Application of a Thermodynamic Nearest-Neighbor Model to Estimate Nucleic Acid Stability and Optimize Probe Design: Prediction of Melting Points of Multiple Mutations of Apolipoprotein B-3500 and Factor V with a Hybridization Probe Genotyping Assay on the LightCycler

PCR-based mutation detection is prone to methodological errors, e.g., in restriction length fragment polymorphism (RFLP) and allele-specific amplification (ASA), false PCR results may occur because of technical faults or atypical new mutations.We investigated the ability of a genotyping assay based on hybridization of labeled oligonucleotides to detect and discriminate known and as yet unknown mutations in the factor V and apolipoprotein B-100 genes. Expected melting points were calculated using a nearest-neighbor model for nucleic acid duplex stability and compared with experimental findings derived from LightCycler melting curves. A method for genotyping the apolipoprotein B-100 G10699A and C10698T mutations is presented.All mismatches tested for in the probed sequence could be detected with a single probe. The measured melting points were in good agreement with their values predicted using the nearest-neighbor model (r = 0.96; y = 0.98x + 1.18; S(y|x) = 0.96; n = 24).This procedure not only allows the identification of the mutation of interest but also enables the discrimination from other potential mutations in the vicinity of the former. The nearest-neighbor model is valid for hybridization probe assays on the LightCycler and should be of general value in setting up such assays. We have shown for two clinically relevant genotyping examples that the LightCycler mutation detection system has superior discriminatory performance compared with conventional RFLP or ASA PCR-based methods for molecular diagnostic purposes. With this method, in every hybridization probe assay, all mutations under a properly designed probe should be detectable, but they will not necessarily be discriminated from each other in all cases.