Genotyping of frequent BRCA1/2 SNPs with unlabeled probes: a supplement to HRMCA mutation scanning, allowing the strong reduction of sequencing burden

Abstract

We previously validated mutation scanning for BRCA1 and 2 using high-resolution melting curve analysis (HRMCA). Due to recurrent single nucleotide polymorphisms (SNPs), a considerable amount of sequencing work remains after HRMCA, as melting curves for SNPs and deleterious mutations may be similar. Here, we present a simple approach for the optimization of SNP genotyping with HRMCA using unlabeled probes. Protocols were optimized for 14 frequent SNPs in BRCA1 and 2. Two probes contained an additional mismatch to detect a rare polymorphism a few nucleotides upstream. PCR was performed in the presence of LCgreenPlus and analyzed on a Lightscanner. Genotyping assays were optimized with five wild-type, heterozygous, and homozygous mutant samples. Sensitivity and specificity of the assays were evaluated with a blind screening of 95 samples. All unlabeled probes correctly genotyped the SNPs. A 1:5 asymmetric primer ratio produced sufficient probe-strand duplexes to accurately genotype the SNP of interest. The most important parameter to optimize was the number of PCR cycles. By complementing our BRCA1/2 HRMCA with 14 unlabeled probe assays, we reduced the sequencing burden by three-fold. Our simple approach for optimization can be used as a blueprint to design genotyping assays for other genes. This is one of the largest studies reported to date and the first that presents an approach combining genotyping and mutation scanning of two large polymorphic genes.

Figure 1

Melting profiles of an unlabeled probe assay allowing the simultaneous identification of five distinct alleles from two BRCA1 SNPs, separated by a few bp (BRCA1 c.3113 A>G and c.3119G>A). Samples were analyzed with amplicon scanning and unlabeled probe genotyping. With the expert scanning module (left panel) only three distinct clusters can be detected. Samples with alleles heterozygous for only one of both SNPs generate similar melting curves. Furthermore, samples with alleles homozygous for rs16941 (BRCA1 c.3113 A>G) cannot be distinguished from wild-type samples. Genotyping of these samples with our unlabeled probe assay allowed simultaneous identification of all five distinct alleles by melting analysis (right panel).