Characterization of the role of a trimeric protein phosphatase complex in recovery from cisplatin-induced versus noncrosslinking DNA damage ´ Cristina Vazquez-Martin, John Rouse and Patricia T W Cohen Medical Research Council Protein Phosphorylation Unit, College of Life Sciences, University of Dundee, UK Keywords cisplatin; histone 2AX; methylmethanesulfonate; PPH3; protein phosphatase Correspondence P T W Cohen, MRC Protein Phosphorylation Unit, College of Life Sciences, Sir James Black Centre, University of Dundee, Dow Street, Dundee DD1 5EH, UK Fax: +44 1382 223778 Tel: +44 1382 384240 E-mail: p.t.w.cohen@dundee.ac.uk (Received 29 April 2008, revised 11 May 2008, accepted 23 June 2008) doi:10.1111/j.1742-4658.2008.06568.x Cisplatin (cis-diamminedichloroplatinum) and related chemotherapeutic DNA-crosslinking agents are widely used to treat human cancers Saccharomyces cerevisiae with separate deletions of the genes encoding the trimeric protein serine ⁄ threonine phosphatase (Pph)3p–platinum sensitivity (Psy)4p–Psy2p complex, are more sensitive than the isogenic wild-type (WT) strain to cisplatin We show here that cisplatin causes an enhanced intra-S-phase cell cycle delay in these three deletion mutants The C-terminal tail of histone 2AX (H2AX) is hyperphosphorylated in the same mutants, and Pph3p is found to interact with phosphorylated H2AX (cH2AX) After cisplatin treatment is terminated, pph3D, psy4D and psy2D mutants are delayed as compared with the WT strain in the dephosphorylation of Rad53p In contrast, only pph3D and psy2D cells are more sensitive than WT cells to methylmethanesulfonate, a noncrosslinking DNA-alkylating agent that is known to cause a Rad53p-dependent intra-S-phase cell cycle delay Dephosphorylation of Rad53p and the recovery of chromosome replication are delayed in the same mutants, but not in psy4D cells By comparison with their mammalian orthologues, the regulatory subunit Psy4p is likely to inhibit Pph3p catalytic activity The presence of a weak but active Pph3p–Psy2p complex may allow psy4D cells to escape from the Rad53pmediated cell cycle arrest Overall, our data suggest that the trimeric Pph3p– Psy4p–Psy2p complex may dephosphorylate both cH2AX and Rad53p, the differences lying in the more stable interaction of the Pph3 phosphatase with cH2AX as opposed to a transient interaction with Rad53p The chemotherapeutic agents cisplatin (cis-diamminedichloroplatinum) and oxaliplatin are currently used for the treatment of tumours in a variety of tissues, such as testis, ovary, lung, bowel, head, and neck These platinum-containing agents form adducts with DNA that produce intrastrand and interstrand nucleotide crosslinks [1], and are thought to be effective because the DNA damage triggers apoptosis of cancerous cells In response to these chemotherapeutic agents, ionizing radiation and chemical mutagens, activation of the DNA damage response pathways causes arrest of the cell cycle to allow time for cells to repair the DNA before the cell cycle resumes If the DNA cannot be repaired or the damage bypassed, apoptosis is initiated Protein phosphorylation plays a key role in the DNA damage response, but the protein phosphatases Abbreviations ATM, ataxia telangiectasia mutated; ATR, ataxia telangiectasia and RAD53 related; 4NQO, 4-nitroquinoline 1-oxide; FACS, fluorescent activated cell sorter; H2AX, histone 2AX; MMS, methylmethanesulfonate; PFGE, pulsed-field gel electrophoresis; Pph, Saccharomyces cerevisiae protein serine ⁄ theonine phosphatase; Ppp ⁄ PP, protein serine ⁄ threonine phosphatase in mammals and Drosophila; Ppp4c, protein phosphatase catalytic subunit (also termed PP4, PPX); Psy, platinum sensitivity; WT, wild-type; cH2AX, phosphorylated histone 2AX FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS 4211 ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response that participate in these pathways are not well delineated Several studies have implicated complexes of protein serine ⁄ threonine phosphatase (Pph)3p in the DNA damage response pathways in the budding yeast Saccharomyces cerevisiae Cells lacking Pph3p (Ydr075w) and an interacting protein Yln201c [2,3], the yeast orthologues of mammalian protein phosphatase catalytic subunit (Ppp4c) and its regulatory subunit, R3, respectively [4], were reported to be more sensitive to the chemical mutagen methylmethanesulfonate (MMS), which alkylates DNA [5] These mutants were also found to be more sensitive to cisplatin and oxaliplatin, so Yln201c was designated platinum sensitivity (Psy2)p [6] Surprisingly, Ybl046w was not identified being as sensitive to DNA-damaging agents in this screen, although it had been identified in Pph3p complexes by systematic analyses of the yeast proteome [2,3], and its putative mammalian homologue R2 had been identified as a core regulatory subunit in complexes with Ppp4c [7] However, subsequent investigations showed that strains deleted for Pph3p, Psy2p and Ybl046w were all sensitive to cisplatin [4,8], and Ybl046w was designated Psy4p [9] One of the earliest events in the cellular response to many DNA-damaging agents is the phosphorylation of histone 2AX (H2AX) at Ser129 in its C-terminal tail and the accumulation of the phosphorylated histone (cH2AX) at the sites of DNA damage [10,11] In the deletion mutants pph3D, psy4D and psy2D, Keogh et al [12] found that H2AX was hyperphosphorylated as compared with the wild-type (WT) strain in both the absence and the presence of ionizing radiation that caused a G2 ⁄ M cell cycle arrest, implicating Ph3p– Psy4p–Psy2p in the dephosphorylation of cH2AX and efficient recovery from the checkpoint Substitution of the H2AX Ser129 by Ala restored the ability of the pph3D strain to turn off checkpoint signalling in a timely manner and re-enter the cell cycle after DNA repair of the double-strand break [12] Although the studies of Hanway et al [5] showed that pph3D and psy2D cells were more sensitive to MMS than were WT cells, Keogh et al [12] reported that the pph3D, psy4D and psy2D cells were only more sensitive than WT cells to MMS if the cells also carried deletions of genes involved in recombination and repair of DNA While our studies were in progress, O’Neill et al [13] found that, during recovery from MMS-induced DNA damage, Rad53p dephosphorylation and resumption of DNA synthesis are delayed in pph3D and psy2D cells as compared with WT cells but not in psy4D cells Consistent with this report, earlier studies had reported that Rad53p, a key controller of the DNA damage response pathways leading to intra-S-phase cell cycle 4212 arrest, interacted with Psy2p in a yeast two-hybrid screen [14] Here we examine the roles of Pph3p, Psy2p and Psy4p in response to the DNA-crosslinking agent cisplatin and the noncrosslinking, alkylating agent MMS, and show that recovery after cisplatin-induced DNA damage is delayed in pph3D, psy4D and psy2D mutant cells, whereas it is only delayed in pph3D and psy2D cells after MMS-induced DNA damage Results Role of Pph3p and its regulatory subunits Psy4p and Psy2p in the cell cycle arrest induced by cisplatin and MMS In order to determine how the members of Pph3p complex affect the S cerevisiae cell cycle on treatment with cisplatin, we subjected the WT diploid BY4743 cells and the pph3D, psy4D and psy2D mutant cells to analysis on a fluorescent activated cell sorter (FACS) machine after incubation with mm cisplatin for various times Figure 1A shows that at 90 and 120 min, the pph3D, psy4D and psy2D mutants were appreciably affected by the drug, with an increased fraction of cells accumulating in S-phase, whereas the WT cells were largely unaffected Although the evidence for an intraS-phase cell cycle delay is less conclusive in psy4D cells than in pph3D and psy2D cells, examination of growth on YPD plates demonstrates that deletion of any component of Pph3p–Psy4p–Psy2p decreases cell proliferation in the presence of cisplatin, and that this trimeric Pph3p complex confers resistance to cisplatin (Fig 1B) [4,8,9] MMS slows progression through S-phase in WT S cerevisiae by methylation of DNA on N7-deoxyguanine and N3-deoxyadenine [15,16] The differing reports as to whether pph3D, psy4D and psy2D mutants were more sensitive than WT cells to MMS [5,12] led us to compare the sensitivity of the mutants to increasing concentrations of MMS, and our results are in line with those of Hanway et al [5] We demonstrated that pph3D and psy2D mutants were more sensitive than the WT strain when inoculated onto YPD plates containing 0.03% MMS, whereas the psy4D mutant was no more sensitive than the WT strain to 0.03% MMS on YPD plates (Fig 1C) We also treated the cells in liquid culture (0.03% MMS) and plated them on YPD plates, obtaining the same results (data not shown) In addition, we investigated the sensitivity to another DNA-damaging agent, 4-nitroquinoline 1-oxide (4NQO), which also affects S-phase The pph3D and psy2D mutants were more sensitive than the WT strain to the UV mimetic 4NQO (20 lgỈmL)1 on plates), FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response A pph3 psy4 psy2 2C 4C WT 2C 4C 2C 4C Cisplatin 120 90 60 30 Untreated DNA content: 2C 4C pph3 B WT MT WT MT Fig Pph3p complexes confer resistance to DNA-damaging agents (A) The diploid WT strain BY4743 and mutants pph3D, psy4D and psy2D were grown to mid-exponential phase in liquid culture and incubated in the presence of cisplatin (2 mM) or left untreated Samples for FACS analysis were taken at the indicated time points (B, C) The sensitivities of the same WT and mutant (MT) cells to cisplatin and MMS were examined Serial dilutions of independent colonies (WT 1, and 3) on all plates are from the control BY4743 (Y20000) strain Serial dilutions of independent colonies (MT 4, and 6) are from strains pph3D (Y34010, left columns), psy4D (Y33072, middle columns), and psy2D (Y32011, right columns) The cisplatin and MMS concentration in each row of plates is indicated on the right Plates were incubated at 30 °C for 48 h WT MT WT MT whereas the psy4D mutant showed a similar sensitivity to that of the WT strain (data not shown) These results appear to support a role for Pph3p–Psy2p in conferring resistance to MMS and 4NQO, and are in contrast to the effects of cisplatin, which imply a role for Pph3p–Psy4p–Psy2p in conferring resistance to cisplatin (Fig 1B) [8,9] S cerevisiae cells often respond to DNA damage by activating a signal transduction pathway that leads to phosphorylation of Rad9p by Mec1p Rad9p phosphorylation allows recruitment of Rad53p to Mec1– Rad9, facilitating Rad53p phosphorylation by Mec1 Rad53p autophosphorylation then leads to an active Cisplatin concentration mM mM pph3 MT psy2 6 C WT psy4 psy4 psy2 MMS (v/v) concentration 0% 0.01% 0.03% 10-fold yeast dilution Rad53p that is multiply phosphorylated [17], and the phosphorylation can be detected by SDS ⁄ PAGE and immunoblotting as bands that migrate more slowly than Rad53p As the protein kinase Rad53p was reported to interact with Psy2p [18], we investigated whether cisplatin treatment caused the activation of Rad53p in WT, pph3D, psy4D and psy2D cells, but we found that Rad53p was not phosphorylated in response to mm cisplatin [9], although this concentration could induce a cell cycle delay (Fig 1A) Use of higher concentrations can be a problem due to cisplatin insolubility, but we have now been able to demonstrate phosphorylation of Rad53p after treat- FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS 4213 ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response ment of yeast cells with mm cisplatin (Fig 2A, top left panel) On examining the recovery after different times, we could see a slight decrease in the Rad53p phosphorylation in the WT strain after h, a pronounced decrease at h, and a return to the unphosphorylated basal state by h In contrast, Rad53p was phosphorylated at h and remained at least partially phosphorylated in pph3D, psy4D and psy2D mutants during recovery up to h The experiment was performed five times In response to MMS treatment, DNA damageinduced phosphorylated forms of Rad53p occurred in the WT strain and in pph3D, psy4D and psy2D mutant cells (Fig 2B, top left panel) On examining the recovery after different times, we saw a pronounced decrease in Rad53p phosphorylation in the WT and psy4D strains at 5–7 h In contrast, phosphorylated forms of Rad53p remained in the pph3D and psy2D strains during recovery between h and h This experiment was performed four times Our data suggest that a complex of Pph3p and Psy2p may dephosphorylate Rad53p after it has been activated and phosphorylated in response to MMS, but that cisplatin-induced activation and phosphorylation of Rad53p may require a complex of Pph3, Psy4p and Psy2p for dephosphorylation to occur as rapidly as in WT cells Considering that Pph3p dephosphorylates Rad53p, we investigated the interaction between Pph3p–Psy4p–Psy2p dephosphorylates cH2AX The form of H2AX phosphorylated at Ser129 (termed cH2AX), which is induced in response to DNA-damaging agents, is normally removed before resumption of the cell cycle, and Keogh et al [12] have presented evidence that Pph3p–Psy4p–Psy2p is involved in this process In untreated pph3D, psy4D and psy2D cells, we found that the phosphorylation of H2AX was markedly elevated as compared with the barely detectable levels of cH2AX in WT cells (Fig 3A,B) Treatment with cisplatin or MMS elevated cH2AX in WT cells and led to a further increase in the pph3D, psy4D and psy2D cells (Fig 3A, cisplatin treatment; Fig 3B, MMS h recovery) After removal of cisplatin and MMS, the cH2AX levels returned to the near-zero basal levels in the WT cells after several hours but remain elevated in the pph3D, psy4D and psy2D cells (Fig 3A, h; Fig 3B, 12 h) These results indicate that the three subunits of Pph3p–Psy4p–Psy2p acting psy2 psy4 pph3 WT Recovery h psy2 psy4 psy2 WT pph3 Recovery h psy4 WT Cisplatin psy2 psy4 WT pph3 UN pph3 A Rad53p and Pph3p–Psy4p–Psy2p by coimmunoprecipitation In untreated AY925 cells, or cells treated with 0.03% MMS to induce Rad53p phosphorylation, Rad53p was not coimmunoadsorbed with HA3–Pph3p and Psy4p–MYC13 (Table 1, and data not shown), indicating that this interaction may be transient or too weak to withstand the isolation protocol Rad53P Rad53 Rad53P Rad53 psy2 psy4 pph3 Recovery h WT psy2 psy4 pph3 WT Recovery h Rad53P Rad53 B psy2 psy4 pph3 Recovery h WT psy2 psy4 WT psy2 psy4 pph3 WT pph3 Recovery h MMS psy2 psy4 WT pph3 UN Rad53P Rad53 4214 psy2 psy4 pph3 WT Recovery 10 h psy2 psy4 WT Rad53P Rad53 pph3 Recovery h psy2 psy4 pph3 Recovery h WT psy2 psy4 pph3 WT Recovery h Rad53P Rad53 Rad53P Rad53 Fig Relationship between Rad53p and the Pph3p complex The diploid WT strain BY4743 and mutants pph3D, psy4D and psy2D were grown to mid-exponential phase in liquid culture and left untreated (UN) or incubated in the presence of mM cisplatin (A) or 0.03% MMS (B) for 90 Cells were filtered, washed free of cisplatin, and incubated at 30 °C in YPD, and samples were taken at the indicated times during recovery Trichloroacetic acid extracts were prepared and subjected to immunoblot analysis with antibodies against Rad53p, which detect the unphosphorylated Rad53p (92 kDa) and more slowly migrating phosphorylated forms FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response psy2 psy4 WT pph3 Recovery h psy2 psy4 WT pph3 Recovery h psy2 psy4 pph3 WT CDDP psy2 psy4 UN pph3 WT A H2AX H2AX H2A/H2B psy2 psy4 WT pph3 Recovery h psy2 psy4 pph3 WT Recovery h H2A/H2B H2AX H2A/H2B psy2 psy4 WT psy2 MMS psy4 WT pph3 UN H2AX H2A/H2B psy2 psy4 pph3 Recovery 12 h WT psy2 psy4 pph3 Recovery h WT psy2 psy4 pph3 Recovery h WT Fig The Pph3p complex is required for dephosphorylation of cH2AX The diploid WT strain BY4743 and mutants pph3D, psy4D and psy2D were grown to mid-exponential phase in liquid culture and left untreated (UN) or incubated in the presence of mM cisplatin (CDDP) (A) or 0.03% MMS (B) for 90 Cells were filtered, washed free of cisplatin, and incubated at 30 °C in YPD, and samples were taken at the indicated times during recovery Trichloroacetic acid extracts were prepared and subjected to immunoblot analysis with antibodies against cH2AX (H2AXphosphoSer129) Lower panels are blots probed for total H2A ⁄ H2B as a control (C) The Pph3p complex associates with cH2AX Lysates from control WT AY925 cells and AY925 HA3–Pph3p cells untreated or treated with MMS at the indicated concentrations were immunoadsorbed with antibodies to HA and probed for cH2AX Lysate (L), 50 lg; the supernatant (S, same volume as lysate) and pellet (P, from mg of lysate) were obtained by centrifugation following the immunoadsorption pph3 B H2AX H2A/H2B C together normally dephosphorylate cH2AX and are essential for its complete dephosphorylation Taking into account a role for Pph3p–Psy4p–Psy2p in dephosphorylation of cH2AX, we investigated whether there was any interaction between the phosphatase and the histone Employing lysates from the and S cerevisiae cells AY925 PSY4–MYC13 AY925HA3–PPH3, cH2AX was found in the anti-HA immunopellets (Fig 3C bottom panel) and in the antiMYC immunopellets (data not shown), but not in the immunopellets from the control cells that did not express the tagged proteins (Fig 3C, top panel) HA3–Pph3p interacted with increasing amounts of cH2AX generated in the presence of increasing doses of MMS (Fig 3C) and even with the low levels of cH2AX that were sometimes present in yeast cell lysates in the absence of DNA-damaging agents (data not shown) Most of the cH2AX is bound to the phosphatase even after treatment with 0.5% MMS (Fig 3C, bottom panel) Given that endogenous Pph3p–Psy4p–Psy2p is also present in the cells, it appears likely that all of the cH2AX is bound to the phosphatase No increases in the levels of the catalytic subunit HA3–Pph3p or the regulatory subunit Psy4p– MYC13 (data not shown) were observed after DNA damage induced by MMS FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS 4215 ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response Treatment for 90 MMS UN MMS CDDP UN CDDP pph21 WT H2AX H2A/H2B UN MMS CDDP UN H2AX MMS pph21 WT MMS UN MMS CDDP CDDP pph21 WT UN Recovery h CDDP Recovery h H2AX H2A/H2B H2A/H2B Pph21p and Pph22p are protein phosphatases that are highly related to Pph3p, and in some cases show overlapping function with Pph3 [19] S cerevisiae cells with deletions of the genes encoding both Pph21p and Pph22p have extremely low viability As the most abundant isoform is Pph21p, we examined pph21D cells in order to determine whether Pph21p may be partly responsible for the dephosphorylation of cH2AX in yeast We treated WT and pph21D cells with mm cisplatin or 0.03% MMS, and also examined the recovery after washing the cultures extensively The levels of cH2AX phosphorylation revealed no differences in the dephosphorylation of cH2AX in pph21D cells as compared with the WT cells, either before or after treatment or during recovery (Fig 4) Fig Pph21p (PP2Ac ortholog) is not involved in cH2AX dephosphorylation in yeast The WT strain BY4743 and the pph21D mutant were grown to mid-exponential phase in liquid culture and incubated in the presence of cisplatin (CDDP) (2 mM) or MMS (0.03%), or left untreated (UN), for 90 Cells were filtered, washed free of the drug, and incubated at 30 °C in YPD, and samples were taken at the indicated times during recovery Trichloroacetic acid extracts were prepared and subjected to immunoblot analysis with antibodies against cH2AX Lower panels: blots were stripped and immunostained for total H2A ⁄ H2B as a control Role of Pph3p and its regulatory subunits Psy4p and Psy2p in recovery of chromosome replication following MMS-induced DNA damage Fig The Pph3p complex is required for recovery of chromosome replication (A) The WT haploid strain BY4741 and mutants pph3D, psy4D and psy2D were grown to mid-exponential phase, arrested in G1 with a-factor, released into S-phase, and treated with 0.05% MMS for 90 Cells were filtered, washed extensively, and incubated in YPD at 30 °C for h Samples for PFGE were taken after a-factor treatment (20 lgỈmL)1) (control), after MMS treatment and following removal of MMS after recovery for 2, and h cH2AX is believed to play a central role in the recruitment and ⁄ or retention of DNA repair factors at the sites of DNA damage [20] In order to investigate whether a Pph3p complex is required for the recovery of chromosome replication following removal of DNA-damaging agents, we employed pulsed-field gel electrophoresis (PFGE) Cells were arrested in G1 with a-factor, released into S-phase, and then treated with MMS for 90 The drug was washed away and cells were allowed to recover At various times, chromosomes were prepared from WT, pph3D, psy4D and psy2D cells, and separated by PFGE These gels resolved a characteristic ladder of bands corresponding to the 16 S cerevisiae chromosomes, visualized after ethidium bromide staining (Fig 5) Treatment of cells with MMS resulted in loss of chromosome bands (Fig 5, WT, pph3D, psy4D and psy2D, h) due to the presence of forks and replication bubbles that prevent entry of the chromosomes into the gel [21,22] Treatment with MMS also sometimes resulted in the appearance of a ‘smear’ of low molecular mass DNA species that may represent some chromosome degradation (Fig 5, psy4D, h) When WT cells were washed free of the DNA-damaging agents and allowed to recover, the intact chromosomes started to reappear after h of recovery and were clearly visible at h (Fig 5, WT), indicating that the S-phase arrest had been overcome and chromosome replication had 4216 FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS ´ C Vazquez-Martin et al resumed and gone to completion When pph3D and psy2D cells were allowed to recover from exposure to MMS, intact chromosomes had barely reappeared after h of recovery, indicating that these mutants were slower in the recovery of chromosome replication (Fig 5: pph3D, h; psy2D, h) In contrast, recovery in psy4D cells was similar to that in WT cells (Fig 5, psy4D, MMS h) Thus, a correlation was observed between delayed completion of chromosome replication and prolonged MMS-induced Rad53p phosphorylation in pph3D and psy2D cells Similar results were obtained in three independent experiments Further analysis of chromosome replication following cisplatin-induced damage suggested that pph3D, psy4D and psy2D cells were all slightly delayed as compared with WT cells, but the small difference [at concentrations of cisplatin (3 mm) near its maximum solubility] was difficult to confirm (data not shown) Discussion In S cerevisiae, Pph3p has been shown to interact with regulatory subunits Psy4p and Psy2p, and these three proteins and their interactions have been found to be conserved through evolution to Drosophila and mammals [4] We have previously demonstrated that deletion of any component of yeast Pph3p–Psy4p– Psy2p causes hypersensitivity to the antitumour drug cisplatin, indicating that all three proteins may operate as a functional unit in vivo, playing a role in the cisplatin response [9] Cisplatin interferes with DNA function by causing intrastrand and interstrand crosslinking of nucleotide bases and, in replicating cells, DNA damage usually induces an intra-S-phase cell cycle arrest Accordingly, we show in this article that treatment of pph3D, psy4D and psy2D mutant cells with cisplatin causes enhanced accumulation of cells in S-phase as compared with WT cells, although the effect is less conclusive in the case of the psy4D cells Nevertheless, the data suggest that a correlation is observed between the slow growth of pph3D, psy4D and psy2D cells in the presence of cisplatin [9] and accumulation of cells in S-phase Delayed S-phase progression is usually associated with activation of the intra-S-phase checkpoint mediated by Rad53 phosphorylation, and indeed, treatment of cells with mm cisplatin for 90 resulted in phosphorylation of Rad53p (Fig 2A), although slightly lower concentrations did not activate Rad53p [9] Notably, recovery from cisplatin-induced Rad53p phosphorylation was delayed in all three mutants: pph3D, psy4D and psy2D Cisplatin-induced DNA damage response At sites of DNA damage, the Ser129-phosphorylated H2AX derivative, cH2AX, forms foci for the recruitment of factors involved in repair of DNA damage and maintenance of the cell cycle arrest H2AX was found to be hyperphosphorylated in pph3D, psy4D and psy2D strains in both the absence and the presence of ionizing radiation [12], and in the present study, in the absence and the presence of cisplatin In addition, we showed that Pph3p directly or indirectly binds to cH2AX, indicating that Pph3p–Psy4p–Psy2p forms a stable complex with H2AX when Ser129 is phosphorylated and is therefore likely to be the phosphatase complex dephosphorylating the histone C-terminal tail Keogh et al [12] have provided evidence that cH2AX is removed from the site of DNA damage before it is dephosphorylated If this is the case, removal from the action of the ataxia telangiectasia mutated (ATM) ⁄ ataxia telangiectasia and RAD53 related (ATR) kinases at the site of DNA damage may decrease the kinase ⁄ phosphatase ratio and allow the phosphatase to dephosphorylate cH2AX In mammalian cells, PP2A isoforms, the orthologues of S cerevisiae Pph21p and Pph22p, have been reported to dephosphorylate cH2AX [23], and in some cases, e.g in the mammalian target of rapamycin pathway, Pph21p and Pph22p have overlapping functions with Pph3p [24] It was therefore important to examine whether Pph21 ⁄ 22p might play a role in the dephosphorylation of cH2AX Cells with deletion of the most abundant isoform, Pph21p, exhibit depletion of phosphatase activity to 35% of the total attributable to Pph21 and Pph22 [25] In addition, these pph21D cells showed a significantly lower budding index and slightly slower growth than WT cells on nonfermentable carbon sources However, the pph21D cells showed no hyperphosphorylation of cH2AX and no delay in recovery from cisplatin and MMS as compared with WT cells, supporting the idea that in S cerevisiae, cH2AX is dephosphorylated by Pph3p and not by Pph21p and its closely related isoform Pph22p In contrast to the effects of cisplatin, which crosslinks DNA strands, our studies in which yeast cells were treated with MMS, which mainly alkylates DNA but can cause double-strand breaks [26], showed delayed dephosphorylation of Rad53p and delayed recovery of chromosome replication only in pph3D and psy2D cells as compared with WT cells, but not in the psy4D mutant The results are in line with the competitive growth assays that identified the pph3D and psy2D cells as MMS hypersensitive [5] Recently, O’Neill et al [13] found delayed recovery from MMS-induced Rad53 phosphorylation and decreased progression of replication forks along the DNA in pph3D and psy2D FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS 4217 ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response cells These authors conclude that a dimeric complex, Pph3p–Psy2p, is responsible for the dephosphorylation of Rad53p, and they further suggest that Rad53p dephosphorylation at the replication fork is necessary to allow the resumption of DNA synthesis Rad53p has been shown to interact with the lagging strand replication apparatus regulating the phosphorylation of DNA polymerase a-primase complex [27], and Psy2p was found to interact with proteins at stalled replication forks in a yeast two-hybrid screen [28] Our studies indicate that different Pph3 phosphatase complexes or more likely different phosphatases are responsible for the dephosphorylation of Rad53p after cisplatin and MMS-induced DNA damage Phosphorylation of Rad53p is detected by a mobility shift after gel electrophoresis, and multiple phosphorylation sites in Rad53p are suggested by the several bands that can sometimes be separated [13,29] Different phosphorylations of Rad53p may be triggered by the two DNAdamaging agents, and the different phosphorylation sites may then be dephosphorylated by different phosphatases The Mg2+-dependent phosphatases, Ptc2p and Ptc3p, have been implicated in the dephosphorylation of Rad53p after a G2 ⁄ M arrest in response to irreparable double-strand breaks in the DNA [29] Keogh et al [12] have suggested that dephosphorylation of Rad53p during recovery from a repairable double-strand break depends on the prior dephosphorylation of cH2AX by Pph3p–Psy4p–Psy2p [12] Our results on recovery from cisplatin-induced DNA damage are consistent with Rad53p phosphorylation at a particular site being maintained by cH2AX, and when the latter is dephosphorylated, Rad53p may be dephosphorylated by Ptc2p and Ptc3p (Fig 6) The question of whether Pph3p–Psy2p or Pph3p– Psy4p–Psy2p is involved in the dephosphorylation of Rad53p after DNA damage by MMS is interesting An alternative explanation is presented by considering the mammalian complex Ppp4c–R2–R3, which is orthologous to Pph3p–Psy4p–Psy2p The isolation of endogenous Ppp4c–R2 revealed that R2 inhibited Ppp4c, and suggested that R2 may be a core regulatory subunit that facilitates binding of further regulatory subunits to Ppp4c [7] The interaction of R3 with Ppp4c was then shown to require prior preassembly of Ppp4c and R2 [8] These observations suggest that if the complexes are conserved through evolution, Psy4p may be present in the complex that dephosphorylates Rad53p In addition, by comparison with R2, the yeast orthologue Psy4p may be an inhibitory regulatory subunit for Pph3p, so that in psy4D cells, the active Pph3p, weakly associated with Psy2p, may dephosphorylate Rad53p In psy4D cells, phosphorylated 4218 Rad53p would therefore not be present at replication forks to stall DNA synthesis Our MMS sensitivity studies (Fig 1B) could also be explained by this mechanism If we consider Psy4p as an inhibitory regulatory subunit of Pph3p, psy4D cells could escape from the Rad53p checkpoint, and grow similarly to WT cells; pph3D and psy2D cells would show slow growth because of the activation of the checkpoint The stable interaction of Pph3p and Psy2p with cH2AX may require the presence of Psy4p The interaction of Rad53p with a Pph3p complex, which we could not detect by coimmunoadsorption, would appear to be transient The different strengths of the interactions of the phosphatase complex with its substrates may underlie the nonessential nature of Psy4p for dephosphorylation of Rad53p by Pph3p in psy4D cells, although we cannot completely exclude the existence of functional dimeric complexes in WT cells Overall, our studies are consistent with a role for Pph3p–Psy4p–Psy2p in the dephosphorylation of both cH2AX and Rad53p The complex may have an additional function at stalled replication forks, but the data not necessitate such a role, as dephosphorylation of both cH2AX and Rad53p is likely to be a prerequisite for chromosome replication to resume A working model for the roles of Pph3p–Psy4p–Psy2p in the recovery from DNA damage induced by the crosslinking anticancer drug cisplatin and the noncrosslinking agent MMS is presented in Fig Our data suggest that the sites phosphorylated on Rad53p and dephosphorylated by different phosphatases may be dependent on the type of the DNA damage However, different levels of DNA damage cannot be entirely excluded, because cisplatin does not readily enter the yeast cell, and therefore it is possible that the overall amount of MMS-induced DNA damage may be higher than that caused by cisplatin Experimental procedures Yeast strains and general methods All methods for the manipulation of yeast and preparation of media were performed according to standard protocols [30] The growth conditions for the yeast strains and drug sensitivity studies were as described previously [9] The strains used in this study are listed in Table The S cerevisiae strain AY925, in which Psy4p bears a C-terminal MYC13 epitope tag and Pph3p an N-terminal HA3 tag, was constructed by a PCR-based method as described in [9] The haploid and diploid strains with deletions of genes PPH3, PSY4 and PSY2 encoding ORFs YDR075w, YBL046w and YNL201c, respectively, were from Euroscarf FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response Cisplatin DNA damage DNA damage H2AX- P ( H2AX) Recovery H2AX H2AX- P ( H2AX) Psy4 Pph3 MMS Recovery Pph3 Psy2 ATM/ATR kinases recruited Psy2 ATM/ATR kinases recruited Rad53- P1 H2AX Psy4 Rad53- P2 Rad-53 Rad-53 Ptc2/Ptc3 Psy4 Cell cycle delay Cell cycle delay Pph3 Psy2 Fig Schematic for the roles of Pph3p–Psy4p–Psy2p in recovery from cisplatin- and MMS-induced DNA damage Pph3p–Psy4p–Psy2p forms a stable complex with cH2AX via Psy4p During recovery from cisplatin- or MMS-induced DNA damage, cH2AX may be removed from the site of action of the ATM ⁄ ATR kinases, allowing Pph3p–Psy4p–Psy2p to dephosphorylate cH2AX In the case of the cisplatin-induced DNA damage response, the site(s) phosphorylated (P1) on Rad53p are dephosphorylated by the phosphatases Ptc2 and Ptc3 In the case of the MMS-induced DNA damage response, the site(s) phosphorylated (P2) on Rad53p are distinct and are dephosphorylated by a transient interaction with Pph3p–Psy4p–Psy2p Psy4p is not absolutely essential for this dephosphorylation, allowing a weakly interacting Pph3p– Psy2p complex to dephosphorylate the P2 site(s) in the psy4D cells (European Saccharomyces cerevisiae Archive for Functional Analysis), Institute for Microbiology, Johann Wolfgang Goethe-University Frankfurt, Germany Yeast extracts, immunoblotting, and immunoprecipitation Extracts for immunoblot analysis were prepared either as described previously [9] or by a slightly modified trichloroacetic acid-lysis method [31] Briefly, in the trichloroacetic acid method, the cells were washed with 20% trichloroacetic acid (v ⁄ v) and were then disrupted with 0.5 mL of 20% trichloroacetic acid (v ⁄ v) ⁄ 0.7 · 107 cells by vortexing for in the presence of glass beads in a mini-bead beater The lysates, separated from the beads, were centrifuged for at 13 000 g The pellets were resuspended in 200 lL of sample buffer adjusted to 0.3 m Tris ⁄ HCl with m Tris-HCl pH 8.8, boiled for 10 min, and clarified by centrifugation for at 13 000 g Proteins in the extracts were subjected to SDS ⁄ PAGE and transferred to nitrocellulose membranes In coimmunoprecipitation experiments, aliquots of lysates (2 mg of protein) prepared in the absence of trichloroacetic acid from cells expressing Psy4p–MYC13 and HA3–Pph3p were incubated with either anti-c-MYC or anti-HA agarose beads (Sigma-Aldrich, Poole, UK) on a shaking platform at °C for h After centrifugation for at 13 000 g, the beads were washed two times in lysis buffer containing 0.15 m NaCl and twice in 50 mm Tris ⁄ HCl (pH 7.5) and 0.1 mm EGTA The beads were boiled for in SDS sample buffer, and released proteins were subjected to SDS ⁄ PAGE (4–12% polyacrylamide) and immunoblotting with either of the monoclonal antibodies anti-MYC (Roche Diagnostics, Indianapolis, IN, USA) or anti-HA (produced in the Division of Signal Transduction Therapy, University of Dundee) Rad53p immunoblots were performed using a mixture of two antibodies (yN-19 and yC-19, Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) Antibodies to histones H2A ⁄ H2B and phosphoSer129-H2AX were from Abcam (Cambridge, UK) FACS analysis Cells (1 · 107) were resuspended in 70% (v ⁄ v) ethanol and left for at least h at °C The cells were then washed in 50 mm Tris ⁄ HCl (pH 7.8), treated with 0.2 mgỈmL)1 RNaseA (Sigma-Aldrich) at 37 °C overnight, and washed with 200 mm Tris ⁄ HCl (pH 7.5), 211 mm NaCl, and 78 mm MgCl2; propidium iodide was then added to give a concentration of 50 lgỈmL)1 in the same buffer at least h prior FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS 4219 ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response Table S cerevisiae strains used in this study Accession numbers for Euroscarf strains are cited at http://www.uni-frankfurt.de/ fb15/mikro/euroscarf/data Strain Genotype Source AY925 MATa, ade2-1, his3-11, leu2-3, trp1-1, ura3-1, can1-100 AY925 PSY4–MYC13, HA3–PPH3 BY4741 (Y00000, haploid) Y04010 Y03072 Y02011 BY4743 (Y20000, diploid) AY925 PSY4–MYC13–HISMX6, HA3–PPH3 FernandezSarabia et al [32] Hastie et al [9] Y34010 Y33072 Y32011 Y33831 MATa; his3D1; leu2D0; met15D0; ura3D0 BY4741 pph3D::kanMX4 BY4741 psy4D::kanMX4 BY4741 psy2D::kanMX4 MATa ⁄ MATa; his3D1 ⁄ his3D1; leu2D0 ⁄ leu2D0; met15D0 ⁄ MET15; LYS2 ⁄ lys2D0; ura3D0 ⁄ ura3D0 BY4743pph3D::kanMX4 ⁄ pph3D::kanMX4 BY4743psy4D::kanMX4 ⁄ psy4D::kanMX4 BY4743psy2D::kanMX4 ⁄ psy2D::kanMX4 BY4743pph21D::kanMX4 ⁄ pph21D::kanMX4 Euroscarf Euroscarf Euroscarf Euroscarf Euroscarf Euroscarf Euroscarf Euroscarf Euroscarf to FACS analysis in a Becton Dickinson FACSort machine, managed by R Clarke (University of Dundee, UK) Analyses of chromosomes by PFGE Cells were grown to early log phase (A600 nm of 0.5) in YPD at 30 °C and arrested in G1 by addition of a-factor (20 lgỈmL)1) When budded cells accounted for < 5% of the population (confirmed by FACS analysis), the cells were released from the G1 arrest by filtration and extensive washing, and this was followed by incubation in prewarmed YPD for 30 to allow entry into S-phase before addition of MMS (0.05%) or cisplatin (3 mm) After 90 in MMS or cisplatin, cells were filtered, washed extensively with YPD containing 5% (w ⁄ v) sodium thiosulfate, and incubated in YPD at 30 °C At the times indicated, · 108 cells were removed and fixed in 70% ethanol at °C overnight before preparation of chromosomes, exactly as described in the CHEF DRII instruction manual (BioRad, Hemel Hempstead, UK) PFGE was carried out using the BioRad CHEF DRII apparatus at 14 °C in a 1% agarose (pulsed field-certified BioRad) gel in 89 mm Tris, 89 mm boric acid and mm EDTA (pH 8) for 24 h at VỈcm)1 using a 120° included angle with a 6.8–158 s switch time ramp Gels were stained with lgỈmL)1 ethidium bromide for 30 and washed for h in water before the DNA was visualized 4220 Acknowledgements We thank the Medical Research Council, UK for financial support References Eastman A (1985) Interstrand cross-links and sequence specificity in the reaction of cis-dichloro(ethylenediamine)platinum(II) with DNA Biochemistry 24, 5027– 5032 Gavin A-C, Bosche M, Krause R, Grandi P, Marzioch ă M, Bauer A, Schultz J, Rick JM, Michon A-M, Cruciat C-M et al (2002) Functional organisation of the yeast proteome by systematic analysis of protein complexes Nature 415, 141–147 Ho Y, Gruhler A, Heilbut A, Bader GD, Moore L, Adams S-L, Millar A, Taylor P, Bennet K, Boutilier K et al (2002) Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry Nature 415, 180–183 ´ Cohen PTW, Philp A & Vazquez-Martin C (2005) Protein phosphatase – from obscurity to vital functions FEBS Lett 579, 3278–3286 Hanway D, Chin JK, Xia G, Oshiro G, Winzeler EA & Romesberg FE (2002) Previously uncharacterized genes in the UV- and MMS-induced DNA damage response in yeast Proc Natl Acad Sci USA 99, 10605–10610 Wu HI, Brown JA, Dorie MJ, Lazzeroni L & Brown JM (2004) Genome-wide identification of genes conferring resistance to the anticancer agents cisplatin, oxaliplatin and mitocin C Cancer Res 64, 3940–3948 Hastie CJ, Carnegie GK, Morrice N & Cohen PTW (2000) A novel 50 kDa protein forms complexes with protein phosphatase and is located at centrosomal microtubule organizing centres Biochem J 347, 845– 855 Gingras AC, Caballero M, Zarske M, Sanchez A, Hazbun TR, Fields S, Sonenberg N, Hafen E, Raught B & Aebersold R (2005) A novel, evolutionarily conserved protein phosphatase complex involved in cisplatin sensitivity Mol Cell Proteomics 4, 1725–1740 ´ Hastie CJ, Vazquez-Martin C, Philp A, Stark MJR & Cohen PTW (2006) The Saccharomyces cerevisiae orthologue of human protein phosphatase core regulatory subunit R2 confers resistance to the anticancer drug cisplatin FEBS J 273, 3322–3334 10 Stucki M & Jackson SP (2006) cH2AX and MDC1: anchoring the DNA-damage-response machinery to broken chromosomes DNA Repair (Amst) 5, 534–543 11 Nussenzweig A & Paull T (2006) DNA repair: tails of histones lost Nature 439, 406–407 12 Keogh MC, Kim JA, Downey M, Fillingham J, Chowdhury D, Harrison JC, Onishi M, Datta N, Galicia S, Emili A et al (2006) A phosphatase complex that FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS ´ C Vazquez-Martin et al 13 14 15 16 17 18 19 20 21 22 dephosphorylates cH2AX regulates DNA damage checkpoint recovery Nature 439, 497–501 O’Neill BM, Szyjka SJ, Lis ET, Bailey AO, Yates JR 3rd, Aparicio OM & Romesberg FE (2007) Pph3–Psy2 is a phosphatase complex required for Rad53 dephosphorylation and replication fork restart during recovery from DNA damage Proc Natl Acad Sci USA 104, 9290–9295 Uetz P, Giot L, Cagney G, Mansfield TA, Judson RS, Knight JR, Lockshon D, Narayan V, Srinivasan M, Pochart P et al (2000) A comprehensive analysis of protein–protein interactions in Saccharomyces cerevisiae Nature 403, 623–627 Tercero JA & Diffley JF (2001) Regulation of DNA replication fork progression through damaged DNA by the Mec1 ⁄ Rad53 checkpoint Nature 412, 553–557 Chang M, Bellaoui M, Boone C & Brown GW (2002) A genome-wide screen for methyl methanesulfonate-sensitive mutants reveals genes required for S phase progression in the presence of DNA damage Proc Natl Acad Sci USA 99, 16934–16939 Pellicioli A & Foiani M (2005) Signal transduction: how rad53 kinase is activated Curr Biol 15, R769–R771 Hazbun TR, Malmstrom L, Anderson S, Graczyk BJ, Fox B, Riffle M, Sundin BA, Aranda JD, McDonald WH, Chiu CH et al (2003) Assigning function to yeast proteins by integration of technologies Mol Cell 12, 1353–1365 Ronne H, Carlberg M, Hu G-Z & Nehlin JO (1991) Protein phosphatase 2A in Saccharomyces cerevisiae: effects on cell growth and bud morphogenesis Mol Cell Biol 11, 4876–4884 Foster ER & Downs JA (2005) Histone H2A phosphorylation in DNA double-strand break repair FEBS J 272, 3231–3240 Hennessy KM, Lee A, Chen E & Botstein D (1991) A group of interacting yeast DNA replication genes Genes Dev 5, 958–969 Desany BA, Alcasabas AA, Bachant JB & Elledge SJ (1998) Recovery from DNA replicational stress is the essential function of the S-phase checkpoint pathway Genes Dev 12, 2956–2970 Cisplatin-induced DNA damage response 23 Chowdhury D, Keogh MC, Ishii H, Peterson CL, Buratowski S & Lieberman J (2005) cH2AX dephosphorylation by protein phosphatase 2A facilitates DNA double-strand break repair Mol Cell 20, 801–809 24 Bertram PG, Choi JH, Carvalho J, Ai W, Zeng C, Chan T-F & Zheng XFS (2000) Tripartite regulation of Gln3p by TOR, Ure2p and phosphatases J Biol Chem 275, 35727–35733 25 Sneddon AA, Cohen PTW & Stark MJR (1990) Saccharomyces cerevisiae protein phosphatase 2A performs an essential cellular function and is encoded by two genes EMBO J 9, 4339–4346 26 Bilsland E & Downs JA (2005) Tails of histones in DNA double-strand break repair Mutagenesis 20, 153– 163 27 Pellicioli A, Lucca C, Liberi G, Marini F, Lopes M, Plevani P, Romano A, Di Fiore PP & Foiani M (1999) Activation of Rad53 kinase in response to DNA damage and its effect in modulating phosphorylation of the lagging strand DNA polymerase EMBO J 18, 6561– 6572 28 O’Neill BM, Hanway D, Winzeler EA & Romesberg FE (2004) Coordinated functions of WSS1, PSY2 and TOF1 in the DNA damage response Nucleic Acids Res 32, 6519–6530 29 Leroy C, Lee SE, Vaze MB, Ochsenbien F, Guerois R, Haber JE & Marsolier-Kergoat MC (2003) PP2C phosphatases Ptc2 and Ptc3 are required for DNA checkpoint inactivation after a double-strand break Mol Cell 11, 827–835 30 Amberg DC, Burke DJ & Strathern JN (2005) Methods in Yeast Genetics Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY 31 Foiani M, Liberi G, Piatti S & Plevani P (1999) Saccharomyces cerevisiae as a model system to study DNA replication In Eukaryotic DNA Replication A Practical Approach (Cotterill S, ed), pp 185–200 Oxford University Press, Oxford 32 Fernandez-Sarabia MJ, Sutton A, Zhang T & Arndt KT (1992) SIT protein phosphatase is required for the normal accumulation of SW14, CLN1, CLN2 and HSC26 RNAs during late G1 Genes Dev 6, 2417–2428 FEBS Journal 275 (2008) 4211–4221 ª 2008 The Authors Journal compilation ª 2008 FEBS 4221 ... and RAD53 related (ATR) kinases at the site of DNA damage may decrease the kinase ⁄ phosphatase ratio and allow the phosphatase to dephosphorylate cH2AX In mammalian cells, PP 2A isoforms, the. .. 2008 The Authors Journal compilation ª 2008 FEBS ´ C Vazquez-Martin et al Cisplatin-induced DNA damage response Cisplatin DNA damage DNA damage H2AX- P ( H2AX) Recovery H2AX H2AX- P ( H2AX) Psy4... MMS-induced DNA damage, cH2AX may be removed from the site of action of the ATM ⁄ ATR kinases, allowing Pph3p–Psy4p–Psy2p to dephosphorylate cH2AX In the case of the cisplatin-induced DNA damage response,