Download Does premature aging of the mtDNA mutator mouse prove that

Aging Cell (2006) 5, pp279–282
Doi: 10.1111/j.1474-9726.2006.00209.x
Blackwell Publishing Ltd
COMMENTARY
Does premature aging of the mtDNA mutator mouse
prove that mtDNA mutations are involved in natural
aging?
Konstantin Khrapko,1 Yevgenya Kraytsberg,1
Aubrey DNJ de Grey,2 Jan Vijg3 and Eric A. Schon4
1
Gerontology Division, Beth Israel Deaconess Medical Center and
Harvard Medical School, Boston, MA, USA
2
Department of Genetics, University of Cambridge, Cambridge, UK
3
University of Texas Health Science Center at San Antonio, San
Antonio, TX, USA
4
Departments of Neurology and of Genetics and Development,
Columbia University, New York, NY, USA
Summary
Recent studies have demonstrated that transgenic mice
with an increased rate of somatic point mutations in mitochondrial DNA (mtDNA mutator mice) display a premature aging phenotype reminiscent of human aging. These
results are widely interpreted as implying that mtDNA
mutations may be a central mechanism in mammalian
aging. However, the levels of mutations in the mutator
mice typically are more than an order of magnitude
higher than typical levels in aged humans. Furthermore,
most of the aging-like features are not specific to the
mtDNA mutator mice, but are shared with several other
premature aging mouse models, where no mtDNA mutations are involved. We conclude that, although mtDNA
mutator mouse is a very useful model for studies of phenotypes associated with mtDNA mutations, the aging-like
phenotypes of the mouse do not imply that mtDNA mutations are necessarily involved in natural mammalian aging.
On the other hand, the fact that point mutations in aged
human tissues are much less abundant than those causing
premature aging in mutator mice does not mean that
mtDNA mutations are not involved in human aging. Thus,
mtDNA mutations may indeed be relevant to human
aging, but they probably differ by origin, type, distribution, and spectra of affected tissues from those observed
in mutator mice.
Key words: aging; human; mouse; mtDNA mutation.
Correspondence
Konstantin Khrapko, Gerontology Division, Beth Israel Deaconess Medical
Center and Harvard Medical School, Boston, MA, USA.
e-mail: [email protected]
Accepted for publication 8 February 2006
Two recent studies reported the creation of transgenic mice
with proofreading-deficient mitochondrial DNA polymerase
gamma (Trifunovic et al., 2004; Kujoth et al., 2005). In these mice,
high levels of mutations (i.e. over 10 mutations per 10 000 bp
or 15 per coding region of mitochondrial genome) accumulate
by the age of 25 weeks in all tissues studied (summarized in
Fig. 1, black bars). Intriguingly, both mouse lines present similar
premature aging phenotypes including osteoporosis, hair loss,
cardiomyopathy, anemia, sarcopenia, fertility problems, and
shortened lifespan. The authors of the reports concluded that
their data provided a ‘causative link between mtDNA mutations
and aging phenotypes in mammals’ (Trifunovic et al., 2004) and
that ‘mtDNA mutations … may be a central mechanism driving
mammalian aging’ (Kujoth et al., 2005).
Intact mitochondrial function is vital for the normal functioning of any cell. The presence of more than 15 random point
mutations per 15-kb-long coding region of the mitochondrial
genome (Fig. 1) almost entirely constituted of essential coding
sequences is likely to severely affect mitochondrial physiology.
This expectation is confirmed by a substantial drop of mitochondrial enzyme activities observed in heart tissue of the mutator mouse (Trifunovic et al., 2004). High levels of mutations are
particularly toxic to dividing cells: mtDNA mutator cell lines have
been shown to enter crisis at about 10 mutations per genome
(Spelbrink et al., 2000). Thus the severe multisystem phenotype
of the mutator mouse is not surprising.
To further explore whether premature aging in mutator mice
corroborates the hypothesis that mtDNA mutations are involved
in mammalian aging in general and in human aging in particular, we surveyed the data regarding levels of mtDNA mutations
in older humans (Fig. 1, white bars). We limited our search to
studies which used, similar to Trifunovic et al. (2004) and Kujoth
et al. (2005), the clone-by-clone sequencing approach to
measure the mutations (see online supplementary material for
details of the data analysis and validation). As shown in Fig. 1,
levels of mtDNA mutations in human tissues are more than an
order of magnitude lower than in mutator mice. In our opinion,
this huge gap in mtDNA mutant fractions makes it difficult to
conclude that the same types of mutations are causally related
to human aging.
One potential caveat in our argument is that mtDNA mutations may be so infrequent in old humans because they have
already fulfilled their role in the aging process by killing the cells
that had carried them, and thus had become ‘self-eliminated’.
The data, however, do not support this scenario. Most likely only
© 2006 The Authors
Journal compilation © Blackwell Publishing Ltd/Anatomical Society of Great Britain and Ireland 2006
279
280 Mutator mouse: a model of human aging? K. Khrapko et al.
Fig. 1 Fractions of somatic point mutations in the coding region of mutator
mice mtDNA as compared to human. The mutator mice data were measured
at mid-life (5–6 months) in duodenum and heart (right bar) (Kujoth et al.,
2005), and brain, heart (left bar), and liver (Trifunovic et al., 2004). The data
for adult and old humans (over 30 years old, mostly over 70 years old) have
been adapted from published reports and our unpublished data: leukocytes
(Monnat & Loeb, 1985a), colon (Taylor et al., 2003), skeletal muscle (Del Bo
et al., 2003a; Wanrooij et al., 2004), retina (Bodenteich et al., 1991), brain
(Simon et al., 2004; Smigrodzki et al., 2004), and heart (Kraytsberg &
Khrapko, unpublished); 95% confidence intervals are shown where available.
Note that the low levels of mtDNA mutations in humans cannot be explained
by preferential apoptosis of mutant cells (see text).
a small fraction of mtDNA mutations are capable of causing cell
death. Hence, their self-elimination should not have significantly
affected the total mutant fraction. If, on the other hand, the set
of deadly mutations were broad enough to cause a significant
reduction in the overall mutant fraction, then this should have
resulted in an equally significant enrichment of silent mutations
(i.e. mutations that do not cause an amino acid change and thus
are neutral). Normally silent mutations constitute about 25%
of all mutations in a protein-coding sequence. Elimination of a
large number of non-silent mutations sufficient for a severalfold reduction in overall mutant fraction should have increased
the fraction of silent mutations considerably, but this is not
observed in human tissues. For example, in colonic crypts (Taylor
et al., 2003), all five mutations in the protein-coding sequences
that were found in normal crypts change highly conserved amino
acids. Although more data would be desirable, those available
are clearly incompatible with a significant elimination of mutations
by apoptosis. Note that colon is a tissue with rapid cellular turn-
over, which is expected to be particularly susceptible to mtDNA
mutation-driven apoptosis. Similarly, no bias in favor of silent
mutations was observed in the much larger database of brain
mtDNA mutations (Smigrodzki et al., 2004): 127 silent mutations (24%) and 411 amino acid changing mutations (76%).
Analysis of the data in Fig. 1 could potentially be taken one
step further. The levels of mtDNA mutations associated with
premature aging in mice are so much higher than the levels of
point mutations present in aged human tissues that it seems
reasonable to ask whether mtDNA mutator mice actually demonstrate that mtDNA mutations might not be involved in human
aging! Such an idea would be consistent with an observation
that non-transgenic littermates of mutator mice accumulate five
4
mutations per 10 bp at 40 weeks (i.e. over five times more than
aged humans) and they still show no signs of aging (Trifunovic
et al., 2005) (see online supplementary material for discussion
of conflicting estimates of mutant fractions in wild-type mice).
This conclusion, however, would probably be incorrect, in part
because the types and distribution of mtDNA mutations in aging
humans may be very different from those in mutator mice,
which may allow them to play a role in aging despite their overall low abundance, as discussed below.
In the mutator mouse, the major type of mtDNA mutations
(i.e. polymerase proofreading errors) start to accumulate rapidly
early in embryogenesis [presumably immediately following
blastocyst implantation, when organellar division and mtDNA
replication resume (Smith & Alcivar, 1993)], and thus distribute
at high levels among most if not all tissues (Trifunovic et al., 2004;
Kujoth et al., 2005). In contrast, in humans the major sources
and the types of mutations (e.g. large-scale deletions vs. point
mutations, or proofreading errors vs. errors at chemically modified nucleotides), and thus the age- and tissue-specificity of
mutation accumulation, are probably different from those in the
mutator mouse. In accord with this, the mutational spectra of
mtDNA mutator human cells are different from the spectra of
human polymorphisms (Spelbrink et al., 2000).
The spectra of affected tissues and the relative timing of the
onset and mechanisms of pathology may be very different, too.
For example, in humans, the highest proportions of cells with
defects of mitochondrial function (about 20%) were found in
the substantia nigra in the brain (Itoh et al., 1996). Moreover,
these defects were caused not by point mutations, but by
large-scale mtDNA deletions (Kraytsberg et al., 2006), which
also account for a majority of the mitochondrial defects in
skeletal muscle (Gokey et al., 2004). Deletions, unlike point
mutations in the mutator mouse, accumulate much later in life,
and are extremely tissue-specific (Soong et al., 1992).
Finally, the intracellular distribution on mtDNA mutations may
differ between mutator mice and aged humans. Each potentially detrimental mtDNA mutation alone is benign below the
threshold of about 90% per cell, and several different mutations,
when present in the same cell, may complement each other’s
defects. mtDNA mutations are known to have a tendency to
expand clonally within cells, but the rate of this process is not
known. Admittedly, mutations may have had a better chance
© 2006 The Authors
Journal compilation © Blackwell Publishing Ltd/Anatomical Society of Great Britain and Ireland 2006
Mutator mouse: a model of human aging? K. Khrapko et al. 281
to expand clonally above the threshold within about 70 years
in an aged person compared to 6 months in mutator mouse.
Thus, we believe that the rapid generation of mtDNA mutations in a relatively small number of progenitor cells beginning
at the earliest stages of mouse embryonic development has a
disproportionately large effect on the mutant load in postnatal
life, and probably does not mimic the slow and tissue-specific
accumulation of mutations seen in normal human aging. As an
example, the presence of anemia in almost all of the mutator
mice stands in contrast to the lack of anemia in most (although
not all) patients with authentic mitochondrial disease harboring
high levels of mutant mtDNA in blood [e.g. the A8344G mutation in MERRF, myoclonus epilepsy with ragged-red fibers
(Silvestri et al., 1993)]. Perhaps a mouse with a conditional postnatal expression of an error-prone mtDNA polymerase, or one
expressing a less aggressively error-prone enzyme variant, or
a collection of mice with tissue-specific elevations of mtDNA
mutation rates, similar to that reported in cardiac tissue (Zhang
et al., 2000; Mott et al., 2005), might provide a model of natural
aging in mammals. The remarkably low abundance of mtDNA
mutations reported by the latter group in association with
severe cardiac dysfunction reinforces our view that there is
much more to be learned about the species- and tissuespecificity of pathologies associated with mitochondrial
mutations (see also online supplementary material for discussion
of mutant fraction estimates in mice). Another very recent example
of a mild mutator is the ‘deletor’ transgenic mouse expressing
a defective human Twinkle gene (Tyynismaa et al., 2005).
The above arguments are supported further by the independent observation that osteoporosis, curvature of the spine, hair
and weight loss, reduced fertility, anemia, and enhanced cell death
– i.e. those phenotypes that collectively make aging of the
mtDNA mutator mice particularly similar to human aging – are
not specific to these mice, as they are also present in other premature aging mouse models, for example in those with impaired
nuclear (but not mitochondrial) DNA metabolism (Hasty et al., 2003;
Chang et al., 2004; Lombard et al., 2005). It follows that while
the mutator mouse does show that large amounts of mtDNA
mutations can cause certain aging-like phenotypes, it does not
follow that mtDNA mutations actually cause similar phenotypes
in natural aging. Particular caution should be exercised in using
mutator mice ‘to test compounds intended for treating or
staving off osteoporosis, sarcopenia, and other age-related conditions’, as has been suggested in the media (Travis, 2004).
It is worth emphasizing that mutator mice, although they
are not, in our opinion, an appropriate test of the involvement
of mtDNA mutations in natural aging, are certainly an indispensable instrument for in vivo studies of the effects of somatic
point mutations in mtDNA. This is illustrated by the demonstration that somatic mtDNA mutations may not cause oxidative
stress in mutator mice (Kujoth et al., 2005; Trifunovic et al.,
2005), which is a powerful argument against the influential
‘vicious cycle’ hypothesis.
In conclusion, prematurely aging mtDNA mutator mice,
however important they are for the studies of somatic mtDNA
mutations, do not prove that mtDNA mutations are involved
in natural aging, although these mice do not exclude this
possibility either. We hope that this discussion will stimulate
new experiments to test the relationship between mtDNA
mutations and the aging process.
Acknowledgments
This work was supported in part by NIH grants ES 11343 and
AG 19787 to K.K.
References
Bodenteich A, Mitchell LG, Merril CR (1991) A lifetime of retinal light
exposure does not appear to increase mitochondrial mutations. Gene
108, 305–310.
Chang S, Multani AS, Cabrera NG, Naylor ML, Laud P, Lombard D,
Pathak S, Guarente L, DePinho RA (2004) Essential role of limiting
telomeres in the pathogenesis of Werner syndrome. Nat. Genet. 36,
877–882.
Del Bo R, Bordoni A, Sciacco M, Di Fonzo A, Galbiati S, Crimi M, Bresolin N,
Comi GP (2003a) Remarkable infidelity of polymerase gammaA
associated with mutations in POLG1 exonuclease domain. Neurology
61, 903–908.
Gokey NG, Cao Z, Pak JW, Lee D, McKiernan SH, McKenzie D, Weindruch
R, Aiken JM (2004) Molecular analyses of mtDNA deletion mutations
in microdissected skeletal muscle fibers from aged rhesus monkeys.
Aging Cell 3, 319–326.
Hasty P, Campisi J, Hoeijmakers J, van Steeg H, Vijg J (2003) Aging and
genome maintenance: lessons from the mouse? Science 299, 1355–
1359.
Itoh K, Weis S, Mehraein P, Muller-Hocker J (1996) Cytochrome c oxidase
defects of the human substantia nigra in normal aging. Neurobiol.
Aging 17, 843–848.
Kujoth GC, Hiona A, Pugh TD, Someya S, Panzer K, Wohlgemuth SE,
Hofer T, Seo AY, Sullivan R, Jobling WA, Morrow JD, van Remmen
H, Sedivy JM, Yamasoba T, Tanokura M, Weindruch R, Leeuwenburgh
C, Prolla TA (2005) Mitochondrial DNA mutations, oxidative stress,
and apoptosis in mammalian aging. Science 309, 481–484.
Lombard DB, Chua KF, Mostoslavsky R, Franco S, Gostissa M, Alt FW
(2005) DNA repair, genome stability, and aging. Cell 120, 497–512.
Monnat R Jr, Loeb LA (1985a) Nucleotide sequence preservation of
human mitochondrial DNA. Proc. Natl Acad. Sci. USA 82, 2895–2899.
Mott JL, Zhang D, Zassenhaus HP (2005) Mitochondrial DNA mutations,
apoptosis, and the misfolded protein response. Rejuvenation Res. 8,
216–226.
Kraytsberg Y, Kudryavtseva E, McKee AC, Geula C, Kowall NW,
Khrapko K (2006) DOI: 10.1038/ng1778.
Silvestri G, Ciafaloni E, Santorelli FM, Shanske S, Servidei S, Graf WD,
Sumi M, DiMauro S (1993) Clinical features associated with the
A → G transition at nucleotide 8344 of mtDNA (‘MERRF mutation’).
Neurology 43, 1200–1206.
Simon DK, Lin MT, Zheng L, Liu GJ, Ahn CH, Kim LM, Mauck WM, Twu F,
Beal MF, Johns DR (2004) Somatic mitochondrial DNA mutations
in cortex and substantia nigra in aging and Parkinson’s disease.
Neurobiol. Aging 25, 71–81.
Smigrodzki R, Parks J, Parker WD (2004) High frequency of mitochondrial
complex I mutations in Parkinson’s disease and aging. Neurobiol.
Aging 25, 1273–1281.
Smith LC, Alcivar AA (1993) Cytoplasmic inheritance and its effects on
development and performance. J. Reprod. Fertil. Suppl. 48, 31–43.
© 2006 The Authors
Journal compilation © Blackwell Publishing Ltd/Anatomical Society of Great Britain and Ireland 2006
282 Mutator mouse: a model of human aging? K. Khrapko et al.
Soong NW, Hinton DR, Cortopassi G, Arnheim N (1992) Mosaicism for
a specific somatic mitochondrial DNA mutation in adult human brain.
Nat. Genet. 2, 318–323.
Spelbrink JN, Toivonen JM, Hakkaart GA, Kurkela JM, Cooper HM,
Lehtinen SK, Lecrenier N, Back JW, Speijer D, Foury F, Jacobs HT (2000)
In vivo functional analysis of the human mitochondrial DNA
polymerase POLG expressed in cultured human cells. J. Biol. Chem.
275, 24818–24828.
Taylor RW, Barron MJ, Borthwick GM, Gospel A, Chinnery PF, Samuels DC,
Taylor GA, Plusa SM, Needham SJ, Greaves LC, Kirkwood TB, Tumbull
DM (2003) Mitochondrial DNA mutations in human colonic crypt stem
cells. J. Clin. Invest. 112, 1351–1360.
Travis J (2004) Dying before their time. Sci. News 166, 26–28.
Trifunovic A, Hansson A, Wredenberg A, Rovio AT, Dufour E, Khvorostov I,
Spelbrink JN, Wibom R, Jacobs HT, Larsson NG (2005) From the cover:
somatic mtDNA mutations cause aging phenotypes without affecting
reactive oxygen species production. Proc. Natl Acad. Sci. USA 102,
17993–17998.
Trifunovic A, Wredenberg A, Falkenberg M, Spelbrink JN, Rovio AT,
Bruder CE, Bohlooly YM, Gidlof S, Oldfors A, Wibom R, Tomell J,
Jacobs HT, Larsson NG (2004) Premature ageing in mice expressing
defective mitochondrial DNA polymerase. Nature 429, 417–423.
Tyynismaa H, Mjosund KP, Wanrooij S, Lappalainen I, Ylikallio E,
Jalanko A, Spelbrink JN, Paetau A, Suomalainen A (2005) Mutant
mitochondrial helicase Twinkle causes multiple mtDNA deletions and
a late-onset mitochondrial disease in mice. Proc. Natl Acad. Sci. USA
102, 17687–17692.
Wanrooij S, Luoma P, van Goethem G, van Broeckhoven C, Suomalainen
A, Spelbrink JN (2004) Twinkle and POLG defects enhance agedependent accumulation of mutations in the control region of
mtDNA. Nucleic Acids Res. 32, 3053–3064.
Zhang D, Mott JL, Chang SW, Denniger G, Feng Z, Zassenhaus HP
(2000) Construction of transgenic mice with tissue-specific acceleration of mitochondrial DNA mutagenesis. Genomics 69, 151–161.
Supplementary material
The following supplementary material is available for this article
online from http://www.blackwellsynergy.com
Appendix S1 Comments regarding Fig. 1, and Mutant fractions in non-mutator mice.
© 2006 The Authors
Journal compilation © Blackwell Publishing Ltd/Anatomical Society of Great Britain and Ireland 2006