Stephen Maren, PhD - US grants

Learned relationships between environmental stimuli (contexts) and reinforcers are important in a number of clinical syndromes. For instance, contextual fear conditioning in humans with panic disorder promotes avoidance of contextual stimuli (typically places) associated with panic attacks. Recent studies of contextual fear conditioning in rats have provided evidence that the hippocampal formation, a structure known to be important in human memory, is critical for this form of learning. For example, electrolytic lesions made in the dorsal hippocampus (DH) before contextual fear conditioning attenuate the subsequent acquisition of conditional freezing (a fear response in the rat characterized by immobility) to contextual stimuli paired with footshock. Similarly, electrolytic lesions of the entorhinal cortex (EC), the primary cortical afferent of the DH, produce deficits in the acquisition of conditional freezing. Given that the amygdala is essential for fear conditioning to both contextual and discrete stimuli, these data suggest that anatomical projections from the EC to the DH are part of a pathway by which contextual information reaches the amygdala for association with footshock. Although the involvement of the DH in contextual fear conditioning has been established, the route by which information processed in the DH reaches the amygdala is unknown. One possibility is that projections from the ventral subiculum (vSUB) to the amygdala constitute this pathway, however this hypothesis has yet to be tested. Therefore, the objective of the present project is to explore the impact of neurotoxic vSUB lesions on contextual fear conditioning in rats, and to examine the hypothesis that direct projections from the vSUB to the amygdala mediate the acquisition of contextual fear conditioning. The project will consist of three experiments. The first experiment will examine the acquisition and retention of contextual fear conditioning in rats with neurotoxic vSUB lesions; rats with neurotoxic EC or DH lesions will be included for comparison. The second experiment will anatomically disconnect the vSUB and amygdala to assess the contribution of this pathway to contextual fear conditioning. The third experiment will assess the impact of posttraining neurotoxic vSUB lesions on the expression of contextual fear. These experiments will provide important new information concerning the role of the vSUB and the vSUB-BLA projection in contextual fear conditioning in rats, and will provide further insight into the role of the hippocampal formation in learning and memory.

DESCRIPTION (provided by applicant): Pavlovian fear conditioning is a ubiquitous form of learning that may underlie disorders of fear and anxiety in humans. Many studies indicate that nuclei in the amygdala including the basolateral complex (BLA) and the central nucleus (CEA) are required for both the acquisition and expression of fear conditioning in mammals. Notably, neurons in both areas exhibit learning-related increases in spike firing during fear conditioning and pharmacological inhibition of both regions reduces fear conditioning. This suggests a role for BLA and CEA neurons in elaborating memories of aversive experiences, although the precise function of these areas in Pavlovian fear conditioning remains unclear. In the present proposal, it is hypothesized that BLA and CEA neurons encode distinct associative representations during fear conditioning. More specifically, we suggest that BLA neurons encode CS-US (S-S) representations that are sensitive to the current value of the US, whereas CEA neurons encode CS-CR (S-R) associations that are insensitive to US value. Moreover, we hypothesize that the relative contribution of these associations, and hence the BLA and CEA, to conditional fear varies as a function of training. To exam these hypotheses, we will use single-unit recordings, reversible pharmacological lesions and sophisticated behavioral designs to isolate the function of amygdala neurons in fear memory processes. The first specific aim will use inflation and devaluation procedures to examine the role of BLA and CEA neurons in representing US "value", which will be indexed by measuring conditional freezing to an auditory CS that has been paired with an aversive US. The second aim will examine the associative basis (S-S or S-R) for conditional fear after overtraining. The third aim will examine whether the role for synaptic plasticity in the BLA and CEA changes as a function of training. Collectively, these experiments represent the first systematic attempt to define parallel, associative functions of BLA and CEA neurons in aversive conditioning. The outcome of these experiments will have an important impact on current conceptions of both amygdaloid function and on clinical disorders of fear that depend on emotional learning and memory.

DESCRIPTION (provided by applicant): One of the goals of modern neuroscience is to understand the nature and mechanisms underlying complex cognitive phenomena, including learning and memory. This effort is particularly germane to understanding disorders of memory, including Alzheimer's disease. In this regard, considerable effort has been directed at understanding the mechanisms whereby memories are encoded and stored in the brain. We have focused on the analysis of a simple form of associative learning in rats known as Pavlovian fear conditioning. In this type of learning, rats learn to fear a conditional stimulus (CS), such as a tone, that has been paired with an aversive unconditional stimulus (US), such as a foot shock. Fear is expressed by a number of responses, including freezing behavior. Considerable progress has been made in elucidating the neural mechanisms of fear conditioning. For example, it is clear that the amygdala plays an essential role in encoding and storing CS-US associations during conditioning, and the synaptic and cellular mechanisms underlying this process are beginning to be understood. Although the neural mechanisms for CS-US association in fear conditioning are well understood, less is known concerning higher-order memory processes in Pavlovian learning paradigms. For example, memory retrieval is not only related to the strength of the CS-US association, but also to the similarity of the retrieval context to the context of memory encoding. Moreover, stimuli often have ambiguous meanings and context is essential for disambiguating these meanings during memory retrieval. The context specificity of memory relies on two interrelated processes that we term: contextual encoding and contextual retrieval. Contextual encoding indexes memory to the context in which it is encoded, and this enables contextual retrieval to occur in the encoding context. An extensive body of literature indicates an important role for the hippocampus in contextual learning and memory. We hypothesize that the hippocampus plays a critical role in the encoding and retrieval of context-specific fear memories. To this end, the present proposal has three specific aims: 1) to determine the role of the hippocampus in contextual encoding using reversible brain lesions, 2) to investigate the neural circuit by which the hippocampus mediates contextual retrieval using a combination of reversible brain lesions and single-unit electrophysiology, and 3) to determine the nature and scope of the context-specific memories regulated by hippocampal circuitry. These experiments will provide essential information regarding the neural substrates of contextual memory and have broad implications for our understanding of normal and pathological memory more generally.

DESCRIPTION (provided by applicant): This application seeks continuation of support for nine predoctoral positions on our training grant, 'Early Stage Training in the Neurosciences'. This training grant serves as the centerpiece in an integrated plan of support for pre-candidate students in the Interdepartmental Neuroscience Graduate Program at the University of Michigan. The first two years of this training program ensure that all students graduate as broadly trained neuroscientists. In the first year, students take an intensive lab course, an interdisciplinary course in the principles of neuroscience, human neuroanatomy, statistics, and a course in research responsibility and ethics, in addition to performing two laboratory research rotations. During their second year, students take elective courses, participate in a seminar course in which they give an oral presentation, complete their lab rotations and begin the initial work on the doctoral thesis. Once students complete pre-candidate training, they perform full-time doctoral research in the laboratory of one of 98 different training faculty members. Approximately half of these faculty members hold appointments in basic science departments, and the other half hold appointments in clinical departments. In both the classroom and laboratory, our students are exposed a broad range of research topics including Cellular and Molecular Neurobiology, Developmental Neurobiology and Regeneration, Clinical and Translational Neuroscience, Sensory and Computational Neuroscience, Behavioral, Affective, and Integrative Neuroscience, and Cognitive Neuroscience and Neuropsychiatry. Upon completion of their training, our graduates are poised to tackle a host of public health issues from the molecular basis of neurodegenerative disorders to brain circuit abnormalities in psychiatric disease. The University of Michigan is proud of its history of recruiting and training underrepresented minority students. The Neuroscience Graduate Program strives to increase the diversity of its trainees, and we describe our accomplishments and a detailed plan for recruiting and retaining underrepresented minorities. We also present detailed plans for the evaluation of the program and for instruction in the responsible conduct of research. This training grant is critical to the success of the Neuroscience Program's training mission. It has provided support for roughly 40% of the students admitted to the program in the last 4 years, and has supported roughly half of the current students in the program at one point in their training. The training grant supports a variety of trainee-related activities including classroom laboratory training, student travel to scientific conferences, and speakers at neuroscience symposia and colloquia. This training grant is key to our recruitment of outstanding students and their educational success. PUBLIC HEALTH RELEVANCE: This is an application to support Early Stage Neuroscience training for students in the University of Michigan Neuroscience Graduate Program. One of the oldest of its kind, this interdisciplinary training program is essential to for the development of the next generation of laboratory scientists capable of using state-of-the-art methodologies to understand the normal and disordered function of the brain and nervous system. Trainees in this program are poised to tackle a host of public health issues from the molecular basis of neurodegenerative disorders to the brain circuit abnormalities in psychiatric disease.

DESCRIPTION (provided by applicant): The long-term goals of this project are to understand the neural systems that mediate the context-dependence of memory, particularly memory for traumatic events. To address these questions, we use auditory fear conditioning to study the acquisition and extinction of aversive memory. Extensive evidence indicates that learned fear generalizes broadly across contexts (both in space and time), whereas memories of extinction are tightly bound to the time and place of extinction. Work in previous years of the project has identified a critical role for the hippocampus in the context-dependence of fear extinction to an auditory conditioned stimulus (CS). Further, hippocampal modulation of neuronal activity in the amygdala appears to be involved in gating the expression of fear memories after they have undergone extinction. This proposal seeks to characterize the neural circuits and cellular mechanisms by which the hippocampus gates the expression of fear and safety memories encoded in the amygdala. We hypothesize that the hippocampus modulates amygdala circuits involved in fear expression via the ventromedial prefrontal cortex. We will test this hypothesis by using a combination of electrophysiological recordings and permanent or reversible brain lesions. The first specific aim of the project will investigate the contribution of the medial prefrontal cortex to the context-specificity of extinction using both permanent and reversible lesions. The second specific aim will determine the essential connections between the hippocampus, prefrontal cortex, and amygdala that are required for the context-dependence of neural and behavioral responses to an extinguished auditory CS. The third specific aim will characterize the context-dependence of cellular firing in the hippocampus, prefrontal cortex, and amygdala during the retrieval of extinction memories. Parallel recordings in all three areas will seek to uncover correlated neuronal activity in the circuit, and the latency at which context-dependent neuronal responses to auditory conditioned stimuli emerge in each area. This work has broad significance for understanding flexible memory representations in the brain and potential clinical interventions to increase the generality and permanence of fear inhibition after extinction-like therapies (e.g., exposure therapy). PUBLIC HEALTH RELEVANCE The long-term goals of this project are to understand the neural systems that mediate the context-dependence of memory, particularly memory for traumatic events. We have proposed that the context-dependent expression of fear memories involves an interaction of the hippocampus and amygdala. An attractive hypothesis is that this interaction involves a hippocampo-prefrontal cortical projection, which in turn modulates neuronal activity in the amygdala. The goal of the present proposal is to test this model using a combination of techniques, including high-density single-unit recordings, sophisticated behavioral designs, and temporary brain lesions. This work will not only yield critical information concerning the essential neural substrates underlying fear and extinction memory, but will also yield insight into the mechanisms underlying the context-dependence of memory, which is a key feature of many memory systems. Moreover, this work will have far-reaching implications for the design of therapeutic interventions for disorders of fear and anxiety in humans.

Project Summary Clinical disorders of fear and anxiety, including trauma- and stressor-related disorders, represent an enormous public health burden. Cognitive-behavioral therapies, such as prolonged exposure therapy, have proven to be remarkably effective in reducing pathological fear in patients with these disorders. Nonetheless, there are a number of factors that limit the efficacy of exposure therapy. In particular, stress undermines exposure-based therapies by impairing extinction learning and promoting fear relapse. Despite years of work elucidating the neural circuitry for extinction, the neural mechanisms responsible for stress-induced extinction impairments remain poorly understood. One possibility is that stress dysregulates neuronal activity in the medial prefrontal cortex (mPFC), a brain area that is critical for extinction learning. In support of this possibility, we have recently shown that footshock stress causes lasting decreases in the spontaneous firing of neurons in the infralimbic (IL) division of the mPFC in rats. Decreases in IL firing were associated with an ?immediate extinction deficit? (IED), an extinction impairment that occurs when extinction is performed soon after fear conditioning (a stressor). Importantly, systemic administration of propranolol, a ß-noradrenergic receptor antagonist, prevented both the stress-induced depression of IL firing and the IED, suggesting a role for locus coeruleus norepinephrine (LC-NE) in this phenomenon. Although these data reveal that noradrenergic transmission is involved in the stress-induced depression of mPFC firing, the neural circuit by which stress perturbs mPFC firing is unknown. Interestingly, we have found that propranolol rescues the IED when delivered to the basolateral amygdala (BLA), but not the IL. Based on this work, we propose a novel hypothesis that stress-induced NE release from the LC recruits an inhibitory BLA->IL circuit that dampens activity in IL principal neurons to impair the acquisition and retention of long-term extinction memories. We propose three specific aims to test this hypothesis using a combination of in vivo electrophysiology, functional circuit tracing, and pharmacogenetic manipulations (e.g., `designer receptors exclusively activated by designer drugs' or DREADDs). The first specific aim of the project examines whether LC-NE projections to the IL or BLA are necessary and sufficient for stress-induced changes in mPFC firing and extinction learning deficits. The second specific aim examines explores whether BLA neurons projecting to the IL or PL mediate these effects. The third specific aim determines whether parvalbumin interneurons (PV-INs) in the mPFC are recruited by LC- NE activation and mediate the immediate extinction deficit through feed forward inhibition by BLA afferents. The outcomes of these aims will advance a novel circuit mechanism for stress-induced extinction impairments. Understanding this mechanism will facilitate the development of novel pharmacotherapeutic approaches that optimally engage mPFC circuits to facilitate extinction learning under stress.