Original Contribution

MnSOD OVEREXPRESSION EXTENDS THE YEAST CHRONOLOGICAL (G0)LIFE SPAN BUT ACTS INDEPENDENTLY OF Sir2p HISTONE

DEACETYLASE TO SHORTEN THE REPLICATIVE LIFE SPAN OFDIVIDING CELLS

NICHOLAS HARRIS,* V ITOR COSTA,†,‡ MORAG MACLEAN,* M EHDI MOLLAPOUR,* PEDRO MORADAS-FERREIRA,†,‡

and PETER W. PIPER**Department of Biochemistry and Molecular Biology, University College London, London, England;†Institute of Molecular and

Cellular Biology, and‡Abel Salazar Institute of Biomedical Science, University of Porto, Porto, Portugal

(Received 28 January 2003;Revised 17 March 2003;Accepted 21 March 2003)

Abstract—Studies inDrosophila andCaenorhabditis elegans have shown increased longevity with the increased freeradical scavenging that accompanies overexpression of oxidant-scavenging enzymes. This study used yeast, anothermodel for aging research, to probe the effects of overexpressing the major activity protecting against superoxidegenerated by the mitochondrial respiratory chain. Manganese superoxide dismutase (MnSOD) overexpression increasedchronological life span (optimized survival of stationary (G0) yeast over time), showing this is a survival ultimatelylimited by oxidative stress. In contrast, the same overexpression dramatically reduced the replicative life span ofdividing cells (the number of daughter buds produced by each newly born mother cell). This reduction in thegenerational life span by MnSOD overexpression was greater than that generated by loss of the major redox-responsiveregulator of the yeast replicative life span, NAD�-dependent Sir2p histone deacetylase. It was also independent of thelatter activity. Expression of a mitochondrially targeted green fluorescent protein in the MnSOD overexpressor revealedthat the old mother cells of this overexpressor, which had divided for a few generations, were defective in segregationof the mitochondrion from the mother to daughter. Mitochondrial defects are, therefore, the probable reason thatMnSOD overexpression shortens replicative life span. © 2003 Elsevier Inc.

Keywords—Yeast, Aging, Superoxide dismutase, Replicative senescence, Chronological life span, Sir2p histonedeacetylase, Free radicals

INTRODUCTION

Studies in model organisms indicate that aging can beslowed with the increased scavenging of reactive ox-ygen species (ROS) that accompanies the overexpres-sion of antioxidant enzymes.Drosophila shows anincreased life span when the levels of the cytosoliccopper,zinc superoxide dismutase (Cu,ZnSOD) or themitochondrial manganese SOD (MnSOD) are in-creased inadult flies [1–3]. It also lives longer withthe expression of human Cu,ZnSOD in its motorneu-rons [4]. Long-livedDrosophila lines, obtained by theselective breeding of individuals with long life span,

as well as long-lived mutants of the nematodeCaeno-rhabditis elegans obtained by direct selection, displayelevated stress resistance [5–7]. This is consistent withnatural selection operating to provide the levels ofprotective activities that will optimize survival of anorganism under the conditions defined by rates of extrin-sic mortality, not the higher stress resistances that wouldmaximize life spans [8]. High antioxidant defense mustcounteract aging by providing an increased protectionagainst oxidative damage. Often oxidant-scavenging en-zymes, and enzymes catalyzing the reduction of oxidizedthiols in protective molecules like glutathione and thethioredoxins, are stress inducible [9]. Therefore, an in-creased stress resistance may both reduce the amount ofdamage inflicted by ROS and increase the levels of repairactivities, with commensurate reduction in the age-re-lated accumulation of damaged cell components.

Address correspondence to: Dr. Peter W. Piper, University CollegeLondon, Department of Biochemistry and Molecular Biology, GowerStreet, London WC1E 6BT, UK; Tel:�44 (20) 7679 2212; Fax:�44(20) 7679 7193; E-Mail: piper@bsm.bioc.ucl.ac.uk.

Free Radical Biology & Medicine, Vol. 34, No. 12, pp. 1599–1606, 2003Copyright © 2003 Elsevier Inc.

Printed in the USA. All rights reserved0891-5849/03/$–see front matter

doi:10.1016/S0891-5849(03)00210-7

1599

Yeast can provide a useful model for investigating theinterplay among stress resistance, levels of damaged cellcomponents, and aging. Moreover, it is possible to in-vestigate the effects of any discrete genetic change ontwo life spans: (i) the replicative (budding, nonchrono-logical) life span, measured as the number of daughtersproduced by each actively dividing mother cell [10]; and(ii) the chronological life span, measured as the ability ofstationary (G0-arrested) cultures to maintain viabilityover time [11–14]. The increases in chronological lifespan of Drosophila and C. elegans with elevations tostress resistance (discussed above) represent effects thatare mainly exerted on the postmitotic cells and tissues ofthe adult. Thus, it may be the chronological life span ofG0-arrested yeast, rather than the yeast replicative lifespan, that most closely resembles these aging processesin Drosophila and C. elegans.

In this study, we constructed a yeast strain that over-expresses the mitochondrial MnSOD. This was, in part,to investigate the effects on these two life spans ofoverexpressing the enzyme thought to be the key protec-tion against the superoxide generated by the respiratorychain [9,15]. It was also to provide a model for investi-gating the more detrimental effects when MnSOD isoverexpressed in higher eukaryotes [16,17].

MATERIALS AND METHODS

Strains and media

Saccharomyces cerevisiae strains used are listed inTable 1. Cells were grown aerobically at 30°C in liquidYP medium [1% (w/v) Difco yeast extract, 2% Bactopeptone, 20 mg/l adenine], containing as carbon sourceeither 2% glucose (YPD) or 3% glycerol (YPGlycerol).

Strain constructions

A 0.675 kb EcoR1-Xba1 fragment, containing theADH2 promoter and multiple cloning site of pWYG2L[18], was ligated into EcoR1 plus Xba1-cleaved pRS403and pRS406 [19] so as to give pRS403(ADH2) andpRS406(ADH2), respectively. The SOD2 coding regionswere then PCR amplified from yeast genomic DNAusing primers flanked with BamH1 or Not1 restrictionsites (sequences available on request). These PCR prod-ucts were then digested with BamH1 and Not1, prior toligation to BamH1/Not1-cleaved pRS403(ADH2) andpRS406(ADH2). This yielded pRS403(ADH2-SOD2)and pRS406(ADH2-SOD2). These vectors and controlpRS403/6 were cleaved at their unique Pst1 site prior tointegrative transformation into S. cerevisiae FY1679-28c, selecting for histidine prototrophy (pRS403-derivedplasmids) or uracil prototrophy (pRS406-derived plas-mids). sod1�, sod2�, and ctt1� mutants were preparedusing standard kanMX4 cassette gene knockout technol-ogy [20].

Life span analyses

Chronological life spans were determined as de-scribed earlier [13,14], using cells grown to early sta-tionary phase on YPGlycerol and then transferred towater. At the indicated times during subsequent 30°C or37°C aerobic maintenance, three serial dilutions wereprepared; these were plated and the viable cells deter-mined by colony counting after 5 d growth at 30°C. Atleast 103 cells from each dilution were analyzed, the datashowing the mean and SD of the three separate serialdilutions. Replicative life spans were measured by stan-dard procedures of counting the number of buds pro-duced by 60 virgin cells; these buds were removed asthey were formed by micromanipulation [10,21].

Protein extract preparation, SOD, and catalase assay

Yeast extracts were prepared in 50 mM potassiumphosphate buffer (pH 7.0) containing a cocktail of pro-tease inhibitors, and protein concentrations were deter-mined using the Bio-Rad protein determination kit (Bio-Rad, Richmond, CA, USA) and bovine serum albumin asstandard. Catalase activity was determined using themethod described elsewhere [22]. SOD activity wasmeasured either from its ability to inhibit the reduction ofcytochrome c by xanthine oxidase or its ability to inhibitreduction of nitro blue tetrazolium to formazan in gels[23]. Distinction of MnSOD from Cu,ZnSOD was based,in the former assay, on the selective capability of 2 mMcyanide to inhibit Cu,ZnSOD [23] and, in the latterassay, on the different gel migrations of MnSOD andCu,ZnSOD.

Table 1. Yeast Strains Used in This Study

Strain Genotype Source

FY1679-28c MATa his3-�200, ura3-52,leu2-�1, trp1-�63

[54]

FYsod1� FY1679-28c sod1�kanMX4 [54]FYsod2� FY1679-28c sod2�kanMX4 [54]FYctt1� FY1679-28c ctt1�kanMX4 This studyFY wild type FY1679-28c HIS3:: [pRS403],

URA3::[pRS406]This study

sir2 FY1679-28c HIS3::[pRS403],URA3::[pRS406], sir2::kanMX4

This study

SOD1 FY1679-28c HIS3::[pRS403],URA3::[pRS406(ADH2-SOD1)]

This study

SOD2 FY1679-28c HIS3::[pRS403],URA3::[pRS406(ADH2-SOD2)]

This study

CTT1 FY1679-28c HIS3::[pRS403(ADH2-CTT1)], URA3::[pRS406]

This study

SOD2 sir2 FY1679-28c HIS3::[pRS403],URA3::[pRS406(ADH2-SOD2)],sir2::kanMX4

This study

1600 N. HARRIS et al.

Isolation of mtGFP-labeled old mother cells

Strains were transformed with a LEU2 vector for TPIpromoter-driven mtGFP expression (pX232 [24]). Thesetransformants were then biotinylated, grown five gener-ations on minus leucine dropout medium [25], afterwhich the biotinylated old mother cells were reisolatedusing Dynal M-280 magnetic beads (Dynal BiotechASA, Oslo, Norway) [10]. They were fixed, their budscars counted (after calcoflour white staining), and theirmitochondria examined by fluorescence microscopy.

RESULTS

Construction of MnSOD and catalase T-overexpressingyeast strains

SODs, catalases, and peroxidases constitute the firstline of antioxidant defense in all aerobic cells [26]. S.cerevisiae possesses both a Cu,ZnSOD (SOD1p), abun-dant in the cytosol and in the intermembrane space of themitochondrion, as well as an MnSOD (SOD2p) that isactive only in the lumen of the mitochondrion[9,15,27,28]. S. cerevisiae also has two catalases, thoughof these the cytosolic catalase T encoded by the CTT1gene is generally the most important (the peroxisomalcatalase A encoded by CTA1 is only induced on carbonsources whose assimilation requires the function of theperoxisome [9,29]).

To determine the effects of increasing the SOD andcytosolic catalase activities of S. cerevisiae, we con-structed a set of isogenic strains with additional genes foroxidant-scavenging enzymes integrated into their ge-nomes (see Materials and Methods and Table 1). Theseadditional genes for the overexpression of Cu,ZnSOD(SOD1), MnSOD (SOD2), and cytoplasmic catalase(CTT1) were placed under the control of the ADH2

promoter since this is a promoter strongly derepressed inrespiratory cells, the cultures that display the longestchronological life spans [13,14].

Consistent with the ADH2 promoter being active onlyin respiratory cells, the ADH2-SOD2 and ADH2-CTT1cassettes caused no increased MnSOD or catalase activ-ity, respectively, during growth on glucose. Increases (2-to 3-fold) in these activities, though, were apparent at thediauxic shift that occurs in response to glucose exhaus-tion (data not shown). Moreover, in cultures efficientlyadapted to respiratory growth on glycerol, these sametwo overexpression cassettes generated 7- or 3-fold in-creases, respectively, in the activities of MnSOD andcatalase (Figs. 1A and 1C). It is improbable that any ofthis overexpressed MnSOD is active in the cytosol sincethe formation of the active MnSOD tetramer requiresmitochondrial import of the MnSOD precursor togetherwith removal of the N-terminal leader sequence from thisprecursor [28]. The ADH2-SOD2 overexpression cas-sette also unexpectedly increased the cytosolic Cu,Zn-SOD and catalase activities of glycerol-grown cells(Figs. 1B and 1C). This may reflect a stress induction ofthe chromosomal genes for Cu,ZnSOD (SOD1) and cata-lase T (CTT1) [9], possibly as a consequence of in-creased H2O2 production in mitochondria with enhancedMnSOD activity. A mitochondrial peroxidase (Cct1p) isa vital component of the sensing of H2O2 stress in yeast[30].

In contrast to these results with the MnSOD overex-pression cassette (ADH2-SOD2), the cassette generatedfor Cu,ZnSOD overexpression (ADH2-SOD1) did notincrease the Cu,ZnSOD activity of glycerol-grown cellsin our initial experiments (Fig. 1B), even though itspresence generated increased catalase activity (Fig. 1C).However, this ADH2-SOD1 cassette was causing in-creased levels of the Cu,ZnSOD apoprotein and, when

Fig. 1. MnSOD (a), Cu,ZnSOD (b), and catalase (c) activities assayed in total protein extracts from midexponential YPGlycerolcultures of the FY wild-type (WT), SOD1, SOD2, CTT1, and ctt1� mutant strains (see also Table 1).

1601MnSOD overexpression and yeast aging

integrated into the sod1� mutant, it could revert the poorrespiratory growth of the latter mutant [31]. Others alsohave experienced problems when attempting to overex-press active Cu,ZnSOD in yeast [32]. We have sincefound that this ADH2-SOD1 cassette can generate anelevated Cu,ZnSOD activity when the respiratory growthmedium is supplemented with a high level of copper orwhen the cells are engineered to simultaneously overex-press the chaperone dedicated to loading the Cu,ZnSODapoprotein with Cu2� ions, Ccs1p [33]. These variouseffects of overexpressing the SOD1p apoprotein, both inthe absence and presence of a simultaneous Ccs1p over-expression, will be described in a later communication(manuscript in preparation).

MnSOD overproduction lengthens the chronologicallife span of yeast cells optimized for G0 maintenancebut leads to dramatic Sir2p-independent shortening ofthe replicative life span

Maximization of the S. cerevisiae chronological lifespan requires adaptation to efficient respiratory mainte-nance, achieved by prior growth of the yeast to earlystationary (G0) phase on respiratory medium [13,14].Cultures prepared this way can be maintained undernormoxic conditions for up to a month at 30°C withoutappreciable viability loss [13]. Often, though, we mea-sure their survival at more stressful temperatures (35°Cor 37°C) so that we can conduct the survival measure-ments over a time scale of days rather than weeks [13,14](Fig. 2). As with aging in Drosophila [34], survival ismuch shorter at the higher temperature [13]. MnSODoverexpression increased the chronological life span ofthese cells optimized for stationary phase survival, bothduring maintenance at 30°C and at 37°C (Fig. 2). Thiscontrasts with the effects of cytosolic catalase overex-pression (the ADH2-CTT1 cassette), which exerted onlysmall or neutral effects on this survival (Fig. 2).

Next we determined the effects of the MnSOD over-expression cassette (ADH2-SOD2) on the yeast replica-tive life span. Mutants that lack SOD or catalase en-zymes are sensitized to oxidative stress and display shortreplicative life spans [35–37]. We found that the MnSODoverexpression due to ADH2-SOD2 cassette expressionin glycerol-grown cells also causes a shortened replica-tive life span, whereas an overexpression of the cytosoliccatalase (the ADH2-CTT1 cassette) exerted only smalleffects on this life span (Fig. 3).

We investigated this dramatic shortening of the rep-licative life span with MnSOD overexpression in greaterdetail. The life span reduction was not apparent when thelife spans were determined on glucose (not shown), acarbon source that does not induce high levels of expres-sion of the ADH2-SOD2 cassette. It is also unlikely that

the shortened replicative life span represents overexpres-sion of the MnSOD metalloenzyme causing the cells tosuffer severe manganese insufficiency, since this over-expression has no effects on vegetative growth rates.

A major determinant of S. cerevisiae replicative se-nescence is genetic instability, specifically homologousrecombination within the 120 tandemly repeated copies

Fig. 2. Chronological life spans of the FY wild-type (WT), SOD2, andCTT1 strains. Cells were grown on YPGlycerol at 30°C to earlystationary phase, transferred to water at time zero, and their viabilitywas determined at the indicated times of 30 or 37°C aerobic mainte-nance.

Fig. 3. Replicative life spans (30°C, YPGlycerol) of the FY wild-type,SOD2, sir2, and SOD2, sir2 strains.

1602 N. HARRIS et al.

of the ribosomal RNA-encoding DNA (rDNA) on chro-mosome XII, and the subsequent proliferation of theextrachromosomal rDNA circles (ERCs) that are gener-ated through this recombination [10,38]. ERCs replicateat each cell division but preferentially stay within themother cell at cytokinesis. Thus, they accumulate to veryhigh levels within this old mother just a few generationsafter the initial ERC excision event. This process causesthe mother cell-specific aging that is so characteristic ofyeast senescence, the cessation of budding being pre-ceded by a loss of transcriptional silencing and by ste-rility in the old mother cell [39]. Recombination withinthe rDNA is suppressed through the activity of the Sir2pNAD�-dependent histone deacetylase and, therefore, iscritically sensitive to the redox state of the cell [40,41].An absence of this activity in the sir2 mutant leads toincreased recombination within rDNA and increasedrates of ERC excision. This, in turn, shortens the repli-cative life span [42].

To determine whether Sir2p is also an important in-fluence over the life span reduction that we had observedfor the SOD2 MnSOD overexpressor (Fig. 3), we deletedthe SIR2 gene in both this strain and the isogenic FY wildtype (see Materials and Methods and Table 1). Replica-tive life span determinations for these sir2� mutant de-rivatives (Fig. 3) revealed the reduction in life span withMnSOD overexpression to be completely unaffected bythe absence of Sir2p (compare SOD2 and SOD2, sir2;Fig. 3). This life span reduction was also slightly greaterthan that generated through the loss of Sir2p (comparewild-type, sir2 and SOD2 cells; Fig. 3). Hence, theshortening of replicative life span by MnSOD overex-pression occurs by a mechanism totally independent ofthe Sir2p NAD�-dependent histone deacetylase, amechanism that is probably not dependent on the initialERC excision event [40–42].

MnSOD overproduction is associated with aberrantmitochondrial morphology and segregation in oldmother cells

Upon finding that the shortening of replicative lifespan with MnSOD overexpression is independent ofSir2p histone deacetylase (Fig. 3), we sought an alterna-tive explanation for this life span reduction. We haveshown recently that the yeast replicative life span can beshortened by mutations that compromise efficient mito-chondrial inheritance [21,43]. So, we transformed theMnSOD overexpressor and wild type (Table 1) with avector for the expression of a mitochondrially targetedgreen fluorescent protein (mtGFP, see Materials andMethods). Expressing mtGFP in yeast facilitates the vi-sualization of mitochondria by fluorescence microscopyand, therefore, the detection of aberrant mitochondrial

morphology and segregation [24]. We next used a bioti-nylation and magnetic bead procedure [44] to preparemtGFP-labeled old mother cells of the wild type andSOD2 overexpressor after these had budded for fivegenerations (see Materials and Methods). We found that,after five generations, 35% of the old mother cells of theSOD2 strain divide no more, while almost all the oldmother cells of the wild type still bud actively (Fig. 3).

Yeast mitochondria are generally distributed aroundthe periphery of the cell, with a spherical or tubular,sometimes snake-like morphology (see wild-type cells,Fig. 4). The mitochondria in the young (unsorted) cellsof the SOD2 overexpressor appeared generally less tu-bular and more ovoid than those of the young, wild-typecontrol cells (Fig. 4). However, it was in the mitochon-dria of old mother cells that the major effects of MnSODoverexpression were apparent. A large fraction of theSOD2 strain, five generation old mothers (60% of thoseexamined), exhibited a total loss of any discrete mtGFP-stained organelles, these appearing to have segregatedtotally to the daughter (Fig. 4). Since the presence of amitochondrion is essential for viability [45], this loss ofmitochondria ensures the demise of these mother cells. Inother old SOD2 cells (30%), it was the attached daughterthat totally lacked any mtGFP staining (Fig. 4), an indi-cation that mitochondria had failed to segregate to thisdaughter. Since the completion of cytokinesis requiresthe completion of mitochondrial inheritance [46,47], thisapparent failure of mitochondria to segregate may alsocontribute to the shortened life span. A much smallerfraction of the old mother SOD2 overexpressor cells(10%) displayed a massive proliferation of mitochondria(Fig. 4, lower right-hand image).

These observations of the old cells of the MnSODoverexpressor contrasted with those of the control, fivegeneration old mother cells prepared from the wild-typestrain. The mtGFP flourescence of the latter indicatedmitochondrial segregation to the daughter was essen-tially similar in these old cells as in young (unsorted)cells (Fig. 4). Typically, the mtGFP staining of both theyoung (unsorted) and five generation old (sorted) cellsfrom the FY wild type revealed the mitochondria migrat-ing into buds when these buds were still very small (Fig.4, left-hand images).

These pronounced differences in the mtGFP stainingof old mother, SOD2 overexpressor, and control wild-type cells indicate that the massive overproduction ofMnSOD in the mitochondrial matrix of the former hascaused defects in the segregation of mitochondria fromthe mother to the daughter, most apparent in the oldmother cells (Fig. 4). This defective mitochondrial seg-regation is, in turn, a plausible explanation for the short-ened replicative life span (Fig. 3).

1603MnSOD overexpression and yeast aging

DISCUSSION

Our earlier work focused on identifying the physio-logical conditions that maximize the yeast chronologicallife span, since we considered the maximization of thislife span to be a desirable prerequisite for the use of yeastas a model of the chronological aging of the nondividingcells and tissues in higher organisms [13,14]. As shownhere, elevated MnSOD activity increases longevity insuch cells optimized for G0 survival (Fig. 2). Losses ofSOD activities markedly decrease the stationary survivalof Escherichia coli [48]. Losses of oxidant-scavengingenzymes also decrease the stationary survival of yeastpregrown on glucose [49,50], while increasing theCu,ZnSOD in the intermembrane space of the mitochon-

drion or the loss of the Sch9 protein kinase generatesmall or larger increases in this survival, respectively[12,15]. These earlier demonstrations of an increasedyeast chronological life span were all for cells pregrownto stationary phase on glucose, cultures that display areduced stationary survival compared to respiratory cul-tures [13,14]. In contrast, this study focused exclusivelyon respiratory cultures, on the premise that aging studiesare more firmly grounded when they address the longestlife spans for the organism under study. This is, there-fore, the first time that an increased chronological lifespan has been shown for yeast adapted to efficient respi-ratory maintenance, cells optimized for G0 survival[13,14]. It shows that oxidative stress limits the maximal

Fig. 4. Phase and mtGFP-stained images of young (unsorted) and five generation old (sorted) cells from exponentially growingYPGlycerol cultures of the FY wild-type and SOD2 overexpressor strains. The mitochondria of wild-type old mothers retain amorphology and inheritance similar to that seen in young cells. In contrast, in old cells of the SOD2 overexpressor the mitochondriaare largely lost completely to the daughter (third and fourth from top right-hand images), not inherited by the daughter (second frombottom right-hand image), or seem to undergo a massive proliferation (bottom right-hand image). The phase-dark, spherical particlesof uniform size and fluorescence in the old mother cell images are all magnetic beads used in the isolation of these cells.

1604 N. HARRIS et al.

chronological life span of nondividing yeast. Oxidativestress is also a major factor in the chronological aging ofadult Drosophila [1–3] and C. elegans [7,51], whereagain its effects are exerted primarily on nondividingcells and tissues.

MnSOD, rather than Cu,ZnSOD, is thought to be themajor protection against the superoxide generated by theyeast respiratory chain [9,15]. However, while overex-pression of MnSOD increased the chronological life span(Fig. 2), it dramatically shortened the replicative poten-tial of the cells (Fig. 3). This reduction in replicative lifespan occurred independently of the Sir2p NAD�-depen-dent histone deacetylase (Fig. 3). Probably the reductionis not mediated through an effect on rates of recombina-tion within the rDNA; it appears instead to be the resultof mitochondrial defects in old mother cells (Fig. 4).

Figures 3 and 4 essentially indicate a trade-off be-tween the beneficial and detrimental effects of MnSODoverexpression, such that this overexpression is benefi-cial for the survival maintenance of nondividing cells(Fig. 2) but detrimental for replicative potential (Figs. 3and 4). These findings have remarkable parallels withthose in more complex systems, where again the detri-mental influences of overexpressing SOD activities ap-pear to relate mainly to dividing cells and tissues. Ad-verse effects of MnSOD overexpression have beendescribed in cultured NIH-3T3 [16] and rat glioma cells[17]. It appears, though, that MnSOD overexpressionalso may be capable of exerting protective effects innondividing cells and tissues, as it slows the aging ofadult Drosophila [1,2,4] and protects against drug-in-duced cardiotoxicity in transgenic mice [52]. It has yet tobe shown that increased SOD activity alone can slowaging in mammals, though an increased general resis-tance to oxidative stress is associated with increasedlongevity in mice [53].

Acknowledgements — We are indebted to M. Romanos and B. West-ermann for gifts of plasmids. This project was supported by a BBSRCstudentship (to N. H.).

REFERENCES

[1] Sun, J.; Tower, J. FLP recombinase-mediated induction of Cu,Znsuperoxide dismutase transgene expression can extend the lifespan of adult Drosophila melanogaster flies. Mol. Cell. Biol.19:216–228; 1999.

[2] Tower, J. Transgenic methods for increasing Drosophila life span.Mech. Ageing Dev. 118:1–14; 2000.

[3] Sun, J.; Folk, D.; Bradley, T. J.; Tower, J. Induced overexpressionof mitochondrial Mn superoxide dismutase extends the life spanof adult Drosophila melanogaster. Genetics 161:661–672; 2002.

[4] Parkes, T. L.; Elia, A. J.; Dickinson, D.; Hilliker, A. J.; Phillips,J. P.; Boulianne, G. L. Extension of Drosophila life span byoverexpression of human SOD1 in motorneurons. Nature Gen.19:171–174; 1998.

[5] Service, P. M.; Hutchinson, E. W.; MacKinley, M. D.; Rose,M. R. Resistance to environmental stress in Drosophila melano-

gaster selected for postponed senescence. Physiol. Zool. 58:380–389; 1985.

[6] Lithgow, G. J. Invertebrate gerontology: the age mutations ofCaenorhabditis elegans. Bioessays 18:809–815; 1996.

[7] Taub, J.; Lau, J. F.; Ma, C.; Yang, H. H.; Hoque, R.; Rothblatt, J.;Chalfie, M. A cytosolic catalase is needed to extend adult life spanin C. elegans daf-C and clk-1 mutants. Nature 399:162–166;1999.

[8] Kirkwood, T. B. L. Evolution of ageing. Mech. Ageing Dev.123:737–745; 2002.

[9] Moradas-Ferreira, P.; Costa, V. Adaptive response of the yeastSaccharomyces cerevisiae to reactive oxygen species: defenses,damage, and death. Redox Rep. 5:277–285; 2000.

[10] Sinclair, D. A.; Guarente, L. Extrachromosomal rDNA circles: acause of aging in yeast. Cell 91:1033–1042; 1997.

[11] Muller, I.; Zimmerman, M.; Becker, D.; Flouner, M. Calender lifespan versus budding life span of Saccharomyces cerevisiae.Mech. Ageing Dev. 12:47–52; 1980.

[12] Fabrizio, P.; Fabiola, P.; Pletcher, S. D.; Gendron, C. M.; Longo,V. D. Regulation of longevity and stress resistance by Sch9 inyeast. Science 292:288–290; 2001.

[13] MacLean, M.; Harris, N.; Piper, P. W. Chronological life span ofstationary phase yeast cells: a model for investigating the factorsthat might influence the ageing of postmitotic tissues in higherorganisms. Yeast 18:499–509; 2001.

[14] MacLean, M. J.; Aamodt, R.; Harris, N. L.; Alseth, I.; Seeberg,E.; Bjørås, M.; Piper, P. W. Base excision repair activities neededfor yeast to attain its full chronological life span. Aging Cell2:93–104; 2003.

[15] Sturtz, L. A.; Diekert, K.; Jensen, L. T.; Lill, R.; Culotta, V. C. Afraction of the yeast Cu,Zn superoxide dismutase and its metal-lochaperone, CCS, localize to the intermembrane space of mito-chondria: a physiological role for SOD1 in guarding againstmitochondrial oxidative damage. J. Biol. Chem. 276:38084–38089; 2001.

[16] Li, N.; Oberley, T. D.; Oberley, L. W.; Zhong, W. Inhibition ofcell growth in NIH/3T3 fibroblasts by overexpression of manga-nese superoxide dismutase: mechanistic studies. J. Cell Physiol.175:359–369; 1998.

[17] Zhong, W.; Oberley, L. W.; Oberley, T. D.; Yan, T.; Domann,F. E.; St. Clair, D. K. Inhibition of cell growth and sensitizationto oxidative damage by overexpression of manganese superoxidedismutase in rat glioma cells. Cell Growth Differ. 7:1175–1186;1996.

[18] Johnston, J. R Molecular genetics of yeast: a practical approach.Oxford, England: I.R.L. Press; 1994.

[19] Sikorski, R. S.; Hieter, P. A system of shuttle vectors and yeasthost strains designed for efficient manipulation of DNA in Sac-charomyces cerevisiae. Genetics 122:19–27; 1989.

[20] Wach, A. PCR-synthesis of marker cassettes with long flankinghomology regions for gene disruptions in S. cerevisiae. Yeast12:259–265; 1996.

[21] Piper, P. W.; Bringloe, D. Loss of prohibitins, though it shortensthe replicative life span of yeast cells undergoing division, doesnot shorten the chronological life span of G0-arrested cells. Mech.Ageing Dev. 123:287–295; 2002.

[22] Aebi, H. Catalase in vitro. Methods Enzymol. 105:121–126; 1984.[23] Flohe, L.; Otting, F. Superoxide dismutase assays. Methods En-

zymol. 105:93–104; 1984.[24] Westermann, B.; Neupert, W. Mitochondria-targeted green fluo-

rescent proteins: convenient tools for the study of organelle bio-genesis in Saccharomyces cerevisiae. Yeast 16:1421–1427; 2000.

[25] Adams, A.; Gottschling, D. E.; Kaiser, C. A.; Stearns, T. Methodsin yeast genetics. Cold Spring Harbor, New York: Cold SpringHarbor Laboratory Press; 1997.

[26] Halliwell, B.; Gutteridge, J. M. C. Free radicals in biology andmedicine, third edition. Oxford, England: Oxford UniversityPress; 1998.

[27] Westerbeek-Marres, C. A.; Moore, M. M.; Autor, A. Regulationof manganese superoxide dismutase in Saccharomyces cerevisiae.

1605MnSOD overexpression and yeast aging

The role of respiratory chain activity. Eur. J. Biochem. 174:611–620; 1988.

[28] Gralla, E. B.; Kosman, D. J. Molecular genetics of superoxidedismutases in yeasts and related fungi. Adv. Genet. 30:251–319;1992.

[29] Ruis, H.; Schuller, C. Stress signaling in yeast. Bioessays 17:959–965; 1995.

[30] Charizanis, C.; Juhnke, H.; Krems, B.; Entian, K.-D. The mito-chondrial cytochome c peroxidase Ccp1 of Saccharomyces cer-evisiae is involved in conveying an oxidative stress signal to thetranscription factor Pos9 (Skn7). Mol. Gen. Genet. 262:437–447;1999.

[31] Gralla, E. B.; Valentine, J. S. Null mutants of Saccharomycescerevisiae Cu,Zn superoxide dismutase: characterization andspontaneous mutation rates. J. Bacteriol. 173:5918–5920; 1991.

[32] Stenlund, P.; Tibell, L. A. Chimeras of human extracellular andintracellular superoxide dismutase. Analysis of structure andfunction of the individual domains. Protein Eng. 12:319–325;1999.

[33] Culotta, V. C.; Klomp, L. W.; Strain, J.; Casareno, R. L.; Krems,B.; Gitlin, J. D. The copper chaperone for superoxide dismutase.J. Biol. Chem. 272:23469–23472; 1997.

[34] Zwaan, B. J.; Bijlsma, R.; Hoekstra, R. F. On the developmentaltheory of ageing. I. Starvation resistance and longevity in Dro-sophila melanogaster in relation to pre-adult breeding conditions.Heredity 66:29–39; 1991.

[35] Barker, M. G.; Brimage, L. J. E.; Smart, K. A. Effect of Cu, Znsuperoxide dismutase disruption mutation on replicative senes-cence in Saccharomyces cerevisiae. FEMS Microbiol. Lett. 177:199–204; 1999.

[36] Wawryn, J.; Krzepilko, A.; Mytszka, A.; Bilinski, T. Deficiencyin superoxide dismutases shortens life span of yeast cells. ActaBiochim. Pol. 46:249–253; 1999.

[37] Nestelbacher, R.; Laun, P.; Vondrakonva, D.; Pichova, D.; Schul-ler, C.; Breitenbach, M. The influence of oxygen toxicity on yeastmother cell-specific aging. Exp. Gerontol. 35:63–70; 2000.

[38] Defossez, P.-A.; Prusty, R.; Kaeberlein, M.; Lin, S.-J.; Ferrigno,P.; Silver, P. A.; Keil, R. L.; Guarente, L. Elimination of repli-cation block protein Fob1 extends the life span of yeast mothercells. Mol. Cell. 3:447–455; 1999.

[39] Smeal, T.; Guarente, L. Mechanisms of cellular senescence. Curr.Opin. Gen. Dev. 7:281–287; 1997.

[40] Lin, S. J.; Kaeberlein, M.; Andalis, A. A.; Sturtz, L. A.; Defossez,P. A.; Culotta, V. C.; Fink, G. R.; Guarente, L. Calorie restrictionextends Saccharomyces cerevisiae life span by increasing respi-ration. Nature 418:344–348; 2002.

[41] Kaeberlein, M.; Andalis, A. A.; Fink, G. R.; Guarente, L. Highosmolarity extends life span in Saccharomyces cerevisiae by amechanism related to calorie restriction. Mol. Cell. Biol. 22:8056–8066; 2002.

[42] Lin, S.-J.; Defossez, P.-A.; Guarente, L. Requirement of NADand SIR2 for life-span extension by calorie restriction in Saccha-romyces cerevisiae. Science 289:2126–2128; 2000.

[43] Piper, P. W.; Jones, G. W.; Bringloe, D.; Harris, N.; MacLean,M.; Mollapour, M. The shortened replicative life span of prohib-itin mutants of yeast appears to be due to defective mitochondrialsegregation in old mother cells. Aging Cell 1:149–157; 2002.

[44] Sinclair, D. A.; Mills, K.; Guarente, L. Accelerated aging andnucleolar fragmentation in the yeast sgs1 mutant. Science 277:1313–1316; 1997.

[45] Lill, R.; Kispal, G. Maturation of cellular Fe-S proteins: anessential function of mitochondria. Trends Biochem. Sci. 25:352–356; 2000.

[46] Yaffe, M. P. The machinery of mitochondrial inheritance andbehavior. Science 283:1493–1497; 1999.

[47] Catlett, N. L.; Weisman, L. S. Divide and multiply: organellepartitioning in yeast. Curr. Opin. Cell Biol. 12:509–516; 2000.

[48] Benov, L.; Fridovich, I. A superoxide dismutase mimic protectssodA sodB Escherichia coli against aerobic heating and station-ary-phase death. Arch. Biochem. Biophys. 322:291–294; 1995.

[49] Longo, V. D.; Gralla, E. B.; Valentine, J. S. Superoxide dismutaseis essential for stationary phase survival in S. cerevisiae: mito-chondrial production of toxic oxygen species in vivo. J. Biol.Chem. 271:12275–12280; 1996.

[50] Longo, V. D.; Liou, L. L.; Valentine, J. S.; Gralla, E. B. Loss ofmitochondrial superoxide dismutase decreases yeast survival instationary phase. Arch. Biochem. Biophys. 365:131–142; 1999.

[51] Melov, S.; Ravenscroft, J.; Malik, S.; Gill, M. S.; Walker, D. W.;Clayton, P. E.; Wallace, D. C.; Malfroy, B.; Doctrow, S. R.;Lithgow, G. J. Extension of life span with superoxide dismutase/catalase mimetics. Science 289:1567–1569; 2000.

[52] Yen, H. C.; Oberley, T. D.; Vichitbandha, S.; Ho, Y. S.; St. Clair,D. K. The protective role of manganese superoxide dismutaseagainst adriamycin-induced acute cardiac toxicity in transgenicmice. J. Clin. Invest. 98:1253–1260; 1996.

[53] Migliaccio, E.; Giorgio, M.; Mele, S.; Pelicci, G.; Reboldi, P.;Pandolfi, P. P.; Lanfrancone, L.; Pelicci, P. G. The p66shc adaptorprotein controls oxidative stress response and life span in mam-mals. Nature 402:309–313; 1999.

[54] Piper, P. W. Yeast superoxide dismutase mutants reveal a pro-oxidant action of weak organic acid food preservatives. FreeRadic. Biol. Med. 27:1219–1227; 1999.

1606 N. HARRIS et al.