Polymerase gamma becomes a mutator polymerase, accelerating the accumulation of point mutations, frameshifts, and deletions in mitochondrial DNA. The steady accumulation of mutations is what accounts for the late onset and progressive nature of PEO. As a result, researchers now know that an errant DNA polymerase can cause disease.
It is unexpected news. "Polymerases are so essential, it was predicted long ago that no one would ever see natural mutations affecting fidelity," says Copeland. Reducing fidelity was presumed lethal. But if a mutator polymerase were possible, it would be a mutant polymerase gamma, Copeland believed, reasoning that the hundreds, sometimes thousands, of copies of mitochondrial DNA (mtDNA) inside cells provided a margin of safety against accumulating mutations. Even as mutator polymerases inactivated one gene after another, plenty of undamaged mitochondria would remain to permit normal cellular function.
What Twofold Can Do
The experiments, led by Mikhail V. Ponamarev, studied catalytic rate, binding affinity, replication fidelity, and proofreading of base pair mismatches. The catalytic rate was normal, as was the intrinsic 3'-5' exonuclease activity for proofreading. Y955C reduced binding affinity, but at nucleotide concentrations typical inside mitochondria, polymerase efficiency was comparable to wild type.
Realizing that proofreading would mask the full extent of replication errors, the researchers genetically deleted the proofreading exonuclease from the mutant polymerase. In doing this they discovered that Y955C increased the overall error rate 10-fold. With exonuclease activity intact, the error rate was predictably lower, but still twice as high as wild type.
"Twofold doesn't sound like much," says Copeland, but over a lifetime it is enough to do real damage. Thirty-seven genes on the mitochondrial chromosome (13 proteins, 20 tRNAs, 2 rRNAs, but not polymerase gamma, which is on chromosome 15) supply 13 polypeptides for the final steps of oxidative phosphorylation, the last of an elaborate chain of reactions that climax when ATP synthetase converts ADP into ATP. The large numbers of mitochondria in each cell delay the reckoning, but as genes inactivated by mutator polymerases mount up, a tipping point arrives: So many cells are starved for ATP that tissues become dysfunctional, and clinical symptoms begin to appear. But "it takes a lot of miscopying to do that," Copeland says. "That's why it takes so long for the disease to show up." The disease usually manifests in patients between 30 and 40 years of age. Copeland's point applies only to heterozygotes, however, since Van Broeckhoven's group did not find family members with PEO who were homozygous for the mutation; two Y955C mutations may be lethal to the fetus.
Mitochondrial failure in muscles produces muscle weakness. In eye muscles like the lateral rectus, mitochondria comprise 60% of the cell volume, which suggests both high ATP requirements and sensitivity to declines in availability of ATP. This may explain why the mutator polymerase, though it works throughout the body, strikes first by paralyzing the eyes. Later, problems develop in the brain, kidneys, and other tissues that require high levels of ATP.
The researchers suggest that the higher rate of base substitution by Y955C polymerase explains another PEO hallmark: multiple large-scale deletions in mtDNA. While the mutator polymerase creates both single-base mutations and deletions, it is by deletions that PEO is known, because deletions are the only mitochondrial DNA mutations clinicians readily detect. Finding point mutations requires cloning and sequencing of individual mitochondrial genomes, which is "very difficult," says Copeland; detecting deletions with PCR and Southern hybridization is far easier.
MtDNA deletions in PEO are commonly several kilobases long and occur primarily between short, direct repeats of 10 to 13 base pairs. Analysis of a 4,977-base pair deletion common in PEO provided a clue to the deletion mechanism. The deletions were associated with point mutations caused by T·dTMP mispairing, an event occurring 110 times more frequently with the mutator polymerase than wild type. To Copeland's team this suggested that when mispairing occurs following replication of a direct repeat, proofreading failure or failure to extend bases beyond the mismatch initiates DNA strand slippage. In turn, slippage allows transfer of the replication complex from one direct repeat to another, deleting sequences in between. Copeland observes that this model is consistent with analysis of deletions caused by a mutator polymerase (Klenow fragment) of Escherichia coli.
The reports from Van Broeckhoven and Copeland are the latest of several that nuclear genes can cause disease by introducing mutations in mtDNA. Another cause of PEO is a mutant mitochondrial helicase encoded by the gene called Twinkle. Adenine nucleotide translocator 1 (ANT1) and thymidine phosphorylase may decrease mtDNA stability by restricting intramitochondrial nucleotides.
A Vast Unknown
These reports only scratch the surface of what remains to be discovered. Mitochondrial diseases already number several dozen, including common illnesses and rarities like PEO. Mitochondria are beginning to be implicated in unusual cardiac problems, baffling kidney disorders, and unique cases of diabetes. The complexity of mitochondrial pathology derives not only from the interplay of different genomes, but also from the mitochondrial differences specific to cell type. By incorporating as many as 3,000 different proteins, mitochondrial function becomes tissue-specific; detoxification in the liver by mitochondrial enzymes is a well-known example. Notwithstanding the Copeland group's accomplishments, current research has illuminated only a small corner within the murky world of mitochondrial disease.
1. G. Van Goethem et al., "Mutation of POLG is associated with progressive external ophthalmoplegia characterized by mtDNA deletions," Nature Genetics, 28: 211-2, 2001.
2. M.V. Ponamarev et al., "Active site mutation in DNA polymerase associated with progressive external ophthalmoplegia causes error-prone DNA synthesis," Journal of Biological Chemistry, 277:15225-8, May 3, 2002.