Development of a Premature Stop Codon-detection method based on a bacterial two-hybrid system

BACKGROUND The detection of Premature Stop Codons (PSCs) in human genes is very useful for the genetic diagnosis of different hereditary cancers, e.g. Familial Breast Cancer and Hereditary Non-Polyposis Colorectal Cancer (HNPCC). The products of these PSCs are truncated proteins, detectable in vitro by the Protein Truncation Test and in vivo by using the living translation machinery of yeast or bacteria. These living strategies are based on the construction of recombinant plasmids where the human sequence of interest is inserted upstream of a reporter gene. Although simple, these assays have their limitations. The yeast system requires extensive work to enhance its specificity, and the bacterial systems yield many false results due to translation re-initiation events occurring post PSCs. Our aim was to design a recombinant plasmid useful for detecting PSCs in human genes and resistant to bacterial translation re-initiation interferences. RESULTS A functional recombinant plasmid (pREAL) was designed based on a bacterial two-hybrid system. In our design, the in vivo translation of fused fragments of the Bordetella pertussis adenylate cyclase triggers the production of cAMP giving rise to a selectable bacterial phenotype. When a gene of interest is inserted between the two fragments, any PSC inhibits the enzymatic activity of the product, and translation re-initiation events post-PSC yield separated inactive fragments. We demonstrated that the system can accurately detect PSCs in human genes by inserting mutated fragments of the brca1 and msh2 gene. Western Blot assays revealed translation re-initiation events in all the tested colonies, implying that a simpler plasmid would not be resistant to this source of false negative results. The application of the system to a HNPCC family with a nonsense mutation in the msh2 gene correctly diagnosed wild type homozygous and heterozygous patients. CONCLUSION The developed pREAL is applicable to the detection of PSCs in human genes related to different diseases and is resistant to translation re-initiation events. The diagnosis steps are easy, have a low cost, detect only pathologic mutations, and allow the analysis of separated alleles.