Detection of single-nucleotide polymorphism.

Identification of polymorphism is the key for genetic mapping, diagnosis and marker-assisted selection. Restriction fragment-length polymorphisms (RFLPs) and short tandem repeats (STRs), also known as microsatellites, are commonly used genetic markers. Until discovery of STRs, RFLPs were used for genetic mapping, but bi-allelic polymorphism of RFLP limited its use as a marker. Although STRs are more informative because they are multi-allelic, they occur on average once per 42 kb of canine genome (6). The need for more closely spaced polymorphic markers has been fulfilled by detection of single-nucleotide polymorphisms (SNPs), which occur more frequently in the mammalian genome. In domestic and agricultural animals, where inbreeding is a common practice, quite often a known marker in a candidate gene might not be informative in a pedigree available for study of the disease. In such a situation, identification of a new marker, informative within the pedigree and associated with the gene of interest, is required. This example underscores one among many other needs for detection of SNPs. Identification of SNPs can be done by different methods (2). The most common among these are single-strand conformation polymorphism (SSCP; Reference 5), heteroduplex analysis (3,4) and direct sequencing of DNA (1). SSCP and heteroduplex analysis are relatively time-consuming and cumbersome experiments. On the other hand, even if it is done using automated machines, sequencing is relatively more expensive unless a region of the test DNA fragment is first identified to contain heterogeneity by one of the first two methods. The classical polymerase chain reaction (PCR)-RFLP technique involves: (i) amplification of the target sequence in the template DNA from multiple samples by PCR using specific primers, (ii) digestion of the PCR products with different restriction enzymes and (iii) separation of the fragments by agarose or polyacrylamide gel electrophoresis (PAGE), depending on the size of the DNA fragments in the digest. In conjunction with DNA sequencing and heteroduplex analysis, we use the PCR-RFLP technique routinely in our laboratory for searching gene mutations and polymorphisms. To maximize the likelihood of detecting RFLP in the test pedigrees, we use a large number of restriction enzymes to digest target sequences. However, because detection of an SNP by a restriction enzyme would be determined by specificity of a minimum of 3 neighboring nucleotide sequences (because a restriction site consists of a sequence at least 4 nucleotides long), it is expected that a large number of SNPs would remain undetected by this strategy. We report a simple modification of the PCR-RFLP protocol to identify SNPs that do not alter the restriction site for the enzyme chosen arbitrarily in search of an RFLP. We typically amplify genomic DNA fragments (1–5 kb) by PCR, digest amplified DNA with 15–20 restriction enzymes and electrophorese the DNA digest in agarose and polyacrylamide gels to efficiently separate the entire range of the DNA fragments from 50 bp to >1 kb and facilitate comparison and easy identification of any RFLP. To identify SNPs, we have made two modifications in our protocol that we have termed “heat and run”. First we heat the DNA digest at 99°C for 10 min and cool it slowly to room temperature (10–20 min); then we electrophorese the DNA digest in two 6% nondenaturing polyacrylamide gels (19:1 acrylamide/bisacrylamide) at 300 V (constant), one for 1.5–2.0 h and the other for >2 h. The DNA bands in the gels are visualized by staining with ethidium bromide (0.5 μg/mL). We have observed that these 2 steps quite often detect the heteroduplexes present in DNA fragments ranging in size from 150 to 400 bp. Heteroduplexes are formed between 2 complementary DNA strands when a mismatch occurs between them because of the presence of noncomplementary nucleotides in an identical location or insertion of one or more nucleotide(s) in one of the 2 strands. We observed that heat denaturation and slow cooling of the DNA samples increase the formation of the heteroduplex, and a longer run in PAGE separates the heteroduplex from a homoduplex more efficiently. The initial short run in PAGE identifies any RFLP and occasionally heteroduplexes in smaller (<300 bp) DNA fragments. If the short run reveals any polymorphism, a longer run is not necessary. To identify an allelic difference in the canine apolipoprotein H (APOH) gene among 3 individuals in a canine pedigree, we amplified by PCR a 1.6-kb DNA fragment of the gene (GenBank Accession No. X72933) from the genomic DNA templates using a forward primer (5′-GAATCTTAGAAAATGGAGCTGTACGCT-3′) and a reverse primer (5′-CATTTTCCTTCCTCGGTGCAT-3′). The amplification was done at 1.5 mM MgCl2 concentration with 35 cycles at 94°C for 30 s, 60°C for 30 s and 72°C for 2.5 min. Amplified DNA fragments were digested with 17 different restriction enzymes (AvaI, BsmAI, BsrI, BstNI, FokI, HaeIII, HhaI, HinfI,