TaqMan-nuclease assays are widely used for the qualitative detection of single nucleotide polymorphisms (SNPs) and the determination of biallelic states in pooled or heterozygous DNA samples. These assays are highly specific, reproducible, and suitable for high-throughput approaches. A crucial limitation of this method, and others, is the detection of minor allele frequencies with detection limits of generally 3% to 9% for minor allele contributions. Here we describe the combination of customized TaqMan-nuclease assay and allele-specific restriction to increase the sensitivity of this method, allowing the qualitative detection of allele contributions as low as 0.05%.
Single nucleotide polymorphisms (SNPs) are estimated to occur at a frequency of one SNP per 1–2 kb (1). Due to this high abundance throughout the genome, SNPs find a wide range of applications as molecular markers in, for example, forensics (2), medical diagnostics (3), and molecular epidemiology (4). A wide range of techniques have emerged to score SNPs qualitatively or quantitatively (5) including TaqMan-nuclease assay (6), molecular beacon assay (7), fluorescent-labeled locked nucleic acids (LNA) (8), pyrosequencing (9), multiplex quencher extension (10), and polymorphism ratio sequencing (11). Although these methods allow the detection of alleles in pooled or heterozygous samples if present at frequencies of 3%-9%, the sensitivity of these methods is limited and often insufficient for many applications. In disease-association studies in which allele contributions in whole-tissues of <2% can be crucial (12,13) or in biological studies where the contribution of minor alleles in heteroplasmic mitochondrial DNA samples can be as low as 0.01% (14), such allele frequencies cannot be detected without a tradeoff in the experimental design and sample throughput (9,10,11).
TaqMan-nuclease assays are well established and commonly used for SNP detection and genotyping studies (15,16). Using this methodology, the detection of SNPs goes hand-in-hand with the amplification of a single short DNA sequence containing the SNP site. Fluorescent probes differentiating between the two alleles bind at the SNP site located between the two primer sites and are degraded during amplification by the exonuclease activity of Taq DNA polymerase, releasing a fluorescent reporter dye. The amplification of a single DNA sequence to subsequently differentiate among allelic states is convenient and allows high throughput, but the downside of this design is the use of the same primer set for the amplification of both alleles. The use of just one primer pair leads automatically to a competing amplification between these alleles. If one allele is present at a lower frequency than the other, the amplification is likely to be dominated by the allele at the higher frequency. Thus, the amplification of the second allele with lower copy number is suppressed and often not detectable.
To minimize this effect, we combined allele-specific restriction with the TaqMan-nuclease assay (Applied Biosystems, Foster City, CA, USA), in order to suppress the amplification of the dominant allele and to allow the qualitative detection of small allele contributions. To evaluate the power of this combined approach, we compared the detection limit for alleles occurring at low frequencies of the conventional TaqMan-nuclease assay to the detection limit of the combined TaqMan/restriction approach.
The investigated SNP was a G/A polymorphism (in which A = wild-type and G = mutant) in the salmon ND1 gene at nucleotide position 3957 of the mitochondrial genome (GenBank accession no. NC_002980). This SNP has been detected in previous work within individuals of a hatchery population of chinook salmon (data not published).
To evaluate the sensitivity of both approaches, we extracted DNA from fin clips of 20 salmon using the DNeasy Blood & Tissue kit (Qiagen GmbH, Hilden, Germany). We amplified and sequenced a 314-bp fragment of the mitochondrial genome of these samples using the primers ND1aF and ND1R1 to identify individuals harboring either the mutant (G) or the wild-type version (A) of the SNP (for primer sequences see (Table 1)). DNA purity was checked using a ND-1000 NanoDrop (NanoDrop Technologies, Wilmington, DE, USA), and 2-fold serial dilutions were prepared for mutant DNA samples (starting concentration: 4.5 ng/µL, 13 dilution steps), aliquoted, and stored at −20°C. As a measure of accuracy, we derived a standard curve from realtime PCR amplifications applying the TaqMan-nuclease assay on this dilution series. The amplification was performed with an efficiency of 98.5% (data not shown), confirming the accurate dilution of template DNA, which was then used for DNA pool construction. Mutant/wild-type DNA pools were prepared by adding the 2-fold dilution series of mutant DNA to a constant remaining DNA solution of 9 ng wild-type DNA, resulting in ratios of 1:2 to 1:8192.Table 1. Primer and Probe Sequences
The restriction enzyme capable of cutting the SNP in an allele-specific manner (AvaII, motif: GGWCC, with W = A or T) was identified using NEBcutter V2.0 (New England Biolabs, Ipswich, MA, USA). Digests were performed for 2 h at 37°C using 1 U enzyme/1 µg DNA, followed by 20 min heat inactivation at 65°C. Digests were applied on both wild-type and mutant DNA samples and were performed prior to the preparation of dilution series and pool construction.