Reconstituting the pathway to mitochondrial division

Abstract

Mitochondria are double-membrane organelles in eukaryotes, whose primary function is energy production through oxidative phosphorylation. Mitochondria have a diverse morphology, which is maintained by the cycles of division and fusion. As mitochondria cannot be synthesized de novo, they rely on fission for successful organelle-inheritance. Moreover, mitochondrial division is also essential for the maintenance of the cellular homeostasis, during cell death and is used as a tool to segregate damaged mitochondria. Dynamin-related protein 1 (Drp1), a member of the dynamin superfamily of proteins, is a crucial player involved in the mitochondrial division process. In the absence of Drp1, mitochondria show an elongated phenotype, which is reminiscent of a defective division process. Apart from its role in the mitochondrial division, Drp1 is also shown to be involved in the division of peroxisomes. The domain architecture of Drp1 is similar to that of classical dynamins, which consists of a GTPase domain at the N-terminus, bundle signaling element and the stalk domain. However, instead of the lipid-binding pleckstrin homology domain found in classical dynamins, Drp1 has a variable, unstructured 100 amino acid loop called the B-insert, which is involved in binding to the mitochondrial lipid cardiolipin. Drp1 self-assembles into helical scaffolds and utilizes energy-derived form GTPhydrolysis to remodel GUVs and liposomes to tubular intermediates. Current literature indicates that Drp1 is involved in membrane remodeling to facilitate fission but its direct involvement in the fission process remains debated. Drp1 is predominantly cytosolic and relies on mitochondrial adaptor proteins (Mff, MiD49, and MiD51) for its recruitment to the mitochondria. Studies demonstrate that these adaptor proteins can act independently to recruit Drp1. However, their contribution, beyond recruiting Drp1 to mitochondrial and to the division process, in general, remains unknown. Recent reports suggest the involvement of the endoplasmic reticulum (ER) in causing mitochondrial constriction prior to Drp1-recruitment thus marking the site of mitochondrial division. A study by Voeltz and colleagues revealed the involvement of the classical dynamin-2 (Dnm2) in the mitochondrial division. Depletion of Dnm2 also led to significant mitochondrial elongation, with Drp1 remaining accumulated on constricted mitochondria. Thus, the current model proposes cooperation between mitochondrial and classical dynamins, and that neither alone is sufficient for the mitochondrial fission. However, recent reports have questioned this model as mitochondrial fission occurs even in the absence of Dnm2, thus necessitating a re-evaluation of the contribution of each of these proteins to the mitochondrial division. This thesis utilizes a bottom-up approach of reconstitution of the mitochondrial fission process and aims to understand the intrinsic functions of individual components, with a focus on Drp1’s involvement in mitochondrial fission. Chapter 1 of the thesis gives an introduction to proteins involved in the regulation of mitochondrial division, briefly summarizing the known components involved in the fission machinery. Chapter 2 introduces the Supported Membrane Templates (SMrT), where the membrane is organized as a planar sheet and curved tubes resting on a passivated glass coverslips covalently modified with polyethylene glycol (PEG). These membrane topologies displayed on SMrTs mimic a non-constricted and constricted states of the mitochondria. This facile and robust assay system allows the use of various membrane lipid compositions and screens for protein function on a membrane surface displaying a range of curvatures. The mitochondrial-specific lipid cardiolipin can also be incorporated into SMrTs to closely mimic mitochondria. In Chapter 3, using the SMrTs, I describe results indicating that Drp1 is sufficient to catalyze membrane fission. Fission is robust, with Drp1 capable of severing tubes as wide as 250 nm in radius. Although dynamin-2 can catalyze fission, it appeared to be severely restricted in its ability to severe wide tubes. Drp1 preferentially binds tubes over the supported lipid bilayer. This preference can be mapped to the B-insert region of Drp1. Stage-specific reconstitution reveals that unlike classical dynamins, which constrict membrane tubes in the absence of GTP, Drp1 requires GTP binding for membrane constriction. Drp1 requires GTP hydrolysis for causing further constriction of the membrane tube finally leading to fission. Together, our results indicate Drp1 to be self-sufficient in membrane fission and prompt a reevaluation of its involvement in the mitochondrial fission pathway. Various adaptor proteins on the outer mitochondrial membrane govern Drp1 recruitment to the mitochondria. In Chapter 4 effects of different adaptor proteins on Drp1-catalyzed membrane fission are probed using SMrTs. The presence of adaptor proteins Mff, MiD49, and MiD51 independently enhance Drp1 recruitment and fission. We also report a novel tendency of Mff to self-oligomerize and remodel membrane tubes, which in turn could determine the site of fission on mitochondria and peroxisomes. The effect of Mff on Drp1’s binding and fission activity is systematically studied by varying cardiolipin concentration. Chapter 5 discusses functional differences among Drp1 Isoforms. Surprisingly, Drp1’s longest Isoform (Isoform 1) is found to be impaired in membrane binding and fission, which is altogether different in behavior compared to Isoform 2 and Isoform 3. Since the various Isoforms are expressed in a tissue-specific manner, these results indicate a contribution from cell physiology to the evolution and selection of specific Isoforms of Drp1 that are different in their biochemical attributes.