An accurate method for comparing transcript levels of two alleles or highly homologous genes: application to fibrillin transcripts in Marfan patients' fibroblasts.

We introduce here a novel and generally applicable, solid-phase minisequencing-based approach for rapid estimation of relative levels of transcripts with high sequence homology. This study was undertaken to screen for the consequences of different fibrillin-1 mutations on the transcript levels in patients with the Marfan syndrome (MFS). This dominantly inherited, connective tissue disorder is characterized by pleiotrophic symptoms in cardiovascular, skeletal, and ocular systems. A spectrum of disease mutations in the gene encoding fibrillin-1 (FBN1), a glycoprotein component of extracellular matrix microfibrils, has been identified in MFS patients, but the mechanisms by which mutations result in different phenotypic manifestations are still unknown to a large extent. Our data from the quantitation of FBN1 transcripts provide support for the hypothesis that mutations causing premature stop codons result in a milder phenotype than classical MFS by reducing the stability of the mutant transcript and, consequently, decreasing the interference of mutant polypeptide in the formation of fibrillin fibers. We also applied this mRNA quantitation method to determine the relative ratio between transcripts from the genes coding for two highly homologous microfibrillar components, FBN1 and FBN2, in control fibroblast cultures as well as in fibroblasts from MFS patients. Interestingly, these data show large variations between the levels of the two transcripts in fibroblast cultures, but these variations do not correlate either with the nature of the disease mutation or to the clinical MFS phenotype.