A TP53 Mutation Detected in Cells Established From an Osteosarcoma, but Not in the Retinoblastoma of a Patient with Bilateral Retinoblastoma and Multiple Primary Osteosarcomas 0

•

Eivind Hovig, Ashlld Andreassen, BjOrn M. Fangan, and Anne-Lise BOrresen

ABSTRACT: A patient with bilateral retinoblastoma and subsequent multiple primary osteosarcomas

has been described previously. Osteosarcoma cell lines established from this patient were shown to express a shortened RB1 mRNA transcript and no detectable normal Rb protein. We now show that the osteosarcoma cell lines have lost one TP53 allele and contain a mutation in exon 8 codon 286 [GAA to A A A (Glu to Lys)] in the remaining allele. Consequently, the osteosarcoma cell lines have no normal Rb protein and no normal p53 protein. Neither constitutional DNA nor DNA extracted from a retinoblastoma of the left eye of the patient contained the TP53 mutation, suggesting that the TP53 mutation in the osteosarcoma cells may represent a tumor-promoting mutation, which confers a selective growth advantage. If both RB1 and TP53 are involved in the initiation of osteosarcoma, the mechanisms for development of the retinoblastoma and osteosarcoma tumors are different.

INTRODUCTION Both TP53 and RB1 are tumor suppressor genes, and the respective proteins are most likely involved in cell cycle regulation [1]. Both proteins are nuclear phosphoproteins, with changing levels of protein phosphorylation throughout the cell cycle [2, 3]. Both proteins have been shown to complex with other cellular proteins, as well as tumor virus oncoproteins. Mutations in these genes have been observed with high frequency in several tumor types, but the mutational spectra in tumors for these genes vary greatly. In the RB1 gene, fewer than 10 amino acid substitution mutations have been noted, whereas several hundred such mutations have been reported in the TP53 gene. All retinoblastomas yet examined extensively have been shown to carry RB1 mutations. Most commonly found defects result in the absence of expressed mRNA through deletions, methylation, or promoter defects, or in shortened mRNA transcripts. In a few cases, mutated mRNA has been observed leading to missense incorporation of amino acids in the protein. The lack of normal Rb protein in retinoblastoma has

From the Departments of Genetics (E. H., B. M. F., A.-L. B.) and T u m o r Biology (A. A., B. M. E ), Institute of Cancer Research, The

Norwegian Radium Hospital, Oslo, Norway. Address reprint requests to: E. Hovig, M.D., Department of Genetics, Institute for Cancer Research, The Norwegian Radium Hospital, 0310 Oslo, Norway. Received January 27, 1992; accepted July 31, 1992.

been considered to illustrate that mutations of both alleles of the RB1 gene are a prerequisite for generation of a retinoblastoma, i.e., the initiation event. The importance of the RB1 gene in the initiation versus progression of other tumors is still unclear. TP53 mutations were recently demonstrated to be important in the rare Li-Fraumeni syndrome [4-7]. Osteosarcomas are common malignancies in this syndrome. The biologic significance of TP53 mutations in other tumor types is less clear, with certain codons more frequently involved than others in mutations in specific tumor types. However, because TP53 mutations occur in more than half of all tumors so far examined, this high frequency apparently may be explained at least in part by a selection for TP53 mutations in the progression process, since established cell lines show even higher frequencies than primary tumors. The involvement of TP53 in progression is also supported by the finding that osteosarcomas are among the tumor types showing significant changes in loss of heterozygosity (LOH) frequencies for both the RB1 and TP53 loci [8, 9]. To our knowledge, no retinoblastoma has yet been shown to be affected at the TP53 locus. We previously developed an assay for screening of mutations in the conserved domains of the TP53 gene [10], based on constant denaturant gel electrophoresis (CDGE) [11]. Using this assay, we investigated whether TP53 mutations could be detected in a patient from whom both a paraffin-fixed eye with retinoblastoma, independent isolations of osteosarcoma cells, and normal fibroblasts were available.

178 Cancer Genet Cytogenet 64:178-182 (1992) 0165-4608/92/$05.00

© 1992 Elsevier Science Publishing Co., Inc. 655 Avenue of the Americas, New York, NY 10010

TP53 Mutation in an Osteosarcoma MATERIALS AND METHODS

Source of DNA The patient, a male caucasian with no family history of retinoblastoma or osteosarcoma, was described previously [12]. Both eyes were enucleated at age 10 months because of locally advanced bilateral retinoblastoma. His second primary cancer was diagnosed at age 14 years, at which time extensive and multiple osteosarcomatous manifestations were observed. Whether the multiple tumor localizations were metastases from a single primary or represented a case of multiple primaries could not be established. For the present examinations, DNA was extracted from various sublines of two independent isolates of cell lines derived from an osteosarcoma of the patient. A detailed characterization of these isolates was reported previously [13]. DNA from fibroblasts derived from the patient was also extracted. These extractions were performed according to standard procedures (phenol/ chloroform extraction and ethanol precipitation) [14]. DNA from the paraffin-embedded left eye with retinoblastoma from this patient was extracted essentially as described [15].

Polymerase Chain Reaction Conditions The strategy and conditions for the generation of PCR products suitable for analysis of TP53 mutations were described previously [10]. PCR was performed using 100-300 ng template DNA in 50 mM Tris.HC1, (pH 8.6) with 10 mM KC1, 0.8-3 mM MgC12, 0.2 mM each dNTP, 25 or 100 pmol each primer [25 of the purified and 100 of the unpurified primer), and 2.5 U Taq polymerase (AmpliTaq, Cetus). The 100-~1 PCR mixture was incubated in a Perkin Elmer/Cetus thermocycler for 35 cycles at 94°C [1 minute), 55°C (1 minute), and 72°C (1 minute). The primer sets amplify across the four conserved regions where ~>95% of TP53 mutations have been identified [16]. All PCR products were analyzed for purity by Z5 % polyacrylamide gel electrophoresis (PAGE), followed by staining with ethidium bromide. Denaturing Gel Electrophoresis Perpendicular denaturing gradient gels (10 x 8 x 0.1 cm) contained 12.5% acrylamide in TAE [0.04 M Tris-acetate, 0.001 M EDTA, pH 8.0), DATD (N'N'-diallyl tartardiamide) as cross-linker [0,55 g/100 ml), and varying denaturant concentrations consisting of urea and formamide [100% denaturant corresponds to 7 M urea and 40% vol/vol formamide). The gels were polymerized with ammonium persulfate [0.8 mg/gel solution) and TEMED [N,N,N;N'-tetramethylethylenediamine, 1.6 pl/ml gel solution]. The gradient gels were cast with a gravitational gradient mixer. All reagents used were of electrophoretic grade. Gels were run submerged in TAE buffer at 56°C at 80 V constant, using self-constructed cell adapted to the mini-protean electrophoresis cell system (Bio-Rad). The modification allowed the glass plates surrounding the gels to be in direct contact with buffer on both sides. Extensive circulation of the buffer was provided during the runs. Running time was I hour 45 minutes. After electrophoresis, the gels were stained for a few minutes in EtBr (2 rag/1 TAE) and photographed by ultraviolet transilluminator. Parallel gels used in CDGE contained the same chemicals as the perpendicular denaturing gradient gels, but with

179 a uniform denaturant concentration through each gel. Running conditions were the same as those used for the perpendicular gels, but with running times varying from 2 to 4 hours. LOH Studies DNA was digested with appropriate restriction endonucleases. Digested DNA was separated by agarose gel electrophoresis and transferred to nylon membrane. The blots were hybridized with 32p-labeled probes, prepared according to the random oligolabeling method [17], and autoradiographed. Southern blotting was performed using alkaline blotting, a modification of the method of Southern [18]. The probes for D17S34 (VNTR locus] and D17S5 [VNTR locus) were obtained from ATCC. pBHP53 [polymorphic with restriction enzyme BamHI) was provided by Bj~rn H~yheim (Utah).

DNA Sequencing Sequencing was performed by direct sequencing of PCR products. Amplification was performed with one of the primers biotinylated. The products were then purified by agarose gel electrophoresis and isolated from gel with GeneClean (BIO 101). The biotinylated PCR products were subsequently sequenced directly using a standard dideoxy sequencing reaction with Dynabeads M-280, streptavidin [Dynal AS) as solid support [19].

RESULTS AND DISCUSSION We previously reported chromosome 13 findings of patient O.H. [13, 19]. This patient had multiple primary osteosarcomas and a previous history of bilateral retinoblastoma, with no affected family members. Two separate isolates of osteosarcoma cells were established in vitro; one through soft agar cultivation (OHS cells) and the other through xenografting in nude mice [OHX cells). Both isolates showed multiple chromosome 13 alterations. (A review of these cell lines and their chromosome 13 alterations, was provided by Hovig et al. [13]). Previous studies demonstrated that the osteosarcoma cells contained RB1 mRNA of lower molecular weight [20], LOH was evident for parts of chromosome 13q [13, 21-23], and although there were common chromosome 13q deletions in the osteosarcoma cell lines, distinct features were also evident, including deletions and duplications of parts of chromosome 13q. No normal Rb protein could be detected [23]. With respect to chromosome 17 alterations, Diller et al. [24] demonstrated the presence of mutant p53 protein in OHS cells through use of antibodies. This protein binds hsc70, a feature of mutant protein. DNA from the fibroblasts of patient OH were heterozygous for three of the tested loci at the distal part of chromosome 17p: D17S34, D12S5, and pBHP53. The osteosarcoma cell lines from both isolates had lost one allele at all loci [data not shown). The polymorphism for pBHP53 is closely linked to the TP53 locus at chromosome 17p13. We also analyzed a PCR restriction fragment length polymorphism (RFLP) in the TP53 locus (codon 72), but the patient proved not to be informative for this locus. These data indicate that the osteosarcoma cell lines had lost one copy of chromosome 17 band

180

E. Hovig et al.

Figure 1 Constant denaturant gel electrophoresis of PCR-amplified DNA from OHF (fibroblasts), OHS (of agar-established osteosarcoma cells), OHX (xenograft-established osteosarcoma cells), and OHE (retinoblastoma of the left eye} showing a TP53 mutation in OHS and OHX cells. Running conditions: gel temperature 56°C at 80 V for 2 hours. Positive photograph of EtBr-stained gel. C: wild-type control. p13. DNA extracted from a paraffin-embedded eye containing retinoblastoma tumor material obtained from the patient could be used only for PCR-based analysis because the DNA was severely degraded. LOH for chromosome 17p is fre-

quently observed in tumors with mutated TP53. These include adrenocortical carcinomas [25], lung carcinomas [26], neurofibrosarcomas [27], hepatocellular carcinomas [28], bladder carcinomas [29], breast carcinomas [30], and others. TP53 is the most frequently altered gene in h u m a n tumors. LOH of chromosome 17p is often accompanied by missense mutations in the remaining TP53 allele. The LOH data therefore prompted us to search for base mutations at the TP53 locus in the DNA from the cell lines. We tested all domains covered by our CDGE assay [10, 11] to detect possible deviations from the wild-type TP53 sequence in all tumor DNAs. We detected aberrantly migrating CDGE bands in the PCR fragments covering codons 265 to 301 (Fig. 1) from all osteosarcoma sublines tested. No mutations were detected in DNA from fibroblasts or the primary retinoblastoma. To verify this finding, we sequenced all samples examined in CDGE analysis. Figure 2 shows sequencing results for DNA from fibroblasts, the retinoblastoma DNA, and two osteosarcoma sublines. All osteosarcoma sublines displayed the same TP53 point mutation in exon 8: a GAA to AAA in codon 286 (Glu~Lys). This mutation falls outside the regions most frequently reported to be mutated in other tumors. In the literature, only one other mutation involving this codon has been described. It was reported to occur in a case of primary colon carcinoma [31]. None of the mutated samples showed any trace of wildtype sequence, neither in the CDGE assay nor when sequenced. This strengthens the assumption that one TP53

Figure 2 Part of autoradiogram of sequenced PCR products showing parts of the TP53 gene with the sequence from codon 271 to codon 289. Wild-type DNA and amino acid sequences (left). Sites of mutation in codon 286 and the resultant amino acid changes (arrows). DNA designations as shown in Figure 1.

p

I

3 3

3 _3 3 A ]286 A [Lys 3 3 A I

3"

TP53 Mutation in an Osteosarcoma

allele had indeed been lost, as indicated by the LOH examination. Because the primary osteosarcoma is not available for analysis, the TP53 mutation observed in the osteosarcoma cell lines may have been a result of i n vitro establishment. Considering the wealth of sequenced mutations at the TP53 locus and their respective position frequencies, however, the chance that two i n d e p e n d e n t l y established sublines would display the exact same base mutation is highly unlikely unless the mutation was present in the primary osteosarcoma, either as a change close to tumor initiation or as the result of clonal expansion. That both genomic DNA and retinoblastoma DNA only display wild-type TP53 sequence demonstrates that the patient does not have an inherited germline TP53 mutation and that TP53 mutations apparently played no role in development of the retinoblastoma. We do not know whether the retinoblastoma has any LOH for chromosome 17. None of the studies of patient OH permit definite conclusions with regard to the relative importance of the two proteins pl05-Rb and p53 in genesis and/or progression of the osteosarcoma, because it is not possible to perform retrospective analysis on the osteosarcoma primary tumor. Leukemic cell lines can harbor as many as four different mutations in the TP53 gene in the same cell line [32], indicating a potential growth advantage, p53 mutant proteins with different missense mutations have different oncogenic potential [16], but whether the codon 286 mutation confers a growth advantage on the tumor cells is not known. The TP53 mutation may be a secondary, progression-related mutation. The c o m b i n e d loss of the two genes may be important in the genesis of osteosarcoma. Patients showing the hereditary bilateral form of retinoblastoma have a 200-fold increased incidence of developing secondary sarcomas, especially osteosarcomas [33]. In a recent study of IADHof osteosarcomas, Scheffer et al. [34] showed a complete association between LOH for chromosome 13 and 17 in sporadic primary tumors from five patients. They imply that TP53 may be involved in initiation of osteosarcoma. The data for patient O.H. are compatible with their findings. The RB1 gene may be operational in initiation of both tumor types. Having mutated TP53 may be one of several necessary genetic events required in addition to the RB1 mutation for osteosarcoma to develop. The authors thank Sigrid Lystadand Frank Karlsen for excellent technical assistance, Dr. Alexandra Kat for critical reading of the manuscript, and the Norwegian Research Council for Science and the Humanities and the Nordic Fund for Technology and Industrial Development for financial support.

181

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16. 17.

18. 19.

REFERENCES 1. Levine AJ, Momand J (1990): Tumor suppressor genes-the P53 and retinoblastoma sensitivity genes and gene products. Biochim Biophys Acta 1032:119-136. 2. Chen P-L, Scully P, Shew J-Y,Wang JYJ, Lee W-H (1989): Phosphorylation of the retinoblastoma gene product is modulated during the cell cycle and cellular differentiation. Cell 58: 1193-1198. 3. BischoffJR, Friedman PN, Marshak DR, Prives C, Beach D (1990):

20.

21.

22.

Human P53 is phosphorylated by P6-cdc2 and cyclin-B-Cdc2. Proc Natl Acad Sci USA 87:4766-4770. Malkin D, Li FP, Strong LC, Fmumeni JF, Nelson CE, Kim DH, Kassel J, Gryka/VIA,Bischoff FZ, Tainsky MA, Friend SH (1990): Germ line P53 mutations in a familial syndrome of breast cancer, sarcomas, and other neoplasms. Science 250;1233-1238. Srivastava S, Zou ZQ, Pirollo K, Blattner W, Chang EH (1990): Germ-line transmission of a mutated P53 gene in a cancer-prone family with Li-Fraumeni syndrome. Nature 348:747-749. Law JC, Strong LC, Chidambaram A, Ferrell RE (1991): A germ line mutation in exon 5 of the p53 gene in an extended cancer family. Cancer Res 51:6385-6387. Santib~tfiez-KorefMF, Birch JM, Hartley AL, Jones PHM, Craft AW, Eden T, Crowther D, Kelsey AM, Harris M (1991):p53 germline mutations in Li-Fmumeni syndrome. Lancet 338:1490-1491. Toguchida J, Ishizaki K, Sasaki MS, Ikenaga M, Masayuki MS, Sugimato M, Kotoura Y, Yamamuro T (1988): Chromosomal reorganization for the expression of recessive mutation of retinoblastoma susceptibility gene in the development of osteosarcoma. Cancer Res 48:3939-3943. Miller CW, Also A, Tay C, Slamon D, Ishizaki K, Toguchida J, Yamamuro T, Lampkin B, Koeffler HP (1990): Frequency and structure of p53 rearrangements in human osteosarcoma. Cancer Res 50:7950-7954. Berresen A-L, Hovig E, Smith-Sarensen B, Malkin D, Lystad S, Andersen TI, Nesland JM, Isselbacher KJ, Friend SH (1991):Constant denaturant gel electrophoresis as a rapid screening technique for p53 mutations. Proc Natl Acad Sci USA 88:8405-8409. Hovig E, Smith-Sarensen B, Bragger A, B~rresen A-L (1991): Constant denaturant gel electrophoresis, a modification of denaturing gel electrophoresis, in mutation research. Mutat Res 267:63-67. Fodstad ~, Br¢gger A, Bruland g), Solheim ~P, Nesland JM, Pihl A (1986): Establishment and cytogenetic analysis of a cell line established from a patient with multiple osteosarcoma, occurring 13 years after treatment for bilateral retinoblastoma. Int J Cancer 38:33-40. Hovig E, Lethe RA, Farmnts G, Bragger A, Fodstad 0, B¢rresen A-L (1991):Chromosome 13 alterations in osteosarcoma cell lines derived from a patient with previous retinoblastoma. Cancer Genet Cytogenet 57:31-40. Kunkel LM, Smith KD, Boyer SH, Bomonkar DS, Wachtel SS, Miller OJ, Breg WR, Jones HWJ, Rary JM (1977):Analysis of human Y-chromosome-specific reiterated DNA in chromosome variants. Proc Natl Acad Sci USA 74:1245-1249. Mies C, Houldsworth J, Chaganti RS (1991): Extraction of DNA from paraffin blocks for Southern blot analysis. Am J Surg Pathol 15:169-174. Hollstein M, Sidransky D, Vogelstein B, Harris CC (1991): p53 Mutations in human cancers. Science 253:49-53. Feinberg AP, Vogelstain B (1983): A technique for radiolabelling DNA restriction endonuclease fragments to high specific activity. Anal Biochem 132:6-13. Southern EM (1975):Detection of specific sequences among DNA fragments separated by gel electrophoresis. Mol Bio198:503-517. Hultman T, St~l S, Hornes E, Uhl~n M (1989): Direct solid phase sequencing of genomic and plasmid DNA using magnetic beads as solid support. Nucleic Acids Res 17:4937-4946. Hovig E, Winberg J-O, Fodstad 0, Bragger A, Gedde-Dahl T (1987): Chromosome 13 instabilityand estemse D expression in an osteosarcoma cell line. Cancer Genet Cytogenet 24:327-334. Fung Y-KT, Murphree AL, T~ng A, Qian J, Hinrichs SH, Benedict, WF (1987): Structural evidence for the authenticity of the human retinoblastoma gene. Science 236:1657-1661. Hansen MF, Koufos A, Gallie BL, Phillips RA, Fodstad ~), Bragger A, Gedde-Dahl T, CaveneeWK (1985):Osteosarcoma and

182

23.

24.

25.

26.

27.

E. Hovig et al.

retinoblastoma: A shared chromosomal mechanism revealing recessive predisposition. Proc Natl Acad Sci USA 82:6216-6220. Shew J-Y,Ling N, Yang X, Fodstad q), Lee W-H (1989): Antibodies detecting abnormalities of the retinoblastoma susceptibility gene product (ppll0Rb) in osteosarcomas and synovial sarcomas. Oncogene Res 1:205-214. Diller L, Kassel J, Neslon CE, Gryka MA, Litwak G, Gebhardt M, Bressac B, Ozturk M, Baker SJ, Vogelstein B, Friend SH (1990): P53 functions as a cell cycle control protein in osteosarcomas. Mol Cell Biol 10:5772-5781. Yano T, Linehan M, Anglard P, Lerman MI, Daniel LN, Stein CA, Robertson CN, LaRocca R, Zbar B (1989): Genetic changes in human adrenocortical carcinomas. J Natl Cancer Inst 81: 518-523. Yokota J, Wada M, Shimosato Y, Terada M, Sugimura T (1987): Loss of heterozygosity on chromosomes 3, 13, and 17 in smallcell carcinoma and on chromosome 3 in adenocarcinoma of the lung. Proc Natl Acad Sci USA 84:9256-9260. Menon AG, Anderson KM, Riccardi VM, Chung RY, Whaley JM, Yandell DW, Farner GE, Freiman RN, Lee JK, Li FP, Barker DE Ledbetter DH, Kleider A, Martuza RL, Gusella JE Seizinger BR (1990): Chromosome-17p deletions and P53-gene mutations associated with the formation of malignant neuroofibrosarcomas in yon Recklinghausen neurofibromatosis. Proc Natl Acad Sci USA 87:5435-5439.

28. Fujimori M, Tokino T, Himo O, Kitagawa T, Imamura T, Okamoto E, Mitsunobu M, Ishikawa T, Nakagama H, Harada H, Yagura M, Matsubura K, Nakamura Y (1991): Allelotype study of primary hepatocellular carcinoma. Cancer Res 51:89-93. 29. Tsai YC, Nichols PW, Hiti AL, Williams Z, Skinner DG, Jones PA (1990): Allelic losses of chromosomes 9, 11, and 17 in human bladder cancer. Cancer Res 50:44-47. 30. Cropp CS, Lidereau R, Campbell G, Champene MH, Callahan R (1990): Loss of heterozygosity on chromosome-17 and chromosome-18 in breast carcinoma- 2 additional regions identified. Proc Nail Acad Sci USA 87:7737-7741. 31. Baker SJ, Markowitz S, Fearon ER, Willson JKV, Vogelstein B (1990): Suppression of human colorectal carcinoma cell growth by wild-type-P53. Science 249:912-915. 32. Cheng J, Scully P, Shew JY, Lee W-H, Vila V, Haas M (1990): Homozygous deletion of the retinoblastoma gene in an acute lymphoblastic leukemia-(T) cell line. Blood 75:730-735. 33. Kitchin FD, Ellsworth RM (1974): Pleiotropic effects of the gene for retinoblastoma. J Med Genet 11:244-246. 34. Scheffer H, Kruize YCM, Osinga J, Kuiken G, Oosterhuis JW, Leeuw JA, Koops HS, Buys CHCM (1991): Complete association of loss of heterozygosity of chromosome 13 and chromosome 17 in osteosarcoma, Cancer Genet Cytogenet 53:45-55.