Alterations in the p16INK4a and p53 tumor suppressor genes ... - Nature

Oncogene (2004) 23, 3116–3121

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

SHORT REPORTS

Alterations in the p16INK4a and p53 tumor suppressor genes of hTERT-immortalized human fibroblasts Jane R Noble1, Ze-Huai Zhong1, Axel A Neumann1, John R Melki2, Susan J Clark2 and Roger R Reddel*,1 1 Children’s Medical Research Institute, 214 Hawkesbury Rd, Westmead, Sydney NSW 2145, Australia; 2Sydney Cancer Centre, Medical Foundation Building, K25, University of Sydney, NSW 2006, Australia

Exogenous expression of the catalytic subunit of telomerase, hTERT, in a normal human foreskin fibroblast cell strain resulted in telomerase activity and an extended proliferative lifespan prior to a period of crisis. Three immortalized cell lines with stably maintained telomere lengths were established from cells that escaped crisis. Each of these cultures underwent a significant downregulation of p16INK4a expression due to gene deletion events. One cell line also acquired mutations in both alleles of the p53 tumor suppressor gene. Downregulation of p16INK4a and loss of wild-type p53 expression occurred after escape from crisis, so these mutations are most likely not required for immortalization of these cells but rather were selected for during continuous growth in vitro. These findings emphasize the need for caution in the use of hTERT-immortalized cells in studies of normal cell biology or in tissue engineering and the need to monitor for genetic instability and the accumulation of mutations in both the p16INK4a/pRb and p53 pathways. Oncogene (2004) 23, 3116–3121. doi:10.1038/sj.onc.1207440 Published online 26 January 2004 Keywords: telomerase; hTERT; immortalization p16INK4a; CDKN2A; TP53

Normal human cells undergo a limited number of cell divisions when cultured in vitro before entering a state of permanent growth arrest, known as senescence (Hayflick and Moorhead, 1961) or M1 (Wright et al., 1989). Throughout the life span of normal somatic cells their telomeres shorten by approximately 50–200 base pairs (bp) per cell division (Harley, 1991). It has been proposed that senescence is triggered when telomeres become sufficiently short (Olovnikov, 1973; Harley, 1991). The limited proliferative capacity of normal cells is thought to act as a barrier to malignant transformation and, conversely, acquiring the ability to proliferate an unlimited number of times (immortalization) may *Correspondence: RR Reddel; E-mail: [email protected] Received 28 August 2003; revised 7 December 2003; accepted 15 December 2003

be critically important for carcinogenesis (Reddel, 2000; Campisi, 2001). During the process of lifespan extension and immortalization, whether by in vitro transformation with the oncogenes of DNA tumor viruses (reviewed in Bryan and Reddel, 1994) or by disruption of the p53 and/or pRb tumor suppressor gene pathways, telomere shortening continues (Counter et al., 1992; Rogan et al., 1995) and the cells eventually enter a state referred to as crisis (Girardi et al., 1965) or M2 (Wright et al., 1989). Approximately 1 in 107 cells may spontaneously acquire a mutation that enables them to escape crisis and become immortalized (Huschtscha and Holliday, 1983; Shay and Wright, 1989). All cells that escape from crisis activate a telomere maintenance mechanism (Colgin and Reddel, 1999). The majority of immortal cells activate telomerase, a reverse transcriptase that maintains telomere ends by the addition of TTAGGG repeats, and the remainder maintain their telomeres by an alternative mechanism for lengthening telomeres (ALT). Most normal human cells have levels of telomerase activity that are low or undetectable by standard assays, but telomerase activity can be induced by the expression of exogenous hTERT cDNA (Weinrich et al., 1997). Exogenous expression of hTERT cDNA in otherwise normal fibroblasts and retinal pigment cells has been reported to permit senescence bypass (Bodnar et al., 1998; Vaziri and Benchimol, 1998) and immortalization (Jiang et al., 1999; Morales et al., 1999; Vaziri et al., 1999). In contrast, hTERT expression does not affect the proliferative life span of a variety of epithelial cells grown under certain culture conditions unless the pRb/ p16INK4a pathway is inactivated (Kiyono et al., 1998; Farwell et al., 2000; Kim et al., 2002; Toouli et al., 2002). Likewise, for some strains of lung (MacKenzie et al., 2000; Franco et al., 2001), mammary (O’Hare et al., 2001) and periodontal ligament (Tsutsui et al., 2002) fibroblasts, hTERT-transduced cells required additional mutational events including the loss of p16INK4a expression prior to immortalization. hTERT transfection of HFF5 In this study we investigated the ability of exogenous hTERT expression to immortalize a karyotypically normal human foreskin fibroblast (HFF5) cell strain.

p16INK4a and p53 alterations in hTERT-immortalized fibroblasts JR Noble et al

3117

HFF5 cells were transfected at PD34 with the expression plasmid, pCI-hTERT, encoding the catalytic subunit of telomerase, or pCI-Neo control plasmid. Transfected cells were selected and grown continuously in 50 mg/ml G418. Transfection of HFF5 with pCI-hTERT resulted in three independent immortal mass cultures: HFF5/ hTERT-1, -2, and -3. The growth curves of these cell lines are shown in Figure 1a. Approximately 32 PD after transfection, all three cell lines entered a period of crisis that ranged in duration from 42 days for HFF5/ hTERT-2 to 102 days for HFF5/hTERT-3, after which the cells increased their growth rate. HFF5/hTERT-3 cells had a second period of slow growth starting at approximately PD50, which lasted for 30 days, before recommencing growth at an increased rate. Exogenous expression of hTERT conferred in vitro telomerase activity on the three immortal cell lines at all time points

examined (Figure 1b). HFF5 cells transfected with the control plasmid pCI-Neo senesced 23PD after transfection (data not shown). HFF5/hTERT-6 that expressed telomerase activity at PD21 after transfection but lost telomerase activity by PD28 (data not shown) ceased proliferation at PD29. Telomere lengths were examined by terminal restriction fragment (TRF) analysis and this revealed a distinct pattern for each culture (Figure 1c). The mean telomere lengths of the HFF5/hTERT-1 cells were similar to those of the parental HFF5 cells at PD40, and after escape from crisis (PD57) were slightly shorter and remained so until PD144. Precrisis HFF5/hTERT-2 cells (PD32 and 35) had mean telomere lengths that were significantly greater than those of the parental cells, but the cells that escaped from crisis had markedly shorter telomeres that were stable in length from PD51 to

Figure 1 Growth of HFF5 normal human fibroblasts transfected with pCI-hTERT. (a) HFF5 (PD34) cells were transfected at day 0 using FuGENE 6 transfection reagent (Roche) with the plasmids pCI-hTERT (Colgin et al., 2000) and pCI-Neo (empty vector control). Transfected cells were selected in Dulbecco’s modified essential medium (DMEM Life Technologies) with 10% fetal bovine serum (Trace Biosciences) and 50 mg/ml G418 (Sigma). Three cell lines designated HFF5/hTERT-1, -2, -3 were established as mass cultures from separate transfections and cumulative PDs were calculated at each passage. The G418-resistant control clones senesced 23 PDs after transfection. (b) Telomerase activity was assayed using the Telomere Repeat Amplification Protocol (TRAP) (Kim et al., 1994) with the modifications previously described (Perrem et al., 1999) in untransfected HFF5 cells at PD36 and cultures HFF5/ hTERT-1, -2 and -3 at increasing PDs. (c) Telomere length was determined using pulsed-field gel electrophoresis as previously described (Bryan et al., 1995) in HFF5 cells at PD40 and the HFF5/hTERT-1, -2 and -3 cultures at increasing PDs. (d) Growth of hTERT-expressing HFF5 cells in 5% O2. In all hTERT cultures, the PD levels indicated refer to post-transfection growth Oncogene


3118

PD141. The mean telomere length of culture HFF5/ hTERT-3 was similar to that of parental HFF5 cells at PD29 (precrisis), but increased substantially by PD45 (postcrisis) and remained stable to PD111. For each of the three cell lines, these data are consistent with the telomere lengths being maintained or increased by the expression of telomerase prior to crisis, and the emergence from crisis of a single clone in which the telomeres are also maintained but at a length that differs from that of the multiclonal precrisis population. As few cells are exposed in vivo to atmospheric oxygen tension, it is likely that growth of human cells under the standard culture conditions of 5% CO2 and 20% O2 results in oxidative stress. It has been proposed that this may cause accelerated shortening of the telomeres and stress-induced premature senescence (Von Zglinicki, 2002), or failure to become immortalized in the case of hTERT-transfected cells (Wright and Shay, 2002). Figure 1d depicts the growth curves for cultures HFF5/hTERT-1, -2, and -3 grown in a lowered O2 environment (5% CO2 and 5% O2). All three cell lines entered a period of crisis when grown at 5% O2, which resembled their behavior under standard conditions. HFF5/hTERT-1 entered crisis at PD38 that lasted 67 days, HFF5/hTERT-2 entered crisis at PD24 that lasted 85 days, and HFF5/hTERT-3 entered crisis at PD24 and continued to grow slowly for the rest of its life span. Therefore, growth in low O2 did not alter whether or not the cells exhibited a period of crisis. Thus, in contrast to some fibroblasts, such as the BJ strain, that are immortalized by the expression of exogenous hTERT alone (Jiang et al., 1999; Morales et al., 1999; Vaziri et al., 1999), and similar to others that exhibited a period of crisis prior to becoming immortalized (MacKenzie et al., 2000), the HFF5 normal human foreskin fibroblast strain was not immortalized directly by transfection with an hTERT plasmid. Although transduction with hTERT plasmid induced telomerase activity that was sufficient to result in the maintenance of telomere length for 19 or 29 PD (HFF5/hTERT-1 and -3, respectively), or in increased telomere length (HFF5/hTERT-2 at PD32 and 35), in each case the cultures subsequently ceased proliferation for periods ranging from 42 to 102 days (Figure 1). The changes permitting these cells to recommence proliferation after crisis have not yet been identified. Loss of p16INK4a expression In order to establish if the Rb pathway was intact in the hTERT immortalized cells, we performed Western analysis on the three immortal cell lines and the cell strain that failed to immortalize, HFF5/hTERT-6, to examine the expression of p16INK4a and the retinoblastoma gene product pRb. The p16INK4a expression underwent a significant decrease in each of the three immortal cell lines, but with differences in timing: it had become undetectable in HFF5/hTERT-1 and -2 cells by PD104 and 48, respectively; and it was downregulated in HFF5/hTERT-3 cells around PD77 (Figure 2a). The pRb level was increased in the absence of p16INK4a Oncogene

Figure 2 Western analysis of p16INK4a and pRb expression in hTERT expressing HFF5 cells. (a) p16INK4a and pRb expression at increasing PDs in cultures HFF5/hTERT-1, -2 and –3, and in -6 at PD29. Western analysis was performed as previously described (Toouli et al., 2002) using the primary mouse monoclonal antibodies pRb (BD Biosciences) and p16INK4a (Neomarkers). The secondary antibody used was horseradish peroxidase-conjugated goat anti-mouse IgG (DAKO). To confirm equal loading of protein lysates following the detection of the proteins of interest, the membranes were hybridized with rabbit polyclonal anti-actin (Sigma) and the secondary horseradish peroxidase conjugated swine anti-rabbit IgG (DAKO). (b) PCR analysis of the promoter and exons 1, 2, and 3 of the p16INK4a gene in HFF5 cells at PD38 and at increasing PDs in cultures HFF5/hTERT-1, -2 and -3. The promoter and exons 1, 2 and 3 were amplified as previously described (Harland et al., 2000; Holland et al., 1995; Noble et al., 1996). (c) Metaphase spread showing the deletion of chromosome 9p material in HFF5/hTERT-3 cells at PD103. Chromosomes were prepared from HFF5/hTERT-3 according to standard cytogenetic methods using QFQ banding (Rooney and Czepulkowski, 1992)

(Figure 2a). The immortal cell lines grown at low O2 (Figure 1d; HFF5/hTERT-1 and -2) also lost the expression of p16INK4a (by PD106 and 172, respectively; data not shown). In order to determine whether loss of the p16INK4a gene correlated with loss of p16INK4a expression, PCR analysis of exons 1, 2 and 3 and the promoter region was performed on genomic DNA. The results of this analysis


3119

indicated that cultures HFF5/hTERT-1 (PD142) and HFF5/hTERT-2 (PD49) had undergone a large homozygous deletion, which resulted in the removal of all three exons and the promoter region of p16INK4a (Figure 2b). Multiplex amplification of GAPDH was performed to confirm the quality of the genomic DNA and as a PCR control in the negative p16INK4a reactions. The p16INK4a expression was not completely absent in culture HFF5/hTERT-3 at the latest time point shown in Figure 2a, but was significantly reduced at PD77 and later PDs. In contrast to HFF5/hTERT-1 and -2, PCR products of correct sizes corresponding to the p16INK4a exons and upstream region were amplified from HFF5/ hTERT-3 genomic DNA at all PD levels tested (Figure 2b). As the p16INK4a gene is frequently silenced by methylation, bisulfite genomic sequencing (Clark et al., 1994; Paul and Clark, 1996) was used to analyse the methylation status of 28 CpG sites (spanning sites 13 to þ 15 relative to the ATG codon) in the core CpG island region of p16INK4a using conditions previously described (Huschtscha et al., 1998). No methylation of the p16INK4a CpG island was observed in HFF5/hTERT-3 at either early (PD27) or late (PD99) time points. As the decrease in p16INK4a expression in HFF5/ hTERT-3 cells with increasing PDs was not the result of promoter methylation or gene deletion, cytogenetic analysis using QFQ banding was performed to detect chromosomal abnormalities. Cytogenetic analysis at PD103 indicated that one copy of chromosome 9 had a large deletion in the short arm most likely involving the 9p21 region (Figure 2c). Therefore, p16INK4a hemizygosity may account for the decreased p16INK4a expression in HFF5/hTERT-3 cells. Further analysis of HFF5/hTERT-3 cells revealed that occasional metaphase spreads (B1%) had two intact copies of chromosome 9 (data not shown), indicating that the chromosomal rearrangement that led to the loss of chromosome 9p material occurred in a subset of the cells, and that these cells presumably had a growth advantage which allowed them to overgrow the cells without a 9p deletion. The gradual reduction of p16INK4a expression (Figure 2a) would correlate with the accumulation of cells that had become hemizygous for p16INK4a. Apart from the loss of the 9p material the majority of HFF5/hTERT-3 cells did not exhibit any other chromosomal abnormalities. The loss of p16INK4a expression observed in cultures HFF5/hTERT-1, -2, and -3 did not coincide with escape from crisis, but was a later event in each case. Our results are in agreement with a study of periodontal fibroblasts (Tsutsui et al., 2002) in which two separately established hTERT-transduced cell strains lost expression of p16INK4a and an earlier study (Kiyono et al., 1998) which demonstrated that human foreskin fibroblasts expressing hTERT had a lower level of p16INK4a protein expression compared to late passage control fibroblasts. This is in contrast to previously published studies of BJ foreskin fibroblasts that found no change in p16INK4a or pRb levels in hTERT-immortalized cells (Jiang et al., 1999; Morales et al., 1999). These

contrasting results may be explained by differences between the fibroblast cell strains used in the various studies. Altered p53 and p21 expression In addition to documenting disruption of the pRb/ p16INK4a pathway in the HFF5/hTERT-1, -2, and -3 cultures, we wanted to determine whether the p21/p53 pathway was functioning normally. Western analysis of p21 levels in the three immortal cell lines and the cell strain HFF5/hTERT-6 indicated that expression increased slightly with increasing time in culture for both HFF5/hTERT-1 and -2 cells but decreased around PD83 for HFF5/hTERT-3. The p21 expression was similar in the parental untransfected HFF5 cells and HFF5/hTERT-6 at PD29 (Figure 3). The p53 expression remained constant at all time points examined for HFF5/hTERT-1 and -2 and was expressed at similar levels in the early PDs of HFF5/hTERT-3. Around PD77, the p53 expression increased in HFF5/hTERT-3 cells, coincident with the decrease of p21 expression (Figure 3). Overexpression of p53 is frequently associated with loss of wild-type function and mutation of the p53 gene (Lane and Benchimol, 1990). DNA damage response To assess the response of p53 in the immortalized cell lines to DNA damage cultures HFF5/hTERT-1, -2, and -3 and the parental cells, HFF5, were exposed to the DNA intercalating agent, actinomycin D. Western analysis indicated that after exposure to 7.5 nM actinomycin D for 24 h, the level of p53 expression increased dramatically in the parental HFF5 cells and the HFF5/ hTERT-1 and -2 cell lines. The p53 expression levels were increased in the HFF5/hTERT-3 line following actinomycin D treatment but to a much lesser extent (Figure 4a). The expression of p21 also substantially increased in parental HFF5 cells and the HFF5/ hTERT-1 and -2 cells after actinomycin D treatment but only slightly in culture HFF5/hTERT-3 (Figure 4a). Flow cytometry of propidium iodide-stained cells was

Figure 3 Western analysis of p21 and p53 expression in hTERT expressing HFF5 cells. The p21 (Cip1/WAF1; BD Biosciences) and p53 (Ab-8; Neomarkers) expression in HFF5 cells at PD35, at increasing PDs in cultures HFF5/hTERT-1, -2 and -3 and in HFF5/hTERT-6 at PD29. Actin expression was used as a loading control for both Westerns Oncogene


3120

Figure 4 Response of parental and hTERT expressing HFF5 cells to DNA damage. Cells were exposed to 7.5 nM actinomycin D (Sigma) for 24 h. (a) Western analysis of p53 and p21 expression in HFF5 cells at PD40 and at the PDs indicated in cultures HFF5/ hTERT-1, -2 and -3. (b) Cell cycle analysis was performed on propidium iodide (Sigma) stained cells as previously described (Toouli et al., 2002). The percentage of cells in S phase is indicated

performed to assess whether the immortal cells maintained normal G1 checkpoint control following the treatment with actinomycin D. After actinomycin D treatment there was a substantial reduction in the percentage of HFF5, HFF5/hTERT-1 and -2 cells in S phase, but there was no effect on HFF5/hTERT-3 cells (Figure 4b). The normal progression of HFF5/hTERT-3 cells into S phase and the absence of a significant increase in p53 expression following treatment with actinomycin D indicate that these cells no longer had normal G1 checkpoint control. Sequence analysis of p53 To determine whether the substantial increase in p53 expression at PD77 and the lack of G1 checkpoint response following treatment of HFF5/hTERT-3 cells with actinomycin D was due to a p53 mutation, genomic DNA was isolated from parental HFF5 and HFF5/ hTERT-3 cells and exons 5–9 of p53 were sequenced in both directions using primers and conditions previously described (McGregor et al., 2002). Two heterozygous mutations were identified in HFF5/hTERT-3 cells: the Oncogene

first a transversion (G to T) at codon 135 in exon 5 resulting in the amino-acid substitution of Cys to Phe and the second a nonsense mutation arising from a transition (C to T) at codon 213 in exon 6, resulting in replacement of Arg by a stop codon. Both of these mutations have been previously identified in numerous tumors (Olivier et al., 2002). Sequence analysis of p53 in the parental HFF5 cells revealed it to be wild type. Cloning and sequencing of PCR product containing both exons 5 and 6 revealed that the mutations in exon 5 and 6 were on separate alleles. Mutation of the p53 gene as observed in immortal HFF5/hTERT-3 cells is a novel finding not previously reported in hTERT-immortalized cell lines. As expected, cells containing this mutant p53 were unable to arrest in G1 following exposure to the DNA intercalating agent actinomycin D. Therefore the p53/p21-dependent DNA damage response pathway was disrupted in these cells, which could lead to further genomic instability. Evidence presented in this study suggests that introduction of hTERT into HFF5 cells did facilitate a large extension of the cells’ life span, allowing these cells to bypass senescence and potentially become immortal, but it is important to note that this was followed by additional genetic changes in both the pRb/p16INK4a pathway and the p53 pathway. Interestingly, in a previous study with this cell strain, it was found that immortalization by hTERT was facilitated by coexpression of a gene, mot-2/mthsp70/GRP75, that disrupts p53 function (Kaul et al., 2003). However, in contrast to the genetic changes that are presumably required for the cells to escape from crisis, the changes that we identified in the p16INK4a and p53 genes occurred later. They are therefore most likely not required for immortalization of these cells but rather are selected for during continuous growth in vitro, presumably by providing a growth advantage. Previous reports of cells that have been immortalized by hTERT in the absence of any p16INK4a or p53 mutations also suggest that these changes are not essential for immortalization. Functional loss of these genes increases the probability that cells continue cycling when their normal counterparts undergo growth arrest, and may also permit the accumulation of other genetic changes that result in an increased growth rate. The demonstration in this study and in previous studies that p16INK4a expression may be disrupted in some hTERT-transduced human fibroblast cultures, and now the finding that more rarely p53 mutations may occur emphasizes the need for caution in the use of such cells for studies of normal biology or for tissue engineering. hTERT-transduced cells need to be monitored for genetic instability and the accumulation of mutations in both the p16INK4a/pRb and p53 pathways. Acknowledgements This work was supported by a project grant from the National Health and Medical Research Council of Australia, and the Carcinogenesis Fellowship of the Cancer Council New South Wales. We thank Christine Smyth for FACscan analysis, Alessandra Muntoni for advice on p53 sequencing, and Helen Rizos for advice on p16INK4a PCR analysis.


3121 References Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu C-P, Morin GB, Harley CB, Shay JW, Lichtsteiner S and Wright WE. (1998). Science, 279, 349–352. Bryan TM, Englezou A, Gupta J, Bacchetti S and Reddel RR. (1995). EMBO J., 14, 4240–4248. Bryan TM and Reddel RR. (1994). Crit. Rev. Oncogenesis, 5, 331–357. Campisi J. (2001). Trends Cell Biol., 11, S27–S31. Clark SJ, Harrison J, Paul CL and Frommer M. (1994). Nucleic Acids Res., 22, 2990–2997. Colgin LM and Reddel RR. (1999). Curr. Opin. Genet. Dev., 9, 97–103. Colgin LM, Wilkinson C, Englezou A, Kilian A, Robinson MO and Reddel RR. (2000). Neoplasia, 2, 426–432. Counter CM, Avilion AA, LeFeuvre CE, Stewart NG, Greider CW, Harley CB and Bacchetti S. (1992). EMBO J., 11, 1921–1929. Farwell DG, Shera KA, Koop JI, Bonnet GA, Matthews CP, Reuther GW, Coltrera MD, McDougall JK and Klingelhutz AJ. (2000). Am. J. Pathol., 156, 1537–1547. Franco S, MacKenzie KL, Dias S, Alvarez S, Rafii S and Moore MAS. (2001). Exp. Cell Res., 268, 14–25. Girardi AJ, Jensen FC and Koprowski H. (1965). J. Cell Comp. Physiol., 65, 69–84. Harland M, Holland EA, Ghiorzo P, Mantelli M, BianchiScarra G, Goldstein AM, Tucker MA, Ponder BA, Mann GJ, Bishop DT and Newton BJ. (2000). Genes Chromosomes Cancer, 28, 45–57. Harley CB. (1991). Mutat. Res., 256, 271–282. Hayflick L and Moorhead PS. (1961). Exp. Cell Res., 25, 585–621. Holland EA, Beaton SC, Becker TM, Grulet OMC, Peters BA, Rizos H, Kefford RF and Mann GJ. (1995). Oncogene, 11, 2289–2294. Huschtscha LI and Holliday R. (1983). J. Cell Sci., 63, 77–99. Huschtscha LI, Noble JR, Neumann AA, Moy EL, Barry P, Melki JR, Clark SJ and Reddel RR. (1998). Cancer Res., 58, 3508–3512. Jiang XR, Jimenez G, Chang E, Frolkis M, Kusler B, Sage M, Beeche M, Bodnar AG, Wahl GM, Tlsty TD and Chiu CP. (1999). Nat. Genet., 21, 111–114. Kaul SC, Yaguchi T, Taira K, Reddel RR and Wadhwa R. (2003). Exp. Cell Res., 286, 96–101. Kim H, Farris J, Christman SA, Kong BW, Foster LK, O’Grady SM and Foster DM. (2002). Biochem. J., 365, 765–772. Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PLC, Coviello GM, Wright WE, Weinrich SL and Shay JW. (1994). Science, 266, 2011–2015. Kiyono T, Foster SA, Koop JI, McDougall JK, Galloway DA and Klingelhutz AJ. (1998). Nature, 396, 84–88.

Lane DP and Benchimol S. (1990). Genes Dev., 4, 1–8. MacKenzie KL, Franco S, May C, Sadelain M and Moore MAS. (2000). Exp. Cell Res., 259, 336–350. McGregor F, Muntoni A, Fleming J, Brown J, Felix DH, MacDonald DG, Parkinson EK and Harrison PR. (2002). Cancer Res., 62, 4757–4766. Morales CP, Holt SE, Ouellette M, Kaur KJ, Yan Y, Wilson KS, White MA, Wright WE and Shay JW. (1999). Nat. Genet., 21, 115–118. Noble JR, Rogan EM, Neumann AA, Maclean K, Bryan TM and Reddel RR. (1996). Oncogene, 13, 1259–1268. O’Hare MJ, Bond J, Clarke C, Takeuchi Y, Atherton AJ, Berry C, Moody J, Silver ARJ, Davies DC, Alsop AE, Neville AM and Jat PS. (2001). Proc. Natl. Acad. Sci. USA, 98, 646–651. Olivier M, Eeles R, Hollstein M, Khan MA, Harris CC and Hainaut P. (2002). Hum. Mutat., 19, 607–614. Olovnikov AM. (1973). J. Theor. Biol., 41, 181–190. Paul CL and Clark SJ. (1996). Bio Techniques, 21, 126–133. Perrem K, Bryan TM, Englezou A, Hackl T, Moy EL and Reddel RR. (1999). Oncogene, 18, 3383–3390. Reddel RR. (2000). Carcinogenesis, 21, 477–484. Rogan EM, Bryan TM, Hukku B, Maclean K, Chang ACM, Moy EL, Englezou A, Warneford SG, Dalla-Pozza L and Reddel RR. (1995). Mol. Cell. Biol., 15, 4745–4753. Rooney DE and Czepulkowski, BH. (1992). Human Cytogenetics: A Practical Approach. Vol I. Constitutional Analysis. Oxford University Press: Oxford. Shay JW and Wright WE. (1989). Exp. Cell Res., 184, 109–118. Toouli CD, Huschtscha LI, Neumann AA, Noble JR, Colgin LM, Hukku B and Reddel RR. (2002). Oncogene, 21, 128–139. Tsutsui T, Kumakura SI, Yamamoto A, Kanai H, Tamura Y, Kato T, Anpo M, Tahara H and Barrett JC. (2002). Carcinogenesis, 23, 2111–2117. Vaziri H and Benchimol S. (1998). Curr. Biol., 8, 279–282. Vaziri H, Squire JA, Pandita TK, Bradley G, Kuba RM, Zhang H, Gulyas S, Hill RP, Nolan GP and Benchimol S. (1999). Mol. Cell. Biol., 19, 2373–2379. Von Zglinicki T. (2002). Trends Biochem. Sci., 27, 339–344. Weinrich SL, Pruzan R, Ma L, Ouellette M, Tesmer VM, Holt SE, Bodnar AG, Lichtsteiner S, Kim NW, Trager JB, Taylor RD, Carlos R, Andrews WH, Wright WE, Shay JW, Harley CB and Morin GB. (1997). Nat. Genet., 17, 498–502. Wright WE, Pereira-Smith OM and Shay JW. (1989). Mol. Cell. Biol., 9, 3088–3092. Wright WE and Shay JW. (2002). Nat. Biotechnol., 20, 682–688.

Oncogene