Analysis of masked mutations in familial adenomatous polyposis

Abstract

Familial adenomatous polyposis (FAP) is an autosomal-dominant disease characterized by the development of hundreds of adenomatous polyps of the colorectum. Approximately 80% of FAP patients can be shown to have truncating mutations of the APC gene. To determine the cause of FAP in the other 20% of patients, MAMA (monoallelic mutation analysis) was used to independently examine the status of each of the two APC alleles. Seven of nine patients analyzed were found to have significantly reduced expression from one of their two alleles whereas two patients were found to have full-length expression from both alleles. We conclude that more than 95% of patients with FAP have inactivating mutations in APC and that a combination of MAMA and standard genetic tests will identify APC abnormalities in the vast majority of such patients. That no APC expression from the mutant allele is found in some FAP patients argues strongly against the requirement for dominant negative effects of APC mutations. The results also suggest that there may be at least one additional gene, besides APC, that can give rise to FAP.

Figure 1

Schematic of MAMA. The hamster cell line, UCW-56, is fused to an FAP lymphoblastoid line, and clones are subsequently selected at 39°C. Only clones that retain human chromosome 5 can grow at this temperature. During the expansion process, hybrids retain at least one human chromosome 5 and usually lose the other copy. Clones then are genotyped to determine which chromosome 5 they contain. Proteins from the fusion clones are used for Western blotting with antibodies reactive with the amino or carboxyl ends of the APC protein. The N-terminal antibody (Human/Rodent APC) reacts with human and rodent APC protein and serves as a loading control. The C-terminal antibody (Human APC) reacts against human but not hamster APC and is used to determine whether full-length APC expression occurs.

Figure 2

DNA typing of MAMA clones. Primers found to be polymorphic in the germ line of the patient are used to genotype the hybrids. Hybrids found to be monoallelic are propagated and genotyped. Shown are three such examples from three separate patients. In each case the patient is polymorphic and each clone is monoallelic for chromosome “A” or “B.” An overexposure of the autoradiograph is made to confirm that no residual chromosomes containing the “lost” allele remain in the hybrids.

Figure 3

Protein analysis of fusion clones. The Western blots in A were derived from a patient who had a truncating mutation of APC detected with the IVSP assay. Clones 1, 2, and 4 retained allele A and expressed full-length APC whereas clones 3 and 5 retained allele “B” and did not express full-length APC. The blots in B–D were derived from patients in whom no truncating proteins could be detected with IVSP assays. The blots in B were derived from patient 4, who had no full-length APC expression from allele “A” (clones 2 and 4), but who expressed full-length protein from allele “B” (clones 1, 3, and 5). The blots in C were from patient 7, who had reduced expression of APC protein from all fusion clones containing allele “A” compared with those containing allele “B.” The blots in D were derived from patient 9, who had comparable amounts of full-length APC protein expressed from clones retaining either allele “A” or “B.” The asterisk (∗) represents a cross-reactive band present in all hamster cells that is detected with the C-terminal antibody.