<<

(2003) 22, 8952–8955 & 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc

Repression of mRNA for the PLK after DNA damage requires BRCA1

Anne Hansen Ree*,1,2,A˚ se Bratland1, Ragnhild V Nome1, Trond Stokke3 and Øystein Fodstad1

1Department of Tumor Biology, The Norwegian Radium Hospital, 0310 Oslo, Norway; 2Department of Oncology, The Norwegian Radium Hospital, 0310 Oslo, Norway; 3Department of Biophysics, The Norwegian Radium Hospital, 0310 Oslo, Norway

DNA damage activates the G2 cell cycle checkpoint to complex. Upon mitotic entry, these inhibitory phos- allow time for DNA repair before mitotic entry. The phorylations are removed by the Cdc25C . mechanism involves inhibition of the enzymatic activity for The activation of Cdc25C requires positive regulatory polo-like 1 (), rendering Cdc25C with a basal , accomplished by polo-like kinase 1 phosphatase activity that is insufficient for converting (Plk1) (Roshak et al., 2000). DNA damage, activating Cdc2 to the fully active G2/M transition kinase. We found G2 checkpoint ATM (for ‘ataxia telangiectasia that cell cycle arrest at the G2/M boundary after ionizing mutated’) kinase signaling (Bakkenist and Kastan, radiation (IR) of breast carcinoma cells may involve 2003) communicated through the downstream CHEK1 repression of the gene for Plk1, PLK, mediated by the and CHEK2 (Matsuoka et al., 1998; Yarden tumor-suppressor BRCA1. The -defective MT- et al., 2002), leads to inhibition of Plk1 activity (Smits 1 cell line had an apparent accumulation of G2/M phase et al., 2000; van Vugt et al., 2001) and, consequently, cells 12 h after irradiation. This response was preceded by disruption of Cdc25C interaction with Cdc2 (Sanchez a transient downregulation of PLK mRNA expression et al., 1997; Matsuoka et al., 1998; Lopez-Girona et al., with a barely detectable level 6 h after exposure to IR but 1999). recovered after 12 h. A significantly lower fraction of Polo-like kinases are evolutionary conserved À/À irradiated BRCA1 HCC1937 cells arrested in the G2/ that are involved at several stages throughout mitotic M phase after 12 h, and the transient response of PLK progression (Nigg, 2001). Most of the evidence im- mRNA was also considerably impaired. After reconstitu- plicating polo-like kinase in the activation of Cdc25C tion of wild-type BRCA1 in the HCC1937 cells however, has evolved from work on Xenopus meiotic maturation downregulation of PLK mRNA as well as Plk1 protein (Kumagai and Dunphy, 1996). Still, recent data support expression after IR was restored. Moreover, the suppres- a regulatory role of the mammalian Plk1 at the cell cycle sion of PLK mRNA expression 6 h after irradiation was G2 checkpoint (Smits et al., 2000; van Vugt et al., 2001), completely abolished by the specific CHEK1 kinase although the finding that Cdc25cÀ/À mice did not display inhibitor UCN-01, further indicating that the effector any obvious abnormalities in cellular response to DNA mechanism of DNA damage on PLK signals through damage nor ability to activate Cdc2a (Chen et al., 2001) BRCA1 and its downstream CHEK1. Our observations may argue against an essential function of Cdc25C in provide new information about the diversity of regulatory mammalian G2/M transition. The Plk1 was mechanisms governed by BRCA1 in DNA damage identified on basis of its high expression levels in rapidly checkpoint control. proliferating cells (Holtrich et al., 1994). More recent Oncogene (2003) 22, 8952–8955. doi:10.1038/sj.onc.1207000 data have suggested that elevated Plk1 expression in the primary tumor is associated with an unfavorable clinical Keywords: BRCA1; polo-like kinase 1; cell cycle con- outcome (Knecht et al., 1999; Kneisel et al., 2002), and trol; DNA damage; breast experimental silencing of Plk1 has been shown to inhibit proliferative capacity of tumor cells (Spa¨ nkuch-Schmitt et al., 2002; Elez et al., 2003). Efficient tumor cell cycle arrest at the G2/M boundary following IR exposure The effector mechanism of (IR) in requires intact function of the tumor-suppressor protein cancer therapy involves a temporary cell cycle delay at BRCA1 (Scully and Livingston, 2000), and here we the G2/M boundary until DNA is repaired or cell death report that the mechanism may involve repression of succeeds. In the orderly dividing cell, the transition from PLK . the to is inhibited through phosphor- Breast carcinoma cell lines were irradiated with a ylations of the Cdc2 kinase of the Cdc2/ B single dose of 8.0 Gy, representing a therapeutic option in radiation treatment of advanced cancer disease. Cell lines with defective p53 checkpoint mechanism were *Correspondence: AH Ree, Department of Tumor Biology, The Norwegian Radium Hospital, Montebello, 0310 Oslo, Norway; selected for analysis of specific G2 phase events. The E-mail: [email protected] MT-1 cell line (Engebraaten and Fodstad, 1999) Received 16 May 2003; revised 25 June 2003; accepted 15 July 2003 displayed complete absence of wild-type (wt) TP53- PLK regulation by BRCA1 AH Ree et al 8953 specific characteristics, such as radiation-induced mRNA expression of CDKN1A and (data not shown) as well as cell cycle arrest at the G1/S boundary, but had instead an apparent accumulation of G2/ M phase cells (B70% of the total cell counts) already 12 h after the radiation exposure (Figure 1). This response was preceded by a transient downregulation of PLK mRNA expression, with a barely detectable level Figure 2 Different responses of mRNAs for PLK and CCNB1 6 h after cellular exposure to the DNA-damaging after IR exposure in breast carcinoma cell lines with intact or defective BRCA1 constitution. In addition to the cell lines assayed treatment but almost recovered after 12 h (Figure 2). A in Figure 1, an HCC1937 derivative cell line stably infected with a significantly lower fraction (B40% of the total cell retroviral vector carrying a green fluorescent protein (GFP) but no counts) of irradiated BRCA1À/À HCC1937 cells (Tom- BRCA1 (the HCC1937/GFP cells; Scully et al., 1999) was linson et al., 1998), in which TP53 is mutated analyzed. Similarly to the HCC1937/BRCA1wt cell line, emission (Tomlinson et al., 1998) and G checkpoint signaling of green fluorescence by this control infected cell line, also kindly 2 provided by Dr David M Livingston, was assayed before use in via the CHEK1 kinase is abrogated (Yarden et al., 2002), experiments. The four tumor cell lines were cultured and were arrested in the G2/M phase after 12 h (Figure 1). In subsequently treated ( þ ) with 8.0 Gy of IR or left untreated (À), these cells, the response of the PLK mRNA to IR was as described in the legend to Figure 1. Following this, the cells were also considerably impaired, with essentially a slight incubated for the indicated time periods before total RNA was extracted and analyzed by standard Northern blotting technique. repression of the expression level after 6 h only (Figure 2). The human cDNA clones for PLK and CCNB1 were obtained The HCC1937 cell line is hemizygous with respect to from RZPD Deutsches Ressourcenzentrum fu¨ r Genomforschung the BRCA1 disease-mutated allele 5382insC, and GmbH (Berlin, Germany). The Northern blot membrane was also synthesizes a truncated BRCA1 protein that lacks parts hybridized to an oligonucleotide probe complementary to nucleo- of the extreme C-terminal BRCT repeats (Tomlinson tides 287–305 of human 18S rRNA, used as RNA loading control et al., 1998; Scully et al., 1999), which directly interact with transcriptional corepressors (Deng and Brodie, 2000). Consistently, in the derivative HCC1937/ BRCA1wt cell line, containing retrovirally expressed wt BRCA1 (Scully et al., 1999), the response of transient downregulation of PLK mRNA to IR was completely restored (Figure 2), as was essentially also the cell cycle G2/M phase arrest 12 h after IR exposure (B55% G2/ M phase cells counted) (Figure 1). Although the parental HCC1937 cells also showed radiation-induced accumu- Figure 1 Effect of IR on cell cycle profiles in breast carcinoma cell lation in G2/M phase, the kinetics was much slower than lines with intact or defective BRCA1 constitution. The MT-1 cells in the MT-1 cells (Figure 3). In irradiated HCC1937/ (Engebraaten and Fodstad, 1999) and the HCC1937 cells BRCA1wt cells, the more immediate G /M phase (Tomlinson et al., 1998; obtained from the American Type Culture 2 Collection), as well as the HCC1937/BRCA1wt cells, representing accumulation was closely similar to that of the MT-1 the HCC1937 parental cell line containing retrovirally expressed wt cells, but, interestingly, the delayed G2/M phase arrest of BRCA1 (Scully et al., 1999), were analyzed. The efficiency of the parental BRCA1À/À HCC1937 cell line was also plasmid uptake and resulting full-length BRCA1 protein expres- preserved after reversal of the nonfunctional BRCA1 sion in the HCC1937/BRCA1wt cell line, kindly provided by Dr David M Livingston (Dana Farber Cancer Institute, Boston, MA, (Figure 3). Residual characteristics of the USA), have been validated in the reference report (Scully et al., parental HCC1937 cells in the HCC1937/BRCA1wt 1999). Moreover, we assayed emission of green fluorescence, cells were also reported by the investigators who encoded by the retroviral vector, before the infected cell line was established the derivative cell line (Scully et al., 1999). used in experiments. All cell lines were routinely grown in RPMI Though, the HCC1937/BRCA1wt cell line has shown 1640 medium supplemented with 10% fetal bovine serum and 2.0 mm glutamine. At 24 h before start of the experiments, this valid for a variety of phenotypic properties directly medium was replaced by medium containing 5% serum and the related to an intact BRCA1 function (Scully et al., 1999; glutamine, and this experimental medium was again changed at Ganesan et al., 2002). time 0, when the cells were subjected to irradiation from a 60Co Whether Plk1 protein expression might correspond source. The delivery of 8.0 Gy of IR to the treated cells ( þ ) was done in room temperature at a rate of approximately 1.0 Gy/min, with the DNA damage-induced repression of PLK whereas the unirradiated control cells (À) simultaneously were mRNA in a BRCA1-dependent manner was assayed placed in room temperature to obtain exactly comparable by comparing the HCC1937/BRCA1wt and HCC1937 conditions. The cells were further incubated for 12 h before the cell lines 12 h after IR exposure. As seen from Figure 4, resulting cell cycle profiles were determined by flow cytometry downregulation of the Plk1 protein level was observed analysis, according to the following procedure: The cells were harvested by trypsinization and resuspended in ice-cold PBS. After in the cell line with intact BRCA1 constitution only and centrifugation, the cell pellets were fixed and stored in 100% seemed to be directly correlated to the mRNA response. methanol at À201C. The cells were subsequently stained with PLK has been shown to be among several , Hoechst 33258 in PBS and measured for their DNA content in a encoding mitotic regulators, of which the mRNA FACStar þ laser flow cytometer with excitation at 488 nm. On the histograms displayed, the peaks labeled 2N and 4N represent cells expression levels are downregulated following activa- with DNA contents characteristic for the G1 and G2/M phases of tion of the G2 cell cycle checkpoint (Crawford and the cell cycle, respectively Piwnica-Worms, 2001). The mechanism of PLK mRNA

Oncogene PLK regulation by BRCA1 AH Ree et al 8954

Figure 3 The pattern of redistribution of cell cycle phases after IR Figure 5 The IR-induced PLK repression is dependent on intact exposure is a function of the cellular BRCA1 constitution in cell CHEK1 signaling. The MT-1 and HCC1937/BRCA1wt cell lines lines with defective p53 checkpoint mechanism. The four cell lines were treated as previously described. The selective CHEK1 were cultured and subsequently treated ( þ ) with 8.0 Gy of IR or inhibitor UCN-01 (National Cancer Institute, Bethesda, MA, left untreated (À). Following this, the cells were incubated for the USA) was added in a final concentration of 100 nm to the cell media indicated time periods before cellular DNA contents were 15 min before irradiation, and the cells were further incubated in determined by flow cytometry analysis, as described in the legend the absence (À) or presence ( þ ) of the IR exposure (8.0 Gy) or to Figure 1. The distribution of cell cycle phases is displayed as UCN-01, or both, as indicated. Expression of PLK mRNA after columns representing the phase-specific percentages of each total the indicated time periods was determined by Northern blotting cell counts: shaded, cells; white, cells; black, G2/ analysis, and 18S rRNA was again used as RNA loading control, M phase cells as described in the legend to Figure 2

reduced following reconstitution of intact BRCA1 function in the HCC1937/BRCA1wt cell line. Further- more, some additional repression of CCNB1 mRNA was found in both the MT-1 and HCC1937/BRCA1wt cells 6 h after IR exposure (Figure 2). The DNA damage-induced ATM kinase signals its downstream activity via BRCA1 (Cortez et al., 1999) Figure 4 Repression of Plk1 expression after IR exposure is and the CHEK1 and CHEK2 kinases (Sanchez et al., dependent on intact BRCA1 constitution. The HCC1937/ 1997; Matsuoka et al., 1998), leading to arrest in the BRCA1wt and parental HCC1937 cell lines were cultured and G phase of the cell cycle until DNA repair is complete. treated ( þ ) with 8.0 Gy of IR or left untreated (À), as previously 2 described, and subsequently incubated for 12 h and harvested. The It was recently suggested that BRCA1 directly mediates cell pellets were lysed by the addition of ice-cold lysis buffer (50 mm the ATM activity onto CHEK1 (Yarden et al., 2002). Tris-HCl, pH 7.4, 150 mm NaCl, 1% NP-40, 0.25% sodium Hence, to strengthen a possible association between m m m m deoxycholate, 1 m EGTA, 1 m PMSF, 1 m Na3VO4,1m BRCA1 and PLK, expression of PLK mRNA after IR NaF, 1 mg/ml each of aprotinin, leupeptin, and pepstatin) and rocking for 30 min at 41C. The lysates were cleared by centrifuga- exposure was analyzed in the presence of the selective tion. Aliquots of 100 mg of total protein were separated on 10% CHEK1 inhibitor UCN-01 (100 nm). In both the MT-1 SDS–polyacrylamide gels and analyzed by standard Western and HCC1937/BRCA1wt cell lines, the suppression of blotting technique. A rabbit polyclonal antibody to Plk1 was PLK mRNA expression 6 h after IR exposure was obtained from Upstate Biotechnology (Lake Placid, NY, USA). completely abolished by UCN-01 (Figure 5). We also The filter was also hybridized to a mouse antihuman a-tubulin monoclonal antibody (Amersham Pharmacia Biotech, Uppsala, observed that these cell lines no longer arrested at the Sweden), used as a loading control G2/M checkpoint after IR in the presence of UCN-01 (data not shown), precisely as reported previously for HCC1937 cells reconstituted with wt BRCA1 (Yarden downregulation after IR exposure may comprise a et al., 2002). decrease in mRNA stability, as has been suggested for In clinical radiation therapy, a minimal tumor cell other G2 phase genes (Datta et al., 1992; Maity et al., delay in the cell cycle G2 phase is desirable to limit the 1995), or transcriptional repression possibly mediated probability of double-strand DNA break repair before by a repressor element in the PLK promoter (Uchiumi mitotic entry. Our results, indicating that the inhibition et al., 1997). We have screened a variety of irradiated of Plk1 activity in DNA damage checkpoint control cell lines on a high-density cDNA array to identify (Smits et al., 2000; van Vugt et al., 2001) principally may candidate radiation-responsive genes (Stokke et al., in comprise a gene repression mechanism, may suggest a preparation). Among cDNAs encoding cell cycle genes therapeutic benefit of PLK gene replacement therapy in on the array were the cDNAs for CCNB1, CCNA2, tumor cell radiosensitization. Pharmacological targeting CDC25C, and PLK considered directly G2 phase- of the signaling pathways involved may also be a related, and among these PLK and CCNB1 seemed to therapeutic strategy, and the specific inhibitor UCN-01 be consistently regulated by IR. Expression of cyclin B1 is among several novel drugs, designed to target the has been reported to decrease upon ectopic introduction G2 phase of the cell cycle, which are currently in pipeline of BRCA1 in cell lines with intact as well as defective for testing in early-phase clinical trials. p53 checkpoint mechanism (MacLachlan et al., 2000). In agreement with this finding, as seen from Figure 2, Acknowledgements the level of CCNB1 mRNA seemed to be somewhat We thank Dr David M Livingston for kindly providing the higher in the HCC1937 cells compared with the MT-1 HCC1937/BRCA1wt and HCC1937/GFP cell lines, as well as cells, and the higher expression level was significantly the Drug Synthesis & Chemistry Branch, Developmental

Oncogene PLK regulation by BRCA1 AH Ree et al 8955 Therapeutics Program, Division of Cancer Treatment and This work was supported by the Norwegian Research Council Diagnosis, National Cancer Institute for the gift of UCN-01. (the ‘Metastasis Research Grant’).

References

Bakkenist CJ and Kastan MB. (2003). Nature, 421, 499–506. MacLachlan TK, Somasundaram K, Sgagias M, Shifman Y, Chen M-S, Hurov J, White LS, Woodford-Thomas T and Muschel RJ, Cowan KH and El-Deiry WS. (2000). J. Biol. Piwnica-Worms H. (2001). Mol. Cell. Biol., 21, 3853–3861. Chem., 275, 2777–2785. Cortez D, Wang Y, Qin J and Elledge SJ. (1999). Science, 286, Maity A, McKenna WG and Muschel RJ. (1995). EMBO J., 1162–1166. 14, 603–609. Crawford DF and Piwnica-Worms H. (2001). J. Biol. Chem., Matsuoka S, Huang M and Elledge SJ. (1998). Science, 282, 276, 37166–37177. 1893–1897. Datta R, Hass R, Gunji H, Weichselbaum R and Kufe D. Nigg EA. (2001). Nat. Rev. Mol. Cell. Biol., 2, 21–32. (1992). Cell. Growth Differ., 3, 637–644. Roshak AK, Capper EA, Imburgia C, Fornwald J, Scott G Deng C-X and Brodie SG. (2000). BioEssays, 22, 728–737. and Marshall LA. (2000). Cell. Signal., 12, 405–411. Elez R, Piiper B, Kronenberger B, Kock M, Brendel M, Sanchez Y, Wong C, Thoma RS, Richman R, Wu Z, Hermann E, Pliquett U, Neumann E and Zeuzem S. (2003). Piwnica-Worms H and Elledge SJ. (1997). Science, 277, Oncogene, 22, 69–80. 1497–1501. Engebraaten O and Fodstad Ø. (1999). Int. J. Cancer, 82, Scully R and Livingston DM. (2000). Nature, 408, 429–432. 219–225. Scully R, Ganesan S, Vlasakova K, Chen J, Socolovsky M and Ganesan S, Silver DP, Greenberg RA, Avni D, Drapkin R, Livingston DM. (1999). Mol. Cell, 4, 1093–1099. Miron A, Mok SC, Randrianarison V, Brodie S, Salstrom J, Smits VAJ, Klompmaker R, Arnaud L, Rijksen G, Nigg EA Rasmussen TP, Klimke A, Marrses C, Marahrens Y, Deng and Medema RH. (2000). Nat. Cell Biol., 2, 672–676. C-X, Feunteun J and Livingston DM. (2002). Cell, 111, Spa¨ nkuch-Schmitt B, Bereiter-Hahn J, Kaufmann M 393–405. and Strebhardt K. (2002). J. Natl. Cancer Inst., 94, Holtrich U, Wolf G, Bra¨ uninger A, Karn T, Bo¨ hme B, 1863–1877. Ru¨ bsamen-Waigmann H and Strebhardt K. (1994). Proc. Tomlinson GE, Chen TTL, Stastny VA, Virmani AK, Natl. Acad. Sci. USA, 91, 1736–1740. Spillman MA, Tonk V, Blum JL, Schneider NR, Wistuba Knecht R, Elez R, Oechler M, Solbach C, von Ilberg C and II, Shay JW, Minna JD and Gazdar AF. (1998). Cancer Strebhardt K. (1999). Cancer Res., 59, 2794–2797. Res., 58, 3237–3242. Kneisel L, Strebhardt K, Bernd A, Wolter M, Binder A and Uchiumi T, Longo DL and Ferris DK. (1997). J. Biol. Chem., Kaufmann R. (2002). J. Cutan. Pathol., 29, 354–358. 272, 9166–9174. Kumagai A and Dunphy WG. (1996). Science, 273, van Vugt MATM, Smits VAJ, Klompmaker R and Medema 1377–1380. RH. (2001). J. Biol. Chem., 276, 41656–41660. Lopez-Girona A, Furnari B, Mondesert O and Russell P. Yarden RI, Pardo-Reoyo S, Sgagias M, Cowan KH and (1999). Nature, 397, 172–175. Brody LC. (2002). Nat. Genet., 30, 285–289.

Oncogene