Bd

FGD

a

ARRAA

KhBADD

1

n(ciceTps(IMrala

1d

The International Journal of Biochemistry & Cell Biology 43 (2011) 1754– 1763

Contents lists available at ScienceDirect

The International Journal of Biochemistry& Cell Biology

journa l h o me page: www.elsev ier .com/ locate /b ioce l

RCA1 is required for hMLH1 stabilization following doxorubicin-induced DNAamage

rancesco Romeo, Lucia Falbo, Maddalena Di Sanzo, Roberta Misaggi, Maria C. Faniello,iuseppe Viglietto, Giovanni Cuda, Francesco Costanzo ∗, Barbara Quaresima

epartment of Experimental and Clinical Medicine, “Magna Græcia” University of Catanzaro, “Salvatore Venuta” Campus, Viale Europa, 88100 Catanzaro, Italy

r t i c l e i n f o

rticle history:eceived 1 April 2011eceived in revised form 29 July 2011ccepted 9 August 2011vailable online 16 August 2011

eywords:MLH1RCA1TM/ATR

a b s t r a c t

Human DNA mismatch repair (MMR) is involved in the removal of DNA base mismatches that arise eitherduring DNA replication or are caused by DNA damage. In this study, we show that the activation of theMMR component hMLH1 in response to doxorubicin (DOX) treatment requires the presence of BRCA1and that this phenomenon is mediated by an ATM/ATR dependent phosphorylation of the hMLH1 Ser-406 residue. BRCA1 is an oncosuppressor protein with a central role in the DNA damage response and itis a critical component of the ATM/ATR mediated checkpoint signaling. Starting from a previous findingin which we demonstrated that hMLH1 is able to bind to BRCA1, in this study we asked whether BRCA1might be the bridge for ATM/ATR dependent phosphorylation of the hMLH1 molecular partner. We foundthat: (i) the negative modulation of BRCA1 expression is able to produce a remarkable reversal of hMLH1

oxorubicinNA damage

stabilization, (ii) BRCA1 is required for post-translational modification produced by DOX treatment onhMLH1 which is, in turn, attributed to the ATM/ATR activity, (iii) the serine 406 phosphorylatable residueis critical for hMLH1 activation by ATM/ATR via BRCA1.

Taken together, our data lend support to the hypothesis suggesting an important role of this onco-suppressor as a scaffold or bridging protein in DNA-damage response signaling via downstreamphosphorylation of the ATM/ATR substrate hMLH1.

. Introduction

DNA mismatch repair (MMR) system contributes to the mainte-ance of the genomic stability in both prokaryotes and eukaryotesKolodner and Marsischky, 1999) through the correction of repli-ation errors, the suppression of recombination between nondentical, but homologous sequences, and the activation of cellycle arrest and apoptosis in response to DNA damage (Buermeyert al., 1999a; Kunkel and Erie, 2005; Schofield and Hsieh, 2003).he MMR system is composed by a group of highly conservedroteins that recognize and repair base–base mismatches andmall insertion–deletion mispairs generated during DNA synthesisKolodner and Marsischky, 1999; Harfe and Jinks-Robertson, 2000).n prokaryotes, the MMR is made out of the three subunits MutS,

utL, and Mut H. In eukaryotes, the system is more complex andequires heterodimeric complexes between MutS homolog (hMSH)

nd MutL homolog (hMLH) families. Of the five MutS homo-ogues that have been identified in human cells, hMSH2, hMSH3nd hMSH6 participate in MMR. hMSH2 forms heterodimers with

∗ Corresponding author. Tel.: +39 0961 3694097; fax: +39 0961 3694112.E-mail address: fsc@unicz.it (F. Costanzo).

357-2725/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.oi:10.1016/j.biocel.2011.08.011

© 2011 Elsevier Ltd. All rights reserved.

hMSH6 (MutS�) or hMSH3 (MutS�), and studies on the specificityof MMR have led to the conclusion that MutS� primarily recog-nizes and corrects single base-pair mismatches (Palombo et al.,1995; Acharya et al., 1996; Gradia et al., 1997), while MutS� actsto correct insertion/deletion loops (Acharya et al., 1996; Palomboet al., 1996). The MutL homologues expressed in human cells arehMLH1, hMLH3, hPMS1 (post-meiotic segregation protein-1) andhPMS2. hMLH1 can dimerize with hPMS2 (MutL�), hPMS1 (MutL�)or hMLH3 (MutL�). The complex composed by hMLH1 and hPMS2(MutL�) has the most important role in MMR. This heterodimershows endonuclease activity (Kadyrov et al., 2006), interacts withMutS–DNA complex and with other enzymes required for DNA syn-thesis, and is involved in meiotic recombination. Similarly, MutL�,that is predominantly involved in meiotic recombination, seemsto have a backup role in repairing mispairs (Jiricny, 2006). On thecontrary, MutL� does not have significant MMR activity in in vitroassays (Raschle et al., 1999).

Recent evidence supports a potential role of MMR in sig-naling the DNA damage response; in particular, induction of

hMLH1 in response to several DNA-damaging agents, includingcisplatin, doxorubicin (Luo et al., 2004) and N-methyl-N′-nitro-N′-nitrosoguanidine has been observed (Kim et al., 2007). Moreover,stabilization of the remaining hMutL proteins (hPMS1 and hPMS2)

Bioche

aae2

tAsg(E

npaMhotidhpQ

BtmsPbts

2

2

Efiwh(Rncc2

eFb

2

hTamQh5

F. Romeo et al. / The International Journal of

nd their nuclear compartmentalization in response to DNA dam-ge has been shown to require the induction of hMLH1 (Raschlet al., 1999; Buermeyer et al., 1999b; Chang et al., 2000; Leung et al.,000).

The major DNA damage-signaling proteins are ATM (ataxiaelangiectasia mutated) and ATR (ATM and Rad3-related). ATM andTR are serine/threonine protein kinases with overlapping sub-trate specificities that trigger, in response to different types ofenotoxic stimuli, the induction of other proteins such as p53Banin et al., 1998; Canman et al., 1998; Keramaris et al., 2003),2F1 (Lin et al., 2001) BRCA1 (Gatei et al., 2001).

BRCA1 is a nuclear phosphoprotein that regulates multipleuclear processes including DNA repair and recombination, check-oint control of the cell cycle, transcription, chromatin remodellingnd ubiquitination (Monteiro, 2000; Lane, 2004; Harkin et al., 1999;acLachlan et al., 2000; Somasundaram et al., 1997). Moreover, it

as been proposed that BRCA1 may act as a scaffold protein thatrganizes many of the DNA repair proteins and of the known ATMargets in a complex called BASC (Wang et al., 2000). Therefore, its conceivable that BRCA1 might be the bridge for ATM/ATR depen-ent phosphorylation of BRCA1 molecular partners. On the otherand, a direct interaction between DNA repair proteins or MMRroteins and BRCA1 has been reported (Durant and Nickoloff, 2005;uaresima et al., 2006; Wang et al., 2001).

The aim of this study was to shed light on the potential role ofRCA1 in the induction of hMLH1 in response to doxorubicin (DOX)reatment. DOX, a key chemotherapeutic drug for cancer treat-

ent, is known to act as topoisomerase II poison and to block theynthesis of DNA by intercalating into the DNA strand (Granados-rincipal et al., 2010). We show here that the induction of hMLH1y DOX treatment is due to an ATM/ATR-dependent phosphoryla-ion of hMLH1 Ser-406 and that BRCA1 is required for the effectivetabilization and accumulation of the active MMR protein.

. Materials and methods

.1. Cell culture and DNA transfections

MCF7 (human sporadic breast cancer cells) and 293T (Humanmbryonic Kidney cells) cells were grown in Dulbecco’s modi-ed Eagle’s medium (DMEM) (Life Technologies) supplementedith 10% FBS and 1% streptomycin/penicillin. HCC1937−/− (humanereditary breast cancer cells) and human lymphoblastoid GM21842184ATM+/+) and GM1526 (1526ATM−/−) cell lines were grown inoswell Park Memorial Institute medium (RPMI-1640) (Life Tech-ologies) containing 15% FBS and 1% streptomycin/penicillin. Stablelones of HCC1937−/−/pcDNA3.1/empty and HCC1937−/−/wtBRCA1ells, were generated as previously described (Quaresima et al.,008). All cells were grown at 37 ◦C in a 5% CO2 atmosphere.

293T cells, lacking expression of the MutL� complex (Trojant al., 2002), were co-transiently transfected with pcDNA3.1-lagATMwt and either pcDNA3-hMLH1wt or pcDNA3-hMLH1S406A

y using calcium phosphate precipitation method.

.2. Plasmids

The entire open reading frame (ORF) of hMLH1 (pcDNA3-MLH1), cloned into pcDNA3 vector (Invitrogen Lifeechnologies) and pcDNA3.1-FlagATMwt were already avail-ble in our laboratory (Quaresima et al., 2003). Site-directed

utagenesis of the hMLH1 insert was obtained using theuickChange Kit (Stratagene). The hMLH1 mutant (pcDNA3-MLH1S406A) was generated with the following primers: forward′-gagcaaacccctgtccGCCcagccccaggccattg-3′ and reverse 5′-

mistry & Cell Biology 43 (2011) 1754– 1763 1755

caatggcctggggctgGGCggacaggggtttgctc-3′. The construct wasverified by DNA sequencing.

2.3. Drug and inhibitor treatment

Doxorubicin (DOX) (Sigma–Aldrich) treatments were per-formed by removing media from cultures of logarithmicallygrowing cells. The cells were washed twice with ice-cold PBS1Xand fresh media added. Then, DOX was added to the culture mediaat a concentration of 600 ng/ml for indicated time points. Cyclohex-imide (CHX) was added (50 �g/ml final concentration) 4 h followedDOX treatment for the time periods as indicated.

2.4. Preparation of lentiviral supernatants and transduction ofMCF7 cells

5 × 106 293T cells were grown on 10-cm plates to 70–80%confluence and co-transfected with 10 �g of each shRNAlentiviral specific plasmid [plK0.1/shRNA (scramble vector) orplK0.1/shBRCA1 (Sigma–Aldrich)], 2 �g VSV-G plasmid DNA and18 �g packaging viral CMV-�8.9 plasmid, using the calcium phos-phate precipitation method. After the addition of fresh culturemedium 8 h later, the cells were cultured for an additional 2 days.The medium was harvested 48 h post-transfection, and filteredthrough a 0.45 �m filter. The supernatant from 293T cell cul-tures were used to cross-transduce MCF7 cells in the presence of8 �g/ml polybrene (Sigma–Aldrich). The efficiency of transductionwas calculated by analyzing expression of GFP 1/2 days after infec-tion using a fluorescent microscope. Subsequently the clones wereselected with puromycin (1 �g/ml) (Sigma–Aldrich).

2.5. Immunoprecipitation assays

Approximately 1 × 107 cells were treated with 600 ng/ml of DOXfor 4 h. After incubation, cells were harvested, washed with ice-coldPBS1X and lysed in 1X lysis Buffer (Hepes 20 mM pH 7.9; 150 mMNaCl; 5 mM EDTA; 0.1% NP-40; 1 mM PMSF; 1 mM Na–vanadate;5 mM NaF; 1 �g/ml Aprotinin; 1 �g/ml Leupeptin). After preclear-ing, the lysates were adjusted for equal protein content, and 5 �g ofmouse anti-hMLH1 antibody (clone G168-728; BD Pharmigen) wasadded. After 1hr incubation on ice, 30 �l of protein A/G beads (SantaCruz) was added, and the incubation was continued over nightwith rocking platform at +4 ◦C. Immunocomplexes were washedthree times with 1× lysis buffer and finally resuspended in 1× SDS-PAGE loading buffer and subjected to immunoblot analysis. Thesame conditions were used for immunoprecipitation of hMLH1wt

or hMLH1S406A proteins expressed in 293T cells.

2.6. Western blot analysis

Western blot analysis for detection of BRCA1 eitherhMLH1 or PMS2 and �-tubulin proteins were performedon MCF7, MCF7/shRNA, MCF7/shBRCA1, HCC1937−/−,HCC1937−/−/pcDNA3.1/empty, HCC1937−/−/wtBRCA1, 2184ATM+/+,1526ATM−/−, 293T/hMLH1wt and 293T/hMLH1S406A cell lines. 40 �gtotal protein extract was separated by SDS-PAGE gel electrophore-sis and transferred to nitrocellulose membranes (ImmobilonTM P,Millipore). After addition of the blocking mixture, the membraneswere incubated with specific antibodies: mouse anti-BRCA1(D-9), rabbit anti-hMLH1 (H300), rabbit anti-PMS2 (C-20) (allantibodies were obtained from Santa Cruz Biotechnology). Thesignal was detected using anti-mouse or anti-rabbit horseradish

peroxidase-conjugated secondary antibodies and ECL® (Santa CruzBiotechnology). Moreover, the membrane was incubated with agoat anti-�-tubulin antibody HRP-conjugated (C-20, Santa Cruz),to ensure uniform gel loading.

1756 F. Romeo et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1754– 1763

Fig. 1. (A) Western blot analysis of BRCA1, hMLH1 and hPMS2 proteins. MCF7 cells were treated with 600 ng/ml DOX for indicated times. Total protein extracts were analyzedby Western blot using anti-BRCA1, anti-hMLH1 and anti-PMS2 antibodies. (B) Expression of the hMLH1 gene. Real-time PCR was performed on total RNA from MCF7 cellsf f hMLw cted bw

owhma((npB

2

pcocDuga5Aco

ollowing treatment with 600 ng/ml DOX for the indicated times. (C) The half-life oere exposed to cycloheximide at 50 �g/ml as indicated. hMLH1 protein was deteas performed to ensure equal levels of protein loading.

Western blot analysis, for detection of phosphorylated formsf hMLH1, was performed on total extract from cells treatedith DOX for 4 h, immunoprecipitated with a mouse anti-MLH1 antibody. After addition of the blocking mixture, theembranes were incubated with specific antibodies [mouse

nti-phospho-serine (Santa Cruz), mouse anti-phospho-threonineSanta Cruz), rabbit anti-phospho-(Ser/Thr) ATM/ATR substrateCell Signaling) or anti-hMLH1 (H300, Santa Cruz)]. The sig-al was detected using anti-mouse or anti-rabbit horseradisheroxidase-conjugated secondary antibodies and ECL® (Santa Cruziotechnology).

.7. RNA extraction and quantitative real-time PCR

Total RNA extraction for semi-quantitative reverse transcriptaseolymerase chain reaction was done from MCF7 cells at 60–70%onfluence, treated and untreated with DOX at final concentrationf 600 ng/ml for indicated time points, with TRIzol Reagent (Gib-oBRL) according to the manufacturer’s protocol. A total of 5 �gNase-treated RNA was reverse transcribed into first-strand cDNAsing the SuperscriptTM First-Strand Synthesis System (Invitro-en) according to the manufacturer’s protocol. cDNA (2 �l) wasmplified for hMLH1 gene with the following primers: forward

′-gctgatgttaggacactacc-3′ and reverse 5′-gttgattctaccagacgatgg-3′.

human glyceraldehyde 3-phosphate dehydrogenase (GAPDH)DNA fragment was amplified as the internal control for the amountf cDNA in the PCR.

H1 is prolonged by DOX. MCF7 cells, either treated or untreated with DOX for 4 h,y Western blot analysis using a specific antibody. �-Tubulin immunoblot analysis

3. Results

3.1. BRCA1 is required for stabilization of hMLH1 in response toDOX

We have previously demonstrated that BRCA1 may physicallyinteract with hMLH1 (Quaresima et al., 2006). Moreover, BRCA1has been shown to act as a scaffold protein for most of the DNArepair proteins (Wang et al., 2000). On this basis, we investi-gated the potential role of the BRCA1-dependent induction ofhMLH1 by DOX treatment, in the human sporadic breast cancercell line MCF7. Western blot analysis demonstrated that BRCA1 andhMLH1 expression is induced in this cell line, following DOX treat-ment in a time-dependent manner up to 8 h (Fig. 1A). Similarly,hPMS2 was also activated by DOX treatment in a time-dependentfashion as demonstrated by Luo et al. (2004) (Fig. 1A). The DOX-induced hMLH1 over-expression appears to be mediated by apost-transcriptional mechanism, since the hMLH1 mRNA levels,measured by real-time PCR, did not significantly change up to 8 hafter treatment (Fig. 1B). As shown by experiments with the proteinsynthesis inhibitor cycloheximide, the accumulation of hMLH1 inMCF7 cells is due to an increased stability of the protein (Fig. 1C).

We postulated that the mechanism leading to hMLH1 accu-

mulation might require BRCA1. To verify this hypothesis MCF7cells, in which BRCA1 expression was silenced with specific shRNA(MCF7/shBRCA1)(see Fig. 2A), were treated with DOX. In theMCF7/shBRCA1 cells the hMLH1 levels were not influenced by the

F. Romeo et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1754– 1763 1757

Fig. 2. BRCA1 is required for stabilization of hMLH1 in response to DOX treatment in MCF7 cells. (A) Total protein extracts of MCF7, MCF7/shRNA, MCF7/shBRCA1 cells wereanalyzed by Western blot using a mouse anti-BRCA1 antibody. Western blot analysis of hMLH1 protein in the MCF7/shBRCA1 (B) and MCF7/shRNA (D) cells, after treatmentw loadim

d(it

bweHvrAm(HmmF

3B

o

ith 600 ng/ml DOX for indicated times. �-Tubulin was used as control for proteinRNA expression levels of hMLH1.

rug both at protein (see Western blot in Fig. 2B) and mRNA levelssee real-time PCR in Fig. 2C). On the other hand, DOX treatmentn MCF7/shRNA cells produced a behaviour similar to that of wildype MCF7 (Fig. 2D and E).

Taken together, these data strongly suggest that hMLH1 sta-ilization may be intimately linked to BRCA1. On this basis,e evaluated the expression of this MMR protein by West-

rn blot and real-time PCR analysis in the BRCA1-defectiveCC1937 cell line (HCC1937−/−), as well as in the empty-ector-transfected (HCC1937−/−/pcDNA3.1/empty) and in BRCA1-econstituted (HCC1937−/−/wtBRCA1) cells, after DOX treatment.s shown in Fig. 3, hMLH1 expression (both at protein andRNA level) did not significantly change either in HCC1937−/−

3A) or in HCC1937−/−/pcDNA3.1/empty (3B) cells. Conversely, inCC1937−/−/wtBRCA1 cells, hMLH1 protein was stabilized, whileRNA levels did not show significantly change up to 8 h after treat-ent (3C), consistently with data obtained using MCF7 cells (see

ig. 1A and B).

.2. DOX treatment induces hMLH1 phosphorylation in a

RCA1-dependent manner

To shed light on the mechanisms underlying stabilizationf hMLH1 after DOX treatment and to demonstrate BRCA1

ng. Real-time PCR analysis was performed on the same cells (C and E) to evaluate

involvement in this process, we first asked whether hMLH1 isphosphorylated in response to DNA damage. hMLH1 protein wasimmunoprecipitated from MCF7/shRNA and MCF7/shBRCA1 cellsafter DOX treatment, and protein phosphorylation was evaluatedby Western blot analysis with specific anti-phospho serine or anti-phospho threonine antibodies. As shown in Fig. 4A, upper panel,the levels of phosphorylated hMLH1 detected in MCF7/shRNA cellschallenged with anti-phospho-serine antibody (lane 3) were con-sistently higher compared to those observed in MCF7/shBRCA1 cells(lane 4). On the other hand, changes in phospho-threonine levelswere much less evident in MCF7 cells interfered for BRCA1 (Fig. 4B,upper panel, lanes 3 and 4). Western blot analysis with anti-hMLH1antibody was performed as control (Fig. 4A and B, lower panels).

Analogous experiments were performed in naturally occurringBRCA1-defective HCC1937 cells. We found that serine phospho-rylation of MLH1 was almost completely abolished in HCC1937−/−

cells (Fig. 5A, upper panel, lane 4) while threonine phosphorylationwas consistently reduced. The difference between the changes inphosphorylation levels observed in the two cellular systems usedmight be ascribed to the fact that while HCC1937−/− cells are com-

pletely devoid of BRCA1 residual BRCA1 expression can be stillobserved in MCF7/shBRCA1 cells.

Next, we set out to seek evidence of the activation of ATMand ATR protein kinases, based on the knowledge that BRCA1 is

1758 F. Romeo et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1754– 1763

F C1937−/− −/−

H h 600t

rshtcpbhlt

teatdslaihAp(cc

ig. 3. BRCA1-mediated stabilization of hMLH1 in response to DOX treatment in HCCC1937−/−/pcDNA3.1/empty (B) and HCC1937−/−/wtBRCA1 (C) cells, after treatment wit

o ensure equal levels of protein loading.

equired for the phosphorylation of a number of ATM and ATRubstrates (Foray et al., 2003), and because of the ATM-mediatedMLH1 stabilization during DNA damage (Luo et al., 2004). Tohis end, MCF7/shRNA and MCF7/shBRCA1 cell lysates, immunopre-ipitated with an anti-hMLH1 specific antibody (�-hMLH1), wererobed with an anti-phospho-(Ser/Thr) ATM/ATR substrate anti-ody (S*/T*Q). As shown in Fig. 6 upper panel, the phosphorylatedMLH1 form was only detected in MCF7/shRNA cells (lane 3). Fig. 6,

ower panel, shows that the ATM/ATR substrate immunoprecipi-ated protein identified was indeed hMLH1 (lanes 3 and 4).

To enforce the idea that phosphorylation of hMLH1 in responseo DNA damage is mediated by ATM/ATR and requires the pres-nce of BRCA1, we performed Western blot analysis in 2184ATM+/+

nd 1526ATM−/− cell lines. Following DOX treatment, we observedhat hMLH1 protein accumulated only in 2184ATM+/+ cells in a timeependent manner (Fig. 7A), while no changes in hMHL1 expres-ion were detected in 1526ATM−/− cells (Fig. 7C); as expected, mRNAevels were not affected by DOX exposure in neither cell line (Fig. 7Bnd D, respectively). Moreover, cell lysates from both cell lines weremmunoprecipitated with an anti-hMLH1 specific antibody (�-MLH1), and subsequently probed with an anti-phospho-(Ser/Thr)TM/ATR substrate antibody (S*/T*Q). As shown in Fig. 7E, upper

anel, the phosphorylated hMLH1 form detected in 2184ATM+/+ cellslane 3), was stronger than that observed in the 1526ATM−/− cell lineounterpart. Finally, anti-hMLH1 antibody used on the same blotonfirmed that the ATM/ATR substrate immunoprecipitated pro-

cells. Western blot analysis and real-time PCR of hMLH1 in the HCC1937 (A), ng/ml DOX for the indicated times. �-Tubulin immunoblot analysis was performed

tein identified was indeed hMLH1 (Fig. 7E, lower panel, lanes 3 and4).

3.3. Serine 406 phosphorylation by ATM/ATR is critical for hMLH1stabilization

To define the potential phosphorylation site(s) on hMLH1 pro-tein targeted by ATM/ATR following DOX treatment, we performedamino acid sequence alignment of the MutL family (human, rat,Saccharomyces cerevisiae) and identified Ser–Gln (SQ) and Thr–Gln(TQ) motifs. hMLH1 contains a single SQ (Ser406–Gln) and TQ(Thr345–Gln) sequences, which are conserved among all the alignedMutL homologues. An additional TQ motif (Thr148–Gln) is onlyfound in humans. Accordingly, BRCA1 silencing in MCF7 cells(MCF7/shBRCA1) produces a certain decrease of the hMLH1/Ser-phosphorylated form, with less evident changes seen in thehMLH1/Thr-phosphorylated counterpart. On the other hand, theresults from HCC1937−/− cells confirm that the lack of BRCA1impairs Ser-phosphorylation and suggest that additional phospho-rylation sites (threonine residues) may exist since a decrease in thethreonine phosphorylated form was observed in these cells.

Therefore, we speculated that Serine 406 phosphorylation might

be directly involved in BRCA1-mediated hMLH1 accumulationvia ATM/ATR. To confirm this hypothesis, hMLH1 protein wasimmunoprecipitated with an anti-hMLH1 antibody from 293Tcells – previously subjected or not to DOX exposure – transiently

F. Romeo et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1754– 1763 1759

Fig. 4. DOX treatment induces hMLH1 phosphorylation in a BRCA1-dependent manner. MCF7/shRNA and MCF7/shBRCA1 cells were treated with DOX (600 ng/ml) andh MLH1a anti-e

copa

Faae

arvested 4 h after drug exposure. Lysates were immunoprecipitated with anti-hnalyzed by Western blot analysis with anti-phospho serine (A, upper panel) andnsure equal levels of protein loading (A and B, lower panels).

o-transfected with ATM and either wild type (293T/hMLH1wt)r mutant hMLH1 (293T/hMLH1S406A). Western blot analysis,erformed with an anti-phospho-(Ser/Thr) ATM/ATR substratentibody, demonstrated an increase in hMLH1 phosphorylation

ig. 5. BRCA1-dependent hMLH1 phosphorylation after DOX exposure in HCC1937−/− celnd harvested 4 h after drug exposure. Lysates were immunoprecipitated with anti-hMLHnalyzed by Western blot analysis with anti-phospho serine (A, upper panel) and anti-nsure equal levels of protein loading (A and B, lower panels).

(�-hMLH1) antibody or non-immune antiserum (IgG). Reactions products werephospho threonine (B, upper panel) antibodies. Anti-MLH1 antibody was used to

in 293T/hMLH1wt cells (Fig. 8A, upper panel) compared to the293T/hMLH1S406A counterpart (Fig. 8B, upper panel). The identifi-cation and the amount of hMLH1 in both cell lines were confirmedusing an anti-hMLH1 antibody (Fig. 8A and B, lower panels).

ls. HCC1937−/−/wtBRCA1 and HCC1937−/− cells were treated with DOX (600 ng/ml)1 (�-hMLH1) antibody or non-immune antiserum (IgG). Reactions products were

phospho threonine (B, upper panel) antibodies. Anti-MLH1 antibody was used to

1760 F. Romeo et al. / The International Journal of Bioche

Fig. 6. BRCA1 modulates hMLH1 phosphorylation through an ATM/ATR-mediatedmechanism. MCF7/shRNA and MCF7/shBRCA1 cells were treated with 600 ng/ml DOXand harvested 4 h after drug. Lysates were immunoprecipitated with anti-hMLH1al(

Toa

4

ieicEnpa(Ab(tut

baatr

ottph(

ntibody (�-hMLH1) or non-immune antiserum (IgG). Reaction products were ana-yzed by Western blot analysis with anti-phospho-(Ser/Thr) ATM/ATR substrateupper panel) or anti-MLH1 (lower panel) antibodies.

hese results suggest that ATM is able to phosphorylated MLH1n Serine 406, which may account for DOX-dependent MLH1ccumulation.

. Discussion

The regulation of cellular responses to DNA damage is extremelymportant for the maintenance of genome stability. The cell hasvolved with the ability to detect and propagate a global signal-ng network called “DNA damage response” (DDR), that includesell cycle arrest, DNA repair, senescence and apoptosis (Zhou andlledge, 2000). Among others, ATM and ATR are important sig-aling components of this network in mammalian cells. Theserotein kinases can be activated by double-strand breaks (DSBs)nd replication protein A (RPA)-coated single-strand DNA (ssDNA)Cimprich and Cortez, 2008; Bartek and Lukas, 2007; Shiloh, 2003).TM/ATR control the DNA damage by recruiting repair factors andy triggering, both transcriptionally and/or post-transcriptionallyvia phosphorylation, acetylation, ubiquitylation or SUMOylation),he activity of DNA-repair proteins (Huen and Chen, 2008). The fail-re of these mechanisms contributes to genomic instability andumorigenesis.

Recently, Luo et al. (2004) found that the ATM-mediated sta-ilization of the human MMR protein hMutL is critical for p53ctivation following DNA damage. hMLH1, one of the four humannalogs of bacterial MutL, appears therefore as a major regula-or and a sensor of the apparatus involved in the DNA damageesponse.

In a previous study, we have demonstrated that hMLH1, as a partf the multimeric complex called BASC (Wang et al., 2000), is ableo bind to the tumor suppressor BRCA1 at specific sites and that

hree distinct mutations in one of these interaction domains mayroduce, in vitro, a microsatellite instability phenotype, one of theallmarks of an imbalance in the mismatch DNA repair machineryQuaresima et al., 2006).

mistry & Cell Biology 43 (2011) 1754– 1763

BRCA1 is an oncosuppressor protein with a central role inthe DNA damage response of mammalian cells and it is a crit-ical component of the ATM/ATR mediated checkpoint signaling.In fact, BRCA1 serves as a scaffold that facilitates ATM/ATR phos-phorylation of a subset of downstream substrates (Foray et al.,2003) and, through interactions with other DNA repair proteins,participates in pathways that signal DNA damage (Ting and Lee,2004).

The aim of this study was to elucidate the potential role of BRCA1in the activation of hMLH1 in response to DOX treatment. DOX, ananti-neoplastic drug widely used in clinical practice, is able to blockDNA synthesis by intercalating into the DNA strand (Granados-Principal et al., 2010). In this report, we demonstrate that BRCA1is required for the effective stabilization and accumulation of thehMLH1 protein and that this phenomenon is mediated, at least inpart, by an ATM/ATR dependent phosphorylation of the hMLH1Ser-406 residue.

We started from the observation that DOX exposure pro-duces an increase of hMLH1 protein levels in MCF7 cells thatis not paralleled by an analogous change at mRNA level. Thisfinding, suggesting that hMLH1 accumulation is a phenomenontranscription-independent, is supported by data obtained treatingMCF7 cells with the protein synthesis inhibitor CHX. Under theseconditions, exposure to DOX produces a prolongation of hMLH1half-life.

To shed light on the molecular mechanisms underlying thisphenomenon and to evaluate the role of BRCA1 in hMLH1 sta-bilization in response to DOX, we applied the shRNA strategyto silence BRCA1 expression in these cells. Western blot analy-sis confirmed that the negative modulation of BRCA1 expressioninhibited hMLH1 protein induction in MCF7 cells, while no changeswere detected at mRNA level. As a further control, hMLH1 pro-tein levels after DOX exposure were evaluated in the BRCA1-lessHCC1937−/− hereditary breast cancer cell line, as well as inthe reconstituted HCC1937−/−/wtBRCA1 counterpart. As expected,hMLH1 did not significantly change at protein level in HCC1937−/−

cells, while its expression increased in HCC1937−/−/wtBRCA1cells. No changes were found in mRNA levels in both celllines.

Among the mechanisms leading to protein stabilization, post-translational modifications can be taken in to account. To this end,we focused our attention on potential changes produced by DOXtreatment on hMLH1, which might be attributed to the ATM/ATRactivity. A serine and threonine phosphorylation of hMLH1 wasindeed detected in MCF7 cells after DOX exposure; remarkably,silencing of BRCA1 in these cells (MCF7/shBRCA1) significantlyreduced serine phosphorylation, with very slight changes in thre-onine phosphorylation levels. Analogous results were obtainedin the BRCA1-deficient (HCC1937−/−) vs. BRCA1 reconstituted(HCC1937−/−/wtBRCA1) HCC1937 cell lines. In these cells, threoninephosphorylation was also consistently reduced.

These findings suggest that Ser406, specifically identified bybioinformatic analysis as one the two potential S/T–Q motifs tar-geted by ATM/ATR on the hMLH1 sequence, might be critical forhMLH1 stabilization by BRCA1 via ATM/ATR following DOX treat-ment.

The involvement of the ATM/ATR pathway in BRCA1-mediatedhMLH1 stabilization was further confirmed by the utilizationof a specific anti-phospho-(Ser/Thr) ATM/ATR substrate antibody(S*/T*Q), which demonstrated in MCF7/shBRCA1 cells, that hMLH1phosphorylation is indeed a BRCA1-dependent phenomenon.

To reinforce this finding, expression of hMLH1, both at mRNA

and protein level, was evaluated in cell harbouring wild-type ATM(2184ATM+/+), as well as in a cell line bearing an ATM homozygousnull-mutation (1526ATM−/−). As shown by Luo et al. (2004), DOXexposure produced a time-dependent hMLH1 protein accumula-

F. Romeo et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1754– 1763 1761

Fig. 7. ATM-dependent hMLH1 phosphorylation in 2184ATM+/+ cells. 2184ATM+/+ (A) and 1526ATM−/− (C) cells were treated with 600 ng/ml DOX for indicated times. Totalprotein extracts were analyzed by Western blot using anti-hMLH1 antibody. �-Tubulin immunoblot analysis was performed to ensure equal levels of protein loading.Expression of hMLH1 gene. Real-time PCR was performed on total RNA from 2184ATM+/+ (B) and 1526ATM−/− (D) cells following treatment with 600 ng/ml DOX for theindicated times. (E) hMLH1 phosphorylation is an ATM mediated mechanism. 2184ATM+/+ and 1526ATM−/− cells were treated with 600 ng/ml DOX and harvested 4 h after drug.Lysates were immunoprecipitated with anti-hMLH1 antibody (�-hMLH1) or non-immune antiserum (IgG). Reaction products were analyzed by Western blot analysis witha el) an

tA

b

fi1

htv

nti-phospho-(Ser/Thr) ATM/ATR substrate (upper panel) or anti-MLH1 (lower pan

ion in 2184ATM+/+ cells, while no changes could be detected in theTM-less cell line counterpart.

On the other hand, hMLH1 mRNA expression levels, evaluatedy real-time PCR, did not show any modification in both cell lines.

The role of ATM in hMLH1 phosphorylation was further con-rmed by the significant reduction of phospho-hMLH1 in the526ATM−/−cell line.

Finally, transfection of either wild type or mutated (S406A)MLH1 in 293T cells over-expressing ATM, demonstrated thathis phosphorylatable residue is indeed critical for hMLH1 acti-ation by ATM/ATR via BRCA1. Interestingly, a hMLH1 S406N

tibodies.

missense mutation has been reported to segregate in Heredi-tary Non-Polyposis Colorectal Cancers (HNPCC) (Genuardi et al.,1999).

Taken together, our data lend support to previous findings sug-gesting a critical role played by BRCA1 in ATM/ATR-dependentphosphorylation of downstream substrates, with a special focus onthe cascade of events triggered by modulation of the MMR protein

hMLH1.

Our model, in which BRCA1 is required for hMLH1 stabiliza-tion in response to DOX, appears in perfect agreement with therole of this oncosuppressor as a scaffold or bridging protein in

1762 F. Romeo et al. / The International Journal of Biochemistry & Cell Biology 43 (2011) 1754– 1763

Fig. 8. Serine 406 phosphorylation is critical for hMLH1 stabilization. 293T cells were transiently co-transfected with pcDNA3.1-FlagATMwt and either pcDNA3-hMLH1wt (A)or pcDNA3-hMLH1S406A (B). 48 h after transfection, cells were treated with DOX at a concentration of 600 ng/ml for 4 h and thus harvested. Lysates were immunoprecipitatedw roducs

Do

A

v

R

A

B

B

B

B

C

C

C

D

F

G

G

G

G

H

ith anti-hMLH1 (�-hMLH1) antibody or non-immune antiserum (IgG). Reaction pubstrate (upper panels) or anti-MLH1 (lower panels) antibodies.

NA-damage response signaling via downstream phosphorylationf ATM/ATR substrates.

cknowledgement

This work was supported by grant from MIUR (Ministero Uni-ersità e Ricerca Scientifica e Tecnologica) COFIN 2008.

eferences

charya S, Wilson T, Gradia S, Kane MF, Guerrette S, Marsischky GT, et al. hMSH2forms specific mispair-binding complexes with hMSH3 and hMSH6. Proc NatlAcad Sci USA 1996;93:13629–34.

anin S, Moyal L, Shieh S, Taya Y, Anderson CW, Chessa L, et al. Enhanced phos-phorylation of p53 by ATM in response to DNA damage. Science 1998;281:1674–7.

artek J, Lukas J. DNA damage checkpoints: from initiation to recovery or adaptation.Curr Opin Cell Biol 2007;19:238–45.

uermeyer AB, Deschenes SM, Baker SM, Liskay RM. Mammalian DNA mismatchrepair. Annu Rev Genet 1999a;33:533–64.

uermeyer AB, Wilson-Van Patten C, Baker SM, Liskay RM. The human MLH1 cDNAcomplements DNA mismatch repair defects in Mlh1-deficient mouse embryonicfibroblasts. Cancer Res 1999b;59:538–41.

anman CE, Lim DS, Cimprich KA, Taya Y, Tamai K, Sakaguchi K, et al. Activationof the ATM kinase by ionizing radiation and phosphorylation of p53. Science1998;281:1677–9.

hang DK, Ricciardiello L, Goel A, Chang CL, Boland CR. Steady-state regula-tion of the human DNA mismatch repair system. J Biol Chem 2000;275:29178.

imprich KA, Cortez D. ATR: an essential regulator of genome integrity. Nat Rev MolCell Biol 2008;9:616–27.

urant ST, Nickoloff JA. Good timing in the cell cycle for precise DNA repair by BRCA1.Cell Cycle 2005;4:1216–22.

oray N, Marot D, Gabriel A, Randrianarison V, Carr AM, Perricaudet M, et al. A sub-set of ATM- and ATR-dependent phosphorylation events requires the BRCA1protein. EMBO J 2003;22:2860–71.

atei M, Zhou BB, Hobson K, Scott S, Young D, Khanna KK. Ataxia telangiectasiamutated (ATM) kinase and ATM and Rad3 related kinase mediate phospho-rylation of Brca1 at distinct and overlapping sites. In vivo assessment usingphospho-specific antibodies. J Biol Chem 2001;276:17276–80.

enuardi M, Carrara S, Anti M, Ponz de Leon M, Viel A. Assessment of pathogenicitycriteria for constitutional missense mutations of the hereditary nonpoly-posis colorectal cancer genes MLH1 and MSH2. Eur J Hum Genet 1999;7:778–82.

radia S, Acharya S, Fishel R. The human mismatch recognition complexhMSH2–hMSH6 functions as a novel molecular switch. Cell 1997;91:995–1005.

ranados-Principal S, Quiles JL, Ramirez-Tortosa CL, Sanchez-Rovira P, Ramirez-Tortosa MC. New advances in molecular mechanisms and the prevention of adri-amycin toxicity by antioxidant nutrients. Food Chem Toxicol 2010;48:1425–38.

arfe BD, Jinks-Robertson S. DNA mismatch repair and genetic instability. Annu RevGenet 2000;34:359–99.

ts were analyzed by Western blot analysis with anti-phospho-(Ser/Thr) ATM/ATR

Harkin DP, Bean JM, Miklos D, Song YH, Truong VB, Englert C, et al. Induction ofGADD45 and JNK/SAPK-dependent apoptosis following inducible expression ofBRCA1. Cell 1999;97:575–86.

Huen MS, Chen J. The DNA damage response pathways: at the crossroad of proteinmodifications. Cell Res 2008;18:8–16.

Jiricny J. The multifaceted mismatch-repair system. Nat Rev Mol Cell Biol2006;7:335–46.

Kadyrov FA, Dzantiev L, Constantin N, Modrich P. Endonucleolytic function of Mut-Lalpha in human mismatch repair. Cell 2006;126:297–308.

Keramaris E, Hirao A, Slack RS, Mak TW, Park DS. Ataxia telangiectasia-mutatedprotein can regulate p53 and neuronal death independent of Chk2 in responseto DNA damage. J Biol Chem 2003;278:37782–9.

Kim WJ, Rajasekaran B, Brown KD. MLH1- and ATM-dependent MAPK signalingis activated through c-Abl in response to the alkylator N-methyl-N′-nitro-N′-nitrosoguanidine. J Biol Chem 2007;282:32021–31.

Kolodner RD, Marsischky GT. Eukaryotic DNA mismatch repair. Curr Opin Genet Dev1999;9:89–96.

Kunkel TA, Erie DA. DNA mismatch repair. Annu Rev Biochem 2005;74:681–710.Lane TF. BRCA1 and transcription. Cancer Biol Ther 2004;3:528–33.Leung WK, Kim JJ, Wu L, Sepulveda JL, Sepulveda AR. Identification of a second MutL

DNA mismatch repair complex (hPMS1 and hMLH1) in human epithelial cells. JBiol Chem 2000;275:15728–32.

Lin WC, Lin FT, Nevins JR. Selective induction of E2F1 in response to DNA damage,mediated by ATM-dependent phosphorylation. Genes Dev 2001;15:1833–44.

Luo Y, Lin FT, Lin WC. ATM-mediated stabilization of hMutL DNA mismatchrepair proteins augments p53 activation during DNA damage. Mol Cell Biol2004;24:6430–44.

MacLachlan TK, Somasundaram K, Sgagias M, Shifman Y, Muschel RJ, Cowan KH,et al. BRCA1 effects on the cell cycle and the DNA damage response are linkedto altered gene expression. J Biol Chem 2000;275:2777–85.

Monteiro AN. BRCA1: exploring the links to transcription. Trends Biochem Sci2000;25:469–74.

Palombo F, Gallinari P, Iaccarino I, Lettieri T, Hughes M, D’Arrigo A, et al. GTBP, a160-kilodalton protein essential for mismatch-binding activity in human cells.Science 1995;268:1912–4.

Palombo F, Iaccarino I, Nakajima E, Ikejima M, Shimada T, Jiricny J. hMutS�, a het-erodimer of hMSH2 and hMSH3, binds to insertion/deletion loops in DNA. CurrBiol 1996;6:1181–4.

Quaresima B, Alifano P, Tassone P, Avvedimento EV, Costanzo FS, Venuta S.Human mismatch-repair protein MutL homologue 1 (MLH1) interacts withEscherichia coli MutL and MutS in vivo and in vitro: a simple genetic systemto assay MLH1 function. Biochem J 2003;371:183–9.

Quaresima B, Faniello MC, Baudi F, Crugliano T, Cuda G, Costanzo F, et al. In vitroanalysis of genomic instability triggered by BRCA1 missense mutations. HumMutat 2006;27:715.

Quaresima B, Romeo F, Faniello MC, Di Sanzo M, Liu CG, Lavecchia A, et al. BRCA15083del19 mutant allele selectively up-regulates periostin expression in vitroand in vivo. Clin Cancer Res 2008;14:6797–803.

Raschle M, Marra G, Nystrom-Lahti M, Schar P, Jiricny J. Identification of hMutL, a

heterodimer of hMLH1 and hPMS1. J Biol Chem 1999;274:32368–75.

Schofield MJ, Hsieh P. DNA mismatch repair: molecular mechanisms and biologicalfunction. Annu Rev Microbiol 2003;57:579–608.

Shiloh Y. ATM and related protein kinases: safeguarding genome integrity. Nat RevCancer 2003;3:155–68.

Bioche

S

T

T

F. Romeo et al. / The International Journal of

omasundaram K, Zhang H, Zeng YX, Houvras Y, Peng Y, Zhang H, et al. Arrestof the cell cycle by the tumour-suppressor BRCA1 requires the CDK inhibitorp21WAF1/CiP1. Nature 1997;389:187–90.

ing NS, Lee WH. The DNA double-strand break response pathway: becoming moreBRCAish than ever. DNA Repair (Amst) 2004;3:935–44.

rojan J, Zeuzem S, Randolph A, Hemmerle C, Brieger A, Raedle J, et al. Func-tional analysis of hMLH1 variants and HNPCC-related mutations using a humanexpression system. Gastroenterology 2002;122:211–9.

mistry & Cell Biology 43 (2011) 1754– 1763 1763

Wang Y, Cortez D, Yazdi P, Neff N, Elledge SJ, Qin J. BASC, a super complex ofBRCA1-associated proteins involved in the recognition and repair of aberrantDNA structures. Genes Dev 2000;14:927–39.

Wang Q, Zhang H, Guerrette S, Chen J, Mazurek A, Wilson T, et al. Adenosinenucleotide modulates the physical interaction between hMSH2 and BRCA1.Oncogene 2001;20:4640–9.

Zhou BB, Elledge SJ. The DNA damage response: putting checkpoints in perspective.Nature 2000;408:433–9.