Successful recovery from neuronal damage requires a huge energy supply, which is provided by mitochondria. However, the physiological relevance of mitochondrial dynamics in damaged neurons in vivo is poorly understood. To address this issue, we established unique bacterial artificial chromosome transgenic (BAC Tg) mice, which develop and function normally, but in which neuronal injury induces labelling of mitochondria with green fluorescent protein (GFP) and expression of cre recombinase. GFP-labelled mitochondria in BAC Tg mice appear shorter in regenerating motor axons soon after nerve injury compared with mitochondria in non-injured axons, suggesting the importance of increased mitochondrial fission during the early phase of nerve regeneration. Crossing the BAC Tg mice with mice carrying a floxed dynamin-related protein 1 gene (Drp1), which is necessary for mitochondrial fission, ablates mitochondrial fission specifically in injured neurons. Injury-induced Drp1-deficient motor neurons show elongated or abnormally gigantic mitochondria, which have impaired membrane potential and axonal transport velocity during the early phase after injury, and eventually promote neuronal death. Our in vivo data suggest that acute and prominent mitochondrial fission during the early stage after nerve injury is an adaptive response and is involved in the maintenance of mitochondrial and neuronal integrity to prevent neurodegeneration.

Mitochondrial DNA (mtDNA) mutator mice are proposed to express premature aging phenotypes including kyphosis and hair loss (alopecia) due to their carrying a nuclear-encoded mtDNA polymerase with a defective proofreading function, which causes accelerated accumulation of random mutations in mtDNA, resulting in expression of respiration defects. On the contrary, transmitochondrial mito-miceΔ carrying mtDNA with a large-scale deletion mutation (ΔmtDNA) also express respiration defects, but not express premature aging phenotypes. Here, we resolved this discrepancy by generating mtDNA mutator mice sharing the same C57BL/6J (B6J) nuclear background with that of mito-miceΔ. Expression patterns of premature aging phenotypes are very close, when we compared between homozygous mtDNA mutator mice carrying a B6J nuclear background and selected mito-miceΔ only carrying predominant amounts of ΔmtDNA, in their expression of significant respiration defects, kyphosis, and a short lifespan, but not the alopecia. Therefore, the apparent discrepancy in the presence and absence of premature aging phenotypes in mtDNA mutator mice and mito-miceΔ, respectively, is partly the result of differences in the nuclear background of mtDNA mutator mice and of the broad range of ΔmtDNA proportions of mito-miceΔ used in previous studies. We also provided direct evidence that mtDNA abnormalities in homozygous mtDNA mutator mice are responsible for respiration defects by demonstrating the co-transfer of mtDNA and respiration defects from mtDNA mutator mice into mtDNA-less (ρ(0)) mouse cells. Moreover, heterozygous mtDNA mutator mice had a normal lifespan, but frequently developed B-cell lymphoma, suggesting that the mtDNA abnormalities in heterozygous mutator mice are not sufficient to induce a short lifespan and aging phenotypes, but are able to contribute to the B-cell lymphoma development during their prolonged lifespan.