Multiple molecular mechanisms contribute to radiation ... - Nature

Oncogene (2003) 22, 7905–7912

& 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc

Multiple molecular mechanisms contribute to radiation sensitivity in mantle cell lymphoma R M’kacher2, A Bennaceur1,2, F Farace1,2, A Lauge´4, LF Plassa5, E Wittmer5, J Dossou2, D Violot2, E Deutsch2, J Bourhis2, D Stoppa-Lyonnet4, V Ribrag1,2, P Carde2, C Parmentier2, A Bernheim3 and AG Turhan*,1,2 1

Translational Research-Cell Therapy Laboratory, Inserm U362, Villeujuif, France; 2Departments of Medicine, Clinical Biology, UPRES EA 27-10, Villejuif, France; 3Cytogenetic and Oncologic Genetic Laboratory, IGR, Villejuif, France; 4Service de Genetique Oncologique, Institut Curie, France; 5Laboratoire de Biochimie B, Hospital St Louis, Paris, France

Mantle cell lymphomas (MCL) are characterized by their aggressive behavior and poor response to chemotherapy regimens. We report here evidence of increased in vitro radiation sensitivity in two cell lines that we have generated from two MCL patients (UPN1 and UPN2). However, despite their increased radiation sensitivity, UPN2 cells were totally resistant to apoptotic cell death, whereas UPN1 cells underwent massive apoptosis 6 h after irradiation. The frequency of induced chromosomal abnormalities was higher in UPN1 as compared to UPN2. Distinct mechanisms have been found to contribute to this phenotype: a major telomere shortening (UPN1 and UPN2), deletion of one ATM allele and a point mutation in the remaining allele in UPN2, mutation of p53 gene (UPN1 and UPN2) with absence of functional p53 as revealed by functional yeast assays. After irradiation, Ku70 levels in UPN1 increased and decreased in UPN2, whereas in the same conditions, DNA-PKcs protein levels decreased in UPN1 and remained unchanged in UPN2. Thus, irradiation-induced apoptotic cell death can occur despite the nonfunctional status of p53 (UPN1), suggesting activation of a unique pathway in MCL cells for the induction of this event. Overall, our study demonstrates that MCL cells show increased radiation sensitivity, which can be the result of distinct molecular events. These findings could clinically be exploited to increase the dismal response rates of MCL patients to the current chemotherapy regimens. Oncogene (2003) 22, 7905–7912. doi:10.1038/sj.onc.1206826 Keywords: DNA repair; ATM; P53; mantle cell lymphoma; irradiation

Introduction Mantle cell lymphomas (MCL) represent a specific subtype of non-Hodgkin’s lymphomas (NHL) derived *Correspondence: AG Turhan, Translational Research-Cell Therapy Laboratory, Inserm U362, IGR, Villejuif, France; E-mail: [email protected] Received 20 June 2002; revised 12 May 2003; accepted 30 May 2003

from naive CD5 þ B cells residing in the primary follicles or in the mantle zones of secondary follicles (Tolksdorf et al., 1980; Weisenburger and Armitage, 1996). MCL are clinically characterized by their poor prognosis, due to either resistance to chemotherapy regimens used in other types of aggressive NHL or to rapid relapse after an initial response. One of the major characteristics of MCL is the presence of a deregulated cyclin D1 overexpression. The t(11;14)(q13;q32), which leads to juxtaposition of CCDN1 gene and immunoglobulin heavy-chain promoter, is thought to be at the origin of the deregulation of the CCDN1 gene coding for cyclin D1 expression, a gene which is in fact 120 kb away from the major translocation region (Bosch et al., 1994; Dreyling et al., 1997). The mechanisms of this long-distance activation remain undetermined at the present time. It is, however, thought that the cyclin D1 gene overexpression alone is not sufficient for hematopoietic transformation as transgenic mice for cyclin D1 do not develop lymphoma (Bodrug et al., 1994; Lovec et al., 1994). The unregulated cyclin D1 expression seems to be, however, a major event in the pathogenesis and progression of MCL, especially in the presence of an inefficient p53 expression due to a deletion or a mutation (Campo et al., 1999). Deletions involving bands 11q22–q23 are also commonly observed in MCL as well as in other B-cell lymphoproliferative disorders such as chronic lymphocytic leukemia (Stankovic et al., 1999; Schaffner et al., 2000). These deletions affect the genomic region of the ATM gene, which is known to play a role in chromosomal instability, radiation sensitivity and defective cell cycle checkpoint activation. ATM has been shown to phosphorylate directly the c-Abl and p53 in response to DNA damage (Baskaran et al., 1997; Shafman et al., 1997; Banin et al., 1998; Canman et al., 1998). ATM inactivation has also been found to be a major molecular event in MCL, either mutation or deletion at the genetic level (Schaffner et al., 2000). Recently, high-dose myeloablative radioimmunotherapy using 131I-labeled anti-CD20 antibodies has

Radiation sensitivity in MCL R M’kacher et al

7906

been shown to induce high response rates and moderate toxicity in patients with relapsed or refractory MCL (Behr et al., 2002; Gopal et al., 2002). This would suggest that MCL cells could be predisposed to irradiation-induced cell death, as compared to normal hematopoietic cells. In this study, we evaluated the in vitro radiation sensitivity of two MCL cell lines established in our laboratory. We report here, for the first time to our knowledge, that MCL cells exhibit an increased sensitivity to in vitro irradiation. Generated data suggest that several molecular mechanisms could contribute to this phenotype, including mutation of p53 and ATM genes, abnormalities of the DNA repair proteins as well as an abnormal telomerase status.

Results Immunophenotypic and cytogenetic characterization of UPN1 and UPN2 cell lines MCL cell lines have been generated as previously described (M’kacher et al., 2003). Cell lines grew independent from growth factors. Immunophenotype analysis showed the presence of B-cell features including CD19 and CD20 expression, whereas both cell lines were negative for CD23 expression. Morphologically, cell lines displayed features identical to those of primary cells. The presence of hyperexpression of cyclin D1 was confirmed using RT–PCR in both primary cells and the cell lines. Morphological analysis revealed a very complex pattern of chromosomal abnormalities that appeared monoclonal, with few variations, in both cell lines. R-banding, Mfluorescence in situ hybridization (FISH), chromosomal painting and specific probes were combined in order to characterize these chromosomal abnormalities. UPN1 cell line was hypodiploid, with a mode at 43 chromosomes. R-banding, M-FISH and even chromosomal painting did not show the classical t(11;14) in UPN1. A FISH with specific CCND1 and IGH gene probes showed the presence of two fusion signals in chromosome 14 indicating the fact that the translocation t(11;14) existed, but in a masked state. Chromosomal abnormalities found in the UPN1 cell line involved chromosomes 2, 4, 8, 11, 13, 16, 17, 18, 19, 21, 22, X and Y. These rearrangements were detected in all cells examined. UPN2 had grown into a hypotetraploid cell line, containing 83 chromosomes. Clonal structural abnormalities involving chromosomes 1, 2, 3, 8, 9, 10, 11, 12, 13, 14, 15, 22, X and Y were also detected. A specific t(11;14)(q13;q32) was observed (M’kacher et al., 2003). Chromosome 17 abnormalities were not identified in UPN2 cell line. However, in UPN1 cell line, using chromosome painting, der(17)t(2;17) was found to be present. Using specific loci of P53, UPN2 cell line showed the deletion of one locus, but in UPN1 cell line the presence of two loci were observed. Oncogene

MCL cell lines exhibit an increased sensitivity to ionizing radiation As shown in Figure 1, both cell lines exhibited a significant increase of the radiation sensitivity as compared to controls. The surviving fraction at 2 Gy (SF2) was 24% in UPN1. In UPN2, SF2 was 26% during the primary passage. In the early passage of the cell line, we found a heterogeneous cell population including hypodiploid and hyperdiploid clones (50 and 30%, respectively) as well as normal appearing cells as evidenced by FISH assay (20%). After further passages, hyperdiploid clone containing 83 chromosomes became predominant and exhibited an increased radiosensitivity (SF2 ¼ 7%). This increase in radiation sensitivity was comparable to that observed in AT 3153 cell line, which originates from a patient with ataxia-telangectasia syndrome (SF2 ¼ 4%). These results were also confirmed by the use of two additional MCL cell lines (kindly provided by Dr N Andersen, Department of Hematology, Copenhagen, Denmark) showing a significant radiation sensitivity (SF2 for Granta 519 was 24% and 12% for NCE) (data not shown). Assessment of DNA repair by scoring of chromosomal aberrations (FISH) Ionizing radiation induces double-strand breaks (DSB) activating the DNA repair system. Chromosome painting (FISH) assessed this DNA repair function in both MCL cell lines after irradiation. UPN1 cell line showed

Figure 1 Clonogenic survival assay demonstrating the presence of a significant radiation sensitivity of both UPN1 and UPN2 cell lines as compared to HT29 (radiation-resistant cell line) and AT3153 (radiation-sensitive cell line with ATM mutation)


7907

Figure 3 The sensitivity of UPN1 and UPN2 cell lines to radiation-induced apoptosis was tested using TUNEL essay. UPN1 shows a significant sensitivity to radiation-induced apoptosis, whereas UPN2 cell line was resistant to similar doses. Cell lines were irradiated at 2 and 4 Gy and collected 0, 3, 6 and 24 h later Figure 2 Evaluation of the frequency of chromosomal aberrations after in vitro irradiation in MCL cell lines and AT3153 control cell line as detected by chromosome 1, 3 and 4 painting (FISH). The frequency of chromosomal abnormalities was extremely low in the UPN2 cell line due to an arrest in G1/S as compared to AT3153. On the other hand, UPN1 cell line showed a major increase of chromosomal abnormalities in the same conditions

a high frequency of chromosomal aberrations (dicentrics, translocations, insertions and acentric fragments) induced by in vitro irradiation (Figure 2). The frequency of chromosomal aberrations after irradiation was very low in UPN2 reflecting the lower rate of mitosis after irradiation. In fact, at 4 Gy, no metaphases were seen in slides. The same observation was found in the AT 3153 cell line. Assessment of apoptosis induced by ionizing radiation The UPN1 cell line exhibited high sensitivity to radiation-induced apoptosis reflecting the major DNA repair deficiency. The maximal level of apoptotic cell death in UPN1 occurred at 6 h after irradiation (Figure 3). However, UPN2 was resistant to radiationinduced apoptosis. These two cell lines exhibited, therefore, an opposite behavior in terms of the mode of cell death as UPN1 clearly underwent apoptotic cell death whereas UPN2 did not. To study this differential behavior, we have performed molecular analysis, evaluating in particular p53, ATM gene status and expression DNA repair proteins. Identification of p53 mutations To determine the functional status of p53, a functional yeast assay was performed as described (Flaman et al., 1995). This analysis showed the presence of 100% yeast colonies of red color in both cell lines, indicating nonfunctional status of p53 and also clonal homogeneity of both cell populations. Sequencing of p53 cDNA confirmed the presence of the 175C (Arg-His) mutation in UPN2 and 286C (Glu-Lys) in UPN1. Interestingly, p53 and phospho-specific p53 expressions were detectable by Western blot in the UPN1 cell

Figure 4 Western blot analysis showing the differential response of UPN1 and UPN2 cell lines to irradiation for 4 Gy after 3, 6 and 24 h. (a and b) Comparative analysis of p53 protein expression before and after irradiation in UPN1 and UPN2 cell lines using a p53 antibdy (a) and a phospho-specific p53 antibody (b). (c) The UPN2 expressed low baseline Ku70 levels (0 h) as compared to UPN1 in the presence of appropriate actin controls (d). Upon irradiation, Ku70 levels in the UPN1 cell line were increased, as expected, whereas the UPN2 cell line did not exhibit any increase of Ku70 protein. (e) Both cell lines expressed comparable amounts of DNA-PKcs at 0 h irradiation, which led to decreased expression of DNA-PKcs in UPN1 cell line. Whereas, the DNA-PKcs expression remained constant in UPN2 cells

line, but had not increased after in vitro irradiation. In contrast the UPN2 cell line did not express significant levels of p53 (Figure 4a and b) possibly due to a deletion of one locus of p53 as confirmed by FISH. Thus, these results demonstrated the absence of p53 function in these cells, due to a mutation (UPN1 and UPN2) and deletion of one allele (UPN2). Discrepancy of ATM gene status in MCL cell lines In the UPN1 cell line as well as in the primary cells, both alleles of the ATM gene were found be present. However, a missense mutation 998 C-T was found to be present in one of the alleles, with an expected serine-phenylalanine change at the protein level in the p53 interaction domain of this molecule. This missense Oncogene


7908

mutation has been found in some atypical ATM patients as well as in normal controls, suggesting a DNA polymorphism. ATM polymorphic variants have also been identified in MCL (Camacho et al., 2002). The UPN2 cell line and the primary cells of the UPN2 patient showed a typical t(11;14)(q11;q32) and one copy of chromosomal region 11q22–23, suggesting the presence of a single ATM allele. Molecular analysis confirmed the presence of a single ATM gene, which has been found to be mutated at 8174 A-T leading to a missense mutation (Asp-Val). This mutation has previously been reported in a patient with T-PLL (Vorechovsky et al., 1997). The same missense mutation has also recently been found to be present in a patient with MCL (Camacho et al., 2002). MCL cell lines have short telomeres and reduced telomerase activity We explored the relationship between cellular response to ionizing irradiation and telomere length by TRF using Southern blot analysis. In both MCL cell lines, telomeres were severely shortened, exhibiting an average length of 3.2 kb for UPN1 and 3 kb for UPN2 (Figure 5) as compared to control cells. Telomerase activity measured by TRAP assay was slightly reduced in UPN1, whereas a major decrease was found to be present in UPN2 as compared to normal cells (Table 1). Relationship between telomere shortening and Ku expression Ku plays a role in telomere maintenance as well as in DNA DSB repair. To assess the relationship between telomere shortening and Ku expression, we have evaluated by Western blots the levels of Ku70 protein. Both UPN1 and UPN2 cell lines exhibited similar levels of Ku70 protein (Figure 4c). After irradiation, Ku70 levels increased in UPN1. In contrast, in the UPN2 cell line, the baseline levels of Ku70 did not increase after irradiation (Figure 4c). These data therefore suggest that Ku function appears to be normal in UPN1, whereas it is deficient in UPN2.

irradiation of the UPN2 cell lines analysed up to 24 h after irradiation (Figure 4e).

DNA-PKcs expression in MCL cell lines

Cell cycle checkpoints of MCL cell lines

DNA-PKcs, which could serve either a structural or enzymatic protein, plays a major role in DNA DSB repair in mammalian cells (Gilley et al., 2001). To explain the telomere shortening in UPN1, we studied DNA-PKcs expression by Western blot. In vitro irradiation induced a major decrease in DNA-PKcs expression in UPN1 cell line (Figure 4e), correlating with an overall increase in induced chromosomal aberrations detected by FISH technique. These data suggested that DNA-PKcs deficiency greatly increases illegitimate recombination involving telomeres, which is an important source of spontaneous cytogenetic instability. On the other hand, Western blot analysis did not show any decrease of DNA-PKcs expression after

To study the status of cell cycle in MCL cell lines, the distribution of cell cycle profiles of the exponentially growing cells was analysed by flow cytometry before and after irradiation. As shown in Figure 6, UPN1 cell line exhibits G2 cell cycle arrest after irradiation and inhibition to the G2/M transition. These results suggest that UPN1 might carry radiation-induced chromosomal aberrations and are presumably doomed to die. This finding was therefore consistent with the result of radiation sensitivity obtained in clonogenic survival assays. On the other hand, there was a discrepancy with regard to cell cycle status between UPN1 and UPN2, due to the fact that the latter exhibited G1/S phase arrest 24 h after irradiation (Figure 6).

Oncogene

Figure 5 TRF analysis in UPN1 and UPN2 cell lines. Control: Peripheral blood mononuclear cells of normal DNA from two children (o1 year old) (lines 1 and 2). Telomere length was drastically decreased in both MCL cell lines


7909 Table 1 Comparative analysis of DNA repair mechanisms in UPN1 and UPN2 before and after irradiation UPN1

p53 function p53 mutation ATM mutation Telomere length (kb) Telomerase activityc Ku70d DNA-PKcsd p53 phosd Mode of cell death

UPN2

Baseline

Post-RXa

Baseline

Post-RXa

Not functional 286C (Glu-Lys) Not mutatedb 3.2 90 Increased Normal Increased NA

NA NA NA NA NA Increased Decreased Unchanged Apoptotic cell death

Not functional 175C(Arg-His) 8174A-T 3 59 Decreased Normal Absent NA

NA NA NA NA NA Decreased Unchanged Unchanged Mitotic cell death

a

Expression in response to ionizing radiation (2 Gy); bIdentification of polymorphism in UPN1; cUnits quantified as detailed in MM section; Evaluated by Western blot analysis performed before and after irradiation (3, 6 and 24 h); NA: not applicable

d

Figure 6 Flow cytometric analysis of MCL cell lines after in vitro irradiation (4 Gy) and fixed at different time intervals. The population of cells present in G1, S, and G2/M was determined using a FACScan flow cytometer (Becton Dickinson). The percentage of cells in G1/S, and G2/M was calculated using MODIFIT software

Discussion We demonstrate in this work that MCL cell lines present an increased sensitivity to ionizing radiation. This radiation sensitivity appeared to be related to different molecular mechanisms in both cell lines, underlining the heterogeneity of the disorder at the molecular level as summarized in Table 1. Our data suggest, in fact, that the radiation sensitivity was associated in the UPN1 with an irradiation-induced apoptosis and chromosomal aberrations, whereas in the second cell line, UPN2, this phenomenon seemed to be due essentially to the inactivation of the ATM gene. The common phenotypic abnormality observed in both cell lines was the presence of shortened telomeres, which is known to contribute significantly albeit not exclusively, to increased radiation sensitivity. Many radiation-sensitive eucaryotic cells show defects in telomere maintenance (Ahmed and Hodgkin, 2000; d’Adda di Fagagna et al., 1999; Smilenov et al., 1999; Metcalfe et al., 1996). Telomere shortening could also explain the complex chromosomal

rearrangement detected in these MCL cell lines in addition to t(11;14) (M’kacher et al., 2003). The mechanisms of shortened telomeres and low levels of telomerase in our MCL cell lines are currently under investigation. The second common abnormality observed in both cell lines was the mutation of p53 gene with absence of p53 function demonstrated by functional assays, despite the immunodetection of p53 at variable levels. Deletions/mutations of p53 have previously been found to be a poor prognostic factor in MCL (Campo et al., 1999). Interestingly, despite the presence of a well-documented p53 mutation, the UPN1 cell line exhibited a high sensitivity to radiation-induced apoptosis reflecting the major DNA repair deficiency. In addition, after in vitro irradiation, UPN1 showed highest frequency of chromosomal aberrations (dicentrics, translations and acentric fragments). The expression of Ku70 increased after irradiation in this cell line, which has been found to be unable to show efficient DNA repair after irradiation (Figure 4) possibly due to the low levels of DNA-PKcs. A major decrease in DNA-PKcs detected in the UPN1 cell line could also contribute to DNA repair abnormalities and telomere shortening (Gilley et al., 2001) (Table 1). Cell cycle checkpoints allow the maintenance of an ordered series of events in cell cycle progression. These checkpoints are also activated in response to DNA damage to allow time for repair and to maintain stability of the genome. The importance of p53 in the activation of the G1/S checkpoint after irradiation have been well established (Meyer et al., 1999). This event therefore requires the presence of normal levels of a functional p53 protein, which can then activate its target gene at the transcriptional level. In our cell lines, p53 genes were found to be mutated and the function of p53 was found to be absent in yeast assays. Despite this, we found that the UPN1 exhibited a G2 cell cycle arrest and underwent apoptosis, as this has usually been reported after irradiation (Iavarone and Massague, 1997; Meyer et al., 1999). In our study, UPN2 cell line underwent Sphase arrest after irradiation. This defect in cell cycle could explain its resistance to irradiation-induced apoptosis and could explain the lower frequency of Oncogene


7910

induced chromosomal abnormalities and metaphases in response to in vitro irradiation. The major gene involved in radiation sensitivity is the ATM. Although the inactivation of ATM gene has been reported in MCL patients (Schaffner et al., 2000), no data are available, to our knowledge, to link this event to the radiation sensitivity. Our data suggest that the increased radiation sensitivity of the UPN2 cell line was due essentially to the ATM gene mutation. ATM is a key regulator of the cellular response to DNA DSB induced by irradiation (Rotman and Shiloh 1998; Brown et al., 1999). ATM encodes a protein containing a phosphatidylinositol-3-kinase-like domain (Savitsky et al., 1995) related to a family of proteins involved in cell cycle control and DNA damage recognition (Lavin et al., 1995; Zakian, 1995). Moreover, ATM has been shown to phosphorylate proteins such as p53, c-Abl, RPA and BRCA1 (Baskaran et al., 1997; Banin et al., 1998; Canman et al., 1998; Gately et al., 1998; Khanna et al., 1998; Gatei et al., 2000). The demonstration of the phenomenon of irradiation sensitivity in the presence of a mutated p53 could suggest that the ATM mutation plays on upstream role as compared to p53 to control the DNA repair system in MCL. Thus, p53 mutation could potentially contribute to the resistance of the UPN2 cell line to radiation-induced apoptosis. The mechanism by which ATM regulates DNA repair is largely unknown. Our results suggest that increased radiation sensitivity of this ATM-mutated cell line (UPN2) is more likely a direct consequence of mitotic cell death. In addition, telomere shortening in AT cells has previously been reported (Pandita et al., 1999; Pandita, 2002; Metcalfe et al., 1996). We have detected ATM mutation in the UPN2 cell line as well in the primary lymphoma cells of the patient suggesting, in UPN2 cells, a genuine role of ATM in the DNA repair abnormalities of MCL. The question of whether ATM inactivation as observed in UPN2 cell line plays a role in cyclin D1 hyperexpression remains to be studied. The absence of ATM mutation combined with shortening telomeres and a higher number of chromosomal abnormalities in the UPN1 cell line suggests that other genes may be involved in the pathogenesis of MCL. Remarkably, the phenotype of radiation sensitivity was achieved in both cell lines by the use of probably different molecular events. Both cell lines represent unique models of further study of the DNA repair in MCL. UPN2 cell line represents, to our knowledge, the first MCL cell line with well-characterized ATM and p53 mutations, suggesting its future usefulness in the study of MCL. In contrast, UPN1 exhibits a single p53 mutation with absence of p53 function and despite this undergoes an apoptotic cell death, indicating the involvement of a different pathway for the induction of this phenomenon. In summary, in this study, we have established two MCL cell lines that exhibit an increased radiation sensitivity. To our knowledge, this is the first study documenting the molecular basis of this phenomenon. These findings suggest that the increased radiation sensitivity could be clinically exploited to improve the Oncogene

dismal response rates of MCL patients to the current chemotherapy regimens. Indeed, recent studies demonstrated the potential clinical efficacy of supplanting total body radiation with high-dose radioimmunotherapy in an autologous transplantation setting for patients with relapsed or refractory MCL (Behr et al., 2002; Gopal et al., 2002).

Material and methods Patients UPN1 was a 52-year old male, who presented with MCL stage IV medullary disease initially treated by four cycles of CHOP regimen. The pathological examination of the initial lymph nodes was consistent with blastic variant MCL (CD5 þ , CD23, CD19 þ , CD3 þ ). After relapse, he received three cycles of VP 16Endoxan, achieved remission, but 3 months later, he presented with a leukemic phase and died. UPN2 was a 57-year old male presented initially in 1999 with anemia, thrombocytopenia and massive splenomegaly and hyperlymphocytosis. The initial diagnosis of CLL led to the administration of chlorambucil during which the patient progressed. Immunophenotyping of circulating lymphocytes revealed a phenotype consistent with MCL, as the cells were CD5 þ , CD19 þ , CD20 þ , CD23, CD10. Bone marrow was heavily infiltrated with B cells of the same phenotype. A splenectomy was performed, and the patient was treated with seven cycles of CHOP that induced a partial response. Upon progression with extensive pleural involvement and ascite, he failed to respond to further therapy and died. Establishment of cell lines Both cell lines were obtained from pleural effusions of the patients at the leukemic stage of their disease. Mononuclear cells were seeded at a density of 105 cells / ml in RPMI medium with 10% fetal calf serum (FCS). A half medium change was performed once a week and cultures were maintained for several weeks. At week þ 8, autonomous growth was evident in UPN1 cell line, whereas the UPN2 cell line exhibited relatively slow growth. Cell samples were frozen at the initial passages of their growth and both cell lines were then used for cytological and molecular characterization. Irradiation procedure Mantle cell lines were irradiated in vitro with 1.35 Gy/ min 137Cs g-rays at different doses (2 and 4 Gy). During irradiation, cell lines were kept at room temperature. The same procedure was applied to control cells. Clonogenic assays After irradiation, cells were cultured in methylcellulose (Stem cell Technologies) at a concentration of 105/ml at 371C and 5% CO2. The surviving fraction was


7911

determined by measuring the viability of colonies generated before and after irradiation. Only colonies with more than 50 cells were counted. Each data point represents the average and s.d. from three independent experiments performed in triplicate.

dilutions of HL60 cell line protein extract in CHAPS buffer. One arbitrary unit (a.u.) was defined as the telomerase activity of 1 ng of HL60 protein extract.

Chromosome preparation

This was performed by Southern blot TRF analysis. Genomic DNA (5 mg) was digested with 10 U each of HinfI and RsaI (Roche, Meylan, France). After agarose gel electrophoresis, 0.7% DNA was transferred onto nylon membranes for hybridization to a digoxigeninlabeled probe (TTAGGG)3. Bands were revealed by chemiluminescence detection.

UPN1 and UPN2 cell lines were cultured in the presence of RPMI 1640 supplemented with 10% FCS antibiotics (penicillin and streptomycin). Colcemid (0.1 mg/ml) was added 2 h before harvesting and slides with chromosomes in metaphase were prepared following the standard methanol/acetic acid (3/1,v/v) procedure. The slides were stored at 201C until use. Fluorescence in situ hybridization FISH was accomplished by using a combination of standard procedures of the recommended protocols for chromosome analysis. In all, 100–400 metaphases were scored per in vitro dose of irradiation. Chromosome 1, 3 and 4 painting (Appligene Oncor) has been used to score induced chromosomal aberrations (translocations, dicentrics and breaks). To further determine the p53 status, FISH analysis was performed using a specific probe for p53 gene (17q13.1) (Appligene Oncor) was used. Apoptosis assays To evaluate the sensitivity of cell lines to apoptosis, 4 106 cells were irradiated at 2 and 4 Gy. Aliquots of 106 cells were harvested immediately after or 3, 6 and 24 h after irradiation and they were incubated for 10 min at 41C in 1% para-formaldehyde, followed by a fixation in 70% ethanol. Nonirradiated cells were kept as control. Apoptosis was measured using the Apoptag Kit from Oncor, based on the detection of 30 OH extremity DNA fragments due to the activated endonuclease released during apoptosis. 30 OH extremities are labeled by a nucleotide triphosphate bearing digoxigenin, using dioxynucleotidyle-transferase (TDT). This complex is detected by fluorescein-labeled antibodies directed against digoxigenin. DNA is labeled with propidium iodide. The percentage of apoptotic cells was then evaluated using flow cytometry with a FACScan (Becton Dickinson) with excitation by an argon laser at 488 nm. Telomerase activity assays Telomerase activity was measured by a modified TRAP assay, using TRAPeze kit (Intergen, Oxford, UK). Semiquantitative measurement of telomerase activity was performed according to the manufacturer’s instructions after polyacrylamide gel electrophoresis using the Storm 840 instrument (Amersham Lifescience, Bondoufle, France). Real-time quantitative TRAP using ABI PRISM 7700 Detection System (Perkin-Elmer Applied Biosystems) was performed as described (Hou et al., 2001). A standard curve was established using serial

Telomere length analysis

p53 functional assay for screening cell lines To study the functionality of p53 gene, a functional assay was performed (Flaman et al., 1995). The assay tests the entire DNA-binding domain (aa 102–292) as the p53 expression vector is linearized at codons 67 and 346. Briefly, p53 mRNA is reverse transcribed, amplified fy PCR and cotransfected into yeast with a linearized expression vector carrying the 50 and 30 ends of the p53 open reading frame. Gap repair of the plasmid with the PCR product results in the constitutive expression of human p53 protein. Yeasts that have repaired the plasmid are selected on a medium lacking leucine. The medium contains sufficient adenine for growth of Ade þ cells, leading to the generation of white colonies. Thus, colonies containing wild-type p53 are white (Ade2 þ ) and colonies containing mutant p53 are red (Ade2). To define exactly the p53 mutation, a direct sequencing has been performed using an automated sequencer. Fluorescence-assisted mismatch analysis To determine the molecular status of ATM in our cell lines, fluorescence-assisted mismatch analysis (FAMA) was chosen as the most appropriate method for mutation detection (Verpy et al., 1994). The entire coding region of ATM gene was examined in both MCL cell lines. Briefly, after reverse transcription of the mRNA with a set of eight overlapping primer pairs covering the coding sequence of the ATM gene, each cDNA primer pair covering the coding sequence of the ATM gene, each cDNA fragment was screened for mutation by FAMA. To favor heteroduplex formation, a control DNA without ATM mutation was mixed to each cDNA fragment. cDNA fragments showing mismatches were sequenced. The absence of Ser333Phe and Asp2734Val was verified on the control DNA. Western blots DNA-PKcs, Ku70, p53 and phosphospecific p53 expression were tested by Western blotting on lysates UPN1 and UPN2 cell lines using anti-DNAPKcs (Neomarker, Fremont, CA, USA), anti-Ku70 (Oncogene, Cambridge, MA, USA), anti-p53 (Santa Cruz Biotechnology) and anti-p53 phosphospecific Oncogene


7912

(Novocastra Laboratories) antibodies. Cells were sonicated in 500 ml of a buffer containing 8 mol/urea , 150 mmol/l b-mercaptoethanol, 50 mmol/l Tris-HCl (pH 7.2) and centrifuged for 30 min at 41C to remove cellular debris. Samples were submitted to electrophoresis on 12% (p53 and Ku) or 6% (DNA-PKcs) SDS–polyacrylamide gels, blotted onto nitrocellulose membranes and developed using the ECL system (Amersham, Uppsala, Sweden). To verify that equivalent amounts of each sample were loaded, the filters were additionally probed with anti-actin antibody (AC74,

Sigma). Densitometry was performed to evaluate the intensity of Ku70, p53, DNA-PKcs and actin bands. Acknowledgements We thank Pr H de The´ (Laboratoire de biochimie B, Hopital Saint-Louis, Paris) for p53 investigation, Dr N Andersen (Department of Hematology, Copenhagen, Denmark) for providing MCL cell lines (Granta and NCE) and Dr M Gueret-Amor for providing AT cell line. We also thank ML Bonnet and M Guillier for their excellent technical help.

References Ahmed S and Hodgkin J. (2000). Nature, 403, 159–164. Banin S, Moyal L, Shieh S, Taya Y, Anderson CW, Chessa L, Smorodinsky NI, Prives C, Reiss Y, Shiloh Y and Ziv Y. (1998). Science, 281, 1674–1677. Baskaran R, Wood LD, Whitaker LL, Canman CE, Morgan SE, Xu Y, Barlow C, Baltimore D, Wynshaw-Boris A, Kastan MB and Wang JY. (1997). Nature, 387, 516–519. Behr TM, Griesinger F, Riggert J, Gratz S, Behe M, Kaufmann CC, Wormann B, Brittinger G and Becker W. (2002). Cancer, 94, 1363–1372. Bodrug SE, Warner BJ, Bath ML, Lindeman GJ, Harris AW and Adams JM. (1994). EMBO J., 13, 2124–2130. Bosch F, Jares P, Campo E, Lopez-Guillermo A, Piris MA, Villamor N, Tassies D, Jaffe ES, Montserrat E and Rozman C. (1994). Blood, 84, 2726–2732. Brown KD, Barlow C and Wynshaw-Boris A. (1999). Am. J. Hum. Genet., 64, 46–50. Camacho E, Hernandez L, Hernandez S, Tort F, Bellosillo B, Bea S, Bosch F, Montserrat E, Cardesa A, Fernandez PL and Campo E. (2002). Blood, 99, 238–244. Campo E, Raffeld M and Jaffe ES. (1999). Semin. Hematol., 36, 115–127. Canman CE, Lim DS, Cimprich KA, Taya Y, Tamai K, Sakaguchi K, Appella E, Kastan MB and Siliciano JD. (1998). Science, 281, 1677–1679. d’Adda di Fagagna F, Hande MP, Tong WM, Lansdorp PM, Wang ZQ and Jackson SP. (1999). Nat. Genet., 23, 76–80. Dreyling MH, Bullinger L, Ott G, Stilgenbauer S, MullerHermelink HK, Bentz M, Hiddemann W and Dohner H. (1997). Cancer Res., 57, 4608–4614. Flaman JM, Frebourg T, Moreau V, Charbonnier F, Martin C, Chappuis P, Sappino AP, Limacher IM, Bron L, Benhattar J, Tada M, Van Meir EG, Esteicher A and Iggo RD. (1995). Proc. Natl. Acad. Sci. USA, 92, 3963–3967. Gatei M, Scott SP, Filippovitch I, Soronika N, Lavin MF, Weber B and Khanna KK. (2000). Cancer Res., 60, 3299–3304. Gately DP, Hittle JC, Chan GK and Yen TJ. (1998). Mol. Biol. Cell, 9, 2361–2374. Gilley D, Tanaka H, Hande MP, Kurimasa A, Li GC, Oshimura M and Chen DJ. (2001). Proc. Natl. Acad. Sci. USA, 98, 15084–15088. Gopal AK, Rajendran JG, Petersdorf SH, Maloney DG, Eary JF, Wood BL, Gooley TA, Bush SA, Durack LD, Martin PJ, Matthews DC, Appelbaum FR, Bernstein ID and Press OW. (2002). Blood, 99, 3158–3162.

Oncogene

Hou M, Xu D, Bjorkholm M and Gruber A. (2001). Clin. Chem., 47, 519–524. Iavarone A and Massague J. (1997). Nature, 387, 417–422. Khanna KK, Keating KE, Kozlov S, Scott S, Gatei M, Hobson K, Taya Y, Gabrielli B, Chan D, Lees-Miller SP and Lavin MF. (1998). Nat. Genet., 20, 398–400. Lavin MF, Khanna KK, Beamish H, Spring K, Watters D and Shiloh Y. (1995). Trends Biochem. Sci., 20, 382–383. Lovec H, Grzeschiczek A, Kowalski MB and Moroy T. (1994). EMBO J., 13, 3487–3495. M’kacher R, Farace F, Bennaceur A, Violot D, Clausse B, Dossou J, Valent A, Parmentier C, Ribrag V, Bosq J, Carde P, Turhan AG and Bernheim A. (2003). Cancer Genet. Cytogenet., 142, 32–38. Metcalfe JA, Parkhill J, Campbell L, Stacey M, Biggs P, Byrd PJ and Taylor AM. (1996). Nat. Genet., 13, 350–353. Meyer KM, Hess SM, Tlsty TD and Leadon SA. (1999). Oncogene, 18, 5795–5805. Pandita TK. (2002). Oncogene, 21, 611–618. Pandita TK, Westphal CH, Anger M, Sawant SG, Geard CR, Pandita RK and Scherthan H. (1999). Mol. Cell. Biol., 19, 5096–5105. Rotman G and Shiloh Y. (1998). Hum. Mol. Genet., 7, 1555–1563. Savitsky K, Bar-Shira A, Gilad S, Rotman G, Ziv Y, Vanagaite L, Tagle DA, Smith S and Uziel T and Sfez S. et al. (1995). Science, 268, 1749–1753. Schaffner C, Idler I, Stilgenbauer S, Dohner H and Lichter P. (2000). Proc. Natl. Acad. Sci. USA, 97, 2773–2778. Shafman T, Khanna KK, Kedar P, Spring K, Kozlov S, Yen T, Hobson K, Gatei M, Zhang N, Watters D, Egerton M, Shiloh Y, Kharbanda S, Kufe D and Lavin MF. (1997). Nature, 387, 520–523. Smilenov LB, Dhar S, Pandita TK. (1999). Mol. Cell Biol., 19, 6963–6971. Stankovic T, Weber P, Stewart G, Bedenham T, Murray J, Byrd PJ, Moss PA and Taylor AM. (1999). Lancet, 353, 26–29. Tolksdorf G, Stein H and Lennert K. (1980). Br. J. Cancer, 41, 168–182. Verpy E, Biasotto M, Meo T and Tosi M. (1994). Proc. Natl. Acad. Sci. USA, 91, 1873–1877. Vorechovsky I, Luo L, Dyer MJ, Catovsky D, Amlot PL, Yaxley JC, Foroni L, Hammarstrom L, Webster AD and Yuille MA. (1997). Nat. Genet, 17, 96–99. Weisenburger DD and Armitage JO. (1996). Blood, 87, 4483–4494. Zakian VA. (1995). Science, 270, 1601–1607.