Translational and Clinical Research Institute, Faculty of Medical Sciences, Newcastle University, Newcastle-upon-Tyne, UK.
Electron Microscopy Research Services, Newcastle University, Newcastle-upon-Tyne, UK.
Methods Mol Biol. 2024;2831:145-177. doi: 10.1007/978-1-0716-3969-6_11.
Neurons contain three compartments, the soma, long axon, and dendrites, which have distinct energetic and biochemical requirements. Mitochondria feature in all compartments and regulate neuronal activity and survival, including energy generation and calcium buffering alongside other roles including proapoptotic signaling and steroid synthesis. Their dynamicity allows them to undergo constant fusion and fission events in response to the changing energy and biochemical requirements. These events, termed mitochondrial dynamics, impact their morphology and a variety of three-dimensional (3D) morphologies exist within the neuronal mitochondrial network. Distortions in the morphological profile alongside mitochondrial dysfunction may begin in the neuronal soma in ageing and common neurodegenerative disorders. However, 3D morphology cannot be comprehensively examined in flat, two-dimensional (2D) images. This highlights a need to segment mitochondria within volume data to provide a representative snapshot of the processes underpinning mitochondrial dynamics and mitophagy within healthy and diseased neurons. The advent of automated high-resolution volumetric imaging methods such as Serial Block Face Scanning Electron Microscopy (SBF-SEM) as well as the range of image software packages allow this to be performed.We describe and evaluate a method for randomly sampling mitochondria and manually segmenting their whole morphologies within randomly generated regions of interest of the neuronal soma from SBF-SEM image stacks. These 3D reconstructions can then be used to generate quantitative data about mitochondrial and cellular morphologies. We further describe the use of a macro that automatically dissects the soma and localizes 3D mitochondria into the subregions created.
神经元包含三个隔室，即胞体、长轴突和树突，它们具有不同的能量和生化需求。线粒体存在于所有隔室中，调节神经元的活动和存活，包括能量产生和钙缓冲，以及其他作用，包括促凋亡信号和类固醇合成。它们的动态性允许它们根据不断变化的能量和生化需求发生持续的融合和裂变事件。这些事件，称为线粒体动力学，影响它们的形态，神经元线粒体网络中存在多种三维（3D）形态。在衰老和常见的神经退行性疾病中，形态学特征的扭曲和线粒体功能障碍可能首先发生在神经元胞体中。然而，3D 形态不能在平坦的二维（2D）图像中全面检查。这突出表明需要在体积数据中分割线粒体，以提供健康和患病神经元中线粒体动力学和线粒体自噬过程的代表性快照。自动化高分辨率体积成像方法（如连续块面扫描电子显微镜（SBF-SEM））的出现以及一系列图像软件包允许这样做。我们描述并评估了一种从 SBF-SEM 图像堆栈中随机生成的神经元胞体感兴趣区域中随机采样线粒体并手动分割其整个形态的方法。然后可以使用这些 3D 重建来生成有关线粒体和细胞形态的定量数据。我们进一步描述了使用宏自动解剖胞体并将 3D 线粒体定位到创建的子区域的用途。
