Temporal regulation of the Mus81-Mms4 endonuclease ensures ...digital.csic.es › ... › 1 ›...

Temporal regulation of the Mus81-Mms4endonuclease ensures cell survival underconditions of DNA damageIrene Saugar1 Marıa Victoria Vazquez1 Marıa Gallo-Fernandez1

Marıa Angeles Ortiz-Bazan1 Monica Segurado1 Arturo Calzada2 and

Jose Antonio Tercero1

1Centro de Biologıa Molecular Severo Ochoa (CSICUAM) Cantoblanco 28049-Madrid Spain and2Centro Nacional de Biotecnologıa (CSIC) Cantoblanco 28049-Madrid Spain

Received May 14 2013 Revised June 25 2013 Accepted June 30 2013

ABSTRACT

The structure-specific Mus81-Eme1Mms4 endo-nuclease contributes importantly to DNA repairand genome integrity maintenance Here usingbudding yeast we have studied its function andregulation during the cellular response to DNAdamage and show that this endonuclease is neces-sary for successful chromosome replication and cellsurvival in the presence of DNA lesions that interferewith replication fork progression On the contraryMus81-Mms4 is not required for coping with repli-cative stress originated by acute treatment withhydroxyurea (HU) which causes fork stallingDespite its requirement for dealing with DNAlesions that hinder DNA replication Mus81-Mms4activation is not induced by DNA damage at replica-tion forks Full Mus81-Mms4 activity is only acquiredwhen cells finish S-phase and the endonucleaseexecutes its function after the bulk of genome rep-lication is completed This post-replicative mode ofaction of Mus81-Mms4 limits its nucleolytic activityduring S-phase thus avoiding the potentialcleavage of DNA substrates that could causegenomic instability during DNA replication At thesame time it constitutes an efficient fail-safe mech-anism for processing DNA intermediates thatcannot be resolved by other proteins and persistafter bulk DNA synthesis which guarantees thecompletion of DNA repair and faithful chromosomereplication when the DNA is damaged

INTRODUCTION

Coping with DNA damage during chromosome replica-tion is a major challenge for cells An effective DNAdamage response is essential for faithful chromosome du-plication and genome integrity maintenance and requiresthe coordination of checkpoints with multiple pathwaysand different proteins that are involved in DNA repair(1ndash5) Among them structure-specific endonucleasescontribute importantly to genomic stability by cleavingsome DNA secondary intermediates that are generatedduring replication-associated repair and need nucleolyticresolution (6ndash10)Mus81-Eme1Mms4 is an evolutionarily conserved

structure-specific endonuclease involved in the cellularresponse to DNA damage (811) Yeast cells lacking thecatalytic (Mus81) or the non-catalytic subunit (Mms4 inSaccharomyces cerevisiae Eme1 in Schizosaccharomycespombe and human cells) of this complex show sensitivityto a variety of DNA-damaging agents that interfere withDNA replication (12ndash17) Likewise mammalian cells de-ficient in MUS81 or EME1 are sensitive to some agentsthat cause DNA lesions (18ndash21) and this endonuclease isnecessary for the repair of broken DNA replication forks(RFs) (2223) Genetic analyses showed that Mus81-Eme1Mms4 is required for cell viability in the absenceof components of the complex formed by the RecQ-helicase Sgs1 Top3 and Rmi1 in budding yeast (BLM-TOPIIIa-RMI1-RMI2 in human cells) (13151624ndash28)indicating overlapping functions between Mus81 andRecQ-helicase complexes Mus81-Mms4 also has func-tional overlap with the Yen1 resolvase during DNArepair (29ndash32) The synthetic lethality between Mus81and RecQ complexes is suppressed by eliminating early

To whom correspondence should be addressed Tel +34 91 1964818 Fax +34 91 1964420 Email jatercerocbmuamesPresent addressMonica Segurado Instituto de Biologıa Funcional y Genomica (CSICUSAL) 37007-Salamanca Spain

Published online 30 July 2013 Nucleic Acids Research 2013 Vol 41 No 19 8943ndash8958doi101093nargkt645

The Author(s) 2013 Published by Oxford University PressThis is an Open Access article distributed under the terms of the Creative Commons Attribution License (httpcreativecommonsorglicensesby30) whichpermits unrestricted reuse distribution and reproduction in any medium provided the original work is properly cited

steps of homologous recombination (1624ndash26) whichtogether with its interaction with Rad54 (14) suggested arole for Mus81-Eme1Mms4 during recombination-mediated DNA repair (11) Nevertheless this nuclease isnot required for the repair of double-strand breaksthrough homologous recombination as mus81mms4mutants are not sensitive to g-irradiation (12ndash14)Although it is still unclear which are the substrates ofMus81-Eme1Mms4 in the cell a number of biochemicalstudies have shown that it can cleave different branchedDNA structures in vitro all of which are its potentialtargets in vivo Thus it is able to act with differentaffinity on model RFs 30-flaps (30-FLs) D-loopsX- and Y-shaped structures (15162433ndash41)The Mus81-Eme1Mms4 endonuclease has a significant

role in the maintenance of genome stability but itsnucleolytic activity needs to be strictly regulated toavoid the undesired opposite effect The unrestrainedcleavage of potential substrates such as RFs and otherDNA intermediates would have negative consequencesfor chromosome replication and could also lead to theformation of chromosomal rearrangements or high levelsof chromatid exchanges Recent studies in budding yeasthave uncovered key features of the regulation of this endo-nuclease showing that in an unperturbed mitotic cell cycleits activity is controlled by Cdc28CDK1- and Cdc5PLK1-dependent phosphorylation of the non-catalytic subunitMms4 (4243) This phosphorylation is required for thefull nuclease activity of the complex and restricts thefunction of Mus81-Mms4 to a narrow period in anormal cell cycle just before chromosome segregation(4243) A recent report has confirmed this mode of regu-lation for Mus81-Mms4 and has provided further evidenceof its importance for the maintenance of genomic stabilityshowing that the uncontrolled activation of this endo-nuclease causes deleterious mitotic crossovers and prema-ture processing of DNA intermediates (44) HumanMus81-Eme1 is also regulated through phosphorylationof the non-catalytic subunit like in budding yeast (42)In addition it has been proposed that in human cellsMus81 is negatively controlled by Wee1 (45) Interes-tingly a recent work has shown that in S pombe thecell cyclendashdependent phosphorylation of Eme1Mms4 isCdc2CDK1-dependent but Plo1PLK1-independent (46)unlike in budding yeast and human cells thus showingdifferences among organisms in the way that Mus81-Eme1Mms4 is regulated Moreover in fission yeastEme1Mms4 hyperphosphorylation is stimulated when cellsare treated with camptothecin or bleomycin This induced-hyperphosphorylation is Rad3-dependent and increasesthe activity of S pombe Mus81-Eme1 (46)Despite the recent findings on Mus81-Eme1Mms4

regulation and its established function in the resistanceto some DNA-damaging agents that hinder chromosomereplication it is still largely unknown how this endonucle-ase operates when cells need to cope with DNA damage atRFs Mus81-Eme1Mms4 is known to be crucial fordealing with DNA lesions such as those induced by themodel DNA-damaging agent methyl methanesulfonate(MMS) which interfere with RF progression Yetbecause this endonuclease can cleave DNA RFs its

activity must be carefully controlled somehow underthese situations of DNA damage Although above-men-tioned works have shown that Mus81-Mms4 is activatedwhen cells finish S-phase in the absence of damaged DNA(4243) and in S pombe its activity is enhanced by someDNA-damaging drugs in G2-cells (46) it is currentlyunclear how Mus81-Mms4 is regulated in response toDNA lesions occurring during replication Here wehave analysed the role of budding yeast Mus81-Mms4under conditions of DNA damage during S-phase andshow that this endonuclease is necessary for successfulchromosome replication when cells are exposed toMMS However our data indicate that Mus81-Mms4 ac-tivation is not induced by the presence of DNA lesionsand that its function to respond to DNA damage whichallows cell survival is carried out after completion of bulkgenome replication

MATERIALS AND METHODS

Strains media cell cycle synchronization cell viability

The yeast strains used in this work were constructed bystandard procedures and are listed in SupplementaryTable S1 The pYM (47) and the pML (48) plasmidseries were used as templates for polymerase chain reac-tion Yeast cells were grown at 30C in YP medium (1yeast extract 2 bacto peptone) with 2 glucose Bactoagar (2) was added for solid medium For gene expres-sion using the GAL1-10 promoter the YP medium wassupplemented with 2 raffinose or 2 galactose The a-factor mating pheromone was added to a final concentra-tion of 5ndash10 mgml to synchronize cells in G1-phase HUwas used at 02M to block cells in early S-phaseNocodazole was used at 5 mgml to block cells in G2MSamples for flow cytometry were collected and processedas described (49) and analysed using a FACSCalibur flowcytometer (BD Biosciences) For fluorescence microscopyanalysis cells were grown in minimal medium supple-mented with yeast synthetic drop out (Sigma-Aldrich)Living cells were analysed using an Axiovert200 Zeissfluorescence microscope with a 63 oil immersion object-ive Cell viability was determined by plating cells in trip-licate onto YP-glucose (YPD) plates and counting colony-forming units after 3 days of incubation at 30C

Immunoblotting and in situ kinase assay

Protein extracts for immunoblotting analysis wereprepared from trichloroacetic acid-treated cells asdescribed (50) HA-tagged proteins were detected withthe 12CA5 antibody (CBMSO) using horseradish perox-idasendashcoupled anti-mouse (Vector Labs) as a secondaryantibody Rad53 was detected with the JDI48 antibody (agift from Dr J Diffley Cancer Research UK) and thehorseradish peroxidasendashcoupled protein A (Invitrogen) asa secondary antibody Immunoreactive bands werevisualized using enhanced chemiluminescence (ECLprime GE Healthcare) The Rad53 in situ kinase assaywas performed essentially as described (51) For theauthophosphorylation reaction the membranes (HybondP GE Healthcare) were incubated for 1 h at room

8944 Nucleic Acids Research 2013 Vol 41 No 19

temperature (RT) in kinase buffer containing 10 mCiml[g-32P]ATP

Drug sensitivity assays

Logarithmic cultures growing at 30C were normalized to1 107 cellsml and 10-fold serial dilutions were spottedonto YP plates containing either glucose or galactose anddifferent concentrations of MMS (Sigma-Aldrich) Theplates were incubated at 30C for 48ndash72 h

Dense isotope transfer

Dense isotope substitution experiments were based on (52)and performed as described (53) The cells were grown foreight generations at 30C in minimal medium with 0113C glucose and 001 15N (NH4)2SO4 (CK GasProducts) (lsquoheavy mediumrsquo) They were then synchronizedin G1 transferred to lsquolightrsquo minimal medium [2 12Cglucose and 001 15N (NH4)2SO4] and released intofresh minimal medium either with or without MMS(0033) The DNA was digested with the restrictionenzymes ClaI and SalI and the digestion products wereseparated on CsCl gradients They were fractionated andthe fractions were blotted and hybridized with specificDNA probes recognizing the ClaISalI restriction frag-ments from chromosome VI (54) DNA replication wasseen as the transfer from the heavyndashheavy (HH) peak(unreplicated DNA) in which the two DNA strands aresubstituted with dense isotopes to the heavyndashlight (HL)peak (replicated DNA) where only the parental strandhas the heavy isotopes The extent of replication wascalculated from the following equation replica-tion=100[05 HL(HH+05 HL)]

Pulse field gel electrophoresis

Genomic DNA samples were obtained from 108 cells andprepared in low melting point agarose plugs as described(55) Yeast chromosomes were separated in a 1 agarose-TBE gel by pulsed-field gel electrophoresis (PFGE) at14C using a Gene Navigator System from PharmaciaBiotech The electrophoresis were carried out at 180V(6Vcm) for 24 h with 90 s 105 s and 125 s pulses for96 6 and 84 h respectively The gels were stained withethidium bromide and scanned on a Gel Doc 2000(Biorad) Quantification of the chromosome bands wasperformed using the Quantity One program (Biorad)

Nuclease activity assays

The oligonucleotides used to make the DNA substrateswere based on (24) For the construction of a model RFthe oligonucleotides were RF-1 50-GACGCTGCCGAATTCTGGCGTTAGGAGATACCGATAAGCTTCGGCTTAAG RF-2 50-ATCGATGTCTCTAGACAGCACGAGCCCTAACGCCAGAATTCGGCAGCGTC RF-350-CTTAAGCCGAAGCTTATCGGTATCT and RF-450-GCTCGTGCTGTCTAGAGACATCGAT For theconstruction of a 30-FL the oligonucleotides were RF-1RF-2 and RF-4 For the formation of the syntheticstructures used as the substrates in the nuclease assaysthe oligonucleotide RF-1 was 50-32P-labelled using

[g-32P]ATP (Perkin Elmer) and T4 Polynucleotide kinase(New England Biolabs) and then annealed with an excessof their complementary oligonucleotides The annealingwas carried out by heating the DNA molecules for10min at 80C in 200mM NaCl plus 60mM TrisndashHCl(pH 75) buffer followed by cooling to RTImmunoaffinity purification and nuclease assays were

based on a previously described method (56) HA-Mms4was immunoaffinity purified from 5 108 cells whichwere disrupted using glass beads in 800 ml of bindingbuffer (40mM TrisndashHCl pH 75 100mM AcK 4glycerol 01 NP40 5mM DTT 5mM NaF 5mmNa4P2O7 protease inhibitors cocktail (2) from RochemdashComplete) and cleared by 40min centrifugation at 4C topspeed The supernatant was incubated for 180min at 4Cwith 12CA5 antibody and followed by 60min incubationwith 15 ml of Protein A-Sepharose (Sigma-Aldrich) TheSepharose-bound proteins were centrifuged washed exten-sively and used directly for the reactions The reaction mix-tures for the nuclease activity assays (25 ml) containedlabelled DNA substrate (40 fmol) in 100mM NaCl50mM TrisndashHCl pH 8 3mM MgCl2 05ndash1 mg poly[dI-dC] DTT 02mM plus the affinity purified tagged-Mms4The reactions were incubated for 60min at 30C andstopped with denaturing stop buffer (final concentration19 formamide 4mM EDTA 001 xylene-cyanol and001 bromophenol) The 32P-labelled products wereanalysed by electrophoresis through 125 denaturinggels containing 7M urea Gels were run at constant 23W

RESULTS

Mus81-Mms4 is important for the cellular response toDNA damage that originates during S-phase

Cells lacking Mus81-Eme1Mms4 are sensitive to thetreatment with different DNA-damaging agents likethe alkylating compound MMS (13ndash15) To determinethe role of this endonuclease specifically in replicationover DNA lesions we restricted DNA damage to asingle S-phase and analysed the consequences of theabsence of Mus81-Mms4 We first made null mutants ofthe genes encoding the subunits of the Mus81-Mms4complex in budding yeast mus81 and mms4 as wellas a double mutant mus81mms4 The strains weretested for growth on solid medium with MMS Allmutants grew like wild-type cells in medium withoutdrug but were much more sensitive to MMS than thecontrol as expected (Figure 1A) We next studiedS-phase progression in cells treated with MMS andexamined its viability throughout the experiment Forthis wild-type and mutant cells were synchronized in G1using a-factor and then released into fresh medium eitherwithout or with MMS at several concentrations In allcases the cells finished replication by 60min in mediumwithout MMS as shown by flow cytometry and pro-gressed later to a new cell cycle (Figure 1B) In contrastin the presence of MMS (for simplicity only the resultswith 0033 MMS are shown) the cells progressed slowlythrough S-phase due to the DNA lesions as shown before(5457) Flow cytometry shows that the slow S-phase in

Nucleic Acids Research 2013 Vol 41 No 19 8945

the three mutants treated with MMS was similar to that ofthe control strain (Figure 1B) suggesting that bulkgenome replication is not affected by the absence ofMus81-Mms4 when the DNA is damaged Howeverunlike the wild-type control all mutants showed a highloss of viability during the experiment when treated withMMS with no differences among the strains (Figure 1C)Therefore although apparently S-phase progression issimilar in wild-type and mus81mms4 cells exposed toMMS cells lacking Mus81-Mms4 might neverthelesshave important chromosome replication defects underthese conditions that severely affect cell survivalTo exclude that the loss of viability observed was an

indirect effect due to a defective S-phase checkpoint inmus81mms4 cells which could yield similar results(5457) we monitored checkpoint activation As theresults obtained in the previous experiments with the

three mutants were identical we used from now mus81as the mutant lacking the endonuclease First we analysedthe phosphorylation and kinase activity of the checkpointprotein Rad53 in mus81 cells during S-phase in the pres-ence or absence of MMS The immunoblot (Figure 1D)showed that Rad53 was only hyperphosphorylated whenmus81 cells were treated with MMS Moreover Rad53phosphorylation correlated with the acquisition of kinaseactivity by this protein as shown by an in situ assay(Figure 1D) that allows measuring its autophosphory-lation (51) Second spindle elongation was analysed inS-phase cells treated with MMS using TUB1-GFP Thedata (Figure 1D) show that like the wild-type control andunlike rad53 cells which cannot restrain mitosis despiteincomplete DNA replication spindle elongation was in-hibited in mus81 cells indicating that the checkpointprevents mitosis in this mutant In summary cells

D

A

B

mus81Δ

BF

wt rad53Δ

GFP-TUB1

C

mus81Δmms4Δ

MUS81+MMS4+

mus81Δmms4Δ

no MMS 0005 MMS 00075 MMS 00025 MMS

Time after release from G1 (min)

01

1

10

100

0 60 120 180

no MMS

Via

bilit

y (

log

)

01

1

10

100

0020 MMS 0033 MMS

01

1

10

100

0015 MMS

0 60 120 180

0 60 120 180 0 60 120 180

Via

bilit

y (

log

)

mus81Δmms4Δmus81Δmms4Δ

MUS81+

MMS4+


Immunoblotmus81Δ

Rad53

In situ kinase assay

Rad53-P Rad53

30 60 90 120 150 180 210 rel S phase (+ MMS) (min)

240 G1 (αF)

Loading control (Pounceau S)

30 60 90 120 150 180 210 rel S phase (- MMS) (min)

240 G1 (αF)

01

1

10

100

Figure 1 Mus81-Mms4 is necessary to respond to DNA damage occurring during S-phase (A) Sensitivity to MMS Serial dilutions (10-fold) ofnormalized log-phase cultures were spotted onto YPD plates containing the indicated amounts of MMS and incubated at 30C for 48ndash72 hMUS81+MMS4+ (W303-1a strain) mus81 (YMV39 strain) mms4 (YMV48 strain) mus81 mms4 (YMV49 strain) (B) Cells were synchronizedin G1 and released in medium plusmnMMS (0033) The cells were collected at the indicated time points and the DNA content was determined by flowcytometry (C) Cell viability after MMS treatment during S-phase The different MMS concentrations used are indicated at the top of each panel (D)Checkpoint activation Upper panel immunoblot analysis of Rad53 Middle panel in situ autophosphorylation assay for Rad53 Bottom panelfluorescence microscopy G1-blocked cells were released into medium with MMS (0033) for 2 h and spindle elongation was analysed Wild type(YJT126 strain) mus81 GFP-TUB1 (YMV22 strain) rad53 GFP-TUB1 (YJT127 strain)


lacking the endonuclease are checkpoint proficient andthe loss of viability after exposure to MMS duringS-phase is due to the absence of Mus81-Mms4 and notto a defective checkpoint response Additionally theseresults indicate that in the absence of DNA damagelack of Mus81-Mms4 did not induce detectable check-point activation

Mus81-Mms4 is necessary for successful completion ofchromosome replication under DNA-damaging conditions

The experiments above suggested that cells lackingMus81-Mms4 might accumulate severe problems duringreplication in the presence of DNA damage as theyshowed a significant loss of viability with respect to thewild-type control despite an apparent similar S-phase pro-gression (Figure 1B and C) Therefore it was necessary toanalyse the dynamics of chromosomal replication usinghigher resolution approaches than flow cytometry

We first studied the role of this endonuclease in thereplication of damaged DNA by analysing chromosomesby PFGE after treating cells with MMS (Figure 2AndashC)MUS81+and mus81 cells were synchronized in G1 witha factor and then released into S-phase in medium con-taining MMS After 60min the MMS was removed andthe cells were released into fresh medium and allowed toprogress through S-phase Nocodazole was added toavoid progression through a new cell cycle stoppingcells at G2M Flow cytometry (Figure 2A) showed thatafter 60min with MMS both MUS81+ and mus81 cellsremained with less than 2C DNA content After removingthe MMS both strains progressed similarly throughS-phase and reached the 2C DNA peak indicating com-pletion of bulk DNA replication Consistent with theresults in Figure 1B and C the viability of MUS81+

cells was high throughout the experiment whereas it wasreduced in the mus81 mutant (Figure 2B)

The PFGE technique resolves linear chromosomes fromagarose-embedded cells but DNA containing replicationbubbles remains trapped in the loading wells (58) In bothMUS81+ and mus81 cells intact chromosomal DNAfrom G1-arrested cells was separated as discrete bandsAfter 60min MMS treatment in S-phase no chromosomebands were detected indicating ongoing replication inboth strains (Figure 2C) consistent with flow cytometry(Figure 2A) In MUS81+ cells full-length chromosomesstarted re-entering the gel after 2 h recovery in MMS-free medium and by 4 h a high proportion of chromo-somal DNA appeared as clear bands (about 80 withrespect to the G1-signal) indicating that in most cellsthe chromosomes were recovered from the MMS-induced lesions and completed replication On thecontrary in mus81 cells most of the chromosomalDNA was retained in the wells 4 h after recovery fromMMS treatment (only 15 with respect to the G1-signalappeared as discrete bands) showing that it was not com-pletely replicated Thus these results indicate that Mus81-Mms4 is required for successful chromosome replicationwhen the DNA is damaged even when S-phase progres-sion seems to be completed in both wild-type and mus81cells (Figure 2A) which in turn explains why cells lacking

the endonuclease lose viability under these conditions(Figures 1C and 2B) These data are in agreement with aprevious report showing similar results (30) where it wasproposed a function for Mus81 in resolving stalled forksor recombination intermediates after DNA damage toallow DNA synthesisAlthough the PFGE data indicate a role for Mus81 in

facilitating the completion of replication under DNA-damaging conditions this technique does not showwhetherhow fork progression is affected and does notallow an estimation of the extent of the replicationdefects To study replication in more detail we used asecond approach based on dense isotope transfer whichprovides a method to study ongoing DNA synthesis at aparticular replicon by analysing the progression of DNARFs (525960) Wild-type and mus81 cells were grown inmedium with lsquoheavyrsquo isotopes synchronized in G1 andreleased into medium containing lsquolightrsquo isotopes eitherwith or without MMS The replication of six DNA restric-tion fragments was analysed in a replicon of 75 kb ofchromosome VI (54) from the early origin ARS607 tothe end of the chromosome (Figure 2DndashE) In bothMUS81+ (Figure 2D) and mus81 (Figure 2E) strainsall DNA fragments were in the HH (unreplicated DNA)peak in G1-cells (top row) In both strains the fragmentsshifted to the HL (replicated DNA) peak by 60min afterrelease from G1 medium without MMS (bottom row)which indicates that they had been replicated WhenMUS81+ cells were released into S-phase in mediumwith MMS the fragment 1 containing ARS607 largelyshifted to the HL position by 30min and at this timethere was some HL DNA in fragment 2 (Figure 2D)Replication progressed rightward fragment 3 shifted tothe HL peak by 60min and the replication of fragments4ndash6 proceeded progressively at later time points (the dataare quantified at the bottom of Figure 2D) Thus as pre-viously described (5461) DNA RFs move slowly butefficiently through a damaged repliconIn mus81 cells the fragment 1 also shifted to the HL

peak by 30min after release from G1 in medium withMMS (Figure 2E) indicating that the firing of ARS607occurs like in the wild-type control Moreover replicationprogressed rightward as seen by the shifting of the frag-ments to the HL peak However the estimation of theextent of replication along the replicon indicated thatchromosome replication was compromised in theabsence of Mus81 Thus at 120min the percentage ofreplicated (HL) DNA in fragments 3ndash6 was reducedwith respect to the wild type For example at this timepoint the number of forks that had passed completelyfragments 5ndash6 in the mus81 mutant was about threetimes less than in the control strain (quantification isshown at the bottom of Figure 2DndashE) Later at240min only 43 30 and 20 of forks in mus81 cellshad passed completely fragments 4 5 and 6 respectivelyshowing defects in the replication of the chromosomewhen compared with the MUS81+ strain in which 6357 and 44 of forks passed the same DNA fragmentsTherefore although bulk DNA replication seems similarin MUS81+ and mus81 cells (Figures 1B and 2A) moresensitive assays indicate that chromosome replication in


the presence of DNA damage is perturbed in cells lackingMus81-Mms4 The dense isotope substitution experimentsalso show that the absence of this endonuclease does notcause extensive fork collapse or persistent fork stalling

when cells are treated with MMS In fact forks progressconsiderably along the studied replicon before clearchromosomal replication defects are observed at latetime points when DNA lesions accumulate due to the

MUS81+ mus81Δ

rel SMMon(+noc) (min)

0

20

60

80

0 +MMS 120180

Cel

lvia

bilit

y (

)

24060

40

100

MUS81+

mus81Δ

B

XVIXIII

Chrom

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

I

G1 MMS recovery

C

G1 MMS recovery

D

fractions

HH HL HH HL HH HL HH HL HH HL HH HL

rel

coun

ts

release + MMS

G1 (α factor)

releaseno MMS

r

eplic

atio

n 6 5 4 3 2 1

20

40

60

80100

030 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

MUS81+

Chromosome VI ARS607 10 20 30 40 50 60 70 kb

6 5 4 3 2 1

Fork movement

X

Time in MMS(min)

A

30 60 120 240


rel

coun

ts

fractions

20

40

60

80100

0

6 5 4 3 2 1

E

r

eplic

atio

n

release+ MMS

G1 (α factor)

releaseno MMS

Time in MMS(min) 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

mus81Chromosome VI

ARS607 10 20 30 40 50 60 70 kb

6 5 4 3 2 1 X

Fork movement

20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40

Figure 2 Mus81-Mms4 is required for successful chromosome replication in the presence of DNA damage (AndashC) PFGE analysis MUS81+

(W303-1a strain) and mus81 (YMV39 strain) cells were treated in S-phase with MMS (0033) for 60min The MMS was then removed andthe cells progressed through S-phase in medium with nocodozale (A) Cell cycle progression was monitored by flow cytometry (B) Cell viabilitythroughout the experiment (C) Ethidium-bromidendashstained pulse-field gel Chromosomes are labelled with Roman numerals (DndashE) Density transferanalysis MUS81+ (YJT110 strain) and mus81 (YMV17 strain) cells were blocked in G1 in medium with heavy isotopes and released then inmedium with light isotopes plusmnMMS (0033) RF progression in MUS81+ (D) and mus81 (E) cells was followed in a replicon of chromosome VIusing DNA probes recognizing the ClaISalI fragments 1ndash6 Probe numbers correspond to fragment numbers The relative amounts of radioactivityin the hybridized DNA are plotted against the gradient fraction number The positions of unreplicated (HH) and fully replicated (HL) peaks areindicated The position of the initial HH peak is shown for comparison (grey area) Quantification of replication in the presence of MMS for everyfragment and time point is shown at the bottom


continued exposure to MMS Nevertheless these replica-tion problems due to the absence of Mus81 are importantenough to account for the incompletion of chromosomereplication observed by PFGE (Figure 2C) as well as forthe reduced viability of mus81 cells under the assayedDNA-damaging conditions (Figures 1C and 2B)

Mus81-Mms4 is not required to resume DNA replicationafter replicative stress caused by acute treatment with HU

We next asked whether Mus81-Mms4 is also necessary tocope with other kinds of DNA replication perturbationsthat like MMS cause RF stalling but unlike this drugnot necessarily DNA damage Thus we studied thepossible involvement of this endonuclease in replicationresumption after replicative stress originated in a singleS-phase by the ribonucleotide reductase inhibitor HUThere are some apparently contradictory data on therole of Mus81-Eme1Mms4 in facilitating DNA replica-tion after fork stalling caused for HU For example mam-malian Mus81 causes double-strand breaks (DSB) afterprolonged exposure to replicative stress induced by HU(6263) and a role for this protein in RF restart after HUtreatment has been proposed (62) However while mouseES cells lacking Mus81 increase their sensitivity to HU(62) MUS81-depleted U2OS cells exhibit increased resist-ance to HU (63) On the other hand although S pombeand S cerevisiae cells lacking Mus81 show sensitivity tochronic treatment with HU (132930) S pombe mus81cells are not significantly sensitive to acute exposure toHU (17) and Mus81 does not have an apparent role inthe resumption of DNA replication after treatment withthis drug in fission yeast (17) These diverse data couldlikely reflect distinct biological responses in differentsystems to the same challenge or could be the conse-quence of different experimental conditions that lead todifferent results Therefore we decided to analyse therequirement of budding yeast Mus81-Mms4 for the com-pletion of chromosome replication after early S-phaseblock caused by acute treatment with HU

MUS81+ and mus81 cells were arrested in G1 with afactor and then released into S-phase in medium contain-ing 02M HU It is well established that under the assayedconditions DNA replication initiates from early originsand forks stall within 10 kb from an origin (64) After90min the cells were released into medium without HUand allowed to progress through S-phase Flow cytometry(Figure 3A) shows that HU-blocked cells remained with aDNA content close to 1C More than 95 of these cellswere budded confirming release from G1 After removingHU MUS81+ and mus81 cells progressed throughS-phase reached the 2C DNA peak similarly and starteda new cell cycle indicating that DNA replication andmitosis proceeded normally Moreover both control andmutant cells maintained high viability after HU treatment(Figure 3B) These results suggest that the Mus81-Mms4endonuclease is not necessary for DNA resumption afterblocking forks with HU The same results were obtainedeven if the HU treatment was extended (SupplementaryFigure S1) To discard that the completion of DNA rep-lication after HU block in the absence of Mus81 was just

due to new replication initiation events that could rescueremaining stalled forks we monitored the activation of theRad53 checkpoint protein (Figure 3C) As shown in theimmunoblot Rad53 appeared hyperphosphorylated whenMUS81+ and mus81 cells were treated with HU whichcorrelated with its kinase activity (Figure 3C) In bothcases when HU was removed Rad53 became dephospho-rylated and consistently no kinase activity was observedIf there were remaining blocked or collapsed forks owingto the absence of Mus81-Mms4 after HU treatment thecheckpoint would continue to be activated which in turnwould block new replication initiation from late replica-tion origins (64) However this was not the case as Rad53did not remain activated once HU was removedTherefore although new replication initiation is expectedto occur in the experiment this result reinforces the dataabove indicating that the absence of Mus81-Mms4 doesnot impede normal fork resumption after HU treatmentTo show more precisely that Mus81-Mms4 was not

required for the response to fork stalling caused by HUwe analysed chromosomes by PFGE after treatingMUS81+ and mus81 cells with this drug Both strainswere blocked in G1 and then released into S-phase inmedium with HU which was subsequently removedNocodazole was added to stop cells in G2M Flowcytometry (Figure 3D upper panel) showed that bothcontrol and mutant cells reached a 2C DNA contentafter release from the HU block indicating that bulkDNA synthesis had finished PFGE (Figure 3D lowerpanel) showed that intact chromosomes from G1-blocked cells were separated as discrete bands In bothstrains chromosomal DNA did not enter the gel afterHU treatment indicating ongoing replication In bothMUS81+ and mus81 cells the chromosomes re-enteredthe gel after 60min recovery in medium without HU andappeared as discrete bands with no differences betweenstrains showing that they recovered from the HU blockand finished replicationThus completely different to the situation in which the

DNA is damaged with MMS Mus81-Mms4 is notrequired for the completion of chromosome replicationwhen RFs stall due to acute treatment with HU Thisstrongly suggests that under these conditions at least inbudding yeast Mus81-Mms4 is not involved in thecleavage of stalled forks that could facilitate recombin-ation-dependent fork restart

Mus81-Mms4 activation occurs when cells finish S-phaseeven in the presence of DNA lesions that interfere withchromosome replication

The results presented in this work indicate that Mus81-Mms4 is necessary for chromosome replication underDNA-damaging conditions At first glance this mightseem paradoxical as the regulation of Mus81-Mms4through Mms4 phosphorylation restrains its normalfunction to a narrow period of the cell cycle just beforechromosome segregation and maintains the nucleaseactivity low during S-phase (4243) However Mus81-Mms4 regulation was previously only studied in an unper-turbed cell cycle in the absence of DNA damage This


raises the question of whether DNA lesions that interferewith chromosome replication and reduce cell survival inthe absence of Mus81-Mms4 can induce or modifysomehow the regulation of this endonucleaseTo answer this question we first analysed Mms4 phos-

phorylation in the presence of DNA damage andcompared it with its previously known modification inan unperturbed cell cycle (Figure 4AndashC) HA-MMS4cells were synchronized in G1 and then released inmedium either without or with MMS (Figure 4AndashC)The cells progressed through S-phase and finished thecell cycle in medium without MMS (Figure 4A) Theimmunoblot (Figure 4A) shows that Mms4 underwentcell cyclendashdependent phosphorylation as shown before(4243) Thus Mms4 was hyperphosphorylated at 45minafter release from G1 when the cells reached a 2C DNAcontent as shown by an electrophoretic mobility shift andbecame dephosphorylated later when they progressedthrough a new cell cycle However in medium withMMS (Figure 4B) Mms4 did not present changes detect-able by immunoblotting even after 120min in thepresence of MMS a treatment that would have caused areduction in cell viability and affected DNA replication inthe absence of Mus81-Mms4 (Figures 1 and 2) The lack

of visible Mms4 phosphorylation suggests that Mus81-Mms4 remained with low activity during S-phasedespite the presence of DNA damage Finally cells werereleased from G1-arrest in medium with MMS for 60minand the MMS was then removed so that the culture couldrecover and progress through the cell cycle (Figure 4C)Phosphorylation of Mms4 was detected 75min afterremoving the MMS when most cells had a 2C DNAcontent (Figure 4C) suggesting that Mus81-Mms4becomes active only at the end of S-phase or in G2Mdespite the previous presence of damaged DNA Togetherthese results indicate that Mms4 phosphorylation is notinduced by DNA damage and that only occurs when cellsfinish S-phase

To confirm directly that DNA lesions during S-phase donot induce Mus81-Mms4 activation we performednuclease assays (Figure 4DndashF) An HA-MMS4 culturewas blocked in G1 and split into three parts one partwas released in medium without MMS for 20min (cellsin S-phase no DNA damage) another one in mediumwithout MMS but containing nocodazole for 60min(cells in G2M no DNA damage) and the third one inmedium with MMS for 60min (cells in S-phase DNAdamage) In the latter the MMS was subsequently

I

XVIXIII

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

Chrom HU

mus81ΔMUS81+

G1

recovery

G1 HU recovery

0

50

100MUS81+ mus81Δ

Cel

lvia

bilit

y (

)

t=0HU t=0HU

B D

Rad53

HU

Immunoblotmus81ΔMUS81+

G1 (αF) HU Rel-HU G1

(αF) Rel-HU

Rad53Rad53-P

C

Loading lortnoc(Pounceau S)

HU

In situ kinase assaymus81ΔMUS81+


(αF) Rel-HU

A

2C 1C 2C 1C

MUS81+ mus81

Log

G1 ( F)

30

60

90 120

rel ndash

HU

(min

)

S phase+HU

rel -

HU

(+ n

oc)

(m

in)

MUS81+ mus81

2C 1C 2C 1C Log

G1 ( F)

30

60

90 120

S phase+HU

Figure 3 Mus81-Mms4 is not required to resume replication after an HU block (A) MUS81+ (W303-1a strain) and mus81 (YMV39 strain) cellswere blocked in G1 released into medium with 02M HU for 90min and then allowed to progress through the cell cycle in medium without thedrug Samples were taken at the indicated time points and the DNA content was analysed by flow cytometry (B) Cell viability after HU treatment(C) Immunoblot analysis of Rad53 (upper panel) and in situ kinase assay for Rad53 (bottom panel) (D) PFGE analysis The cells were blocked inG1 released into medium containing 02M HU for 90min and then released from the HU block in medium with nocodazole Cell cycle progressionwas monitored by flow cytometry (upper panel) The ethidium-bromidendashstained pulse-field gel is shown at the bottom Chromosomes are labelledwith Roman numerals


removed and the cells were allowed to progress for120min to reach the 2C DNA content All cultures weremonitored by flow cytometry (Figure 4D) and buddingindex estimation As shown in the immunoblot(Figure 4E) Mms4 was not hyperphosphorylated in cellsin S-phase regardless of having been treated with MMSin agreement with data in Figure 4AndashC However whenMMS-treated cells were released from the drug andallowed to reach the end of S-phase Mms4 was hyperpho-sphorylated consistent with Figure 4C and like inG2M-blocked cells that were used as a positive controlIn each case HA-Mms4 was immunoprecipitated fromthe cell extracts (Figure 4E) and the nuclease activitywas tested (Figure 4F) For the assays we used two

different 32P-labelled synthetic structures that are sub-strates of Mus81-Mms4 and are generated during DNAreplication or DNA replication-associated repair a modelRF and a 30-FL Mms4 is a non-catalytic subunit but itwas shown that the immunoprecipitation (IP) of Mms4from extracts yields nuclease activity and therefore theMus81-Mms4 complex immunoprecipitates and is func-tional under these conditions (65) Figure 4F shows thatthe nuclease activity of Mus81-Mms4 determined by theappearance of clear labelled products was only high whenMms4 was immunoprecipitated from extracts obtainedfrom cells that had finished S-phase after MMS treatmentThis activity was similar to that of Mus81-Mms4 takenfrom G2M arrested cells that had not been treated

E

HA-Mms4

[

IP

WCE

HA-Mms4

[

ImmunoblotD

F IPs

HA-Mms4

Nuclease assay

RF

IPsHA-Mms4

Nuclease assay

G1 (αF) 15 30 45 60 90 105 120 75

release + MMS (min)

HA-Mms4-P HA-Mms4

B


C

G1 (αF) +M

MS

15 30 45 60 90 105 120 75

no MMS (min)

HA-Mms4-P HA-Mms4


G1 (αF) 15 30 45 60 90 105 120 75

release no MMS (min)

HA-Mms4-P HA-Mms4


A

Figure 4 Mus81-Mms4 activation is not induced by DNA damage and occurs when cells finish S-phase (A) HA-MMS4 cells (YMV33 strain) weresynchronized in G1 and then released into fresh medium Cell cycle progression was monitored by flow cytometry (left) Mms4 phosphorylation wasanalysed by immunoblot (right) (B) HA-MMS4 cells were blocked in G1 and then released into medium with MMS (002) Left Flow cytometryRight Immunoblot analysis of Mms4 (C) HA-MMS4 cells were synchronized in G1 and then released into S-phase in medium with MMS (002)for 60min The MMS was then removed and the cells were allowed to progress through S-phase Left Flow cytometry Right Immunoblot analysisof Mms4 (D) Cell cycle experiments for the nuclease assays HA-MMS4 cells were blocked in G1 and the culture was then divided in three Oneculture was released in medium without MMS for 20min another one in medium without MMS plus nocodazole for 60min the third one wasreleased into S-phase for 60min in medium with MMS (002) The MMS was then removed and the cells were allowed to progress through the cellcycle Cell cycle progression was followed by flow cytometry (E) The extracts were prepared from cells taken at the indicated time points and HA-Mms4 was immunoaffinity purified Mms4 was analysed by immunoblot in the whole cell extracts (WCE) and the yield of the IP was estimatedAbout 2 of the total amount of the IP protein used for the nuclease assays was loaded (F) The nuclease activity was assayed by the resolution of amodel RF (left panel) and a 30-FL (right panel) An arrow indicates the labelled product resulting from the cleavage of each substrate The controlswere a reaction without extract and a nuclease assay using IP-extracts from untagged cells blocked in G2M


previously with DNA damage However when Mms4 wasimmunoprecipitated from S-phase cells treated withMMS or from untreated S-phase cells the nucleaseactivity was low These results show that althoughMus81-Mms4 is necessary for cells to cope with DNAlesions originated during S-phase by MMS DNAdamage does not induce the activation of the nucleasewhose full activity is restrained until cells finish bulkgenome replication like in an unperturbed cell cycle

The absence of Sgs1 or Yen1 does not have an effect onMus81-Mms4 regulation

As mentioned previously Mus81-Mms4 functionallyoverlaps with Sgs1-Top3-Rmi1 and is required for cell via-bility in the absence of this RecQ-helicase complex(131524ndash26) It is possible that most common substratesfor Mus81-Mms4 and Sgs1-Top3-Rmi1 are resolved by dis-solution by the latter but that they need to be processed byMus81 when the RecQ complex is not functional In factsome DNA intermediates that persist after MMS treatmentin Rmi1-deficient cells or in the absence of Sgs1 can beremoved by Mus81-Mms4 (6644) Likewise the sensitivityof sgs1 cells to some DNA-damaging agents like MMSincreases when Mus81-Mms4 is not fully active (43)Mus81-Mms4 has also overlapping functions with theYen1 resolvase during DNA repair and cells lacking bothMus81 and Yen1 show hypersensitivity to DNA-damagingdrugs (29ndash32) Due to these functional links we askedwhether the absence of Sgs1-Top3-Rmi1 and Yen1 has aneffect on Mus81-Mms4 regulation as it was possible thatthis endonuclease could be required under these patho-logical circumstances at times at which normally is notTo address this question sgs1HA-MMS4 cells were

blocked in G1 and released into medium without orwith MMS (Figure 5A) Flow cytometry indicates thatthese cells progressed normally through S-phase in theabsence of the drug and slowly when exposed to MMSThe immunoblot shows that in the absence of MMS thecell cyclendashdependent phosphorylation of Mms4 was like inSGS1+ cells (Figure 4A) Likewise like in SGS1+ cells(Figure 4B) Mms4 did not show an electrophoreticmobility shift when the sgs1 cells were treated withMMS in S-phase We also analysed Mms4 phosphoryl-ation under the same conditions in sgs1yen1HA-MMS4 cells (Figure 5B) G1-blocked cells were releasedinto medium without or with MMS These cells pro-gressed in S-phase similarly to SGS1+YEN1+ (Figure 4Aand B) and sgs1 cells (Figure 5A) both in the absence orthe presence of MMS The immunoblot (Figure 5B) showsthat when the sgs1yen1 cells were not treated withMMS Mms4 phosphorylation occurred as inSGS1+YEN1+ cells (Figure 4A) whereas Mms4 did notpresent electrophoretic mobility changes indicative ofphosphorylation when the cells were exposed to MMS inS-phase These results indicate that the absence of Sgs1 orYen1 does not cause changes in the dynamics of activationof this endonuclease even in the presence of DNA damageduring chromosome replication Moreover these data alsoshow that the hyperphosphorylation of Mms4 does notincrease in budding yeast cells lacking Sgs1 with respect

to SGS1+ cells (compare Figures 4A and 5A) unlikeEme1Mms4 phosphorylation in S pombe (46) Mms4 phos-phorylation does not increase either in the absence of bothSgs1 and Yen1 (compare Figures 4A and 5B)

Next to confirm the above data we assayed Mus81-Mms4 nuclease activity (Figure 5CndashE) The sgs1HA-MMS4 and sgs1yen1HA-MMS4 cells were arrested inG1 the cells were then released from the block for 60mineither in the presence of MMS (cells in S-phase DNAdamage) or in medium without MMS plus nocodazole(cells in G2M no DNA damage) The cultures were moni-tored by flow cytometry (Figure 5C) and microscopy Theimmunoblots (Figure 5D) show that Mms4 was nothyperphosphorylated in the extracts of S-phase cellstreated with MMS in agreement with data in Figure 5AandB On the contraryMms4was hyperphosphorylated inboth strains in cells in G2M In each case HA-Mms4 wasimmunoprecipitated from the cell extracts (Figure 5D) andthe nuclease activity was analysed Figure 5E shows that insgs1 and sgs1yen1 strains the nuclease activity ofMus81-Mms4 was low when the complex was immunopre-cipitated from S-phase cells treated with MMS Howeverwhen Mms4 was immunoprecipitated from G2M-cellsMus81-Mms4 activity was more robust as shown by theappearance of significant amount of labelled products fromthe substrates used Therefore the mode of regulation ofMus81-Mms4 is not apparently altered when cells lack Sgs1or Yen1 even in the presence of damaged DNA during S-phase Thus even when the function of Mus81-Mms4becomes more important to respond to DNA damagedue to the absence of Sgs1 (43) the full activity of the endo-nuclease is still prevented until bulk DNA synthesis iscompleted

Mus81-Mms4 functions after completion of bulk DNAreplication to respond to DNA damage present in S-phase

As Mus81-Mms4 activation occurs only at the end ofS-phase despite the previous originated DNA damagewe reasoned that the problems caused by MMS duringreplication in the absence of this endonuclease should bereversed by new expression of the complex before the cellswere plated for viability analysis To test this hypothesiswe made a strain in which the endogenous MUS81 andMMS4 were placed under control of the GAL1-10promoter to allow their conditional expression To char-acterize this strain the sensitivity of PGAL1-10-MUS81PGAL1-10-MMS4 cells to MMS was analysed in mediumwith glucose or galactose Figure 6A shows that these cellsgrew normally in medium without MMS and behaved likea mus81mms4 mutant in medium with MMS andglucose (GAL-110 promoter OFF) However inmedium with galactose (GAL-110 promoter ON) thesecells increased notably their resistance to MMS withrespect to mus81mms4 This result indicates that theinduced expression of Mus81 and Mms4 in the same strainreconstitutes a functional complex at least to an extentthat allows our experimental approach

To analyse the reversibility of the problems caused byMMS in S-phase cells lacking Mus81-Mms4 PGAL1-10-MUS81 PGAL1-10-MMS4 cells were grown in medium


with raffinose (GAL1-10 promoter inactive) synchronizedin G1 and then released into medium containing raffinoseeither without or with MMS After 90min the culturecontaining MMS was divided one part was held inmedium with raffinose plus MMS in the others theMMS was removed and the cells were transferred tomedium either with raffinose or with galactose The induc-tion with galactose was also carried out 3 and 4 h aftereliminating the MMS (a scheme of the experiment isdrawn in Figure 6B) Cell cycle progression wasfollowed by flow cytometry (Figure 6C) and the expres-sion of Mus81 and Mms4 was monitored by immunoblot(Figure 6D) Cell viability was estimated at the indicatedtime points (Figure 6E)

Cells lacking Mus81 and Mms4 progressed normallythrough several cell cycles after release from the G1block in medium without MMS (Figure 6C panel 1) Incontrast these cells progressed slowly through S-phase inthe presence of MMS (Figure 6C panel 2) which was

accompanied by a high loss of viability (Figure 6E)Thus PGAL1-10-MUS81 PGAL1-10-MMS4 cells under re-pressive conditions for the GAL1-10 promoter behavedsimilarly to a mus81mms4 strain in a medium withglucose (Figure 1) After MMS removal in the continuedabsence of Mus81 and Mms4 the cells progressed throughS-phase and reached the 2C DNA content (Figure 6Cpanel 3) However the cells did not recover viabilityand even 8 h after removing the MMS did notcomplete the cell cycle When PGAL1-10-MUS81 PGAL1-10-MMS4 cells were transferred to galactose medium lackingMMS (Figure 6C panel 4) Mus81 and Mms4 wereinduced within an hour (Figure 6D) At different timeson release into galactose medium lacking MMS cell via-bility was monitored on YPGal medium (SupplementaryFigure S2A) or on YPD medium that repressed expres-sion of Mus81 and Mms4 once again (Mus81 and Mms4were degraded rapidly under such conditions as shown inSupplementary Figure S2B) Importantly the viability

A

C

D

E

B

Figure 5 Mus81-Mms4 regulation is not modified in cells lacking Sgs1 or Yen1 (A) sgs1HA-MMS4 cells (YMG21 strain) were synchronized inG1 and then released into fresh medium plusmnMMS (002) Cell cycle progression was monitored by flow cytometry (left) Immunoblot analysis ofMms4 (right) (B) G1-blocked sgs1yen1HA-MMS4 cells (YSG56 strain) were released into fresh medium plusmnMMS (002) Left Flow cytometryRight Immunoblot analysis of Mms4 (C) Cell cycle experiments for the nuclease assays Cells were synchronized in G1 and released in medium withMMS (002) for 60min or in medium without MMS (plus nocodazole) (D) The extracts were prepared from cells taken at the indicated timepoints and HA-Mms4 was immunoaffinity purified Mms4 was analysed by immunoblot in the whole cell extracts (WCE) and the yield of the IP wasestimated About 2 of the total amount of the IP protein used for the nuclease assays was loaded (E) The nuclease activity was assayed by theresolution of a model RF and a 30-FL The arrows indicate the labelled products resulting from the cleavage of each substrate The controls were areaction without extract and a nuclease assay using IP-extracts from untagged cells blocked in G2M


was only restored when Mus81 and Mms4 were expressedseveral hours after release from MMS around the timethat cells reached the end of S-phase (Figure 6C panel 4)This strongly suggests that Mus81-Mms4 only carries outits function in the response to DNA damage after bulkDNA synthesisTo further demonstrate that Mus81-Mms4 works after

bulk DNA replication to respond to DNA damage presentin S-phase its expression was induced when the cells had a2C DNA content rather than just after eliminating theMMS Thus 3 and 4 h after MMS removal (Figure 6Cpanels 5 and 6 respectively) part of the culture growing in

raffinose medium was transferred to medium with galact-ose The immunoblot (Figure 6D) shows that bothsubunits were expressed within an hour Flow cytometry(Figure 6C) and microscopy analysis indicated that inboth cases cells finished S-phase and started dividingRegardless of the time of Mus81-Mms4 induction afterMMS removal cell viability was significantly restored(Figure 6E) Moreover the recovery of viabilityoccurred progressively but started as soon as the expres-sion of Mus81-Mms4 was induced unlike the case when itwas expressed just after eliminating the MMS Likely thisis because now the cells were already at the end of S-phase

A

B

C

D E

Figure 6 Mus81-Mms4 functions after bulk DNA synthesis to respond to DNA damage originated in S-phase (A) Sensitivity to MMS Serialdilutions (10-fold) of normalized log-phase cultures were spotted onto YP-Glucose or YP-Galactose plates containing the indicated amounts of MMSand incubated at 30C for 48ndash72 h MUS81+MMS4+ (W303-1a strain) mus81 mms4 (YMV49 strain) 3HA-MUS81 3HA-MMS4 (YSG24 strain)GAL-3HA-MUS81 GAL-3HA-MMS4 (YSG23 strain) (B) Scheme of the experiment Numbers on the right correspond to those on the flowcytometry panels and in the graphic of cell viability (C) Cell cycle progression was followed by flow cytometry Roman numerals indicate thetime points at which protein samples were taken for immunoblot analysis (D) (E) Cell viability analysis


when Mus81-Mms4 executes its function to respond toDNA damage according to our data

Together these results indicate that the consequences ofthe absence of the Mus81-Mms4 endonuclease during rep-lication of a damaged DNA template are largely reversibleby new expression of this complex Furthermore thesedata show that Mus81-Mms4 operates after bulk DNAsynthesis to allow cells to cope with DNA damage thatoriginated during chromosomal replication consistentwith the results shown in Figures 4 and 5 It thusappears that Mus81-Mms4 function is temporally separ-able from bulk genome replication

DISCUSSION

We have shown in this work that the structure-specificMus81-Mms4 endonuclease has a key role in the success-ful completion of chromosome replication under condi-tions of DNA damage Furthermore we have foundthat this crucial function for cell survival is strictlycontrolled and that Mus81-Mms4 responds to DNAlesions that occur during S-phase and hinder RF progres-sion after the bulk of genome replication has beenfinished

Our data indicate that the problems for finishing repli-cation in DNA-damaged cells lacking Mus81-Mms4 arenot due to massive fork collapse or lasting and extendedfork stalling Instead they more likely reflect the result ofrelatively infrequent events Nevertheless these are quali-tatively important as they account for the incompletion ofchromosome replication and the high loss of viability inmus81 mutants after treatment with a DNA-damagingagent like MMS in a single S-phase

Our results also show that budding yeast Mus81-Mms4is not required for coping with every kind of perturbationthat causes fork block We have shown that Mus81-defi-cient cells resume DNA replication normally after forkstalling originated by HU without loss of viabilityTherefore it is unlikely that at least in S cerevisiaeMus81-Mms4 is involved in the cleavage of stalledforks to initiate recombination-dependent fork restartThis is consistent with the fact that recombination is notnecessary to complete S-phase after HU arrest (67)Furthermore in agreement with these data Mus81-Mms4 shows low activity in cells blocked with HU (43)and in S pombe Mus81 is displaced from chromatin afterHU treatment in a Cds1-dependent manner (17) Theseresults are apparently in contradiction with data frommammalian cells showing that Mus81-Eme1 cleavesstalled forks following HU exposure (6263) which isproposed to allow fork restart (62) The different resultsmay reflect distinct responses to stalled RFs between yeastand higher eukaryotes They may also likely indicate dif-ferences between an acute short HU treatment in ourexperiments and prolonged HU exposure in those withmammalian cells In fact long but not short HU treat-ment can increase the amount of DNA damage and fork-associated DSB (6869) Mus81-dependent fork cleavageafter HU exposure has also been found under pathological

situations (70ndash72) but these are not present in our experi-mental approachMMS causes DNA lesions that trigger fork pausing

(54617374) The fact that Mus81-Mms4 is not requiredfor restart after fork stalling caused by HU makes alsounlikely that its role to respond to MMS-induced lesionsduring replication involves fork cleavage Instead it ismore probable a function for Mus81-Mms4 in processingDNA intermediates that originate during DNA replica-tion or DNA replication-associated repair the resolutionof which is necessary for finishing chromosome duplica-tion Several pathways including base excision repairhomologous recombination and post-replication repairare required for replication through DNA lesionsproduced by MMS (61) These pathways generate poten-tial Mus81-Mms4 substrates during the removal of DNAdamage some of which need probably to be resolved atsome point by this endonuclease to allow completion ofDNA repair and chromosome replicationThe data presented in this work indicate that DNA

lesions produced during S-phase do not modify themode of activation of Mus81-Mms4 that occurs in anormal cell cycle Mms4 phosphorylation and acquisitionof full endonuclease activity in cells treated with MMSduring S-phase remain restrained until they finish bulkgenome replication and therefore the regulation ofMus81-Mms4 under DNA-damaging conditions thatinterfere with DNA replication is like in an unperturbedcell cycle (4243) Moreover the absence of the RecQ-helicase Sgs1 or the Yen1 resolvase proteins with whichMus81-Mms4 functionally interacts does not have anapparent effect on Mus81-Mms4 regulation Thereforethe regulation of Mus81-Mms4 through Mms4 phosphor-ylation is independent of the presence of DNA damageduring chromosome replication even if the number ofDNA lesions could accumulate due to pathological situ-ations originated by the absence of proteins with which ithas overlapping functions A recent work has shown thatin S pombe Eme1Mms4 phosphorylation is increased inG2 by damage induced by camptothecin or bleomycinwhich is accompanied by enhanced activity of Mus81-Eme1 (46) However our work indicates that MMS doesnot induce or modify the phosphorylation of buddingyeast Mms4 and therefore the activity of Mus81-Mms4Furthermore in budding yeast the treatment with otherDNA-damaging drugs during S-phase or in G2M doesnot have a detectable effect on the cell cyclendashdependentphosphorylation of Mms4 (Supplementary Figure S3)These results together with the fact that Eme1-phosphor-ylation in S pombe is Plk1-independent (46) unlike Mms4in budding yeast and human cells (42ndash44) clearly showdifferences in the mode of regulation of Mus81-Eme1Mms4 among different organisms They reflect distinctsuccessful evolutionarily strategies and make importantthe study of these mechanisms in different systemsOur results have indicated that the function of Mus81-

Mms4 during the response to DNA damage is onlyexecuted after the bulk of genome replication iscompleted Supporting this conclusion we have addition-ally shown that new expression of Mus81-Mms4 signifi-cantly restores cell viability that is lost in the absence of


this complex after MMS treatment and that the functionof the endonuclease is only carried out when the cells areat the end of S-phase regardless of when the complex isnewly synthesized The post-replicative function ofMus81-Mms4 has a number of advantages that help tomaintain genome stability Thus preventing its nucleolyticaction during S-phase avoids the counterproductivecleavage of DNA structures during chromosome replica-tion like RFs Furthermore the low Mus81-Mms4activity during S-phase even when the DNA isdamaged avoids the undesired cleavage of DNA inter-mediates during DNA repair that can lead to chromo-somal rearrangements and high levels of sister chromatidexchange Indirectly the absence of high Mus81-Mms4activity in S-phase favours the action of the non-nucleolytic resolution pathway mediated by Sgs1-Top3-Rmi1 which does not originate crossovers or potentialchromosome reorganizations Likewise timely activationof Mus81-Mms4 provides a safeguard mechanism thatguarantees just before chromosome segregation the reso-lution of DNA intermediates that cannot be resolved byother proteins and need to be processed for proper com-pletion of DNA repair and chromosome replication Thisnucleolytic action can still put cells at some risk ofgenomic instability but it constitutes the necessary lastresort to resolve persistent intermediates that in anycase under non-pathological situations should not behigh in number Finally as the Mus81-Mms4 function isuncoupled from bulk genome replication it is alsoexpected that the proposed role of Mus81-Eme1Mms4in post-replication gap repair (11) is facilitatedAs the key aspects of the regulation of Mus81-Eme1

Mms4 in an unperturbed cell cycle seem to be conservedbetween budding yeast and human mitotic cells (42) it islikely that the main results presented in this work can beextrapolated to human cells suffering DNA damage Thecorrect function of Mus81-Eme1Mms4 is necessary toprevent genomic instability a hallmark of cancer (75)Although it is unclear whether Mus81 is important toavoid tumorigenesis (1920) Mus81-associated defectshave been linked to some types of cancers (7677) andthis endonuclease has been involved in oncogene-inducedgenotoxicity (7879) Therefore it would be interesting tostudy deeper whether the regulation of this endonucleasefails in some tumour cells Future work will be necessaryto explore the possible potential of Mus81-Eme1 as atumour marker and to gain information that may helpin the improvement of anti-cancer therapy

SUPPLEMENTARY DATA

Supplementary Data are available at NAR Onlineincluding [43474861]

ACKNOWLEDGEMENTS

We thank Crisanto Gutierrez for critical reading of themanuscript and Jose Perez-Martın for facilitating the useof the Gene Navigator System for PFGE

FUNDING

Spanish Ministry of Economy and Competitiveness[BFU2010-16989 and Consolider Ingenio CSD2007-00015 to JAT] Fundacion Ramon Areces(Institutional Grant to the Centro de Biologıa MolecularSevero Ochoa) Spanish Ministry of Economy andCompetitiveness (predoctoral fellowships to MVV andMAO-B) Universidad Autonoma de Madrid(predoctoral fellowship to MG-F) Consejo Superior deInvestigaciones Cientıficas (JAE-Doc contract to MS)Funding for open access charge Spanish Ministry ofEconomy and Competitiveness [BFU2010-16989 andConsolider Ingenio CSD2007-00015]

Conflict of interest statement None declared

REFERENCES

1 BranzeiD and FoianiM (2008) Regulation of DNA repairthroughout the cell cycle Nat Rev Mol Cell Biol 9 297ndash308

2 FriedelAM PikeBL and GasserSM (2009) ATRMec1coordinating fork stability and repair Curr Opin Cell Biol 21237ndash244

3 SeguradoM and TerceroJA (2009) The S-phase checkpointtargeting the replication fork Biol Cell 101 617ndash627

4 ZegermanP and DiffleyJF (2009) DNA replication as a targetof the DNA damage checkpoint DNA Repair (Amst) 81077ndash1088

5 AguileraA and Gomez-GonzalezB (2008) Genome instability amechanistic view of its causes and consequences Nat RevGenet 9 204ndash217

6 MartiTM and FleckO (2004) DNA repair nucleases Cell MolLife Sci 61 336ndash354

7 NishinoT IshinoY and MorikawaK (2006) Structure-specificDNA nucleases structural basis for 3D-scissors Curr OpinStruct Biol 16 60ndash67

8 CicciaA McDonaldN and WestSC (2008) Structural andfunctional relationships of the XPFMUS81 family of proteinsAnnu Rev Biochem 77 259ndash287

9 SchwartzEK and HeyerWD (2011) Processing of jointmolecule intermediates by structure-selective endonucleases duringhomologous recombination in eukaryotes Chromosoma 120109ndash127

10 RouseJ (2009) Control of genome stability by SLX proteincomplexes Biochem Soc Trans 37 495ndash510

11 OsmanF and WhitbyMC (2007) Exploring the roles of Mus81-Eme1Mms4 at perturbed replication forks DNA Repair (Amst)6 1004ndash1017

12 XiaoW ChowBL and MiloCN (1998) Mms4 a putativetranscriptional (co)activator protects Saccharomyces cerevisiaecells from endogenous and environmental DNA damage MolGen Genet 257 614ndash623

13 BoddyMN Lopez-GironaA ShanahanP InterthalHHeyerWD and RussellP (2000) Damage tolerance proteinMus81 associates with the FHA1 domain of checkpoint kinaseCds1 Mol Cell Biol 20 8758ndash8766

14 InterthalH and HeyerHD (2000) MUS81 encodes a novelHelix-hairpin-Helix protein involved in the response to UV- andmethylation-induced DNA damage in Saccharomyces cerevisiaeMol Gen Genet 263 812ndash827

15 DoeCL AhnJS DixonJ and WhitbyMC (2002) Mus81-Eme1 and Rqh1 involvement in processing stalled and collapsedreplication forks J Biol Chem 277 32753ndash32759

16 Bastin-ShanowerSA FrickeWM MullenJR and BrillSJ(2003) The mechanism of Mus81-Mms4 cleavage site selectiondistinguishes it from the homologous endonuclease Rad1-Rad10Mol Cell Biol 23 3487ndash3496

17 KaiM BoddyMN RussellP and WangTS (2005)Replication checkpoint kinase Cds1 regulates Mus81 to preserve


genome integrity during replication stress Genes Dev 19919ndash932

18 AbrahamJ LemmersB HandeMP MoynahanMEChahwanC CicciaA EssersJ HanadaK ChahwanRKhawAK et al (2003) Eme1 is involved in DNA damageprocessing and maintenance of genomic stability in mammaliancells EMBO J 22 6137ndash6147

19 McPhersonJP LemmersB ChahwanR PamidiA MigonEMatysiak-ZablockiE MoynahanME EssersJ HanadaKPoonepalliA et al (2004) Involvement of mammalian Mus81 ingenome integrity and tumor suppression Science 304 1822ndash1826

20 DendougaN GaoH MoecharsD JanicotM VialardJ andMcGowanCH (2005) Disruption of murine Mus81 increasesgenomic instability and DNA damage sensitivity but does notpromote tumorigenesis Mol Cell Biol 25 7569ndash7579

21 HanadaK BudzowskaM ModestiM MaasA WymanCEssersJ and KanaarR (2006) The structure-specificendonuclease Mus81-Eme1 promotes conversion of interstrandDNA crosslinks into double-strands breaks EMBO J 254921ndash4932

22 RoseaulinL YamadaY TsutsuiY RussellP IwasakiH andArcangioliB (2008) Mus81 is essential for sister chromatidrecombination at broken replication forks EMBO J 271378ndash1387

23 Munoz-GalvanS TousC BlancoMG SchwartzEKEhmsenKT WestSC HeyerWD and AguileraA (2012)Distinct roles of Mus81 Yen1 Slx1-Slx4 and Rad1 nucleases inthe repair of replication-born double-strand breaks by sisterchromatid exchange Mol Cell Biol 32 1592ndash1603

24 KaliramanV MullenJR FrickeWM Bastin-ShanowerSAand BrillSJ (2001) Functional overlap between Sgs1-Top3 andthe Mms4-Mus81 endonuclease Genes Dev 15 2730ndash2740

25 MullenJR KaliramanV IbrahimSS and BrillSJ (2001)Requirement for three novel protein complexes in the absence ofthe Sgs1 DNA helicase in Saccharomyces cerevisiae Genetics 157103ndash118

26 FabreF ChanA HeyerWD and GangloffS (2002) Alternatepathways involving Sgs1Top3 Mus81 Mms4 and Srs2 preventformation of toxic recombination intermediates from single-stranded gaps created by DNA replication Proc Natl Acad SciUSA 99 16887ndash16892

27 HartungF SuerS BergmannT and PuchtaH (2006) The roleof AtMUS81 in DNA repair and its genetic interaction with thehelicase AtRECQ4A Nucleic Acids Res 34 4438ndash4448

28 TrowbridgeK McKimK BrillSJ and SekelskyJ (2007)Synthetic lethality of Drosophila in the absence of the MUS81endonuclease and the DmBlm helicase is associated with elevatedapoptosis Genetics 176 1993ndash2001

29 BlancoMG MatosJ RassU IpSC and WestSC (2010)Functional overlap between the structure-specific nucleases Yen1and Mus81-Mms4 for DNA-damage repair in S cerevisiae DNARepair (Amst) 9 394ndash402

30 HoCK MazonG LamAF and SymingtonLS (2010) Mus81and Yen1 promote reciprocal exchange during mitoticrecombination to maintain genome integrity in budding yeastMol Cell 40 988ndash1000

31 TayYD and WuL (2010) Overlapping roles for Yen1 andMus81 in cellular Holliday junction processing J Biol Chem285 11427ndash11432

32 AgmonN YovelM HarariY LiefshitzB and KupiecM(2011) The role of Holliday junction resolvases in the repair ofspontaneous and induced DNA damage Nucleic Acids Res 397009ndash7019

33 BoddyMN GaillardPH McDonaldWH ShanahanPYatesJR 3rd and RussellP (2001) Mus81-Eme1 are essentialcomponents of a Holliday junction resolvase Cell 107 537ndash548

34 ChenXB MelchionnaR DenisCM GaillardPH BlasinaAVan de WeyerI BoddyMN RussellP VialardJ andMcGowanCH (2001) Human MUS81-associated endonucleasecleaves Holliday junctions in vitro Mol Cell 8 1117ndash1127

35 CicciaA ConstantinouA and WestSC (2003) Identificationand characterization of the human Mus81-Eme1 endonucleaseJ Biol Chem 278 25172ndash25178

36 WhitbyMC OsmanF and DixonJ (2003) Cleavage of modelreplication forks by fission yeast Mus81-Eme1 and budding yeastMus81-Mms4 J Biol Chem 278 6928ndash6935

37 OsmanF DixonJ DoeCL and WhitbyMC (2003)Generating crossovers by resolution of nicked Hollidayjunctions a role for Mus81-Eme1 in meiosis Mol Cell 12761ndash774

38 FrickeWM Bastin-ShanowerSA and BrillSJ (2005) Substratespecificity of the Saccharomyces cerevisiae Mus81-Mms4endonuclease DNA Repair (Amst) 4 243ndash251

39 EhmsenKT and HeyerWD (2008) Saccharomyces cerevisiaeMus81-Mms4 is a catalytic DNA structure-selective endonucleaseNucleic Acids Res 36 2182ndash2195

40 TaylorER and McGowanCH (2008) Cleavage mechanism ofhuman Mus81-Eme1 acting on Holliday-junction structures ProcNatl Acad Sci USA 105 3757ndash3762

41 SchwartzEK WrightWD EhmsenKT EvansJEStahlbergH and HeyerWD (2012) Mus81-Mms4 functions as asingle heterodimer to cleave nicked intermediates inrecombinational DNA repair Mol Cell Biol 32 3065ndash3080

42 MatosJ BlancoMG MaslenS SkehelJM and WestSC(2011) Regulatory control of the resolution of DNArecombination intermediates during meiosis and Mitosis Cell147 158ndash172

43 Gallo-FernandezM SaugarI Ortiz-BazanMA VazquezMVand TerceroJA (2012) Cell cycle-dependent regulation of thenuclease activity of Mus81-Eme1Mms4 Nucleic Acids Res 408325ndash8335

44 SzakalB and BranzeiD (2013) Premature Cdk1Cdc5Mus81pathway activation induces aberrant replication and deleteriouscrossover EMBO J 32 1155ndash1167

45 Domınguez-KellyR MartınY KoundrioukoffSTanenbaumME SmitsVAJ MedemaRH DebatisseM andFreireR (2011) Wee1 controls genomic stability duringreplication by regulating the Mus81-Eme1 endonuclease J CellBiol 194 567ndash579

46 DehePM CoulonS ScaglioneS ShanahanP TakedachiAWohlschlegelJA YatesJR III LlorenteB RussellP andGaillardPH (2013) Regulation of Mus81-Eme1 Hollidayjunction resolvase in response to DNA damage Nat Struct MolBiol 20 598ndash603

47 JankeC MagieraMM RathfelderN TaxisC ReberSMaekawaH Moreno-BorchartA DoengesG SchwobESchiebelE et al (2004) A versatile toolbox for PCR-basedtagging of yeast genes new fluorescent proteins more markersand promoter substitution cassettes Yeast 21 947ndash962

48 LongtineMS McKenzieA III DemariniDJ ShahNGWachA BrachatA PhilippsenP and PringleJR (1998)Additional modules for versatile and economical PCR-based genedeletion and modification in Saccharomyces cerevisiae Yeast 14953ndash961

49 LabibK DiffleyJFX and KearseySE (1999) G1-phase andB-type cyclins exclude the DNA-replication factor Mcm4 fromthe nucleus Nat Cell Biol 1 415ndash422

50 FoianiM MariniF GambaD LucchiniG and PlevaniP(1994) The B subunit of the DNA polymerase alpha-primasecomplex in Saccharomyces cerevisiae executes an essential functionat the initial stage of DNA replication Mol Cell Biol 14923ndash933

51 PellicioliA LuccaC LiberiG MariniF LopesM PlevaniPRomanoA Di FiorePP and FoianiM (1999) Activation ofRad53 kinase in response to DNA damage and its effect inmodulating phosphorylation of the lagging strand DNApolymerase EMBO J 18 6561ndash6572

52 McCarrollRM and FangmanWL (1988) Time of replication ofyeast centromeres and telomeres Cell 54 505ndash513

53 TerceroJA (2009) Density transfer as a method to analyze theprogression of DNA replication forks Methods Mol Biol 521203ndash213

54 TerceroJA and DiffleyJF (2001) Regulation of DNAreplication fork progression through damaged DNA by the Mec1Rad53 checkpoint Nature 412 553ndash557

55 LengronneA PaseroP BensimonA and SchwobE (2001)Monitoring S phase progression globally and locally using BrdU


incorporation in TK(+) yeast strains Nucleic Acids Res 291433ndash1442

56 RassU and WestSC (2006) Synthetic junctions as tools toidentify and characterize Holliday junction resolvases MethodsEnzymol 408 485ndash501

57 PaulovichAG and HartwellLH (1995) A checkpoint regulatesthe rate of progression through S phase in S cerevisiae inresponse to DNA damage Cell 82 841ndash847

58 MesnerLD CrawfordEL and HamlinJL (2006) Isolatingapparently pure libraries of replication origins from complexgenomes Mol Cell 21 719ndash726

59 ReynoldsAE McCarrollRM NewlonCS and FangmanWL(1989) Time of replication of ARS elements along yeastchromosome III Mol Cell Biol 9 4488ndash4494

60 TerceroJA LabibK and DiffleyJF (2000) DNA synthesis atindividual replication forks requires the essential initiation factorCdc45p EMBO J 19 2082ndash2093

61 VazquezMV RojasV and TerceroJA (2008) Multiplepathways cooperate to facilitate DNA replication fork progressionthrough alkylated DNA DNA Repair (Amst) 7 1693ndash1704

62 HanadaK BudzowskaM DaviesSL van DrunenEOnizawaH BeverlooHB MaasA EssersJ HicksonID andKanaarR (2007) The structure-specific endonuclease Mus81contributes to replication restart by generating double-strandDNA breaks Nat Struct Mol Biol 14 1096ndash1104

63 FuggerK Kit ChuW HaahrP Nedergaard KousholtABeckH PayneMJ HanadaK HicksonID and StorgaardSorensenC (2013) FBH1 co-operates with MUS81 in inducingDNA double-strand breaks and cell death following replicationstress Nat Commun 4 1423

64 SantocanaleC and DiffleyJFX (1998) A Mec1- and Rad53-dependent checkpoint controls late-firing origins of DNAreplication Nature 395 615ndash618

65 IpSC RassU BlancoMG FlynnHR SkehelJM andWestSC (2008) Identification of Holliday junction resolvasesfrom humans and yeast Nature 456 357ndash361

66 AshtonTM MankouriHW HeidenblutA McHughPJ andHicksonID (2011) Pathways for Holliday junction processingduring homologous recombination in Saccharomyces cerevisiaeMol Cell Biol 31 1921ndash1933

67 DiffleyJF BoussetK LabibK NotonEA SantocanaleCand TerceroJA (2000) Coping with and recovering fromhydroxyurea-induced replication fork arrest in budding yeastCold Spring Harb Symp Quant Biol 65 333ndash342

68 SaintignyY DelacoteF VaresG PetitotF LambertSAverbeckD and LopezBS (2001) Characterization of

homologous recombination induced by replication inhibition inmammalian cells EMBO J 20 3861ndash3870

69 PetermannE OrtaML IssaevaN SchultzN and HelledayT(2010) Hydroxyurea-stalled replication forks become progressivelyinactivated and require two different RAD51-mediated pathwaysfor restart and repair Mol Cell 37 492ndash502

70 FrogetB BlaisonneauJ LambertS and BaldacciG (2008)Cleavage of stalled forks by fission yeast Mus81Eme1 in absenceof DNA replication checkpoint Mol Biol Cell 19 445ndash456

71 FranchittoA PirzioLM ProsperiE SaporaO BignamiMand PichierriP (2008) Replication fork stalling in WRN-deficientcells is overcome by prompt activation of a MUS81-dependentpathway J Cell Biol 183 241ndash252

72 ShimuraT TorresMJ MartinMM RaoVA PommierYKatsuraM MiyagawaK and AladjemMI (2008) Bloomrsquossyndrome helicase and Mus81 are required to induce transientdouble-strand DNA breaks in response to DNA replication stressJ Mol Biol 375 1152ndash1164

73 LundinC NorthM ErixonK WaltersK JenssenDGoldmanAS and HelledayT (2005) Methyl methanesulfonate(MMS) produces heat-labile DNA damage but no detectablein vivo DNA double-strand breaks Nucleic Acids Res 333799ndash3811

74 LarsonK SahmJ ShenkarR and StraussB (1985)Methylation-induced blocks to in vitro DNA replication MutatRes 150 77ndash84

75 NegriniS GorgoulisVG and HalazonetisTD (2010) Genomicinstabilityndashan evolving hallmark of cancer Nat Rev Mol CellBiol 11 220ndash228

76 WuF ShirahataA SakurabaK KitamuraY GotoTSaitoM IshibashiK KigawaG NemotoH SanadaY et al(2010) Down-regulation of Mus81 as a potential markerfor the malignancy of gastric cancer Anticancer Res 305011ndash5014

77 WuF ShirahataA SakurabaK KitamuraY GotoTSaitoM IshibashiK KigawaG NemotoH SanadaY et al(2011) Downregulation of Mus81 as a novel prognosticbiomarker for patients with colorectal carcinoma Cancer Sci102 472ndash477

78 NeelsenKJ ZaniniIM HerradorR and LopesM (2013)Oncogenes induce genotoxic stress by mitotic processing ofunusual replication intermediates J Cell Biol 200 699ndash708

79 MurfuniI NicolaiS BaldariS CrescenziM BignamiMFranchittoA and PichierriP (2013) The WRN and MUS81proteins limit cell death and genome instability followingoncogene activation Oncogene 32 610ndash620


steps of homologous recombination (1624ndash26) whichtogether with its interaction with Rad54 (14) suggested arole for Mus81-Eme1Mms4 during recombination-mediated DNA repair (11) Nevertheless this nuclease isnot required for the repair of double-strand breaksthrough homologous recombination as mus81mms4mutants are not sensitive to g-irradiation (12ndash14)Although it is still unclear which are the substrates ofMus81-Eme1Mms4 in the cell a number of biochemicalstudies have shown that it can cleave different branchedDNA structures in vitro all of which are its potentialtargets in vivo Thus it is able to act with differentaffinity on model RFs 30-flaps (30-FLs) D-loopsX- and Y-shaped structures (15162433ndash41)The Mus81-Eme1Mms4 endonuclease has a significant

role in the maintenance of genome stability but itsnucleolytic activity needs to be strictly regulated toavoid the undesired opposite effect The unrestrainedcleavage of potential substrates such as RFs and otherDNA intermediates would have negative consequencesfor chromosome replication and could also lead to theformation of chromosomal rearrangements or high levelsof chromatid exchanges Recent studies in budding yeasthave uncovered key features of the regulation of this endo-nuclease showing that in an unperturbed mitotic cell cycleits activity is controlled by Cdc28CDK1- and Cdc5PLK1-dependent phosphorylation of the non-catalytic subunitMms4 (4243) This phosphorylation is required for thefull nuclease activity of the complex and restricts thefunction of Mus81-Mms4 to a narrow period in anormal cell cycle just before chromosome segregation(4243) A recent report has confirmed this mode of regu-lation for Mus81-Mms4 and has provided further evidenceof its importance for the maintenance of genomic stabilityshowing that the uncontrolled activation of this endo-nuclease causes deleterious mitotic crossovers and prema-ture processing of DNA intermediates (44) HumanMus81-Eme1 is also regulated through phosphorylationof the non-catalytic subunit like in budding yeast (42)In addition it has been proposed that in human cellsMus81 is negatively controlled by Wee1 (45) Interes-tingly a recent work has shown that in S pombe thecell cyclendashdependent phosphorylation of Eme1Mms4 isCdc2CDK1-dependent but Plo1PLK1-independent (46)unlike in budding yeast and human cells thus showingdifferences among organisms in the way that Mus81-Eme1Mms4 is regulated Moreover in fission yeastEme1Mms4 hyperphosphorylation is stimulated when cellsare treated with camptothecin or bleomycin This induced-hyperphosphorylation is Rad3-dependent and increasesthe activity of S pombe Mus81-Eme1 (46)Despite the recent findings on Mus81-Eme1Mms4

regulation and its established function in the resistanceto some DNA-damaging agents that hinder chromosomereplication it is still largely unknown how this endonucle-ase operates when cells need to cope with DNA damage atRFs Mus81-Eme1Mms4 is known to be crucial fordealing with DNA lesions such as those induced by themodel DNA-damaging agent methyl methanesulfonate(MMS) which interfere with RF progression Yetbecause this endonuclease can cleave DNA RFs its

activity must be carefully controlled somehow underthese situations of DNA damage Although above-men-tioned works have shown that Mus81-Mms4 is activatedwhen cells finish S-phase in the absence of damaged DNA(4243) and in S pombe its activity is enhanced by someDNA-damaging drugs in G2-cells (46) it is currentlyunclear how Mus81-Mms4 is regulated in response toDNA lesions occurring during replication Here wehave analysed the role of budding yeast Mus81-Mms4under conditions of DNA damage during S-phase andshow that this endonuclease is necessary for successfulchromosome replication when cells are exposed toMMS However our data indicate that Mus81-Mms4 ac-tivation is not induced by the presence of DNA lesionsand that its function to respond to DNA damage whichallows cell survival is carried out after completion of bulkgenome replication

MATERIALS AND METHODS

Strains media cell cycle synchronization cell viability

The yeast strains used in this work were constructed bystandard procedures and are listed in SupplementaryTable S1 The pYM (47) and the pML (48) plasmidseries were used as templates for polymerase chain reac-tion Yeast cells were grown at 30C in YP medium (1yeast extract 2 bacto peptone) with 2 glucose Bactoagar (2) was added for solid medium For gene expres-sion using the GAL1-10 promoter the YP medium wassupplemented with 2 raffinose or 2 galactose The a-factor mating pheromone was added to a final concentra-tion of 5ndash10 mgml to synchronize cells in G1-phase HUwas used at 02M to block cells in early S-phaseNocodazole was used at 5 mgml to block cells in G2MSamples for flow cytometry were collected and processedas described (49) and analysed using a FACSCalibur flowcytometer (BD Biosciences) For fluorescence microscopyanalysis cells were grown in minimal medium supple-mented with yeast synthetic drop out (Sigma-Aldrich)Living cells were analysed using an Axiovert200 Zeissfluorescence microscope with a 63 oil immersion object-ive Cell viability was determined by plating cells in trip-licate onto YP-glucose (YPD) plates and counting colony-forming units after 3 days of incubation at 30C

Immunoblotting and in situ kinase assay

Protein extracts for immunoblotting analysis wereprepared from trichloroacetic acid-treated cells asdescribed (50) HA-tagged proteins were detected withthe 12CA5 antibody (CBMSO) using horseradish perox-idasendashcoupled anti-mouse (Vector Labs) as a secondaryantibody Rad53 was detected with the JDI48 antibody (agift from Dr J Diffley Cancer Research UK) and thehorseradish peroxidasendashcoupled protein A (Invitrogen) asa secondary antibody Immunoreactive bands werevisualized using enhanced chemiluminescence (ECLprime GE Healthcare) The Rad53 in situ kinase assaywas performed essentially as described (51) For theauthophosphorylation reaction the membranes (HybondP GE Healthcare) were incubated for 1 h at room













RESULTS







D

A

B

mus81Δ

BF

wt rad53Δ

GFP-TUB1

C

mus81Δmms4Δ

MUS81+MMS4+

mus81Δmms4Δ



01

1

10

100

0 60 120 180

no MMS

Via

bilit

y (

log

)

01

1

10

100

0020 MMS 0033 MMS

01

1

10

100

0015 MMS

0 60 120 180

0 60 120 180 0 60 120 180

Via

bilit

y (

log

)


MUS81+

MMS4+


Immunoblotmus81Δ

Rad53


Rad53-P Rad53


240 G1 (αF)



240 G1 (αF)

01

1

10

100















MUS81+ mus81Δ


0

20

60

80

0 +MMS 120180

Cel

lvia

bilit

y (

)

24060

40

100

MUS81+

mus81Δ

B

XVIXIII

Chrom

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

I

G1 MMS recovery

C

G1 MMS recovery

D

fractions


rel

coun

ts

release + MMS

G1 (α factor)

releaseno MMS

r

eplic

atio

n 6 5 4 3 2 1

20

40

60

80100

030 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

MUS81+


6 5 4 3 2 1

Fork movement

X

Time in MMS(min)

A

30 60 120 240


rel

coun

ts

fractions

20

40

60

80100

0

6 5 4 3 2 1

E

r

eplic

atio

n

release+ MMS

G1 (α factor)

releaseno MMS

Time in MMS(min) 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

mus81Chromosome VI

ARS607 10 20 30 40 50 60 70 kb

6 5 4 3 2 1 X

Fork movement

20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40


















I

XVIXIII

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

Chrom HU

mus81ΔMUS81+

G1

recovery

G1 HU recovery

0

50

100MUS81+ mus81Δ

Cel

lvia

bilit

y (

)

t=0HU t=0HU

B D

Rad53

HU



(αF) Rel-HU

Rad53Rad53-P

C


HU



(αF) Rel-HU

A

2C 1C 2C 1C

MUS81+ mus81

Log

G1 ( F)

30

60

90 120

rel ndash

HU

(min

)

S phase+HU

rel -

HU

(+ n

oc)

(m

in)

MUS81+ mus81

2C 1C 2C 1C Log

G1 ( F)

30

60

90 120

S phase+HU





E

HA-Mms4

[

IP

WCE

HA-Mms4

[

ImmunoblotD

F IPs

HA-Mms4

Nuclease assay

RF

IPsHA-Mms4

Nuclease assay

G1 (αF) 15 30 45 60 90 105 120 75

release + MMS (min)

HA-Mms4-P HA-Mms4

B


C

G1 (αF) +M

MS

15 30 45 60 90 105 120 75

no MMS (min)

HA-Mms4-P HA-Mms4


G1 (αF) 15 30 45 60 90 105 120 75


HA-Mms4-P HA-Mms4


A
















A

C

D

E

B






A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES

































































































RESULTS







D

A

B

mus81Δ

BF

wt rad53Δ

GFP-TUB1

C

mus81Δmms4Δ

MUS81+MMS4+

mus81Δmms4Δ



01

1

10

100

0 60 120 180

no MMS

Via

bilit

y (

log

)

01

1

10

100

0020 MMS 0033 MMS

01

1

10

100

0015 MMS

0 60 120 180

0 60 120 180 0 60 120 180

Via

bilit

y (

log

)


MUS81+

MMS4+


Immunoblotmus81Δ

Rad53


Rad53-P Rad53


240 G1 (αF)



240 G1 (αF)

01

1

10

100















MUS81+ mus81Δ


0

20

60

80

0 +MMS 120180

Cel

lvia

bilit

y (

)

24060

40

100

MUS81+

mus81Δ

B

XVIXIII

Chrom

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

I

G1 MMS recovery

C

G1 MMS recovery

D

fractions


rel

coun

ts

release + MMS

G1 (α factor)

releaseno MMS

r

eplic

atio

n 6 5 4 3 2 1

20

40

60

80100

030 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

MUS81+


6 5 4 3 2 1

Fork movement

X

Time in MMS(min)

A

30 60 120 240


rel

coun

ts

fractions

20

40

60

80100

0

6 5 4 3 2 1

E

r

eplic

atio

n

release+ MMS

G1 (α factor)

releaseno MMS

Time in MMS(min) 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

mus81Chromosome VI

ARS607 10 20 30 40 50 60 70 kb

6 5 4 3 2 1 X

Fork movement

20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40


















I

XVIXIII

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

Chrom HU

mus81ΔMUS81+

G1

recovery

G1 HU recovery

0

50

100MUS81+ mus81Δ

Cel

lvia

bilit

y (

)

t=0HU t=0HU

B D

Rad53

HU



(αF) Rel-HU

Rad53Rad53-P

C


HU



(αF) Rel-HU

A

2C 1C 2C 1C

MUS81+ mus81

Log

G1 ( F)

30

60

90 120

rel ndash

HU

(min

)

S phase+HU

rel -

HU

(+ n

oc)

(m

in)

MUS81+ mus81

2C 1C 2C 1C Log

G1 ( F)

30

60

90 120

S phase+HU





E

HA-Mms4

[

IP

WCE

HA-Mms4

[

ImmunoblotD

F IPs

HA-Mms4

Nuclease assay

RF

IPsHA-Mms4

Nuclease assay

G1 (αF) 15 30 45 60 90 105 120 75

release + MMS (min)

HA-Mms4-P HA-Mms4

B


C

G1 (αF) +M

MS

15 30 45 60 90 105 120 75

no MMS (min)

HA-Mms4-P HA-Mms4


G1 (αF) 15 30 45 60 90 105 120 75


HA-Mms4-P HA-Mms4


A
















A

C

D

E

B






A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES

























































































D

A

B

mus81Δ

BF

wt rad53Δ

GFP-TUB1

C

mus81Δmms4Δ

MUS81+MMS4+

mus81Δmms4Δ



01

1

10

100

0 60 120 180

no MMS

Via

bilit

y (

log

)

01

1

10

100

0020 MMS 0033 MMS

01

1

10

100

0015 MMS

0 60 120 180

0 60 120 180 0 60 120 180

Via

bilit

y (

log

)


MUS81+

MMS4+


Immunoblotmus81Δ

Rad53


Rad53-P Rad53


240 G1 (αF)



240 G1 (αF)

01

1

10

100















MUS81+ mus81Δ


0

20

60

80

0 +MMS 120180

Cel

lvia

bilit

y (

)

24060

40

100

MUS81+

mus81Δ

B

XVIXIII

Chrom

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

I

G1 MMS recovery

C

G1 MMS recovery

D

fractions


rel

coun

ts

release + MMS

G1 (α factor)

releaseno MMS

r

eplic

atio

n 6 5 4 3 2 1

20

40

60

80100

030 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

MUS81+


6 5 4 3 2 1

Fork movement

X

Time in MMS(min)

A

30 60 120 240


rel

coun

ts

fractions

20

40

60

80100

0

6 5 4 3 2 1

E

r

eplic

atio

n

release+ MMS

G1 (α factor)

releaseno MMS

Time in MMS(min) 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

mus81Chromosome VI

ARS607 10 20 30 40 50 60 70 kb

6 5 4 3 2 1 X

Fork movement

20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40


















I

XVIXIII

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

Chrom HU

mus81ΔMUS81+

G1

recovery

G1 HU recovery

0

50

100MUS81+ mus81Δ

Cel

lvia

bilit

y (

)

t=0HU t=0HU

B D

Rad53

HU



(αF) Rel-HU

Rad53Rad53-P

C


HU



(αF) Rel-HU

A

2C 1C 2C 1C

MUS81+ mus81

Log

G1 ( F)

30

60

90 120

rel ndash

HU

(min

)

S phase+HU

rel -

HU

(+ n

oc)

(m

in)

MUS81+ mus81

2C 1C 2C 1C Log

G1 ( F)

30

60

90 120

S phase+HU





E

HA-Mms4

[

IP

WCE

HA-Mms4

[

ImmunoblotD

F IPs

HA-Mms4

Nuclease assay

RF

IPsHA-Mms4

Nuclease assay

G1 (αF) 15 30 45 60 90 105 120 75

release + MMS (min)

HA-Mms4-P HA-Mms4

B


C

G1 (αF) +M

MS

15 30 45 60 90 105 120 75

no MMS (min)

HA-Mms4-P HA-Mms4


G1 (αF) 15 30 45 60 90 105 120 75


HA-Mms4-P HA-Mms4


A
















A

C

D

E

B






A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES


































































































MUS81+ mus81Δ


0

20

60

80

0 +MMS 120180

Cel

lvia

bilit

y (

)

24060

40

100

MUS81+

mus81Δ

B

XVIXIII

Chrom

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

I

G1 MMS recovery

C

G1 MMS recovery

D

fractions


rel

coun

ts

release + MMS

G1 (α factor)

releaseno MMS

r

eplic

atio

n 6 5 4 3 2 1

20

40

60

80100

030 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

MUS81+


6 5 4 3 2 1

Fork movement

X

Time in MMS(min)

A

30 60 120 240


rel

coun

ts

fractions

20

40

60

80100

0

6 5 4 3 2 1

E

r

eplic

atio

n

release+ MMS

G1 (α factor)

releaseno MMS

Time in MMS(min) 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240 30 60 120 240

mus81Chromosome VI

ARS607 10 20 30 40 50 60 70 kb

6 5 4 3 2 1 X

Fork movement

20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40 20 60 0 40


















I

XVIXIII

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

Chrom HU

mus81ΔMUS81+

G1

recovery

G1 HU recovery

0

50

100MUS81+ mus81Δ

Cel

lvia

bilit

y (

)

t=0HU t=0HU

B D

Rad53

HU



(αF) Rel-HU

Rad53Rad53-P

C


HU



(αF) Rel-HU

A

2C 1C 2C 1C

MUS81+ mus81

Log

G1 ( F)

30

60

90 120

rel ndash

HU

(min

)

S phase+HU

rel -

HU

(+ n

oc)

(m

in)

MUS81+ mus81

2C 1C 2C 1C Log

G1 ( F)

30

60

90 120

S phase+HU





E

HA-Mms4

[

IP

WCE

HA-Mms4

[

ImmunoblotD

F IPs

HA-Mms4

Nuclease assay

RF

IPsHA-Mms4

Nuclease assay

G1 (αF) 15 30 45 60 90 105 120 75

release + MMS (min)

HA-Mms4-P HA-Mms4

B


C

G1 (αF) +M

MS

15 30 45 60 90 105 120 75

no MMS (min)

HA-Mms4-P HA-Mms4


G1 (αF) 15 30 45 60 90 105 120 75


HA-Mms4-P HA-Mms4


A
















A

C

D

E

B






A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES




































































































I

XVIXIII

well

IVXII

VIIXV

II XIVX

XI VVIII

IX III VI

Chrom HU

mus81ΔMUS81+

G1

recovery

G1 HU recovery

0

50

100MUS81+ mus81Δ

Cel

lvia

bilit

y (

)

t=0HU t=0HU

B D

Rad53

HU



(αF) Rel-HU

Rad53Rad53-P

C


HU



(αF) Rel-HU

A

2C 1C 2C 1C

MUS81+ mus81

Log

G1 ( F)

30

60

90 120

rel ndash

HU

(min

)

S phase+HU

rel -

HU

(+ n

oc)

(m

in)

MUS81+ mus81

2C 1C 2C 1C Log

G1 ( F)

30

60

90 120

S phase+HU





E

HA-Mms4

[

IP

WCE

HA-Mms4

[

ImmunoblotD

F IPs

HA-Mms4

Nuclease assay

RF

IPsHA-Mms4

Nuclease assay

G1 (αF) 15 30 45 60 90 105 120 75

release + MMS (min)

HA-Mms4-P HA-Mms4

B


C

G1 (αF) +M

MS

15 30 45 60 90 105 120 75

no MMS (min)

HA-Mms4-P HA-Mms4


G1 (αF) 15 30 45 60 90 105 120 75


HA-Mms4-P HA-Mms4


A
















A

C

D

E

B






A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES
























































































E

HA-Mms4

[

IP

WCE

HA-Mms4

[

ImmunoblotD

F IPs

HA-Mms4

Nuclease assay

RF

IPsHA-Mms4

Nuclease assay

G1 (αF) 15 30 45 60 90 105 120 75

release + MMS (min)

HA-Mms4-P HA-Mms4

B


C

G1 (αF) +M

MS

15 30 45 60 90 105 120 75

no MMS (min)

HA-Mms4-P HA-Mms4


G1 (αF) 15 30 45 60 90 105 120 75


HA-Mms4-P HA-Mms4


A
















A

C

D

E

B






A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES



































































































A

C

D

E

B






A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES

























































































A

B

C

D E





DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES
























































































DISCUSSION











SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES
























































































SUPPLEMENTARY DATA


ACKNOWLEDGEMENTS


FUNDING



REFERENCES






















































































Temporal regulation of the Mus81-Mms4 endonuclease ensures ...digital.csic.es › ... › 1 ›...

Documents

Transcript of Temporal regulation of the Mus81-Mms4 endonuclease ensures ...digital.csic.es › ... › 1 ›...