基于显微光学断层成像技术研究AD小鼠神经元形态结构

发布时间：2018-07-18 09:25

【摘要】：人类的神经系统是目前已知的最复杂的功能系统,这一系统是基于数以百亿计的基本结构单元——神经元相互连接组成神经纤维网络。目前,由于技术限制,对阿尔茨海默病(Alzheimer’s disease,AD)发展进程中,神经元形态的精细变化仍缺乏了解。大脑形态各异的神经元相互连接形成网络,使全脑成为一个复杂的系统,各部分协同工作。为了研究神经元在病变时精细的形态变化,需要利用分辨率更高的全脑切片成像技术。通过这样的技术我们可以研究正常和AD鼠之间神经元形态的差异。本文借助MOST技术的高分辨率全脑成像能力,对三转基因AD模型小鼠(3×TgAD)的全脑进行Golgi染色,实施全脑切片和扫描,首次获得在细胞精度层面AD小鼠的全脑图谱,对所获得的每只鼠高达1TB的数据图片进行三维重建。在三维空间中构建并显示神经元的几何分布,追踪并提取数以百计的神经元重建信息。通过聚类分析、树突分析和Sholl分析,来分析和检测AD小鼠脑部神经元的发育及病理变化。为了获得神经元数据,需要将神经元形态信息从图片像素中分离出来,目前主要采用Amira软件实现半自动的追踪方法。但这一方法需要大量人工参与,追踪一个神经元至少需要半个小时,而每只小鼠有超过千万的神经元。在目前的文献报道中,已有的算法都不能有效实现对复杂图像数据的神经元追踪。为了能够实现对复杂数据更为有效的追踪方法,本文在OpenSnake追踪算法的基础上提出了改进方案,并用C++语言初步实现了这一算法及其三维成像。从我们的结果可以看出,阿尔茨海默病对小鼠的不同脑区产生不同的影响,其中影响最大的是海马CA1区。WT和AD小鼠随年龄增长神经元的纤维丰度增加,但病变鼠的增加幅度低于正常鼠,表明AD会减缓神经元的生长,同时AD会促使有活性的神经元比例减小。我们的研究结果和方法为AD的研究提供了重要的理论和方法基础。
[Abstract]:The human nervous system is the most complex functional system known at present. This system is based on tens of billions of basic structural units-neurons connected to each other to form a neural fiber network. At present, due to technical constraints, the fine changes of neuron morphology in the development of Alzheimer's disease (AD) are still lack of understanding. The neurons of different forms of the brain are connected to each other to form a network, which makes the whole brain a complex system and all parts work together. In order to study the fine morphological changes of neurons in pathological changes, a higher resolution whole brain slice imaging technique is needed. This technique allows us to study the differences in neuronal morphology between normal and AD mice. In this paper, the whole brain of three transgenic AD mice (3 脳 TgAD) was stained with Golgi staining, and the whole brain was sliced and scanned with the help of the high resolution global brain imaging ability of the most-technique. The whole brain map of AD mice at the cell precision level was obtained for the first time. Three-dimensional reconstruction was carried out on each mouse up to 1 TB of data. The geometric distribution of neurons is constructed and displayed in 3D space, and hundreds of neuronal reconstruction information are traced and extracted. Cluster analysis, dendritic analysis and Sholl analysis were used to analyze and detect the development and pathological changes of brain neurons in AD mice. In order to obtain neuron data, it is necessary to separate the morphological information from the image pixels. At present, Amira software is mainly used to realize semi-automatic tracking method. But this method requires a lot of artificial involvement, tracking a neuron for at least half an hour, and each mouse has more than 10 million neurons. In the current literature reports, the existing algorithms can not effectively realize the neuronal tracking of complex image data. In order to realize a more effective tracking method for complex data, this paper proposes an improved algorithm based on OpenSnake tracing algorithm, and realizes the algorithm and its 3D imaging with C language. From our results, we can see that Alzheimer's disease has different effects on different brain regions in mice. The most significant effect is the increase of fiber abundance of neurons in hippocampal CA1 region, WT and AD mice with age. However, the increase of AD was lower than that of normal mice, indicating that AD could slow down the growth of neurons, and AD could induce the proportion of active neurons to decrease. Our results and methods provide an important theoretical and methodological basis for AD research.
【学位授予单位】：深圳大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R749.16

【参考文献】