Sci. Aging Knowl. Environ., 13 November 2002
Vol. 2002, Issue 45, p. vp6
[DOI: 10.1126/sageke.2002.45.vp6]


Error Catastrophe in Mutant Mitochondria

Laura L. Mays Hoopes

The author is in the Department of Molecular Biology/Biology at Pomona College, Claremont, CA 91711, USA. E-mail: Lhoopes{at};2002/45/vp6

Key Words: error catastrophe • DNA polymerase gamma • DNA replication • progressive external ophthalmoplegia • mitochondria


Leslie Orgel proposed in 1963 that aging might result from the inaccuracy permitted by information transfer processes, including DNA synthesis, transcription, and translation, even though the level of fidelity of these processes is high (1). According to this hypothesis, the processes would produce very few defective proteins early in life, but at some point, an inaccurate DNA polymerase would be produced that would introduce errors into DNA during synthesis, resulting in inaccuracies in an increasingly large set of mRNAs and their encoded proteins. Eventually, the catalytic proteins for information transfer would become so riddled with errors that almost no accurate protein synthesis would be possible, and thus catastrophe would result. The error catastrophe hypothesis attracted a good deal of attention in the two decades after it was proposed, but the idea that it could have a role in the usual processes of aging received little experimental support. Interestingly, however, a new finding implicates a process resembling error catastrophe in mitochondria in a human genetic disease.

Initial Studies

In the early investigations seeking evidence for or against the error catastrophe hypothesis, some investigators examined the information transfer apparatus directly. For example, DNA polymerases were shown by Loeb and Martin and their colleagues to exhibit the same level of fidelity whether they were isolated from aging humans and mice or from young members of these species (2-4). However, Linn (5) reported that DNA polymerase that had been semipurified from human fibroblasts late in replicative senescence (a state in which proliferation has slowed and will soon stop) produced more errors during DNA synthesis than an equivalent preparation from early-passage cells of the same cell line. With respect to translation, surprisingly, the adaptor molecules for protein synthesis--transfer RNAs (tRNAs)--were found to discriminate between amino acid analogs and the natural amino acid better when they were isolated from the livers of aging rather than young rats (6). However, changes in the proportions of tRNA isoacceptors (different tRNAs that recognize the same amino acid) also occur as rats age, implying either changes in tRNA gene expression or in posttranscriptional processing (7) and provide a possible explanation for the observation described above. Distinct tRNA isoacceptors have reproducible chromatographic differences, which can result either from expression of different tRNA genes or from different posttranscriptional modifications of the products of the same gene. If the isoacceptors are derived from different genes, they can usually recognize different mRNA codons, and changes in their proportions could potentially affect the efficiency of translation. Thus, the overall conclusion from studies such as these was that information transfer was not defective in aging animals, although it might be affected during replicative senescence.

Other investigators directly examined individual proteins produced by the information transfer process (8). For example, if the error catastrophe hypothesis held true, aging animals might be expected to produce abnormal proteins with altered size, charge, or other characteristics, which would appear as immunoreactive cross-reactive species in experiments involving antibodies that recognize the wild-type protein. It was found that when particular abundant enzymes were purified, such as triose phosphate isomerase from the nematode Turbatrix aceti (9) and creatine kinase from human muscle (10), no more immunoreactive cross-reacting material per unit of enzyme activity could be detected in samples isolated from old versus young organisms. This type of information was not always obtained, however. For example, aldolase preparations derived from the livers of old rats evidently consisted of both normal and heat-labile molecules, although the aldolase from the old rats did not exhibit an increase in immunologically cross-reacting material (11). More recently, this type of change has been attributed to age-dependent increases in the concentrations of proteins with posttranslational modifications, as well as increases in the concentrations of partially denatured proteins, probably caused by deficiencies in protein degradation (12, 13). In addition, isoelectric focusing was used to examine the charge distribution of purified enzymes (14, 15), and these studies did not uncover abnormally charged species in samples isolated from older organisms. Although some variants would indubitably be lost during purification, one would certainly expect to find more variability in the enzymes synthesized by old organisms if error catastrophe indeed played a role in the process of aging, and that result was not found.

One of the most interesting approaches used in these early investigations was to cause the relevant cells to synthesize viral particles, which could then be examined to determine whether the constituent proteins resembled those produced in young cells. These experiments were done using human diploid fibroblasts in replicative senescence. Tomkins et al. (16), for example, used polio virus (an RNA virus), as well as herpes simplex (a DNA virus), and found no changes in the proteins in either case, as assayed by acrylamide gel electrophoresis. Holland et al. (17) found that the same two viruses, plus vesicular stomatosis virus, were produced in normal quantities by the late-passage cells. This experiment supported the idea that the viral proteins, produced by viral pirating of the usual cellular machinery, were normal. In addition, because the proteins were functional by the criterion of forming part of infective virions, this experiment also implied that there was no selection for normal virions in the Tomkins et al. experiment. In other words, because normal numbers of virions were produced, the cell was not assembling a proportion of defective proteins that interfered with the function of the virion into these virus particles. This series of experiments, performed in the laboratories of well-regarded investigators, convinced most people that protein synthesis was normal in the cells in replicative senescence.

Many scientists interested in mechanisms of aging gave up the error catastrophe hypothesis with some reluctance; it proposed an attractive mechanism that looked general as well as testable. The alternatives often were organism-specific and vague, so that they were much less exciting to investigate. With time, the error catastrophe hypothesis ceased to attract investigations, having generated a great deal of information but no convincingly supportive data.

Error Catastrophe and Progressive External Ophthalmoplegia

That was then, this is now: Error catastrophe may really be occurring in people with progressive external ophthalmoplegia (PEO), a genetic disease. The catch is that the catastrophe is occurring in the mitochondria (18), meaning that the catastrophe is not general to the cell but specific to the information transfer processes occurring inside the mitochondria. PEO was recently discovered to result from mutation of the gene encoding DNA polymerase gamma, the polymerase that replicates mitochondrial DNA (mtDNA) circles (19). The symptoms of this disease begin with a restricted ability to use the muscles that would normally move the eye, and they progress to the point where the patient must move his or her head to follow a moving object. This disorder is benign at early ages, but after a late onset, PEO is progressive with age.

Ponamarev et al. (18) reasoned that because the Tyr955->Cys955 mutation found in DNA polymerase gamma in individuals with PEO affects a highly conserved tyrosine residue that is critical for recognition of the correct incoming nucleotides, the fidelity of the enzyme might be affected. These investigators purified both the normal and the mutant enzyme and found that the mutant enzyme retained a normal catalytic rate. However, this enzyme was decreased 45-fold in its binding affinity (or Michaelis constant, Km) for the incoming nucleoside triphosphate that pairs with the template strand nucleotide in the active site. Even though this enzyme does have an intrinsic proofreading (3' exonuclease) activity, the mutant version of the enzyme is overall two times less accurate than the normal version. The investigators were able to inactivate the proofreading exonuclease and thus found that the intrinsic mutator effect (without proofreading) was elevated 10- to 100-fold. Their findings provide an explanation for the accumulation of mtDNA point mutations in patients with PEO. One hypothetical outcome of these point mutations is shown in Fig. 1.

View larger version (25K):
[in this window]
[in a new window]
Fig. 1. Inaccurately synthesized membrane proteins could cause PEO mitochondria to explode, as depicted in this hypothetical model. Green circles represent accurately synthesized proteins, and red circles indicate inaccurately synthesized proteins. DNA (black lines) and DNA polymerase gamma (red ovals) are shown at the left in the mitochondrial matrix.

Not all PEO is attributed to mutations in this gene; for example, mutations in the Twinkle gene, which encodes a putative mtDNA helicase, also result in this phenotype (20). Ponamarev et al. suggest that the PEO-associated variant helicases might also lead to mutagenesis of the mitochondrial genome, because mutations affecting the DnaB helicase from E. coli are known to stall replication forks. These stalled forks could cause the induction of one of several error-prone repair pathways that are involved in postreplication repair of DNA. Of course, as we all know, there are human genetic diseases that many would argue lead to the acceleration of aspects of aging, which result from mutation in DNA helicase-encoding genes. However, the gene (WRN) that is affected in Werner syndrome, as well as the genes mutated in other such disorders, seem to encode helicases of the nucleus, not the mitochondria (see Fry Review and "Of Hyperaging and Methuselah Genes").

So the error catastrophe hypothesis lives after all, but as far as we now know, the phenomenon appears to be restricted to mitochondria in individuals with certain genetic disorders. Interestingly, Ponamarev et al. suggest that even wild-type DNA polymerase gamma has an error rate, implying that the same mechanism may be important in normal aging. Changes in mtDNA sequences have been found to occur during aging of normal humans. For example, Michikawa et al. (21) found that point mutations are present in the mtDNA of fibroblasts from normal old people but not from normal young people. Further, the number of mutations was found to increase when one individual was resampled as aging progressed, providing longitudinal data on the phenomenon. The authors particularly noted a T414G transversion that occurred in 8 of 14 individuals aged over 65 years, but not in 13 younger individuals. In addition, Wang et al. (22), using a specific assay for particular point mutations, found that in the postmitotic muscle tissue, most people tested between the ages of 53 and 92 had two specific point mutations in the control sites for DNA replication. Stay tuned, more mitochondrial error catastrophes may be on the horizon.

November 13, 2002
  1. L. Orgel, The maintenance of the accuracy of protein synthesis and its relevancy to aging. Proc. Natl. Acad. Sci. U.S.A. 49, 517-521 (1963).[Free Full Text]
  2. S. Agarwal, M. Tuffner, L. A. Loeb, DNA replication in human lymphocytes during aging. J. Cell Physiol. 96, 235-243 (1978).[CrossRef][Medline]
  3. M. Fry, J. Silber, L. A. Loeb, G. M. Martin, Delayed and reduced cell replication and diminishing levels of DNA polymerase-alpha in regenerating liver of aging mice. J. Cell Physiol. 118, 225-232 (1984). [CrossRef][Medline]
  4. K. Subba Rao, G. M. Martin, L. A. Loeb, Fidelity of DNA polymerase-beta in neurons from young and very aged mice. J. Neurochem. 45, 1273-1278 (1985). [CrossRef][Medline]
  5. S. Linn, M. Kairis, R. Holliday, Decreased fidelity of DNA polymerase activity isolated from aging human fibroblasts. Proc. Natl. Acad. Sci. U.S.A. 73, 2818-2822 (1976).[Abstract/Free Full Text]
  6. L. Mays-Hoopes, G. Cleland, J. Bochantin, D. Kalunian, J. Miller, W. Wilson, M. K. Wong, D. Johnson, O. K. Sharma, Function and fidelity of aging tRNA: In vivo acylation, analog discrimination, synthetase binding, and in vitro translation. Mech. Ageing Dev. 22, 135-149 (1983).[CrossRef][Medline]
  7. L. Mays, A. E. Lawrence, R. W. Ho, S. Ackley, Age-related changes in function of transfer ribonucleic acid of rat livers. Fed. Proc. 38, 1984-1988 (1979).[Medline]
  8. M. Rothstein, in CRC Handbook of Biochemistry in Aging, J. Florini, Ed. (CRC Press, Boca Raton, FL, 1981), pp. 103-111.
  9. S. Gupta, M. Rothstein, Triosephosphate isomerase from young and old Turbatrix aceti. Arch. Biochem. Biophys. 174, 333-338 (1976).[CrossRef][Medline]
  10. E. Steinhagen-Thiessen, H. Hilz, The age-dependent decrease in creatine kinase and aldolase activities in human striated muscle is not caused by an accumulation of faulty proteins. Mech. Ageing Dev. 5, 447-457 (1976).[CrossRef][Medline]
  11. A. Weber, C. Gregori, F. Schapira, Aldolase B in the liver of senescent rats. Biochim. Biophys. Acta. 444, 810-815 (1976).[Medline]
  12. B. S. Berlett, E. R. Stadtman, Protein oxidation in aging, disease, and oxidative stress. J. Biol. Chem. 272, 20313-20316 (1997).[Free Full Text]
  13. I. Petropoulos, M. Conconi, X. Wang, B. Hoenel, F. Bregegere, Y. Milner, B. Friguet, Increase of oxidatively modified protein is associated with a decrease of proteasome activity and content in aging epidermal cells. J. Gerontol. Ser. A Biol. Sci. Med. 55, B220-B227 (2000).[Abstract/Free Full Text]
  14. H. Sharma, S. Gupta, M. Rothstein, Age-related alterations in purified fructose-1,6-diphosphate aldolase from the nematode, Turbatrix aceti. Arch. Biochem. Biophys. 174, 324-332 (1976).[CrossRef][Medline]
  15. U. Reiss, D. Gershon, Rat liver superoxide dismutase: purification and age-related modifications. Eur. J. Biochem. 63, 617-623 (1976).[Medline]
  16. G. Tomkins, E. Stanbridge, L. Hayflick, Viral probes of aging in the human diploid cell strain WI-38 (38110). Proc. Soc. Exp. Biol. Med. 146, 385-390 (1974).[Abstract/Free Full Text]
  17. J. Holland, D. Kohne, M. Doyle, Analysis of virus replication in ageing human fibroblast cultures. Nature 245, 316-329 (1973).[CrossRef][Medline]
  18. M. Ponamarev, M. Longley, D. Nguyen, T. Kunkel, W. Copeland, Active site mutation in DNA polymerase associated with progressive external ophthalmoplegia causes error-prone DNA synthesis. J. Biol. Chem. 277, 15225-15228 (2002).[Abstract/Free Full Text]
  19. G. Van Goethem, Mutation of POLG is associated with progressive external ophthalmoplegia characterized by mtDNA deletions. Nature Genet. 28, 211-222 (2001).[CrossRef][Medline]
  20. J. Spellbrink, F. Li, V. Tiranti, K. Nikali, Q. Yuan, M. Tariq, S. Wanroodij, N. Garrido, G. Comi, L. Morandi, L. Santoro, A. Toscano, G. Farizi, H. Somer, R. Crozen, D. Beeson, J. Poulton, A. Suomalainen, H. Jacobs, M. Zeviani, C. Larsson, Human mitochondrial DNA deletions associated with mutation in the gene encoding Twinkle, a phage T7 gene4-like protein localized in mitochondria. Nature Genet. 28, 223-231 (2001).[CrossRef][Medline]
  21. Y. Michikawa, F. Mazzucchelli, N. Bresolin, G. Scarlato, G. Attardi, Aging-dependent large accumulation of point mutations in the human mtDNA control region for replication. Science 286, 774-779 (1999). [Abstract/Free Full Text]
  22. Y. Wang, Y. Michikawa, C. Mallidis, Y. Bai, L. Woodhouse, K. Yarasheski, C. Miller, V. Askanas, W. Engel, S. Bhasin, G. Attardi, Muscle-specific mutations accumulate with aging in critical human mtDNA control sites for replication. Proc. Natl. Acad. Sci. U.S.A. 98, 4022-4027 (2001).[Abstract/Free Full Text]
Citation: L. L. M. Hoopes, Error Catastrophe in Mutant Mitochondria. Science's SAGE KE (13 November 2002),;2002/45/vp6

Science of Aging Knowledge Environment. ISSN 1539-6150