[CANCER RESEARCH 51.6194-6198. November 15. 19911 Advances in Brief Mutations in the p53 Gene in Primary Human Breast Cancers1 R. J. Osborne,2 G. R. Merlo, T. Mitsudomi, T. Venesio, D. S. Liscia, A. P. M. Cappa, I. Chiba, T. Takahashi, M. M. Ñau, R. Callahan, and J. D. Minna3 .\avy Medical Oncology Branch [R. J. O.. T. M.. I. O, T. T.. M. M. N.. J. D. M.] and Oncogenetics Section /G. R. M.. T. l'.. R. C.J National Cancer Institute, Bethesda. Maryland 20X92, and Ospedale S. Giovanni Vecchio. Turin. Italy ¡D.S. L, A. P. M. C.J Abstract Twenty-six primary breast tumors were examined for mutations in the p53 tumor suppressor gene by an KNase protection assay and nucleotide sequence analysis of PCR-amplified p53 complementary DNAs. Each method detected p53 mutations in the same three tumors (12%). One tumor contained two mutations in the same alÃ-ele. Single strand confor mation polymorphism analysis of genomic DNA and complementary DNA proved more sensitive in the detection of mutations. Combining this technique with the other two a total of 12 mutations in the p5i gene were demonstrated in 11 tumors (46%), and a polymorphism at codon 213 was detected in another tumor. Loss of heterozygosity on chromosome 17p was detected by Southern blot analysis in 30% of the tumor DNAs. Not all of the tumors containing a point mutation in p53 also had loss of heterozygosity of the remaining alÃ-ele, suggesting that loss of heterozy gosity may represent a later event. Introduction There is now substantial evidence that genetic abnormalities affecting the p53 gene are frequently associated with the path- ogenesis of several neoplasias, particularly solid tumors such as breast, colon, and lung carcinomas (1). The mutations appear to cluster in highly conserved regions of the gene (2) and in some cases appear to inactivate the growth-regulatory functions of the protein (3, 4). This has been taken as evidence that p53 is a tumor suppressor gene which when mutated, contributes to the malignant progression of the tumor. The frequency of p53 mutations in primary breast carcinomas has been directly de termined in only a limited number of cases (5-9). However, using immunohistochemical detection of abnormal p53 protein in primary breast tumors (7, 10, 11) it has been inferred that about 50% of the tumors contain a p53 mutation. Similarly, examination of primary breast tumor DNAs has revealed LOH4 on the short arm of chromosome 17, the location of the p53 gene, in approximately 50% of cases (6, 12-15). In the present study we have undertaken a comprehensive molecular analysis to directly determine the frequency of genetic abnormalities affecting the p53 locus in 26 primary breast tumors as well as to determine how this is related to LOH on chromosome 17p. Received 9/3/91; accepted 9/27/91. The costs of publication of this article »eredefrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1This work was supported in part by a grant from the Associazione Italiana Ricerca sul Cancro. 2 Present address: Department of Clinical Oncology. Addenbrookes's Hospital. Cambridge. CB2 2QQ, England. 3To whom requests for reprints should be addressed, at Simmons Cancer Center. University of Texas Southwestern Medical Center. 5323 Harry Hiñes Blvd.. Dallas. TX 75235-8590. 4 The abbreviations used are : LOH. loss of hctero/ygosity; RFLP. restriction fragment length polymorphism: cDNA. complementary DNA; PC'R. polymerase chain reaction; SSCP. single strand conformation polymorphism; ORF. open reading frame. Materials and Methods Sample Acquisition and Preparation. Infiltrating ductal carcinomas were obtained from 27 patients at the S. Giovanni Vecchio Hospital (Turin, Italy) who had not undergone treatment prior to surgery. Macroscopically normal mammary tissue was manually removed, and the tumor tissue was quickly frozen and embedded in OCT compound (Miles Scientific, Kankakee, IL) for intraoperative diagnosis. Areas of the specimen with a predominant neoplastic component were collected and stored at —¿70°C for further analysis. Lymphocytes from the same patients were isolated from 20 ml heparinized blood by density gradient centrifugation on LSM Ficoll medium (Organon Teknica Co.. Durham, NC) according to the manufacturer's instructions. DNA and RNA Preparation. High-molecular-weight DNA was pre pared from frozen tissues and lymphocytes as described (15). After ethanol precipitation, the DNA samples were dissolved in IO mM Tris/ 1 mM EDTA (pH 7.4) and stored at -20°C. Total RNA was extracted as described (16), redissolved in 10 mM Tris buffer. pH 7.5, and stored precipitated at —¿70°C. DNA Probes and Analysis of Genomic DNA and RNA. The following probes were used: pYNZ22.1 (chromosome 17pl3.3, D17S5, ATCC 57575; Ref. 17) was obtained from the American Type Culture Collec tion (Rockville, MD) and identifies a variable number tandem repeat RFLP in BamHl- and Piil-digested human genomic DNA; and pBHP53 (18), a genomic p53 fragment, identifies a BamHl RFLP within the p53 gene. The p53 locus was analyzed by two independent methods: (a) Southern blot analysis using the pBHP53 DNA probe and (h) PCR-based amplification of a 250-base pair genomic fragment containing the Tha\ RFLP at codon 72, as described (19). The DNA probes were labeled with [<;P]dCTP (Amersham, Arlington Heights, IL) using random primers (Stratagene, La Jolla, CA). Genomic DNAs (10 Mgeach) were digested to completion with the restriction enzyme of choice (Gibco-BRL, Gaithersburg. MD), subjected to electrophoresis in lcc agarose gels, and transferred to GenatranJ5 Nylon membranes (Plasco, Woburn. MA). Hybridizatio" and autoradiography were car ried out as previously described ( 15). Membranes were stripped at 65°C and rehybridized with control probes (pDFl.8-P.s7l, chr. Iq; Erb-A2, chr. 3p) to confirm equivalent DNA loading. Northern blot analysis and RNase protection assays were performed on total RNA samples as previously described (20), using three overlapping probes which covered the entire coding region of p53. SSCP Analysis of Genomic and cDN A. The PCR/SSCP method (21 ) was modified to screen for point mutations in the p53 gene (22). Genomic DNA/PCR fragments (438 base pairs spanning exons 5 and 6, or 670 base pairs containing exons 7 and 8) and cDNA/PCR fragments spanning exons 4 to 7 (codons 116 to 242) or exons 7 to 9 (codons 242 to 327) were amplified using 100 ng of genomic DNA or 250 ng of random-primed cDNA as templates and 0.5 n\ of [-"PJdCTP (Amersham) in 10-¿il reaction volumes. To localize the possible point mutations to a specific exon, 1 /jl of the PCR product was digested with Aatl (USB Corp., Cleveland. OH) for the exon 5/6 fragment or Dra\ (Bethesda Research Laboratory. Gaithersburg. MD) for the exon 7/8 fragment, and .4/n'NI (New England Biolabs, Beverly, MA) for the cDNA/PCR product. The reaction was diluted 1:5 with loading buffer (95% formamide. 2 mM EDTA. pH 8.3). Two ^1 of each diluted sample were denatured (90°Cfor 5 min) and loaded onto a 6rr nondenaturing acrylamide gel in 89 mM Tris-borate, 2 mM EDTA. pH 8.3. and electrophoresed for 5 h at 4°Cat 25 \V. The gels were subsequently dried and autoradiographed. 6194 on June 29, 2015. © 1991 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from