Correlated light and ultrastructural imaging of learning-related synaptic plasticity

点击翻译按钮获取中文摘要

Learning involves the reorganization of the complex connectivity schemes between neurons in the brain. Critical to this re-wiring is the formation of new synaptic connections, and thus the study of the rules governing synapse formation remains a central focus of neuroscience. Over the past decade, it has become increasingly clear that synapses follow specific spatial organizational principles, such that functionally similar synapses tend to “cluster” on dendrites. This spatial patterning likely affords computational advantages that allow neurons to efficiently construct representations of their input repertoire. However, it is unknown whether and how new synapses that form during learning contribute to such functional clustering of synapses. We propose to use a combination of cutting-edge imaging techniques to investigate new spine formation - and their potential clustering patterns - over learning, integrating a detailed description of the local synaptic activity profiles with a deep interrogation of the cellular and subcellular anatomy of the surrounding tissue. We will do this by first applying longitudinal functional imaging of dendritic spines in vivo in mice learning a motor skill over 2 weeks, followed by 3D electron microscopy of the volume imaged in vivo for a high-resolution reconstruction of relevant structures, including those that are not labeled for in vivo imaging. This approach will reveal how both the structural and functional environment of neuronal dendrites relates to the formation of new synapses. By focusing on new synapses whose activity becomes coherent with other nearby synapses on the same dendrite, we will provide a thorough description of how spinogenesis during learning contributes to functional synaptic clustering. Furthermore, by reconstructing the nearby cellular structures, we will provide heretofore inaccessible details, such as whether such functionally related synapses share the same axonal inputs. Finally, by using a well characterized model of learning, this work will also allow a quantitative description of how new spines and the clusters that they form relate to specific features of a learned behavior. The experimental paradigm established in the proposed project will be widely applicable to the studies of neural circuits underlying other types of learning and behavior. The prevalence of neurological diseases affecting neuronal connectivity highlights the importance this pursuit.