Mitochondrial fission is necessary for the maintenance of the mitochondrial network, and relies on the GTPase dynamin-related protein 1 (DRP1; also known as DNM1L, dynamin-1-like protein) 1, 2. DRP1 forms helical oligomers that wrap around the mitochondrial outer membrane and scission it 3, 4. Recently, it was proposed that DRP1 is not sufficient to execute mitochondrial fission, and that another GTPase, dynamin-2 (DNM2, also known as DYN2), was an essential component of the mitochondrial division machinery 5. Here we report that mouse fibroblasts lacking all three mammalian dynamin proteins (DNM1, DNM2 and DNM3) 6 (dynamin triple-knockout cells), as well as cells with knockdown of DNM2 only, do not display defects in mitochondrial fission or mitochondrial hyperfusion, which were readily detected after knockdown of DRP1, even in the dynamin triple-knockout cells. The same was observed when examining peroxisomal fission (mitochondria and peroxisomes share the fission machinery 7). Thus, our data show that DRP1 is essential for mitochondrial and peroxisomal fission, whereas DNM1–DNM3 are dispensable. Using wild-type and dynamin triple-knockout fibroblasts (Extended Data Fig. 1a, b; absence of DNM1–DNM3 is demonstrated in Extended Data Fig. 1c, d), we first characterized the steady-state morphology of the mitochondrial network by transfecting cells with green fluorescent protein (GFP) targeted to the mitochondrial matrix (mito-GFP)(Fig. 1a). The systematic analysis of the mitochondrial network revealed a similar number of individual mitochondria per region-of-interest (ROI) in wild-type and triple-knockout cells (Fig. 1b; defective fission would result in less individual mitochondria). The mitochondrial area (Fig. 1c) and perimeter (Fig. 1d) were similar between wild-type and triple-knockout cells, and no constricted areas were observed. To exclude interference of mito-GFP expression with mitochondria fission or fusion, we also characterized the morphology of the mitochondrial network by immunocytochemistry, using an antibody against the outer mitochondrial membrane protein TOM20 (Extended Data Fig. 1e). No significant differences were observed between wild-type and dynamin triple-knockout cells in the mitochondria number per ROI (Extended Data Fig. 1f), area (Extended Data Fig. 1g) or perimeter (Extended Data Fig. 1h). These results suggest that mitochondrial gross morphology and dynamics are independent of dynamin proteins. Nevertheless, it was theoretically possible that the mitochondrial fission rate would be decreased in the absence of dynamin proteins, with the steady-state mitochondrial morphology remaining unaffected owing to a compensatory decrease in fusion 8. To address this possibility, we measured the levels of several proteins involved in mitochondrial fission and fusion by western blot analysis. We found no difference in the levels of the fission protein DRP1 or mitochondrial fission factor (MFF), which recruits DRP1 to the mitochondrial outer membrane 9 (Fig. 1e), between wild-type and triple-knockout cells. The phosphorylation of DRP1 residues Ser616 (which promotes mitochondrial localization 10) and Ser637 (which excludes DRP1 from mitochondria 2) was not altered (Fig. 1e). In addition, the protein levels of the mitochondrial fusion proteins optic atrophy 1 (OPA1; both short and long isoforms), mitofusin 1 (MFN1), mitofusin 2 (MFN2), or the mitochondrial outer membrane protein voltage-dependent anion channel (VDAC; a marker of mitochondrial mass) were also indistinguishable between wild-type and dynamin triple-knockout cells (Fig. 1f).

To determine whether acute loss of …