Structural analysis of dynamins involved in mitochondrial morphology

Dynamin蛋白家族由独特的GTP酶组成，参与整个细胞的膜分裂和融合事件。我们的目标是了解这些蛋白质的动态结构特性，并将它们与不同的细胞功能联系起来。动力蛋白对细胞内吞和囊泡活动是必不可少的。更多的动力蛋白家族成员参与了多种基本的细胞过程，包括线粒体的分裂和融合、抗病毒活性、细胞板的形成和叶绿体的生物发生。在这些蛋白质中，自组装和齐聚成有序结构是一个共同的特征，对大多数蛋白质来说，是其功能所必需的。虽然有大量关于动力素的信息，但对动力素相关蛋白质的结构性质知之甚少。为了确定动力蛋白家族成员之间是否存在共同的作用机制，我们研究了参与线粒体分裂的酵母动力蛋白家族成员DNM1和参与线粒体融合的人动力蛋白(OPA1和Mfn1)的结构和功能。 线粒体膜融合和分裂之间的平衡是正常线粒体形态和功能所必需的。在哺乳动物细胞中，Mitofusins被证明介导了线粒体外膜的融合，而OPA1被证明在线粒体内膜的融合中起作用。这两种哺乳动物的丝裂原丝裂原蛋白在其N端都含有一个大的胞质GTP酶结构域，然后是两个跨膜结构域和一个短的C末端结构域。像其他动力蛋白家族成员一样，GTP结合和水解会驱动丝裂原蛋白的构象变化，从而介导膜融合。然而，丝裂原介导的膜融合的确切机制仍不清楚。目前尚不清楚线粒体外膜融合是否通过拴系、对接、融合和拆卸等规范步骤进行。与SNARE蛋白复合体类似，我们预计有丝分裂素将在相对的膜上相互作用，经历一种构象变化，使膜足够接近，以克服融合的激活能障碍，并在融合后解体，以便下一轮融合。为了深入了解导致膜融合的构象变化，我们正在研究脂质双层中的丝裂原蛋白的结构。与其他跨膜蛋白一样，丝裂原蛋白在其全长状态下很难表达和纯化，不适合用于结晶学研究。然而，我们已经优化了全长丝裂原蛋白1的纯化和脂质体掺入策略，这是第一次可以用冷冻-EM进行可视化。这些电子显微照片显示紧密连接的蛋白质脂质体网络，有电子致密的缝隙。这表明，丝裂霉素可以通过形成寡聚体，在合成的膜之间相互作用，从而将蛋白质脂质体捆绑在一起。此外，我们正在与John Hammer和吴旭峰博士合作，通过荧光显微镜显示细胞中的有丝分裂素，并制造出更脆的有丝分裂素1和2敲入细胞。在未来，我们计划确定丝裂原蛋白组装突变体对线粒体融合的影响。 此前，我们与加州理工大学的David Chan博士合作，通过负染和低温电子显微镜检查了OPA1的结构。OPA1(常染色体显性遗传性视神经萎缩)突变可导致遗传性视网膜神经节细胞病变。在细胞中，OPA1已被证明是线粒体膜内融合所必需的，但其作用机制尚不清楚。将OPA1添加到含有心磷脂的脂质体中可提高GTP的水解率，并促进OPA1在脂质周围自组装成螺旋阵列，形成蛋白质-脂质管。在过去的一年里，我们用CryoEM方法计算了OPA1在apo和GTP过渡态下与脂质结合的两个高分辨结构。这些结构提供了OPA1如何在脂质双层上组装成螺旋阵列的新的分子细节，以及OPA1的高阶组装对线粒体形态和显性视神经萎缩(DOA)病理的重要性。在与吴旭峰博士和Marie-Paule Strub博士(NHLBI)的合作中，我们检查了与我们细胞中的高阶组装界面相关的DOA突变体，并在每个案例中观察到线粒体碎裂。 最近，我们开始探索Neurolastin的结构，这是一种大脑特有的动力蛋白相关蛋白，与线粒体相关，也在突触传递中发挥作用。在进化系统树中，Neurolastin与atlastin最为相似，后者是一种参与内质网形态的动力蛋白相关蛋白。此前，我们与Tina Lee博士(卡内基甜瓜大学)合作，揭示了atlastins通过冷冻EM连接脂质体的能力，这是膜融合之前的关键步骤(Crosby等人，2021年)。在未来，我们计划通过低温电子显微镜和冷冻电子断层扫描方法来检测神经抑素在体外和细胞内系膜和融合膜的能力。

The dynamin family of proteins consists of unique GTPases involved in membrane fission and fusion events throughout the cell. Our goal is to understand the dynamic structural properties of these proteins and correlate them with their diverse cellular functions. Dynamin is essential for endocytosis and vesiculation events in the cell. Additional dynamin family members have been implicated in a variety of fundamental cellular processes, including mitochondrial fission and fusion, anti-viral activity, cell plate formation and chloroplast biogenesis. Among these proteins, self-assembly and oligomerization into ordered structures is a common characteristic and, for the majority, is essential for their function. Although there is a wealth of information regarding dynamin, little is known about the structural properties of dynamin-related proteins. To determine if a common mechanism of action exists among the dynamin family members, we examined the structure and function of Dnm1, a yeast dynamin family member involved in mitochondria fission and human dynamins involved in mitochondria fusion (Opa1 & Mfn1). A balance between mitochondrial membrane fusion and fission is required for normal mitochondrial morphology and function. Mitofusins have been shown to mediate mitochondrial outer membrane fusion in mammalian cells while Opa1 has been shown to play a role in the fusion of the inner mitochondrial membrane. Both mammalian mitofusins contain a large cytosolic GTPase domain at their N-terminus followed by two transmembrane domains and a short C-terminal domain. Like other dynamin family members, it is predicted that GTP binding and hydrolysis drive mitofusin conformational changes that mediate membrane fusion. However, the precise mechanism for mitofusin-mediated membrane fusion remains unclear. It is still unknown if mitochondrial outer membrane fusion proceeds through the canonical steps of tethering, docking, fusion, and disassembly. Similar to the SNARE protein complex, we expect that mitofusins will interact with each other on opposing membranes, undergo a conformational change that drives the membranes close enough to overcome the activation energy barrier for fusion, and after fusion disassemble to be available for the next round of fusion. To gain insight into the conformational changes that lead to membrane fusion we are examining the structure of mitofusins in a lipid bilayer. As with other transmembrane proteins, mitofusins are difficult to express and purify in their full-length state and are poor candidates for crystallography. Nevertheless, we have optimized a purification and liposome-incorporation strategy for full-length mitofusin 1, which can be visualized for the first time by cryo-EM. These electron micrographs show networks of tightly tethered proteoliposomes with electron dense seams. This suggests that mitofusin can tether proteoliposomes by forming oligomers that interact in trans between synthetic membranes. In addition, we are collaborating with Drs. John Hammer and Xufeng Wu to visualize mitofusins in cells by fluorescent microscopy and made mitofusin 1 and 2 CRISPER knock-in cells. In the future we plan to determine the effects of mitofusin assembly-mutants on mitochondrial fusion. Previously, in collaboration with Dr. David Chan from Cal Tech, we examined the structure of OPA1 by negative stain and cryo electron microscopy. Mutations in OPA1 (autosomal dominant optic atrophy) can lead to an inherited neuropathy of the retinal ganglion cells. In the cell, OPA1 has been shown to be essential for the fusion of the inner mitochondrial membranes, but its mechanism of action remains poorly understood. Addition of OPA1 to liposomes containing cardiolipin results in enhanced GTP hydrolysis rate and promotes OPA1 to self-assemble into helical arrays around the lipid, forming protein-lipid tubes. This past year we calculated two high-resolution structures of OPA1 bound to lipid in an apo and GTP transition state by cryoEM methods. The structures provide novel molecular details of how OPA1 assembles into a helical array on a lipid bilayer and the importance of OPA1 higher order assembles to mitochondrial morphology and dominant optic atrophy (DOA) pathologies. In collaboration with Drs. Xufeng Wu and Marie-Paule Strub (NHLBI), we examined DOA mutants that correlated to our higher order assembly interfaces in cells and in each case observed mitochondrial fragmentation. Recently we started exploring the structure neurolastin, a brain-specific dynamin-related protein associated with the mitochondria and also plays a role in synaptic transmission. In the evolutionary phylogenetic tree, neurolastin is most similar to atlastin, a dynamin-related protein involved in ER morphology. Previously, in collaboration with Dr. Tina Lee (Carnegie Melon U), we revealed atlastins ability to tether liposomes by cryoEM, a crucial step prior to membrane fusion (Crosby et al, 2021). In the future we plan to examine neurolastins ability to tether and fuse membranes in vitro and within cells by cryoEM and cryo-electron tomography methods.