P53 abnormalities in primary prostate cancer: single-strand conformation polymorphism analysis of complementary DNA in comparison with genomic DNA. The Cooperative Prostate Network.

Abstract

The reported frequency of mutation of the p53 tumor suppressor gene (also known as TP53) in human carcinomas of the prostate has varied widely, ranging from 3% to 42%. This variability may be a consequence of tumor heterogeneity and/or the use of different methods of analysis. Since p53 mutation has been associated with clinical outcome for a number of cancer types, determination of its true frequency in primary carcinomas of the prostate is important. The principal aims of this study were as follows: 1) to validate the utility of detecting p53 gene mutations by means of polymerase chain reaction-single-strand conformation polymorphism (PCR-SSCP) analysis of complementary DNA (cDNA) (synthesized from prostate tissue RNA and 2) to study the concordance of RNA- and DNA-based PCR-SSCP assays in detecting p53 mutations in individual tumor fragments. RNA and genomic DNA were isolated by means of standard techniques from specimens of 19 carcinomas of the prostate, selected on the basis of p53 data obtained in a previous analysis of cDNA (indicating that 14 were mutant and five were wild-type). RNA was converted into cDNA by means of reverse transcription (RT); the cDNA was then amplified by means of nonisotopic (i.e., nonradioactive) PCR, and the PCR products were subjected to SSCP analysis in polyacrylamide gels (RT-PCR-SSCP analysis). Genomic DNA was examined by means of SSCP analysis of isotopically labeled (32PO4) PCR products (DNA-PCR-SSCP analysis). In both approaches, the protein coding region of the p53 gene was divided into multiple, smaller fragments for study. PCR products exhibiting abnormal migration in SSCP gels were subjected to direct nucleotide sequencing or to cloning and sequencing of multiple clones. RT-PCR-SSCP and DNA-PCR-SSCP identified p53 gene abnormalities in 15 of the 19 selected carcinomas, including one previously reported to be wild-type for p53. Overall, PCR-SSCP analysis identified 18 p53 fragments with abnormalities; three carcinomas showed two abnormalities each. Six (33%) of the 18 abnormalities were detected by both RT-PCR-SSCP and DNA-PCR-SSCP, 10 (56%) were detected by RT-PCR-SSCP alone, and two (11%) were detected by DNA-PCR-SSCP alone. The 18 abnormalities were caused by 20 changes in the sequence of the p53 gene; in one carcinoma, double mutations in two individual p53 exons were identified. PCR-SSCP analysis of both RNA and DNA allows the detection of more mutations than the analysis of either alone. Some primary carcinomas of the prostate contain more than one altered p53 gene, consistent with the possibility of intratumoral heterogeneity of mutation of this gene. For comprehensive analysis of p53 mutations in carcinomas of the prostate, and perhaps in other tumor tissues, SSCP analysis of cDNA should be used in combination with SSCP analysis of genomic DNA.