 will determine the role of CRF-R1 and CRF-R2 in modulating DA release in the ventral striatum. Specific aim 3 will determine the role of CRF-R1- and CRF-R2-dependent pathways in inhibiting footshock- induced relapse to cocaine seeking. Finally, specific aim 4 will take a deep mechanistic look at the CRF-BP. The results from this grant will produced a deep mechanistic and behavioral understanding of the various effects of CRF on VTA neurons. Our results will likely create new therapeutic leads toward agents that disrupt the CRF-R1- and CRF- R2-dependent effects on VTA neurons and thus stress-induced cocaine seeking. PUBLIC HEALTH RELEVANCE: Stress increases addictive behavior. However, the mechanism by which stress-released molecules exert their negative effects on drug-seeking are poorly understood. The main goal of this project is to elucidate the role of CRFR1 and CRFR2 in promoting stress-enhanced relapse to cocaine seeking. Relapse to drug seeking is a major health problem that still has no cure. The results from this proposal could enable us to create new therapeutic targets aimed at inhibiting the ability of stressful event to increase relapse to substance abuse.

My future research will build on these findings to explore possible involvement of VTA non-DA efferents in updating relationship between stimulus and conditioned response. The goals are to determine which putative targets innervated by VTA non-DA neurons are critically required in discrimination learning, whether chronic use of substance of abuse will dampen discrimination learning and impact selective VTA non-DA efferents and develop an optical intervention technique to rescue learning deficits. Beside my search, I also participate in Dr. Dong Wangs project Median raphe nucleus regulates hippocampal ripple oscillation and memory consolidation. Hippocampal sharp-wave ripples have been the most studied neuronal oscillations and occur primarily during slow-wave sleep and have been shown to play an essential role underlying memory consolidation. A widely held view is that hippocampal ripple oscillations reactivate newly-formed memory traces for long-term storage. Although numerous studies have aimed at addressing the interaction between cortical activities and hippocampal ripple oscillations, causal relationship remains poorly defined. Moreover, little research attention is paid to address whether hippocampal ripple oscillation is subjected to sub-cortical regulation. Here, by combining optogenetics and in vivo simultaneous electrophysiology recordings in freely-moving mice, our study uncovered a novel, unexpected regulatory role of subcortical median raphe nucleus to hippocampal ripple oscillation as well as memory consolidation. Specifically, we first observe ripple-related neuronal activities in median raphe nucleus. To investigate causal relationship, we present evidence that optical stimulation of median raphe suppresses hippocampal ripple activities. Finally, given the causal role of hippocampal ripple oscillation in memory consolidation, we show that optical stimulation of median raphe during sleep impairs hippocampus-dependent fear memory. These data should be of great interest to a wide variety of scientists, as it not only sheds light on the first causal regulatory link from subcortical median raphe toward hippocampal ripple oscillation, but also attribute median raphe the novel function as gatekeeper for ripples and memory consolidation to occur. This work was submitted to Nature and is under revision.

Current and future research Prelimbic stimulation in addicted rats decreases compulsive cocaine-seeking behavior. A hallmark of addiction is continued use of drugs despite negative consequences. This compulsive drug-seeking behavior is thought to result, in part, from drug-induced hypofunction in cortical regions, including the PFC. Because the PFC has critical function in exerting control over unwanted behaviors2, dysfunctions in this area may render an addict incapable of resisting the urge to seek drugs. Consistent with this hypothesis, my on-going research has uncovered significant decreases in neuronal activity in the PL in rats with a long history of cocaine self-administration. Moreover, in Addicted rats, photo-stimulation of channelrhodopsin (ChR2)-expressing PL neurons during self-administration sessions significantly reduced cocaine-seeking behaviors, but only after the rats have learned that lever pressing for cocaine is associated with foot shocks. Prior to the pairing of foot shocks with cocaine, photo-stimulation of ChR2-expressing PL neurons has no effect on cocaine-seeking behavior. The PL has been shown to guide goal-directed behaviors and in the absence of any punishment, the goal of the rat is to seek cocaine. Therefore the rat does not need to evaluate the outcome of the decision and activation of the PL has no effect. However, when future cocaine rewards are paired with punishments, a rat is faced with two conflicting goals, to get cocaine or to avoid shocks. In these situations, the rat must evaluate the value of cocaine against foot shocks. My data suggests that in the face of foot shocks, decreased PL activity biases the rats behavior towards continued cocaine seeking. My future research program will examine two major PL circuits that may be altered by long-term voluntary consumption of cocaine to allow the execution of addictive behaviors. I will focus on the PL-BLA and the PL-NAc circuit. The PL-BLA and the PL-NAc projections are implicated in promoting goal-directed and drug-seeking behaviors3,7, but their role in regulating compulsive behaviors remains poorly understood. To examine the effects of cocaine-induced neuroadaptations in these two PL output pathways and identify their function in compulsive cocaine seeking, the following experiments will be performed: In vivo and in vitro patch-clamp recording in neurons from Addicted and Non-addicted rats. To identify cocaine-induced changes in the function of PL circuits and the underlying cellular mechanisms, in vivo and in vitro whole-cell patch clamp recordings will be made in Addicted and Non-addicted rats. Of particular interests are adaptations present in Addicted but not Non-addicted rats, which may uniquely permit compulsive cocaine seeking. In vivo patch-clamp recordings will be performed to examine the intrinsic electrophysiological properties of PL neurons in the context of a fully connected network. In addition, because the loss of inhibitory control in Addicted rats may reflect impaired PL responses to punishment, synaptic responses of PL neurons to foot-shocks will also be examined. Lastly, in vivo recordings will be made at various stages of self-administration training. This will provide a longitudinal analysis of cocaine-induced neuroadaptations at various stages of the addiction process. Changes in cocaine-induced network function can be mediated by molecular changes at the cellular and synaptic level. Among the possible neuroadaptations include function and or number of receptors and ion channels as well as release probability of transmitters (i.e. glutamate and GABA). To precisely analyze these parameters in Addicted and Non-addicted rats, in vitro patch-clamp recording in brain slices containing PL, BLA, and NAc will be made. In addition, I will use ChR2 to selectively examine changes in the excitatory strengths of projections from the PL- to BLA or NAc in the two groups. Together, in vivo and in vitro electrophysiological experiments will provide a thorough analysis of cocaine-induced changes at both the circuit and the cellular level. The identification of differences between Addicted and Non-addicted rats may reveal critical adaptations that facilitate compulsive, cocaine-seeking behavior. Optogenetic control of PL circuits to decrease compulsive cocaine seeking. I show that hypofunction of the PL has a critical role in permitting compulsive cocaine-seeking behaviors and that this behavior can be reversed with photo-activation of PL neurons. To understand the function of PL-BLA and PL-NAc circuits in compulsive cocaine seeking, I will use optogenetics to investigate afferent-specific excitatory synaptic function in the BLA or NAc in exerting inhibitory control over cocaine seeking. I will transduce PL projection neurons with adeno-associated virus encoding either ChR2 or halorhodopsin (NpHR) to stimulate or inhibit, respectively, afferent-specific neural activity in the BLA or NAc. Results from in vitro slice experiments will guide in vivo optogenetic experiments. For example, if the strength of a particular PL circuit is significantly reduced in the Addicted rats, ChR2 will be employed to selectively enhance these glutamatergic projections during cocaine seeking. In contrast, if an increase in synaptic strength in one or both of these circuits is observed, NpHR will be used to decrease its activity during cocaine seeking. Real-time modulation of neuronal circuits with optogenetics will establish a causal relationship between PL-circuits and its role in regulating compulsive behavior. Summary The loss of inhibitory control, despite knowing the negative consequences of this behavior, lies at the heart of compulsive drug use and addiction. My research seeks to identify cocaine-induced neuroadaptations, starting from individual synapses to progressing to specific neuronal circuits that contribute to decision-making impairments and allow maladaptive, compulsive cocaine seeking. The integration of rat addiction model with in vivo and in vitro electrophysiology and optogenetic targeting of specific cortical circuits provides the ideal platform to identify neural circuits and cellular mechanisms underlying addictive behaviors.

Sensory information can be used as clue for identifying objects, both rewarding objects and aversive objects. However, learning process would be required to use that information as clue. The association between sensory information and rewarding outcome would modify neural circuits and probably contribute to change value of the sensory cue. Although many brain areas are involving eating behavior, it is poorly known which neurons in the circuit mediate the ability to identify edible foods. One important region that may supply valence information is the insular cortex. The insular cortex is a brain region that integrates incoming sensory information including somatosensory and visceral information and is also known as the taste cortex. Furthermore, recent findings suggest that through the integration of information coming from visceral cues and sensory component of rewarding objects, insular cortex neurons can powerfully modulate appetitive behaviors. Therefore, I hypothesize that neural circuit that involve in cue-reward learning will also be made and refined in the insular cortex during cue-reward association learning training. Major purpose of this research is to confirm whether synaptic inputs into insular cortex neurons would be potentiated during cue-reward association learning training. To get more precise anatomical information about neuronal connection in the insular cortex, observation of afferent inputs into insular cortex by labeling neurons retrogradely using tracer called Fluorogold was performed. Pavlovian conditioning task was used as a model behavior of cue-reward learning. Mice learned association between sensory cue and food reward within one week. To confirm insular cortex is involved to this behavior, pharmacological experiments were performed. Electrophysiological approach was used to measure synaptic inputs into insular cortex neurons.

The sigma-1 receptor (Sig-1R), an endoplasmic reticulum (ER) chaperone protein, is an interorganelle signaling modulator that potentially plays a role in drug-seeking behaviors. However, the brain site of action and underlying cellular mechanisms remain unidentified. We found that cocaine exposure triggers a Sig-1R-dependent upregulation of D-type K+ current in the nucleus accumbens (NAc) that results in neuronal hypoactivity and thereby enhances behavioral cocaine response. Combining ex vivo and in vitro studies, we demonstrated that this neuroadaptation is caused by a persistent protein-protein association between Sig-1Rs and Kv1.2 channels, a phenomenon that is associated to a redistribution of both proteins from intracellular compartments to the plasma membrane. In conclusion, the dynamic Sig-1R-Kv1.2 complex represents a mechanism that shapes neuronal and behavioral response to cocaine. Functional consequences of Sig-1R binding to K+ channels may have implications for other chronic diseases where maladaptive intrinsic plasticity and Sig-1Rs are engaged.

My research in the Bonci lab generally all deals with studying inputs to the VTA. In the last year I have focused on the dorsal raphe nucleus as a VTA input, because this region is generally reported as providing the greatest density of inputs to the VTA compared with other afferent regions, and the function of these inputs remains largely uncharacterized. My first project, a collaboration with the lab of Dr. Marisela Morales, asked the question of whether serotonergic terminals projecting from the dorsal raphe nucleus to the VTA co-release glutamate along with serotonin. For this project I learned slice electrophysiology here at NIDA in the Bonci lab, and performed it in experimental mice. We used transgenic mice with cre recombinase expression restricted to serotonergic neurons to turn on expression of virally introduced Channelrhodopsin-2 protein in serotonergic cells. I did find that serotonergic terminals indeed co-release glutamate, and we are preparing the manuscript for submission to Neuron. A second project of mine involves studying non-serotonergic neurons of the dorsal raphe nucleus. It has been long-established that the dorsal raphe contains a greater number of non-serotonergic than serotonergic neurons. I have preliminary evidence that the non-serotonin neurons of the dorsal raphe project heavily to the VTA, release glutamate, and are highly rewarding in vivo. I am currently using a variety of techniques to characterize the role of these neurons in regulating the VTA and influencing behavior. I am working on this project in collaboration with the lab of Dr. Marisela Morales. In addition to the above research, I have contributed as a co-author to a protocol for Current Protocols in Neuroscience on using optogenetics in mice. I also performed experiments and am co-author on a project published in the Bonci lab studying inputs to the nucleus accumbens; this paper has currently been provisionally accepted with revisions.

Optogenetic control of behavior. Data collected from the slice experiments described above will direct our optogenetic experiments in awake-behaving mice. Using bilateral channel- and halorhodopsin stimulation in vivo, we are presently determining which behavioral tasks are influenced by the selective activation or inhibition of PFC inputs to the VTA. The selective activation of PFC inputs to the VTA neither influences locomotor behavior nor induces a conditioned place preference, but preliminary experiments suggest an impact of this stimulation on reversal learning. Research described below, in Project #2, extends this idea and examines the role of PFC activity in compulsive drug seeking. Overall, the data generated by Project #1 will elucidate the strength of specific neural circuits involving the VTA and NAc. Specific pathways will be examined in their basal state as well as following repeated exposure to cocaine. Drug-induced changes identified in slice experiments will direct our efforts in awake-behaving experiments, where we will use optogenetics to try and counteract drug-induced neural changes and, ultimately, compulsive drug seeking.

Data collected so far demonstrate that vHipp fibers are uniquely concentrated in the medial NAc shell. In this region vHipp input is predominant and selectively strengthened following cocaine injections. Bidirectional optogenetic manipulations in vivo have shown that vHipp input to the NAc drives cocaine-induced locomotion. The most notable finding thus far is that optical stimulations designed to offset the differential potency of each afferent pathway have proved that each excitatory input can reinforce instrumental behavior. These results suggest that the specific excitatory pathway activated is not as important as how much glutamate is released into the NAc, at least in terms of generating motivated behaviors. We have recently finished revising a manuscript detailing this work for the journal Neuron. We anticipate this work will be published in Neuron 2012.

In order to define the basic properties of microglia within the basal ganglia (BG) of the adult CNS, we used CX3CR1-EGFP transgenic mice to visualize microglia within the ventral tegmental area (VTA), nucleus accumbens (NAc), substantia nigra pars compacta (SNc), and substantia nigra pars reticulata (SNr). Previously we found that microglia populate the VTA at significantly lower density and exhibit sparse branching compared to other BG regions. In contrast, microglia within the SNr are present at a dramatically high density and both SNr and NAc microglia display highly-ramified morphologies. In the past year, we have expanded these analyses to show that there is no clear correlation between BG microglial density and neuronal density, as assessed by immunostaining for NeuN. However, similar analysis of overall cell density through DAPI staining showed that there is a consistent microglia-to-neuron ratio in the VTA, SNc, and NAc, suggesting that overall cell density may be one factor that regulates microglial cell numbers in these brain regions. In addition, we performed 3D whole-cell morphological reconstructions that demonstrate that NAc microglia have much greater process branching complexity and total process length compared to other BG microglia, which could suggest a heightened interaction with nearby synapses or other surrounding structures. We are currently performing analysis of glutamatergic synapse density within the basal ganglia through immunostaining for VGlut1, 2, and 3 to determine if microglial process branching complexity is related to the number of excitatory synapses within the surrounding tissue. To determine if these regional differences in microglial structure and distribution are accompanied by distinctions in functional status, we analyzed the abundance and localization of lysosomes within BG microglia through immunostaining for CD68. This analysis indicated that SNr microglia have elevated lysosome content compared to other BG microglia, suggesting that they differ from their counterparts in their phagocytotic or metabolic status. Differences in microglial functional status have been associated with shifts in their membrane properties. Previously, we used electrophysiological recording from microglial cells to show that microglia in the VTA and the adjacent SNr differed significantly in their membrane capacitance, resting potential, and the expression of voltage-gated potassium channels. We have expanded this analysis to include microglia in the SNc and have used pharmacology to more accurately demonstrate the identity of the potassium channels expressed by most SNr microglia. To complete our analysis of regional differences in microglial functional status, we have developed a workflow that allows whole transcriptome RNA sequencing of microglial cells isolated from distinct BG regions as well as the cortex (Ctx). Known microglial genes were expressed at high levels in all samples and genes expressed by neurons, astrocytes, and oligodendrocyte lineage cells were absent, confirming that pure populations of microglia had been isolated from each brain region. RT-PCR analysis of purinergic receptor expression from an independent cohort of animals showed the same pattern of expression as that observed by RNAseq, supporting the quality and reliability of the RNAseq dataset. While a core set of genes were expressed by microglia in all the analyzed brain regions, 34% of genes were expressed by microglia in only one brain region. In particular, VTA microglia exhibited the greatest number of uniquely expressed genes as well as the greatest number of genes that were expressed at significantly higher or lower levels than those observed in microglia from other brain regions. We are currently working to use independent techniques to validate some of the specific genes that appear to be uniquely expressed by microglia in specific BG nuclei. Together, these findings indicate that there are functional differences among microglia in different regions of the BG and raise questions about whether BG microglia will exhibit variable responses to pathological changes in BG circuit activity. Future experiments will use this RNA sequencing approach to determine how BG microglia and astrocytes are altered following both acute and chronic exposure to cocaine. To further define whether these cells influence the membrane properties and synaptic transmission of neurons within the BG, we are using transgenic strategies to ablate microglia within the CNS (CX3CR1-CreER;rfs-DTA mice). Recently, we have carried out important control experiments to demonstrate that this approach does in fact kill microglial cells. We used immunostaining for caspase3 in the early stages of ablation to show that this marker of programmed cell death can be detected within BG microglia. In addition, if microglia are allowed to repopulate the CNS following ablation and mice are treated with BrdU during this repopulation, numerous BrdU+ microglia are observed, indicating that the cells had been eliminated and were undergoing cell division to reinstate their numbers, as opposed to having temporarily downregulated key microglial cell markers. Experiments aimed at determining whether microglial elimination influences the membrane properties of BG neurons or drug-induced plasticity and behaviors are ongoing. Aging is a major risk factor for numerous neurodegenerative diseases. The changes that microglia undergo during normal aging and whether they influence the degenerative susceptibility of specific neuronal populations has not been well explored. To determine whether the regional differences in microglia phenotype that we observed in the young adult are preserved throughout life, we examined BG microglia in 18 month and 22-24 month old mice. Significant increases in microglial density were observed in the NAc, VTA, SNc, and SNr by 22 months of age. However, the magnitude of this increase was much greater in the VTA and significant increases in microglial density were already apparent in the VTA at 18 months of age. Regional differences in microglial process branching complexity are maintained at 18 months of age, although overall tissue coverage by microglial cell processes and somas is increased in all analyzed regions. Future analysis will address whether this increase in tissue coverage is due to increased soma size, increased soma number, increased process thickness, or increased process branching. Intracellular inclusions of undegradable proteins and lipids, known as lipofuscin, are observed in cells throughout the BG by 18 months of age. Preliminary analysis indicates that the majority of this material is accumulated in either neurons or microglia, and not other CNS cells. Future analysis will quantify whether there are regional differences in the extent of neuronal or microglial lipofuscin accumulation and whether this is related to the degree of increased microglial density.

The Office of the Clinical Director (OCD), Intramural Research Program (IRP), National Institute on Drug Abuse (NIDA), NIH provides research support to two clinical research branches at the IRP. The scope of the research program is broad and is strongly represented in neuroimaging, as well as behavioral and pharmacological treatments for substance abuse disorders and chemistry / toxicology. The Office of the Clinical Director has a staff of an Administrative Assistant who provides primary support to the Acting Clinical Director and Acting Deputy Clinical Director. This Administrative Assistant also coordinates with the IRP Pharmacy, Matthews Media Group, Inc. (MMG), Medical Records Department, the Mid-Level Providers and Nursing. The three full time Mid-Level Providers (Federal) medically screen potential participants as well as assist in the day to day running of various protocols. Four full time RNs provide the staffing for running the protocols. Additional Clinical staff supporting the NIDA/IRP programs includes 4 full time Research Associates. These positions are part of the clinical services provided through the Johns Hopkins Bayview Medical Contract with NIDA/IRP. This contract also provides essential infrastructure and services including professional physician consultations, laboratory medicine and staff support as well as overnights for subjects at Clinical Research Unit (CRU) located on the Bayview campus. The Addictions IRB, while under the direction of the NIH Office of Human Subjects Research Protections (OHSRP) is supported by the OCD. Staffing includes an IRB Administrator (Federal) with assistance from two other staff members (1 Federal and 1 contractor). The office handles approximately 400 IRB related actions a year, from 58 active protocols. The Addictions IRB serves both NIDA and NIAAA. The OCD oversees a contract with MMG, an outside organization. MMG represents NIDA/IRP, recruiting and screening research participants in the Baltimore Washington area. At this time MMG employs 7 screening specialists, 2 participant counselors, 1 medical assistant and a 2 member management team. There is a yearly budget of $246,000 for print, web, radio and television advertising and $35,000 for materials for recruiting. The Medical Records Department at NIDA/IRP is designed to maintain department compliance of Policy and Procedures while safekeeping the Privacy of over 6300 electronic (HuRIS System) medical records annually. Hardcopy documents of the medical records currently include signed consents and outside medical records; otherwise everything is electronic. Some older charts are available on microfilm. This is done according to the National Institute of Health, Federal and State Rules and Regulations (Including the Privacy Act of 1974 and HIPPA). The IRP Pharmacy employs two full time pharmacists and one pharmacy technician. The IRP pharmacy currently supports about 16 clinical studies including 3 Archway studies and 100 researchers/labs for nonclinical studies. One pharmacist devotes about 80% of her time supporting the clinical studies and 20 % nonclinical studies. The other pharmacist spends about 50 % effort on clinical and 50 % effort on nonclinical studies. Clinical research support includes reviewing, preparing, compounding, and dispensing the study medications. Nonclinical support includes ordering, compounding, dispensing, and laboratory auditing. In addition, the pharmacy monitors drug inventories, and meets all DEA and FDA regulatory requirements including licensing and IND reporting.

Experience-dependent plasticity at excitatory synapses within dopamine neurons of the ventral tegmental area (VTA), a key region for a broad range of motivated behaviors, is thought to be a fundamental cellular mechanism that enables adaptation to a dynamic environment. Synaptic plasticity is mainly described by long-term potentiation (LTP) and depression (LTD) of excitatory synaptic transmission, that are widespread phenomena within the mammalian brain (Malenka and Bear ; 2004). The synaptic insertion or removal of AMPA receptors (AMPAR) plays critical roles in the fine regulation of synaptic activity reflected in LTP and LTD. The cellular events underlying this fundamental processes occurring into DA neurons of the VTA are still poorly understood. To dissect out the functional significance of synaptic plasticity (LTP and LTD) within dopaminergic neurons of the VTA we developed a new mouse line carrying a conditional deletion of the AAA+ ATPase Thorase protein in a DATcre dependent fashion. AAA+ ATPase Thorase protein mediates, in an ATPase-dependent manner, the internalization of AMPAR by disassembling the AMPAR-GRIP1 complex (Zhang et al. 2011). Following genetic deletion of Thorase protein in DATcre expressing cells, the internalization of AMPAR is substantially reduced, leading to several perturbations of synaptic plasticity. My preliminary data suggest that DATcre x Thorase KO mice display reduced LTD and enhanced AMPA/NMDA ratio. This findings parallels and increased amplitude of mEPSCs for DATcrexThorase while the amplitude of mIPSCs is unaffected. We suspected that this peculiar synaptic phenotype is capable of producing perturbations to the physiological expression of general associative learning and memory. To address the latter questions we designed behavioral experiments looking for associative learning related to aversion: fear conditioning test. At this point my preliminary data suggest that cKO animals exhibited a propensity to learn a fearful experience (fear conditioning) compared to their respective littermates controls.

We conducted ex-vivo whole-cell electrophysiological recordings from mice expressing channelrhodopsin-2 in nucleus accumbens afferents, including the ventral hippocampus and amygdala. Thus far, our preliminary results indicate a dissociation in the modulation of nucleus accumbens afferents by kappa-opioid and mu-opioid receptors. Kappa-opioid receptor activation inhibits synaptic transmission in the amygdala to nucleus accumbens, but not hippocampus to nucleus accumbens, pathway. Conversely, mu-opioid receptor activation inhibits hippocampal inputs to a subset of nucleus accumbens neurons, an effect not observed on the amygdala to nucleus accumbens pathway. Future studies are aimed at understanding the cell-types (D1 vs D2) modulated by mu- and kappa-opioid receptors, how different NAcc afferents can drive post-synaptic nucleus accumbens neurons to release opioid peptides, and the contribution of opioid receptors on nucleus accumbens afferent terminals to motivated behavior.

We have found that neurons in the nucleus accumbens inhibit VTA dopamine neurons through GABAB receptor activation while inhibiting VTA GABA neurons through GABAA receptors. The activation of GABAB receptors by nucleus accumbens inputs is sufficient to pause spiking in VTA dopamine neurons. Using optogenetic manipulations, cell-type specific lesions, and slice electrophysiology, we have shown that dynorphin-containing neurons in the nucleus accumbens are both necessary and sufficient for the activation of GABAB receptors in dopamine neurons. We have also shown that multiple neuromodulators can increase or decrease the strength of inhibitory transmission in this pathway. Currently, we are attempting to delete GABAB receptors to examine their role in naturalistic and addictive behaviors.

Fear learning is a fundamental behavioral process shared by rodents and humans and requires dopamine (DA) release. Experience-dependent synaptic plasticity, a cellular form of learning and memory, occurs on DA neurons while an organism is engaged in aversive experiences. However, whether synaptic plasticity on DA neurons is causally involved in aversion learning is unknown. Here, we took advantage of the ability of the ATPase Thorase in regulating the internalization of AMPA receptors (AMPARs) in order to artificially manipulate glutamatergic synaptic plasticity in DA neurons. Genetic ablation of Thorase in DAT+ neurons produced increased AMPAR surface expression and impaired the induction of both long-term depression (LTD) and long-term potentiation (LTP). Strikingly, animals lacking Thorase in DAT+ neurons expressed greater associative learning in a fear-conditioning paradigm without showing fear generalization or anxiety. In conclusion, our data provides a novel, causal link between induction of synaptic plasticity onto DA neurons and fear learning. In the current project we characterize the molecular and physiological role of Thorase protein in respect to the internalization of AMPAR and in the occurrence of an activity-dependent glutamatergic synaptic plasticity in the context of DA neurons. Additionally we show that animals lacking Thorase from DA neurons display an increase propensity toward learning and memory aversive experiences.

It is well-documented that midbrain dopamine neurons express D2 autoreceptors. These receptors are Gi-coupled, resulting in hyperpolarizing potassium currents when activated. Furthermore, these receptors are endogenously active in vivo: pharmacological inhibition causes increased firing rates of VTA dopamine neurons, and selective genetic knockout of D2 autoreceptors results in hyperactive mice. Despite these observations, the source of dopamine that activates these autoreceptors is not understood. Golgi stains have revealed that the VTA does not contain robust dopaminergic axon collaterals, which would suggest an alternative source rather than local classical synaptic release. The consensus view within the field is that the dopamine that activates D2 autoreceptors comes from dendritic release of neurotransmitter. However, a major problem with this theory is that electron microscopic studies have noted an absence of dopamine-containing synaptic vesicles in dendrites within the VTA. We are currently using a combination of electrophysiological, optogenetic, pharmacological, and genetic techniques to explore alternative possible sources of dopamine onto these receptors.

In the past year, we have made significant progress towards realizing the goals of this project. As of August, we have completed a set of slice recording experiments comparing the basic electrophysiological properties of astrocytes in midbrain regions with those of astrocytes in the cortex and hippocampus. These experiments also allow us to determine the extent to which astrocytes and oligodendrocytes are coupled within each brain region by visualizing the diffusion of small molecule tracers (e.g. biocytin) loaded into a single astrocyte. Since astrocytes exhibit complex patterns of calcium signaling, weve also profiled the calcium activity of midbrain and hippocampal astrocytes in response to a number of agonists. In addition to these functional assays, we have worked out a successful protocol for acutely isolating astrocytes from adult brain tissue via a combination of microdissection and fluorescence activated cell sorting (FACS). As of right now, astrocyte RNA transcripts from several brain regions have been sequenced and analysis of these data is ongoing. Finally, we are working to secure mouse lines that will enable us to probe the function of an enzyme involved in glutamate homeostasis that is selectively abundant within midbrain oligodendrocytes.

I combine optogenetic stimulation with multi-optetrode recordings in freely moving mice to study how reward associative learning and neural activities in the ventral tegmental area (VTA) are modulated by VTA afferents. The VTA plays an essential role in reward, motivation and associative learning. The mesolimbic dopamine (DA) system helps establish incentive value for neutral stimulus in acquisition of stimulus-reward associations. Dopamine neurons fire phasic burst activity during presentation of reward or reward predictive cues. It is believed that DA neuron burst activity depends on afferent inputs. Anatomical and physiological evidence lead me to study the regulatory roles of the pedunculopontine tegmental nucleus (PPTg) in VTA circuits and its role in modulating behaviors. I employed optogenetic approach to selectively inhibit PPTg glutamatergic or cholinergic inputs to the VTA in order to examine the contribution of these inputs in Pavlovian appetitive conditioning. Furthermore, I performed in vivo optetrode recording in freely moving mice as they performed a Pavlovian task to examine the regulation of VTA circuits by PPTg glutamatergic or cholinergic afferents. The results show that PPTg-to-VTA excitatory inputs play a critical role in the acquisition of stimulus-reward associations by dampening cue discrimination. Surprisingly, the results from in vivo recording experiments show that cue-elicited burst activity of DA neurons is not affected by photoinhibition of PPTg-to-VTA excitatory inputs. Instead, I observe that photoinhibition significantly impacts cue-elicited neural activity in non-DA neurons. This raised a possibility that VTA non-DA neurons may play a role in acquisition of cue-reward association. To examine this possibility, VTA GABAergic or glutamatergic neural activity is inhibited with optogenetic inhibition approach during Pavlovian conditioning learning. The results show that inhibition of VTA GABAergic or glutamatergic neural activity dampens cue-reward associative learning. In conclusion, my study reveals an unprecedented role of PPTg-to-VTA pathways in modulating the activity of VTA non-DA neurons and cue-reward associative learning. My future research will build on these findings to explore possible involvement of VTA non-DA efferents in cue-reward associative learning. The goals are to determine which putative targets innervated by VTA non-DA neurons are critically required in discrimination learning, whether chronic use of substance of abuse will dampen discrimination learning and impact selective VTA non-DA efferents and develop an optical intervention technique to rescue learning deficits.

As KORs inhibit both glutamate release onto NAcc medium-sized spiny neurons (MSNs) via a presynaptic site of action, we first determined whether KORs differentially regulated glutamatergic inputs to the NAcc.. Moreover, we find that KORs inhibit basolateral amygdala (BLA), but not ventral hippocampal afferents to MSNs. KORs are directly situated on BLA terminals to NAcc as genetic ablation of BLA KORs blocks prevents KOR agonist inhibition of glutamate release from BLA terminals and increases the basal probability of glutamate release. We next determined whether KORs were differentially inihibiting glutamate and GABA release onto the two projection neurons of the NAcc, D1 and D2 MSNs, which play opposing roles in reward and mood. KORs inhibit glutamatergic afferents to D1 MSNs an effect that is not evident in D2 MSNs. KORs also inhibit glutamate release from BLA afferents onto D1 MSNs but not D2 MSNs. Intriguingly, KORs inhibit GABA release onto both D1 and D2 MSNs. Furthermore, D1 MSN GABAergic inputs to other MSNs are also inhibited by KORs. Thus, KOR signaling changes excitation and inhibition to differentially change in D1 and D2 MSNs. We are also actively investigating how endogenous dynorphin release may impact synaptic transmission in the NAcc. Strong, but not weak, optogenetic activation of populations D1 MSNs, which express dynorphin, causes the release of dynorphin to act on KORs glutamatergic afferents. As dynorphin-mediated inhibition of glutamate release is present in cells activated and not activated by optogenetic stimulation and strong activation of a single D1 MSN with similar stimulation patterns is not sufficient to inhibit presynaptic glutamate release, KOR-mediated inhibition of glutamate release may not be a consequence of dynorphin release from the recorded neuron. Interestingly the same optogenetic stimulation pattern that causes dynorphin release onto glutamatergic synapses does not modify GABAergic transmission evoked electrically or optogenetically from D1 MSNs to other MSNs. Taken together, these results suggest that released endogenous dynorphins inhibit glutamatergic transmission via volume transmission, but are restricted with regards to GABAergic synapses. Collectively, here we show that the endogenous dynorphin/kappa-opioid receptor system modulates information processing in the NAcc by differentially regulating glutamatergic and GABAergic synapses onto D1 and D2 MSNs. Moreover, we also provide the first demonstration that D1 MSNs utilize endogenous opioid systems to influence synaptic transmission

Afferent inputs to the ventral tegmental area (VTA) control reward-related behaviors through regulation of dopamine neuron activity. The nucleus accumbens (NAc) provides one of the most prominent projections to the VTA, however recent studies have provided conflicting evidence regarding the functional role of these inhibitory inputs. Using optogenetics, cell-specific ablation, whole cell patch-clamp, and immuno-electron microscopy, we show that NAc inputs synapse directly onto dopamine neurons, preferentially activating GABAB receptors. We demonstrate that GABAergic inputs from the NAc and local VTA GABA neurons are differentially modulated and activate separate receptor populations within dopamine neurons. Genetic deletion of GABAB receptors from dopamine neurons in adult mice did not affect general or morphine-induced locomotor activity, but profoundly increased cocaine-induced locomotion. Collectively, our findings demonstrate remarkable selectivity in the inhibitory architecture of the VTA and suggest that long-range GABAergic inputs to dopamine neurons fundamentally regulate behavioral responses to cocaine.

We discovered that oligodendrocytes in the midbrain and hindbrain express high levels of the glutamate metabolizing enzyme glutamine synthetase (GS). In addition to this striking observation, we also detected functional glutamate transporter currents in midbrain oligodendrocytes, thus demonstrating that oligodendrocytes can both take up and metabolize extracellular glutamate. As such, they are in a position to regulate the concentration and localization of extracellular glutamate, which may have implications for both excitatory synaptic transmission as well as protection from excitotoxicity. We are now crossing a floxed GS mouse line with an oligodendrocyte-specific cre line to determine whether oligodendrocyte GS is required for proper regulation of extracellular glutamate.

Synaptic plasticity involves several orchestrated changes of synaptic efficacy supporting experience-dependent modifications of brain function, physiological foundations of learning and memory. The latter process is mainly described by long-term potentiation (LTP) and depression (LTD) of excitatory synaptic transmission, that are widespread phenomena within the mammalian brain (Malenka and Bear ; 2004). The synaptic insertion or removal of AMPA receptors (AMPAR) plays critical roles in the fine regulation of synaptic activity reflected in LTP and LTD. The cellular events underlying this fundamental processes occurring into DA neurons of the VTA are still poorly understood. To dissect out the functional significance of synaptic plasticity (LTP and LTD) within dopaminergic neurons of the VTA we developed a new mouse line carrying a conditional deletion of the AAA+ ATPase Thorase protein in a DATcre dependent fashion. AAA+ ATPase Thorase protein mediates, in an ATPase-dependent manner, the internalization of AMPAR by disassembling the AMPAR-GRIP1 complex (Zhang et al. 2011). Following genetic deletion of Thorase protein in DATcre expressing cells, the internalization of AMPAR is substantially reduced, leading to several perturbations of synaptic plasticity. My preliminary data suggest that DATcre x Thorase KO mice display reduced LTD and enhanced AMPA/NMDA ratio. This findings parallels and increased amplitude of mEPSCs for DATcrexThorase while the amplitude of mIPSCs is unaffected. Surprisingly, the impact of this synaptic configuration at the behavioral level is quite remarkable. We tested our animals with a fear conditioning paradigm, an associative learning and memory essay. We discovered that cKO animals exhibited a propensity to learn from a fearful experience (fear conditioning) compared to their respective littermates controls. It is tempting to speculate that a rigid synaptic state within DA neurons residing into the VTA is affecting the way animals cope with stressful events. This scientific project is now translating into a clinical direction since our collaborators Ted and Valina Dawson at Johns Hopkins University discovered 3 different point mutations of Thorase protein into families of Ashkenazi Jewish origin were schizophrenia cases were recurrent. We are now taking advantage of mouse genetics and electrophysiology to understand, in the mouse model, the molecular underpinnings of the observed point mutations at the functional level.

Taste helps establish food preference and the neural processing of taste aspect of food reward starts from the gustatory system. The gustatory information via cranial nerves reaches the brain first in the nucleus of the solitatory tract (NST). Parabrachial nucleus (PBN) receives the excitatory glutamatergic input from the NST and has been shown to control feeding. In addition, PBN is also necessary to convey hedonic information of taste stimuli. Given that PBN sends substantial projections to midbrain dopamine neurons, it raises the possibility that taste stimuli may engage dopamine neurons in the ventral tegmental area (VTA) via PBN to modulate food intake. Thus, in the present study, we set out to study the functional connection between PBN and VTA in food consumption. We first carried out in situ hybridization to verify the neuronal types in PBN. In the past, locally electrical stimulation or pharmacological manipulation has the limitation to characterize defined connections. Thus, we would employ optogenetic approach by selectively express channelrhodopsin (ChR2) in specific Cre transgenic mouse line to target specific subtype of PBN neurons. Littermates of the same mouse line were injected with cre-dependent virus encoding EYFP alone into PBN to serve as control group to rule out confounding variables such as surgery procedures. We specifically tweaked PBN-to-VTA afferent input by delivering 473 nm blue light to the VTA via fiber optics while mice were performing behavioral task. The preliminary results showed that VTA receives functional inputs from the PBN and photostimulation of PBN-to-VTA input affects food consumption. In addition to my own project, I also contribute to another research with Dr. Dong Wang and Dr. Ikemoto Satoshi. We focused on studying modulation of neuronal oscillation in conditioned fear paradigm. By combining optogenetics and in vivo simultaneous electrophysiology recordings in freely-moving mice, our study exhibited that optical stimulation of subcortical median raphe nucleus to hippocampal ripple oscillation during sleep impairs hippocampus-dependent fear memory. This work was published in Nature Neuroscience in 2015.

We found that: 1) under basal conditions (i.e., before MPTP/6-OHDA injection), there is no significant change in midbrain TH-immunostaining, DA cell counting or basal levels of locomotion. However, methamphetamine-induced increase in locomotion was significantly reduced in Vglut2-cKO mice than in Vglut2-het control mice; 2) MPTP (20 mg/kg 4 with 2-hr injection interval, s.c.) caused more DA neuron death/loss in the SNc in Vglut2-cKO mice (80%) than in Vglut2-het control mice (40%), as assessed by TH-immunostaining, DA neuron counting, and tissue DA/DOPAC contents in both the striatum and SNc; Similarly, 3) intra-striatal microinjection of 6-OHDA also caused more DA terminal degeneration in the striatum in Vglut2-cKO mice than in Vglut2-het mice, as assessed by TH-immunostaining; Furthermore, 4) behavioral assays suggest that MPTP caused more severe impairment in rotarod locomotor performance and a larger reduction in locomotor activity in Vglut2-cKO mice than in Vglut2-het mice. These findings suggest that deletion of Vglut2 in DA neurons increases DA neuron susceptibility (toxicity) to the neurotoxins MPTP or 6-OHDA. To further confirm this finding, we will use transgenic techniques to increase Vglut2 expression in midbrain DA neurons in the future study to determine whether increased Vglut2 expression in midbrain DA neurons will produce a neuroprotective effect against MPTP- or 6-OHDA-induced DA neuron injury.

Serotonin is produced in neurons of the raphe nuclei. Ascending serotonergic neurons of the brain are restricted to the dorsal raphe nucleus (DRN) and median raphe nucleus (MRN). Neurons in the DRN and MRN project widely throughout the brain, where they regulate diverse physiological and behavioral processes. Several electrophysiological studies have demonstrated that serotonergic neurons alter their firing rates in response to reward-predictive cues and consumption of food and liquid rewards; however, the causal role of these neurons in controlling reward learning remains unknown. To characterize the role of serotonin in these processes, we have undertaken a multifaceted behavioral approach. First, we explored whether mice would self-stimulate serotonergic neurons using the light-activated ion channel Channelrhodopsin-2 (ChR2) delivered virally into serotonergic neurons of the DRN or MRN. In both cases, mice failed to self-stimulate these brain regions, suggesting that serotonin from these neurons does not convey a reinforcement signal. We next performed loss-of-function experiments knocking out the enzyme tryptophan hydroxylase-2 (tph2), the rate-limiting enzyme in serotonin synthesis. Knockout was induced by virally injecting virus expressing cre recombinase into either the DRN or MRN of adult mice. Although knockout did not alter general measures of locomotion, it produced opposing effects in a Pavlovian conditioned approach experiment, with knockout in the DRN enhancing measures of reward learning and knockout in the MRN reducing such measures. Interestingly, the opposite pattern of results were observed in a conditioned reinforcement paradigm: DRN knockouts performed normally, whereas MRN knockouts showed greatly enhanced responding. Experiments are currently underway targeting efferent projections of DRN and MRN serotonergic neurons, to test what brain region(s) mediate these effects.

Mechanisms of deep brain stimulation in the Nucleus Accumbens are unknown. Electrical stimulation of this region is complex, activating all cells within the region of stimulation as well as excitatory and modulatory afferents. Therefore, it is necessary to specifically parse out effects on individual cells within the region. To determine the effect of stimulation on cells within the NAc, cell-type specific activation of MSNs was performed by using an adeno-associated virus expressing ChR2 in NAc D1-MSNs of Dynorphin-Cre/Tdtomato mice. D1- and D2-MSNs are known to release different peptides locally and at efferent projections. Results indicate Substance P, exclusively expressed in D1-MSNs, may contribute to the behavioral effects of stimulation. To probe excitatory transmission, slices are prepared and whole-cell patch clamp is performed on D1-MSNs and D2-MSNs under fluorescent guidance. Excitatory input is electrically evoked prior to and following 50Hz (high frequency) optogenetic stimulation. Stimulation of D1-MSNs significantly depresses excitatory transmission on D1-MSNs, but potentiates excitatory transmission on D2-MSNs. Blocking the main receptor for Substance P, NK1 receptors, blocks potentiation on D2-MSNs and induces synaptic excitatory depression to the same degree as the stimulation-induced depression on D1-MSNs. Substance P (50nM) application mimics this potentiation effect on D2-MSNs in the NAc in a striatal dorsolateral to ventromedial gradient. These results suggest activation of NK1 receptors and Substance P release mediates the excitatory potentiation produced by stimulation. Furthermore, these effects are exclusively post-synaptic. Interestingly, using in situ hybridization, NK1 receptor expression is almost exclusively found in interneurons and all cholinergic interneurons express NK1 receptors. This suggests a disynaptic effect through cholinergic interneurons, may mediate the effect of Substance P on neuronal transmission. Indeed, high frequency but not low frequency stimulation causes elevated firing of cholinergic interneurons for up to 15 min following stimulation. Additionally only a single 5 sec train of HFS is necessary to produce elevated cholinergic interneuron firing and excitatory potentiation. Excitatory potentiation on D2-MSNs is muscarinic 1 receptor dependent signifying increased acetylcholine release is necessary for this effect. In these experiments, I demonstrate for the first time D1-MSNs can drive excitatory transmission on D2-MSNs. These results are significant in several ways: 1) D1-MSNs can drive excitation of cholinergic interneurons 2) High frequency stimulation or strong D1-MSN activation can rebalance output of MSN subtypes. Because MSN subtypes often oppositely mediate reward related outcomes (e.g., D1-MSNs promote reward while D2-MSNs blunt reward), we predict this stimulation can blunt opioid-mediated reward. Future experiments will test the potential effect of D1-MSN high frequency stimulation on reward. Additionally, current experiments are examining how stimulation releases Substance P in vivo using microdialysis and in vivo imaging to determine if afferent stimulation can drive Substance P release.

The current study demonstrates that stressful experiences induces a negative affective state characterized by anhedonia and behavioral despair. This phenotype can be reversed independently by depotentiating observed increased synaptic strength of limbic ventral hippocampus (VH) excitatory synapses onto D1 medium spiny neurons (D1-MSNs) in the NAc shell, or restoring depression-induced decreased potassium channel function in hyperexcitable D1-MSNs. Furthermore, in nave animals, mimicking the aforementioned synaptic and intrinsic adaptations in NAc D1-MSNs induced by stressful experiences is sufficient to promote both anhedonia and behavioral despair. Moreover, utilizing a novel disconnection procedure, we demonstrate that strengthening of VH synapses and excitability changes in D1-MSNs induced by potassium channel dysfunction are serial processes that promote anhedonia and behavioral despair after stressful experiences. Lastly, we deconstruct how information gets received from the VH to NAc D1-MSNs and relayed to their downstream targets to drive negative affect. Taken together, our study identifies a novel disynaptic pathway for control of emotional behavior in which VH shapes information flow into NAc D1-MSNs to specific downstream targets. Over the last two to three decades, the dogma in the field of neuroscience is that D1-MSN activity supports appetitive and reinforcing behaviors, due to limited investigation of this cell type in negative affective states. However, here we provide unexpected, novel evidence that D1-MSNs also participate in the context of negative affective states. This evidence will encourage the field of biological psychiatry to revisit and reinterpret much of the present literature and help guide future studies. In conclusion, there is a paucity of effective treatments in the clinic for negative affective states due to a critical knowledge gap in our understanding of the mechanisms promoting such symptoms. Thus, our study provides novel, druggable targets for rationale, evidence-based therapeutic interventions for endophenotypes observed in a plethora of neuropsychiatric disorders.

Glutamine synthetase (GS) is an enzyme that converts glutamate and ammonia into glutamine. It is thought to be essential for regulating levels of glutamate and ammonia in the brain, and changes in GS expression and enzymatic activity have been reported in several neurological disorders, including epilepsy, hepatic encephalopathy, and Alzheimers disease. GS is commonly described as being exclusively expressed in astrocytes, which take up glutamate from the extracellular space, and is widely used as an astrocyte-specific cell type marker. Unexpectedly, we observed widespread GS expression in mature oligodendrocytes throughout the brain. Given the potential implications of this finding on our current understanding of the brain glutamate-glutamine cycle, brain ammonia processing, and oligodendrocyte function, we decided to investigate the role of GS in oligodendrocytes. In the past year, we have characterized mice in which oligodendrocyte GS has been conditionally deleted (cKOs). cKOs show profound reductions in the levels of brain glutamate and glutamine, as well as deficits in glutamatergic neuronal tranmission in the midbrain. In addition, cKOs are impaired in cocaine sensitization, a behavior that is known to require glutamate signaling in the midbrain. Therefore, we conclude that oligodendrocyte GS is required for maintaining glutamate synaptic transmission in the brain.

Using conditional VgluT2-KO techniques, we found that deletion of VgluT2 in a subpopulation of DA neurons abolished glutamate release from these cells, causing a reduction in BDNF and TrkB expression in midbrain DA neurons and an increase in vulnerability to MPTP-induced DA cell death and locomotor impairment. Restoration of VgluT2 expression in VgluT2-cKO mice normalized BDNF/TrkB expression and attenuated MPTP-induced toxicity in DA neurons and locomotor dysfunction. These findings suggest that reduced VgluT2 expression in DA neurons may be a newly-identified risk factor in the development of neurodegenerative diseases, such as PD, and normalization of VgluT2 expression in DA neurons may be useful in preventing and treating PD or other neurodegenerative diseases.

Taste helps establish food preference and the neural processing of taste aspect of food reward starts from the gustatory system. The gustatory information via cranial nerves reaches the brain first in the nucleus of the solitatory tract (NST). Parabrachial nucleus (PBN) receives the excitatory glutamatergic input from the NST and has been shown to control feeding. In addition, PBN is also necessary to convey hedonic information of taste stimuli. Given that PBN sends substantial projections to midbrain dopamine neurons, it raises the possibility that taste stimuli may engage dopamine neurons in the ventral tegmental area (VTA) via PBN to modulate food intake. Thus, in the The release of noradrenaline in midbrain monoaminergic nuclei, including the dorsal raphe nucleus (DR), ventral tegmental area (VTA), and substantia nigra pars compacta (SNc) excites serotonin and dopamine neurons through the activation of excitatory 1-adrenergic receptors, enhancing action potential firing and neurotransmitter release. The release of noradrenaline from axon terminals is through the activation of inhibitory 2-adrenergic receptors. Despite this influential role, the most basic questions regarding noradrenaline-dependent excitation remain unanswered, stemming from an incomplete characterization of 1- and 2-adrenergic receptor signaling at a synaptic level and within the context of the brain circuits and behavior. We have been using wild type mice for acute brain slice electrophysiology to determine the signal transduction pathway and spatial and temporal kinetics of 1-adrenergic receptor signaling in the DR. A manuscript detailing the results of these studies is in preparation. In addition, intracranial microinjections of retrograde neural tracers have been made in the DR of wild type mice, followed by transcardial perfusion and histological experiments in order to determine the origin of noradrenaline axon terminals in the DR. Activation of 1-adrenergic receptors in the DR by dopamine released from midbrain dopamine neuron axon terminals has been examined by intracranial microinjection of viral vectors encoding for the selective expression of light-activated channelrhodopsin 2 (AAV-DIO-ChR2-eYFP) in the SNc/VTA of TH-cre mice. Functional connectivity was assessed by acute brain slice electrophysiology.