We are pleased to announce the awards for the inaugural year of the PNI Innovation Fund. Below are the individuals receiving the awards and the research that will be supported.
Probing Predictive Computations in the Neocortex
The neocortex is one of the most fascinating regions of the brain, both because of its great size in humans and because of its relatively stereotyped microcircuitry. Therefore, understanding the computations carried out in a local neocortical microcircuit is of great interest. One candidate for a fundamental theory of neocortical function is that idea that the neocortex attempts to predict patterns in its spiking inputs, rapidly learning new patterns and encoding them in its local circuitry via learning rules. The Berry and Tank labs will collaborate to explore the predictive capabilities of area V1 in the mouse. They will engage animals in behavioral tasks while stimulating with elementary visual patterns and recording from large neural populations with optical imaging. The goal will be to find neurons that encode predictions either by remaining silent during predictable stimulus patterns or by selectively firing following a violation of the pattern. Experiments will seek to distinguish complex feature selectivity from predictive pattern encoding as well as determine to what degree pattern recognition is learned during the task.
Optogenetics for dissection of neural circuits and dynamics underlying high-level cognition
A critical tool towards understanding how neural circuits give rise to cognition is the ability to rapidly yet transiently inhibit neural activity in both a spatially and temporally specific manner. To achieve this we have developed and married two cutting edge technologies: the fully automated high-throughput training of rats on cognitively demanding tasks; and optogenetic mediated inhibition in relatively large volumes of neural tissue. In order to silence neural activity we first express halorhodopsin eNpHR3.0, a light activated chloride pump, with a series of precisely targeted viral injections. However, because neural tissue is rather opaque, illuminating the infected volume from one side (for example, from the surface of cortex), may fail to provide enough intensity to activate the halorhodopsin throughout the targeted tissue (in this example, the deepest layers of cortex, ~1.8mm away). To circumvent this problem we developed a chemical technique to sharpen one end of an optical fiber, allowing it to both be easily inserted into neural tissue and to deliver light more spherically, thereby inhibiting neurons above, below, and to the side of the fiber tip. The optical fiber implant and connection also had to be engineered to be tough enough not to be damaged by very active rats who would live and work with it for months. With these tools in hand we have first trained numerous rats on two behavioral tasks: a Memory-Guided Orienting Task (MGO) which is designed to probe decision-making and short-term memory, and the Poisson Clicks Task, an evidence accumulation task which is designed to probe the gradual accumulation of information used to form a binary decision. We have then targeted a range of cortical and subcortical areas for optogenetic mediated inhibition. One area in particular, the Frontal Orienting Field, a subregion of the rat's prefrontal cortex, shows evidence of playing distinct roles in the two tasks. In the MGO task it appears to be part of a distributed circuit bridging cortical and subcortical areas involved in both the decision formation and memory maintenance phases of the task. In the Poisson Clicks task, it appears to be part of an ongoing motor preparation response, at each point in time signaling what the optimal response would be given the evidence available so far. In the Poisson Clicks task, the FOF's activity is read out, and therefore necessary, only just prior to the animal initiating its motor response. Data from single site inactivation experiments in these two tasks has led us to develop neural circuit models that we are now testing with multi-site simultaneous inactivations.
Identification of novel neuronal binding partners for MHCI immune proteins
Over the past decade, it has become clear that the normal establishment, function, and plasticity of synaptic connections in the mammalian brain unexpectedly require proteins that were first identified in the immune system: members of the major histocompatibility complex class I (MHCI). A great deal is known about how MHCI performs the immune functions for which it is famous. In contrast, almost nothing is known about the mechanisms underlying MHCI’s newly-discovered, critical neuronal functions. It is particularly urgent to understand how MHCI signals in the brain, given the many genetic, symptomatic, and epidemiological associations between MHCI and neurodevelopmental disorders including autism and schizophrenia. This proposal will break new ground in understanding how MHCI signals in neurons, by identifying neuronal binding partners for MHCI, and defining how MHCI interactions with these proteins affect synaptic transmission and plasticity. In so doing, these studies will provide the first molecular mechanistic insights into how MHCI contributes to brain function and dysfunction. Ultimately, the information obtained in these studies could motivate the development of surprising, immune-based approaches for the diagnosis and/or treatment of MHCI-linked neurological disorders.
Solving Acoustic Communication in Drosophila: Connecting Neural Physiology with Biological Motion
Flies communicate using sounds during their mating ritual – the Murthy lab is focused on solving how these love songs are both produced and processed by the nervous system (via in vivo neural recordings and behavioral experiments). Because both song generation and perception depend heavily on the biomechanics of sensory and motor appendages in the periphery, they plan to use the PNI Innovation Funds to additionally study and model the biomechanics of the receivers and generators of fly acoustic signals (and thereby gain a deeper understanding of the relationship between neural activity and biological motion). Results from this project will contribute to resolving how nervous systems produce and process acoustic and species ‐ specific information.
- Sam Wang and Lynn Enquist
New Horizons for the Cerebellum in Cognitive and Affective Processing
The cerebellum is thought to contribute to cognitive and affective function based on lesion and functional data as well as anatomical tracing. Cerebellar circuitry follows a distinctive, repeating circuit architecture in which granule cell input converges massively onto Purkinje cells, which also receive “teaching” signals from the inferior olive’s climbing fibers. This organization follows a precise map in which defined cerebellar modules correspond to specific neocortical and other forebrain areas. Taken together, these ideas invite the possibility that the cerebellum’s nonmotor role may be analogous to its role in guiding movement. Understanding cerebellum-forebrain interactions has been hampered by lack of knowledge of neuroanatomical pathways across multiple synapses. The cerebellum’s output, the deep nuclei, are quite small, and project to the thalamus and ventral tegmental area (VTA), which are themselves densely organized. Following these pathways to neocortical targets with classical (i.e. nonviral) methods is difficult.
We propose a large-scale mapping of pathways between mouse cerebellum and forebrain. This project will use trans-synaptic methods and conditional expression strategies to identify intermediate structures and specific chains of connectivity. The Enquist lab has initiated studies with two strains of herpes simplex virus type 1 (HSV-1) that are principally anterograde (HSV-H129) or retrograde (HSV-McIntyre) in their transport through circuitry. These strains allow creative experiments for probing CNS pathways. New recombinants expressing various fluorophores and other proteins are being constructed. We also will implement the CLARITY method, in which tissue is cleared by the electrophoretic removal of light-scattering lipids. The Wang lab working with Stephan Thiberge will retrofit a multiphoton microscope for large-scale image acquisition and reconstruction. High-throughput reconstruction will allow us to trace the ontogeny of long-distance cerebellar pathways. This question is of central importance for human health, since early-life cerebellar disruption dramatically increases the likelihood of developmental disorders, most prominently autism. Our results will test whether the cerebellum is in a position to guide the formation of cognitive and affective functions in early life.