John H. Byrne - US grants

DESCRIPTION (provided by applicant): Memory deficits affect a substantial portion of the population. Thus, a key goal of research on memory mechanisms is to develop strategies for improving long-term memory (LTM). Recently we developed a strategy for improving LTM, which relies on 'tuning' training protocols to match the dynamics of the underlying signaling pathways. Computational and empirical studies focused on the molecular pathways involved in long- term synaptic facilitation (LTF; a cellular correlate of LTM). A model of two signaling pathways that are critical for LTF (the cAMP/PKA and Raf/MEK/ERK pathways) was developed. Using simulations, a novel training protocol (the 'Enhanced' protocol) was tuned to match the dynamics of these two pathways and improve the simulated LTF. Empirical studies revealed that the Enhanced protocol indeed increased LTF in sensorimotor cultures, increased phosphorylation of CREB1 (a transcription factor that is critical for LTM formation), and improved LTM following behavioral training. The goals of the current proposal are to investigate the ways in which the Enhanced protocol affects downstream components of the molecular network underlying LTF and to apply this new strategy to restoring memory deficits (via novel training protocols and/or pharmacological interventions) that are induced by molecular lesions. The Specific Aims are: 1) Quantify activation of the cAMP/PKA and Raf/MEK/ERK pathways and the transcription factors CREB1, CREB2, and C/EBP in response to the Enhanced protocol. More detailed information about the dynamics of these pathways and their responses to training can further improve the predictive power of the computational model, and with an improved model, more complex features of LTM can be studied (see Aim 3). 2) Examine whether computationally designed pharmacological manipulations of the PKA and ERK pathways can increase LTF and long-term excitability (LTE, another cellular correlate of LTM). In principle, memory can be improved by administering drugs alone. However, drugs often have unacceptable side effects. Aim 2 will examine whether combinations of lower, less toxic doses of drugs can improve memory. 3) Determine whether computationally predicted training protocols and pharmacological regimens can restore impaired LTF and LTE. Memory impairment is often the result of mutations/deletions in genes controlling signaling pathways involved in LTM. Using RNAi technology, Aim 3 will examine whether memory impairment induced by molecular lesions can be overcome by computer-designed training protocols and/or pharmacological interventions. Although these studies will utilize a simple model system and the molecular defects will be acute rather than congenital, the results will provide important insights into the interactions among training protocols, pharmacological treatments and molecular defects. The success of these studies will help the development of a new paradigm for cognitive enhancement that can have broad generality, including improving human health.

The research program is focused on the neural and molecular mechanisms underlying information storage. Previously Dr. Byrne found that a classical conditioning protocol applied directly to individual tail sensory neurons results in an associative modification of the monosynaptic connections to motor neurons. Sensory neurons receiving a conditioned stimulus (CS, intracellular activation of the sensory neuron) immediately before the unconditioned stimulus (US, tail shock) show significantly more synaptic facilitation than sensory neurons exposed to the US alone or to unpaired CS and US applications. An analog of the classical conditioning paradigm produces a selective amplification of the cAMP content of isolated sensory neuron clusters. These results indicate that a pairing- specific enhancement of cAMP levels may be a biochemical mechanism for associative learning. Experiments proposed here are designed to extend these analyses. Specifically, Dr. Byrne will examine 1) whether Ca2+ serves as the signal for the induction of the associative change, 2) whether cAMP levels in the sensory neurons are increased during simple forms of learning such as sensitization and classical conditioning, 3) the contribution of spike broadening in the sensory neurons to synaptic facilitation and behavioral modification, 4) the properties of the neural circuit elements mediating the reflex and effects of the US, and 5) the coordination and modulation of synergistic defensive responses triggered by tail stimulation. Continued analysis of this system promises to yield much additional information concerning the cellular mechanisms underlying associative and nonassociative learning. The award will foster Dr. Byrne's research development and professional growth. It will allow him to expand his research program by hiring a junior colleague. It will give Dr. Byrne the flexibility to plan interludes to develop new areas of expertise through collaboration and visits with other scientists in the field.

The overall objective of the proposed research is to provide insights into one of the most fundamental problems in the neurosciences - the physiological basis of learning and memory. There are three broad aspects of this topic that will be investigated. The first aim is to continue to elucidate the mechanisms of synaptic plasticity that underlies short-term sensitization. Of particular interest will be the interaction of multiple second messenger pathways. Specific objectives of this first aim include: 1) identify the second messenger(s) that mediates the 5-HT-induced modulation of the voltage-dependent delayed K+ channel; 2) Determine the contribution of CaMKII to FMRFamide-induced spike narrowing; 3) identify the second messenger mediating the inhibitory effect of FMRFamide on mobilization and characterize its role in sensitization; and 4) Characterize the currents modulated by a sensitizing stimulus. A second major goal is to investigate the unique mechanisms underlying long-term sensitization from two perspectives: its induction and its expression. Specific objectives of this second and third aim include: 1) Determine the role of cAMP in the induction of long-term synaptic plasticity of sensorimotor connections; 2) Determine the time window of the requirement for gene transcription and protein translation; 3) Examine the time course of long-term plasticity at points longer than 24 hr after training; 4) Determine the role of apTBL-1 in long-term synaptic plasticity; 5) Determine the role of TGF-beta and other growth factors in long-term facilitation. The final aim of the proposal examines distributed representations of learning and memory and seeks to determine whether mechanisms for induction, maintenance and expression are shared among different sites. Although a great deal is known about plasticity at the sensorimotor synapse, modification of other sites in the circuit will be investigated in order to understand the full expression of the behavioral modification. Specific objectives of this aim include: 1) Determine the contribution of interneurons to short-term sensitization; 2) Determine the contribution of interneurons to long-term sensitization; 3) Examine the effects of modulatory transmitters; and 4) Examine the ionic mechanisms and possible second messengers involved in the short-and long-term changes in the properties of the interneurons and motor neurons.

A fundamental problem in neuroscience is to understand events occurring within Individual neurons and within neural networks that contribute to forms of plasticity underlying learning and memory. This proposal outlines both empirical and modeling studies that will examine the molecular, biochemical and biophysical properties of Identified neurons and the connectivity of neural circuits that have demonstrated capacities for nonassociative and associative plasticity. Specifically, the neural circuit that mediates the tail withdrawal reflex will be analyzed. Many of the sensory neurons, Interneurons, motor neurons and modulatory interneurons that control this behavior have been Identified and are accessible to study. Thus, molecular, biochemical and cellular neurophysiological techniques will be applied to analyze the particular processes that might explain associative and nonassociative learning. Formalisms of the cellular and network processes that underlie these forms of plasticity will be developed and incorporated into quantitative, real-time models of neuron-like elements and neural networks. The ability of these models to fit the experimental data and to predict simple and complex features of learning will be examined. The proposed research will provide for a fairly complete analysis of the mechanisms underlying the Induction, expression and maintenance of simple forms of nonassociative and associative learning as well as help address fundamental questions regarding the mechanistic relationship between short- and long-term memories.

DESCRIPTION (Adapted from applicants abstract): The overall goal of this project is to investigate the cellular processes that underlie two forms of associate learning - classical conditioning and operant conditioning. Studies will focus on consummatory feeding behavior of Aplysia. This behavior can be modified by both classical and operant conditioning. Moreover, many key elements of the neural circuitry have been identified and this circuitry retains many of its functional characteristics in vitro. The project has four specific aims: First, investigate cellular mechanisms underlying an in vitro analogue of operant conditioning. As a first step toward a cellular analysis of operant conditioning, we developed a preparation that retained the essential features of operant conditioning in vitro and was amenable to cellular analyses. The first aim is to identify and characterize the neuronal plasticities underlying this analogue of operant conditioning; identify cells and/or transmitters mediating the reinforcement; and investigate the cellular mechanisms encoding the temporal specificity of contingency-dependent neuronal plasticity. Second, develop an in vitro analogue of classical conditioning. To facilitate the cellular analysis of classical conditioning, we will develop an in vitro preparation that retains the essential features of classical conditioning and that is amenable to cellular analyses. The second aim is to characterize pathways mediating the reinforcement; to establish a stimulation protocol that mimics classical conditioning; and investigate cellular mechanisms underlying this in vitro analogue of associative learning. Third, identify neuronal correlates of classical and operant conditioning. Although the in vitro analogues may retain the essential features of associative learning, these reduced preparations are removed from the behaving animal. Thus, it is critical to explore loci and changes in the nervous systems of animals that have been behaviorally trained. The third aim is to determine whether the cellular loci that are modified by the in vitro analogues of associative learning (both classical and operant conditioning) are also modified by behavioral training; and to determine whether behavioral conditioning involves additional cellular loci. Fourth, compare cellular processes mediating short- (1 hr), intermediate- (~4 hr) and long-term memory (24 hr). The forth aim is to characterize and compare different temporal domains of memory that are induced by classical and operant condition; identify neuronal correlates for each phase of memory; and for each domain of memory, develop an in vitro analogue that will be amenable to cellular analyses. The proposed studies will provide the first major insights into the cellular mechanisms that underlie operant conditioning, as well as provide an opportunity for a mechanistic comparison of classical and operant conditioning.

The main function of the nervous system is to process information in ways that lead to adaptive behavior, and to accomplish this, the excitability of neurons and the strength of their synaptic connections need to be modulated continually. After a neuron or neural system has been analyzed extensively, it becomes possible to ask what information it carries and how it contributes to this plasticity. At this point, there is generally so much data that only computational approaches can explain how individual components of a system interact, however. This Program Project will apply constructionistic computational techniques to several such well- characterized neural systems to achieve a more complete understanding of neuronal information processing and plasticity. The Project will examine multiple levels of organization, ranging from genetic networks within neurons to neural circuit. The individual projects will examine: 1) the dynamic properties and interactions of gene networks and excitable membranes; 2) the contribution of plasticity in individual neurons to associative learning; 3) the computational role of cellular and synaptic plasticity in an oscillatory neural circuit; and 4) the role of dopamine in light and dark adaptation in the primate retina. The individual projects are linked by a common goal of investigating plasticity in neurons and determining its contributions to higher levels of processing. For example, simulation of simple forms of cellular and synaptic plasticity may provide insights into the roles of these distinct mechanisms in the information processing capabilities of larger-scale neural networks such as those controlling feeding behavior. The group will be supported by a Computational Core that serves as a resource for developing models and for the exchange of information among the project groups. Another important goal of the project is to train graduate students and postdoctoral fellows in Computational Neuroscience. Finally, the Projects will further develop general-purpose simulation programs for neuronal and biochemical modeling, which will be used by the Program Project group. These programs will also be widely distributed to other groups who wish to apply computational approaches to analyze the properties of nerve cells and neuronal networks.

The main function of the nervous system is to process information in ways that lead to adaptive behavior, and to accomplish this, the excitability of neurons and the strength of their synaptic connections need to be modulated continual. After a neuron of neural system has been analyzed for years, it becomes possible to ask what information it carries and how it contributes to this plasticity. At this point, computational approaches can greatly assist integrating accumulated data to explain how different components of a system interact. This proposal will, via computation supplemented with experiments, improve the understanding of the dynamics of a key genetic regulatory system necessary for neuronal plasticity, and of the reciprocal interactions that could be expected between such genetic systems and membrane currents. Two distinct levels of organization will be modeled. At the molecular level, a detailed model will be developed for the genetic regulatory system that utilizes CREB and related transcription factors. This system is known from experiments with mammals and invertebrates to be important for synaptic plasticity and long-term memory formation. At the level of the bioelectrical properties of a single neuron, there is abundant experimental evidence for regulation of ion channel densities by electrical activity. The extensively characterized neuron R15 of Aplysia will serve as a model with which to computationally investigate the consequences of this feedback. A conductance-based model of R15 will be improved by adding coupling terms to the CREB genetic model to provide a plausible description of the effects of gene expression on electrical behavior and to describe feedback from electrical activity, via calcium influx, to gene expression. With this combined model, and parameter values derived from experiment or from the literature, we will also investigate whether known kinetic properties of CREB regulation could provide a mechanism for optimal transcription at specific stimulus frequencies-such a mechanism Could help explain experiments that have demonstrated optimal stimulus frequencies for long-term memory formation in invertebrates. Finally, focusing on these systems is expected to further the aim of our Preliminary Studies-to determine how specific observed behaviors of genes arise from general organizational principles of genetic regulatory systems.

DESCRIPTION (Adapted from applicants abstract): The overall goal of this project is to investigate the cellular processes that underlie two forms of associate learning - classical conditioning and operant conditioning. Studies will focus on consummatory feeding behavior of Aplysia. This behavior can be modified by both classical and operant conditioning. Moreover, many key elements of the neural circuitry have been identified and this circuitry retains many of its functional characteristics in vitro. The project has four specific aims: First, investigate cellular mechanisms underlying an in vitro analogue of operant conditioning. As a first step toward a cellular analysis of operant conditioning, we developed a preparation that retained the essential features of operant conditioning in vitro and was amenable to cellular analyses. The first aim is to identify and characterize the neuronal plasticities underlying this analogue of operant conditioning; identify cells and/or transmitters mediating the reinforcement; and investigate the cellular mechanisms encoding the temporal specificity of contingency-dependent neuronal plasticity. Second, develop an in vitro analogue of classical conditioning. To facilitate the cellular analysis of classical conditioning, we will develop an in vitro preparation that retains the essential features of classical conditioning and that is amenable to cellular analyses. The second aim is to characterize pathways mediating the reinforcement; to establish a stimulation protocol that mimics classical conditioning; and investigate cellular mechanisms underlying this in vitro analogue of associative learning. Third, identify neuronal correlates of classical and operant conditioning. Although the in vitro analogues may retain the essential features of associative learning, these reduced preparations are removed from the behaving animal. Thus, it is critical to explore loci and changes in the nervous systems of animals that have been behaviorally trained. The third aim is to determine whether the cellular loci that are modified by the in vitro analogues of associative learning (both classical and operant conditioning) are also modified by behavioral training; and to determine whether behavioral conditioning involves additional cellular loci. Fourth, compare cellular processes mediating short- (1 hr), intermediate- (~4 hr) and long-term memory (24 hr). The forth aim is to characterize and compare different temporal domains of memory that are induced by classical and operant condition; identify neuronal correlates for each phase of memory; and for each domain of memory, develop an in vitro analogue that will be amenable to cellular analyses. The proposed studies will provide the first major insights into the cellular mechanisms that underlie operant conditioning, as well as provide an opportunity for a mechanistic comparison of classical and operant conditioning.

[unreadable] DESCRIPTION (provided by applicant): This project wilt develop a framework based on mathematical modeling that a) describes the mechanism by which synaptic plasticity emerges from molecular processes regulating gene transcription, and b) tests mechanistic hypotheses, such asproposed roles of specific protein kinases. The project builds upon our previous model describing aspects of the gene and protein network responsible for long-term synaptic facilitation (LTF) and the formation of longterm memory (LTM) in the mollusk Aplysia. This model is based on transcriptional regulation by Ca2*/cAMP response element - binding protein (CREB, termed ApCREB1 in Aplysia) and related transcription factors. We will extend this model to incorporate additional elements of gene regulation recently demonstrated to be essential for LTF. In addition, we will develop an analogous model to simulate biochemical events underlying the induction of long-term synaptic potentiation (LTP) in vertebrates. Both LTF and LTP are thought to play essential roles in the formation of LTM, and the LTF induction and LTP induction exhibit mechanistic similarities, such as dependence on MAP kinase activation. Therefore, a modeling framework that can simulate aspects of both LTF and LTP induction is likely to significantly increase the understanding of learning mechanisms. [unreadable] [unreadable] The LTF model variant will incorporate additional transcriptional regulators essential for LTF, such as ApCREB2, and ApC/EBP. Bifurcation analysis and pre-programmed integrations will identify key control parameters which are plausible sites of physiological regulation and which, when varied, have important effects on the dynamics of the model. We will then use the model to simulate the results of experimental protocols in which alterations are made in the activity of the transcriptional regulators listed above. A minimal set of variations in key control parameters will be identified that allows simulation of data from these protocols. This approach is likely to help identify the key mechanisms that determine the amount of LTF induced by different training protocols. The LTP model variant will be used to simulate three common stimulus protocols that induce hippocampal late LTP. These protocols are high-frequency (tetanic) stimulation, theta-burst stimulation, and stimulation by forskolin. Parameters will be optimized to fit experimental time courses of nuclear [Ca 2+] and of kinase and transcription factor activities. The model will then be used to test the hypothesis that CREB kinases other than protein kinase A, such as ribosomal $6 kinase 2, are primarily responsible for CREB phosphorylation and LTP induction. [unreadable] [unreadable] [unreadable]

This project will develop a framework based on mathematical modeling that a) describes the mechanism by which synaptic plasticity emerges from molecular processes regulating gene transcription, and b) tests mechanistic hypotheses, such as proposed roles of specific protein kinases. The project builds upon our previous model describing aspects of the gene and protein network responsible for long-term synaptic facilitation (LTF) and the formation of long-term memory (LTM) in the mollusk Aplysia. This model is based on transcriptional regulation by Ca2+/cAMP response element - binding protein (CREB, termed ApCREBI in Aplysia) and related transcription factors. We will extend this model to incorporate additional elements of gene regulation recently demonstrated to be essential for LTF. In addition, we will develop an analogous model to simulate biochemical events underlying the induction of late long-term synaptic potentiation (L-LTP) in vertebrates. Both LTF and L-LTP are thought to play essential roles in the formation of LTM, and LTF induction and L-LTP induction exhibit mechanistic similarities, such as dependence on MAP kinase activation. Therefore, a modeling framework that can simulate aspects of both LTF and L-LTP induction is likely to significantly increase the understanding of learning mechanisms. The LTF model variant will incorporate additional transcriptional regulators essential for LTF, such as ApCREB2, and ApC/EBP. Bifurcation analysis and pre-programmed integrations will identify key control parameters which are plausible sites of physiological regulation and which, when varied, have important effects on the dynamics of the model. We will then use the model to simulate the results of experimental protocols in which alterations are made in the activity of the transcriptional regulators listed above. A minimal set of variations in key control parameters will be identified that allows simulation of data from these protocols. This approach is likely to help identify the key mechanisms that determine the amount of LTF induced by different training protocols. The L-LTP model variant will be used to simulate three common stimulus protocols that induce hippocampal L-LTP. These protocols are high-frequency (tetanic) stimulation, theta-burst stimulation, and stimulation by forskolin. Parameters will be optimized to fit experimental time courses of nuclear [Ca2+] and of kinase and transcription factor activities. The model will then be used to test the hypothesis that CREB kinases other than protein kinase A, such as ribosomal S6 kinase 2, are primarily responsible for CREB phosphorylation and LTP induction. Preliminary model development and simulations predict that L-LTP induction by a low-frequency burst stimulus protocol does not depend on nuclear CaM kinase activation and consequent CREB phosphorylation.

DESCRIPTION (provided by applicant): The main function of the nervous system is to process information in ways that lead to adaptive behavior. Two different approaches, one theoretical and the other empirical, are being used to explore the role of neuronal plasticity in development, learning, memory, information processing, and other complex brain functions. The theoretical approach simulates and synthesizing brain function with mathematical models based on known and hypothesized principles of neural function. The empirical approach delineates the complex biochemical and biophysical properties of neurons, the rules that determine their connectivity, and the mechanisms through which their properties and connections are modified during development and learning. Although these two approaches have traditionally been used independently, there is a growing realization among neurobiologists, psychologists, and adaptive systems theorists that progress in understanding the brain is dependent on a combination of both approaches. In addition, in many cases, the knowledge of systems has matured to the point where there is not only a sufficient body of information to warrant a computational approach, but further progress in the understanding of the system requires it. The overall goal of the Program Project is to use computational approaches to examine neuronal plasticity at multiple levels of organization, ranging from molecular dynamics within subcellular neuronal compartments, to genetic networks within neurons, to neural network mechanisms. The individual Projects are linked by the common goal of investigating plasticity in neurons in the hippocampus and related structures and determining its contributions to higher levels of processing. The individual Projects will examine: 1) the dynamical properties of gene networks underlying plasticity; 2) the quantitative behavior of the postsynaptic Ca2+/calmodulin signaling pathway that plays an essential role in neuronal plasticity; 3) dynamics of synaptic plasticity at the molecular level and its importance as a substrate for plasticity in the hippocampus; and 4) the neural network mechanisms by which the hippocampus constructs high-order cognitive representations from multimodal inputs. In addition, the individual Projects will be supported by a Computational Core Facility that will serve as a resource for developing computational models and for the exchange of information among the projects.

[unreadable] DESCRIPTION (provided by applicant): This revised application seeks support under the NCRR Shared Instrumentation Grant (SIG) program to update the department's multi-user confocal microscopy imaging facility. This facility currently houses a Bio-Rad 1024 MP scanhead on a fixed stage upright microscope (Olympus BX50) with a Kr/Ar laser (Bio-Rad) used for conventional single-photon confocal microscopy. In 2004, Carl Zeiss acquired the Cell Science division of Bio-Rad, and as a result, support for the hardware and software on our current system is being phased out. By 2007, there will be no service contract and the availability of parts is not guaranteed. Moreover, we have been unable to identify a third party vendor who could offer service and maintenance. Because user time on other confocal imaging systems within the University is either severely limited or unavailable, the acquisition of a new, fully supported confocal imaging system is critical for continued progress on the imaging aspects of the NIH funded research projects in the Department. We identified the Zeiss LSM 510 as an appropriate replacement because it offers state-of- the-art performance and sufficient versatility to meet the needs of multiple users. Finally, Zeiss is offering a substantial discount to encourage Bio-RAD users to migrate to the Zeiss instrument, and we do not know how long this policy will remain in effect. The Department of Neurobiology and Anatomy has committed significant space and support to the confocal imaging suite. In addition, the University has committed funds towards the maintenance contract of the instrument. The current system is housed in a 245 sq ft. room that has been extensively renovated to meet the needs of the instrument. The ventilation system was enhanced to provide constant temperature and low humidity. The electrical system was enhanced to isolate sensitive instrumentation from power supplies that produce line spikes. An adjacent room was remodeled to provide a convenient location for off-line analysis of confocal data. In addition, the Department has an established organizational structure and user-sharing system, which will be continued. Thus, institutional support for the confocal imaging core is already in place. Relevance: The focus of the user group is on signaling cascades and changes in morphology in individual neurons. Confocal microscopy is an essential tool for the investigation of these fundamentally important processes in neurobiology. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Learning and memory are cornerstones of human cognition, and cognitive defects are associated with many brain disorders. It has recently become possible to relate cognitive function to specific molecules. These molecules include the CREB family of transcription factors, which are essential for long-term (LT) synaptic plasticity and memory. Alterations in CREB signaling pathways are associated with diseases that impair cognition, such as Rubinstein-Taybi syndrome, neurofibromatosis, and Coffin-Lowry syndrome. Many of the molecular details of these signaling pathways are known. However, the ways in which these elements quantitatively account for normal and pathological cellular behavior are not well understood because the signaling cascades are embedded in a biochemical and genetic network that includes extensive cross talk and negative and positive feedback loops. To address this issue, the present proposal outlines computational studies that model and simulate CREB signaling pathways and their role in memory. Two well characterized neuronal correlates of memory will be modeled: long-term facilitation (LTF) and long-term potentiation (LTP). The proposed models will use differential equations to simulate molecular processes and will be constrained by empirical data. Aim 1 will test the hypothesis that the dynamics for the induction and consolidation of LTF are governed by the dynamics of the PKA and ERK kinase cascades and by feedback loops within CREB regulated transcription. Simulations will examine the efficacy of training protocols and predict protocols that optimize learning. Aim 2 will test the hypothesis that LTF and LTP share molecular mechanisms and dynamics. Simulations will identify control parameters, which may correspond to pharmacological control points for enhancing learning and cognition. Simulations also will explore LT plasticity impairment due to mutations that affect CREB activity, such as Rubinstein-Taybi syndrome. Finally, the models will be used to predict treatments for ameliorating CREB-related memory deficits and thereby help restore normal plasticity, learning and memory. PUBLIC HEALTH RELEVANCE: Learning and memory are essential to human cognition, and their disruptions contribute to several brain diseases including neurofibromatosis, Rubinstein-Taybi syndrome, and Coffin-Lowry syndrome. This project will advance the understanding of basic memory mechanisms, which will lead to better learning paradigms and help identify molecular targets for pharmacological treatments of brain disorders.

DESCRIPTION (provided by applicant): The long range aim of this research project is to identify the molecular mechanisms underlying the formation of associations in the nervous system. The studies proposed in this grant renewal will provide insights into general principles underlying memory formation following one-trial and multi-trial Pavlovian conditioning, and contribute to the elucidation of events that are essential for the transformation of memory into an enduring form. To pursue these goals we are conducting a proteomic analysis of Pavlovian conditioning in the marine mollusk Hermissenda, a preparation that has been used extensively for over 30 years in biochemical, biophysical, and molecular studies of associative learning. The proposed research will use a combination of proteomic techniques and mass spectrometry to identify proteins whose abundance is regulated by conditioning followed by cloning full length cDNA's based upon identified peptide sequences. Four specific aims form the basis for this renewal application. The first will involve proteomic profiling to examine differences in protein abundance between conditioned groups and unpaired controls at different times following one-trial in vitro conditioning using difference gel electrophoresis (DIGE) and Cy Dye labeling of proteins in lysate samples. Mass spectrometry (MS) will be used to provide identification of proteins that exhibit statistically significant differences in protein abundance. We will also examine long-term changes in protein abundance supporting the maintenance of memory produced by multi-trial Pavlovian conditioning. The second aim will involve detecting changes in protein abundance regulated by one-trial in vitro conditioning in two types of identified neurons shown previously to express intrinsic cellular and synaptic plasticity. Isolated eyes and identified type I interneurons will receive in vitro conditioning followed by DIGE and gel image analysis to identify proteins regulated by one-trial conditioning. The proposed proteomic analysis of identified neuron types will be less likely to produce an overall average result that might mask subtle, localized changes since there will be a homogeneity of cell types studied following conditioning. The third aim will be addressed by the application of quantitative double-label autoradiography and phosphor imaging to identify proteins whose synthesis is regulated by one trial-conditioning at different time points post-conditioning. We will identify proteins in the 2D gel field of all 35S-labeled proteins that are phosphoproteins by incubating lysate samples of conditioned and control nervous systems with both 35S-methionine and 32P-orthophosphate, and imaging the same gel twice, before and after the 35S signal is blocked by a filter. The final aim will assess the role of candidate proteins in memory formation by blocking protein expression with RNAi techniques. Based upon pilot data we will also examine the role of tropomyosin and gelsolin in memory formation. These studies in conjunction with MS analysis will provide evidence for both de novo synthesis and post-translational modifications of identified proteins contributing to the induction and maintenance of long-term memory.

Molecular processes that underlie the induction and consolidation of long-term memory (LTM) are the subjects of intensive research, and studies are providing a wealth of empirical data that relate aspects of memory to specific intracellular signaling pathways. For example, empirical studies are elucidating the roles played by extracellular factors (e.g. growth factors), kinase activity, and transcriptional regulation in induction and consolidation of memory. Due in part to the complexity and nonlinear features of these molecular pathways, it is difficult to develop an intuitive understanding of the ways in which these pathways respond to stimulus protocols or pharmacological manipulations or are affected by single-site molecular lesions. To provide a better understanding of the processes underlying LTM, the present proposal will develop quantitative models of the molecular pathways that underlie two well-characterized models of LTM: i) long-term synaptic facilitation (LTF) and ii) long-term synaptic potentiation (LTP). Parameters will be constrained by empirical data. Parameter sensitivity analysis and a novel cluster analysis will assess model robustness. Aim 1 will extend our model for LTF, which describes the regulation of transcription by PKA and ERK via phosphorylation of the transcription factors CREB1 and CREB2. The extended model will include components of additional intra- and extracellular feedback loops (e.g., TGF?, and ApNT), an additional transcription factor (C/EBP), ribosomal s6 kinase (RSK) and p38 MAP kinase. In Aim 2, this model will be used to predict stimulus protocols, as well as pharmacological treatments, that enhance LTF and that rescue impaired LTF. Aim 3 will extend our current model of LTP, which describes roles of several kinase pathways (e.g., MAPK, PKA, PKC, and CAMKII) and histone acetylation. The model will incorporate a recently delineated BDNF positive-feedback loop, which leads to activation of ERK, phosphorylation of CREB1, and induction of transcription necessary for the consolidation of LTP. We will simulate stimulus protocols and drug effects to predict treatments that could rescue impaired memory mechanisms in Rett syndrome, which is caused by mutations that alter the activity of the transcription factor MeCP2, and that can rescue impaired mechanisms in Rubinstein-Taybi syndrome, which is caused by mutations in CREB binding protein. This proposed approach of using models to predict novel learning paradigms and/or drug treatments that restore normal plasticity, is an innovative methodology that could ultimately lead to the development of new strategies for the treatment of cognitive disorders.

PROJECT SUMMARY/ABSTRACT Although significant advances have been made in elucidating the cellular, biophysical and molecular mechanisms of learning and memory, much less is known about the ways in which mnemonic processes are embedded in neuronal networks, thereby storing and expressing a memory via changes in neural activity. The overall goal of this proposal is to provide insights into the design principles that govern the implementation of memories within the complex environment of a neural circuit. Studies will focus on an established in vitro analogue of operant conditioning (OC) in a relatively complex neural circuit, which is amenable to cellular and biophysical analyses. A combination of intracellular electrophysiological recording techniques and voltage- sensitive dye (VSD) recordings will locate and analyze loci of non-synaptic plasticity and synaptic plasticity. In addition, the project will examine the extent to which short- and long-term memory share plasticity loci. Aim 1 will use intracellular recording techniques to examine loci of OC-induced plasticity. Previous studies of OC in this model system focused primarily on non-synaptic plasticity mechanisms. However, preliminary data indicate that OC also modifies the strength of several synaptic connections in the network. Therefore, Aim 1 will examine OC-induced synaptic and non-synaptic plasticity. Aim 2 will use a combination of intracellular and VSD recordings to identify additional sites of OC-induced plasticity. To date, published studies have examined only five of the ~100 neurons and none of the hundreds of synaptic connections that comprise the neural circuit. To address this shortcoming, large-scale VSD recordings, in combination with intracellular recordings, will be used to identify OC-induced changes in activity and synaptic properties in a substantial proportion of the neurons in the circuit. Aim 3 will determine the extent to which short- and long-term memory share common loci and plasticity mechanisms. Our previous studies indicate that at least one locus of plasticity is common to both short- and long-term memory. Thus, an important question in memory research is to determine the extent to which sites for short-term memory are also sites for long-term memory, or conversely, which sites of plasticity may be unique to long-term memory. By examining these three aims, the project will provide insights into the ways in which the many components of a nervous system orchestrate learning and generate behavior.