Decoding network activity states using sparse optical sampling

点击翻译按钮获取中文摘要

Project Summary This proposal is focused on developing new technological approaches to study how local circuits in cortical brain regions function. The central goal of this project is to develop a new recording method, we term “random aperture photometry”, that can monitor network state optically in combination with fluorescent Ca indicator probes that reflect activity in individual neurons. Unlike conventional photometry approaches where the entire microscope field-of-view is monitored using a sensitive light detector, in the method we propose to split the microscope output into multiple parallel image paths, each passing through a different mask that only allows light transmission through small apertures. By monitoring the total light transmission through the mask, this method generates an optical signal that sparsely samples the field of neurons labeled with fluorescent activity indicators. By employing multiple, independent image paths—each containing a mask with a different random arrangement of apertures—we can generate multiple signals that each reflect the ongoing network activity but are only weakly correlated. Our preliminary experiments and computer simulations suggest that signals from four independent masked pathways will likely be sufficient to decode network states evoked by different stimuli applied to the input layer of the dentate gyrus. The proposed work will use computer simulations, whole-cell patch clamp recording and optical methods to optimize and refine this method. We then will validate the ability of this new optical method to decode network states using bulk-loaded fluorescent Ca imaging while simultaneously assaying neuronal behavior using multiple intracellular recordings. We have used the intracellular recording-based method of assaying network states through a series of four recent publications. As with these previous studies, the proposed experimental work will be conducted using standard acute rodent brain slices. We believe this method is unique in providing non-invasive “whole-circuit” views of circuit function and will be especially useful in identifying specific network states and the dynamics of network state transitions. This method is also complementary with standard multi-electrode array recordings and 2-photon Ca imaging techniques. While these methods also can sample over a wide area of brain tissue, the primary goal of both techniques (and conventional extracellular unit recordings) is to estimate activity in individual neurons. By contrast, our method intentionally sacrifices spatial resolution to generate network state-specific optical signals with high signal-to-noise ratios. Since our method uses standard Ca indicator dyes and single photon excitation, it can easily be applied in vivo once the methodological details are refined and validated through the proposed brain slice experiments.