- Email: [email protected]
Somatic mutations, genome mosaicism, cancer and aging
Available online at www.sciencedirect.com
ScienceDirect Somatic mutations, genome mosaicism, cancer and aging Jan Vijg Genomes are inherently unstable due to the need for DNA sequence variation in the germ line to fuel evolution through natural selection. In somatic tissues mutations accumulate during development and aging, generating genome mosaics. There is little information about the possible causal role of increased somatic mutation loads in late-life disease and aging, with the exception of cancer. Characterizing somatic mutations and their functional consequences in normal tissues remains a formidable challenge due to their low, individual abundance. Here, I will briefly review our current knowledge of somatic mutations in animals and humans in relation to aging, how they arise and lead to genome mosaicism, the technology to study somatic mutations and how they possibly could cause nonclonal disease. Addresses Department of Genetics, Albert Einstein College of Medicine, New York, United States Corresponding author: Vijg, Jan ([email protected], http://www.einstein.yu.edu)
Current Opinion in Genetics & Development 2014, 26:141–149 This review comes from a themed issue on Molecular and genetic bases of disease Edited by Cynthia T McMurray and Jan Vijg
http://dx.doi.org/10.1016/j.gde.2014.04.002 0959-437X/Published by Elsevier Ltd.
Introduction Structure and function of all organisms are ultimately determined by their genome, the complete set of species-specific hereditary information. Genomes are sufficiently stable to continuously provide function over the life time. Yet, they are far from immutable. Indeed, mutations in germline genomes are the substrate of evolution by natural selection and therefore necessary to perpetuate and diversify life. If mutation rates would be zero, all alleles would be fixed and organisms could not adapt [1]. The tendency of genomes to undergo mutations, that is, addition, deletion, or substitution of bits of genetic code, is known as genome instability and the inevitable and irreversible result of errors made during the repair or replication of damaged DNA. Because they generally occur at random, most mutations have adverse effects. Natural selection often eliminates genomes with unfavorable mutations through the demise www.sciencedirect.com
of their carriers. Some mutations, however, convey positive attributes and will be picked up by natural selection to be perpetuated. It is through mutations that Darwinian selection could lead to increasingly complex genomes and the adaptation of their hosts to the various challenges of a continuously changing environment. Genomes are not only subject to variation in the germ cells transmitted by parents to offspring. New variation, through DNA mutations, accumulates between the germ lines in the soma. Like in the germ line, mutations in somatic tissues are generally random and rarely beneficial; most have either no effect or an adverse effect. While in natural environments deleterious mutations can be eliminated from the germ line by selection, in somatic tissues of metazoans selection is less efficacious in removing mutations or not at all possible. It is for this reason that the gradual accumulation of mutations in the genome of somatic cells has been implicated in aging, that is, the timedependent process of deteriorative change that ultimately brings life to a close [2,3–5]. Figure 1 schematically depicts this double-edged sword of genome instability. Genome instability is known to cause disease in humans, most notably cancer. Cancer arises from single cells as a consequence of repeated cycles of mutagenesis and selection for growth advantage and other attributes that promote tumor progression. However, little is known about the effect of randomly induced mutations on the function of individual cells in a tissue, without selection. Depending on the frequency at which mutations occur and their distribution across the genome of individual cells it is conceivable that the resulting increased tissue mosaicism leads to functional deterioration at old age. The role of somatic mutations in diseases other than cancer will be discussed elsewhere in this special issue. Here, I will discuss how somatic mutations arise and accumulate in aging organisms, leading to genome mosaics, the difficulties in studying somatic mutations and genome mosaicism, how somatic mutations cause cancer and, finally, how random mutations can lead to age-related functional decline.
Somatic mutations lead to genome mosaicism The concept that all somatic cells have identical genomes needs to be replaced by a much more dynamic model of an increasingly mosaic genome. The engine behind this is somatic mutagenesis. Somatic mutations are low-frequency events and only directly detectable when they are amplified in a clonal lineage (Figure 2). Using selectable marker genes in transgenic mice, in the past we have conclusively demonstrated the accumulation of mutations during aging in a variety of organs and tissues, Current Opinion in Genetics & Development 2014, 26:141–149
142 Molecular and genetic bases of disease
Figure 1
Mutations Evolutionary change
Germline
Soma
Disease, Aging?
RNA, protein Cellular function Current Opinion in Genetics & Development
The genome harbors all the instructions for providing function to the somatic cells of an organism. Random alterations in its information content in the germ line drive evolutionary change, while similar changes in the somatic cells could be the cause of aging.
including liver, brain, heart, small intestine, and spleen [6–8]. The rate of this increase as well as the spectrum of mutations (basepair substitutions, deletions, translocations) differed greatly from organ to organ. For example, while in the intestine point mutations accumulated rapidly to a very high level at old age, in heart and liver also the frequency of genome rearrangements was significantly higher at old age [8]. Over time, accumulating somatic mutations essentially turns an aging tissue into a mosaic of cells with different genotypes. Owing to the emergence of powerful new
Figure 2
Tumor
Clonal lineage
Development and aging Current Opinion in Genetics & Development
Schematic depiction of mutation accumulation during development and aging. Cancer-causing mutations (red check mark) are actively selected for, resulting in the clonal development of a tumor (beige). Occasionally, a mutation is subject to genomic drift (green check mark) and also leads to clonal amplification (green). Current Opinion in Genetics & Development 2014, 26:141–149
methods for analyzing copy number variation (amplifications and deletions) of DNA segments defined by single nucleotide polymorphisms, deletions were observed in blood and other tissues of humans and mice [9,10]. Such deletions must find their origin in the creation of somatic subpopulations of cells due to expansion of a single, de novo mutation. Indeed, an age-related accumulation of clonally expanded, copy number variants has been found [11–13]. Similarly, the age-related emergence of genomic differences between monozygotic twins who were identical at early age has been observed and interpreted as originating from somatic mutation [14]. These findings are highly likely to be only the tip of an iceberg because even the most recent 1 million-probe array cannot detect deletions smaller than 10 kb by comparative genomic hybridization (CGH) [15]. Moreover, mutations occurring in less than 10% of all cells are no longer detectable. Hence, direct methods can only detect sizable somatic genome alterations and then only as clonally amplified mutations affecting a substantial fraction of the tissue or cell population. There is also a large body of evidence that chromosomal aberrations, which are microscopically detectable using fluorescence in situ hybridization (FISH), capable of detecting alterations affecting no less than 1–5 million basepairs, accumulate in blood lymphocytes of both humans and mice [16,17]. Especially aneuploidy, that is, the loss or gain of whole chromosomes resulting in an abnormal numerical karyotype, appears to be remarkably high, even in postmitotic tissue [18,19]. Chromosomal aneuploidy is a hallmark of pathological conditions and a causal factor of birth defects and cancer. Using advanced FISH technology it has been shown that in the developing nervous system of humans and mice the frequency of aneuploid cells is as high as 33% for all chromosomes combined [20,21]. This extraordinary high level of genome instability during development is in keeping with observations of whole-chromosome aneuploidy and segmental deletions and duplications in individual blastomeres of cleavage-stage embryos after in vitro fertilization (IVF) [22]. Interestingly, although IVF success rates are typically in the order of 25%, in only 9% of these embryos was such genomic variation not present. This indicates extensive selection of non-mutated or less severely mutated blastomeres in the early embryo, which also suggests that the observed high frequency of aneuploidy during development can decline thereafter (through selection) and not be fatal. Indeed, In human or mouse adult brain, in which only few chromosomes were analyzed by interphase FISH, aneuploidy frequencies were found to be much lower than in developmental mouse brain, that is, 10% for all chromosomes combined. However, this is still high and we have shown that in the mouse brain aneuploidy accumulates with age, up until about 50% for all chromosomes combined in 28-month-old animals [19]. www.sciencedirect.com
Genome mosaicism and aging Vijg 143
Somatic mutation rates differ among different types of genomic sequence. Indeed, in regions containing repeat elements, such as mini-satellites and micro-satellites, retrotransposons and telomeres, spontaneous mutations can occur at a much higher rate than in single-copy regions. Mutations at microsatellite loci have been found to occur at rates as high as 1 102 per locus in human blood and increase with age [23]. Retrotransposition has been observed in rodent brain in vivo [24] and deep sequencing has revealed insertions of retrotransposons in Drosophila and human brain [25,26]. In yeast, during replicative aging, increased retrotransposition has been observed concomitant with a loss of nucleosomes, a cause of aging in budding yeast [27]. Telomeres, regions of repetitive DNA protecting chromosome ends from deterioration, significantly shorten with age in mammalian cells and tissues [28], in part due to the end replication problem [29]. In conclusion, somatic mutations accumulate with age in a tissue-specific manner, turning tissues into genome mosaics. Direct evidence for such mosaics has been obtained only for clonally amplified structural variations greater than 10 kb. For elucidating the complete landscape of somatic mutations over the life time of an individual new, advanced methodology is necessary.
Single-cell genomics to determine the in vivo landscape of somatic mutations Rapid progress in nucleotide sequencing technology [30,31] now allows the detection of even minute postzygotic mutational differences between the genomes of monozygotic twins [32]. However, as we have seen, apart from very large chromosomal alterations, mutations present in less than 10% of all cells in a tissue cannot be detected by direct analysis, but requires selectable markers. In the above discussed examples somatic mosaicism could only be detected because the mutation occurred early during development and/or aging, thereby eventually comprising a sizable fraction of cells. Such mutations are the tiny tip of what surely must be a gigantic iceberg of mutations, many unique for a particular individual cell. Therefore, we have no real insight into the true number of genome variants in the different cells comprising a tissue. Indeed, virtually our entire knowledge of the genome has come from studying cells in bulk, that is, as mixtures of cells. In principle one could detect somatic mutations in DNA from bulk cell populations by sequencing thousands of times across the genome and call all possible variants of the consensus sequence. In practice this would be very inefficient and expensive, and also essentially constrained by the high rate of sequencing errors, which can amount to close to 1% [33,34]. Genuine somatic mutations cannot be distinguished from sequencing errors, which are scattered through the reads (Figure 3). The problem is very similar to the situation in tumors in which we are also unable to glimpse behind www.sciencedirect.com
the relatively few mutations common to all or most cells in that tumor [35]. One possible solution for this problem is to sequence single cells after whole genome amplification and analyze combinations of mutations in an integrated manner. The latter is essential for assessing if random mutations are truly capable of adversely affecting cellular function (see below). We recently developed a method to determine somatic mutation frequencies and spectra across the genome of a tissue or cell population by sequencing a representative number of single cells after whole genome amplification [36]. After amplification, mutations unique for each cell will now occur in half of all sequencing reads for a particular locus. Only half of all reads, because a true mutation will affect one allele only, since it is highly unlikely that the other allele will also be hit at that same site (Figure 3). The single-cell protocol we developed was validated by testing if an elevation in mutation frequency in Drosophila S2 cells or mouse embryonic fibroblasts could be detected after treatment with a powerful mutagen (i.e., ethylnitrosourea; ENU). The results indicated a dramatically higher level of mutations in the exposed cells as compared to untreated control cells [36]. This is expected because ENU is a powerful mutagen. The exposed cells also revealed increased levels of A:T ! T:A mutations, which are known to be highly enriched following treatment with alkylating agents. While in these initial experiments we only called single nucleotide variants (i.e., point mutations), it is possible to also identify structural variations in this way using a socalled paired-end mode of sequencing [30,37]. When DNA fragments of a particular size are sequenced from both ends the paired reads should be positioned at a known distance from each other when aligned to a reference sequence (Figure 4). If the distance between the read pairs differs from the distance between the mapped positions on the reference genome, then a deletion or insertion is implied. It is also possible that one of the two read pairs maps to another chromosome, indicating a chromosomal translocation. Polymorphic variants with the reference genome are subtracted by always including a sequencing run from the non-amplified, whole tissue or whole cell population. The procedure allows us to call all possible somatic mutations present in a particular single cell. Genome amplification is often considered a major source of false positive mutations. For example, the most popular whole genome amplification method is multiple displacement amplification (MDA), which has been shown to cause false positive inversion mutations [38]. However, for point mutations such artifacts are unlikely. An error introduced early in MDA would be found in only 12.5% of the sequencing reads at a given locus in a diploid cell on average [39]. Since as shown in Figure 3 Current Opinion in Genetics & Development 2014, 26:141–149
144 Molecular and genetic bases of disease
Figure 3
CAT TTA GT T TG AT GT TGG C TAT C G C C C C C
T
C
T Sequencing errors
CAT TTA GT T TG AT GT TGG C TAT C G C G
Single Nucleotide Polymorphism
G Whole genome amplification
G G
T
C C C
G
C
G True somatic mutation
Current Opinion in Genetics & Development
Single-cell analysis is required for detecting low-abundant, random mutations. The A > G mutation (red check mark) can only be distinguished from sequencing errors after single-cell, whole genome amplification when it shows up in 50% of the reads (red; one mutated allele). Polymorphic variants (SNPs; blue) are also observed in the unamplified control DNA extracted from the bulk cell population. SNPs are variants between the genome of our cells and the reference genome. They are electronically discarded.
true somatic mutations should affect one allele and therefore need to be found in 50% of the reads, amplification errors are easily filtered out. Also, recent developments in whole genome amplification have now greatly diminished concerns about amplification artifacts [40]
and, albeit more difficult than SNVs, structural variations can be detected in single cells [30]. Indeed, recent work demonstrated the detection of copy number variation in single human neurons after whole genome amplification [10].
Figure 4
100bp
100bp
A to C PE seq RefSeq
PE seq RefSeq
500bp
chr2
chr3
PE seq RefSeq
800bp
Paired end sequencing
Paired end mapping
Single base substitution
300 bp deletion
Translocation Current Opinion in Genetics & Development
Paired-end sequencing allows the identification of a large variety of structural variation as discrepancies after aligning the sequencing reads to the reference genome. Current Opinion in Genetics & Development 2014, 26:141–149
www.sciencedirect.com
Genome mosaicism and aging Vijg 145
The great advantage of opening up next-generation sequencing to the analysis of somatic mutations is that, for the first time, all possible mutations, including low-abundant mutations, can be analyzed on a genome-wide basis. The latter is critically important because mutations are not uniformly distributed across the genome, but subject to significant constraints, including, for example, chromatin structure, transcription, the local structure of the genomic regions studied (e.g., repeats versus single-copy DNA) [41]. To systematically determine complete mutation spectra for a large number of single cells or nuclei from multiple somatic tissues at different age levels should allow us to unravel the impact of genome mosaicism on health and disease. Below I will illustrate how somatic mutations are responsible for cancer, a major age-related disease, but possibly also for non-cancer, degenerative aspects of the aging process.
Somatic mutations can cause both cancer and aging While cancer can occur also in young people, the incidence and mortality of most cancers increases exponentially with age [42]. While other factors, such as agerelated changes in immune function [43], apoptosis or autophagy [44], may also contribute, the mechanistic basis of the relationship between cancer and aging most likely involves a lifetime of DNA mutation accumulation as a natural consequence of inevitable errors during repair or replication of damaged DNA [5,45]. Indeed, DNA damage is ubiquitous, as a result of both endogenous processes, such as hydrolysis, oxidation and alkylation, and environmental genotoxicants [46]. While most of that damage is efficiently repaired, errors during replication and repair are inevitable and irreversible once the original template is lost. Exposure to environmental mutagens is known to shift the age-incidence curves of cancer towards accelerated risk, most likely by increasing mutation rates. The best example is tobacco smoke which, increases cancer risk in a clear dose-response relationship [47,48]. Indeed, lung tumors from smokers have an average mutation frequency more than 10-fold higher than lung tumors from non-smokers [49]. The reason that the accumulation of somatic mutations with age is so readily adopted as a likely cause of the agerelated increase in cancer risk is the universally accepted model of cancer initiation and progression. This model, first proposed by Fearon and Vogelstein in 1990 for colorectal cancer [50], posits a series of sequential mutations that first provide a selective growth advantage to the target cell(s) and subsequently other favorable attributes, such as the capacity to invade tissues, suppress immune responses and inactivate other anti-tumor responses, such as apoptosis. Eventually, mutations are also responsible for resistance to anti-cancer drugs. Each mutation is followed by clonal outgrowth in which a particularly advantageous mutation is selected. The www.sciencedirect.com
eventual malignant neoplasm shows a mutation spectrum that reflects its clonal history, but also contains so-called passenger mutations, that is, mutations that were not selected because of a clonal growth advantage and therefore did not contribute to cancer development. Many of those mutations had been accumulated during development and aging before tumor initiation [51]. The tumor mutation spectrum often also reveals the original mutagens responsible for the tumor-inducing mutations, for example, C:G to A:T in lung tumors from smokers and C:G to T:A mutations in the same tumor type from non-smokers. In this way, initially rare somatic mutations can eventually become predominant through natural selection. It is now clear that the number of somatic mutations in tumors is significantly higher when the tumor was derived from an old patient as compared to a young one [52,53]. Indeed, mathematical modeling strongly suggests that half or more of somatic mutations in tumors arise before initiation of the tumor, that is, during development and aging [51]. We still know very little about the mutational heterogeneity of normal tissues. Also intratumor variation in mutation spectra remains an enigma [35]. Nevertheless, the concept of age-related mutation accumulation to explain the age-related increase in cancer is straightforward. Such mutation-driven loss of proliferative homeostasis, giving rise to hyperplasias, benign neoplasms and malignant neoplasms, is a major component of the aging phenotype and likely to be causally related to age-related mutation accumulation in cells from self-renewing tissues [54]. By contrast, a causal role of mutations in tissue functional decline would rely on a cumulated mutation load in many or most cells of an organ or tissue that is high enough to have adverse effects without selection. Recently, a series of observations revealed that defects in genome maintenance pathways cannot only lead to increased cancer but also to multiple symptoms of cell and tissue degeneration that are normally only seen at old age. For example, heritable mutations in RecQ helicase genes cause segmental progeroid syndromes, such as Werner Syndrome [55]. Also mice harboring defects in DNA double-strand break repair or nucleotide excision repair display signs of premature aging, sometimes at very early age [56]. While this strongly suggests that DNA damage and/or mutations are critical factors in aging and its disease sequelae, there is now also recent evidence that cancer survivors who have been treated with DNA damaging chemotherapeutic agents display multiple symptoms of premature aging. Indeed, while in the past surviving cancer was relatively rare, this has now become more common. Interestingly, an ever growing number of adults who survived childhood cancer, is now diagnosed with health problems common to much older people, such as frailty, low muscle mass, slow walking speed, weakness and memory impairment [57]. While this could Current Opinion in Genetics & Development 2014, 26:141–149
146 Molecular and genetic bases of disease
be, at least in part, an effect of the cellular responses to DNA damage, such as apoptosis and cellular senescence, the secondary cancers that are also observed [58] point towards DNA mutations as another possible cause. As described in more detail by Erickson elsewhere in this special issue, both de novo and post-zygotic mutations are increasingly found to contribute to human disease, with the former now a known cause of autism [59,60] and the latter already known from disorders where the manifestations are readily seen, for example, the skin in type I neurofibromatosis [61]. Clearly, somatic mutations occurring early during development are likely to have a greater effect on phenotype expression than mutations occurring later and virtually all examples derive from clonal lineages with the disease-causing somatic mutation occurring during embryogenesis. In many cases the critical somatic mutation is a second mutation turning a recessive germ line mutation in one allele into a dominant phenotype after somatic inactivation of the other allele [62]. An interesting example of genome mosaicism as a cause of neurodegenerative disease is the observed somatic mutation in the presenilin-1 gene in a sporadic earlyonset patient with Alzheimer’s disease [63]. There is currently no direct evidence that somatic mutations causally contribute to aging and late-life disease other than cancer. Somatic mutations have been found in atherosclerotic plaques, but their causal role in atherogenesis and vascular disease development is uncertain [64]. Indeed, the functional impact of somatic mutagenesis other than through clonal amplification is presently unclear. The question is whether random mutations affecting single genes at low frequency can have an effect at the tissue level when not amplified through clonal outgrowth. While the current lack of detailed information about the landscape of somatic mutations in aged individuals prevents answering this question, it is certainly conceivable that random mutations affect cellular function. Indeed, function is not provided by single genes but through often extensive, functional networks consisting of genes and their regulatory sequences. Together, this would comprise a sizable genomic target for mutation. Each and every mutational hit in a sequence that is part of this network has the potential to adversely affect function. For example, for a typical differentiated human cell we can conservatively estimate that about 1% of the diploid genome (60 Mb) is functional [65]. While the germline de novo mutation frequency in humans is only about 60 new mutations per genome (1.2 108/bp or 0.01/ Mb [66]), somatic mutation frequencies are much higher than germline mutation frequencies [67]. Indeed, the mutation frequency in human tumors that have been subjected to whole genome sequencing was found to vary from about 0.1/Mb to almost 200/Mb [52], with Current Opinion in Genetics & Development 2014, 26:141–149
the median at about 6/Mb. As mentioned above, many mutations found in tumors have accumulated in the normal cells before tumor initiation, which is the most likely reason that tumors from children contain fewer mutations than tumors from adults or older individuals [51,68]. Our own unpublished data, obtained using the new, single-cell sequencing procedure (see above), indicate about 1 single nucleotide variant (SNV) per Mb in liver hepatocytes from 27-month old mice. This excludes structural variants (SVs), which are much less frequent, but likely to have more dramatic effects. In the 60 Mb of the genome that presumably is functional in a differentiated cell, an SNV frequency of about 1/Mb would then result in about 60 mutations. A fraction of these mutations could potentially adversely affect functional networks in that cell. Our data indicate that 5% of all somatic mutations in exons are nonsense mutations, with about 60% missense mutations (unpublished). This fraction of non-synonymous mutations is much higher than that in the germline genome because of the absence of selection. Including the SVs, which also accumulate with age [6], it certainly is conceivable that random mutations accumulating during aging do have functional effects other than clonal disease.
Summary and future prospects Like germ line mutations, somatic mutations are part of the logic of life. But while germ line mutagenesis is a sine qua non for the perpetuation of life, mutations in the soma are not needed and could very well lead to an error catastrophe, as originally proposed by Orgel for proteins [69]. In principle, evolution could postpone catastrophe by evolving ever more sophisticated genome maintenance systems to gradually increase species-specific longevity [70]. However, evolution by natural selection is a process not conducive to maximization of life span beyond the age of first reproduction [71]. Somatic mutation has been coopted during evolution to generate the means of creating immunoglobulin diversity, a critical factor in the immune response, and has also been suggested to play a role in generating diversity in the brain to broaden the spectrum of phenotypes that can originate from a single genome [72]. However, somatic mutations are unlikely to play a major role in human physiology and most of them are harmful and could contribute to aging. In summary, somatic mutations appear to be much more frequent than initially thought, reactivating original ideas about the causes of aging and age-related disease [2,3]. Indeed, those hypotheses can now be more firmly outlined. As schematically depicted in Figure 5, mutations inevitably and irreversibly accumulate during development and aging. They become manifest as clonal lineages causing cancer and other cell proliferative disorders. However, conceivably when a critical level is reached, www.sciencedirect.com
Genome mosaicism and aging Vijg 147
Figure 5
Neoplastic or dysplastic lesion
+ : cell depletion
Mutation accumulation, functional decline, cancer
Development and aging
+
+
Diseased or aged tissue
Young tissue
Current Opinion in Genetics & Development
Age-related, cell functional divergence as a consequence of stochastic effects. Functional decline is indicated by increasing darkness. Even in a young tissue, the function of highly differentiated cells in an organ or tissue is never maximized. The old tissue is different in the sense that there are many more cells that have suffered functional decline, with some of them dying (cross) or eliminated (open space). Others have grown into a hyperplastic or neoplastic lesion (not shown) or have been replaced by fibrosis (not shown).
mutations begin to adversely affect cell-specific functional networks, resulting in a mosaic of cells harboring different mutation loads and varying levels of functional deficiency. Depending on the severity of the mutation load, the cell type, genetic and environmental factors, this would lead to functional decline, disease and, possibly, aging.
References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest 1.
Baer CF, Miyamoto MM, Denver DR: Mutation rate variation in multicellular eukaryotes: causes and consequences. Nat Rev Genet 2007, 8:619-631.
2. Failla G: The aging process and carcinogenesis. Ann N Y Acad Sci 1958, 71:1124-1135. The first proposal of DNA mutations as a cause of aging. 3.
Szilard L: On the nature of the aging process. Proc Natl Acad Sci U S A 1959, 45:30-45.
4.
Curtis H: Biological mechanisms underlying the aging process. Science 1963, 141:686-694.
5.
Vijg J: Aging of the Genome. New York: Oxford; 2007.
6.
Dolle´ ME, Giese H, Hopkins CL, Martus HJ, Hausdorff JM, Vijg J: Rapid accumulation of genome rearrangements in liver but not in brain of old mice. Nat Genet 1997, 17:431-434.
7.
Dolle´ ME, Snyder WK, Dunson DB, Vijg J: Mutational fingerprints of aging. Nucleic Acids Res 2002, 30:545-549.
8.
Dolle´ ME, Snyder WK, Gossen JA, Lohman PH, Vijg J: Distinct spectra of somatic mutations accumulated with age in mouse
www.sciencedirect.com
heart and small intestine. Proc Natl Acad Sci U S A 2000, 97:8403-8408. 9.
Piotrowski A, Bruder CE, Andersson R, Diaz de Stahl T, Menzel U, Sandgren J, Poplawski A, von Tell D, Crasto C, Bogdan A et al.: Somatic mosaicism for copy number variation in differentiated human tissues. Hum Mutat 2008, 29:1118-1124. One of the first reports indicating that humans are commonly affected by somatic mosaicism for CNVs in a substantial fraction of cells.
10. McConnell MJ, Lindberg MR, Brennand KJ, Piper JC, Voet T, Cowing-Zitron C, Shumilina S, Lasken RS, Vermeesch JR, Hall IM et al.: Mosaic copy number variation in human neurons. Science 2013, 342:632-637. The authors used single-cell sequencing of endogenous human frontal cortex neurons to demonstrate that genomes in a sizable subset of neurons are marked by de novo CNVs. These results show that mosaic CNV is abundant in human neurons. 11. Jacobs KB, Yeager M, Zhou W, Wacholder S, Wang Z, RodriguezSantiago B, Hutchinson A, Deng X, Liu C, Horner MJ et al.: Detectable clonal mosaicism and its relationship to aging and cancer. Nat Genet 2012, 44:651-658. 12. Laurie CC, Laurie CA, Rice K, Doheny KF, Zelnick LR, McHugh CP, Ling H, Hetrick KN, Pugh EW, Amos C et al.: Detectable clonal mosaicism from birth to old age and its relationship to cancer. Nat Genet 2012, 44:642-650. 13. Forsberg LA, Rasi C, Razzaghian HR, Pakalapati G, Waite L, Thilbeault KS, Ronowicz A, Wineinger NE, Tiwari HK, Boomsma D et al.: Age-related somatic structural changes in the nuclear genome of human blood cells. Am J Hum Genet 2012, 90: 217-228. 14. Bruder CE, Piotrowski A, Gijsbers AA, Andersson R, Erickson S, de Stahl TD, Menzel U, Sandgren J, von Tell D, Poplawski A et al.: Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet 2008, 82:763-771. 15. Przybytkowski E, Ferrario C, Basik M: The use of ultra-dense array CGH analysis for the discovery of micro-copy number alterations and gene fusions in the cancer genome. BMC Med Genom 2011, 4:16. Current Opinion in Genetics & Development 2014, 26:141–149
148 Molecular and genetic bases of disease
16. Ramsey MJ, Moore DH II, Briner JF, Lee DA, Olsen L, Senft JR, Tucker JD: The effects of age and lifestyle factors on the accumulation of cytogenetic damage as measured by chromosome painting. Mutat Res 1995, 338:95-106. The authors demonstrated that chromosomal aberrations in peripheral blood lymphocytes of humans increase with age. 17. Tucker JD, Spruill MD, Ramsey MJ, Director AD, Nath J: Frequency of spontaneous chromosome aberrations in mice: effects of age. Mutat Res 1999, 425:135-141. 18. Iourov IY, Vorsanova SG, Yurov YB: Chromosomal mosaicism goes global. Mol Cytogenet 2008, 1:26. 19. Faggioli F, Wang T, Vijg J, Montagna C: Chromosome-specific accumulation of aneuploidy in the aging mouse brain. Hum Mol Genet 2012, 21:5246-5253. 20. Yurov YB, Iourov IY, Vorsanova SG, Liehr T, Kolotii AD, Kutsev SI, Pellestor F, Beresheva AK, Demidova IA, Kravets VS et al.: Aneuploidy and confined chromosomal mosaicism in the developing human brain. PLoS ONE 2007, 2:e558. 21. Rehen SK, McConnell MJ, Kaushal D, Kingsbury MA, Yang AH, Chun J: Chromosomal variation in neurons of the developing and adult mammalian nervous system. Proc Natl Acad Sci U S A 2001, 98:13361-13366. One of the first demonstrations that neurons in the mammalian nervous system do not possess identical genomes. Evidence is presented that approximately 33% of mouse neuroblasts are aneuploid, about 10 times as much as in adult lymphocytes. In the adult cortex aneuploidy is decreased, possibly because aneuploid cells are prone to cell death. 22. Vanneste E, Voet T, Le Caignec C, Ampe M, Konings P, Melotte C, Debrock S, Amyere M, Vikkula M, Schuit F et al.: Chromosome instability is common in human cleavage-stage embryos. Nat Med 2009, 15:577-583. This study establishes that chromosome instability is also common during early IVF embryogenesis. The results strongly suggest that in addition to selection against chromosomally abnormal embryos, which is well-documented, selection at the cellular level occurs against abnormal blastomeres.
33. Quail MA, Swerdlow H, Turner DJ: Improved protocols for the illumina genome analyzer sequencing system. Curr Protoc Hum Genet 2009. Chapter 18: Unit 18.2. 34. Kinde I, Wu J, Papadopoulos N, Kinzler KW, Vogelstein B: Detection and quantification of rare mutations with massively parallel sequencing. Proc Natl Acad Sci U S A 2011, 108: 9530-9535. 35. Salk JJ, Fox EJ, Loeb LA: Mutational heterogeneity in human cancers: origin and consequences. Ann Rev Pathol 2010, 5: 51-75. 36. Gundry M, Li W, Maqbool SB, Vijg J: Direct, genome-wide assessment of DNA mutations in single cells. Nucleic Acids Res 2012, 40:2032-2040. The authors show that it is possible to directly measure somatic mutation frequencies and spectra in cell populations, not using selectable markers but through genome-wide sequencing of single cells. 37. Campbell PJ, Stephens PJ, Pleasance ED, O’Meara S, Li H, Santarius T, Stebbings LA, Leroy C, Edkins S, Hardy C et al.: Identification of somatically acquired rearrangements in cancer using genome-wide massively parallel paired-end sequencing. Nat Genet 2008, 40:722-729. The authors demonstrate that genome rearrangements can be measured by paired end sequencing. When such paired reads are mapped back to the reference genome sequence and they do not align correctly with respect to each other, that is, have an abnormal distance or orientation, then a genomic rearrangement is suggested. 38. Lasken RS: Genomic DNA amplification by the multiple displacement amplification (MDA) method. Biochem Soc Trans 2009, 37:450-453. 39. Dear PH: One by one: single molecule tools for genomics. Brief Funct Genomic Proteomic 2003, 1:397-416. 40. Zong C, Lu S, Chapman AR, Xie XS: Genome-wide detection of single-nucleotide and copy-number variations of a single human cell. Science 2012, 338:1622-1626.
23. Coolbaugh-Murphy MI, Xu J, Ramagli LS, Brown BW, Siciliano MJ: Microsatellite instability (MSI) increases with age in normal somatic cells. Mech Ageing Dev 2005, 126:1051-1059.
41. Michaelson JJ, Shi Y, Gujral M, Zheng H, Malhotra D, Jin X, Jian M, Liu G, Greer D, Bhandari A et al.: Whole-genome sequencing in autism identifies hot spots for de novo germline mutation. Cell 2012, 151:1431-1442.
24. Muotri AR, Chu VT, Marchetto MC, Deng W, Moran JV, Gage FH: Somatic mosaicism in neuronal precursor cells mediated by L1 retrotransposition. Nature 2005, 435:903-910.
42. Yancik R, Ries LA: Cancer in older persons: an international issue in an aging world. Semin Oncol 2004, 31:128-136.
25. Perrat PN, DasGupta S, Wang J, Theurkauf W, Weng Z, Rosbash M, Waddell S: Transposition-driven genomic heterogeneity in the Drosophila brain. Science 2013, 340:91-95. 26. Evrony GD, Cai X, Lee E, Hills LB, Elhosary PC, Lehmann HS, Parker JJ, Atabay KD, Gilmore EC, Poduri A et al.: Single-neuron sequencing analysis of L1 retrotransposition and somatic mutation in the human brain. Cell 2012, 151:483-496. 27. Hu Z, Chen K, Xia Z, Chavez M, Pal S, Seol JH, Chen CC, Li W, Tyler JK: Nucleosome loss leads to global transcriptional upregulation and genomic instability during yeast aging. Genes Dev 2014, 28:396-408. 28. Aubert G, Lansdorp PM: Telomeres and aging. Physiol Rev 2008, 88:557-579. 29. Olovnikov AM: A theory of marginotomy. The incomplete copying of template margin in enzymic synthesis of polynucleotides and biological significance of the phenomenon. J Theor Biol 1973, 41:181-190. The author proposes that the end replication problem is a major source of telomere shortening over time, which has since then been amply confirmed. 30. Gundry M, Vijg J: Direct mutation analysis by high-throughput sequencing: from germline to low-abundant, somatic variants. Mutat Res 2012, 729:1-15.
43. Ershler WB, Longo DL: Aging and cancer: issues of basic and clinical science. J Natl Cancer Inst 1997, 89:1489-1497. 44. Levine B, Kroemer G: Autophagy in the pathogenesis of disease. Cell 2008, 132:27-42. 45. DePinho RA: The age of cancer. Nature 2000, 408:248-254. 46. Ames BN, Gold LS: Endogenous mutagens and the causes of aging and cancer. Mutat Res 1991, 250:3-16. 47. Doll R, Peto R: Cigarette smoking and bronchial carcinoma: dose and time relationships among regular smokers and lifelong non-smokers. J Epidemiol Commun Health 1978, 32:303-313. 48. Pfeifer GP, Denissenko MF, Olivier M, Tretyakova N, Hecht SS, Hainaut P: Tobacco smoke carcinogens, DNA damage and p53 mutations in smoking-associated cancers. Oncogene 2002, 21:7435-7451. 49. Govindan R, Ding L, Griffith M, Subramanian J, Dees ND, Kanchi KL, Maher CA, Fulton R, Fulton L, Wallis J et al.: Genomic landscape of non-small cell lung cancer in smokers and neversmokers. Cell 2012, 150:1121-1134.
31. Mardis ER: A decade’s perspective on DNA sequencing technology. Nature 2011, 470:198-203.
50. Fearon ER, Vogelstein B: A genetic model for colorectal tumorigenesis. Cell 1990, 61:759-767. The authors present a model for the origin of cancer based on a series of mutations that provide a selective growth advantage.
32. Weber-Lehmann J, Schilling E, Gradl G, Richter DC, Wiehler J, Rolf B: Finding the needle in the haystack: Differentiating ‘‘identical’’ twins in paternity testing and forensics by ultra-deep next generation sequencing. Forensic Sci Int Genet 2014, 9:42-46.
51. Tomasetti C, Vogelstein B, Parmigiani G: Half or more of the somatic mutations in cancers of self-renewing tissues originate prior to tumor initiation. Proc Natl Acad Sci U S A 2013, 110:1999-2004.
Current Opinion in Genetics & Development 2014, 26:141–149
www.sciencedirect.com
Genome mosaicism and aging Vijg 149
52. Lawrence MS, Stojanov P, Polak P, Kryukov GV, Cibulskis K, Sivachenko A, Carter SL, Stewart C, Mermel CH, Roberts SA et al.: Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 2013, 499:214-218. 53. Welch JS, Ley TJ, Link DC, Miller CA, Larson DE, Koboldt DC, Wartman LD, Lamprecht TL, Liu F, Xia J et al.: The origin and evolution of mutations in acute myeloid leukemia. Cell 2012, 150:264-278. 54. Martin GM: Somatic mutagenesis and antimutagenesis in aging research. Mutat Res 1996, 350:35-41. 55. Vijg J, Calder RB: Transcripts of aging. Trends Genet 2004, 20:221-224. 56. Hasty P, Campisi J, Hoeijmakers J, van Steeg H, Vijg J: Aging and genome maintenance: lessons from the mouse? Science 2003, 299:1355-1359. 57. Ness KK, Krull KR, Jones KE, Mulrooney DA, Armstrong GT, Green DM, Chemaitilly W, Smith WA, Wilson CL, Sklar CA et al.: Physiologic frailty as a sign of accelerated aging among adult survivors of childhood cancer: a report from the st jude lifetime cohort study. J Clin Oncol 2013, 31:4496-4503. The authors show that survivors of childhood cancer show multiple signs and symptoms of premature aging later in life. 58. Henderson TO, Whitton J, Stovall M, Mertens AC, Mitby P, Friedman D, Strong LC, Hammond S, Neglia JP, Meadows AT et al.: Secondary sarcomas in childhood cancer survivors: a report from the Childhood Cancer Survivor Study. J Natl Cancer Inst 2007, 99:300-308. 59. O’Roak BJ, Deriziotis P, Lee C, Vives L, Schwartz JJ, Girirajan S, Karakoc E, Mackenzie AP, Ng SB, Baker C et al.: Exome sequencing in sporadic autism spectrum disorders identifies severe de novo mutations. Nat Genet 2011, 43:585-589. 60. Sanders SJ, Ercan-Sencicek AG, Hus V, Luo R, Murtha MT, Moreno-De-Luca D, Chu SH, Moreau MP, Gupta AR, Thomson SA et al.: Multiple recurrent de novo CNVs, including duplications of the 7q11.23 Williams syndrome region, are strongly associated with autism. Neuron 2011, 70:863-885. 61. Erickson R: Somatic gene mutation and human disease other than cancer. Mut Res/Rev Mutat Res 2003, 543:125-136. 62. Garcia-Linares C, Fernandez-Rodriguez J, Terribas E, Mercade J, Pros E, Benito L, Benavente Y, Capella G, Ravella A, Blanco I et al.:
www.sciencedirect.com
Dissecting loss of heterozygosity (LOH) in neurofibromatosis type 1-associated neurofibromas: Importance of copy neutral LOH. Hum Mutat 2011, 32:78-90. 63. Beck JA, Poulter M, Campbell TA, Uphill JB, Adamson G, Geddes JF, Revesz T, Davis MB, Wood NW, Collinge J et al.: Somatic and germline mosaicism in sporadic early-onset Alzheimer’s disease. Hum Mol Genet 2004, 13:1219-1224. 64. Andreassi MG: Coronary atherosclerosis and somatic mutations: an overview of the contributive factors for oxidative DNA damage. Mutat Res 2003, 543:67-86. 65. Bernstein BE, Birney E, Dunham I, Green ED, Gunter C, Snyder M: An integrated encyclopedia of DNA elements in the human genome. Nature 2012, 489:57-74. By systematically mapping biochemically active regions of the human genome the ENCODE project discovered that at least 80% of the genome is functional across tissues. 66. Kong A, Frigge ML, Masson G, Besenbacher S, Sulem P, Magnusson G, Gudjonsson SA, Sigurdsson A, Jonasdottir A, Wong WS et al.: Rate of de novo mutations and the importance of father’s age to disease risk. Nature 2012, 488:471-475. 67. Lynch M: Evolution of the mutation rate. Trends Genet: TIG 2010, 26:345-352. 68. Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA Jr, Kinzler KW: Cancer genome landscapes. Science 2013, 339:1546-1558. 69. Orgel LE: The maintenance of the accuracy of protein synthesis and its relevance to ageing. Proc Natl Acad Sci U S A 1963, 49:517-521. 70. Sacher GA: Evolutionary theory in gerontology. Perspect Biol Med 1982, 25:339-353. The author proposed that species-specific life span is determined by socalled longevity determinant genes, which include genes involved in genome maintenance, antioxidant defense and other biological systems playing a role in maintaining cell structure and function. 71. Kirkwood TB: Understanding the odd science of aging. Cell 2005, 120:437-447. 72. Singer T, McConnell MJ, Marchetto MC, Coufal NG, Gage FH: LINE-1 retrotransposons: mediators of somatic variation in neuronal genomes? Trends Neurosci 2010, 33:345-354.
Current Opinion in Genetics & Development 2014, 26:141–149
ScienceDirect Somatic mutations, genome mosaicism, cancer and aging Jan Vijg Genomes are inherently unstable due to the need for DNA sequence variation in the germ line to fuel evolution through natural selection. In somatic tissues mutations accumulate during development and aging, generating genome mosaics. There is little information about the possible causal role of increased somatic mutation loads in late-life disease and aging, with the exception of cancer. Characterizing somatic mutations and their functional consequences in normal tissues remains a formidable challenge due to their low, individual abundance. Here, I will briefly review our current knowledge of somatic mutations in animals and humans in relation to aging, how they arise and lead to genome mosaicism, the technology to study somatic mutations and how they possibly could cause nonclonal disease. Addresses Department of Genetics, Albert Einstein College of Medicine, New York, United States Corresponding author: Vijg, Jan ([email protected], http://www.einstein.yu.edu)
Current Opinion in Genetics & Development 2014, 26:141–149 This review comes from a themed issue on Molecular and genetic bases of disease Edited by Cynthia T McMurray and Jan Vijg
http://dx.doi.org/10.1016/j.gde.2014.04.002 0959-437X/Published by Elsevier Ltd.
Introduction Structure and function of all organisms are ultimately determined by their genome, the complete set of species-specific hereditary information. Genomes are sufficiently stable to continuously provide function over the life time. Yet, they are far from immutable. Indeed, mutations in germline genomes are the substrate of evolution by natural selection and therefore necessary to perpetuate and diversify life. If mutation rates would be zero, all alleles would be fixed and organisms could not adapt [1]. The tendency of genomes to undergo mutations, that is, addition, deletion, or substitution of bits of genetic code, is known as genome instability and the inevitable and irreversible result of errors made during the repair or replication of damaged DNA. Because they generally occur at random, most mutations have adverse effects. Natural selection often eliminates genomes with unfavorable mutations through the demise www.sciencedirect.com
of their carriers. Some mutations, however, convey positive attributes and will be picked up by natural selection to be perpetuated. It is through mutations that Darwinian selection could lead to increasingly complex genomes and the adaptation of their hosts to the various challenges of a continuously changing environment. Genomes are not only subject to variation in the germ cells transmitted by parents to offspring. New variation, through DNA mutations, accumulates between the germ lines in the soma. Like in the germ line, mutations in somatic tissues are generally random and rarely beneficial; most have either no effect or an adverse effect. While in natural environments deleterious mutations can be eliminated from the germ line by selection, in somatic tissues of metazoans selection is less efficacious in removing mutations or not at all possible. It is for this reason that the gradual accumulation of mutations in the genome of somatic cells has been implicated in aging, that is, the timedependent process of deteriorative change that ultimately brings life to a close [2,3–5]. Figure 1 schematically depicts this double-edged sword of genome instability. Genome instability is known to cause disease in humans, most notably cancer. Cancer arises from single cells as a consequence of repeated cycles of mutagenesis and selection for growth advantage and other attributes that promote tumor progression. However, little is known about the effect of randomly induced mutations on the function of individual cells in a tissue, without selection. Depending on the frequency at which mutations occur and their distribution across the genome of individual cells it is conceivable that the resulting increased tissue mosaicism leads to functional deterioration at old age. The role of somatic mutations in diseases other than cancer will be discussed elsewhere in this special issue. Here, I will discuss how somatic mutations arise and accumulate in aging organisms, leading to genome mosaics, the difficulties in studying somatic mutations and genome mosaicism, how somatic mutations cause cancer and, finally, how random mutations can lead to age-related functional decline.
Somatic mutations lead to genome mosaicism The concept that all somatic cells have identical genomes needs to be replaced by a much more dynamic model of an increasingly mosaic genome. The engine behind this is somatic mutagenesis. Somatic mutations are low-frequency events and only directly detectable when they are amplified in a clonal lineage (Figure 2). Using selectable marker genes in transgenic mice, in the past we have conclusively demonstrated the accumulation of mutations during aging in a variety of organs and tissues, Current Opinion in Genetics & Development 2014, 26:141–149
142 Molecular and genetic bases of disease
Figure 1
Mutations Evolutionary change
Germline
Soma
Disease, Aging?
RNA, protein Cellular function Current Opinion in Genetics & Development
The genome harbors all the instructions for providing function to the somatic cells of an organism. Random alterations in its information content in the germ line drive evolutionary change, while similar changes in the somatic cells could be the cause of aging.
including liver, brain, heart, small intestine, and spleen [6–8]. The rate of this increase as well as the spectrum of mutations (basepair substitutions, deletions, translocations) differed greatly from organ to organ. For example, while in the intestine point mutations accumulated rapidly to a very high level at old age, in heart and liver also the frequency of genome rearrangements was significantly higher at old age [8]. Over time, accumulating somatic mutations essentially turns an aging tissue into a mosaic of cells with different genotypes. Owing to the emergence of powerful new
Figure 2
Tumor
Clonal lineage
Development and aging Current Opinion in Genetics & Development
Schematic depiction of mutation accumulation during development and aging. Cancer-causing mutations (red check mark) are actively selected for, resulting in the clonal development of a tumor (beige). Occasionally, a mutation is subject to genomic drift (green check mark) and also leads to clonal amplification (green). Current Opinion in Genetics & Development 2014, 26:141–149
methods for analyzing copy number variation (amplifications and deletions) of DNA segments defined by single nucleotide polymorphisms, deletions were observed in blood and other tissues of humans and mice [9,10]. Such deletions must find their origin in the creation of somatic subpopulations of cells due to expansion of a single, de novo mutation. Indeed, an age-related accumulation of clonally expanded, copy number variants has been found [11–13]. Similarly, the age-related emergence of genomic differences between monozygotic twins who were identical at early age has been observed and interpreted as originating from somatic mutation [14]. These findings are highly likely to be only the tip of an iceberg because even the most recent 1 million-probe array cannot detect deletions smaller than 10 kb by comparative genomic hybridization (CGH) [15]. Moreover, mutations occurring in less than 10% of all cells are no longer detectable. Hence, direct methods can only detect sizable somatic genome alterations and then only as clonally amplified mutations affecting a substantial fraction of the tissue or cell population. There is also a large body of evidence that chromosomal aberrations, which are microscopically detectable using fluorescence in situ hybridization (FISH), capable of detecting alterations affecting no less than 1–5 million basepairs, accumulate in blood lymphocytes of both humans and mice [16,17]. Especially aneuploidy, that is, the loss or gain of whole chromosomes resulting in an abnormal numerical karyotype, appears to be remarkably high, even in postmitotic tissue [18,19]. Chromosomal aneuploidy is a hallmark of pathological conditions and a causal factor of birth defects and cancer. Using advanced FISH technology it has been shown that in the developing nervous system of humans and mice the frequency of aneuploid cells is as high as 33% for all chromosomes combined [20,21]. This extraordinary high level of genome instability during development is in keeping with observations of whole-chromosome aneuploidy and segmental deletions and duplications in individual blastomeres of cleavage-stage embryos after in vitro fertilization (IVF) [22]. Interestingly, although IVF success rates are typically in the order of 25%, in only 9% of these embryos was such genomic variation not present. This indicates extensive selection of non-mutated or less severely mutated blastomeres in the early embryo, which also suggests that the observed high frequency of aneuploidy during development can decline thereafter (through selection) and not be fatal. Indeed, In human or mouse adult brain, in which only few chromosomes were analyzed by interphase FISH, aneuploidy frequencies were found to be much lower than in developmental mouse brain, that is, 10% for all chromosomes combined. However, this is still high and we have shown that in the mouse brain aneuploidy accumulates with age, up until about 50% for all chromosomes combined in 28-month-old animals [19]. www.sciencedirect.com
Genome mosaicism and aging Vijg 143
Somatic mutation rates differ among different types of genomic sequence. Indeed, in regions containing repeat elements, such as mini-satellites and micro-satellites, retrotransposons and telomeres, spontaneous mutations can occur at a much higher rate than in single-copy regions. Mutations at microsatellite loci have been found to occur at rates as high as 1 102 per locus in human blood and increase with age [23]. Retrotransposition has been observed in rodent brain in vivo [24] and deep sequencing has revealed insertions of retrotransposons in Drosophila and human brain [25,26]. In yeast, during replicative aging, increased retrotransposition has been observed concomitant with a loss of nucleosomes, a cause of aging in budding yeast [27]. Telomeres, regions of repetitive DNA protecting chromosome ends from deterioration, significantly shorten with age in mammalian cells and tissues [28], in part due to the end replication problem [29]. In conclusion, somatic mutations accumulate with age in a tissue-specific manner, turning tissues into genome mosaics. Direct evidence for such mosaics has been obtained only for clonally amplified structural variations greater than 10 kb. For elucidating the complete landscape of somatic mutations over the life time of an individual new, advanced methodology is necessary.
Single-cell genomics to determine the in vivo landscape of somatic mutations Rapid progress in nucleotide sequencing technology [30,31] now allows the detection of even minute postzygotic mutational differences between the genomes of monozygotic twins [32]. However, as we have seen, apart from very large chromosomal alterations, mutations present in less than 10% of all cells in a tissue cannot be detected by direct analysis, but requires selectable markers. In the above discussed examples somatic mosaicism could only be detected because the mutation occurred early during development and/or aging, thereby eventually comprising a sizable fraction of cells. Such mutations are the tiny tip of what surely must be a gigantic iceberg of mutations, many unique for a particular individual cell. Therefore, we have no real insight into the true number of genome variants in the different cells comprising a tissue. Indeed, virtually our entire knowledge of the genome has come from studying cells in bulk, that is, as mixtures of cells. In principle one could detect somatic mutations in DNA from bulk cell populations by sequencing thousands of times across the genome and call all possible variants of the consensus sequence. In practice this would be very inefficient and expensive, and also essentially constrained by the high rate of sequencing errors, which can amount to close to 1% [33,34]. Genuine somatic mutations cannot be distinguished from sequencing errors, which are scattered through the reads (Figure 3). The problem is very similar to the situation in tumors in which we are also unable to glimpse behind www.sciencedirect.com
the relatively few mutations common to all or most cells in that tumor [35]. One possible solution for this problem is to sequence single cells after whole genome amplification and analyze combinations of mutations in an integrated manner. The latter is essential for assessing if random mutations are truly capable of adversely affecting cellular function (see below). We recently developed a method to determine somatic mutation frequencies and spectra across the genome of a tissue or cell population by sequencing a representative number of single cells after whole genome amplification [36]. After amplification, mutations unique for each cell will now occur in half of all sequencing reads for a particular locus. Only half of all reads, because a true mutation will affect one allele only, since it is highly unlikely that the other allele will also be hit at that same site (Figure 3). The single-cell protocol we developed was validated by testing if an elevation in mutation frequency in Drosophila S2 cells or mouse embryonic fibroblasts could be detected after treatment with a powerful mutagen (i.e., ethylnitrosourea; ENU). The results indicated a dramatically higher level of mutations in the exposed cells as compared to untreated control cells [36]. This is expected because ENU is a powerful mutagen. The exposed cells also revealed increased levels of A:T ! T:A mutations, which are known to be highly enriched following treatment with alkylating agents. While in these initial experiments we only called single nucleotide variants (i.e., point mutations), it is possible to also identify structural variations in this way using a socalled paired-end mode of sequencing [30,37]. When DNA fragments of a particular size are sequenced from both ends the paired reads should be positioned at a known distance from each other when aligned to a reference sequence (Figure 4). If the distance between the read pairs differs from the distance between the mapped positions on the reference genome, then a deletion or insertion is implied. It is also possible that one of the two read pairs maps to another chromosome, indicating a chromosomal translocation. Polymorphic variants with the reference genome are subtracted by always including a sequencing run from the non-amplified, whole tissue or whole cell population. The procedure allows us to call all possible somatic mutations present in a particular single cell. Genome amplification is often considered a major source of false positive mutations. For example, the most popular whole genome amplification method is multiple displacement amplification (MDA), which has been shown to cause false positive inversion mutations [38]. However, for point mutations such artifacts are unlikely. An error introduced early in MDA would be found in only 12.5% of the sequencing reads at a given locus in a diploid cell on average [39]. Since as shown in Figure 3 Current Opinion in Genetics & Development 2014, 26:141–149
144 Molecular and genetic bases of disease
Figure 3
CAT TTA GT T TG AT GT TGG C TAT C G C C C C C
T
C
T Sequencing errors
CAT TTA GT T TG AT GT TGG C TAT C G C G
Single Nucleotide Polymorphism
G Whole genome amplification
G G
T
C C C
G
C
G True somatic mutation
Current Opinion in Genetics & Development
Single-cell analysis is required for detecting low-abundant, random mutations. The A > G mutation (red check mark) can only be distinguished from sequencing errors after single-cell, whole genome amplification when it shows up in 50% of the reads (red; one mutated allele). Polymorphic variants (SNPs; blue) are also observed in the unamplified control DNA extracted from the bulk cell population. SNPs are variants between the genome of our cells and the reference genome. They are electronically discarded.
true somatic mutations should affect one allele and therefore need to be found in 50% of the reads, amplification errors are easily filtered out. Also, recent developments in whole genome amplification have now greatly diminished concerns about amplification artifacts [40]
and, albeit more difficult than SNVs, structural variations can be detected in single cells [30]. Indeed, recent work demonstrated the detection of copy number variation in single human neurons after whole genome amplification [10].
Figure 4
100bp
100bp
A to C PE seq RefSeq
PE seq RefSeq
500bp
chr2
chr3
PE seq RefSeq
800bp
Paired end sequencing
Paired end mapping
Single base substitution
300 bp deletion
Translocation Current Opinion in Genetics & Development
Paired-end sequencing allows the identification of a large variety of structural variation as discrepancies after aligning the sequencing reads to the reference genome. Current Opinion in Genetics & Development 2014, 26:141–149
www.sciencedirect.com
Genome mosaicism and aging Vijg 145
The great advantage of opening up next-generation sequencing to the analysis of somatic mutations is that, for the first time, all possible mutations, including low-abundant mutations, can be analyzed on a genome-wide basis. The latter is critically important because mutations are not uniformly distributed across the genome, but subject to significant constraints, including, for example, chromatin structure, transcription, the local structure of the genomic regions studied (e.g., repeats versus single-copy DNA) [41]. To systematically determine complete mutation spectra for a large number of single cells or nuclei from multiple somatic tissues at different age levels should allow us to unravel the impact of genome mosaicism on health and disease. Below I will illustrate how somatic mutations are responsible for cancer, a major age-related disease, but possibly also for non-cancer, degenerative aspects of the aging process.
Somatic mutations can cause both cancer and aging While cancer can occur also in young people, the incidence and mortality of most cancers increases exponentially with age [42]. While other factors, such as agerelated changes in immune function [43], apoptosis or autophagy [44], may also contribute, the mechanistic basis of the relationship between cancer and aging most likely involves a lifetime of DNA mutation accumulation as a natural consequence of inevitable errors during repair or replication of damaged DNA [5,45]. Indeed, DNA damage is ubiquitous, as a result of both endogenous processes, such as hydrolysis, oxidation and alkylation, and environmental genotoxicants [46]. While most of that damage is efficiently repaired, errors during replication and repair are inevitable and irreversible once the original template is lost. Exposure to environmental mutagens is known to shift the age-incidence curves of cancer towards accelerated risk, most likely by increasing mutation rates. The best example is tobacco smoke which, increases cancer risk in a clear dose-response relationship [47,48]. Indeed, lung tumors from smokers have an average mutation frequency more than 10-fold higher than lung tumors from non-smokers [49]. The reason that the accumulation of somatic mutations with age is so readily adopted as a likely cause of the agerelated increase in cancer risk is the universally accepted model of cancer initiation and progression. This model, first proposed by Fearon and Vogelstein in 1990 for colorectal cancer [50], posits a series of sequential mutations that first provide a selective growth advantage to the target cell(s) and subsequently other favorable attributes, such as the capacity to invade tissues, suppress immune responses and inactivate other anti-tumor responses, such as apoptosis. Eventually, mutations are also responsible for resistance to anti-cancer drugs. Each mutation is followed by clonal outgrowth in which a particularly advantageous mutation is selected. The www.sciencedirect.com
eventual malignant neoplasm shows a mutation spectrum that reflects its clonal history, but also contains so-called passenger mutations, that is, mutations that were not selected because of a clonal growth advantage and therefore did not contribute to cancer development. Many of those mutations had been accumulated during development and aging before tumor initiation [51]. The tumor mutation spectrum often also reveals the original mutagens responsible for the tumor-inducing mutations, for example, C:G to A:T in lung tumors from smokers and C:G to T:A mutations in the same tumor type from non-smokers. In this way, initially rare somatic mutations can eventually become predominant through natural selection. It is now clear that the number of somatic mutations in tumors is significantly higher when the tumor was derived from an old patient as compared to a young one [52,53]. Indeed, mathematical modeling strongly suggests that half or more of somatic mutations in tumors arise before initiation of the tumor, that is, during development and aging [51]. We still know very little about the mutational heterogeneity of normal tissues. Also intratumor variation in mutation spectra remains an enigma [35]. Nevertheless, the concept of age-related mutation accumulation to explain the age-related increase in cancer is straightforward. Such mutation-driven loss of proliferative homeostasis, giving rise to hyperplasias, benign neoplasms and malignant neoplasms, is a major component of the aging phenotype and likely to be causally related to age-related mutation accumulation in cells from self-renewing tissues [54]. By contrast, a causal role of mutations in tissue functional decline would rely on a cumulated mutation load in many or most cells of an organ or tissue that is high enough to have adverse effects without selection. Recently, a series of observations revealed that defects in genome maintenance pathways cannot only lead to increased cancer but also to multiple symptoms of cell and tissue degeneration that are normally only seen at old age. For example, heritable mutations in RecQ helicase genes cause segmental progeroid syndromes, such as Werner Syndrome [55]. Also mice harboring defects in DNA double-strand break repair or nucleotide excision repair display signs of premature aging, sometimes at very early age [56]. While this strongly suggests that DNA damage and/or mutations are critical factors in aging and its disease sequelae, there is now also recent evidence that cancer survivors who have been treated with DNA damaging chemotherapeutic agents display multiple symptoms of premature aging. Indeed, while in the past surviving cancer was relatively rare, this has now become more common. Interestingly, an ever growing number of adults who survived childhood cancer, is now diagnosed with health problems common to much older people, such as frailty, low muscle mass, slow walking speed, weakness and memory impairment [57]. While this could Current Opinion in Genetics & Development 2014, 26:141–149
146 Molecular and genetic bases of disease
be, at least in part, an effect of the cellular responses to DNA damage, such as apoptosis and cellular senescence, the secondary cancers that are also observed [58] point towards DNA mutations as another possible cause. As described in more detail by Erickson elsewhere in this special issue, both de novo and post-zygotic mutations are increasingly found to contribute to human disease, with the former now a known cause of autism [59,60] and the latter already known from disorders where the manifestations are readily seen, for example, the skin in type I neurofibromatosis [61]. Clearly, somatic mutations occurring early during development are likely to have a greater effect on phenotype expression than mutations occurring later and virtually all examples derive from clonal lineages with the disease-causing somatic mutation occurring during embryogenesis. In many cases the critical somatic mutation is a second mutation turning a recessive germ line mutation in one allele into a dominant phenotype after somatic inactivation of the other allele [62]. An interesting example of genome mosaicism as a cause of neurodegenerative disease is the observed somatic mutation in the presenilin-1 gene in a sporadic earlyonset patient with Alzheimer’s disease [63]. There is currently no direct evidence that somatic mutations causally contribute to aging and late-life disease other than cancer. Somatic mutations have been found in atherosclerotic plaques, but their causal role in atherogenesis and vascular disease development is uncertain [64]. Indeed, the functional impact of somatic mutagenesis other than through clonal amplification is presently unclear. The question is whether random mutations affecting single genes at low frequency can have an effect at the tissue level when not amplified through clonal outgrowth. While the current lack of detailed information about the landscape of somatic mutations in aged individuals prevents answering this question, it is certainly conceivable that random mutations affect cellular function. Indeed, function is not provided by single genes but through often extensive, functional networks consisting of genes and their regulatory sequences. Together, this would comprise a sizable genomic target for mutation. Each and every mutational hit in a sequence that is part of this network has the potential to adversely affect function. For example, for a typical differentiated human cell we can conservatively estimate that about 1% of the diploid genome (60 Mb) is functional [65]. While the germline de novo mutation frequency in humans is only about 60 new mutations per genome (1.2 108/bp or 0.01/ Mb [66]), somatic mutation frequencies are much higher than germline mutation frequencies [67]. Indeed, the mutation frequency in human tumors that have been subjected to whole genome sequencing was found to vary from about 0.1/Mb to almost 200/Mb [52], with Current Opinion in Genetics & Development 2014, 26:141–149
the median at about 6/Mb. As mentioned above, many mutations found in tumors have accumulated in the normal cells before tumor initiation, which is the most likely reason that tumors from children contain fewer mutations than tumors from adults or older individuals [51,68]. Our own unpublished data, obtained using the new, single-cell sequencing procedure (see above), indicate about 1 single nucleotide variant (SNV) per Mb in liver hepatocytes from 27-month old mice. This excludes structural variants (SVs), which are much less frequent, but likely to have more dramatic effects. In the 60 Mb of the genome that presumably is functional in a differentiated cell, an SNV frequency of about 1/Mb would then result in about 60 mutations. A fraction of these mutations could potentially adversely affect functional networks in that cell. Our data indicate that 5% of all somatic mutations in exons are nonsense mutations, with about 60% missense mutations (unpublished). This fraction of non-synonymous mutations is much higher than that in the germline genome because of the absence of selection. Including the SVs, which also accumulate with age [6], it certainly is conceivable that random mutations accumulating during aging do have functional effects other than clonal disease.
Summary and future prospects Like germ line mutations, somatic mutations are part of the logic of life. But while germ line mutagenesis is a sine qua non for the perpetuation of life, mutations in the soma are not needed and could very well lead to an error catastrophe, as originally proposed by Orgel for proteins [69]. In principle, evolution could postpone catastrophe by evolving ever more sophisticated genome maintenance systems to gradually increase species-specific longevity [70]. However, evolution by natural selection is a process not conducive to maximization of life span beyond the age of first reproduction [71]. Somatic mutation has been coopted during evolution to generate the means of creating immunoglobulin diversity, a critical factor in the immune response, and has also been suggested to play a role in generating diversity in the brain to broaden the spectrum of phenotypes that can originate from a single genome [72]. However, somatic mutations are unlikely to play a major role in human physiology and most of them are harmful and could contribute to aging. In summary, somatic mutations appear to be much more frequent than initially thought, reactivating original ideas about the causes of aging and age-related disease [2,3]. Indeed, those hypotheses can now be more firmly outlined. As schematically depicted in Figure 5, mutations inevitably and irreversibly accumulate during development and aging. They become manifest as clonal lineages causing cancer and other cell proliferative disorders. However, conceivably when a critical level is reached, www.sciencedirect.com
Genome mosaicism and aging Vijg 147
Figure 5
Neoplastic or dysplastic lesion
+ : cell depletion
Mutation accumulation, functional decline, cancer
Development and aging
+
+
Diseased or aged tissue
Young tissue
Current Opinion in Genetics & Development
Age-related, cell functional divergence as a consequence of stochastic effects. Functional decline is indicated by increasing darkness. Even in a young tissue, the function of highly differentiated cells in an organ or tissue is never maximized. The old tissue is different in the sense that there are many more cells that have suffered functional decline, with some of them dying (cross) or eliminated (open space). Others have grown into a hyperplastic or neoplastic lesion (not shown) or have been replaced by fibrosis (not shown).
mutations begin to adversely affect cell-specific functional networks, resulting in a mosaic of cells harboring different mutation loads and varying levels of functional deficiency. Depending on the severity of the mutation load, the cell type, genetic and environmental factors, this would lead to functional decline, disease and, possibly, aging.
References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest 1.
Baer CF, Miyamoto MM, Denver DR: Mutation rate variation in multicellular eukaryotes: causes and consequences. Nat Rev Genet 2007, 8:619-631.
2. Failla G: The aging process and carcinogenesis. Ann N Y Acad Sci 1958, 71:1124-1135. The first proposal of DNA mutations as a cause of aging. 3.
Szilard L: On the nature of the aging process. Proc Natl Acad Sci U S A 1959, 45:30-45.
4.
Curtis H: Biological mechanisms underlying the aging process. Science 1963, 141:686-694.
5.
Vijg J: Aging of the Genome. New York: Oxford; 2007.
6.
Dolle´ ME, Giese H, Hopkins CL, Martus HJ, Hausdorff JM, Vijg J: Rapid accumulation of genome rearrangements in liver but not in brain of old mice. Nat Genet 1997, 17:431-434.
7.
Dolle´ ME, Snyder WK, Dunson DB, Vijg J: Mutational fingerprints of aging. Nucleic Acids Res 2002, 30:545-549.
8.
Dolle´ ME, Snyder WK, Gossen JA, Lohman PH, Vijg J: Distinct spectra of somatic mutations accumulated with age in mouse
www.sciencedirect.com
heart and small intestine. Proc Natl Acad Sci U S A 2000, 97:8403-8408. 9.
Piotrowski A, Bruder CE, Andersson R, Diaz de Stahl T, Menzel U, Sandgren J, Poplawski A, von Tell D, Crasto C, Bogdan A et al.: Somatic mosaicism for copy number variation in differentiated human tissues. Hum Mutat 2008, 29:1118-1124. One of the first reports indicating that humans are commonly affected by somatic mosaicism for CNVs in a substantial fraction of cells.
10. McConnell MJ, Lindberg MR, Brennand KJ, Piper JC, Voet T, Cowing-Zitron C, Shumilina S, Lasken RS, Vermeesch JR, Hall IM et al.: Mosaic copy number variation in human neurons. Science 2013, 342:632-637. The authors used single-cell sequencing of endogenous human frontal cortex neurons to demonstrate that genomes in a sizable subset of neurons are marked by de novo CNVs. These results show that mosaic CNV is abundant in human neurons. 11. Jacobs KB, Yeager M, Zhou W, Wacholder S, Wang Z, RodriguezSantiago B, Hutchinson A, Deng X, Liu C, Horner MJ et al.: Detectable clonal mosaicism and its relationship to aging and cancer. Nat Genet 2012, 44:651-658. 12. Laurie CC, Laurie CA, Rice K, Doheny KF, Zelnick LR, McHugh CP, Ling H, Hetrick KN, Pugh EW, Amos C et al.: Detectable clonal mosaicism from birth to old age and its relationship to cancer. Nat Genet 2012, 44:642-650. 13. Forsberg LA, Rasi C, Razzaghian HR, Pakalapati G, Waite L, Thilbeault KS, Ronowicz A, Wineinger NE, Tiwari HK, Boomsma D et al.: Age-related somatic structural changes in the nuclear genome of human blood cells. Am J Hum Genet 2012, 90: 217-228. 14. Bruder CE, Piotrowski A, Gijsbers AA, Andersson R, Erickson S, de Stahl TD, Menzel U, Sandgren J, von Tell D, Poplawski A et al.: Phenotypically concordant and discordant monozygotic twins display different DNA copy-number-variation profiles. Am J Hum Genet 2008, 82:763-771. 15. Przybytkowski E, Ferrario C, Basik M: The use of ultra-dense array CGH analysis for the discovery of micro-copy number alterations and gene fusions in the cancer genome. BMC Med Genom 2011, 4:16. Current Opinion in Genetics & Development 2014, 26:141–149
148 Molecular and genetic bases of disease
16. Ramsey MJ, Moore DH II, Briner JF, Lee DA, Olsen L, Senft JR, Tucker JD: The effects of age and lifestyle factors on the accumulation of cytogenetic damage as measured by chromosome painting. Mutat Res 1995, 338:95-106. The authors demonstrated that chromosomal aberrations in peripheral blood lymphocytes of humans increase with age. 17. Tucker JD, Spruill MD, Ramsey MJ, Director AD, Nath J: Frequency of spontaneous chromosome aberrations in mice: effects of age. Mutat Res 1999, 425:135-141. 18. Iourov IY, Vorsanova SG, Yurov YB: Chromosomal mosaicism goes global. Mol Cytogenet 2008, 1:26. 19. Faggioli F, Wang T, Vijg J, Montagna C: Chromosome-specific accumulation of aneuploidy in the aging mouse brain. Hum Mol Genet 2012, 21:5246-5253. 20. Yurov YB, Iourov IY, Vorsanova SG, Liehr T, Kolotii AD, Kutsev SI, Pellestor F, Beresheva AK, Demidova IA, Kravets VS et al.: Aneuploidy and confined chromosomal mosaicism in the developing human brain. PLoS ONE 2007, 2:e558. 21. Rehen SK, McConnell MJ, Kaushal D, Kingsbury MA, Yang AH, Chun J: Chromosomal variation in neurons of the developing and adult mammalian nervous system. Proc Natl Acad Sci U S A 2001, 98:13361-13366. One of the first demonstrations that neurons in the mammalian nervous system do not possess identical genomes. Evidence is presented that approximately 33% of mouse neuroblasts are aneuploid, about 10 times as much as in adult lymphocytes. In the adult cortex aneuploidy is decreased, possibly because aneuploid cells are prone to cell death. 22. Vanneste E, Voet T, Le Caignec C, Ampe M, Konings P, Melotte C, Debrock S, Amyere M, Vikkula M, Schuit F et al.: Chromosome instability is common in human cleavage-stage embryos. Nat Med 2009, 15:577-583. This study establishes that chromosome instability is also common during early IVF embryogenesis. The results strongly suggest that in addition to selection against chromosomally abnormal embryos, which is well-documented, selection at the cellular level occurs against abnormal blastomeres.
33. Quail MA, Swerdlow H, Turner DJ: Improved protocols for the illumina genome analyzer sequencing system. Curr Protoc Hum Genet 2009. Chapter 18: Unit 18.2. 34. Kinde I, Wu J, Papadopoulos N, Kinzler KW, Vogelstein B: Detection and quantification of rare mutations with massively parallel sequencing. Proc Natl Acad Sci U S A 2011, 108: 9530-9535. 35. Salk JJ, Fox EJ, Loeb LA: Mutational heterogeneity in human cancers: origin and consequences. Ann Rev Pathol 2010, 5: 51-75. 36. Gundry M, Li W, Maqbool SB, Vijg J: Direct, genome-wide assessment of DNA mutations in single cells. Nucleic Acids Res 2012, 40:2032-2040. The authors show that it is possible to directly measure somatic mutation frequencies and spectra in cell populations, not using selectable markers but through genome-wide sequencing of single cells. 37. Campbell PJ, Stephens PJ, Pleasance ED, O’Meara S, Li H, Santarius T, Stebbings LA, Leroy C, Edkins S, Hardy C et al.: Identification of somatically acquired rearrangements in cancer using genome-wide massively parallel paired-end sequencing. Nat Genet 2008, 40:722-729. The authors demonstrate that genome rearrangements can be measured by paired end sequencing. When such paired reads are mapped back to the reference genome sequence and they do not align correctly with respect to each other, that is, have an abnormal distance or orientation, then a genomic rearrangement is suggested. 38. Lasken RS: Genomic DNA amplification by the multiple displacement amplification (MDA) method. Biochem Soc Trans 2009, 37:450-453. 39. Dear PH: One by one: single molecule tools for genomics. Brief Funct Genomic Proteomic 2003, 1:397-416. 40. Zong C, Lu S, Chapman AR, Xie XS: Genome-wide detection of single-nucleotide and copy-number variations of a single human cell. Science 2012, 338:1622-1626.
23. Coolbaugh-Murphy MI, Xu J, Ramagli LS, Brown BW, Siciliano MJ: Microsatellite instability (MSI) increases with age in normal somatic cells. Mech Ageing Dev 2005, 126:1051-1059.
41. Michaelson JJ, Shi Y, Gujral M, Zheng H, Malhotra D, Jin X, Jian M, Liu G, Greer D, Bhandari A et al.: Whole-genome sequencing in autism identifies hot spots for de novo germline mutation. Cell 2012, 151:1431-1442.
24. Muotri AR, Chu VT, Marchetto MC, Deng W, Moran JV, Gage FH: Somatic mosaicism in neuronal precursor cells mediated by L1 retrotransposition. Nature 2005, 435:903-910.
42. Yancik R, Ries LA: Cancer in older persons: an international issue in an aging world. Semin Oncol 2004, 31:128-136.
25. Perrat PN, DasGupta S, Wang J, Theurkauf W, Weng Z, Rosbash M, Waddell S: Transposition-driven genomic heterogeneity in the Drosophila brain. Science 2013, 340:91-95. 26. Evrony GD, Cai X, Lee E, Hills LB, Elhosary PC, Lehmann HS, Parker JJ, Atabay KD, Gilmore EC, Poduri A et al.: Single-neuron sequencing analysis of L1 retrotransposition and somatic mutation in the human brain. Cell 2012, 151:483-496. 27. Hu Z, Chen K, Xia Z, Chavez M, Pal S, Seol JH, Chen CC, Li W, Tyler JK: Nucleosome loss leads to global transcriptional upregulation and genomic instability during yeast aging. Genes Dev 2014, 28:396-408. 28. Aubert G, Lansdorp PM: Telomeres and aging. Physiol Rev 2008, 88:557-579. 29. Olovnikov AM: A theory of marginotomy. The incomplete copying of template margin in enzymic synthesis of polynucleotides and biological significance of the phenomenon. J Theor Biol 1973, 41:181-190. The author proposes that the end replication problem is a major source of telomere shortening over time, which has since then been amply confirmed. 30. Gundry M, Vijg J: Direct mutation analysis by high-throughput sequencing: from germline to low-abundant, somatic variants. Mutat Res 2012, 729:1-15.
43. Ershler WB, Longo DL: Aging and cancer: issues of basic and clinical science. J Natl Cancer Inst 1997, 89:1489-1497. 44. Levine B, Kroemer G: Autophagy in the pathogenesis of disease. Cell 2008, 132:27-42. 45. DePinho RA: The age of cancer. Nature 2000, 408:248-254. 46. Ames BN, Gold LS: Endogenous mutagens and the causes of aging and cancer. Mutat Res 1991, 250:3-16. 47. Doll R, Peto R: Cigarette smoking and bronchial carcinoma: dose and time relationships among regular smokers and lifelong non-smokers. J Epidemiol Commun Health 1978, 32:303-313. 48. Pfeifer GP, Denissenko MF, Olivier M, Tretyakova N, Hecht SS, Hainaut P: Tobacco smoke carcinogens, DNA damage and p53 mutations in smoking-associated cancers. Oncogene 2002, 21:7435-7451. 49. Govindan R, Ding L, Griffith M, Subramanian J, Dees ND, Kanchi KL, Maher CA, Fulton R, Fulton L, Wallis J et al.: Genomic landscape of non-small cell lung cancer in smokers and neversmokers. Cell 2012, 150:1121-1134.
31. Mardis ER: A decade’s perspective on DNA sequencing technology. Nature 2011, 470:198-203.
50. Fearon ER, Vogelstein B: A genetic model for colorectal tumorigenesis. Cell 1990, 61:759-767. The authors present a model for the origin of cancer based on a series of mutations that provide a selective growth advantage.
32. Weber-Lehmann J, Schilling E, Gradl G, Richter DC, Wiehler J, Rolf B: Finding the needle in the haystack: Differentiating ‘‘identical’’ twins in paternity testing and forensics by ultra-deep next generation sequencing. Forensic Sci Int Genet 2014, 9:42-46.
51. Tomasetti C, Vogelstein B, Parmigiani G: Half or more of the somatic mutations in cancers of self-renewing tissues originate prior to tumor initiation. Proc Natl Acad Sci U S A 2013, 110:1999-2004.
Current Opinion in Genetics & Development 2014, 26:141–149
www.sciencedirect.com
Genome mosaicism and aging Vijg 149
52. Lawrence MS, Stojanov P, Polak P, Kryukov GV, Cibulskis K, Sivachenko A, Carter SL, Stewart C, Mermel CH, Roberts SA et al.: Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 2013, 499:214-218. 53. Welch JS, Ley TJ, Link DC, Miller CA, Larson DE, Koboldt DC, Wartman LD, Lamprecht TL, Liu F, Xia J et al.: The origin and evolution of mutations in acute myeloid leukemia. Cell 2012, 150:264-278. 54. Martin GM: Somatic mutagenesis and antimutagenesis in aging research. Mutat Res 1996, 350:35-41. 55. Vijg J, Calder RB: Transcripts of aging. Trends Genet 2004, 20:221-224. 56. Hasty P, Campisi J, Hoeijmakers J, van Steeg H, Vijg J: Aging and genome maintenance: lessons from the mouse? Science 2003, 299:1355-1359. 57. Ness KK, Krull KR, Jones KE, Mulrooney DA, Armstrong GT, Green DM, Chemaitilly W, Smith WA, Wilson CL, Sklar CA et al.: Physiologic frailty as a sign of accelerated aging among adult survivors of childhood cancer: a report from the st jude lifetime cohort study. J Clin Oncol 2013, 31:4496-4503. The authors show that survivors of childhood cancer show multiple signs and symptoms of premature aging later in life. 58. Henderson TO, Whitton J, Stovall M, Mertens AC, Mitby P, Friedman D, Strong LC, Hammond S, Neglia JP, Meadows AT et al.: Secondary sarcomas in childhood cancer survivors: a report from the Childhood Cancer Survivor Study. J Natl Cancer Inst 2007, 99:300-308. 59. O’Roak BJ, Deriziotis P, Lee C, Vives L, Schwartz JJ, Girirajan S, Karakoc E, Mackenzie AP, Ng SB, Baker C et al.: Exome sequencing in sporadic autism spectrum disorders identifies severe de novo mutations. Nat Genet 2011, 43:585-589. 60. Sanders SJ, Ercan-Sencicek AG, Hus V, Luo R, Murtha MT, Moreno-De-Luca D, Chu SH, Moreau MP, Gupta AR, Thomson SA et al.: Multiple recurrent de novo CNVs, including duplications of the 7q11.23 Williams syndrome region, are strongly associated with autism. Neuron 2011, 70:863-885. 61. Erickson R: Somatic gene mutation and human disease other than cancer. Mut Res/Rev Mutat Res 2003, 543:125-136. 62. Garcia-Linares C, Fernandez-Rodriguez J, Terribas E, Mercade J, Pros E, Benito L, Benavente Y, Capella G, Ravella A, Blanco I et al.:
www.sciencedirect.com
Dissecting loss of heterozygosity (LOH) in neurofibromatosis type 1-associated neurofibromas: Importance of copy neutral LOH. Hum Mutat 2011, 32:78-90. 63. Beck JA, Poulter M, Campbell TA, Uphill JB, Adamson G, Geddes JF, Revesz T, Davis MB, Wood NW, Collinge J et al.: Somatic and germline mosaicism in sporadic early-onset Alzheimer’s disease. Hum Mol Genet 2004, 13:1219-1224. 64. Andreassi MG: Coronary atherosclerosis and somatic mutations: an overview of the contributive factors for oxidative DNA damage. Mutat Res 2003, 543:67-86. 65. Bernstein BE, Birney E, Dunham I, Green ED, Gunter C, Snyder M: An integrated encyclopedia of DNA elements in the human genome. Nature 2012, 489:57-74. By systematically mapping biochemically active regions of the human genome the ENCODE project discovered that at least 80% of the genome is functional across tissues. 66. Kong A, Frigge ML, Masson G, Besenbacher S, Sulem P, Magnusson G, Gudjonsson SA, Sigurdsson A, Jonasdottir A, Wong WS et al.: Rate of de novo mutations and the importance of father’s age to disease risk. Nature 2012, 488:471-475. 67. Lynch M: Evolution of the mutation rate. Trends Genet: TIG 2010, 26:345-352. 68. Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA Jr, Kinzler KW: Cancer genome landscapes. Science 2013, 339:1546-1558. 69. Orgel LE: The maintenance of the accuracy of protein synthesis and its relevance to ageing. Proc Natl Acad Sci U S A 1963, 49:517-521. 70. Sacher GA: Evolutionary theory in gerontology. Perspect Biol Med 1982, 25:339-353. The author proposed that species-specific life span is determined by socalled longevity determinant genes, which include genes involved in genome maintenance, antioxidant defense and other biological systems playing a role in maintaining cell structure and function. 71. Kirkwood TB: Understanding the odd science of aging. Cell 2005, 120:437-447. 72. Singer T, McConnell MJ, Marchetto MC, Coufal NG, Gage FH: LINE-1 retrotransposons: mediators of somatic variation in neuronal genomes? Trends Neurosci 2010, 33:345-354.
Current Opinion in Genetics & Development 2014, 26:141–149