non-homologous end joining dependency of γ-irradiation-induced adaptive frameshift mutation...

Mutation Research 556 (2004) 201–208

Non-homologous end joining dependency of�-irradiation-inducedadaptive frameshift mutation formation in cell cycle-arrested

yeast cells

Erich Heidenreich∗, Herfried Eisler

Institute of Cancer Research, Division of Molecular Genetics, Medical University of Vienna, Borschkegasse 8a, A-1090 Vienna, Austria

Received 26 May 2004; received in revised form 11 August 2004; accepted 18 August 2004

Abstract

There is a strong selective pressure favoring adaptive mutations which relieve proliferation-limiting adverse living conditions.Due to their importance for evolution and pathogenesis, we are interested in the mechanisms responsible for the formation ofsuch adaptive, gain-of-fitness mutations in stationary-phase cells. During previous studies on the occurrence of spontaneousreversions of an auxotrophy-causing frameshift allele in the yeastSaccharomyces cerevisiae, we noticed that about 50% of theadaptive reversions depended on a functional non-homologous end joining (NHEJ) pathway of DNA double-strand break (DSB)repair. Here, we show that the occasional NHEJ component Pol4, which is the yeast ortholog of mammalian DNA polymerasel tici -deficients thway.T ion-inducedD n resting(©

K

f

(

areses

gingcated, the

0

ambda, is not required for adaptive mutagenesis. An artificially imposed excess of DSBs by�-irradiation resulted in a dramancrease in the incidence of adaptive, cell cycle arrest-releasing frameshift reversions. By the use of DNA ligase IVtrains we detected that the majority of the�-induced adaptive mutations were also dependent on a functional NHEJ pahis suggests that the same mutagenic NHEJ mechanism acts on spontaneously arising as well as on ionizing radiatSBs. Inaccuracy of the NHEJ repair pathway may extensively contribute to the incidence of frameshift mutations i

non-dividing) eukaryotic cells, and thus act as a driving force in tumor development.2004 Elsevier B.V. All rights reserved.

eywords:Selection-induced; Starvation-induced; Stationary phase; Replication-independent; Dnl4

∗ Corresponding author. Tel.: +43 1 4277 65212;ax: +43 1 4277 9651.

E-mail address:[email protected]. Heidenreich).

1. Introduction

Spontaneous mutations arising in resting cellslikely to contribute to diverse fundamental processuch as the evolution of microorganisms, cellular aand cancerogenesis. Whereas the mutations impliin cellular aging are detrimental to the affected cells

027-5107/$ – see front matter © 2004 Elsevier B.V. All rights reserved.doi:10.1016/j.mrfmmm.2004.08.006

202 E. Heidenreich, H. Eisler / Mutation Research 556 (2004) 201–208

type of mutations discussed as relevant for the evolutionof microorganisms and cancer of mammalia results ina proliferation advantage for the individual variant cell.

The mechanisms responsible for formation of thelatter type of mutations (termed “adaptive” mutationson the basis of their immediate proliferation advantagefor the affected cells) have so far been studied mainlyin the model organismsEscherichia coli(reviewed byRosenberg and Foster[1,2]) andSaccharomyces cere-visiae. During our studies on adaptive reversions of aLYS2frameshift allele in cell cycle-arrestedS. cere-visiaecells we observed that such reversions occurredpreferentially by simple deletions in mononucleotiderepeats and thus resembled microsatellite instability[3]. In search for the mechanisms behind adaptive mu-tagenesis in yeast we and others noticed that pathwayswhich counteract adaptive mutation and whose im-pairment therefore favors adaptive mutations are read-ily identifiable. The frequencies of adaptive mutationswere increased by deletions of mismatch repair genes[4], deletion of the double-strand break (DSB) repairgeneRAD52 [5], deletion of the nucleotide excisionrepair (NER) genesRAD14, RAD16or RAD26[6], orby elimination of the proofreading activities of DNApolymerase� or � [7].

In contrast, it proved rather cumbersome to identifygenes whose deletions decrease the frequency of adap-tive mutations and whose coded-for proteins conse-quently contribute to the production of mutations. Theerror-prone translesion synthesis (TLS) polymerase�t ptiver ak velo tedw

ath-w p-t ing( heo IVaT air ofs onet es-p

te-i ro-k

DSB end-recognizing and -binding functions of theYku70/Yku80 heterodimer[10], followed by the end-bridging function of the Rad50/Mre11/Xrs2 proteincomplex[11], and finalized by joining of the ends byDNA ligase IV (Dnl4) with its cofactor Lif1. If the endsare not readily connectable due to chemical modifi-cations or sequence incompatibilities, end-processinghas to precede and enable ligation. A thorough under-standing of such processing steps remains to be estab-lished, but Wilson and Lieber[12] reported that theDNA polymerase encoded byPOL4 is required forjoining of aligned intermediates containing 5′ or 3′ ter-minal mismatches. Pol4 physically interacts with Dnl4and this interaction stimulates the DNA synthesis ac-tivity of Pol4 as well as the DNA-joining activity of theDnl4/Lif1 complex[13].

Here, we describe the results of an analysis ofthe role of Pol4 during adaptive mutagenesis in ourframeshift detection system. Furthermore, we reportthe observation of a dramatically increased incidenceof adaptive frameshift reversions following treatmentof cell cycle-arrested cells with ionizing radiation, andwe show that DNA ligase IV considerably contributesto this increased frequency of adaptive mutations.

2. Materials and methods

2.1. Yeast strains

udya s

TS

N

EYYYY

Y

Y

Y

h

urned out to be essential for the increased adaeversion frequencies following UV irradiation ornock-out of NER genes, but not for the basal lef adaptive frameshift reversions in cell cycle-arresild-type cells[6].In another study we succeeded in identifying a p

ay which contributes to the wild-type level of adaive mutation, namely the non-homologous end joinNHEJ) pathway of DSB repair. A deficiency for tbligatory NHEJ components Ku70 or DNA ligasebolished about 50% of the frameshift mutations[5].hese results suggested that a NHEJ-mediated reppontaneously occurring DSBs in resting cells is pro introduction of frameshift mutations at the joint,ecially in regions of monotonous repeats.

The NHEJ pathway repairs DSBs in a templandependent fashion, with simple rejoining of the ben ends (reviewed by Lewis[8,9]). It starts with the

The genotypes of all strains used in this stre listed inTable 1. The null allele of POL4 wa

able 1. cerevisiaestrains

ame Genotypea Source

H150 MATa lys2∆BglII [15]LB4 MATa lys2∆BglII dnl4∆::HIS3 [5]LB4� MATα lys2∆BglII dnl4∆::HIS3 [5]LBP4 MATa lys2∆BglII pol4∆::TRP1 This studyLB44 MATa lys2∆BglII dnl4∆::HIS3

pol4∆::TRP1This study

LBD MATa/� homozygous diploid ofEH150

[5]

LB4D MATa/� homozygous diploid ofYLB4

[5]

LBDa/a MATa/a diploid, derived fromYLBD

[5]

a All strains additionally contain the following markers:trp1-∆,is3-∆200, ura3-52, ade2-1o .

E. Heidenreich, H. Eisler / Mutation Research 556 (2004) 201–208 203

produced by oligo-directed gene disruption using aPCR-derived DNA fragment with a central TRP1marker gene. Primers for amplification of the disrup-tion cassette from plasmid DNA were: Pol4trp1f: 5′-GGATAAACATGCGACCTGTTAGACAAATCGCA-CATGTCTCTAAAGGGTAAATTTTTCGCcaagttgatt-ccattgcggtg-3′ and Pol4trp1r: 5′-CCACCTCTTTC-TTTCTCCATACCCAACAATCTTCGCTGTACGAA-ATCTCTCCACTATTCgcagttatgacgccagatggc-3′. Up-percase characters correspond to target-specific se-quences and the lowercase sequences flank theselective marker gene TRP1. The disruption cassettewas transformed into the parent strain EH150 bystandard techniques[14]. Correct integration wasverified by PCR. The double-deficient strain YLB44was derived from a spore of a heterozygous diploidresulting from mating of strains YLB4� and YLBP4.

2.2. Determination of adaptive mutationfrequencies

The experimental procedure was described previ-ously [15]. Briefly, several subcultures of the respec-tive strains were grown in liquid complete medium toobtain the necessary cell numbers. Then, the cells wereharvested and transferred to solid selective medium(synthetic complete medium lacking lysine, at a plat-ing density of 108 cells per plate). Each experimen-tal series consisted of 13 subpopulations plated ontofour plates each. Owing to thelys2∆BglII frameshifta fterc l cy-c dyht ndf fterpt tantsr ingf uset latet ribedp iesf t 4d in-c rise.T ter-e

are not just the result of slowly growing pre-existingrevertants, but are the consequence of adaptive rever-sion events which occurred during starvation-inducedcell cycle arrest.

2.3. �-Irradiation procedure

For a determination of adaptive mutation frequen-cies following�-irradiation the above procedure wasaccordingly modified. Liquid cultures of the strains tobe tested were used to inoculate four identical selec-tive plates (synthetic complete medium lacking lysine)with 5 × 107 cells each. After 5 days of starvation forlysine (thus at a time when all pre-existing revertantsalready have produced colonies), two of the plates wereexposed to 25 Gy of�-irradiation from a60Co source(Theratron 60). Five hours after�-ray treatment smallaliquots of cells (Ø 9 mm discs) were removed and theviability was examined (by determining the number ofcolony-forming units on complete medium) in compar-ison to the starvation survival of cells from the identi-cal non-irradiated plates.�-Irradiation-induced rever-sions, as well as a background of spontaneous rever-sions of thelys2∆BglII allele permitted a restart of pro-liferation, detectable as colony formation. Newly aris-ing lysine-prototrophic colonies on the irradiated platesand the non-irradiated controls were counted daily forthe next five days. The cumulative colony counts werenormalized to the given number of viable cells perstrain and treatment.�-Irradiation-induced adaptive re-v oc-c s in-d n-ti nta-n

3

thes e ofp pu-lh rva-t se-q nt ifer-

llele all strains are auxotrophic for lysine and aonsumption of internal lysine reserves enter a celle arrest (predominantly in G1). Only cells alreaarboring spontaneous reversions oflys2∆BglII at the

ime of plating are able to continue proliferation aorm colonies on the lysine-free medium right alating. We know from reconstruction growth tests[3]

hat such replication-dependent pre-existing reveresult in readily visible prototrophic colonies appearrom days 2 to 3. Adding day 4 as a tolerance wehe cumulative colony counts of days 2 to 4 to calcuhe replication-dependent mutation rates as descreviously[6]. Following the appearance of colon

ormed by pre-existing mutants (i.e. after the firsays of selection for prototrophy), during prolongedubation of the plates further colonies continue to ahese late-arising colonies are the matter of our inst because we know from previous work[3] that they

ertant colonies could not be distinguished fromasional spontaneous adaptive revertant colonieividually, but the comparison of irradiated with u

reated populations substantiated that the�-irradiation-nduced reversions clearly outnumbered the spoeous reversions (seeSection 3.2).

. Results and discussion

The experimental setup we routinely use fortudy of adaptive mutation relies on the emergencrototrophic founder cells of colonies among a po

ation of lysine-auxotrophicS. cerevisiaecells whichave arrested their cell cycle due to protracted sta

ion for lysine. The auxotrophy for lysine is a conuence of thelys2∆BglII frameshift allele. A reversio

o prototrophy and subsequent resumption of prol


ation is possible by the occurrence of compensatingframeshift mutations in a certain region of thelys2gene[3]. While during the first 2 to 4 days of starvation pre-existing revertant cells form colonies, further coloniescontinue to arise as a result of adaptive reversion eventsduring starvation-induced cell-cycle arrest.

3.1. Disruption of the POL4 gene does not reducethe incidence of adaptive frameshift reversions

The yeastPOL4gene encodes a DNA polymeraseformerly regarded as a Pol�. In the light of recent find-ings, nowadays this polymerase is classified as an or-tholog of mammalian DNA polymerase� [16]. Unlikemammalian Pol�, yeast Pol4 probably does not par-ticipate in base excision repair[17] but is required forNHEJ of not fully compatible DNA ends[12]. Since wepreviously observed that about 50% of the frameshiftmutations arising in cell cycle-arrested cells were de-pendent on the NHEJ pathway[5], we were interested ifPol4 might play a role during this mutagenesis. There-fore, we introduced a null allele ofPOL4 into a wild-type as well as into a DNA ligase IV-deficient strain. Adetermination of the mutation rate during proliferationof the resulting strains yieldedlys2∆BglII reversionrates of 1.01± 0.24× 10−8 and 0.95± 0.25× 10−8

for thepol4∆and thednl4∆pol4∆strains, respectively.These values were not significantly different from themutation rates of the wild-type and thednl4∆ strain[5], pointing to a negligible influence of Pol4 on muta-g

dur-i atedf nicw itiaa

r EJ-d ithinm ew riseb t se-q lip-p turea ire ap trim

Fig. 1. Frequencies of spontaneous adaptive Lys+ revertant coloniesin haploid wild-type and DNA ligase IV-deficient strains with eitheran intact or a disruptedPOL4gene. A time-course starting with day4 after plating is shown because earlier arising colonies were con-sidered as originating (surely or potentially) from pre-existing rever-tants. Cumulative colony counts were normalized to the numbers ofcells on the plates. The mean of three experiments is shown togetherwith standard error bars.

protruding single-stranded ends. Such a dispensabil-ity of a polymerase is in agreement with our results.However, right from the start we could not rule outan involvement of Pol4, especially in consideration ofthe unusual high−1 base deletion error rate typicalfor human Pol� [18]. Furthermore, Wilson and Lieber[12] reported that Pol4 is stringently required for re-moval of 5′ or 3′ terminal mismatches during NHEJ,and we presently do not know the constitution of theDNA ends, e.g. if the terminal nucleotides are dam-aged. Therefore, it was important to test Pol4-deficientyeast cells and it was interesting to learn that in ourframeshift detection system Pol4 was not required formutation formation.

Actually, as evident fromFig. 1, Pol4 rather seemsto antagonize adaptive mutation, since the disruptionof thePOL4 gene increased the reversion frequency.To find out if the presence of Pol4 possibly enhancesthe fidelity of NHEJ and thereby reduces adaptive mu-tations in the wild type, we tested adnl4∆pol4∆ strain(Fig. 1). As the adaptive reversion frequency of thisdouble-deficient strain was somewhat higher than thatof thednl4∆ strain which is completely incapable ofNHEJ, we assume that the antagonistic effect of Pol4towards adaptive mutation is partially due to an in-

enesis during proliferation.In contrast, the frequency of adaptive reversions

ng prolonged cell cycle arrest was moderately elevor the Pol4-deficient strain if compared to the isogeild-type strain (Fig. 1). On the basis of this increase

s unlikely that activities of yeast polymerase� (Pol4)re required for adaptive reversions of thelys2∆BglIIllele.

We know from a comparison of thelys2∆BglIIeversion sequence spectra that the typical NHependent reversion events are small deletions wononucleotide repeats[5]. Mechanistically, onould expect that such frame-shifting mutations ay a (possibly staggered) DSB within such a repeauence, followed by a misalignment or an inward sage event prior to ligation. An intermediate strucccordingly misaligned would not necessarily requolymerase activity, but instead some nuclease to


Table 2Cell viability at day 5 with or without�-irradiation (mean of sixexperiments)

Strain Survival without�-irradiation (%)

Total survival after�-irradiation (%)

Wild-type hap-loid

43.8 29.1

dnl4∆ haploid 41.3 22.1Wild-type

diploidMATa/a

28.9 29.9

Wild-typediploidMATa/�

32.9 29.4

dnl4∆ diploid 39.4 38.5

creased fidelity of NHEJ, but also comprises a smallunknown component — probably related to the so farunidentified mechanism which produces the remaining50% of adaptive frameshift mutations.

3.2. �-Irradiation provokes a substantial increasein the formation of adaptive frameshift mutations

Previous results indicated a significant contributionof the repair of spontaneously arising DSBs to the in-cidence of frameshift mutations in cell cycle-arrestedcells[5]. Therefore, we asked if this effect could be am-plified and accordingly reproduced by a systematic in-troduction of strand breaks. We employed�-irradiationas a potent source of DSBs, and treated the cells afterfive days of starvation for lysine to be sure that all pre-existing revertants had already formed colonies. A lowdose of�-radiation was chosen to induce DSBs withoutsubstantially reducing the number of viable cells. The25 Gy applied did not kill more than 50% of the cellsviable before�-ray treatment. Diploid cells were nearlyunaffected by this dose (Table 2). To account for differ-ences in starvation survival and�-irradiation survivalbetween the different strains, the revertant yield wasnormalized to the numbers of viable cells present after�-ray treatment, and at the same time for the untreatedcontrol, respectively.

In spite of the weak impact on cell viability, the ef-fect of 25 Gy�-irradiation on the incidence of adap-tive, cell cycle arrest-releasing frameshift mutationsw con-t tivem arti-fi tion

Fig. 2. Comparison of spontaneous and�-irradiation-induced adap-tive reversion frequencies in haploid and diploid wild-type strains.The graph shows the accumulation of revertant colonies in parallelcell populations, either unirradiated (triangles) or�-irradiated on day5 (squares). Cumulative colony counts were normalized to the num-ber of viable cells (present 5 h after�-treatment of half of the plates).The mean of six experiments is shown together with standard errorbars.

of frameshift mutations in cell cycle-arrested cells and,based on our previous results[5], it is likely that thesemutations were generated by mutagenic repair of theDSBs.

As expected, the diploid wild-type strain was lessprone to�-induced adaptive mutation than the hap-loid wild type. Unlike stationary-phase haploid cellsdiploid cells are able to take advantage of the addi-tional chromosome set as a template for DSB repair byhomologous recombination[19].

3.3. In cell cycle-arrested DNA ligase IV-deficientcells the formation of�-induced frameshiftreversions is significantly reduced

To answer the question if the�-induced increase inadaptive mutagenesis may be effected by NHEJ, we ex-posed Dnl4-deficient haploid and diploid strains to thesame dose of�-irradiation and compared the incidencesof revertant colonies to the data of the isogenic wild-type strains. Concerning the haploid strains, the major-

as substantial in comparison to the untreatedrol (Fig. 2). Due to the observed increase in adaputation frequencies we have to conclude that an

cial excess of DSBs massively induced the forma


Fig. 3. Frequencies of�-irradiation-induced adaptive frameshift re-versions in wild type and DNA ligase IV-deficient strains. Wild-typea/a is aMATa-homozygous diploid strain. The bars represent themeans (±standard error) of the cumulative frequencies of revertantcolonies arising within five days after a 25 Gy�-irradiation (derivedfrom six experiments each).

ity (about 75%) of the�-induced reversions turned outto depend on Dnl4 (Fig. 3). Thus, a functional NHEJmachinery is most likely necessary for the formationof �-induced frameshift mutations. Thereby, this resultis in agreement with our model worked out for spon-taneous adaptive mutagenesis[5], implying that theNHEJ pathway introduces frameshift mutations duringDSB repair in stationary-phase cells.

With diploid strains, the situation inS. cerevisiaeismore complex. NHEJ is down-regulated in diploid cellsby a mechanism which senses mating type heterozy-gosity and prevents nuclear localization of the obliga-tory Dnl4-cofactor Lif1[20,21]. Accordingly, a knock-out of theDNL4gene had less effect in the diploid strain(in comparison to the wild type of the same ploidy)than in the haploid strain (Fig. 3). However, the adap-tive mutation frequencies of both diploid strains werehigher than those of the haploiddnl4∆ strain. Thiscould be caused by a higher probability that one ofthe twolys2∆BglII genes in a diploid genome reverts,which is sufficient for proliferation. This assumption issupported by the data obtained with a diploidMATa/astrain. In such strains, which are able to express only

one type of the two alternative mating types, NHEJis not repressed in spite of diploidy[22,23]. Consis-tent (i), with our hypothesis that NHEJ produces mostof the �-induced frameshift reversions and (ii), withour assumption that the chance of reversion should behigher in a diploid strain, the reversion frequency oftheMATa/a strain was not only higher than that of thediploidMATa/α but also than that of the haploid strain(Fig. 3).

Although �-induced adaptive mutagenesis is re-duced substantially in DNA ligase IV-deficient strains,it is not abolished. The level of revertants in�-irradiateddnl4∆ strains still is significantly higher than the levelof spontaneous adaptive frameshift reversions ofdnl4∆strains without exposure to ionizing radiation (deter-mined in an earlier study[5]). The most probable ex-planation of this effect is that�-irradiation — besidesmultiplying the level of NHEJ-mediated frameshifts —also stimulates the so far unknown mechanism(s) re-sponsible for the remaining 50% of spontaneous adap-tive reversion. Ionizing radiation not only causes DSBs,but also single-strand breaks and diverse types of ox-idative DNA damage[24,25]. Prompted by the presentknowledge, we guess that the latter two types of DNAdamage might be additional triggers of mutagenic re-pair.

With respect to cell viability, we noted that (thoughthe haploiddnl4∆ strain exhibited the lowest via-bility following 25 Gy �-irradiation) the impact ofa deletion of theDNL4 gene on�-induced killingw ildt ofs ibuteo hel m-b liest SBsw arem Bs.A ayslc n-ss ffer-e k ofr

andD ts

as relatively weak compared to the haploid wype (Table 2). Therefore, in the present situationtarvation-induced G1 arrest NHEJ seems to contrnly to a minor extent to cell survival, in spite of t

acking option of DSB repair by homologous recoination. This effect could be explained if one imp

hat either radiation damage other than DSBs, or Dith aberrant ends which cannot be ligated by NHEJore fatal to resting cells than NHEJ-substrate DSrechanneling of DSBs to alternative repair pathw

ike, e.g. single-strand annealing[26,27] could alsoontribute to a higher viability by partially compeating for the NHEJ deficiency in the haploiddnl4∆train. However, the exact interdependency of dint modes of error-free or error-prone repair, a lacepair and cell death remains to be established.

In summary, the response of repair-proficientNA ligase IV-deficient cells to�-irradiation sugges


that the same pathway, NHEJ, which in stationary cellsaccomplishes mutagenic repair of endogenously oc-curring DSBs, also is active in mutagenic repair ofDSBs induced by exogenous sources. Experimentalwork with mammalian cells established that NHEJacts as a caretaker of the genome (by repairing DSBs)but may also actively produce chromosomal rearrange-ments [28]. Such genome rearrangements like, e.g.translocations are just as relevant for cancerogenesisas protein truncations caused by frameshift mutationsare[29,30]. Therefore, in the future it will be interestingto see with the aid of a translocation detection system,if not only NHEJ-mediated frameshift mutations, butalso NHEJ-mediated translocations might arise spon-taneously in cell cycle-arrested yeast cells.

Acknowledgements

We thank V. Weinberger for technical assistance, R.Kodym for providing access to the Theratron 60, andU. Wintersberger for helpful discussions and commentson the manuscript. This work was supported by a grantfrom the Herzfeldersche Familienstiftung.

References

[1] S.M. Rosenberg, Evolving responsively: adaptive mutation,Nat. Rev. Genet. 2 (2001) 504–515.

on,

of a

. 473

thesis in)

U.rtant. 22

pen-otide04)

i,ase-

[8] L.K. Lewis, M.A. Resnick, Tying up loose ends: nonhomolo-gous end-joining inSaccharomyces cerevisiae, Mutat. Res. 451(2000) 71–89.

[9] S.E. Critchlow, S.P. Jackson, DNA end-joining: from yeast toman, Trends Biochem. Sci. 23 (1998) 394–398.

[10] A.J. Doherty, S.P. Jackson, DNA repair: how Ku makes endsmeet, Curr. Biol. 11 (2001) 920–924.

[11] L. Chen, K. Trujillo, W. Ramos, P. Sung, A.E. Tomkin-son, Promotion of Dnl4-catalyzed DNA end-joining by theRad50/Mre11/Xrs2 and Hdf1/Hdf2 complexes, Mol. Cell. 8(2001) 1105–1115.

[12] T.E. Wilson, M.R. Lieber, Efficient processing of DNA endsduring yeast nonhomologous end joining. Evidence for a DNApolymerase beta (Pol4)-dependent pathway, J. Biol. Chem. 274(1999) 23599–23609.

[13] H.M. Tseng, A.E. Tomkinson, A physical and functional inter-action between yeast Pol4 and Dnl4-Lif1 links DNA synthesisand ligation in nonhomologous end joining, J. Biol. Chem. 277(2002) 45630–45637.

[14] R.D. Gietz, R.A. Woods, Transformation of yeast bylithium acetate/single-stranded carrier DNA/polyethylene gly-col method, Meth. Enzymol. 350 (2002) 87–96.

[15] E. Heidenreich, U. Wintersberger, Replication-dependent andselection-induced mutations in respiration-competent andrespiration-deficient strains ofSaccharomyces cerevisiae, Mol.Gen. Genet. 260 (1998) 395–400.

[16] P.M. Burgers, E.V. Koonin, E. Bruford, L. Blanco, K.C. Burtis,M.F. Christman, W.C. Copeland, E.C. Friedberg, F. Hanaoka,D.C. Hinkle, C.W. Lawrence, M. Nakanishi, H. Ohmori, L.Prakash, S. Prakash, C.A. Reynaud, A. Sugino, T. Todo, Z.Wang, J.C. Weill, R. Woodgate, Eukaryotic DNA polymerases:proposal for a revised nomenclature, J. Biol. Chem. 276 (2001)43487–43490.

[17] M. McInnis, G. O’Neill, K. Fossum, M.S. Reagan, Epistaticanalysis of the roles of theRAD27andPOL4gene products in

)

[ hepli-90.

[ n in-e

[ ypegase

[ -urr.

[ teaksiol. 9

[ tion

[ Mu-

[2] P.L. Foster, Adaptive mutation: implications for evolutiBioassays 22 (2000) 1067–1074.

[3] E. Heidenreich, U. Wintersberger, Adaptive reversionsframeshift mutation in arrestedSaccharomyces cerevisiaecellsby simple deletions in mononucleotide repeats, Mutat. Res(2001) 101–107.

[4] A. Halas, H. Baranowska, Z. Policinska, The influence ofmismatch-repair system on stationary-phase mutagenethe yeastSaccharomyces cerevisiae, Curr. Genet. 42 (2002140–146.

[5] E. Heidenreich, R. Novotny, B. Kneidinger, V. Holzmann,Wintersberger, Non-homologous end joining as an impomutagenic process in cell cycle-arrested cells, EMBO J(2003) 2274–2283.

[6] E. Heidenreich, V. Holzmann, H. Eisler, Polymerase zeta dedency of increased adaptive mutation frequencies in nucleexcision repair-deficient yeast strains, DNA Repair 3 (20395–402.

[7] N. Babudri, Y.I. Pavlov, N. Matmati, C. Ludovisi, A. AchillStationary-phase mutations in proofreading exonucledeficient strains of the yeastSaccharomyces cerevisiae, Mol.Genet. Genom. 265 (2001) 362–366.

DNA base excision repair inS. cerevisiae, DNA Repair 1 (2002311–315.

18] K. Bebenek, M. Garcia-Diaz, L. Blanco, T.A. Kunkel, Tframeshift infidelity of human DNA polymerase lambda. Imcations for function, J. Biol. Chem. 278 (2003) 34685–346

19] F. Paques, J.E. Haber, Multiple pathways of recombinatioduced by double-strand breaks inSaccharomyces cerevisia,Microbiol. Mol. Biol. Rev. 63 (1999) 349–404.

20] M. Frank-Vaillant, S. Marcand, NHEJ regulation by mating tis exercised through a novel protein, Lif2p, essential to the liIV pathway, Genes Develop. 15 (2001) 3005–3012.

21] A. Kegel, J.O. Sjostrand, S.U.Astrom, Nej1p, a cell typespecific regulator of nonhomologous end joining in yeast, CBiol. 11 (2001) 1611–1617.

22] S.E. Lee, F. Paques, J. Sylvan, J.E. Haber, Role of yeasSIRgenes and mating type in directing DNA double-strand brto homologous and non-homologous repair paths, Curr. B(1999) 767–770.

23] S.U.Astrom, S.M. Okamura, J. Rine, Yeast cell-type regulaof DNA repair, Nature 397 (1999) 310.

24] E.C. Friedberg, G.C. Walker, W. Siede, DNA Repair andtagenesis, ASM Press, Washington, DC, 1995.


[25] J.F. Ward, DNA damage produced by ionizing radiation in mam-malian cells: identities, mechanisms of formation, and repara-bility, Prog. Nucl. Acids Res. Mol. Biol. 35 (1988) 95–125.

[26] A. Pastink, J.C. Eeken, P.H. Lohman, Genomic integrity andthe repair of double-strand DNA breaks, Mutat. Res. 480–481(2001) 37–50.

[27] L.S. Symington, Role ofRAD52epistasis group genes in homol-ogous recombination and double-strand break repair, Microbiol.Mol. Biol. Rev. 66 (2002) 630–670.

[28] K. Rothkamm, M. Kuhne, P.A. Jeggo, M. Lobrich, Radiation-induced genomic rearrangements formed by nonhomologousend-joining of DNA double-strand breaks, Cancer Res. 61(2001) 3886–3893.

[29] W.B. Coleman, G.J. Tsongalis, The role of genomic insta-bility in human carcinogenesis, Anticancer Res. 19 (1999)4645–4664.

[30] C. Lengauer, K.W. Kinzler, B. Vogelstein, Genetic instabilitiesin human cancers, Nature 396 (1998) 643–649.

non-homologous end joining dependency of γ-irradiation-induced adaptive frameshift mutation...

Documents