Background: Reliable PCR amplification of DNA fragments is the prerequisite for most genetic assays. We investigated the impact of G-quadruplex- or i-motif-like sequences on the reliability of PCR-based genetic analyses.
Methods: We found the sequence context of a common intronic polymorphism in the MEN1 gene (multiple endocrine neoplasia I) to be the cause of systematic genotyping errors by inducing preferential amplification of one allelic variant [allele dropout (ADO)]. Bioinformatic analyses and pyrosequencing-based allele quantification enabled the identification of the underlying DNA structures.
Results: We showed that G-quadruplex- or i-motif-like sequences can reproducibly cause ADO. In these cases, amplification efficiency strongly depends on the PCR enzyme and buffer conditions, the magnesium concentration in particular. In a randomly chosen subset of candidate single-nucleotide polymorphisms (SNPs) defined by properties deduced from 2 originally identified ADO cases, we confirmed preferential PCR amplification in up to 50% of the SNPs. We subsequently identified G-quadruplex and i-motifs harboring a SNP that alters the typical motif as the cause of this phenomenon, and a genomewide search based on the respective motifs predicted 0.5% of all SNPs listed by dbSNP and Online Mendelian Inheritance in Man to be potentially affected.
Conclusions: Undetected, the described phenomenon produces systematic errors in genetic analyses that may lead to misdiagnoses in clinical settings. PCR products should be checked for G-quadruplex and i-motifs to avoid the formation of ADO-causing secondary structures. Truly affected assays can then be identified by a simple experimental procedure, which simultaneously provides the solution to the problem.