Mitochondrial fission and fusion

In animals and fungi, mitochondrial shape is set by a balance between fission and fusion carried out by three classes of dynamin-family GTPases. Fission of the outer membrane is driven by a dynamin-related protein (DNM1L/Drp1 in mammals, Dnm1 in budding yeast, DRP-1 in nematodes) that is recruited from the cytosol to constriction sites on the outer membrane. There it assembles into helices and uses GTP hydrolysis to constrict and sever the organelle. The same dynamin also divides peroxisomes. Recruitment depends on outer-membrane receptors whose identity differs between lineages: the tail-anchored adaptor MFF in animals, and Fis1 with the adaptors Mdv1 and Caf4 in budding yeast. In nematodes, the two MFF paralogs promote fission but are not essential for recruiting DRP-1. Fusion happens in two membrane-specific steps. Mitofusins (MFN1/MFN2, Fzo1, FZO-1) tether and fuse outer membranes. OPA1 and its orthologs (Mgm1 in yeast, EAT-3 in nematodes) fuse inner membranes; human L-OPA1 does this together with cardiolipin on the opposing membrane. When fusion is lost, unopposed fission fragments the mitochondrial network, and this fragmentation needs the fission dynamin. When fission is lost, the network becomes hyperconnected. The balance between the two is used in stress responses and in segregating damaged mitochondria for mitophagy.

Boundary. This module covers the membrane-remodeling reactions that divide and join mitochondria: dynamin-related-protein outer-membrane fission with its outer-membrane receptor, mitofusin outer-membrane fusion, and OPA1-family inner-membrane fusion. Selective removal of mitochondria is MODULE:mitophagy, which uses MFF/DRP-1 fission upstream of the FIS-1/FIS-2-dependent disposal step in worm muscle. Cristae shaping by OPA1 oligomers together with MICOS is a separable OPA1 function (S-OPA1 keeps cristae without supporting fusion, PMID:32245890) and is out of scope here. Mitochondrial transport and positioning (MIRO/TRAK) are also out of scope. Taxon scope. The context is Opisthokonta (animals and fungi), the clade in which all three GTPase classes are represented by reviewed members: worm, human and yeast for the fission dynamin, worm and human for mitofusin and OPA1. Fission dynamins are pan-eukaryotic: PAINT places mitochondrial fission on an animal DRP1 node (PTN002480309) and on a fungal Dnm1 node (PTN002480318), and plant DRP3-type proteins carry their own IBDs in PTHR11566. PAINT places the mitofusin GTPase/fusion node at Eukaryota (PTN002573358, taxon:2759), but no reviewed non-opisthokont mitofusin is included here. The OPA1-family fusion IBDs are on an animal node (PTN007514526, seeded by fly, mouse, human and worm) and, separately, on a fungal Mgm1 node (PTN008973437, seeded by S. pombe and S. cerevisiae). The fusion parts are therefore asserted only for opisthokonts, and a plant or protist version of this module would need different fusion parts. Fission receptor lineages. Receptors are the part that differs most between lineages. Animals use the tail-anchored adaptor MFF; vertebrates also use MiD49/MiD51, which can recruit Drp1 independently of Mff and Fis1 (PMID:23921378). Budding yeast uses Fis1 with the WD40 adaptors Mdv1 and Caf4 (PMID:18445678). No yeast FIS1, MDV1 or CAF4 gene review exists in this repository, so the receptor is modeled as a single MFF-family annoton rather than as a taxon variant set. A Fis1-Mdv1/Caf4 variant should be added once those yeast proteins are reviewed. In worms the two MFF paralogs act redundantly, and the mff-1 mff-2 double mutant has a fission defect close to that of drp-1, but MFF is not absolutely required: DRP-1 still reaches worm mitochondria without it (PMID:24196833). The worm reviews therefore keep the parent term molecular adaptor activity (GO:0060090) for MFF-1/MFF-2, whereas the human MFF review supports the more specific protein-membrane adaptor activity (GO:0043495) from direct MFF-Drp1 binding. FIS1 homologs in worms. Worm FIS-1 and FIS-2 are not fission factors: neither single nor double mutants have a detectable fission or morphology defect (PMID:24196833, PMID:18722182). Both gene reviews give them mitophagy and disposal roles, so they are members of MODULE:mitophagy. FIS-1 overexpression can still fragment mitochondria, even in the Mff double mutant, which is not evidence for an endogenous role. Unreviewed orthologs. Human MFN1, human FIS1, yeast Fzo1 (FZO1) and yeast Mgm1 (MGM1) have no gene review in this repository and are not used as representatives. Yeast Fzo1 and Mgm1 are the classic outer- and inner-membrane fusion GTPases and should be added as representatives once reviewed. PANTHER families. The fission dynamins fall in different PTHR11566 (DYNAMIN) subfamilies in each species (worm SF21, human SF39, yeast SF235), and PTHR11566 also contains OPA1 and the endocytic dynamins, so no family term is asserted for the fission-dynamin role; the PAINT nodes and representatives carry the grounding. Mitofusins share PTHR10465, MFF-family adaptors share PTHR16501 (named after fly Tango11), and worm EAT-3 and human OPA1 share the subfamily PTHR11566:SF67. Family assignments are from the UniProt PANTHER cross-references of the representatives. BCL-2-family regulation in worms. CED-9 binds FZO-1 and, when bound to EGL-1, binds and recruits DRP-1 (PMID:21949250). Another group found that loss or gain of egl-1 or ced-9 had no effect on mitochondrial connectivity or size (PMID:19327994). CED-9 and EGL-1 are therefore not modeled as members, and the conflict is recorded as a knowledge gap. Members left out. MIRO-1 (Q94263) is an outer-membrane Rho GTPase that, in other animals, links mitochondria to microtubule motors. Its worm review supports GTPase, calcium-binding and kinesin-binding activities, but no cached primary paper shows a worm transport or distribution defect: the miro-1 mutant has about half the normal amount of mitochondria and a long lifespan (PMID:27064409). SPG-7 (Q9N3T5) is the worm AFG3L2-type m-AAA protease subunit of the inner membrane. Its review places it in mitochondrial protein processing, respiratory-chain assembly and the UPRmt, not in the dynamics machinery, and no worm evidence links it to EAT-3 processing. MOMA-1 affects cristae morphology and is also left out.