Cell, Vol. 120, 497–512, February 25, 2005, Copyright ©2005 by Elsevier Inc. DOI 10.1016/j.cell.2005.01.028

ReviewDNA Repair, GenomeStability, and Aging

David B. Lombard,1,2 Katrin F. Chua,1

Raul Mostoslavsky,1 Sonia Franco,1

Monica Gostissa,1 and Frederick W. Alt1,*1Howard Hughes Medical InstituteThe Children’s HospitalDepartment of GeneticsHarvard Medical School andThe CBR Institute for Biomedical ResearchBoston, Massachusetts 021152Department of PathologyBrigham and Women’s HospitalBoston, Massachusetts 02115

Aging can be defined as progressive functional de-cline and increasing mortality over time. Here, we re-view evidence linking aging to nuclear DNA lesions:DNA damage accumulates with age, and DNA repairdefects can cause phenotypes resembling prematureaging. We discuss how cellular DNA damage re-sponses may contribute to manifestations of aging.We review Sir2, a factor linking genomic stability, me-tabolism, and aging. We conclude with a general dis-cussion of the role of mutant mice in aging researchand avenues for future investigation.

IntroductionAlthough aging is nearly universally conserved amongeukaryotic organisms, the molecular mechanisms un-derlying aging are only beginning to be elucidated. Auseful conceptual framework for considering the prob-lem of aging is the Disposable Soma model (Kirkwoodand Holliday, 1979). This model proposes that organ-isms only invest enough energy into maintenance of thesoma to survive long enough to reproduce. Aging oc-curs at least in part as a consequence of this imperfectmaintenance, rather than as a genetically programmedprocess. Although aging may involve damage to vari-ous cellular constituents, the imperfect maintenance ofnuclear DNA likely represents a critical contributor toaging. Unless precisely repaired, nuclear DNA damagecan lead to mutation and/or other deleterious cellularand organismal consequences. Damage to both nuclearDNA, which encodes the vast majority of cellular RNAand proteins, and mitochondrial DNA have been pro-posed to contribute to aging (Karanjawala and Lieber,2004). The reader is referred to the review by Balabanet al. in this issue of Cell concerning the potential roleof mitochondrial DNA damage in aging (Balaban et al.,2005). Nuclear DNA is an attractive target for aging-related changes since it must last the lifetime of thecell, unlike other cellular constituents, which can be re-placed. In addition, the nuclear genome is present atonly two to four copies per cell, rendering it potentiallyvery vulnerable to lesions; by contrast, the mito-chondrial genome is present at several thousand cop-

*Correspondence: alt@enders.tch.harvard.edu

ies per cell. Many mutants with phenotypes that resem-ble premature aging possess defects in nuclear DNArepair. Therefore, we focus on the role of nuclear DNAdamage in mammalian aging, emphasizing recent workin mouse models.

DNA Damage, Reactive Oxygen Species, and AgingA large body of evidence argues that DNA damage andmutations accumulate with age in mammals (Vijg,2000). Cells harboring mutations at defined loci havebeen shown to increase with age in humans and mice.Cytogenetically visible lesions such as translocations,insertions, dicentrics, and acentric fragments also ac-cumulate in aging mammalian cells. Mice with inte-grated reporter arrays have allowed estimates of theage-related occurrence of both point mutations andlarger genomic rearrangements. Such analyses haverevealed considerable variation in mutation spectra be-tween different tissues. These differences likely reflectfunctional characteristics of those tissues, such as mi-totic rate, transcriptional activity, metabolism, and theaction of specific DNA repair systems.Reactive Oxygen Species: An Important Sourceof Age-Related DNA DamageThere are many sources of DNA damage. In addition toexternal sources, such as ionizing radiation and geno-toxic drugs, there are also cell-intrinsic sources, suchas replication errors, spontaneous chemical changes tothe DNA, programmed double-strand breaks (DSBs) (inlymphocyte development), and DNA damaging agentsthat are normally present in cells. The latter categoryincludes reactive oxygen species (ROS), such as super-oxide anion, hydroxyl radical, hydrogen peroxide, nitricoxide, and others. Major sources of cellular ROS pro-duction are the mitochondria, peroxisomes, cytochromep450 enzymes, and the antimicrobial oxidative burst ofphagocytic cells. ROS can cause lipid peroxidation,protein damage, and several types of DNA lesions: sin-gle- and double-strand breaks, adducts, and cross-links. The situation in which ROS exceed cellular anti-oxidant defenses is termed oxidative stress. As normalbyproducts of metabolism, ROS are a potential sourceof chronic, persistent DNA damage in all cells and maycontribute to aging (Sohal and Weindruch, 1996). TheROS theory of aging is discussed in depth in this issueof Cell by Balaban et al. (2005). In brief, longer-livedspecies generally show higher cellular oxidative stressresistance and lower levels of mitochondrial ROS pro-duction than shorter-lived species. Caloric restriction,an intervention that extends life span in many organ-isms, likely decreases ROS production (Barja, 2004). Inmodel organisms, many mutations that promote lon-gevity concomitantly increase oxidative stress resis-tance (Finkel and Holbrook, 2000). In addition, levels of8-oxoguanine (oxo8dG), a major product of oxidativedamage to DNA, accumulate with age (Hamilton et al.,2001).

The potential importance of oxidative damage toDNA in age-related dysfunction is highlighted by a re-

Cell498

cent study of postmortem human brain tissue (Lu et al., pc2004), which found that many nuclear genes involved

in critical neural functions show reduced expression af- apter age 40, concomitant with elevated levels of oxo8dG

and DNA damage in their promoters. However, the high cRlevels of oxidative DNA damage found by these investi-

gators is at odds with other much lower estimates of aIthe amount of oxidatively damaged DNA in cells (Hamil-

ton et al., 2001). This highlights the difficulty of accu- ccrately measuring ROS and oxidative damage experi-

mentally. Additional unresolved issues concerning the tcROS theory of aging exist that will require clarification.

If ROS are indeed an important source of aging-associ- mwated damage, it is currently unclear whether nuclear or

mitochondrial DNA is the most relevant functional totarget (Barja and Herrero, 2000; Hamilton et al., 2001).

Among long-lived mutants, the correlation with oxida- lftive stress resistance is a frequent but not universal

one: in some mutants, longevity occurs despite un- aichanged resistance to ROS or other forms of stress.

Additionally, mice heterozygous for a mutation in the pimitochondrial enzyme that processes superoxide, Sod2,

live out a normal lifespan, despite accumulating higher odlevels of nuclear and mitochondrial 8oxodG with age

(Van Remmen et al., 2003). Thus, while several lines of rtevidence argue for important role for ROS in aging,

many questions remain regarding this hypothesis. In cethis regard, it is possible that additional mutations may

be required to fully unveil potential effects of ROS fg(see below).ooCellular Senescence: A Link between Cellular(Damage and Aging?aROS and many other DNA-damaging agents can causeccells to enter a state of irreversible cell cycle arrest ac-(companied by characteristic morphologic and func-

tional alterations, termed senescence (Ben-Porath andpWeinberg, 2004). The induction of senescence dependsron pathways involving the p53 and Rb proteins (Figuree1). Cellular senescence has been best characterized in2cultures of human fibroblasts and mouse embryo fibro-

blasts (MEFs), which cease expanding after repeated t

Figure 1. Multiple Pathways to Senescence

These pathways are employed in a cell- andorganism-specific fashion. In MEFs, activa-tion of p19ARF leads to stabilization of p53.In certain human cell types (e.g., fibroblastsand keratinocytes), telomere attrition can ac-tivate p53 via the ATM and (potentially) ATRkinases. Senescence can also be triggeredin human cells via p16 expression (Itahana etal., 2004).Activated p53 induces senescence via acomplex gene expression program that in-cludes induction of p21. The role of Rb insenescence involves repression of E2F targetgenes as well as alterations in chromatinstructure. Senescent cells may contribute toaging via depletion of stem cell pools and/or elaboration of factors that interfere withtissue function. Factors elaborated by se-nescent cells may also stimulate the growthof epithelial tumors (Krtolica and Campisi,2002).

assage in culture, a process termed replicative senes-ence. Replicative senescence has been employed ascellular model for aging; many mutations in DNA re-

air genes that cause premature aging phenotypes alsoonfer premature replicative senescence (Table 1).eplicative Senescence Differs between Humannd Mouse Cellsn many primary human cell lines, replicative senes-ence occurs secondary to attrition of the ends of thehromosomes, the telomeres, in a process termed “in-rinsic senescence” (Itahana et al., 2004). In mammalianells, the ends of the chromosomes consist of a ter-inal 3# single-stranded tail, the G strand overhang,hich is buried into adjacent double-stranded repeti-

ive telomeric DNA, forming a protective “t loop” higher-rder structure (de Lange, 2002). This t loop is stabi-

ized by a “D loop,” or displacement loop: the regionormed between the invading end of the telomere intodjacent double-stranded DNA. The G strand overhang

s the substrate for the enzyme telomerase, which em-loys an RNA template, Terc, to extend telomeres dur-

ng S phase, thereby countering the natural shorteningf telomeres that would otherwise occur with each cellivision. Telomerase access to its substrate is, in turn,

egulated by telomeric proteins, which also modulatehe conformational changes required for telomere repli-ation and subsequent reestablishment of a protectivend structure. In many types of human cells, includingibroblasts, telomeres shorten with each successiveeneration. Critically short telomeres trigger the onsetf senescence through a process that may involve lossf the t loop structure and/or loss of protective proteins

referred to as “uncapping”). Such uncapped telomeresre then recognized by the cell cycle checkpoint ma-hinery as DNA damage, leading to cell cycle arrestBen-Porath and Weinberg, 2004).

In contrast to primary human fibroblasts, which ex-ress very low levels of telomerase activity, MEFs de-ived from M. musculus possess long telomeres andasily detectable telomerase activity (Itahana et al.,004). Thus, MEFs ordinarily do not senesce due toelomeric attrition. Instead, senescence of wild-type

Review499

Table 1. Features of Selected Models of Premature Aging in the Mouse

AcceleratedIncreased Rate of Fibroblast

Mutant Cellular Process Affected Tissues Affected Cancer? Senescence? Citations

Atm DSB signaling/repair Cerebellum, Gonad, Yes Yes (Ito et al., 2004;Hematopoietic Shiloh and Kastan,organs, Thymus 2001)

Bub1bH/H Spindle assembly Bone, Lens, Skin, No Yes (Baker et al., 2004)checkpoint Gonad

BRCA1�11/�11/ DSB repair, other Bone, Eye, Heart, Yes Yes (Cao et al., 2003b)p53+/− Intestine, Liver,

Lymphocytichyperplasia, Testes+others

DNA-PKcs NHEJ, other Bone, Intestine Yes No (Espejel et al., 2004b)Ercc1, XPF Nucleotide excision repair, Liver (Ercc1, XPF) + No Yes (Ercc1) (McWhir et al., 1993;

crosslink repair, other Brain, Kidney, Skin, Tian et al., 2004;Spleen—Ercc1 Weeda et al., 1997)

Ku80 NHEJ, other Bone, Liver, Skin No Yes (Vogel et al., 1999)Dysfunctional p53 DNA damage response Bone, Hair, Lymphoid No ND (Tyner et al., 2002)

(p53m/+) tissue, SkinDysfunctional p53 DNA damage response Bone, Testes No Yes (Maier et al., 2004)

(p44 Tg)PASG/Lsh DNA methylation Bone, Hair, Kidney, No Yes (Sun et al., 2004)

Skin, ThymusPolgA Mitochondrial DNA Bone, Hair, Heart, No ND (Trifunovic et al.,

polymerase Hematopoietic 2004)organs, Skin,Testes

Rad50S/S DSB repair Hematopoietic Yes No (Bender et al., 2002)organs, Testes

Terc Telomere maintenance Hematopoietic Conflicting results Yes (Espejel et al., 2004a;organs, Hair, Heart, Lee et al., 1998)Intestine,Myometrium, Skin,Testes

Terc/Atm Telomere maintenance/ Bone, Brain, Hair, No Yes (Wong et al., 2003)DSB signaling/repair Hematopoietic

organs, IntestineTerc/DNA-PKcs Telomere maintenance/ Intestine, Various No ND (Espejel et al., 2004a)

NHEJ, otherTerc/Parp-1 Telomere maintenance/ Various No ND (Espejel et al., 2004a)

DNA repairTerc/Ku80 Telomere maintenance/ Intestine, Various No ND (Espejel et al., 2004a)

NHEJ, otherTerc/Wrn Telomere maintenance/ Bone, Endocrine, Yes Yes (Chang et al., 2004)

DNA repair Gonad, Hair,Intestine, Lens,Skin, Spleen

Terc/Wrn/Blm Telomere maintenance/ Bone, Endocrine, Yes Yes (Du et al., 2004)DNA repair Gonad, Hair,

IntestineTopIIIbeta Topoisomerase Kidney, Lymphocytic No ND (Kwan et al., 2003)

infiltrates,Pancreatic islets,Skin, Testes

Wrn DNA repair None No Yes (Lebel and Leder,1998; Lombard etal., 2000)

XPA/CSB NER, transcription Cerebellum No ND (Murai et al., 2001)XpdTTD NER, transcription Bone, Hair, Ovary, Yes ND (de Boer et al., 2002)

SkinXpdTTD/XPA NER, transcription Bone, Hair, Skin No ND (de Boer et al., 2002)

MEFs occurs primarily in response to oxidative DNAdamage incurred during cell culture (Parrinello et al.,2003) in a process termed “extrinsic senescence” (Ita-hana et al., 2004). These differences between humanand mouse cells with respect to senescence are not as

great as they may seem. Terc deficiency in mice leadsto progressive telomere attrition in successive mousegenerations and rapid senescence in embryonic fibro-blasts derived from late-generation animals (Espejeland Blasco, 2002). Replicative life span of human cells

Cell500

is altered by culture under different oxygen tensions, Tand telomere shortening in these cells is itself ac- Bcelerated by elevated levels of oxidative stress. It is

Hnow clear that senescence can be induced by a varietyAof different types of cellular injury (Figure 1).ACellular Senescence and AgingHWhat is the relationship between cellular senescenceE

and aging? The evidence that cellular senescence ac- Ltually plays a role in aging is correlative: senescent cells L

Maccumulate in vivo in mammals with increasing age andMat sites of pathology (Itahana et al., 2004), and manyOmouse and human models of premature aging are ac-Pcompanied by premature cellular senescence in vitroP

(Table 1). Of note, late-generation Terc-deficient mice Sshow some signs of accelerated aging (Lee et al., 1998; T

URudolph et al., 1999). Two general models have beenproposed to explain how cellular senescence may con- Ltribute to aging (Krtolica and Campisi, 2002; Pelicci,

K2004). First, senescence of progenitor or stem cells Lthemselves could impair tissue renewal. In this regard, Lthe Polycomb group repressor Bmi1 appears to control M

Plevels of hematopoietic stem cells via negatively regu-Slating the induction of senescence specifically in theseGstem cells. Second, senescent cells secrete proteases

and other factors that may disrupt tissue function. In Hthis regard, senescence has a complex relationship O

Swith neoplasia. Senescence has been postulated to oc-Rcur as a tumor suppressor mechanism, whereby cells

that have undergone a genotoxic insult and therefore Bpossess the potential for neoplastic transformation en- Dter a state in which they are incapable of dividing (Krtol- D

Mica and Campisi, 2002). However, senescent stromalPcells can actually promote the growth of epithelial can-

cers, malignancies that occur with increased incidence Nin the elderly. Senescence has been offered as an ex- Hample of “antagonistic pleiotropy,” a process that is Nbeneficial in young organisms but deleterious later in R

Wlife: senescence suppresses cancer by preventing po-tentially tumorigenic cells from dividing but may poten- Otially contribute to organ dysfunction in the aged Athrough a variety of mechanisms, perhaps even con- Ftributing to neoplasia in this setting (Krtolica and Cam- F

Hpisi, 2002). However, a causal relationship between cel-Tlular senescence and organismal aging has yet to be

proved.

DNA Repair and Agingp

The accurate maintenance of nuclear DNA is critical toF

cellular and organismal function, and therefore, numer-t

ous DNA repair systems have evolved. There is someo

evidence that the intrinsic fidelity and activity of suchs

systems in different species may influence the rate oft

age-associated functional decline (Hart and Setlow,v

1974), although such studies need to be performed withmodern methodology. As outlined below, the efficiencyof cellular DNA repair machinery itself may decline with T

Dage. Many different types of DNA lesions exist. DNADSBs are repaired via the nonhomologous end-joining U

p(NHEJ) and homologous recombination (HR) pathways,whereas lesions on a single strand of DNA are repaired c

svia base excision repair (BER) and nucleotide excisionrepair (NER) (and its subpathways). In mice and hu- f

tmans, mutations in certain DNA repair genes lead to

henotypes that, in some respects, caricature aging.or reference, the features of aging as it occurs in wild-ype mice are enumerated in Table 2, and the featuresf mouse DNA repair/metabolism mutants showingome features of aging are listed in Table 1. We nowurn to discussing specific gene defects and their rele-ance to aging.

he ATM-p53 Axis: A Link between the DNAamage Response and Aging?nrepaired or improperly repaired DSBs have seriousotential consequences for the cell: cell death, senes-ence, dysregulation of cellular functions, genomic in-tability, and, in higher organisms, oncogenic trans-ormation. The initial step in DSB repair is detection ofhe lesion, and this step involves the ataxia-telangiecta-

able 2. Common Pathologic Features of Aging in Mice (afterronson and Lipman [1991]; Cao et al. [2003a])

yperplasia/Neoplasia

drenal hyperplasiangiosarcomaarderian gland adenomandometrial hyperplasiaung adenomaymphomaammary gland adenocarcinomaast cell tumorvarian cystadenomaaraovarian cystituitary adenomaarcomahyroid follicular cell hyperplasiaterine leiomyoma/leiomyosarcoma

eukocytic Infiltrates

idneyiverungesentery/omentumerineuriumalivary gland

enitourinary system

ydronephrosisvarian/testicular atrophyeminal vesicle dilationenal tubular dilation

one

ecreased cancellous boneegenerative joint diseaseolar teeth periodontitisroliferations in the head/spine

eurological

ydrocephaluseuronal lipofuscinosisadiculopathyhite matter gliosis

ther

myloidosisatty change of the liverocal myocardial degenerationepatocyte polyploidizationhymic involution

Review501

sia mutated (ATM) kinase, other related phosphatidyl-inositol 3-kinase-like kinases (PIKKs), and other pro-teins (Figure 2; Bassing and Alt, 2004). Activated ATMphosphorylates numerous proteins involved in the G1/S,intra-S, and G2/M checkpoint responses and addition-ally phosphorylates factors involved in DNA repair(Bassing and Alt, 2004).

An important target of ATM is the p53 protein, whichplays a role in the cellular response to numerous geno-toxic insults including DSBs. Phosphorylation of p53 byATM and kinases downstream of ATM is a major mech-anism leading to upregulation of p53 levels and activity,although p53 also can be activated via ATM-indepen-dent mechanisms. Activated p53 stimulates or repressestranscription of many target genes and coordinatescheckpoint, senescence, and apoptosis pathways inresponse to DSBs and other signals. In this regard, p53,through modulation of various downstream targets,triggers arrest/senescence or apoptosis (Meek, 2004).The exact factors that determine the differential out-comes of this complex program are not yet completelyelucidated but vary with the cell type, as well as thekind, intensity, and duration of the damage.ATM and AgingPerturbations in ATM function can lead to symptoms ofaccelerated aging. Patients with mutations in the ATMgene suffer from Ataxia-Telangiectasia (AT), a conditioncharacterized by a prematurely aged (progeroid) ap-pearance, immunodeficiency, cerebellar degeneration,and cancer (Shiloh and Kastan, 2001). ATM deficiencyin mice recapitulates many of these phenotypes, al-though the progeroid features of the mouse models areless prominent than in the human disease. Given themany targets of ATM, it is difficult to trace the progeroidappearance of AT patients to a specific function of this

Figure 2. A Highly Simplified View of the DNA Damage Response

See text for details.

protein although several possible mechanisms are con-ceivable (see below). In this context, the phenotype ofATM-deficient cells may be relevant. In particular, suchcells manifest sensitivity to DSB-inducing agents andhave marked genomic instability; they also grow poorlyand senesce prematurely in culture (Barzilai et al.,2002).

Genomic instability or DNA repair defects of ATM de-ficiency could contribute to the aging-like features ofthis disorder (see below). In this context, the prematurecellular senescence phenotype of ATM deficiency isrescued by p53 deficiency (Xu et al., 1998), suggestinga role for ATM-independent, p53-dependent check-point responses. By extension, such responses alsomight contribute to premature-aging phenotypes asso-ciated with ATM deficiency. Another potential functionfor ATM in suppressing progeroid phenotypes may berelated to reported ATM functions in regulating intracel-lular ROS levels and sensitivity to these molecules (Bar-zilai et al., 2002; Ito et al., 2004). Persistently elevatedROS levels in AT cells might cause chronic damage toDNA and other cellular macromolecules. However,given the lack of an AT-like phenotype of the Sod2+/−

mouse, progeroid features of ATM deficiency likely re-flect more than increased ROS levels. One possibilitywould be elevated ROS levels in conjunction with lossof another ATM function, such as ability to respond toROS-induced DNA damage. Finally, ATM plays an im-portant role in telomere maintenance; AT cells showshortened telomeres and an increased incidence of te-lomeric fusions (Pandita, 2002). Moreover, mice lackingboth ATM and Terc have higher levels of telomeric dys-function than generation-matched Terc-deficient mu-tants, as well as proliferative defects in multiple tissues,decreased survival, and clinical evidence of prematureaging (Wong et al., 2003). Thus, human AT patients mayshow aging-like effects, at least in part, as a conse-quence of telomeric dysfunction, an effect that ordinar-ily may be masked in the mouse due to the long telo-meres of this organism.

The phenotypes of ATM deficiency are influenced byenvironmental conditions. In the ATM-deficient mouse,the onset of T cell lymphoma can be delayed by treat-ment with an antioxidant, suggesting that dietary factorsimpact on the expression of this condition (Schubert etal., 2004). Neurological dysfunction in this model cansimilarly be ameliorated via antioxidant treatment (Browneet al., 2004). Additionally, the frequency of thymic lym-phoma varies among different strains of ATM-deficientmice, a finding that might reflect different backgroundmutations and/or housing conditions and the types ofpathogens present (Petiniot et al., 2002). These obser-vations raise the possibility that other aspects of theATM-deficient phenotype, including the prematureaging observed in human patients, may be influencedby environment. This is a theme that we will return toin our discussion of other premature aging models.p53 and AgingVarious lines of evidence suggest that p53 plays op-posing roles in the aging process. While p53 sup-presses the onset of malignancy and thereby extendslife span, at the same time it promotes cellular senes-cence and apoptosis in response to DNA damage, po-tentially contributing to the clinical changes of aging.

Cell502

Thus, p53 function may display antagonistic pleiotropy Kt(Campisi, 2002). The role of p53 in aging cannot be di-trectly tested using p53-deficient mice, as such animalsPinvariably die of malignancy before age-related changesabecome manifest. However, two mouse strains that ex-tpress C-terminal p53 fragments along with full-length p53ehave been reported to show accelerated aging pheno-Ktypes and a lower incidence of malignancy (Maier et al.,t2004; Tyner et al., 2002). It has been proposed thatothese truncated p53 proteins exert their effects byDmodifying the activity of endogenous wild-type p53Dprotein. Only nonphysiologic activation of p53 leads tofprogeria, as mice expressing extra copies of wild-typecp53 under the control of its own promoter do not showIsigns of premature aging (Garcia-Cao et al., 2002).cFurther arguing for a role for p53 in promoting aging,ia null mutation in a gene functioning downstream of

p53 in the induction of apoptosis, p66Shc, confers oxi-adative stress resistance and extends mouse life spans(Migliaccio et al., 1999). The p66Shc protein shortensnmurine life span via at least two mechanisms: it in-rcreases constitutive intracellular ROS levels, and it pro-nmotes cell death in response to oxidative stress. It isiunclear how p66shc evolved to play such a role in thetcell, since the shorter isoforms of this protein, p52 andAp46, fulfill an entirely dissimilar cellular function, trans-cducing signals from tyrosine kinases to ras.im

A Putative Role for NHEJ in Suppressing Aging aNHEJ in DSB Repair and Telomere Maintenance aWe now turn from a discussion of factors upstream of bDNA repair to a consideration of repair pathways them- Kselves and their involvement in aging. NHEJ is one of tthe two major DSB repair pathways in mammalian cells f(Bassing and Alt, 2004). NHEJ is mediated by at least a

csix core factors (Figure 3). Four of these proteins (Ku80,

Figure 3. NHEJ at General DSBs

Factors associated with progeroid mutantphenotypes (Ku80 and DNA-PKcs) areshown in red. See text for details.

u70, Ligase IV, and XRCC4) are conserved from yeasto mammals; they are indispensable for all NHEJ reac-ions. In contrast, the other two NHEJ factors, DNA-Kcs (DNA-dependent protein kinase catalytic subunit)nd Artemis, have evolved more recently and arehought to be required for joining the subset of DNAnds that require processing prior to ligation. Ku70 andu80 bind as a heterodimer to DSBs, where they are

hought to serve a protective function and to enlistther factors. In this regard, Ku70 and Ku80 recruitNA-PKcs, and together the three proteins form theNA-PK holoenzyme. DNA-PK activates Artemis, which

unctions as an endonuclease to process ends thatannot be directly rejoined. Finally, XRCC4 and LigaseV, which are likely recruited by Ku, function together toatalyze end ligation itself. NHEJ often occurs concom-

tant with loss of a few nucleotides at the site of joining.NHEJ plays a critical role in general DNA DSB repair

nd, correspondingly, in the maintenance of genomictability. In addition, NHEJ plays a role in repairing ge-etically programmed DSBs in the context of antigeneceptor variable region gene assembly (V(D)J recombi-ation) in developing lymphocytes. Mice with targeted

nactivating mutations in NHEJ genes display pheno-ypes that reflect loss of these functions (Ferguson andlt, 2001). All NHEJ-deficient mice suffer from severeombined immunodeficiency as a consequence of an

nability to productively rejoin broken V(D)J gene seg-ents in developing B and T cells. NHEJ deficiency is

lso associated with ionizing radiation-sensitivity andn elevated incidence of spontaneous genomic insta-ility. MEFs deficient for XRCC4, Ligase IV, Ku70, oru80 (but not DNA-PKcs or Artemis) senesce prema-

urely in culture, and mice deficient for these fouractors are very small and show widespread neuronalpoptosis during embryogenesis. Premature senes-ence and neuronal apoptosis, but not small size, are

Review503

relieved by p53 deficiency. Thus, the former pheno-types occur as a response to, rather than as a directconsequence of, unrepaired DSBs (Ferguson and Alt,2001). Mice deficient in NHEJ and p53 invariably suc-cumb to pro-B cell lymphomas as a consequence ofaberrant repair of V(D)J recombination-associatedDSBs in developing B cells, leading to oncogenic trans-locations (Bassing and Alt, 2004).

In addition to their roles in DSB repair, at least threecore NHEJ factors, namely Ku70, Ku80, and DNA-PKcs,localize to telomeres (d'Adda di Fagagna et al., 2001).Deficiency for any of these, as well as for Artemis(Rooney et al., 2003), is associated with an increasedfrequency of end-to-end chromosomal fusions in MEFs,suggesting a role for these proteins in chromosomal endcapping. There is conflicting data on whether NHEJfactors protect against telomere shortening. NHEJ alsopromotes telomeric fusions in some circumstances. In-creased telomere end-to-end fusions are observed in thesetting of the telomeric attrition associated with Terc defi-ciency or inhibition of TRF2, a protein thought to functionin end capping, and these end-to-end fusions are elimi-nated in NHEJ-deficient backgrounds (Espejel et al.,2002a, 2002b; Smogorzewska et al., 2002). Thus, by li-gating uncapped telomeres, NHEJ may actually pro-mote genomic instability. Aside from roles in the NHEJreaction and in telomere maintenance, Ku70, Ku80,DNA-PKcs, and Artemis have other known or sus-pected cellular functions, including DSB or checkpointsignaling, whereas XRCC4 and Ligase IV appear tofunction only in end-ligation during NHEJ. In summary,NHEJ plays crucial roles in general and site-specificDSB repair, in telomere maintenance, and in mainte-nance of genomic stability.A Potential Relationship between Ku80 Deficiency,DNA-PKcs Deficiency, and AgingNHEJ has been proposed to play a causative role in theaging process. DSBs are frequent events in mammaliansomatic cells, where they are also very commonly re-paired by NHEJ. In addition, genetic studies supportthe notion that NHEJ plays an important role in the re-pair of ROS-induced DNA lesions (Karanjawala andLieber, 2004). Since NHEJ can delete a few nucleotidesat sites of DSB repair, this process theoretically couldlead to accumulation of mutations and contribute tocellular decline and aging (Karanjawala and Lieber,2004). Moreover, in the absence of NHEJ, DSBs oftenare repaired concomitant with large deletions and/ortranslocations; thus, absent or even decreased levelsof NHEJ also might contribute to accelerated aging.

The observation that phenotypes resembling ac-celerated aging have been described in one strain eachof Ku80- and DNA-PKcs-deficient mice potentially sup-ports this model. Thus, a line of Ku80-deficient miceprematurely exhibits age-specific changes includingosteopenia, atrophic skin, liver lesions, and shortenedlife span (Vogel et al., 1999). Likewise, a strain of DNA-PKcs-deficient mice recently has been noted to exhibitage-related pathologies, with osteopenia, intestinal at-rophy, thymic lymphoma, and reduced longevity (Es-pejel et al., 2004b). There are several potential explana-tions to account for how NHEJ deficiency could causeaging-like phenotypes. As described above, spontane-ous DSBs in Ku80- or DNA-PKcs-deficient animals

could directly impair cellular function by promotingDNA deletions. In this context, Ku80-deficient mice arefound to have a low rate of mutations at a marker locus(Rockwood et al., 2003); such an assay might not de-tect large deletions or rearrangements, however. Alter-natively, unrepaired DSBs could trigger elevated levelsof cellular senescence and apoptosis. In this regard,p53-dependent responses are critical in mediating thepremature senescence and neuronal apoptosis pheno-types of certain NHEJ mutants (Ferguson and Alt,2001).

Despite the phenotypes of the Ku80- and DNA-PKcs-deficient mice, several observations suggest that anypotential roles for NHEJ in suppressing aging might bemore complicated. Ku70- and Artemis-deficient micethus far have not been reported to show prematureaging phenotypes. If Ku70-deficient mice actually lackan aging phenotype, the apparent Ku80-deficient pro-geroid phenotype must be rationalized in the contextof the finding that targeted ablation of Ku70 results indramatically reduced Ku80 levels (Gu et al., 1997). Inaddition, aging-related phenotypes thus far have notbeen reported in several other independently generatedDNA-PKcs mutant mouse strains. Ligase IV- and XRCC4-deficient mice show embryonic lethality due to wide-spread neuronal apoptosis, precluding aging analyses.However, the lack of a consistent aging-like phenotypein other lines of long-lived Ku-, DNA-PKcs-, or Artemis-deficient mice is difficult to explain if NHEJ plays a gen-eral role in delaying manifestations of aging (Karanjawalaand Lieber, 2004). In this regard, it is conceivable thatsome aging-related phenotypes might have beenmissed due to lack of thorough examination of otherstrains of NHEJ-deficient mice.

Background mutations in various strains of NHEJ-deficient mice, either exacerbating or suppressing theprogeroid manifestations of NHEJ deficiencies, mightcontribute to the apparently discrepant phenotypes ofdifferent lines. This possibility, which is relevant to anygene deficiency/aging model, warrants further exami-nation and is discussed in other contexts below. Also,given that both the Ku80- and DNA-PKcs-deficientmice show evidence of ongoing inflammation, poten-tially indicative of infection, it is conceivable that someaging-like phenotypes might occur directly due to theeffects of chronic infection in the setting of immunode-ficiency, rather than due to impaired DNA repair. Os-teopenia, for example, could result from elevated gluco-corticoid levels induced by chronic physiologic stressfrom infection or malnutrition associated with intestinalatrophy. The degenerative changes in afflicted strainsof DNA-PKcs- and Ku80-deficient mice affect only alimited subset of organs. Therefore, either these NHEJfactors are only involved in suppressing aging-relatedchanges in certain tissues or these degenerative changesoccur for reasons distinct from those that contribute toaging in wild-type animals. Of note, degenerative changesin these models occur in both highly proliferative (intes-tine and skin) and relatively less proliferative (bone andliver) tissues, apparently sparing many other tissues ofboth types. These observations argue against a simplerelationship between mitotic status and dependence onNHEJ in the suppression of aging.

In addition to the above considerations, other models

Cell504

for NHEJ factor functions in suppressing aging may be 2fpostulated. It is notable that efficiency of DSB repairgmay decline with age in budding yeast (McMurray andpGottschling, 2003) and in mammalian tissues (Sedelni-pkova et al., 2004; Singh et al., 2001), as well as in senes-tcent mammalian cells (Seluanov et al., 2004). Theoreti-vcally, such an age-related decline in DSB repair mightrcontribute to increased mutations and genomic re-darrangements preferentially near the end of life, al-mthough such a decline has not been rigorously proven.(Also, loss of other functions of DNA-PKcs or Ku80,asuch as telomere end capping and/or DNA damage sig-analing, could contribute to aging phenotypes. In thisRcontext, targeted inactivation of either Ku80 or DNA-wPKcs, but not in Ligase IV or XRCC4, demonstratesosynthetic lethality with ATM deficiency. These observa-qtions argue for overlapping functions between Ku/DNA-tPKcs and ATM that do not involve classical NHEJ, per-shaps related to telomere maintenance (Sekiguchi et al.,t2001). Moreover, Ku80/Terc and DNA-PKcs/Terc doubleadeficient mice show exacerbation of intestinal atrophytand other aging-like phenotypes of Terc deficiency (Es-gpejel et al., 2004a). Late generation Terc-deficient micePhave an uncharacterized DSB repair defect (Wong etBal., 2000); loss of NHEJ in the context of dysfunctionalTtelomeres would be predicted to further compromisecDSB repair. In this regard, cells deficient in Ligase IVcand ATM or Terc and ATM show high levels of genomicwinstability and very rapid senescence (Sekiguchi et al.,l2001; Wong et al., 2003), pointing to synergies betweenmDSB repair, DNA damage signaling, and telomere main-etenance in genomic maintenance and overall cellular vi-2ability.tIn summary, there are many potential roles for NHEJtin preventing premature aging-like phenotypes, such as(those observed in Ku80- and DNA-PKcs-deficientSmice. Current findings suggest, however, that agingfphenotypes likely do not result from a failure of DSBdrepair alone but instead from a loss of other functionsnor combinations of functions of these proteins, perhapsain concert with environmental factors, such as housingmconditions and/or infection and/or background muta-Ttions. Such complexities also may likely apply to otherr

DNA repair proteins implicated in the suppression off

aging as well, since, as we shall see, many of thesef

proteins play roles in other cellular processes besides EDNA repair and, as with NHEJ, only a subset of mutants vin any given DNA repair pathway show aging-related dphenotypes. d

mPotential Roles for Rad50 and BRCA1 cin HR and Aging vHR in DSB Repair tTwo factors with roles in HR, Rad50 and BRCA1 (breast 2cancer susceptibility gene-1), appear linked to aging- plike phenotypes in mouse models. HR is the other major Bpathway of DSB repair in mammalian cells. Unlike SNHEJ, HR uses the sister or (in some cases) the homol- Dogous chromosome as a template to repair the broken tchromosome. p

Whereas recent data indicate that NHEJ functions in eDSB repair throughout the cell cycle, HR is largely re- (

estricted to late S/G2 (Couedel et al., 2004; Mills et al.,

004; Rothkamm et al., 2003; Takata et al., 1998). Theirst step in HR is processing of DSBs by a nuclease toenerate 3# ssDNA tails, which are coated with RPArotein (Figure 4). The MRN complex, composed of theroteins Mre11, Rad50, and Nbs1, is a candidate forhis nuclease, although other nucleases are likely in-olved as well. The MRN complex plays multiple otheroles in DSB repair: it functions both upstream andownstream of ATM and has been proposed to pro-ote sister chromatid association and recombination

Stracker et al., 2004). The Rad51 protein, assisted bynumber of factors including Rad52, Rad54, BRCA2,

nd the Rad51 paralogs (XRCC2, XRCC3, Rad51B,ad51C, and Rad51D), forms a nucleoprotein complexith the DNA and directs the 3# ssDNA tails to searchut, invade, and pair with undamaged homologous se-uences. DNA polymerases then carry out repair usinghe intact DNA as a template. The processes of DNAtrand exchange and extension generate Holliday junc-ions (HJs), structures in which two dsDNA duplexesre intertwined. In mammalian DSB repair, HJs arehought to be resolved primarily via disengagement andap repair rather than cleavage of the HJ (Valerie andovirk, 2003).RCA1 in HR and Other Cellular Processeshe biology of the BRCA1 protein has proven to be veryomplex; BRCA1 plays roles in multiple fundamentalellular processes. Biochemically, BRCA1, togetherith its partner protein BARD1, possesses E3 ubiquitin

igase activity in addition to binding both DNA andultiple other proteins. Functionally, several lines of

vidence link the BRCA1 protein to HR (Scully et al.,004). BRCA1 is a phosphorylation target of ATM andhe related PIKK, ATR, following DNA damage induc-ion. BRCA1 forms a complex with Rad51 and BRCA2Dong et al., 2003) and can be detected with Rad51 in

phase foci thought to represent stalled replicationorks. BRCA1 foci also form in response to ionizing ra-iation and on chromosomes during meiotic recombi-ation. BRCA1 deficiency leads to impaired HR-medi-ted repair of chromosomal DSBs, hypersensitivity toany DNA-damaging agents, and genomic instability.he biochemical nature of BRCA1’s involvement in HRemains unclear; one possibility is that BRCA1 may per-orm a scaffolding function, potentially coordinating theormation of functional repair complexes at DSBs. The3 ubiquitin ligase activity of BRCA1 may also be rele-ant in this context; BRCA1 may modify other proteinsuring HR to alter their functions or direct their degra-ation. Additionally, BRCA1 has been implicated inany other functions outside HR: transcription, G2/M

heckpoint control, chromatin remodeling, and X inacti-ation (Scully et al., 2004). It has also been proposedhat BRCA1 may play a role in NHEJ (Ting and Lee,004). Thus, BRCA1 plays roles in numerous cellularrocesses, including DNA repair.RCA1 and Rad50 Hypomorphshow Aging Phenotypesefects in Rad50 or BRCA1 cause progeroid pheno-

ypes. Deficiency of Mre11, Nbs1, or Rad50 is not com-atible with cellular survival (Stracker et al., 2004). How-ver, homozygosity for a Rad50 hypomorphic alleleRad50s/s) permits viability; Rad50s/s mice show a short-ned life span, cancer predisposition, and hemato-

Review505

poietic stem cell and spermatogenic failure (Bender etal., 2002). Genomic instability is detectable in cells de-rived from this animal. The attrition of the hema-topoietic and male germ cell lineages occurs in largemeasure due to p53-mediated signaling triggered bygenomic instability (Bender et al., 2002).

Homozyogous inactivation of BRCA1 results in earlyembryonic lethality (Valerie and Povirk, 2003); however,mice homozygous for a BRCA1 hypomorphic allele andhaploinsufficient for p53 (BRCA1�11/�11/p53+/−) are via-ble and have many features reminiscent of acceleratedaging: wasting, skin atrophy, osteopenia, and malig-nancy (Cao et al., 2003b). There is compelling evidencethat this phenotype results from p53-dependent re-sponses to unrepaired DNA damage (Cao et al., 2003b).Baseline p53 protein levels are higher in BRCA1�11/�11/p53+/− mice than in p53+/− control animals. Similar tosome NHEJ-deficient MEFs, BRCA1�11/�11 MEFs showincreased chromosomal abnormalities and prematurecellular senescence, and BRCA1�11/�11 embryos showtissue SA-β-galactosidase activity, a marker of senes-cence. Thus, in both the Rad50s/s and BRCA1�11/�11/p53+/− mice, signs of premature aging occur as a con-sequence of the p53-mediated responses to unrepairedDNA damage. However, since BRCA1 and Rad50 areboth involved in multiple cellular processes, the possi-bility exists that loss of other functions beyond their

Figure 4. HR at DSBs

See text for details. Reproduced by permission with modification from Oncogene (Valerie and Povirk, 2003), copyright 2003 MacmillanPublishers Ltd.

DNA repair roles may contribute to these aging-like mu-tant phenotypes.

Single-Stranded DNA Lesions and AgingNucleotide Excision RepairDNA lesions that affect only one DNA strand are re-paired via BER or NER and its subpathways. AlthoughBER is thought to play a critical role in the repair ofoxidative lesions, mutations in genes involved in thispathway do not produce aging manifestations: they areeither lethal or confer no obvious phenotypes (Hasty etal., 2003). By contrast, lesions in some factors involvedin NER can lead to premature aging syndromes in miceand humans. NER is activated by a wide range of helix-distorting DNA lesions, including UV-induced photopro-ducts, bulky chemical adducts, and certain oxidativelesions. NER can be subdivided into two pathways,global genome NER (GG-NER) and transcription-cou-pled NER (TC-NER), which differ with respect to thelesion detected and some of the factors involved(Mitchell et al., 2003; Figure 5).

The basic NER machinery consists of the proteinsXPA through XPG, the CSA and CSB proteins, andother participants such as the basal transcription factorTFIIH. The GG-NER specific factors, XPC (in a complexwith the HR23B protein) and XPE are responsible fordetecting helix-distorting lesions that occur throughout

Cell506

the genome. The TC-NER specific factors, CSA and fmCSB, are involved in repair specifically in transcribed

regions and, when RNA polymerase II stalls at a lesion, tAcontribute to displacing the stalled polymerase to cre-

ate access for repair machinery. Following recogni- cmtion of the damaged DNA, common NER factors are

recruited in both GG-NER and TC-NER. TFIIH, which dpcontains two DNA helicases, XPB and XPD, unwinds

the DNA flanking the lesion. The single-strand DNA tt(ssDNA) binding protein RPA binds to and stabilizes the

unwound DNA strands, and XPA aids in lesion recogni- Cation. Two structure-specific endonucleases, XPG and

XPF (the latter in a complex with the protein ERCC1), ithen make single-strand incisions on either side of thelesion to release an oligonucleotide. The resulting gap i

mis filled by template-dependent DNA polymerization fol-lowed by ligation. l

lSeveral core NER factors have been implicated inprocesses aside from NER. XPB and XPD are critical in m

dRNA polymerase II transcription; CSB associates withRNA polymerases I and II; and XPF/ERCC1 has been g

dimplicated in repair of interstrand crosslinks, homologydirected repair, and processing of the 3# G strand over- X

vhang at telomeres.Some NER Defects Lead to Premature s

mAging PhenotypesNumerous human patients and mouse strains with de- p

cfects in different NER factors exist, and some have phe-notypes reminiscent of premature aging (Mitchell et al.,2003). Defects in the TC-NER-specific factors CSA or W

DCSB lead to Cockayne syndrome in humans, a severelydebilitating disorder with striking progeroid features. D

hCSA- and CSB-deficient mice show much milder phe-notypes than their human counterparts (Mitchell et al., m

d2003). Patients with specific mutations in XPD sufferfrom trichothiodystrophy (TTD), a disease character- n

sized by photosensitivity, brittle hair, skin defects, anda shortened life span. These patients show defects in t

etranscription of hair- and skin-specific transcripts (andperhaps other mRNAs) (Bergmann and Egly, 2001). In b

paddition, cells derived from TTD patients show NER de-fects, suggesting that multiple functions of XPD are de- i

Figure 5. Global-Genome NER and Tran-scription-Coupled NER

Factors with aging-related mutant pheno-types (XPD, CSA, CSB, ERCC1, and XPF) areshown in red. See text and Table 1 for details.Reprinted with modification from Mitchell et al.(2003) with permission from Elsevier.

ective in these individuals. Mice bearing a targetedutation in XPD that recapitulates a human TTD muta-

ion show similar manifestations to human TTD patients.dditionally, these XPDTTD mice show aging-associatedhanges such as wasting, scoliosis, osteoporosis, andelanocyte loss (de Boer et al., 2002). XPDTTD/XPAouble mutants show a much more rapid degenerativehenotype, suggesting that in the setting of impaired

ranscription and/or TC-NER, a total lack of NER is ex-remely deleterious. Similarly, mice deficient in bothSB and XPA also die within a few weeks after birth,lthough only cerebellar defects have been described

n detail in these animals (Murai et al., 2001).The aging-like phenotypes in human CS patients and

n XPDTTD, XPDTTD/XPA, and CSB/XPA mouse mutantsay be explained by the fact that a failure to repair

esions in transcribed genes can result in cell death,eading to tissue attrition and aging. Stalled RNA poly-

erase provides a signal for activation of p53-depen-ent apoptosis (Ljungman and Lane, 2004). In this re-ard, it will be of interest to determine whether p53eficiency rescues the aging-like features of XPDTTD,PDTTD/XPA, and CSB/XPA mice. Alternatively, the in-olvement of XPD and the CSB proteins in transcriptionuggests that impaired transcription of critical genesay play a role in causing these progeroid phenotypes,erhaps interacting with the repair defects in a compli-ated fashion.

RN and Agingefects in proteins with less well-defined functions inNA repair can lead to aging phenotypes as well. Theuman disease WS represents the best model for pre-ature aging in humans (Goto, 1997). WS patientsevelop premature graying, cataract, loss of subcuta-eous fat, skin atrophy, osteoporosis, diabetes, athero-clerosis, and malignancies. WS cells senesce prema-urely in culture. The gene defective in WS, WRN,ncodes a helicase of the RecQ family (a group definedy its similarity to E. coli RecQ helicase). WRN alsoossesses an exonuclease domain. WRN plays a role

n the maintenance of overall genomic stability, and

Review507

WRN may be involved in multiple DNA repair pathways(Bachrati and Hickson, 2003).

Recent experiments using mice bearing targeted mu-tations in WRN provide evidence that, with respect toaging, the most relevant sites of WRN function are thetelomeres. WRN-deficient mice do not recapitulate hu-man WS (Bachrati and Hickson, 2003). Based on theobservations that human WS cells show telomeric in-stability (Schulz et al., 1996; Tahara et al., 1997) andthat introduction of telomerase into WS fibroblasts canrescue their premature senescence (Wyllie et al., 2000),it was proposed that WRN-deficient mice may not de-monstrate a strong phenotype due to the abundantmouse telomere reserve (Lombard et al., 2000). This hy-pothesis has been proven correct by the generation ofmice deficient in both WRN and Terc (Chang et al.,2004; Du et al., 2004). In the WRN/Terc double deficientanimals, phenotypes reminiscent of human WS that arenot observed in either single mutant are present: os-teopenia, diabetes, and sarcomas. Phenotypes ordinar-ily seen in late-generation Terc-deficient animals alsooccur earlier in the compound mutants and are associ-ated with a greater degree of telomeric dysfunction.Whether or not these phenotypes depend on p53 func-tion is unknown; the premature senescence of humanWS fibroblasts is p53 dependent (Davis et al., 2003).The latter observation suggests that cellular checkpointfunctions may be involved in producing the clinical fea-tures of WS.

The exact role of WRN in telomere maintenance iscurrently unclear. WRN can be detected at the telo-meres in the absence of telomerase function in mam-malian cells (Johnson et al., 2001; Opresko et al., 2004),and Sgs1p, the S. cerevisiae WRN homolog, is requiredfor recombinational telomere maintenance in telom-erase-deficient cells in yeast (Cohen and Sinclair, 2001;Huang et al., 2001; Johnson et al., 2001). WRN also in-teracts with the telomeric protein TRF2 and can unwindand degrade telomeric D loop structures. These obser-vations suggest that WRN may play a role in providingtelomeric access to other factors involved in telomeremaintenance (Machwe et al., 2004; Opresko et al., 2004;Orren et al., 2002). Overall, the phenotype of the WRN/Terc double knockout mouse argues that defectivetelomere maintenance is an important factor in produc-ing the premature aging-like aspects of WS in humans,although it does not exclude a role for other functionsof WRN at nontelomeric sites.

Sir2: A Link between Metabolism, GenomeStability, and Life SpanIn the models discussed above, decreased genomicstability is associated with shortened life span. The Sir2family of proteins provides an example in which in-creased genomic stability extends life span (Blanderand Guarente, 2004). In S. cerevisiae, the chromatinregulatory factor Sir2 (silent information regulator-2)functions as an NAD-dependent histone deacetylase(Imai et al., 2000) to suppress recombination and turnsoff transcription at multiple genomic loci (Blander andGuarente, 2004). One metric of aging in yeast is thenumber of divisions that a single mother cell un-dergoes. The excision and replication of rDNA circles

is an important cause of mortality in yeast when lifespan is measured in this fashion (Sinclair and Guarente,1997). Loss of Sir2 increases rDNA recombination andshortens life span; whereas an extra genomic copy ofSir2, which increases rDNA stability, extends life span(Kaeberlein et al., 1999).

It has been proposed that Sir2 activity ties the energystatus of the yeast cell to longevity (Blander and Guar-ente, 2004; Lin et al., 2000). When nutrients are scarce,yeast cells preferentially employ respiration rather thanfermentation to generate ATP (Lin et al., 2002). Thismetabolic switch alters the metabolism of the cell,increasing the NAD/NADH ratio and/or decreasinglevels of the Sir2 inhibitor nicotinamide, in turn activa-ting Sir2 and increasing rDNA stability (reviewed inBlander and Guarente, [2004]). Strikingly, overexpres-sion or pharmacologic activation of Sir2 in worms andflies also extends life span (Rogina and Helfand, 2004;Tissenbaum and Guarente, 2001; Wood et al., 2004).Sir2-driven increased longevity in C. elegans requiresthe Daf-16 transcription factor (Tissenbaum and Guar-ente, 2001). Daf-16 is a critical mediator in the insulin-like signaling pathway, normally employed by worms toarrest as extremely long-lived larvae under unfavorableenvironmental conditions; certain mutations in thispathway confer longevity upon adult worms. The mech-anisms by which Sir2 extends life span in flies are cur-rently unclear. Sir2 family members may play a generalrole in mediating caloric restriction (Sohal and Wein-druch, 1996), an intervention capable of extending lifespan in many different organisms from yeast to mam-mals (Cohen et al., 2004; Howitz et al., 2003; Lin et al.,2000; Rogina and Helfand, 2004; Wood et al., 2004),although in yeast the involvement of Sir2 in CR appearsto be strain specific (Kaeberlein et al., 2004).

In mammals, there are seven Sir2 family members,designated SIRT1–SIRT7 (Frye, 2000); SIRT1 is themost highly related to S. cerevisiae Sir2. The role ofSIRT1 in mammalian longevity has not yet been directlytested, since on a pure strain background SIRT1-defi-cient animals die very early as a consequence ofmultiple developmental defects (Cheng et al., 2003;McBurney et al., 2003). Unlike yeast Sir2, which has noknown targets aside from histones, SIRT1 possesses alarge and growing list of targets, some of which, includ-ing p53 and forkhead transcription factors (mammalianhomologs of Daf-16), modulate cellular resistance tooxidative and genotoxic stress (Blander and Guarente,2004). Additionally, SIRT1, like Sir2, has recently beenshown to directly modify chromatin and silence tran-scription (Vaquero et al., 2004). It is now important todetermine whether SIRT1, in addition to silencing tran-scription, also suppresses recombination and genomicinstability via chromatin effects and if so, whether suchan activity could be involved in regulating aging inmammals. SIRT1 conditional alleles may allow studiesof the role of this protein in aging.

The true mammalian functional ortholog of Sir2, ifone exists, might also be a different mammalian Sir2family member (or members) than SIRT1. SIRT2 andSIRT3 are unlikely to play this role, because these pro-teins are cytoplasmic and mitochondrial, respectively,rather than chromatin associated (Blander and Guar-ente, 2004). Thus far, no information has been forth-

Cell508

coming regarding the functions of the four remaining ccSIRTs, SIRT4–SIRT7, but these remain candidates forfproteins that may regulate longevity through genomesstabilization.hoConclusionseThe hypothesis that nuclear DNA, a critically importantucellular constituent that cannot be replaced, is an im-

portant target of age-related change is supported bysevidence that nuclear DNA damage and mutations ac-scumulate with age. While ROS are likely to be one im-fportant source of this damage, there are numerousAother cellular and environmental sources of damage,p

and the impact of such lesions may be enhanced byc

age-related compromise of DNA repair. In the latterd

context, most premature aging syndromes are causedt

by mutations in genes encoding proteins involved ini

DNA repair (Karanjawala and Lieber, 2004). Accumula- ation of mutations in critical genes may be one general amechanism by which compromised DNA repair could bcontribute to aging. In addition, p53-mediated senes- 2cence and apoptosis, in response to DNA damage, also clikely contribute to aging (Figure 6). Indeed, the fact that ilesions in several disparate repair systems cause phe- hnotypes that are broadly similar to one another (Table 1) ris consistent with the notion that the specific chemical onature of the accumulated DNA lesions may be less limportant than their ability to activate the common cel- slular checkpoint machinery. m

It remains unclear why only certain DNA repair mu- ctants in particular pathways show progeroid pheno- mtypes. In some cases, it may simply be that some mu- mtants have not been scrutinized sufficiently to reveal ssuch effects. However, genetic background effects al- mmost certainly play an important role in modifying the iaging-like manifestations of DNA repair deficiencies.Also, it must be remembered that many of the DNA re- cpair genes and factors implicated in suppressing aging balso play roles in cellular processes other than DNA re- mpair, and therefore aging-like phenotypes might be en- ahanced by impairment of other cellular functions in d

oconjunction with altered DNA repair. In addition, as dis-

Figure 6. A Model for the Role of UnresolvedDNA Lesions in Aging

Unrepaired DNA lesions can activate the cellcycle checkpoint machinery, leading to se-nescence or apoptosis and subsequent cel-lular attrition and tissue dysfunction. Sometypes of unrepaired DNA damage—like DNADSBs—can trigger genomic instability, whichcan in turn lead to further DNA damage. Inaddition, unrepaired DNA damage can di-rectly compromise cellular processes liketranscription, an effect that could also impairtissue function. Such unrepaired DNA dam-age will be frequent in DNA repair mutantsand lead to accelerated degenerative changes;in wild-type organisms unrepaired damage isinfrequent, owing to efficient repair systems.

ussed above, environmental factors including housingonditions, infectious agents, diet, and many other in-luences, also likely play a significant role in the expres-ion of aging phenotypes in mouse models and, per-aps, in humans as well. Thus, aging is likely theutcome of a complex interplay between the geneticndowment of an organism and the stresses placedpon it by its particular environment.All mouse models that link DNA repair to aging pos-

ess defects in DNA repair and have shortened lifepans. It is important to bear in mind the potential pit-alls of such models (Hasty and Vijg, 2004; Miller, 2004).ging encompasses a wide spectrum of degenerativerocesses, many of which are quite nonspecific, bothlinically and pathologically (Harrison, 1994). Thus, it isifficult to arrive at a strict, experimentally useful defini-ion of aging. Factors implicated in organismal declinen genetic models might not play a role in the normalging processes. A related difficulty is that prematureging models fail to recapitulate all aspects of agingut are instead “segmental progerias” (Hasty and Vijg,004; Miller, 2004); that is, they reproduce in an ac-elerated fashion some but not all aspects of aging as

t occurs in wild-type animals. In this regard, the myriadistopathologic changes of normal aging (Table 2) cor-espond poorly with the changes that occur in modelsf premature aging (Table 1). Mammalian aging is not

ikely a single process but rather the decline of manyomatic functions, heavily influenced by the environ-ent; this is a complex interplay that is extremely diffi-

ult to model accurately. For these reasons, geneticodels of extended life span are likely to be more infor-ative than models with reduced longevity with re-

pect to physiologically relevant causes of decline andortality; in such long-lived organisms, life span-limit-

ng factors must of necessity be counteracted.There are many outstanding questions regarding the

onnection between DNA repair and aging that willenefit from the application of emerging techniques inolecular biology and genetics. In models of premature

ging, the most vulnerable system fails first, leading toeath and precluding gain of insights from effects onther, potentially more relevant, organ systems. This

Review509

difficulty might be overcome with hypomorphic allelesor via conditional knockouts of relevant genes. The lat-ter approach, for example, might allow an evaluation ofthe roles of genes essential during embryogenesis inthe aging process. Such an approach also could permitinsights into the types of DNA lesions that are mostimportant in causing aging in different tissues by allow-ing impairment of specific DNA repair pathways in de-fined cell populations. For example, tissue-specific de-letion of DNA damage-response/checkpoint genes inDNA repair mutants showing evidence of prematureaging might allow the direct effects of these DNAlesions to be distinguished from the cellular responsesto them. Further insights into the basic biology of DNArepair proteins, particularly regarding the exact natureof the DNA lesions that these repair systems respondto and their roles in checkpoints, may also shed morelight on the role of various DNA lesions in contributingto aging.

Eventually, the importance of DNA damage in agingmight be directly tested via the generation of experi-mental organisms with enhanced efficiency of DNAmaintenance, which would be predicted to show re-tarded aging. Although the plethora of different repairsystems makes this a challenging, if not impossible,undertaking, the existence of Sir2, a regulator of ge-nome stability and aging in yeast, offers encourage-ment for those seeking similar global regulators ofgenome maintenance and potentially DNA repair inmammals.

Acknowledgments

The authors would like to thank members of the Alt lab, as well asRoderick Bronson, Ronny Drapkin, Toren Finkel, Lenny Guarente,Marcia Haigis, F. Bradley Johnson, Michael Lieber, Kevin Mills,Ralph Scully, Peter Sorger, Tony Wynshaw-Boris, and especiallyNed Sharpless for helpful discussions and comments on the manu-script. F.W.A. is an Investigator of the Howard Hughes Medical In-stitute. This work was supported by an Ellison Foundation SeniorScholar Award (to F.W.A), a K08 award from NIA/NIH (to D.B.L.),a Pfizer Postdoctoral Fellowship in Immunology/Rheumatology (toK.F.C.), a Senior Postdoctoral Fellowship from The Leukemia andLymphoma Society (to R.M.), and a Long-Term Fellowship from theEuropean Molecular Biology Organization (to S.F.).

The authors would like to apologize to those whose work wasnot cited, due to space constraints. We have referenced other arti-cles in this issue and recent in-depth reviews that provide thesereferences.

References

Bachrati, C.Z., and Hickson, I.D. (2003). RecQ helicases: suppres-sors of tumorigenesis and premature aging. Biochem. J. 374, 577–606.

Baker, D.J., Jeganathan, K.B., Cameron, J.D., Thompson, M., Ju-neja, S., Kopecka, A., Kumar, R., Jenkins, R.B., de Groen, P.C.,Roche, P., and van Deursen, J.M. (2004). BubR1 insufficiencycauses early onset of aging-associated phenotypes and infertilityin mice. Nat. Genet. 36, 744–749.

Balaban, R.S., Shino, N., and Toren, F. (2005). Mitochondria, oxi-dants, and aging. Cell 120, this issue, 483–495.

Barja, G. (2004). Free radicals and aging. Trends Neurosci. 27,595–600.

Barja, G., and Herrero, A. (2000). Oxidative damage to mito-

chondrial DNA is inversely related to maximum life span in the heartand brain of mammals. FASEB J. 14, 312–318.

Barzilai, A., Rotman, G., and Shiloh, Y. (2002). ATM deficiency andoxidative stress: a new dimension of defective response to DNAdamage. DNA Repair (Amst.) 1, 3–25.

Bassing, C.H., and Alt, F.W. (2004). The cellular response to generaland programmed DNA double strand breaks. DNA Repair (Amst.)3, 781–796.

Ben-Porath, I., and Weinberg, R.A. (2004). When cells get stressed:an integrative view of cellular senescence. J. Clin. Invest. 113, 8–13.

Bender, C.F., Sikes, M.L., Sullivan, R., Huye, L.E., Le Beau, M.M.,Roth, D.B., Mirzoeva, O.K., Oltz, E.M., and Petrini, J.H. (2002). Can-cer predisposition and hematopoietic failure in Rad50(S/S) mice.Genes Dev. 16, 2237–2251.

Bergmann, E., and Egly, J.M. (2001). Trichothiodystrophy, a tran-scription syndrome. Trends Genet. 17, 279–286.

Blander, G., and Guarente, L. (2004). The Sir2 family of protein de-acetylases. Annu. Rev. Biochem. 73, 417–435.

Bronson, R.T., and Lipman, R.D. (1991). Reduction in rate of occur-rence of age related lesions in dietary restricted laboratory mice.Growth Dev. Aging 55, 169–184.

Browne, S.E., Roberts, L.J., 2nd, Dennery, P.A., Doctrow, S.R., Beal,M.F., Barlow, C., and Levine, R.L. (2004). Treatment with a catalyticantioxidant corrects the neurobehavioral defect in ataxia-telangiec-tasia mice. Free Radic. Biol. Med. 36, 938–942.

Campisi, J. (2002). Between Scylla and Charybdis: p53 links tumorsuppression and aging. Mech. Ageing Dev. 123, 567–573.

Cao, J., Venton, L., Sakata, T., and Halloran, B.P. (2003a). Expres-sion of RANKL and OPG correlates with age-related bone loss inmale C57BL/6 mice. J. Bone Miner. Res. 18, 270–277.

Cao, L., Li, W., Kim, S., Brodie, S.G., and Deng, C.X. (2003b). Senes-cence, aging, and malignant transformation mediated by p53 inmice lacking the Brca1 full-length isoform. Genes Dev. 17, 201–213.

Chang, S., Multani, A.S., Cabrera, N.G., Naylor, M.L., Laud, P., Lom-bard, D., Pathak, S., Guarente, L., and DePinho, R.A. (2004). Essen-tial role of limiting telomeres in the pathogenesis of Werner syn-drome. Nat. Genet. 36, 877–882.

Cheng, H.L., Mostoslavsky, R., Saito, S., Manis, J.P., Gu, Y., Patel,P., Bronson, R., Appella, E., Alt, F.W., and Chua, K.F. (2003). Devel-opmental defects and p53 hyperacetylation in Sir2 homolog(SIRT1)-deficient mice. Proc. Natl. Acad. Sci. USA 100, 10794–10799.

Cohen, H., and Sinclair, D.A. (2001). Recombination-mediatedlengthening of terminal telomeric repeats requires the Sgs1 DNAhelicase. Proc. Natl. Acad. Sci. USA 98, 3174–3179.

Cohen, H.Y., Miller, C., Bitterman, K.J., Wall, N.R., Hekking, B., Kess-ler, B., Howitz, K.T., Gorospe, M., de Cabo, R., and Sinclair, D.A.(2004). Calorie restriction promotes mammalian cell survival by in-ducing the SIRT1 deacetylase. Science 305, 390–392.

Couedel, C., Mills, K.D., Barchi, M., Shen, L., Olshen, A., Johnson,R.D., Nussenzweig, A., Essers, J., Kanaar, R., Li, G.C., et al. (2004).Collaboration of homologous recombination and nonhomologousend-joining factors for the survival and integrity of mice and cells.Genes Dev. 18, 1293–1304.

d'Adda di Fagagna, F., Hande, M.P., Tong, W.M., Roth, D., Lans-dorp, P.M., Wang, Z.Q., and Jackson, S.P. (2001). Effects of DNAnonhomologous end-joining factors on telomere length and chro-mosomal stability in mammalian cells. Curr. Biol. 11, 1192–1196.

Davis, T., Singhrao, S.K., Wyllie, F.S., Haughton, M.F., Smith, P.J.,Wiltshire, M., Wynford-Thomas, D., Jones, C.J., Faragher, R.G., andKipling, D. (2003). Telomere-based proliferative lifespan barriers inWerner-syndrome fibroblasts involve both p53-dependent andp53-independent mechanisms. J. Cell Sci. 116, 1349–1357.

de Boer, J., Andressoo, J.O., de Wit, J., Huijmans, J., Beems, R.B.,van Steeg, H., Weeda, G., van der Horst, G.T., van Leeuwen, W.,Themmen, A.P., et al. (2002). Premature aging in mice deficient inDNA repair and transcription. Science 296, 1276–1279.

de Lange, T. (2002). Protection of mammalian telomeres. Oncogene21, 532–540.

Cell510

Dong, Y., Hakimi, M.A., Chen, X., Kumaraswamy, E., Cooch, N.S., sgGodwin, A.K., and Shiekhattar, R. (2003). Regulation of BRCC, a

holoenzyme complex containing BRCA1 and BRCA2, by a signalo- Isome-like subunit and its role in DNA repair. Mol. Cell 12, 1087– T1099. dDu, X., Shen, J., Kugan, N., Furth, E.E., Lombard, D.B., Cheung, C., IPak, S., Luo, G., Pignolo, R.J., DePinho, R.A., et al. (2004). Telomere lshortening exposes functions for the mouse werner and bloom 1syndrome genes. Mol. Cell. Biol. 24, 8437–8446.

IEspejel, S., and Blasco, M.A. (2002). Identification of telomere- Ndependent “senescence-like” arrest in mouse embryonic fibro- (blasts. Exp. Cell Res. 276, 242–248. rEspejel, S., Franco, S., Rodriguez-Perales, S., Bouffler, S.D., Cigu- Jdosa, J.C., and Blasco, M.A. (2002a). Mammalian Ku86 mediates Wchromosomal fusions and apoptosis caused by critically short telo- hmeres. EMBO J. 21, 2207–2219. iEspejel, S., Franco, S., Sgura, A., Gae, D., Bailey, S.M., Taccioli, KG.E., and Blasco, M.A. (2002b). Functional interaction between cDNA-PKcs and telomerase in telomere length maintenance. EMBO eJ. 21, 6275–6287. KEspejel, S., Klatt, P., Murcia, J.M., Martin-Caballero, J., Flores, J.M., STaccioli, G., de Murcia, G., and Blasco, M.A. (2004a). Impact of Ptelomerase ablation on organismal viability, aging, and tumorigene- Ksis in mice lacking the DNA repair proteins PARP-1, Ku86, or DNA- MPKcs. J. Cell Biol. 167, 627–638.

KEspejel, S., Martin, M., Klatt, P., Martin-Caballero, J., Flores, J.M., land Blasco, M.A. (2004b). Shorter telomeres, accelerated ageing

Kand increased lymphoma in DNA-PKcs-deficient mice. EMBO Rep.t5, 503–509.C

Ferguson, D.O., and Alt, F.W. (2001). DNA double strand break re-Kpair and chromosomal translocation: lessons from animal models.pOncogene 20, 5572–5579.N

Finkel, T., and Holbrook, N.J. (2000). Oxidants, oxidative stress andLthe biology of ageing. Nature 408, 239–247.s

Frye, R.A. (2000). Phylogenetic classification of prokaryotic and aeukaryotic Sir2-like proteins. Biochem. Biophys. Res. Commun. U273, 793–798.

LGarcia-Cao, I., Garcia-Cao, M., Martin-Caballero, J., Criado, L.M., CKlatt, P., Flores, J.M., Weill, J.C., Blasco, M.A., and Serrano, M. i(2002). “Super p53” mice exhibit enhanced DNA damage response,

Lare tumor resistant and age normally. EMBO J. 21, 6225–6235.

NGoto, M. (1997). Hierarchical deterioration of body systems in Wer- Sner's syndrome: implications for normal ageing. Mech. Ageing Dev.

L98, 239–254.

CGu, Y., Seidl, K.J., Rathbun, G.A., Zhu, C., Manis, J.P., van der eStoep, N., Davidson, L., Cheng, H.L., Sekiguchi, J.M., Frank, K., et tal. (1997). Growth retardation and leaky SCID phenotype of Ku70- Ldeficient mice. Immunity 7, 653–665. gHamilton, M.L., Van Remmen, H., Drake, J.A., Yang, H., Guo, Z.M., LKewitt, K., Walter, C.A., and Richardson, A. (2001). Does oxidative Jdamage to DNA increase with age? Proc. Natl. Acad. Sci. USA 98, e10469–10474. iHarrison, D.E. (1994). Potential misinterpretations using models of Laccelerated aging. J. Gerontol. 49, B245–B246. BHart, R.W., and Setlow, R.B. (1974). Correlation between deoxyribo- bnucleic acid excision-repair and life-span in a number of mamma- Mlian species. Proc. Natl. Acad. Sci. USA 71, 2169–2173. nHasty, P., and Vijg, J. (2004). Accelerating aging by mouse reverse Ogenetics: a rational approach to understanding longevity. Aging MCell 3, 55–65. THasty, P., Campisi, J., Hoeijmakers, J., van Steeg, H., and Vijg, J. o(2003). Aging and genome maintenance: lessons from the mouse? 3Science 299, 1355–1359. MHowitz, K.T., Bitterman, K.J., Cohen, H.Y., Lamming, D.W., Lavu, S., WWood, J.G., Zipkin, R.E., Chung, P., Kisielewski, A., Zhang, L.L., et lal. (2003). Small molecule activators of sirtuins extend Saccharo- gmyces cerevisiae lifespan. Nature 425, 191–196. M

sHuang, P., Pryde, F.E., Lester, D., Maddison, R.L., Borts, R.H., Hick-

on, I.D., and Louis, E.J. (2001). SGS1 is required for telomere elon-ation in the absence of telomerase. Curr. Biol. 11, 125–129.

mai, S., Armstrong, C.M., Kaeberlein, M., and Guarente, L. (2000).ranscriptional silencing and longevity protein Sir2 is an NAD-ependent histone deacetylase. Nature 403, 795–800.

tahana, K., Campisi, J., and Dimri, G.P. (2004). Mechanisms of cel-ular senescence in human and mouse cells. Biogerontology 5,–10.

to, K., Hirao, A., Arai, F., Matsuoka, S., Takubo, K., Hamaguchi, I.,omiyama, K., Hosokawa, K., Sakurada, K., Nakagata, N., et al.

2004). Regulation of oxidative stress by ATM is required for self-enewal of haematopoietic stem cells. Nature 431, 997–1002.

ohnson, F.B., Marciniak, R.A., McVey, M., Stewart, S.A., Hahn,.C., and Guarente, L. (2001). The Saccharomyces cerevisiae WRN

omolog Sgs1p participates in telomere maintenance in cells lack-ng telomerase. EMBO J. 20, 905–913.

aeberlein, M., McVey, M., and Guarente, L. (1999). The SIR2/3/4omplex and SIR2 alone promote longevity in Saccharomyces cer-visiae by two different mechanisms. Genes Dev. 13, 2570–2580.

aeberlein, M., Kirkland, K.T., Fields, S., and Kennedy, B.K. (2004).ir2-independent life span extension by calorie restriction in yeast.LoS Biol. 2, e296. 10.1371/journal.pbio.0020296

aranjawala, Z.E., and Lieber, M.R. (2004). DNA damage and aging.ech. Ageing Dev. 125, 405–416.

irkwood, T.B., and Holliday, R. (1979). The evolution of ageing andongevity. Proc. R. Soc. Lond. B. Biol. Sci. 205, 531–546.

rtolica, A., and Campisi, J. (2002). Cancer and aging: a model forhe cancer promoting effects of the aging stroma. Int. J. Biochem.ell Biol. 34, 1401–1414.

wan, K.Y., Moens, P.B., and Wang, J.C. (2003). Infertility and aneu-loidy in mice lacking a type IA DNA topoisomerase III beta. Proc.atl. Acad. Sci. USA 100, 2526–2531.

ebel, M., and Leder, P. (1998). A deletion within the murine Werneryndrome helicase induces sensitivity to inhibitors of topoisomer-se and loss of cellular proliferative capacity. Proc. Natl. Acad. Sci.SA 95, 13097–13102.

ee, H.W., Blasco, M.A., Gottlieb, G.J., Horner, J.W., 2nd, Greider,.W., and DePinho, R.A. (1998). Essential role of mouse telomerase

n highly proliferative organs. Nature 392, 569–574.

in, S.J., Defossez, P.A., and Guarente, L. (2000). Requirement ofAD and SIR2 for life-span extension by calorie restriction inaccharomyces cerevisiae. Science 289, 2126–2128.

in, S.J., Kaeberlein, M., Andalis, A.A., Sturtz, L.A., Defossez, P.A.,ulotta, V.C., Fink, G.R., and Guarente, L. (2002). Calorie restrictionxtends Saccharomyces cerevisiae lifespan by increasing respira-ion. Nature 418, 344–348.

jungman, M., and Lane, D.P. (2004). Transcription—guarding theenome by sensing DNA damage. Nat. Rev. Cancer 4, 727–737.

ombard, D.B., Beard, C., Johnson, B., Marciniak, R.A., Dausman,., Bronson, R., Buhlmann, J.E., Lipman, R., Curry, R., Sharpe, A.,t al. (2000). Mutations in the WRN gene in mice accelerate mortal-

ty in a p53-null background. Mol. Cell. Biol. 20, 3286–3291.

u, T., Pan, Y., Kao, S.Y., Li, C., Kohane, I., Chan, J., and Yankner,.A. (2004). Gene regulation and DNA damage in the ageing humanrain. Nature 429, 883–891.

achwe, A., Xiao, L., and Orren, D.K. (2004). TRF2 recruits the Wer-er syndrome (WRN) exonuclease for processing of telomeric DNA.ncogene 23, 149–156.

aier, B., Gluba, W., Bernier, B., Turner, T., Mohammad, K., Guise,., Sutherland, A., Thorner, M., and Scrable, H. (2004). Modulationf mammalian life span by the short isoform of p53. Genes Dev. 18,06–319.

cBurney, M.W., Yang, X., Jardine, K., Hixon, M., Boekelheide, K.,ebb, J.R., Lansdorp, P.M., and Lemieux, M. (2003). The mamma-

ian SIR2alpha protein has a role in embryogenesis and gameto-enesis. Mol. Cell. Biol. 23, 38–54.

cMurray, M.A., and Gottschling, D.E. (2003). An age-inducedwitch to a hyper-recombinational state. Science 301, 1908–1911.

Review511

McWhir, J., Selfridge, J., Harrison, D.J., Squires, S., and Melton,D.W. (1993). Mice with DNA repair gene (ERCC-1) deficiency haveelevated levels of p53, liver nuclear abnormalities and die beforeweaning. Nat. Genet. 5, 217–224.

Meek, D. (2004). The p53 response to DNA damage. DNA Repair(Amst) 3, 1049–1056.

Migliaccio, E., Giorgio, M., Mele, S., Pelicci, G., Reboldi, P., Pan-dolfi, P.P., Lanfrancone, L., and Pelicci, P.G. (1999). The p66shcadaptor protein controls oxidative stress response and life span inmammals. Nature 402, 309–313.

Miller, R.A. (2004). ‘Accelerated aging’: a primrose path to insight?Aging Cell 3, 47–51.

Mills, K.D., Ferguson, D.O., Essers, J., Eckersdorff, M., Kanaar, R.,and Alt, F.W. (2004). Rad54 and DNA Ligase IV cooperate to main-tain mammalian chromatid stability. Genes Dev. 18, 1283–1292.

Mitchell, J.R., Hoeijmakers, J.H., and Niedernhofer, L.J. (2003). Di-vide and conquer: nucleotide excision repair battles cancer andageing. Curr. Opin. Cell Biol. 15, 232–240.

Murai, M., Enokido, Y., Inamura, N., Yoshino, M., Nakatsu, Y., vander Horst, G.T., Hoeijmakers, J.H., Tanaka, K., and Hatanaka, H.(2001). Early postnatal ataxia and abnormal cerebellar developmentin mice lacking Xeroderma pigmentosum Group A and Cockaynesyndrome Group B DNA repair genes. Proc. Natl. Acad. Sci. USA98, 13379–13384.

Opresko, P.L., Otterlei, M., Graakjaer, J., Bruheim, P., Dawut, L.,Kolvraa, S., May, A., Seidman, M.M., and Bohr, V.A. (2004). The Wer-ner Syndrome helicase and exonuclease cooperate to resolve telo-meric D loops in a manner regulated by TRF1 and TRF2. Mol. Cell14, 763–774.

Orren, D.K., Theodore, S., and Machwe, A. (2002). The Werner syn-drome helicase/exonuclease (WRN) disrupts and degrades D-loopsin vitro. Biochemistry 41, 13483–13488.

Pandita, T.K. (2002). ATM function and telomere stability. Oncogene21, 611–618.

Parrinello, S., Samper, E., Krtolica, A., Goldstein, J., Melov, S., andCampisi, J. (2003). Oxygen sensitivity severely limits the replicativelifespan of murine fibroblasts. Nat. Cell Biol. 5, 741–747.

Pelicci, P.G. (2004). Do tumor-suppressive mechanisms contributeto organism aging by inducing stem cell senescence? J. Clin. In-vest. 113, 4–7.

Petiniot, L.K., Weaver, Z., Vacchio, M., Shen, R., Wangsa, D., Bar-low, C., Eckhaus, M., Steinberg, S.M., Wynshaw-Boris, A., Ried, T.,and Hodes, R.J. (2002). RAG-mediated V(D)J recombination is notessential for tumorigenesis in Atm-deficient mice. Mol. Cell. Biol.22, 3174–3177.

Rockwood, L.D., Nussenzweig, A., and Janz, S. (2003). Paradoxicaldecrease in mutant frequencies and chromosomal rearrangementsin a transgenic lacZ reporter gene in Ku80 null mice deficient inDNA double strand break repair. Mutat. Res. 529, 51–58.

Rogina, B., and Helfand, S.L. (2004). Sir2 mediates longevity in thefly through a pathway related to calorie restriction. Proc. Natl.Acad. Sci. USA 101, 15998–16003.

Rooney, S., Alt, F.W., Lombard, D., Whitlow, S., Eckersdorff, M.,Fleming, J., Fugmann, S., Ferguson, D.O., Schatz, D.G., and Seki-guchi, J. (2003). Defective DNA repair and increased genomic insta-bility in Artemis-deficient murine cells. J. Exp. Med. 197, 553–565.

Rothkamm, K., Kruger, I., Thompson, L.H., and Lobrich, M. (2003).Pathways of DNA double-strand break repair during the mamma-lian cell cycle. Mol. Cell. Biol. 23, 5706–5715.

Rudolph, K.L., Chang, S., Lee, H.W., Blasco, M., Gottlieb, G.J.,Greider, C., and DePinho, R.A. (1999). Longevity, stress response,and cancer in aging telomerase-deficient mice. Cell 96, 701–712.

Schubert, R., Erker, L., Barlow, C., Yakushiji, H., Larson, D., Russo,A., Mitchell, J.B., and Wynshaw-Boris, A. (2004). Cancer chemo-prevention by the antioxidant tempol in Atm-deficient mice. Hum.Mol. Genet. 13, 1793–1802.

Schulz, V.P., Zakian, V.A., Ogburn, C.E., McKay, J., Jarzebowicz,A.A., Edland, S.D., and Martin, G.M. (1996). Accelerated loss of

telomeric repeats may not explain accelerated replicative declineof Werner syndrome cells. Hum. Genet. 97, 750–754.

Scully, R., Xie, A., and Nagaraju, G. (2004). Molecular functions ofBRCA1 in the DNA damage response. Cancer Biol. Ther. 3, 521–527.

Sedelnikova, O.A., Horikawa, I., Zimonjic, D.B., Popescu, N.C.,Bonner, W.M., and Barrett, J.C. (2004). Senescing human cells andageing mice accumulate DNA lesions with unrepairable double-strand breaks. Nat. Cell Biol. 6, 168–170.

Sekiguchi, J., Ferguson, D.O., Chen, H.T., Yang, E.M., Earle, J.,Frank, K., Whitlow, S., Gu, Y., Xu, Y., Nussenzweig, A., and Alt, F.W.(2001). Genetic interactions between ATM and the nonhomologousend-joining factors in genomic stability and development. Proc.Natl. Acad. Sci. USA 98, 3243–3248.

Seluanov, A., Mittelman, D., Pereira-Smith, O.M., Wilson, J.H., andGorbunova, V. (2004). DNA end joining becomes less efficient andmore error-prone during cellular senescence. Proc. Natl. Acad. Sci.USA 101, 7624–7629.

Shiloh, Y., and Kastan, M.B. (2001). ATM: genome stability, neuronaldevelopment, and cancer cross paths. Adv. Cancer Res. 83, 209–254.

Sinclair, D.A., and Guarente, L. (1997). Extrachromosomal rDNA cir-cles—a cause of aging in yeast. Cell 91, 1033–1042.

Singh, N.P., Ogburn, C.E., Wolf, N.S., van Belle, G., and Martin, G.M.(2001). DNA double-strand breaks in mouse kidney cells with age.Biogerontology 2, 261–270.

Smogorzewska, A., Karlseder, J., Holtgreve-Grez, H., Jauch, A., andde Lange, T. (2002). DNA ligase IV-dependent NHEJ of deprotectedmammalian telomeres in G1 and G2. Curr. Biol. 12, 1635–1644.

Sohal, R.S., and Weindruch, R. (1996). Oxidative stress, caloric re-striction, and aging. Science 273, 59–63.

Stracker, T.H., Theunissen, J.W., Morales, M., and Petrini, J.H.(2004). The Mre11 complex and the metabolism of chromosomebreaks: the importance of communicating and holding things to-gether. DNA Repair (Amst.) 3, 845–854.

Sun, L.Q., Lee, D.W., Zhang, Q., Xiao, W., Raabe, E.H., Meeker, A.,Miao, D., Huso, D.L., and Arceci, R.J. (2004). Growth retardationand premature aging phenotypes in mice with disruption of theSNF2-like gene, PASG. Genes Dev. 18, 1035–1046.

Tahara, H., Tokutake, Y., Maeda, S., Kataoka, H., Watanabe, T., Sa-toh, M., Matsumoto, T., Sugawara, M., Ide, T., Goto, M., et al. (1997).Abnormal telomere dynamics of B-lymphoblastoid cell strains fromWerner's syndrome patients transformed by Epstein-Barr virus.Oncogene 15, 1911–1920.

Takata, M., Sasaki, M.S., Sonoda, E., Morrison, C., Hashimoto, M.,Utsumi, H., Yamaguchi-Iwai, Y., Shinohara, A., and Takeda, S.(1998). Homologous recombination and non-homologous end-join-ing pathways of DNA double-strand break repair have overlappingroles in the maintenance of chromosomal integrity in vertebratecells. EMBO J. 17, 5497–5508.

Tian, M., Shinkura, R., Shinkura, N., and Alt, F.W. (2004). Growthretardation, early death, and DNA repair defects in mice deficientfor the nucleotide excision repair enzyme XPF. Mol. Cell. Biol. 24,1200–1205.

Ting, N.S., and Lee, W.H. (2004). The DNA double-strand break re-sponse pathway: becoming more BRCAish than ever. DNA Repair(Amst.) 3, 935–944.

Tissenbaum, H.A., and Guarente, L. (2001). Increased dosage of asir-2 gene extends lifespan in Caenorhabditis elegans. Nature 410,227–230.

Trifunovic, A., Wredenberg, A., Falkenberg, M., Spelbrink, J.N.,Rovio, A.T., Bruder, C.E., Bohlooly, Y.M., Gidlof, S., Oldfors, A., Wi-bom, R., et al. (2004). Premature ageing in mice expressing de-fective mitochondrial DNA polymerase. Nature 429, 417–423.

Tyner, S.D., Venkatachalam, S., Choi, J., Jones, S., Ghebranious,N., Igelmann, H., Lu, X., Soron, G., Cooper, B., Brayton, C., et al.(2002). p53 mutant mice that display early ageing-associated phe-notypes. Nature 415, 45–53.

Cell512

Valerie, K., and Povirk, L.F. (2003). Regulation and mechanisms ofmammalian double-strand break repair. Oncogene 22, 5792–5812.

Van Remmen, H., Ikeno, Y., Hamilton, M., Pahlavani, M., Wolf, N.,Thorpe, S.R., Alderson, N.L., Baynes, J.W., Epstein, C.J., Huang,T.T., et al. (2003). Life-long reduction in MnSOD activity results inincreased DNA damage and higher incidence of cancer but doesnot accelerate aging. Physiol. Genomics 16, 29–37.

Vaquero, A., Scher, M., Lee, D., Erdjument-Bromage, H., Tempst, P.,and Reinberg, D. (2004). Human SirT1 interacts with histone H1 andpromotes formation of facultative heterochromatin. Mol. Cell 16,93–105.

Vijg, J. (2000). Somatic mutations and aging: a re-evaluation. Mutat.Res. 447, 117–135.

Vogel, H., Lim, D.S., Karsenty, G., Finegold, M., and Hasty, P. (1999).Deletion of Ku86 causes early onset of senescence in mice. Proc.Natl. Acad. Sci. USA 96, 10770–10775.

Weeda, G., Donker, I., de Wit, J., Morreau, H., Janssens, R., Vissers,C.J., Nigg, A., van Steeg, H., Bootsma, D., and Hoeijmakers, J.H.(1997). Disruption of mouse ERCC1 results in a novel repair syn-drome with growth failure, nuclear abnormalities and senescence.Curr. Biol. 7, 427–439.

Wong, K.K., Chang, S., Weiler, S.R., Ganesan, S., Chaudhuri, J.,Zhu, C., Artandi, S.E., Rudolph, K.L., Gottlieb, G.J., Chin, L., et al.(2000). Telomere dysfunction impairs DNA repair and enhancessensitivity to ionizing radiation. Nat. Genet. 26, 85–88.

Wong, K.K., Maser, R.S., Bachoo, R.M., Menon, J., Carrasco, D.R.,Gu, Y., Alt, F.W., and DePinho, R.A. (2003). Telomere dysfunctionand Atm deficiency compromises organ homeostasis and acceler-ates ageing. Nature 421, 643–648.

Wood, J.G., Rogina, B., Lavu, S., Howitz, K., Helfand, S.L., Tatar,M., and Sinclair, D. (2004). Sirtuin activators mimic caloric restric-tion and delay ageing in metazoans. Nature 430, 686–689.

Wyllie, F.S., Jones, C.J., Skinner, J.W., Haughton, M.F., Wallis, C.,Wynford-Thomas, D., Faragher, R.G., and Kipling, D. (2000). Telo-merase prevents the accelerated cell ageing of Werner syndromefibroblasts. Nat. Genet. 24, 16–17.

Xu, Y., Yang, E.M., Brugarolas, J., Jacks, T., and Baltimore, D.(1998). Involvement of p53 and p21 in cellular defects and tumori-genesis in Atm−/− mice. Mol. Cell. Biol. 18, 4385–4390.