Adam Kepecs - US grants

DESCRIPTION (provided by applicant): The long-term goal of our research program is to understand the neural circuit mechanisms underlying motivated behavior. The exquisite neural architecture of microcircuits in prefrontal cortex (PFC) is thought to underlie the flexibility and dynamics responsible for cognition. This proposal aims to understand the role of distinct interneuron types in prefrontal cortical function. Our general approach is predicated on the idea that access to cell-type identity is essential to unlocking the function of neocortical circuits. Because interneurons constitute a highly diverse neural population, some with well-understood anatomical specializations, they represent both an important opportunity to reveal microcircuit function as well as an excellent showcase for demonstrating the use of cell-type identity based functional studies. Our first objective therefore is to develop an optogenetic toolkit for this purpose. We will design and validate a miniature microdrive for combined electrophysiological recordings and fiberoptic stimulation that is light-weight and suitable for chronic recordings from freely behaving mice - enabling us to identify, record and manipulate genetically labeled cell-types. We propose to study three non-overlapping classes of interneurons: the parvalbumin (PV), somatostain (SOM), and vasoactive intestinal peptide (VIP) positive cells using knock-in Cre-driver lines, each with distinct functions. Using these tools we will examine how distinct cortical brain rhythms, signatures of coordinated neural activity, are correlated with the firing of distinct interneuron subtypes. Our optogenetic approach will not only establish the correlation of interneuron subtypes with oscillations but also enable selective control over the activity of distinct interneuron subtypes to pursue the mechanisms for generating different brain rhythms. We will employ a loss-and-gain-of-function approach to abolish and induce different brain rhythms in behaving animals. If successful, the proposed research is expected to result in a detailed understanding of the role of distinct interneuron subtypes in prefrontal cortical function and behavior. Because maladaptive changes inhibitory interneurons have been linked with a diverse set of diseases from epilepsy to schizophrenia and autism, our results will have direct implications for interpreting deficits in these disease states and potentially suggest avenues for remediation.

DESCRIPTION (provided by applicant): The long-term goal of our investigation is to understand how neural circuits in the prefrontal cortex support decision-making. Orbitofrontal cortex (OFC) is a key brain region for decision- making under uncertainty in humans and other animals. Our central hypothesis is that OFC represents confidence information and plays causal role in reporting decision confidence to other brain regions, thereby affecting behavior. To study this issue, we have developed quantitative psychophysical methods for rodents, adapted from human and primate work, which enables the behavioral readout of confidence in a well-controlled decision task. Briefly, in each trial the animal has to decide which of two odor components is in the majority. After moving to the corresponding choice port, the rat waits for a delayed reward that may or may not come. If it doesn't, the rat eventually leaves the port and initiates another trial. The duration rats are willing to wait for a reward can be thought of as a gamble on the outcome of the perceptual decision, and hence it provides a quantitative index for confidence. Thus in each trial there is a perceptual decision (left/right) and a post-decision confidence report (leaving decision). If OFC is causal in this leaving decision, neural activity should be predictive of how long a rat waits before leaving on a trial-by-trial basis (correlation) We will also test the necessity of OFC activity for this process using pharmacological inactivation. These results will be interpreted in a computational framework that links neural and behavioral data to decision confidence. Next we will test the hypothesis that identified subsets of OFC neurons projecting to distinct target areas carry different types of information and thus mediate distinct behavioral functions. Using retrograde viruses to deliver light-sensitive activators (ChR2) and suppressors (Halo), we will optically microstimulate or block a subset of OFC neurons defined not by spatial proximity but by projection target. In combination with electrophysiology we will identify the neural correlates of distinct projection neurons and determine whether they are necessary and sufficient for confidence reports on a trial-to-trial basis. Upon completion of these aims, we expect to establish the role of OFC in decision confidence and determine the neural circuits through which it exerts its actions. By recording from neurons belonging to different long-range projections we also expect to help explain the observed heterogeneity in OFC neural representations. Beyond these mechanistic studies, we hope to inform an improved framework for understanding how impairments in a single prefrontal brain area can lead to such a wide range of psychiatric disorders, including addiction, depression, anxiety, obsessive-compulsive disorder, schizophrenia, and autism.

A central goal in neuroscience is to understand behavior in terms of the activity and connectivity of specific neurons. Anatomical tracings have revealed the mesoscale connections in the brain, but this dataset lacks functional utility without the ability to behaviorally study neurons with specific connections to other brain regions. Recently, a number of technological advances have enabled these types of experiments, including re-engineered viruses to target specific neuron types and deliver genes of interests. This methodology is important because genetic targeting of specific neurons allows one to draw a link between a neurons's anatomy (the regions it projects to) and its function (the information it encodes). This project will explore the engineering of novel tools that will significantly improve existing retrograde tracing techniques and thereby enable mapping behavioral functions to specific neuron types. The resulting tools will be made broadly available to the community to increase their impact and utility.

As systems neuroscience is adopting tools from molecular biology there is an increasing appreciation for the importance of circuit-specific technologies, for instance targeting neuron-types defined by their projections using retrograde viral strategies for anatomical and genetic labeling. These tools have been critical for cell-type specific recordings using genetic activity indicators, or control using optogenetic actuators. Nevertheless, the efficiency and variable tropism (i.e. inability to target all cell types) of existing retrograde viruses presents a challenge for these experimental approaches. To overcome this challenge, two methods are proposed to generate reagents that will provide a non-toxic profile while maximizing efficient labeling of the targeted population. The methods will exploit well-understood molecular mechanisms for viral internalization to overcome tropisms, as well as classic tracers improved using a novel protein ligation technique. These reagents will be suitable for precise and robust anatomical tracing, for induction of opto- and chemicogenetic actuators. Therefore we expect that these improved reagents will allow for more efficient and less variable projection-based targeting of neurons with molecular cargo and facilitate new types of circuit-based mapping experiments.

? DESCRIPTION (provided by applicant): The brain is a massively interconnected network of regions, each of which contains neural circuits that process information related to combinations of sensory, motor and internal variables. Adaptive behavior requires that these regions communicate: sensory and internal information must be evaluated and used to make a decision, which must then be transformed into a motor output. Despite the importance of this question, we know relatively little about the principles of how spiking activity in one region influences activiy in downstream areas, particularly in the context of cognitive operations like decision-making. Here we propose to address this question by focusing on how the ventral striatum (VS), a region critical for motivational control of behavior receives and processes information from two important upstream regions, the orbitofrontal cortex (OFC) and the hippocampus (HP). We have assembled a unique team of scientists with complementary expertise studying the HP (Frank), OFC and VS (Kepecs), using synergistic technologies for large-scale recordings using novel polymer electrodes (Frank/Tolosa) with improved optogenetic identification of projections (Kepecs), and a team of statistical and computational researchers providing complementary analytical expertise in dimensionality reduction (Machens), statistical modeling (Eden/Kramer) and normative models (Ganguli). Our combined expertise will allow us to (1) measure large populations of neurons across the brain regions, (2) identify and (3) manipulate the neurons connecting them in order to (4) test for the first time a range of hypotheses about different modes and circuits for information transmission across regions. Beyond revealing how the OFC, HP and VS communicate during learning and decision-making, our approach will provide new experimental tools and computational methods for systems neuroscience, as well as new insights into the general principles of information transmission across regions.

? DESCRIPTION (provided by applicant): The long-term goal of our investigation is to understand how neural circuits in the orbitofrontal cortex support decision-making in the healthy state and how suboptimal choices after chronic drug abuse. Our experiments in this proposal are designed to delineate how OFC representations are mapped to specific neural circuits and how these are impacted by chronic morphine exposure. Our central hypotheses are that distinct projection-neuron types arising from OFC represent different decision-variables and these signals are routed to different subcortical target structures, and that drugs of abuse such as morphine selectively affect defined pathways to impair decision-making. To study this issue, we have developed quantitative psychophysical methods for rodents, adapted from human and primate work, which enables the behavioral readout of different decision variables, such as reward value and perceptual uncertainty in a well-controlled decision task. First, we will record OFC neurons in a reward-biased perceptual decision task that forces animals to make choices in the context of variable reward size and likelihood, and map distinct decision-variables to single neurons. Second, we will determine how these representations map onto specific classes of OFC pyramidal neurons - in particular, OFC projection neurons targeting VS, VTA, and BLA respectively. To achieve this we developed a technique based on a novel use of optogenetics to identify specific projection cell-types during behavior: by targeting ChR2 using retrograde viruses to specific projection neurons we identify these neurons in electrophysiological recordings by their light-responses. Using this approach we can examine what information OFC->NAc, OFC->VTA and OFC->BLA neurons carry. Third, based on information gleaned from recordings about when and how these pathways are activated, we will perform bidirectional manipulations of neuronal activity to reveal their causal roles in impacting choice behavior. Finally, we will determine how morphine self-administration and withdrawal disrupts choice behavior along different OFC output pathways and attempt to rescue this impairment. Upon completion of these aims, we expect to reach an improved understanding of how decision-variables are computed in the OFC. In addition, we will provide new information about how information within OFC is transmitted in a pathway-specific manner from to its subcortical targets. We expect that our approach will contribute to an improved translational understanding how drugs of abuse can cause the sometimes subtle impairments in decision-making that nevertheless have devastating consequences.

? DESCRIPTION (provided by applicant): Our long-term goal is to understand how the forebrain neuromodulatory region, nucleus basalis (NB) supports cognitive functions. NB is thought to play significant roles in learning and attention, and its degeneration parallels the decline of cognitive functions in patients in a range of dementias. However, identified projection cell-types have never been recorded. Our objective is to determine what information is represented in and signaled by distinct NB long- range cortical projections and establish their causal behavioral function. To study this issue, we have developed a quantitative psychophysical auditory detection task for mice, adapted from human and primate work, enabling us to assess a number of behavioral correlates of NB neurons. In addition, we will use our recently developed optogenetic toolkit to record from all the three known NB projection cell-types: cholinergic, GABAergic and glutamatergic neurons during behavior for the first time. Our central hypothesis is that NB broadcasts distinct cognitive signals in a cell-type-specific manner. First, we operationalized different cognitive variables so that the activity of every neuron encoding a particular cognitive variable can be statistically evaluated. Specifically we will assess the moment-to-moment, trial-to-trial correlations between firing rates and different behavioral measures (e.g. accuracy or reaction time, RT). Second, in each of the three aims we will consider one of the major projection systems and evaluate when they are recruited during behavior and what cognitive variables they encode. Finally, using the temporal information gleaned from these recordings we will test the causal role of different NB projection cell- types using optogenetic gain and loss of function manipulations. Upon completion of these aims, we expect to establish the cell-type specific broadcast signals from NB and their role in defined aspects of cognition.

Project Summary/Abstract The complex architecture of cortical microcircuits is thought to comprise variations of canonical microcircuits that perform elemental computations. In the midst of specialized local computations, cortex also receives global input from a variety of long-range neuromodulatory centers. This proposal investigates the relationship between these two types of circuits. A recently identified cortical circuit motif is controlled by a class of interneurons defined by their expression of vasoactive intestinal polypeptide (VIP), which disinhibit pyramidal cells across four distinct cortical areas, thus defining a canonical cortical circuit. These VIP interneurons in the auditory cortex are recruited in response to reinforcement signals (reward and punishment), which are likely driven by neuromodulatory systems. Therefore the dual objectives of this proposal are to determine both the generality of VIP recruitment by reinforcement signals and the circuit mechanisms responsible for this activity. First we will evaluate the cortex-wide generality of VIP interneuron recruitment by reinforcers using a combination of sophisticated techniques for neuron identification and evaluation of their activity. Second, we will identify which inputs drive reinforcement responses in VIP neurons. We will map all common regions across the brain that provide inputs to cortical VIP neurons, with a focus on the cholinergic and serotonergic neuromodulatory systems and determine the pathway(s) causally responsible. Finally, we will determine how different subtypes of VIP neurons are recruited and driven. Completion of these aims should reveal fundamental principles of how the VIP- controlled cortical microcircuit functions cortex-wide and serves as a conduit for fast neuromodulatory control.

Abstract The goal of this project is to develop an archival data format and a formal task specification language to serve as standards for describing behavioral experiments. Because different laboratories use different behavioral systems, hardware, and software, it has been difficult to communicate behavioral task design, share data, or reproduce experiments. To fill this gap, we have assembled a team that includes neuroscientists with extensive behavioral experience, a computer science expert in real-time formal language design, and scientific software engineers. To meet the needs of the broader community, we have formed a large consortium of laboratories, including BRAIN-initiative grantees that have supplied numerous use cases spanning the range of behavioral tasks. In Aim 1, we will identify requirements for behavioral data formats by convening a large consortium of experimentalists. To represent behavioral data, we will create an extension to the NWB:N format to facilitate analysis alongside other behavior types. To specify behavioral contingencies, we will develop a formal BEhAvioral task Description Language (BEADL) using the latest computer science techniques for real-time language design. In Aim 2, we will develop an open-source software suite for editing, executing and visualizing behavioral tasks. The software will interpret the same behavior into machine code for hardware controllers, or publication figures for humans. To ensure the standards address the needs of the community, we will collaborate with users and consortium labs to implement an extensive battery of published tasks. In Aim 3, we will expand our outreach to broaden adoption of behavioral task description and data formats. We will engage the broader community through web-based tutorials, hackathons, and workshops at major conferences, and will encourage feedback. A publication-ready, graphical BEADL format will be developed to generate summary figures for behavioral tasks that are formal, yet easy to understand.