CRCNS: Deciphering the Dynamical Multi-Scale Structure-Function Relation of Dendritic Spines

描述（由申请人提供）：我们储存和保留新信息的能力取决于大脑巨大的可塑性。实验证据表明，单树突棘水平的形态化学修饰可能有助于学习和记忆，但我们缺乏对棘功能的定量和机械理解。为了解决这一缺陷，拟议的项目旨在探索树突棘的超微结构三维架构如何塑造其电化学信号转导，以及结构变化如何改变转导，从而影响突触功效。国家的最先进的三维电子显微镜（EM）重建赋予精确的离子型受体动力学和准确的生物物理模型将用于构建树突棘的纳米分辨率模型。为了模拟这样一个复杂的多尺度环境中的电化学动力学，先进的数值方案，如有限元离散化和快速多级求解器将被采用。通过在计算机上进行详细的实验，将确定影响离子电流在脊柱中传导的主要因素 然后用于系统地导出适合于精确数学分析的低维脊柱模型。完整和简化模型都将使该联盟能够研究单个棘的亚细胞信息处理能力，并将结果与体内和体外数据进行比较。这些模型将使研究人员能够开发和测试与脊柱相关的实验假设，并在统一的框架内解释健康和疾病修饰组织中记录的数据。目标1：以纳米分辨率在3D中重建树突棘;目标2：建立生物病理学上真实的纳米分辨率脊柱模型;目标3：开发高性能数值方法来模拟脊柱模型;目标4：使用模拟和理论来研究树突棘的计算。 该项目旨在将树突棘的多尺度生物组织与宏观和系统层面的可能功能后果联系起来。在超级计算能力不断增强的时代，任何突触传递和突触后整合的机械模型都应该从清楚地了解生物物理学如何精确地编排信号开始 在最小的尺度上进行转导。破译脊柱动力学的微观结构特征的影响将是理解神经信号传播和突触可塑性的垫脚石，并可能揭示新的亚细胞计算原理。这一发现将加深我们对健康和疾病改善的大脑中神经信息处理的理解，并可能导致神经形态设备的新设计。 脊柱形态的改变见于各种脑部疾病[1]，包括唐氏症？s综合征和脆性X综合征[2]。同样，改变脊柱内钙动力学和稳态已被记录为阿尔茨海默氏症？疾病[3]。更好地理解树突棘的动力学结构和功能之间的关系将有助于开发这些疾病的新疗法。为了实现这一目标，在过去NSF支持的项目的延续中，我们将招募和培训年轻科学家，以应对现代多尺度和多模式数据驱动生物学的跨学科挑战，其中，成功不仅由神经科学驱动，还由工程，数学和物理科学驱动。计算科学和神经信息学美国和德国的三个实验室之间计划的合作将产生国际培训研究生和博士后研究人员的机会，以及研究人员参与鼓励代表性不足的少数民族在STEM学科追求职业道路的计划。

DESCRIPTION (provided by applicant): Our ability to store and retain new information rests upon the brain's immense plastic capabilities. Experimental evidence suggests that morpho-chemical modifications at the level of single dendritic spines may contribute to learning and memory but we lack both a quantitative and a mechanistic understanding of how spines function. To address this deficit, the proposed project aims to explore how the ultra-structural three-dimensional architecture of dendritic spines shapes their electro-chemical signal transduction and how structural changes alter the transduction and thus affect synaptic efficacy. State-of-the-art 3D electron microscope (EM) reconstructions endowed with precise ionotropic receptor kinetics and accurate biophysical models will be used to construct a nanometer-resolution model for dendritic spines. To simulate the electro-chemical dynamics within such a complex multi-scale environment, advanced numeric schemes such as finite-element discretization and fast multi-level solvers will be employed. By performing detailed experiments in silico, the primary factors that influence ionic current conduction in spines will be identified and then used to systematically derive a low-dimensional spine model amenable to exact mathematical analysis. Both, the full and the reduced model will allow the consortium to study the sub-cellular information-processing capabilities of single spines, and to compare the results with in vivo and in vitro data. The models will enable researchers to develop and test spine-related experimental hypotheses and to interpret data recorded in healthy and disease-modified tissue within a unified framework. Objective 1: Reconstruct Dendritic Spines in 3D at Nanometer Resolution; Objective 2: Establish a Biophysically Realistic Nanometer-Resolution Spine Model; Objective 3: Develop High-Performance Numerical Methods to Simulate the Spine Model; Objective 4: Use Simulations and Theory to Study the Computations of Dendritic Spines. The project aims to relate the multi-scale biological organization of dendritic spines to possible functional consequences at the macroscopic and systems level. In the era of ever-increasing super-computing capabilities, any mechanistic model of synaptic transmission and postsynaptic integration should start with a clear insight into precisely how biophysics orchestrates the signal transduction at the smallest scales. Deciphering the impact of micro-structural features on spine dynamics will be a stepping-stone towards understanding neural signal propagation and synaptic plasticity, and likely reveal novel sub-cellular computational principles. The findings wil deepen our understanding of neural information processing in healthy and disease-modified brains and may lead to new designs for neuromorphic devices. Alterations in spine morphology are seen in various brain diseases [1] including Down?s syndrome and fragile X syndrome [2]. Similarly, changes of the intra-spine calcium dynamics and homeostasis have been documented for Alzheimer?s disease [3]. Developing new cures and therapies for these diseases will profit from a better understanding of the relation between the dynamical structure and the function of dendritic spines. To reach this goal, in continuation of past NSF-supported projects, we will recruit and train young scientists to meet the interdisciplinary challenges of modern multi-scale and multi-modal data-driven biology, where progess is driven not only by neuroscience, but also engineering, mathematics and physical sciences, computational science and neuroinformatics The planned collaboration between the three laboratories in the US and Germany will generate international training opportunities for graduate students and postdoctoral researchers, and the participation of researchers in programs that encourage underrepresented minorities to pursue career paths in STEM disciplines.