Mutation detection using Surveyor nuclease

Mutation detection using Surveyor nuclease Peter Qiu, Harini Shandilya, James M. D Alessio, Kevin O Connor, Jeffrey Durocher, and Gary F. Gerard BioTechniques 36:702-707 (April 2004) We have developed a simple and flexible mutation detection technology for the discovery and mapping of both known and unknown mutations. This technology is based on a new mismatch-specific DNA endonuclease from celery, Surveyor nuclease, which is a member of the CEL nuclease family of plant DNA endonucleases. Surveyor nuclease cleaves with high specificity at the 3 side of any mismatch site in both DNA strands, including all base substitutions and insertion/deletions up to at least 12 nucleotides. Surveyor nuclease technology involves four steps: (i) PCR to amplify target DNA from both mutant and wild-type reference DNA; (ii) hybridization to form heteroduplexes between mutant and wild-type reference DNA; (iii) treatment of annealed DNA with Surveyor nuclease to cleave heteroduplexes; and (iv) analysis of digested DNA products using the detection/separation platform of choice. The technology is highly sensitive, detecting rare mutants present at as low as 1 in 32 copies. Unlabeled Surveyor nuclease digestion products can be analyzed using conventional gel electrophoresis or high-performance liquid chromatography (HPLC), while end-labeled digestion products are suitable for analysis by automated gel or capillary electrophoresis. The entire protocol can be performed in less than a day and is suitable for automated and high-throughput procedures. INTRODUCTION The discovery of unknown mutations and polymorphisms in genome sequences is a vital aspect of widespread research areas, including pharmacogenetics in humans [linking single nucleotide polymorphism (SNP) genotype and haplotype information to disease drug treatment] and reverse genetics in model organisms (relating mutant phenotype to genomic DNA sequence information). The method of choice for precise mutation and polymorphism analysis has been DNA sequencing. However, this approach generates large quantities of unnecessary data. Another widely used method for screening DNA mismatches, PCR single-stranded conformational polymorphism (SSCP) analysis, has recently been made much more sensitive by using fluorescence detection and capillary electrophoresis for analyses (1). Denaturing highperformance liquid chromatography (DHPLC) is another sensitive mutation screening method that is used extensively (2). Unfortunately, both of these latter methods are generally limited to the analysis of DNA fragments less than 1000 bp in length, cannot be used to map multiple mutations in a single fragment, and do not easily yield the location of a mutation within a DNA sequence. Recently, a new family of neutral ph DNA endonucleases from plants has been described (3). The enzyme isolated from celery, CEL I nuclease, has been shown to cut DNA with high specificity in one of two strands at the 3 side of basesubstitution mismatches and DNA distortions (3 5). CEL I nuclease has been used to accurately detect a variety of mutations and polymorphisms in the human BRCA1 gene (3 5). It has also been used to perform high-throughput screening of induced point mutations in Arabidopsis (6,7) and Lotus (8). These applications rely on fluorophore end-labeled DNA substrates coupled with fractionation platforms that separate denatured DNA cleavage products. CEL I nuclease has also been used to scan large regions of bacterial genomic DNA for mutations and polymorphisms by cleaving DNA at the 3 side of a mismatch site in both DNA strands (9). The double-stranded cleavage products were fractionated by agarose gel electrophoresis and detected by Southern hybridization with labeled DNA (9). Using an enzyme from the CEL nuclease family, Surveyor nuclease (Transgenomic, Gaithersburg, MD, USA), we have extended the method of mutation discovery that exploits the enzyme s ability to cut both DNA strands at a mismatch site to the analysis of digestion products on several separation/detection platforms. Surveyor nuclease mutation detection technology is a simple, sensitive, and reliable alternative to other screening methods. Here we show with substrates containing known mutations that Surveyor nuclease cleaves both DNA strands at sites of base substitution or insertion/deletion, recognizes and cleaves at multiple mutations in large DNA fragments, and produces detectable cleavage products from mismatch DNA representing only a small proportion of the DNA in a Transgenomic, Gaithersburg, MD, USA 702 BioTechniques Vol. 36, No. 4 (2004)

population. We have also used Surveyor nuclease to identify unknown mutations in cdna, plasmid DNA, and genomic DNA (P. Qiu and G.F. Gerard, manuscript submitted). MATERIALS AND METHODS DNA Substrates Plasmid pqis155 contains a derivative of the gene for CEL I (5), cloned between the XhoI and NdeI sites of pet22b (EMD Biosciences, Madison, WI, USA). Using a Gene- Tailor Site-Directed Mutagenesis System (Invitrogen, Carlsbad, CA, USA), the C at position 605, called the target site, in the CEL I gene of pqis155 was changed to A, G, and T, or insertions of 1, 2, 3, 6, 9, and 12 nucleotides were made at the site. Plasmid DNA was transformed into Escherichia coli DH5α (Invitrogen), and cells were grown in LB medium plus 100 μg/ml ampicillin. Plasmid DNA was isolated using a QIAGEN Plasmid Mini Kit (Qiagen, Valencia, CA, USA). Plasmids were named based on their sequence at the target site (e.g., pqis155g has a G at the target site). A base substitution (C to G) at position 189 in the CEL I gene was introduced into pqis155c to generate plasmid pqis190. A 632-bp fragment was amplified from pqis155 plasmid DNA and its derivatives utilizing the following primers (all from Invitrogen): pcelr (5 -CGCCAAAGAATGATCT- GCGGAGCTT-3 ) and pcel190f (5 -ACACCTGAT- CAAGCCTGTTCATTTGATTAC-3 ). Fluorophore end-labeled DNA fragments were prepared by performing PCR with these primers labeled at the 5 end with FAM (Applied Biosystems, Foster City, CA, USA). PCR primers placzr (5 -GGCCGCGCGAAAACGTTCTTCGGGGCGAAAACT- 3 ) and pet22r2 (5 -CAGTGGGAACGATGCCCTCATT- CAGC-3 ) were used to amplify a 2.95-kb fragment from pqis190 and pqis155g. placzr and pet22r1 (5 - ATTCACCACCCTGAATTGACTCT-3 ) were used to amplify a 2.23-kb fragment from pqis155c and pqis155g. PCR was performed with Optimase polymerase (Transgenomic) utilizing the reaction conditions recommended by the manufacturer. The amount and quality of the amplified DNA produced in a reaction were determined by the visual comparison of the DNA product to a 100-bp DNA ladder (New England Biolabs, Beverly, MA, USA) separated by agarose gel electrophoresis. The 2.95-kb amplified DNA was purified using a QIAquick PCR Purification Kit (Qiagen) before annealing. All other amplified DNA was annealed without purification. Heteroduplexes were formed by annealing equal amounts of amplified DNA that was prepared from two different plasmids. Annealing was performed in a DNA Engine PTC-200 thermal cycler (MJ Research, Waltham, MA, USA) by running the following program: 95 C for 10 min, 95 to 85 C in increments of -2 C/s, 85 to 25 C in increments of -0.1 C/s, and a 4 C hold. Control homoduplex was prepared by annealing one of a pair of amplified DNAs alone. When two different alleles are annealed in a 1:1 mixture, mismatch heteroduplexes are formed approximately 50% of the time. For each base change, two mismatches are formed. Reformed homoduplexes constitute the other 50% of the population. Surveyor Nuclease Surveyor nuclease was purified from celery by a modification of a published procedure (5). Based on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and amino acid sequence analysis of the primary protein in preparations of Surveyor nuclease (H. Shandilya and G.F. Gerard, manuscript in preparation), the Surveyor nuclease corresponds to CEL II (5). DNA Cleavage and Analysis DNA substrates were treated with Surveyor nuclease in reaction mixtures (20 μl) containing 20 mm Tris-HCl, ph 7.4, 25 mm KCl, 10 mm MgCl 2, 200 ng DNA substrate, 100 U DNA ligase, and various amounts of Surveyor nuclease. The total amount of DNA substrate was the same in all reaction mixtures. Reactions were incubated at 42 C for 20 min and terminated by the addition of 2 μl 0.5 M EDTA. Unlabeled cleavage products (20 μl) were fractionated on a 2% agarose gel cast and run in 1 TAE buffer (40 mm Tris-acetate, ph 8.3, 1 mm EDTA). The gel was stained with ethidium bromide and photographed over UV transillumination. When necessary, the bands were quantified using the EDAS imaging system and software (Eastman Kodak, Rochester, NY, USA). Unlabeled cleavage products were also separated on a WAVE System (Transgenomic) equipped with a fluorescence detector and a high sensitivity accessory for post-column DNA intercalation with fluorescent dye. The DNASep cartridge (Transgenomic) was run at 50 C under nondenaturing conditions to separate DNA fragments by size. Cleavage products generated from DNA fragments labeled at the 5 ends with FAM were fractionated using an ABI Prism 3100 Genetic Analyzer following the manufacturer s recommended procedure and GeneScan software (both from Applied Biosystems). RESULTS AND DISCUSSION Surveyor Nuclease Mutation Detection Technology The use of Surveyor nuclease to detect mutations involves four steps: (i) PCR to amplify target DNA from both mutant and wild-type reference DNA; (ii) hybridization to form heteroduplexes between mutant and wild-type reference DNA; (iii) treatment of both heteroduplex and reference homoduplex DNA with Surveyor nuclease to cleave heteroduplexes; and (iv) analysis of DNA products using any suitable separation platform. The quality of the amplified product used as a substrate for Surveyor nuclease digestion is critical to the success of the method. PCR amplification must be optimized to produce single-band quality product in adequate amounts. When the PCR product is not of sufficiently high yield or contains PCR artifacts (e.g., primer dimers or truncated products from primer misannealing), the background after Surveyor nuclease digestion can obscure the signal. Low PCR yield can result in the use of a suboptimal amount of DNA substrate relative Vol. 36, No. 4 (2004) BioTechniques 703

to the enzyme in a reaction mixture, which increases background. In the presence of PCR artifacts, the digestion pattern reflects cleavage fragments derived from the DNA artifacts. Errors introduced by the use of a low-fidelity PCR DNA polymerase, such as Taq DNA polymerase, can also significantly increase Surveyor nuclease digestion background (P. Qiu and G.F. Gerard, unpublished observation). We use a high-fidelity proofreading DNA polymerase to minimize errors and reduce digestion background. Model substrates were used to determine the reaction optima of Surveyor nuclease and to define the range of reaction conditions that produce maximum quantities of desired cleavage products for each of three separation platforms. Plasmids containing genetically engineered mutations in the CEL I gene were used to generate model substrates by annealing 632-bp DNA fragments derived from a reference plasmid and an appropriate derivative plasmid by PCR (see Materials and Methods). Surveyor nuclease digestion of the single mismatch site in heteroduplexes derived from the annealed amplified DNAs yields 217- and 415-bp products. The size separation platforms included agarose gel electrophoresis, nondenaturing HPLC, and capillary electrophoresis. In the first two methods, unlabeled DNA cleavage products were separated under nondenaturing conditions and detected with intercalated fluorescent dye, while in the third method, denatured fluorophore end-labeled DNA was separated and detected. Detection and Cleavage Efficiencies of Surveyor Nuclease at Single-Base Substitutions and Insertions Utilizing synthetic 64-bp double-stranded complementary DNA oligonucleotides containing a single-base mismatch 35 nucleotides from the 5 end of the top strand (3), we determined the single-base mismatch cutting preference of Surveyor nuclease (data not shown). The mismatch cutting preferences fall into four groups from most to least preferred: CT, AC, and CC are preferred equally over TT, followed by AA and GG, and finally followed by the least preferred, AG and GT. The mismatch cleavage preferences of Surveyor nuclease differ somewhat from those reported for CEL nuclease (3). When mutant and reference PCR products are hybridized in equal amounts, two alternate heteroduplex mismatches are formed that represent approximately 50% of the DNA in the mixture. In almost all cases, one and sometimes both of these mismatches are in the group of preferred substrates for Surveyor nuclease. Figure 1 shows that Surveyor nuclease produces a double strand cut in every mismatched pair tested in the 632-bp substrate, including GG and CC, GA and TC, GT and AC, and AA and TT, producing the expected cleavage products migrating at positions consistent with sizes of 217 and 415 bp. Homoduplex and heteroduplex DNA not cleaved by Surveyor nuclease separate as apparent full-length fragments. Similar results were obtained when unlabeled cleavage products were analyzed by WAVE and WAVE high sensitivity detection (HSD) HPLC and when labeled cleavage products were analyzed by denaturing capillary electrophoresis (data not shown). Switching the mismatched base at the top and bottom strands wherever possible gave similar cutting efficiencies. Figure 1 also shows that Surveyor nuclease cleaves efficiently at a site in 632-bp substrate containing an insertion of 1, 2, 3, 6, 9, or 12 nucleotides, which produces the expected cleavage products in each case. Visual comparison of the amounts of products generated by Surveyor nuclease (Figure 1) indicates that the enzyme cleaves insertion/deletions more efficiently than single-base substitution mismatches. Quantification of the bands in the gel (see Materials and Methods) Figure 1. Agarose gel electrophoresis of Surveyor nuclease digestion products of 632-bp DNA containing a variety of single-base mismatches or insertions of different lengths. The mutations were positioned 415 bp from the left end of the PCR product. The lanes from left to right show the digestion products of homoduplex/heteroduplex DNA mixtures containing the mismatches GG and CC, GA and TC, GT and AC, and AA and TT; homoduplex/heteroduplex DNA mixtures containing insertions of 1, 2, 3, 6, 9, and 12 nucleotides; and homoduplex DNA alone. The 100-bp DNA ladder is shown on the far right. Figure 2. Agarose gel electrophoresis (right panel) of Surveyor nuclease digestion products of 2.95-kb heteroduplex DNA containing a GG or CC mismatches at 1.4 and 1.8 kb from the left end. Digestion products resulting from a single cleavage event per molecule are 1.4 and 1.55 kb or 1.8 and 1.15 kb long. Cutting the same molecule at both mismatch sites gives an additional product that is 0.4 kb long. The lanes from left to right show the digestion products of the heteroduplex/homoduplex DNA mixture and homoduplex DNA alone. A 1-kb Plus DNA ladder (Invitrogen) is on the far right. The left panel shows a representation of the digestion products with the asterisks indicating the locations of DNA cleavage. 704 BioTechniques Vol. 36, No. 4 (2004)

indicates that, on average, there is at least 2-fold more (approximately 200 and 400 bp) cleavage products formed from an insertion/deletion than from a single-base mismatch. The digestion of annealed homoduplex DNA alone with Surveyor nuclease produced no discrete products but did generate small amounts of nonspecific breakdown products shorter than full-length substrate (Figure 1, homoduplex lane). This material originates from cleavage at mismatches introduced by PCR errors and/or from nonspecific DNA endonuclease activity associated with Surveyor nuclease preparations (J.M. D Alessio, unpublished observation). Reaction conditions used to generate Surveyor nuclease cleavage products were optimized for each separation platform to maximize signal and reduce this background as much as possible. Mapping Multiple Mutations in Large DNA Fragments with Surveyor Nuclease Mutation detection by SSCP or DHPLC is generally limited to smaller DNA fragments (<1000 bp). It is also difficult to identify the number and location of multiple mutations in a fragment. Using Surveyor nuclease, the upper size range of mutation detection is extended considerably, and multiple mutations can be mapped. Figure 2 shows the Surveyor nuclease digestion products of a 2.95-kb fragment containing two single-base mismatches 1.4 and 1.8 kb from the 5 end. Cleavage products 1.8, 1.55, 1.4, 1.15, and 0.4 kb long were easily resolved in this gel system. Single molecules cut only once by Surveyor nuclease at each of the mismatch sites will generate pairs of cleavage products, the sum of whose lengths total 2.95 kb (i.e., 1.4 + 1.55 kb and 1.8 + 1.15 kb) (Figure 2, left panel). Single molecules digested to completion (two cuts per molecule) will produce limit digest cleavage products 1.4, 1.15, and 0.4 kb long. Studies with a number of different DNA fragments containing two or three single-base mismatch sites show that cut substrate molecules are generally cleaved only once under the reaction conditions used with unpurified PCR fragments (data not shown). Efficient cutting at both mismatch sites of a single 2.95-kb fragment to yield substantial 0.4-kb product required post-pcr DNA cleanup before Surveyor nuclease digestion. In addition to DNA cleanup, increased incubation time and amounts of Surveyor nuclease enhance the complete digestion of heteroduplex DNA with multiple mismatches. Background from nonspecific DNA cleavage by Surveyor nuclease also increases with such changes. Both the mismatch sites in the 2.95-kb fragment in Figure 2 produce GG and CC mismatches when mutant and reference DNA are annealed. The sequence context of each mismatch is different. After taking into consideration differences

in length, there was greater than 2-fold more 1.4-kb product than 1.15-kb product at 20 min, which is consistent with the site 1.4 kb from the 5 end being preferred by Surveyor nuclease over the site at 1.8 kb. This result suggests sequence context influences the Surveyor nuclease digestion rate. Surveyor nuclease can be used to map the general location(s) of a mutation(s) based on the lengths of the digestion products. In an unlabeled DNA fragment, the orientation of the position of a mutation(s) with respect to the fragment ends can be mapped by altering digestion fragment length(s) through the use of an alternate PCR primer. Mapping of the locations of multiple mutations in a DNA fragment with respect to fragment ends can also be achieved by labeling the ends of a PCR product with primers containing two different fluorescent labels. The Sensitivity of Mutation Detection with Surveyor Nuclease The CEL nuclease family of enzymes to which Surveyor nuclease belongs has been used extensively to screen for mutations in PCR product populations derived from pooled individual genomic DNAs (6 8). For example, an 8-fold pooling of individual DNAs (detection of 1 in 16 alleles) has been used in screening Arabidopsis mutants (6,7). To determine the sensitivity of mismatch detection using Surveyor nuclease, substrates containing different proportions of 632 bp (Figure 3A) or 2.23 kb (Figure 3B) single mismatch heteroduplex and homoduplex DNA were digested using Surveyor nuclease and analyzed by agarose gel electrophoresis. The expected digestion products can be seen at 1 in 32 heteroduplex to homoduplex for the 632-bp PCR product, and clearly at 1 in 16 heteroduplex to homoduplex, and even at 1 in 32 for the 2.23-kb PCR product. Utilizing WAVE HSD HPLC and the ABI Prism 3100 capillary electrophoresis with the 632- bp substrate mixture, the detection limits are 1 in 32 and 1 in 8, respectively. In other experiments performed with the WAVE HSD HPLC analysis of the products, similar levels of heteroduplex detection sensitivity were observed with PCRamplified genomic DNA containing GA and TC, AA and TT, TG and CA, or GG and CC mismatched pairs all present in different sequence contexts (data not shown). In conclusion, mutation detection with Surveyor nuclease provides a simple and versatile mutation detection method, based on a new class of mismatch recognition nuclease. We have shown its broad substrate specificity and high sensitivity with model substrates of different lengths. REFERENCES 1.Inazuka, M., H. Wenz, M. Sakabe, T. Tahira, and K. Hayashi. 1997. A streamlined mutation detection system: multicolor post-pcr fluorescence labeling and single-strand conformational polymorphism analysis by capillary electrophoresis. Genome Res. 7:1094-1103. 2.Oefner, P.J. and P.A. Underhill. 1995. Comparative DNA sequencing by denaturing high-performance liquid chromatography (DHPLC). Am. J. Hum. Genet. 57:A266. 3.Oleykowski, C.A., C.R. Bronson Mullins, A.K. Godwin, and A.T. Yeung. 1998. Mutation detection using a novel plant endonuclease. Nucleic Acids Res. 26:4597-4602. 4.Kulinski, J., D. Besack, C.A. Oleykowski, A.K. Godwin, and A.T. Yeung. 2000. CEL enzymatic mutation detection assay. BioTechniques 29:44-48. 5.Yang, B., X. Wen, N.S. Kodali, C.A. Oleykowski, C.G. Miller, J. Kulinski, D. Besack, J.A. Yeung, et al. 2000. Purification, cloning, and characterization of CEL nuclease. Biochemistry 39:3533-3541. 6.Colbert, T., B.J. Till, R. Tompa, S. Reynolds, M.N. Steine, A.T. Yeung, C.M. McCallum, L. Comai, et al. 2001. High-throughput screening for induced point mutations. Plant Physiol. 126:480-484. 7.Till, B.J., S.H. Reynolds, E.A. Greene, C.A. Codomo, L.C. Enns, J.E. Johnson, C. Burtner, A.R. Odden, et al. 2003. Large-scale discovery of induced point mutations with high-throughput TILLING. Genome Res. 13:524-530. 8.Perry, J.A., T.L. Wang, T.J. Welham, S. Gardner, J.M. Pike, S. Yoshida, and M. Parniske. 2003. A TILLING reverse genetics tool and a Web-accessible collection of mutants of the legume Lotus japonicus. Plant Physiol. 131:866-871. 9.Sokurenko, E.V., V. Tchesnokova, A.T. Yeung, C.A. Oleykowski, E. Trintchina, K.T. Hughes, R.A. Rashid, J.M. Brint, et al. 2001. Detection of simple mutations and polymorphisms in large genomic regions. Nucleic Acids Res. 29:e111. Address correspondence to Gary F. Gerard, Transgenomic, Inc., 11 Firstfield Road, Suite E, Gaithersburg, MD 20878, USA. e-mail: ggerard@transgenomic.com Figure 3. Agarose gel electrophoresis of Surveyor nuclease digestion products of (A) a 632-bp or (B) a 2.23-kb heteroduplex/homoduplex DNA. (A) DNA products containing a GG or CC mismatch 415 bp from the left end of the PCR product are shown. The proportion of heteroduplex mixed with homoduplex DNA is 1 in 2 (lane 1) and 1 in 32 (lane 3). Homoduplex DNA alone and a 100-bp DNA ladder are shown in lanes 2 and M, respectively. (B) DNA products containing a GG or CC mismatch 1.08 kb from the left end of the PCR product are shown. The proportion of heteroduplex mixed with homoduplex DNA is 1 in 2 (lane 1), 1 in 16 (lane 3), and 1 in 32 (lane 4). Homoduplex DNA alone and 1-kb Plus DNA ladders are shown in lanes 2 and M, respectively. Vol. 36, No. 4 (2004) BioTechniques 707