18 April 2006. Mutations in mitochondrial DNA (mtDNA) accumulate in aging and Alzheimer disease, but their impact on cell physiology has been open to debate. With thousands of copies of the mitochondrial genome per cell, the fraction of organelles afflicted with a somatic mutation needs to be pretty high (by some accounts, more than 60 percent) before functional changes are seen cell-wide. The dopaminergic neurons of the substantia nigra are an interesting place to look for mitochondrial mutations, because they are exquisitely sensitive to oxidative damage, and their loss causes the symptoms of Parkinson disease. Now, two studies published together in this month’s Nature Genetics online show that these neurons display high levels of mtDNA deletions, affecting an average of 50 percent of the mtDNA copies in neurons from PD brains. The proportion of deleted genomes in individual cells increases with age and the presence of PD, and correlates with loss of the respiratory enzyme cytochrome c oxidase (COX). These studies present the first direct evidence that mitochondrial DNA deletions lead to respiratory defects in human neurons, and suggest that clonal expansion of such deletions could play a role in the onset of age-related sporadic neurodegenerative disease.
In the first paper, Douglass Turnbull and colleagues at the University of Newcastle on Tyne, U.K., set out to discover whether mutations in mitochondrial DNA lay behind the sensitivity of these cells to mitochondrial dysfunction in PD. Using immunohistochemistry, joint first authors Andreas Bender and Kim Krishnan measured the levels of COX and succinate dehydrogenase proteins in substantia nigra (SN) neurons from PD patients and age-matched controls. They observed more COX/SDH-deficient cells in PD patients (approximately 2.5 percent) than controls (less than 1 percent).
Because the COX enzyme is encoded on the mitochondrial genome, its loss is a surrogate for high levels of mtDNA mutations. To test for such mutations, the researchers microdissected COX-deficient neurons from brain tissue of a single patient and examined the mtDNA. When they assayed for deletions, using a long-range PCR protocol, they detected no wild-type DNA, only deleted genomes, in a pool of 50 neurons. Looking at single cells, they found that the deletions were clonal—each cell displayed a single, unique deletion. No deletions were seen in glial cells from the same person.
Next, they used a real-time quantitative PCR method to assess the percentage of deleted mitochondrial genomes per cell. By comparing the accumulation of two different PCR products, one from a region susceptible to deletion versus one from a region that is rarely deleted, they could determine the fraction of mutated mitochondrial genomes in a given cell. Brains from PD patients had slightly higher prevalence of mtDNA deletions—52 percent versus the 43 percent seen in age-matched controls. The fraction of deleted genomes increased with age, and COX-deficient cells tended to have higher deletions (67 percent) than cells with normal COX protein (48 percent). The high level of mtDNA deletions was a property of SN cells, since hippocampal neurons displayed far lower levels. Their results suggested that the clonal expansion of somatic deletions was the cause of COX deficiency in neurons.
The second paper, from the lab of Konstantin Khrapko at Harvard Medical School in Boston, also used a novel PCR-based method to quantify the total burden of mtDNA deletions in SN neurons. Starting with postmortem brain samples from nine people age 33 to 102, first author Yevgenya Kraytsberg and colleagues also stained for COX and then laser-dissected COX-positive and COX-deficient cells. They lysed the cells and performed PCR after limiting dilution to amplify single mtDNA molecules. By this method, they found some cells that gave all full-length, wild-type PCR products, and other neurons that yielded shorter PCR products. For the latter type of cells, they confirmed the presence of clonal deletions. Like the Turnbull group, they found that the proportion of deletions measured in single cells rose with age. Rarely, a young cell would accumulate deletions, but by age 70, nearly all the SN neurons had elevated mtDNA deletions, with the fraction increasing rapidly after that. In addition, they report that deletions are specific to SN neurons, with other parts of the brain containing undetectable levels. They also found that the COX- deficient neurons had a higher percentage of mtDNA deletions than COX-positive cells.
Both sets of results imply that somatic mtDNA deletions are the cause of COX defects, and that these mutations accumulate to high levels selectively in SN neurons. The reason for this is unclear, but could be related to a high mutation rate in the oxidatively active SN neurons, or impaired mtDNA replication. Bender et al. speculate that the accumulation of high levels of mtDNA deletions could lead to neuronal loss in aging, and in sporadic PD. An increase in mtDNA mutations has been found in Alzheimer disease brain as well, and it will be interesting to test whether high levels of mtDNA deletions afflict other areas of the brain and their mitochondrial fitness, too.—Pat McCaffrey.
Bender A, Krishnan KJ, Morris CM, Taylor GA, Reeve AK, Perry RH, Jaros E, Hersheson JS, Betts J, Klopstock T, Taylor RW, Turnbull DM. High levels of mitochondrial DNA deletions in substantia nigra neurons in aging and Parkinson disease. Nat Genet. 2006 Apr 9; [Epub ahead of print] Abstract
Kraytsberg Y, Kudryavtseva E, McKee AC, Geula C, Kowall NW, Khrapko K. Mitochondrial DNA deletions are abundant and cause functional impairment in aged human substantia nigra neurons. Nat Genet. 2006 Apr 9; [Epub ahead of print] Abstract