Mitochondrial Biology

The group of Nils-Göran Larsson focuses on mitochondrial genetics and the impact of mitochondrial dysfunction on disease and ageing.

Mitochondrial dysfunction is heavily implicated in the ageing process. Ageing humans have increased levels of somatic mutations in the mitochondrial genome that tend to undergo clonal expansion to cause mosaic deficiencies in the respiratory chain. The oxidative phosphorylation (OXPHOS) system produces adenosine triphosphate (ATP), the universal source of energy in all tissues. Respiratory chain dysfunction, and thus insufficient supply of ATP, can cause a variety of phenotypes associated with ageing, age-related- and mitochondrial diseases.

Mitochondria harbour their own genome, which is typically approx. 16 kb. Animal mitochondrial DNA (mtDNA) typically encodes 2 ribosomal RNAs (rRNAs) and 22 transfer RNAs (tRNAs). Furthermore, it encodes 13 proteins, which are all components of the oxidative phosphorylation system. The remaining proteins of the OXPHOS system are nuclear encoded. These proteins form complexes that are located in the inner membrane of mitochondria. Transporting electrons through these complexes is coupled to the creation of a proton gradient, which drives the ATP synthase.

In the department of Nils-Göran Larsson, we use the house mouse (Mus musculus) and fruit flies (Drosophila melanogaster) as model organisms to investigate the regulation of the mitochondrial genome and the accumulation of somatic mutations in mtDNA.

Selected projects

Mitochondrial DNA is replicated by the DNA polymerase γ (Polg) in concert with other enzymes of the replication machinery. This enzyme is heterotrimeric and consists of the catalytic subunit PolgA and a dimeric accessory subunit PolgB. The catalytic subunit also harbours a exonucleolytic proofreading activity. In order to investigate the role of mosaic mutations in the mtDNA, we created a knockin mouse that displays a reduced proofreading activity of the Polg enzyme. With ongoing replication cycles during life, the mitochondria accumulate mutations in their genomes. We could show that the mutator mice show 3 to 5 times increased levels of random point mutations per mtDNA molecule, as well as increased levels of linear mtDNA molecules with deletions. Up to the age of 25 weeks these mice appear normal. But thereafter, they develop severe ageing phenotypes with weight loss, reduced subcutaneous fat, alopecia, kyphosis, osteoporosis, anaemia, reduced fertility, heart enlargement, greying of the hair and hearing loss. Furthermore, we could detect a reduced life span and respiratory chain dysfunction (Trifunovic et al., 2004). These findings show that mtDNA mutations can cause a premature ageing phenotype.

To further assess the influence of mutations in mitochondrial DNA (mtDNA) on ageing, and to clarify if these mutations are generated by continuous damage accumulation, we performed ultra-deep sequencing using the SOLiD technology (Ameur et al. 2011). This technology is known to have a low error rate, because its two-base colour code system interrogates each base twice to ensure accuracy. We sequenced the mtDNA of wildtype mice aged 30-84 weeks, as well as homozygous and heterozygous mutator mice and their wild-type siblings at the age of 30 and 40 weeks. We found that the mtDNA mutation load in these mice did not increase with age. Interestingly, the mutational profile was transition-biased which points to replication errors  and not damage to the mtDNA, (such as attach by radical oxygen species) as the source of the mutations . The mutator mice showed, in accordance with the previous findings, increased levels of point mutations. The heterozygous and wildtype siblings of mutator mice also showed elevated levels of clonally expanded point mutations, consistent with a substantial mutation load inherited from their heterozygous mother. We conclude that somatic mutations in mitochondrial DNA are not accumulating over the life span of an individual, but rather mainly are derived from replication errors during the rapid amplification of mtDNA during embryogenesis.

Transcription is needed for mitochondrial gene expression. We determined the basal transcription machinery as consisting of the mitochondrial RNA polymerase (POLRMT), mitochondrial transcription factor B2 (TFB2M) and the mitochondrial transcription factor A (TFAM) (Falkenberg et al., 2002). These factors constitute the basal machinery required for transcription initiation, but additional proteins are likely involved in regulation of this basal machinery for mitochondrial transcription. A family of four putative mitochondrial transcription termination factors (MTERF1-4) are likely of importance in this process.

MTERF1 is known to be a transcription termination factor and MTERF2 has been shown to be involved in regulation of mitochondrial transcription initiation.

By using a knockout mouse model we could show that MTERF3 is a negative regulator of mitochondrial transcription initiation (Park et al., 2007). Mice with a whole body knockout of MTERF3 die during embryogenesis. Mice with a specific knockout in heart and skeletal muscle show a dramatically reduced lifespan of max. 18 weeks. Electron micrographs of heart tissue from terminal-stage Mterf3 knockouts revealed abundant abnormal mitochondria, consistent with severe respiratory chain deficiency. MTERF3 binds in the promotor region of mtDNA and acts as a negative regulator of transcription instead of being a transcription termination factor. Repression of mtDNA transcription may be important in regulating oxidative phosphorylation in response to different physiological demands.

Recently our group was able to demonstrate that MTERF4 acts on the translation of mitochondrially encoded proteins (Camara et al., 2011). Whole body knockout of MTERF4 in mice also leads to embryonic lethality. As well as for MTERF3, a heart and skeletal muscle specific knockout of MTERF4 leads to a shortened life span of 21 weeks. Abnormal mitochondria can be observed in electron microscopy and a severe respiratory chain deficiency was detected in the animals. But besides these similar symptoms to the MTERF3 mice, MTERF4 closely interacts with NSUN4. NSUN4 is a RNA cytosine methyltransferase, present in the mitochondria. MTERF4 targets NSUN4 to the large ribosomal subunit, where it is needed for stable assembly of the ribosomal subunits. Loss of MTERF4, and the complex with NSUN4, leads to a reduction of mitochondrial translation. This finding shows an unexpected, but highly important function of a member of the MTERF family in regulating mitochondrial translation.

Mammalian mtDNA is not naked in the mitochondria, but is instead packaged by proteins in point-shaped complexes called nucleoids. Among others, the mitochondrial transcription factor A (TFAM) colocalizes with mtDNA in all of the nucleoids. In order to get a more detailed picture of the organization of the nucleoids we used super-resolution stimulated emission depletion (STED) microscopy, which has a resolution limit of approx. 40 nm in this experimental setup. The so far used confocal microscopy shows a nucleoid size of ~250nm, which coincides with the resolution limit of this method. Confocal microscopy can therefore not define the true size of the nucleoids. In a collaboration with the group of Prof. Jakobs at MPI for Biophysical Chemistry in Goettingen we used super-resolution microscopy and could narrow down the diameter of the nucleoids to ~100nm, which likely represents the true size of mammalian mitochondrial nucleoids. Furthermore, we could show that there is often only one single copy of mtDNA per nucleoid and that TFAM is the main protein constituent (Kukat and Wurm et al., 2011). This suggests that the TFAM protein is the main factor for packaging and organizing mtDNA into nucleoids. These findings have broad implications for understanding segregation and transmission of mtDNA in disease and ageing.