Enhanced expression of MYCN leads to centrosome ... - Nature

Oncogene (2004) 23, 1005–1009

& 2004 Nature Publishing Group All rights reserved 0950-9232/04 $25.00 www.nature.com/onc

Enhanced expression of MYCN leads to centrosome hyperamplification after DNA damage in neuroblastoma cells Eiji Sugihara1, Masayuki Kanai1, Akira Matsui2, Masafumi Onodera3, Manfred Schwab4 and Masanao Miwa*,1 1 Department of Biochemistry and Molecular Oncology, Institute of Basic Medical Sciences, University of Tsukuba, 1-1-1, Tennoudai, Tsukuba Science City, Ibaraki 305-8575, Japan; 2Department of Pediatrics, Institute of Clinical Medicine, University of Tsukuba, Ibaraki, Japan; 3Department of Hematology, Institute of Clinical Medicine, University of Tsukuba, Ibaraki, Japan; 4Department of Tumour Genetics (B-030), Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 280, Heidelberg 69120, Germany

Centrosomes play important roles in cell polarity, regulation of cell cycle and chromosomal stability. Centrosome abnormality is frequently found in many cancers and contributes to chromosomal instability (including aneuploidy, tetraploidy, and/or micronuclei) in daughter cells through the assembly of multipolar or monopolar spindles during mitosis. It has recently been reported that loss of tumor suppressor genes or overexpression of oncogenes causes centrosome hyperamplification. Amplification and overexpression of the MYCN oncogene is found in a subgroup of neuroblastomas. In this study, we examined whether overexpression of MYCN causes centrosome hyperamplification in neuroblastoma cells. We show that ectopic expression of MYCN alone in a neuroblastoma cell line did not cause centrosome hyperamplification. However, centrosome hyperamplification and micronuclei formation were seen in these cells after DNA damage. These findings suggest that overexpression of MYCN abrogates the regulation of the centrosome cycle after DNA damage. Oncogene (2004) 23, 1005–1009. doi:10.1038/sj.onc.1207216 Published online 1 December 2003 Keywords: damage

MYCN; centrosome; micronuclei; DNA

The centrosome functions as a microtubule organizing center, regulating cell polarity and acting as spindle poles to distribute chromosomes to daughter cells equally (Doxsey, 2001). Since each daughter cell receives only one centrosome, the centrosome must duplicate once per each cell cycle. Centrosome duplication begins near the G1/S transition, and is completed in the G2 phase. The abrogation of the regulatory mechanisms that ensures the coordinated progression of centrosome duplication, that prevents reduplication of centrosome within the same cell cycle, and that ensures progression of cytokinesis, should result in centrosome hyperamplification (Brinkley and Goepfert, 1998; Fukasawa, 2002). *Correspondence: M Miwa; E-mail: [email protected] Received 11 June 2003; revised 5 August 2003; accepted 25 August 2003

Hyperamplified centrosomes lead to increased frequency of multipolar mitotic spindles and imbalanced segregation of chromosomes into daughter cells as observed in cancer cells (Lingle et al., 1998; Pihan et al., 1998; Carroll et al., 1999; Sato et al., 1999). Recent studies have revealed that loss of tumor suppressor genes including p53 (Fukasawa et al., 1996), p21 (Mantel et al., 1999), BRCA1 (Xu et al., 1999), and BRCA2 (Tutt et al., 1999), or overexpression of oncogenes including mdm2 (Carroll et al., 1999), v-ras (Saavedra et al., 1999), Aurora A (Zhou et al., 1998), and the human papilloma virus E6 and E7 oncoproteins (Duensing et al., 2000) cause centrosome hyperamplification and chromosomal instability. Since the above tumor-related genes are involved in the cell cycle and centrosome function, dysregulation of these genes would cause centrosome hyperamplification. Neuroblastoma, a most frequently found solid tumor of early childhood, originates from neural crest-derived cells and is associated with genomic instability, including ploidy changes and gains or losses of certain chromosomal regions (Westermann and Schwab, 2002; Schwab et al., 2003). Amplification of the MYCN oncogene is one of the most prominent genomic abnormalities being found in a subgroup of neuroblastoma (Schwab et al., 1983; Savelyeva and Schwab, 2001). Overexpression of MYCN promotes S phase entry of neuroblastoma cells and postmitotic sympathetic neurons (Lutz et al., 1996, Wartiovaara et al., 2002), over-rides G1 arrest after g-irradiation (Tweddle et al., 2001), and also abrogates p53-mediated cell cycle arrest in S phase-arrested cells (Chernova et al., 1998). Since the centrosome cycle should strictly cooperate with the cell cycle, the above reports raise a possibility that dysregulation of MYCN expression can cause abnormality of the centrosome cycle. In this paper, we examined whether overexpression of MYCN causes centrosome hyperamplificaiton and chromosomal instability. To analyse whether enhanced expression of MYCN causes centrosome hyperamplification, we transduced MYCN cDNA into a neuroblastoma cell line SH-EP that does not show amplification of MYCN (Breit and Schwab, 1989) using the retroviral vector (pGCDNsa-

MYCN and centrosome hyperamplification E Sugihara et al

1006

mIRES-EGFP) derived from a murine stem cell virus (Suzuki et al., 2002). This retroviral vector expresses at the same time a gene of interest and an enhanced green fluorescence protein (EGFP) using internal ribosome entry site (IRES). MCYN-transduced cells and control vector-transduced cells were termed MYCN-EGFP and vector-EGFP cells, respectively. MYCN protein was clearly demonstrated with MYCN-EGFP cell extracts but not with vector-EGFP cell extracts on immunoblot analysis (Figure 1A). In order to count centrosome numbers, vector-EGFP and MYCN-EGFP cells were immunostained for g-tubulin, a centrosome marker (Figure 1B). Quantitative analysis showed that MYCN-EGFP cells did not show an increase of cells with centrosome hyperamplification (1.33%) as compared with vector-EGFP cells (1.56%) or the parental SH-EP cells (1.70%) (Figure 1C). Thus, enhanced expression of MYCN alone could not induce centrosome hyperamplification. Recently, it was reported that g-irradiation causes centrosome hyperamplification in osteosarcoma, colon, pancreatic, breast, and uterine cancer cell lines (Sato et al., 2000). Since overexpression of MYCN abrogates G1 arrest after g-irradiation in neuroblastoma cell lines (Tweddle et al., 2001), we assumed that enhanced expression of MYCN induces centrosome hyperamplification after g-irradiation in neuroblastoma cells. To examine this hypothesis, MYCN-EGFP and vectorEGFP cells were irradiated with 10 Gy of g-irradiation and immunostained for g-tubulin at each time point. During 24–48 h after g-irradiation, centrosome hyperamplification was frequently observed in MYCN-EGFP cells (Figure 2A, a). Both vector-EGFP and MYCNEGFP cells showed increased numbers of cells with centrosome hyperamplification after g-irradiation (Figure 2B). Remarkably, from 24 h after g-irradiation MYCN-EGFP cells showed significant increase of the number of cells with centrosome hyperamplification as compared to vector-EGFP cells (Po0.01, Fisher’s exact probability test). In addition, abnormal tripolar mitotic MYCN-EGFP cells were also observed after 72 h (Figure 2A, b). There were 14 and 3 mitotic cells with multipolar spindles out of 36 and 32 mitotic cells in MYCN-EGFP cells and vector-EGFP cells, respectively, 72 h after g-irradiation (Po0.005). To confirm the above findings, we treated vectorEGFP and MYCN-EGFP cells with aphidicolin (Aph) as another DNA-damaging agent. Although Aph is known to induce S phase arrest, long-term Aph treatment stimulates DNA breaks (Toledo et al., 2000). The vector-EGFP and MYCN-EGFP cells were treated with 0.5 mg/ml of Aph and immunostained with phosphorylated histone H2AX (g-H2AX) as an indicator of DNA damage (Rogakou et al., 1998). Almost all vector-EGFP (95.5%) and MYCN-EGFP (91.8%) cells were positive for dots of g-H2AX 48 h after Aph treatment (Figure 3A, a, b and c). The MYCN-EGFP cells with hyperamplified centrosomes were frequently observed from 48 h after Aph treatment (Figure 3B, a). From 48 h after Aph treatment, MYCN-EGFP cells showed significant increase of the number of cells with centroOncogene

Figure 1 Enhanced expression of MYCN alone using retroviral vector did not cause centrosome hyperamplification. The MYCN cDNA was cloned in a retroviral vector, pGCDNsamIRES-EGFP. The retroviral DNAs were cotransfected with pcDNA3-VSV-G (vesicular stomatitis virus G protein) vector DNA into the packaging cell line 293 gp using Lipofectamine plus according to the manufacturer’s protocol (Invitrogen). After 24 h, the supernatant was recovered and infected to the SH-EP cells with 10 mg/ml of protamine (Sigma). After 24 h of incubation, the supernatant was replaced by fresh medium. EGFP-expressing cells were sorted by FACSsVantage SE (Beckton Dickinson) and expanded for subsequent experiments. (A) Western blot analysis of MYCN (1 : 500, purified mouse anti-human N-myc Antibody, Pharmingen) and EGFP (1 : 100, Living Colorss A.v. Peptide Antibody, Clontech) expressions in vector-EGFP and MYCN-EGFP cells. The cells were lysed with a RIPA buffer (20 mM Tris-HCl (pH 7.5), 5 mM EDTA, 150 mM NaCl, 1% NP-40, 0.025% SDS) and proteinase inhibitors (EDTA-free proteinase inhibitor cocktail, Roche Diagnostics). Total proteins (30 mg) per lane were applied. CBB staining shows the equal amount of proteins per lane. (B) Immunostaining for g-tubulin. Cells grown on coverslips were fixed with methanol for 20 min at 201C, washed with phosphatebuffered saline (PBS), and permeabilized with 1% NP-40 in PBS for 5 min at 251C. Cells were blocked with 10% normal goat serum in PBS for 1 h and probed with anti-g-tubulin monoclonal antibody (1 : 400, GTU-88, Sigma) for 1 h at 251C. The antibody–antigen complexes were detected with TRITC-conjugated goat anti-mouse IgG antibody (1 : 400, Sigma) by incubation for 1 h at 251C. The samples were counterstained with Hoechst No. 33342 (Sigma). These panels show merge of DNA (blue) and g-tubulin (red). Scale bars, 10 mm. (C) Quantitative analysis of centrosome hyperamplification in parental cell line SH-EP, vector-EGFP and MYCNEGFP cells. The number of centrosomes per cell (one and two, or X3) was counted after immunostaining with g-tubulin. For the analysis, 4200 cells were examined. Bars represent the average7s.e. calculated from three independent measurements


1007

Figure 2 g-Irradiation induced centrosome hyperamplification in MYCN-transduced cells. Cells grown on coverslips were irradiated with 10 Gy at 251C using a 137Cs source g-irradiator Gamma Cell 40 (Atomic Energy of Canada Ltd, Ontario, Canada). (A) After girradiation, MYCN-EGFP cells were immunostained for g-tubulin as described in the legend to Figure 1. The left, middle and right panels show DNA (blue), g-tubulin (red) and merge, respectively. A MYCN-EGFP cell 48 h after irradiation is shown in a. Tripolar mitosis in a MYCN-EGFP cell after 72 h is shown in b. Arrowheads indicate centrosomes. Scale bars, 10 mm. (B) Quantitative analysis of centrosome hyperamplification in vector-EGFP and MYCN-EGFP cells after 10 Gy of g-irradiation. Number of centrosomes per cell (one and two, or X3) was counted after immunostaining for g-tubulin. For the analysis, 4200 cells were examined. Bars represent the average7s.e. calculated from three independent measurements

some hyperamplification as compared to vector-EGFP cells (Po0.001) (Figure 3B, b). Collectively, these findings indicate that after DNA damage, enhanced expression of MYCN leads to centrosome hyperamplification. Micronuclei are known to represent extranuclear chromosomal fragments that are not incorporated into the nucleus during mitosis and are produced via various mechanisms (i.e. acentric fragments, multicentric chromosomes, damaged kinetochores and spindle fiber defects) (Mu¨ller et al., 1996). Positive correlation between micronuclei formation and centrosome hyperamplification after DNA damage has been reported (Sato et al., 2000; Roninson et al., 2001). The cells with micronuclei in MYCN-EGFP cells obviously increased

Figure 3 Aph induced centrosome hyperamplification in MYCNtransduced cells. The vector-EGFP and MYCN-EGFP cells grown on coverslips were treated with 0.5 mg/ml of Aph as a DNAdamaging agent. (A) The vector-EGFP and MYCN-EGFP cells that had been treated with Aph were immunostained for phosphorylated histone H2AX (g-H2AX) (1 : 500, antiphosphohistone H2A.X[Ser139] mouse monoclonal antibody, Upstate Biotechnology). Panels a and b show g-H2AX (red) and DNA (blue) in an Aph- and nontreated vector-EGFP cell, respectively. Panel c shows the percentage of g-H2AX positive cells in vectorEGFP and MYCN-EGFP cells after Aph treatment. Scale bar, 10 mm. (B) Centrosome hyperamplification in a MYCN-EGFP cell at 48 h after Aph treatment is shown in a. The left, middle and right panels show DNA (blue), g-tubulin (red) and merge, respectively. Arrowheads indicate centrosomes. Scale bars, 10 mm. Quantitative analysis of centrosome hyperamplification in vector-EGFP and MYCN-EGFP cells after Aph treatment is shown in panel b. Number of centrosomes per cell (one and two, or X3) was counted after immunostaining for g-tubulin. For the analysis, 4200 cells were examined. Data are the mean of two independent experiments

from 12 h after g-irradiation (Figure 4A, a and b). aTubulin staining indicates that there are cells that contain more than one micronuclei per cell. The number of cells with micronuclei in MYCN-EGFP cells significantly increased from 24 h after g-irradiation as compared to those in vector-EGFP cells (Po0.001) Oncogene


1008

Figure 4 g-Irradiation induced micronuclei formation in MYCNtransduced cells. (A) Micronuclei in the MYCN-EGFP cells after girradiation. After 10 Gy of g-irradiation, cells grown on coverslips were immunostained with anti-a-tubulin monoclonal antibody (1 : 400, DM1A, Sigma) (red) and counterstained with Hoechst No. 33342 (blue). Cells with micronuclei 48 h after g-irradiation are shown in a and b. Arrowheads indicate micronuclei that are smaller than the size of normal nucleus. Scale bars, 10 mm. (B) Quantitative analysis of cells with micronuclei in vector-EGFP and MYCNEGFP cells after 10 Gy of g-irradiation. The number of micronuclei was counted after Hoechst staining. For the analysis, 4200 cells were examined. Data are the mean of two independent experiments

(Figure 4B). Similarly, after Aph treatment, the population of cells with micronuclei in MYCN-EGFP cells increased as compared to those in vector-EGFP cells (data not shown). Thus, enhanced expression of MYCN also leads to micronuclei formation after DNA damage.

Centrosomes play an important role in maintaining the fidelity of chromosome distribution during mitosis. Loss of these functions should cause multipolar mitotic spindles, resulting in chromosomal instability including micronuclei. In the present study, we found that enhanced expression of MYCN leads to centrosome hyperamplification after DNA damage in neuroblastoma cells. If the failure of cell division after DNA damage causes centrosome hyperamplification with DNA replication, the fraction of cell population with DNA contents of 8N should increase. However, by FACS analysis, the fraction of MYCN-EGFP cells with 8N at 0 and 48 h after g-irradiation (2.45 and 6.38%) did not increase as compared to that of vector-EGFP cells (4.01 and 6.71%). Since centrosome duplication cycle is frequently uncoupled with DNA replication in the absence of DNA damage-response proteins, such as p53, p21, or PARP-1 (Tarapore et al., 2001; Tarapore and Fukasawa, 2002; Kanai et al., 2003), it is speculated that after DNA damage by Aph treatment, MYCN protein disrupts the coupling between DNA replication and centrosome duplication and directly or indirectly enhances centrosome hyperamplification by suppressing the expression or function of the above DNA damageresponse proteins (Figure 3B, b). It is interesting that p21 expression after DNA damage was suppressed by an MYCN homologue, MYC (Seoane et al., 2002). We have also found that another neuroblastoma cell line, NB19 (Gilbert et al., 1982), with gene amplification and overexpression of MYCN, also showed centrosome hyperamplification after g-irradiation (data not shown). In conclusion, we showed that enhanced expression of MYCN leads to centrosome hyperamplification and micronuclei formation after DNA damage in neuroblastoma cells. Further analysis of molecular mechanism of centrosome hyperamplification by overexpression of MYCN after DNA damage will clarify a novel function of the MYCN oncogene. Acknowledgements This work was supported in part by grant-in-aids for Cancer Research from the Ministry of Health, Labour and Welfare (Japan) and the Japan Society for the Promotion of Sciences (Japan). We thank Dr H Saya (Kumamoto University) for a valuable suggestion. We thank S Hanai, N Uematsu, S Ohashi, Y Nogi and A Kabayama for precious comments and technical assistance.

References Breit S and Schwab M. (1989). J. Neurosci. Res., 24, 21–28. Brinkley BR and Goepfert TM. (1998). Cell Motil. Cytoskeleton, 41, 281–288. Carroll PE, Okuda M, Horn HF, Biddinger P, Stambrook PJ, Gleich LL, Li YQ, Tarapore P and Fukasawa K. (1999). Oncogene, 18, 1935–1944. Chernova OB, Chernov MV, Ishizaka Y, Agarwal ML and Stark GR. (1998). Mol. Cell. Biol., 18, 536–545. Doxsey S. (2001). Nat. Rev. Mol. Cell. Biol., 2, 688–698. Duensing S, Lee LY, Duensing A, Basile J, Piboonniyom S, Gonzalez S, Crum CP and Munger K. (2000). Proc. Natl. Acad. Sci., 97, 10002–10007. Oncogene

Fukasawa K. (2002). Oncogene, 21, 6140–6145. Fukasawa K, Choi T, Kuriyama R, Rulong S and Vande Woude GF. (1996). Science, 271, 1744–1747. Gilbert F, Balaban G, Moorhead P, Bianchi D and Schlesinger H. (1982). Cancer Genet. Cytogenet., 7, 33–42. Kanai M, Tong WM, Sugihara E, Wang ZQ, Fukasawa K and Miwa M. (2003). Mol. Cell. Biol., 23, 2451–2462. Lingle WL, Lutz WH, Ingle JN, Maihle NJ and Salisbury JL. (1998). Proc. Natl. Acad. Sci., 95, 2950–2955. Lutz W, Stohr M, Schurmann J, Wenzel A, Lohr A and Schwab M. (1996). Oncogene, 13, 803–812.


1009 Mantel C, Braun SE, Reis S, Henegariu O, Liu L, Hangoc G and Broxmeyer HE. (1999). Blood, 93, 1390–1398. Mu¨ller WU, Nu¨sse M, Miller BM, Slavotinek A, Viaggi S and Streffer C. (1996). Mutat. Res., 366, 163–169. Pihan GA, Purohit A, Wallace J, Knecht H, Woda B, Quesenberry P and Doxsey SJ. (1998). Cancer Res., 58, 3974–3985. Rogakou EP, Pilch DR, Orr AH, Ivanova VS and Bonner WM. (1998). J. Biol. Chem., 273, 5858–5868. Roninson IB, Broude EV and Chang BD. (2001). Drug Resist. Updat., 4, 303–313. Saavedra HI, Fukasawa K, Conn CW and Stambrook PJ. (1999). J. Biol. Chem., 274, 38083–38090. Sato N, Mizumoto K, Nakamura M, Nakamura K, Kusumoto M, Niiyama H, Ogawa T and Tanaka M. (1999). Clin. Cancer Res., 5, 963–970. Sato N, Mizumoto K, Nakamura M, Ueno H, Minamishima YA, Farber JL and Tanaka M. (2000). Oncogene, 19, 5281– 5290. Savelyeva L and Schwab M. (2001). Cancer Lett., 167, 115– 123. Schwab M, Alitalo K, Klempnauer KH, Varmus HE, Bishop JM, Gilbert F, Brodeur GM, Goldstein M and Trent J. (1983). Nature, 305, 245–248.

Schwab M, Westermann F, Hero B and Berthold F. (2003). Lancet. Oncol., 4, 472–480. Seoane J, Le HV and Massague J. (2002). Nature, 17, 729–734. Suzuki A, Obi K, Urabe T, Hayakawa H, Yamada M, Kaneko S, Onodera M, Mizuno Y and Mochizuki H. (2002). J. Neurochem., 82, 953–960. Tarapore P and Fukasawa K. (2002). Oncogene, 21, 6234– 6240. Tarapore P, Horn HF, Tokuyama Y and Fukasawa K. (2001). Oncogene, 20, 3173–3184. Toledo F, Coquelle A, Svetlova E and Debatisse M. (2000). Nucleic Acids Res., 28, 4805–4813. Tutt A, Gabriel A, Bertwistle D, Connor F, Paterson H, Peacock J, Ross G and Ashworth A. (1999). Curr. Biol., 9, 1107–1110. Tweddle DA, Malcolm AJ, Cole M, Pearson AD and Lunec J. (2001). Am. J. Pathol., 158, 2067–2077. Wartiovaara K, Barnabe HF, Miller FD and Kaplan DR. (2002). J. Neurosci., 22, 815–824. Westermann F and Schwab M. (2002). Cancer Lett., 284, 127– 147. Xu X, Weaver Z, Linke SP, Li C, Gotay J, Wang XW, Harris CC, Reid T and Deng CX. (1999). Mol. Cell., 3, 389–395. Zhou H, Kuang J, Zhong L, Kuo WL, Gray JW, Sahin A, Brinkley BR and Sen S. (1998). Nat. Genet., 20, 189–193.

Oncogene