A Mismatch EndoNuclease Array-Based Methodology (MENA) for Identifying Known SNPs or Novel Point Mutations

Abstract

Accurate and rapid identification or confirmation of single nucleotide polymorphisms (SNPs), point mutations and other human genomic variation facilitates understanding the genetic basis of disease. We have developed a new methodology (called MENA (Mismatch EndoNuclease Array)) pairing DNA mismatch endonuclease enzymology with tiling microarray hybridization in order to genotype both known point mutations (such as SNPs) as well as identify previously undiscovered point mutations and small indels. We show that our assay can rapidly genotype known SNPs in a human genomic DNA sample with 99% accuracy, in addition to identifying novel point mutations and small indels with a false discovery rate as low as 10%. Our technology provides a platform for a variety of applications, including: (1) genotyping known SNPs as well as confirming newly discovered SNPs from whole genome sequencing analyses; (2) identifying novel point mutations and indels in any genomic region from any organism for which genome sequence information is available; and (3) screening panels of genes associated with particular diseases and disorders in patient samples to identify causative mutations. As a proof of principle for using MENA to discover novel mutations, we report identification of a novel allele of the beethoven (btv) gene in Drosophila, which encodes a ciliary cytoplasmic dynein motor protein important for auditory mechanosensation.

Keywords: SNP detection; disease mutation; endonuclease; genetic variation; microarray; mismatch.

Figure 1

General Mismatch EndoNuclease Array (MENA) strategy. Red stars indicate probes with mismatches to the hybridized DNA. Lightning bolts represent mismatches where the endonuclease cleaves the duplex DNA.

Figure 2

Relative signal change (RSC) between pre- and post-treatment arrays with Surveyor. Nucleotides (X-axis) indicate the nucleotides interrogated by the probes (not the nucleotides in the probe). In the sample analyzed, rs9451298 and rs9359876 have “T” and “C”, respectively (forward strand). See text for a detailed description of the RSC measure. Red arrows indicate identity of the SNP.

Figure 3

Summary of genotyping accuracy. Genotyping accuracy of MENA under different experimental conditions. Results shown after combining all homozygous SNPs for each condition. See Table 1 for detailed experimental conditions.

Figure 4

Genotyping accuracy with different number of probes. Percentage of genotyping accuracy based different number of probes interrogating a given SNP. Results shown after combining all homozygous SNPs under experimental conditions 405722 (see Table 1 and Figure 3). Because probes are 60 nucleotides long, analyses based on probes tiled every three nucleotides use the combined information of 20 probes, analyses based on probes tiled every six nucleotides use information of 10 probes, etc.

Figure 5

Comparison of signal change (SC) using MENA between mutant and control samples. SC indicates the reduction in signal after Surveyor treatment in controls and btv² samples. In red, probes showing the strongest (1%) reduction in signal in mutant btv² sample relative to controls and thus putative genetic variants. In black, the probe corresponding to a single bp deletion in exon 22 of the btv gene that results in a frameshift that causes early stop codons. In blue, all remaining probes that were interrogated with MENA. SC shown in Log₂ units.

Figure 6

Identification of a novel allele of the beethoven (btv) gene in Drosophila. Results of PCR and Sanger sequencing around the candidate mutation detected by MENA confirm a single bp (“A”) deletion in btv² relative to a wild type (WT) sequence (A); this deletion maps to exon 22 of btv (red rectangle in B; red arrow in C), which is located in chromosome arm 2L (position 17966613-17966617); and (D) the consequences of the “A” deletion causing a frameshift and early stop codons. Green letters represent the altered amino acids that are translated as a result of the frameshift, which ends with a stop codon indicated by the red x.