Overall objectives were to better understand how sensorimotor information is represented in the brain, study activity patterns generated by cortical pyramidal cells, and explore how plasticity may operate within networks to influence computations. Excitatory and inhibitory...
Overall objectives were to better understand how sensorimotor information is represented in the brain, study activity patterns generated by cortical pyramidal cells, and explore how plasticity may operate within networks to influence computations.
Excitatory and inhibitory neurons form interconnected networks that extract sensorimotor features, combine them with internally generated evidence, and filter out irrelevant information, thereby creating perceptions and guiding behaviors. At the level of individual cells, information processing occurs at synaptic contacts all along their morphologically complex dendritic trees. It has been challenging to monitor this activity because most technologies lack the speed and agility to carry out stable in vivo recordings from small structures like dendrites. In addition, computations take place during various behavioral states in the context of the network. In vivo-like network activity patterns can dramatically alter how inputs are transformed into outputs, but how neurons actually process information in active networks in vivo is unclear. In addition, circuits are endowed with plasticity mechanisms that enable flexible adaptions to the environment. How plasticity impacts dendritic integration remains poorly defined. Resolving these issues will inform our mechanistic understanding of how single neurons process sensory information during naturalistic behaviors, enabling steps forward in our knowledge of normal brain function.
Basic research impacts society by deepening our understanding of how the brain works. Complete explanations of how circuits operate should aid society in the long-run by leading to novel therapies that prevent or treat disease. There is an urgent need since the European Brain Council indicates 38% of the EUâ€™s population (164.8 out of ~500 million people) suffer from mental disorders, which apparently amounts to 798 billion tax-payer euros. Unfortunately, it is difficult to design better treatments for mental disorders because the gap in our knowledge of how genetic and environmental factors affect proteins, synapses, neurons, and networks remains large.
We employed a variety of experimental approaches and made several technical advances that enabled the study of outstanding biological questions related to how neurons integrate and transform sensorimotor signals. In the beginning, we focused more on mechanistic studies of how excitatory neurons in the mouse visual cortex integrate synaptic input to trigger spiking output. Using electrophysiological recordings from single cells in a reduced brain slice preparation, we experimentally manipulated network activity using methods that mimic what happens in an intact brain. Our results suggest increased network activity can increase the likelihood that synaptic input drives spiking output when inhibition is dampened or removed from the network. These results provided some valuable insight into how active networks in vivo may change the rules of neuronal computation across various behavioral states. During this time, we also carried out 3D imaging experiments in vitro and in vivo to characterize the patterns of dendritic activity that can be observed in excitatory neurons from visual and motor cortex during sensorimotor-evoked activity and passive behaviors. To study these activity profiles, we performed rapid 3D calcium imaging experiments (to infer neuronal activity) while simultaneously correcting for brain motion artefacts that can arise from breathing, heart-rate, and locomotion. We made the unexpected observations that both spontaneous and sensorimotor-evoked dendritic activity occur globally across the entire cell, with yet little evidence for local, dendritic branch-specific processing. Our latest in vivo 3D calcium imaging experiments have been focused on relating these single-cell patterns of dendritic activity to the overall level of activity present in the local network. These experiments are providing insight into how dendrites integrate synaptic inputs in vivo and suggest that dendrites may not be endowed with a vast number of flexible computational modes. To date, these results have been discussed and presented internally for scrutiny during lab meetings as well as at international specialist scientific meetings. Our plans for further dissemination and exploitation include additional presentations at scientific meetings to gain critical feedback before publication of the results, as well as preparing manuscripts for consideration for publication in international, peer-reviewed scientific journals. To make these data and results more accessible, they will be published in open-access journals, and relevant data-sets may be made openly available for mining, verification, and reuse.
How individual neurons perform sensorimotor integration in vivo is poorly understood because it is difficult to measure activity within morphologically complex 3D structures, such as dendrites. Another challenge arises from brain-motion artifacts that are inevitably present in experiments on awake-behaving animals, which cause small structures like dendrites to move in and out of the region of interest. To overcome these issues, our laboratory has developed two cutting-edge techniques: 2-photon acousto-optic lens (AOL) 3D microscopy, which achieves ultra-fast recordings and selective imaging of regions of interest distributed within the imaging volume, and real-time, on-line motion compensation, which nearly eliminates brain-motion artifacts during the experiment. Combining these tools with our novel semi-automated 3D dendritic tracing has enabled stable and reliable recordings simultaneously from somata, dendrites, and synaptic spines. We thus carried out, for the first time, selective imaging of neuronal activity from a large fraction of the entire dendritic tree of excitatory pyramidal cells in motor and visual cortices in awake behaving mice. Our results are revealing the spatial and temporal extent of dendritic activity within and across various types of neurons, brain regions, and levels of network activity, providing fundamental insights into how sensorimotor information is integrated by single cells and routed through neural circuits in vivo.
Findings from this research have been and will continue to be shared with scientists and the public. The broader impacts include: advancing neuroscience research and technology, developing collaborations and innovation, and sparking public interest in the brain. The potential socio-economic impacts include: advancing scientific knowledge via presenting data at international specialized scientific conferences and publications in respected international peer-reviewed open-access journals to give broad visibility to scientists and the public; facilitating international collaborations aimed at exploiting 3D imaging technologies; and making pertinent data-sets, metadata, and computer codes available for further exploitation via mining, verification, and reuse. The potential societal impacts include: improving our mechanistic understanding of normal brain function; development of new methods to test and treat brain disorder and disease; new ways of visualizing neuronal computations in 3D for use as educational tools; and inspiring the next generation of scientists through laboratory demonstrations and non-specialist presentations.
More info: https://silverlab.org/.