Alexei Morozov - US grants

Among these groups, cholinergic neurons are considered the key modulators of the oscillatory activities. In the past, the functional role of cholinergic neurons has been studied by the elimination of these neurons with immunotoxins, however this irreversible elimination of neurons brings about irreversible changes compromising interpretation of behavioral experiments. To directly test the role of oscillations in learning, memory and mood, we will reversibly inactivate cholinergic neurons in the mouse brain using regulated expression of the light chain of tetanus toxin. This toxin does not kill neurons, but prevent secretion of neurotransmitter by cleaving synaptobrevin, which is required for the docking of synaptic vesicles. Once the expression of the toxin is turned off, neurons should recover their functions. We will test the role of rhythmic oscillations at different stages of memory formation, consolidation and retrieval taking advantage of the reversibility of the system. In vivo recording and analysis of neuronal activity will be performed by Dr. Buzsaki at Rutgers University. We have completed the design of the scheme for reversible genetic inactivation of cholinergic neurons. The scheme includes generation of 2 lines of genetically modified mice. The first line will express tetracycline transactivator in the cholinergic neurons. It will be produced by targeting cholinergic locus with the construct harboring a gene for tetracycline transactivator. The second line will carry modified inactive tetanus toxin, which could only be activated only in the brain following a withdrawal of doxycycline from mouse diet. During the past fiscal year, we have completed a generation of mice with the tetracyclin transactivator gene expressed from the cholinergic locus promoter. Specifically, as the last step, we have successfully excised a selectable marker used during targeting of the locus. We were continuing work on making a targeting construct containing the tetanus toxin gene.

The main goal of this project is to understand how changes in the hippocampus may cause pathological aggression. We have previously found that mice with the conditional knockout (KO) of Brain Derived Neurotrophic Factor (BDNF) restricted to the hippocampal area CA3 are more aggressive than their wild type (WT) counterparts. These animals had reduced levels of serotonin (5-HT) and increased expression and activity of the serotonin receptor 3 (5-HTr3), and over-activation of 5-HTr3 increases aggression in wild type mice. The decrease in the levels of 5-HT appeared to cause the up-regulation of 5-HTr3 in BDNF KO mice, because chronic administration of fluoxetine, which elevates concentration of extracellular 5-HT, normalized expression and activity of 5-HTr3, it also reduced aggression. In search for physiological properties of the hippocampus that are regulated by 5-HTr3 and possibly correlate with aggression, we found that 5-HTr3 agonist suppresses induction of hippocampal gamma oscillations by carbachol. During the last fiscal year, we focused on two questions: 1) How loss of BDNF in the hippocampus reduces levels of 5-HT? 2) What mechanisms underlie suppression of gamma oscillations through 5-HTr3? To address the first question, we first examined 5-HT reuptake, which is thought to be influenced by BDNF, and found no difference between KO and control WT mice. This result suggested that lower 5-HT release could be responsible for the reduced 5-HT concentration in KO mice. Accurate measurement of 5-HT release in real time remains a technical challenge, because the release sites are compartmentalized and cannot be readily accessed by a human-made measuring tool. To circumvent this problem, we used fast 5-HT transmission mediated by 5-HTr3 as a measure of 5-HT release. 5-HTr3 is the only ionotrophic 5-HT receptor, which it is a cation channel that opens within milliseconds upon 5-HT release. The current across the channel reflects the amounts of 5-HT released by an electrical stimulus. The channel is expressed in a subpopulation of hippocampal inhibitory neurons (5-HT3 cells), which are readily identifiable in transgenic mice containing GFP gene under the control of 5-HTr3a gene promoter. We began whole cell recordings from hippocampal 5-HT3 cells and preliminary data suggest that 5-HT terminals in BDNF KO mice release less 5-HT than the terminals in WT controls. To address the second question about mechanisms whereby 5-HTr3 activation suppresses gamma oscillations, we examined effects of 5-HTr3 agonist m-CPBG on the cellular properties of 5-HT3 neurons and oscillatory properties of parvalbumine (PV) expressing interneurons, which are the main driving force for gamma oscillations. We found that m-CPBG reduced excitability and firing frequency of 5-HT3 cells by increasing afterhyperpolarization (AHP). This increase in AHP was mediated by calcium-activated small conductance potassium (SK) channels, which were activated as a result of 5-HTr3-mediated depolarization and calcium influx. At the same time, m-CPBG attenuated inhibitory currents in PV-interneurons and increased their firing, which became more random and less coupled to the oscillation phase. Based on these observations, we propose that activation of 5-HTr3 interferes with gamma oscillations by attenuating inhibitory input from 5-HT3 cells to PV interneurons and by increasing random firing of PV cells. Our future goals are to understand how oscillatory activity of hippocampus may relate to aggression and whether 5-HTr3 modulates aggression by changing hippocampal oscillations.

Fear behaviors, which are driven by the amygdala, can become maladaptive during mental illness. Our main goal is to understand how amygdala circuitry triggers fear behaviors and how this function changes during transition from normal to pathological states. This knowledge will provide information that will help in developing treatments for mental disorders associated with pathological fear. Amygdala operates by analyzing incoming information and triggering defensive responses. The first question of our investigation is how amygdala distinguishes, at the synaptic level, between signals which arrive from different areas of the brain, those which process sensory information, and those which provide executive control. To address this question, one needs to interrogate a specific input by selectively stimulating fibers coming from a specific brain area. To address this question we established opsin-based techniques for selective activation or silencing of amygdala inputs from cortical area TeA which transmits sensory information, and from the anterior cingulate cortex, which is implicated in affect, pain and cognition. Both inputs target same amygdala neurons and are intermingled inside amygdala. We found significant difference in synaptic plasticity between the two pathways. While long-term potentiation of synaptic transmission (LTP) in the input from perirhinal cortex required suppression of GABAa receptor-mediated inhibition, LTP in the ACC-amygdala pathway did not. Moreover, severing connections between external capsule and amygdala enabled LTP in the input from perirhinal cortex even in the presence of GABAa receptor-mediated inhibition. In addition, we found that these two inputs exhibit differential connectivity to the amygdala inhibitory neurons. The ACC input was more effective in activating interneurons that express serotonin receptor 3, whereas the TeA input was more effective in recruiting the pericapsular cells. These findings have interesting implications: first, dopamine-dependent inhibitory neurons of the external capsule appear to gate plasticity in the amygdala input from perirhinal cortex, whereas serotonin-dependent interneurons inside the basolateral nucleus gate the highly refined information from ACC. We continued to investigate the mechanisms responsible for amygdala disinhibition and focused on dopaminergic modulation of local microcircuits. Using interneuron-specific lines of transgenic Cre-mice, we selectively activated parvalbumin positive neurons in the basolateral amygdala and found that dopamine selectively suppressed GABA release towards principal cells, but not towards interneurons. Our current goal is to determine how neuromodulators, through their synapse-specific effcts, control the balance between inhibition and excitation in the basolateral amygdala nucleus.

DESCRIPTION (provided by applicant): Antisocial personality disorder (ASPD) is ...a pervasive pattern of disregard for, and violation of, the rights of others that begins in childhood or early adolescence and continues into adulthood. ASPD is found in 1.0 % of the U.S. adult population. People with ASPD break social norms, lie repeatedly, place others at risk for their own benefit, and demonstrate a profound lack of remorse. What is referred to as psychopathy or sociopathy is a variant of ASPD characterized by instrumental/proactive aggression and lack of empathy; at the same time, there is no pathological anxiety, depression or cognitive deficits. While such behaviors impose severe burden on human society, ASPD is considered untreatable because of difficulty in clinical management of the patients and lack of intervention strategies. ASPD is predicted in youth with conduct disorder (CD); however, the fundamental mechanisms of CD/ASPD remain unknown and potential treatment strategies have not been investigated, partly because there is no good animal model. The goal of the proposal is to validate and investigate a novel empirical animal model of ASPD, mice with the deletion of BDNF gene in the hippocampal area CA3 (KO mice). These animals are aggressive and appear to have no empathy-like behaviors, but have normal cognition, anxiety and depression-like behaviors. This phenotype is reminiscent of ASPD symptoms in humans and was associated with reduced levels of serotonin and enhanced activity of serotonin receptor 3, whose activation inside hippocampus enhanced aggression. Surprisingly, this receptor attenuated hippocampal gamma oscillations. It is unclear through which neuronal mechanisms BDNF and serotonergic system in the hippocampus control aggression or empathy. The proposal represents a multidisciplinary study and combines behavioral and biochemical analyses, in vitro slice recording and pharmacogenetics to addresses the following questions: Which forms of empathy-like behaviors are compromised in KO mice? What changes in the hippocampus of KO mice may be relevant to the escalated aggression? Can pharmacological intervention with these changes attenuate escalated aggression and increase empathy-like behaviors? Such studies may validate BDNF CA3 KO mice as a model of ASPD and promote development of novel therapies for ASPD.

? DESCRIPTION (provided by applicant): Abnormal function of the prefrontal cortex and the amygdala, which interact with each other to control emotions, memory, and decision making, have been implicated in behavioral traits of major mental disorders. Yet, very little is known about how synaptic transmission between the two structures is altered by environmental factors that lead to mental disease. Psychological trauma is one of such factors. It increases risk of developing stronger fear memories in the future, upon exposure to another traumatic event. Here, we will test a hypothesis that a purely psychological trauma makes synaptic connections between prefrontal cortex and amygdala more prone to facilitation during fear learning and that silent synapses, which are generated after psychological trauma, are responsible for the enhanced facilitation. Our preliminary experiments, in which mice are exposed to a conspecific under distress, revealed that such exposure enhances future formation of fear memory in the passive avoidance paradigm. We also found increased number of silent synapses in dmPFC-BLA pathway, which have NMDA receptor, but do not have functional AMPA receptor. Interestingly, the emergence of silence synapses and enhanced avoidance learning were abolished when mice were treated with sub-anesthetic doze of ketamine immediately after psychological trauma. The objectives of the proposal are to understand the process leading to formation of silent synapses, their removal by ketamine and the role silent synapses may play in plastic changes that occur in dmPFC-BLA connections during avoidance learning. The proposal employs optogenetic stimulation of specific axonal fibers and recording of synaptic responses in amygdala slices prepared from animals exposed to combinations of psychological trauma and avoidance learning paradigms. As a parallel approach, we will use immuno-electron microscopy quantification of glutamate receptors in dmPFC-BLA synapses, identified using anterograde tracing. The following questions will be addressed. Are silent synapses between dmPFC and BLA generated by insertion of NMDAR into cell membrane, or by removal of AMPAR? Do they enhance long-term potentiation in dmPFC-BLA pathway in slice? Are they used during passive avoidance learning? What is the mechanism of their elimination by ketamine? By focusing on dmPFC-BLA connection and BLA microcircuit, this study will help elucidate role of defined cellular and synaptic elements underlying emotional traumatization and validate them as a potential target for novel therapies PTSD, depression and related emotional disorders.

Project summary Remote and local connectivity of the prefrontal cortex is altered in mental disease, but little is known about underlying synaptic mechanisms. Emotional trauma is an environmental factor that increases odds of developing mental disease in the future, upon exposure to another traumatic event. We have recently shown that a rodent analog of emotional trauma, the observational fear paradigm (OF), in which mice are exposed to a conspecific in distress, enables stronger passive avoidance learning at a later time. Our preliminary analysis of synaptic transmission in the dorsomedial prefrontal cortex (dmPFC) of traumatized mice, which is essential for passive avoidance learning, revealed a redistribution of inhibition along the somato-dendritic axis of the layer V pyramidal cells (PCs) via a GABAb receptor-mediated mechanism. The objectives of this proposal are two-fold: first, to determine synaptic and cellular mechanisms of that redistribution by testing a hypothesis that OF uniquely modulates GABAb receptor signaling in genetically-distinct classes of GABAergic neurons, and, second, to examine its consequences for synaptic transmission between the basolateral amygdala (BLA) and dmPFC. Our pilot experiments showed that the two ascending pathways from BLA to dmPFC that arrive in layers II/III and V, can be optogenetically stimulated with high selectively using focal blue light and that such stimulation elicits different patterns of feedforward inhibition in the layer V PCs. A prediction will be tested that OF has opposing effects on inhibition recruited by these pathways. The proposal employs optogenetic stimulation of specific axonal fibers using localized focused light, pharmacogenetic silencing of specific interneuronal populations and recording of synaptic responses in dmPFC slices prepared from animals subjected to OF. The following questions will be addressed. Which types of GABAergic neurons are responsible for the somato-dendritic shift in inhibition? What is the subcellular compartment where GABAb receptor signaling is altered? Does OF alter the balance between excitation and feedforward inhibition elicited in layer 5 PCs by the two BLA inputs? Which interneurons mediate feedforward inhibition recruited by each input? Do glutamatergic inputs in those interneurons play a role in the effects of OF? By identifying cells, synapses and molecules that redistribute inhibition of L5 PC along somato- dendritic axis and by examining resulting changes in transmission via the BLA-dmPFC ascending pathway, this project will help determine mechanisms of the enhanced fear learning after psychological trauma and identify potential targets for novel therapies in PTSD, depression and related emotional disorders.

Project Summary During development, learning or progression towards disease, the brain undergoes plastic changes characterized by gradual increases or decreases in synaptic transmission. Modeling such changes by artificial means is necessary for understanding their functional role. Recently developed optogenetic and chemogenetic techniques allow neuronal activation or suppression, but they act in the all-or-none manner and do not allow modeling of gradual synaptic changes. Meanwhile, such gradual changes can be obtained by inducing long-term potentiation (LTP) or depression, but these techniques do not work reliably in all areas of the brain, particularly in the areas with a strong inhibitory control, which include the basolateral amygdala. In the pilot experiments, we found that a transient chemogenetic or optogenetic suppression of the somatostatin- but not parvalbumin-positive interneurons enables LTP induction in the prefrontal- amygdala pathway. Based on these findings and published data, we hypothesize that a transient suppression of certain classes of the local GABAergic neurons, combined with stimulation of synapses of interest, will provide a universal means for inducing LTP in the remote inputs to the local principal neurons in vivo. We will test this hypothesis in Aim 1 using the prefrontal-amygdala circuit, because its artificial synaptic modulation has been especially difficult to achieve, while the need for such modulation is high given the role of this circuit in the behavioral traits relevant to mental disease. In Aim 2, we will test predictions that synaptic efficacy in the dmPFC-BLA loop determines oscillatory synchronization between the two structures and influences anxiety- like behaviors. These predictions are based on findings that theta oscillations synchrony between BLA and dmPFC increase with innate anxiety in the open field, and photostimulation of BLA axonal terminals in dmPFC acutely increase anxiety-like behaviors in the elevated plus maze and open field. The study is expected to produce techniques for obtaining LTP of a desirable magnitude in glutamatergic synapses connecting principal neurons of dmPFC and BLA. The classes of GABAergic neurons that gate LTP in these pathways will be identified, and methods for their transient suppression to aid LTP induction will be developed. This LTP optimization process will provide a template for developing analogous LTP protocols for other brain areas. The role of synaptic efficacy of the dmPFC-BLA reciprocal projections in oscillatory synchronization and anxiety-relevant traits will be determined, which will inform about potential methods for targeted manipulation of that pathway in emotional disorders.

Project summary Forming stronger aversive memories is a characteristic of PTSD. Meanwhile, psychological trauma is a risk factor for developing PTSD in the future, upon exposure to another adverse event. This project will use mice to investigate how observing fear in others, as a form of social distress, enhances the retention of new inhibitory avoidance (IA) memories. It serves our long- term goal to understand how neuronal plasticity contributes to emotional behaviors and to identify the means for reversing PTSD-relevant behavioral traits by artificial circuit manipulations. We have recently found that a brief exposure to a conspecific receiving electrical footshocks, the observational fear paradigm (OF), enables a stronger inhibitory avoidance (IA) learning in mice. Our preliminary data strongly implicate the pathways between the dorsomedial prefrontal cortex (dmPFC) and basolateral amygdala (BLA) pathway in the enhancement. First, a pharmacogenetic disconnection of these structures during OF prevented the enhancement. Second, OF enabled facilitation of this pathway by IA training, which lasted for several hours. Third, OF generated NMDAR-only (silent) synapses, which we unsilenced by IA training. In addition, OF attenuated GABAbR-mediated depression of the feedforward GABAergic currents, evoked in BLA neurons by a 5 Hz repeated stimulation of the dmPFC inputs. We will test a hypothesis that OF enhances IA by generating silent synapses and by altering the GABAbR- dependent inhibitory balance between PV- and Sst-IN in the prefrontal-amygdala circuit, both of which enable a stronger synaptic facilitation during IA training. We will determine the necessity of dmPFC-BLA synapses at each phase of the OF-IA paradigm (Aim 1), test the causal role of the silent synapses and the transient synaptic facilitation (Aim 2), and identify the micro-circuit mechanism responsible for the abnormally high plasticity (Aim 3). The study may inform about potential targets and methods for early intervention to prevent PTSD in traumatized people.