The Mitochondrial Hypothesis of Ageing

Our aim is to investigate at single-cell level why some cell types are more susceptible to neurodegenerative diseases like Morbus Parkinson than others. After extracting these cells from human post mortem brain tissue by laser microdissection we subject them to molecular-biological examination. In particular, we are looking for damage to mitochondrial DNA (mtDNA). Mitochondria have not been thoroughly researched with single cell analysis so far, although evidence is growing that they could play a major role in neurodegenerative diseases and in the ageing process.

Mitochondria are very special cell organelles – they have their own genetic substance called mtDNA. They produce energy for the cell via the respiratory chain, whereby free radicals are released as a waste product of cell respiration, as it were. These oxidative oxygen compounds then damage the adjacent mtDNA. It is now supposed that during a person’s lifetime there is an accumulation of mtDNA damage which, upon reaching a certain threshold value, damages the cell and eventually leads to apoptosis. The brain is particularly affected, as brain cells are not replaced to a notable extent. Therefore, it is assumed that one of the causes of neurodegenerative diseases is an accelerated death of certain types of cell. In the case of Parkinson’s disease, these are mainly dopamine producing cells of the Substantia nigra – a core complex in the midbrain.

Fig. 1: Molecular vicious circle of the mitochondrial theory of ageing and neurodegeneration. Even under normal conditions, reactive oxygen species (ROS) are formed in the respiratory chain. This oxidative stress damages the adjacent mtDNA and in consequence also mitochondrial proteins. Gradually, the function of the respiratory chain is disturbed, leading to even higher ROS production. The result is a decrease in ATP production and eventually, disturbed cell functions and apoptosis.

Working with human brain samples not only entails special challenges but also restrictions. It’s not only that human post mortem brain tissue is only available in extremely small amounts and only via special brain banks. Part of our work involves gene expression analysis basedon RNA, which greatly depends on the quality of the tissue – the time between the death of the patient, organ removal and freezing and adequate cooling of samples from start to finish. Only a few hours after defrosting, the RNA is degraded by enzymes. If this happens before the experiment is performed, the result of the gene expression is useless.

When a suitable patient who has previously agreed to having tissue samples taken for research purposes dies, the relevant national brain bank is notified without delay. The corpse is transported to a pathological institute as quickly as possible and the brain removed in compliance with a standardised protocol. At least one half of the brain is then instantly shock-frosted and its quality can then be preserved for many years at a temperature of –80 °C. For experiments, the material is heated to –20 °C. Then frozen sections with a thickness of 20 microns areproduced and mounted onto special specimen slides for laser microdissection. They are then immediately stained prior to dissecting the required cells and processing them for gene expression.

In one of our studies we compared post mortembrain sections of Parkinson patients with samples of healthy control subjects of the same age. By staining the samples, we were able to see that a lotof cells in the Substantia nigra had a mitochondrial function disorder. These cells were extracted by laser microdissection to examine their respiratory chain and mtDNA. This examination revealed a high proportion of mutations (deletions) of the mtDNA.

We then compared a large amount of cells of the Parkinson group and the control group and found a significantly higher number of deletions in the Parkinsongroup – a great deal more than was thought in the past, in fact. It has been known for 20 years that there are changes in the mtDNA, but we always thought it was a matter of a few per cent. With single cell technology we were able to prove that 50 to 60 per cent of the mtDNA in Parkinson patients is damaged, which leads to a drastic energy deficit in the affected cells.

What surprised us was that there were a lot of mutations in 60 to 70-year-old control patients as well – although distinctly fewer than in the Parkinson group. In a second step we therefore examined control samples of people of all ages, from a few months to a hundred years old. We found that the mtDNA deletions increase with age. We are born with noor extremely few deletions and at some stage in our life the critical threshold of 50 to 60 per cent of mutations is reached.

Fig. 2: (A) Cytochrome c oxidase (COX) and succinate dehydrogenase (SDH) staining of dopaminergic substantia nigra neurons. Normal neurons are COX-positive (brown). Neurons with high levels of mtDNA deletions are COX-negative (blue). (B) A COX-negative neuron is cut out with laser capture microdissection (middle image) and identified in the lid of a PCR tube (right image). (C) Realtime PCR quantification of mtDNA deletions of individual COX-positive and COX-negative neurons. COX-negative neurons have higher levels of mtDNA deletions. (D) Long range PCR of individual neurons show clonal expansion of mtDNA deletions (bands shorter than the mtDNA wildtype band on the right). (E) Analysis of mtDNA deletion load of dopaminergic neurons in controls shows a steady increase of deletion levels with age.

Laser microdissection plays an extremely important role in my work, as the precise dissection of individual cells ensures that we really examine the cells that interest us. Before the age of laser microdissection, most examinations were done on homogenates. These contained different types of cells, so that the molecular connections we were researching sometimes disappeared in a great background noise. If a disease only affects one particular cell type, as in the case of Parkinson’s, we only obtain meaningful results when we are able to analyse homogeneous cell material. This is the advantage of laser microdissection.

Such therapies are feasible in principle, of course. One approach would be to prevent cells from accumulating mitochondrial damage by using antioxidative substances. On the other hand, extensive research is being conducted on how to make an already damaged cell produce more new, healthy mitochondria. However, it is too early to predict whether and when the results will assume relevance for mitochondrial diseases in everyday hospital routine.

Mitochondrial function disorders are certainly not the only cause of ageing. A recent review compared over a hundred hypotheses on why we grow old. Even if we managed to protect the mitochondria, it would be wrong to conclude that the cell would then lead a long and healthy life. It might trigger new problems. Saving mitochondria as a recipe for the fountain of youth; that is too simple, even though, of course, it would be a nice aspect of our work.

1. Bender A et al.: Dopaminergic midbrain neurons are the prime target for mitochondrial DNA deletions. J Neurol 255 (2008) 1231–5.
2. Reeve AK, Krishnan KJ, Elson JL, Morris CM, Bender A et al.: Nature of mitochondrial DNA deletions in substantia nigra neurons. Am J Hum Genet 82 (2008) 228–35.
3. Bender A et al.: High levels of mitochondrial DNA deletions in substantia nigra neurons in aging and Parkinson disease. Nat Genet 38 (2006) 515–517.