Genotypic mutation analysis by RFLP/PCR.

In most cases somatic mutations in disease-related genes do not give rise to a functional change of the mutated cell which would allow its isolation or expansion in vitro. Therefore, selection of mutated cells on the basis of an altered phenotype has to be replaced by biochemical separation and detection of the altered sequence of the gene of interest. In the RFLP/PCR (RFLP, restriction fragment length polymorphism) approach of 'genotypic' mutation analysis base pair changes, small deletions and insertions are measured which are located in a restriction enzyme recognition sequence and render this site resistant to cleavage by the corresponding endonuclease. The resistant DNA sequence containing the mutated site is amplified by PCR only after wild-type DNA has been eliminated by restriction digestion. Amplified DNA is directly sequenced or cloned into lambda gt10 and mutants are quantitated by oligonucleotide plaque hybridization. Absolute mutation frequencies are estimated relative to an internal 'mutant standard'. The RFLP/PCR protocol has been developed with mixtures of plasmid constructs containing wild-type inserts of the human c-H-ras1 protooncogene or inserts with base pair changes in an MspI site, a PvuII site and a TaqI site of this gene. As few as 1-5 mutated copies could be rescued from 10(7)-10(9) copies of the corresponding wild-type sequence. The RFLP/PCR protocol was successfully applied to the determination of N-ethyl-N-nitrosourea-induced mutations in codon 12 of c-H-ras1 (MspI site 1695-1698) and codon 248 of the p53 tumor suppressor gene (MspI site 14067-14070) in human skin fibroblasts. The RFLP/PCR approach holds promise for molecular toxicology and epidemiology.