A FRET-based analysis of SNPs without fluorescent probes
http://www.100md.com
《核酸研究医学期刊》
Kankyo Engineering Co., Ltd, 2-1-38 Shiohama, Kisarazu, Chiba 292-0838, Japan and 1 Institute for Biological Resources and Functions, National Institute of Advanced Industrial Science and Technology, 1-1-1 Higashi, Tsukuba, Ibaraki 305-8566, Japan
* To whom correspondence should be addressed. Tel: +81 29 861 6026; Fax: +81 29 861 6400; Email: kanagawa-taka@aist.go.jp
ABSTRACT
Fluorescence resonance energy transfer (FRET) is a simple procedure for detecting specific DNA sequences, and is therefore used in many fields. However, the cost is relatively high, because FRET-based methods usually require fluorescent probes. We have designed a cost-effective way of using FRET, and developed a novel approach for the genotyping of single nucleotide polymorphisms (SNPs) and allele frequency estimation. The key feature of this method is that it uses a DNA-binding fluorogenic molecule, SYBR Green I, as an energy donor for FRET. In this method, single base extension is performed with dideoxynucleotides labeled with an orange dye and a red dye in the presence of SYBR Green I. The dyes incorporated into the extended products accept energy from SYBR Green I and emit fluorescence. We have validated the method with ten SNPs, which were successfully discriminated by end-point measurements of orange and red fluorescence intensity in a microplate fluorescence reader. Using a mixture of homozygous samples, we also confirmed the potential of this method for estimation of allele frequency. Application of this strategy to large-scale studies will reduce the time and cost of genotyping a vast number of SNPs.
INTRODUCTION
The completion of the Human Genome Project has set the stage for the screening of genetic mutations to identify disease-related genes on a genome-wide scale. Single nucleotide polymorphisms (SNPs), which occur approximately every 1.9 kb in the human genome (1), are attractive targets for genetic screening and pharmacogenetic studies. There are several different techniques and approaches for SNP genotyping, and each has different advantages and disadvantages (2). However, a precise, cost-effective, high-throughput method is still required for large-scale SNP genotyping involving analysis of large numbers of SNPs across many individuals. A method for allele frequency estimation is also required for efficient and economical identification of disease-associated SNPs without genotyping all individuals.
The current methods for SNP genotyping can be broadly classified into two categories: enzyme-based methods and hybridization-based methods (3,4). Enzyme-based methods use the inherent catalytic specificity of enzymes for their substrates. Among them, single base extension (SBE)-based methods have proved particularly attractive for their robustness, allowing specific genotyping of most SNPs under the same reaction conditions (4), simplicity of primer design (5), accuracy (6,7) and adaptability to various detection formats, including gel and capillary electrophoresis (8,9), solution-phase homogeneous assay (10,11), MALDI-TOF mass-based spectrometry (12), and solid-phase microtiter plates (13) or microarrays (14,15). These features are suitable for high-throughput genotyping. However, SBE-based methods usually require complicated experimental procedures, including size separation of the product or removal of surplus substrates that limit the potential for automation.
Hybridization-based methods depend on single base differences in hybridization stability between a short oligonucleotide probe and the target DNA. Therefore, the probes have to be carefully designed, and the hybridization conditions have to be optimized for each SNP. In general, hybridization-based methods possess low assay fidelity and numerous sequence limitations (4). However, they are attractive because of the simple procedures involved. Several rapid genotyping methods have been developed, such as TaqMan or 5' nuclease assay (16), Molecular Beacon (17,18), iFRET (19) and Hybridization Probe Assay (20). In these methods, two dye molecules that act as an energy donor and acceptor for fluorescence resonance energy transfer (FRET) are used, and the fluorescence emission is measured for SNP discrimination. FRET is a distance-dependent interaction between two dye molecules, and allows allele discrimination in a single procedure without a separation or washing step. However, FRET-based methods are costly, because they require fluorescently labeled oligonucleotide probes that are tailor-made and expensive.
Previously, we reported a cost-effective method for SNP genotyping based on FRET (21). The key feature of this method is that it uses fluorescently labeled mononucleotides as FRET reagents instead of the fluorescent probes that are used by current methods. This approach avoids the time and cost of synthesis of fluorescent probes. In this method, fluorescently labeled mononucleotides are inserted into the strand by the allele-specific primer extension and emit FRET signals. However, this method has three limitations: allele-specific primers have to be carefully designed for selective allele primer extension. Reaction conditions have to be optimized separately for each SNP; consequently, there is no single set of reaction conditions that is optimal for genotyping all SNPs, and the method requires two separate tubes for each pair of allele-specific primers. These features are disadvantageous for high-throughput analysis of large numbers of SNPs.
To overcome these limitations, we have developed a novel SBE-based method that uses FRET in a cost-effective way to identify SNPs. Here, we describe this novel method, which uses a DNA-binding fluorogenic molecule, SYBR Green I, as the energy donor in FRET to excite fluorescently labeled dideoxynucleotides bound to the primer by an SBE reaction, and discriminates SNPs by measuring fluorescence after the SBE reaction. The same detection principle using FRET between SYBR Green I and a fluorescent dye , and the same combination of SBE and FRET (10), have already been reported, but these reported methods require a fluorescent probe or primer, which increases the cost. We combined SBE and FRET in a cost-effective way, retaining most of the advantages of both technologies. We validated our method with ten SNPs, and then investigated the potential of our method for estimation of allele frequency.
MATERIALS AND METHODS
Chemicals
We obtained ddNTPs labeled with Cy5 from Amersham Biosciences, ddNTPs labeled with Rox from Perkin-Elmer, and SYBR Green I from Molecular Probes. We purchased ThermoSequenase DNA polymerase from Amersham Biosciences, Accuprime Supermix II from Invitrogen and shrimp alkaline phosphatase from Roche. We purchased E.coli exonuclease I and lambda exonuclease from New England Biolabs.
Oligonucleotides
All the oligonucleotides were synthesized and purified by Espec Oligo Service. The genotyping primers and PCR primers used in this study are described in Tables 1 and 2, respectively.
Table 1. Genotyping primers used in this study
Table 2. PCR primers used in this study
Real sample preparation by PCR
Genomic DNA was extracted with ISOHAIR (Nippon Gene) from black hair samples of 18 volunteers according to the manufacturer's manual. PCR was performed with 20 ng of genomic DNA template, 200 nM of each primer shown in Table 2 and Accuprime Supermix II in a final reaction volume of 25 μl in an I-Cycler (Bio-Rad Laboratories). The reaction mixture was held at 94°C for 2 min, followed by 37 cycles of denaturation at 94°C for 30 s, annealing at 54°C (CYP2C19) or 60°C (other primers) for 20 s and extension at 68°C for 25 s, and then kept at 68°C for 5 min. The quantity of genomic DNA extracted from the hair samples was not large enough for PCR amplification of all SNP sites shown in Table 1 for all 18 extracts, and therefore four SNP sites were amplified from fewer than 18 extracts.
Pretreatment of PCR product for SBE reaction
The PCR products were treated with 10 units of exonuclease I for 30 min at 37°C to degrade the primers left over from the PCR, then the exonuclease I was inactivated at 90°C for 10 min. The products were treated with 10 units of lambda exonuclease for 30 min at 37°C to generate single-stranded DNA, and were then treated with 2 units of shrimp alkaline phosphatase for 30 min at 37°C to degrade the dNTPs, followed by 10 min at 90°C for enzyme inactivation. Four microliters of the treated PCR product was used as the template for the SBE reaction.
SBE reaction
The SBE reaction mixture contained 50 mM Tris–HCl (pH 9.0), 50 mM KCl, 5 mM MgCl2, 5 mM NaCl, 200 nM each ddNTP and 2.5 U Thermosequenase in a total volume of 50 μl. The two ddNTPs corresponding to the two substitutions of the SNP site were labeled with Rox and Cy5. The two other ddNTPs were unlabeled. Thermal cycling was performed in an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems) under the following conditions: 94°C for 2 min, followed by 45 (genotyping) or 35 (allele frequency estimation) cycles of 93°C for 20 s and 50°C for 60 s. A no-template control (NTC) contained water instead of the template.
Fluorescence measurement
The fluorescence intensity after the SBE reaction was measured with a Spectra Max Gemini Microplate Spectrofluorometer (Molecular Devices) using an excitation wavelength of 495 nm at room temperature. For genotype assignment by end-point readings, emission wavelength/cutoff filter combinations of 605/570 nm (orange) and 670/630 nm (red) were used. A 550/530 nm filter was also used for fluorescence compensation for allele frequency estimation.
Signal ratio calculation
The end-point fluorescence intensity at 605 nm (I605) was divided by the intensity at 667 nm (I667) after correction for the fluorescence intensity of NTC (I605,NTC and I667,NTC), as follows:
Allele frequency estimation
The homozygous samples for each allele of TAP2 and SSPN gene polymorphisms were amplified by PCR, and purified with a QIAquick PCR purification kit (Qiagen) to remove the PCR primers and dNTPs left over from the PCR. Quantification was carried out using an Agilent 2100 Bioanalyzer (Agilent Technologies). The quantified products were then mixed in different proportions (100:0, 95:5, 90:10, 80:20, 50:50, 20:80, 10:90, 5:95 and 0:100) to provide pools with different allele frequencies. The mixed samples were then treated with lambda exonuclease as described above. The SBE reaction was performed with 10 ng of these mixed samples. After the SBE reaction, I550, I605 and I667 were measured using a fluorescence plate reader as described above, and the allele frequency was calculated as follows. The reaction mixture contained SYBR Green I, Rox and Cy5. From their fluorescence emission pattern, we considered I605 to be the sum of the fluorescence from SYBR Green I (I605(SYBR)) and Rox (I605(Rox)), and I667 to be the sum of SYBR Green I and Rox and Cy5 (I667(Cy5)). We considered I550 to arise only from SYBR Green I in all samples, and therefore I605(SYBR) and I667(SYBR) were calculated from the intensities of NTC as follows:
We calculated I605(Rox) and I667(Cy5) as follows:
where Ihomo was the intensity of the homozygous sample with the Rox-labeled nucleotide in the SBE reaction, and I605(Rox),homo and I667(Rox),homo were calculated as follows:
Then, the allele frequency f was calculated from I605(Rox) and I667(Cy5) after they were normalized by the fluorescence of the 50:50 heterozygous sample (I605(Rox),hetero and I667(Cy5),hetero) as follows:
RESULTS AND DISCUSSION
Strategy
A scheme of the method is shown in Figure 1. As an example, we describe the method to identify an SNP involving a C/T substitution. The amplified genomic DNA containing the polymorphic site is incubated with a primer (designed to anneal immediately next to the polymorphic site) in the presence of DNA polymerase, SYBR Green I, ddCTP labeled with orange dye (Rox; max, 605 nm), and ddTTP labeled with red dye (Cy5; max, 667 nm). The primer binds the complementary site and is extended with a single Rox-ddCTP or Cy5-ddTTP. SYBR Green I binds DNAs and emits green fluorescence on excitation with blue light (495 nm). SBE products containing Rox-ddCTP and Cy5-ddTTP can be excited by the green light emitted by SYBR Green I, and emit orange and red fluorescence, respectively, which are measured for SNP discrimination after the SBE reaction.
Figure 1. Schematic presentation of the new method. The SNP shown here as an example is a C/T substitution. Genotyping primers can be extended by a single base that is complementary to the nucleotide at the SNP site, and the corresponding fluorescence is emitted from the dye by FRET.
Effect of thermal cycling on the SBE reaction
In our experiments, the SBE reaction was performed with thermal cycling, which we found enhanced the signal fluorescence. This effect was examined with a DNA fragment of the ALDH2 gene, homozygous for C/C, which was prepared from genomic DNA by PCR. The SBE reaction was carried out with Rox-ddCTP and thermal cycling for 45 cycles, and the real-time fluorescence emission spectra were recorded (Figure 2). There was no peak in the spectrum after cycle 1, but the peak appeared after cycle 5, and the peak height increased as the cycle number increased. Since SYBR Green I bound to single-stranded DNA has been reported to emit fluorescence, although the intensity is one-tenth of that after binding to double-stranded DNA (22), such an increase in the peak height may be due to FRET between Rox in the extended primer and SYBR Green I attached to the primer. Another possible mechanism of this increase is that the extended primer itself forms a double-stranded structure in some regions, and the SYBR Green I attached there transfers energy to Rox in the primer. In subsequent experiments, thermal cycling for 35–45 cycles was performed in SBE reactions, and enhanced the signal fluorescence.
Figure 2. Effect of thermal cycling on the SBE reaction. Thermal cycling, with denaturation at 94°C for 20 s and annealing/extension at 50°C for 60 s, and the measurement of real-time fluorescence emission spectra (550–660 nm) in the annealing phase were performed on an ABI PRISM 7900HT Sequence Detection System.
Validation of the method
We validated the method with a DNA fragment of the ALDH2 gene containing a polymorphic site (C/C, C/T and T/T). The DNA fragment was subjected to the SBE reaction with Rox-ddCTP, Cy5-ddTTP, and unlabeled ddATP and ddGTP. After the reaction, the fluorescence emission spectra were measured by excitation of SYBR Green I (Figure 3). A peak in the orange emission band (610 nm) was observed with homozygous C/C, whereas a peak in the red emission band (670 nm) was observed with homozygous T/T. Two peaks in the red and orange emission bands were observed with heterozygous C/T. These results indicate that the use of two different fluorescently labeled dideoxynucleotides allows the detection of both SNP alleles in a single tube, and the three genotypes can be clearly distinguished from one another by measuring the intensity of orange and red fluorescence after the reaction.
Figure 3. Validation of the method with the three known genotypes of the ALDH2 gene (C/C, C/T, T/T). After the SBE reaction, the fluorescence emission spectra (580–700 nm) were measured with a microplate fluorescence reader. The NTC spectrum was subtracted from all spectra.
SNP genotyping
The new method was then used to type ten different SNPs containing all possible substitutions, i.e. C/T, G/A, C/A, G/T, A/T and G/C (Table 1). After the SBE reaction, the fluorescence intensities at 605 and 667 nm were measured by fluorescence plate reader and used to generate points on an XY scatter plot (Figure 4). For each SNP, homozygous and heterozygous SNP genotypes were discriminated as clusters on the graph, and the negative control was separated from them. The genotypes were then distinguished by comparing their signal ratio calculated as described in Materials and Methods. The signal ratios were 3.31 to 5.59 and –0.64 to 0.02 for the homozygous samples, and 0.87 to 1.55 for the heterozygous samples (Table 3). Thus, the three genotypes were clearly distinguishable. The SNPs were also analyzed in parallel by a conventional PCR–RFLP assay or an allele-specific primer extension method (21) to verify the results of our method.
Figure 4. Genotyping results for ten SNPs. A scatter plot of the end-point 605 nm and 667 nm fluorescence intensities is shown. The fluorescence intensities of NTC were subtracted from sample fluorescence readings.
Table 3. The signal ratio for 10 SNPs
We also measured the fluorescence intensities of all samples with a real-time PCR machine (ABI PRISM 7900) after the SBE reaction. The maximum wavelength measurable by this machine was 660 nm, so we could not calculate the signal ratio using I667. Therefore, we compared (I605 – I580) and (I660 – I645), and we could distinguish the genotypes (data not shown). In principle, real-time PCR machines are suitable for our method if the detection wavelength is suitable.
Some SNPs have three or four possible bases at the same position. In principle, our method is applicable for distinguishing those SNPs using three or four different dyes in one tube. We tried Joe (max, 550 nm)-labeled ddCTP and Cy5.5 (max, 695 nm)-labeled ddATP with synthetic oligonucleotide templates, and observed the fluorescence emission peaks corresponding to these dyes after SBE reaction with the corresponding templates (data not shown). Therefore, the four possible bases would be distinguishable by using Joe, Rox, Cy5 and Cy5.5, or other appropriate combinations of fluorescent dyes.
Allele frequency estimation
Allele frequency estimation is an efficient and economical method of identifying SNPs associated with disease without genotyping all individuals. We investigated the potential of our method for estimation of allele frequency. We mixed homozygous samples of each allele for the TAP2 and SSPN gene polymorphisms, and measured fluorescence intensities after the SBE reaction. The estimated allele frequencies were linearly related to the expected ratios, with correlation coefficients of more than 0.997 (Figure 5), indicating that our method is useful for the estimation of allele frequency.
Figure 5. Allele frequency calculation from analysis of PCR product pools for the TAP2 and SSPN gene polymorphisms. The experiments were performed with three replicates. Predicted allele frequencies are based on the amount of PCR product from each homozygous donor used to construct the pool.
In actual allele frequency estimation, PCR is performed after mixing of DNA samples. There are many reports on PCR bias (29), which may change the ratio of genotypes. Polz and Cavanaugh (30) examined the bias in PCR using a 1:1 to 1:20 mixture of two templates possessing six different nucleotides in the middle of the template. The initial ratio was maintained. Small differences in DNA sequences such as SNPs will not cause bias in PCR amplification. Therefore, we will be able to estimate allele frequency after DNA amplification by PCR.
Advantages of the method
We have demonstrated a novel, cost-effective and simple method for SNP analysis. The method has several major advantages.
The method uses SYBR Green I and fluorescently labeled ddNTPs for the FRET reagent instead of fluorescently labeled oligonucleotides. This can markedly reduce the cost.
Our method, based on the SBE reaction, is robust, allowing specific genotyping of most SNPs under the same reaction conditions, as well as simple primer design. Therefore, the effort required for assay design and optimization is minimal. Indeed, in our experiments, all genotyping primers worked well under the standard assay conditions with no requirements for optimization. This is a considerable advantage when compared with hybridization-based methods such as TaqMan and Molecular Beacon assays, which require optimization of the sequences of probes and hybridization conditions. Optimization of probe sequences is often performed by trial and error using expensive, fluorescently labeled oligonucleotides, and this process is costly and time-consuming. Our method is robust, and therefore very useful when a large number of SNPs—particularly new SNP sites—need to be analyzed.
Genotypes can be determined by end-point fluorescence measurement with a simple optical system. This simple procedure can significantly increase throughput.
All reactions, including PCR, enzyme treatment and SBE, can be performed simply in the same vessel by the addition of solutions and incubation, and the fluorescence of the resulting product can be measured directly in the same vessel. Hence, automation is facilitated, allowing the use of a robotic workstation for high throughput. However, our method needs post-PCR processing by opening the tube, and therefore the total assay time and the risk of carryover contaminations are greater than in the hybridization-based methods, which are closed-tube methods. The risk of carryover contaminants in our method will be reduced by substituting dTTP by dUTP in PCR and treating the reaction mixture with uracil-DNA glycosylase (31) to cut the contaminants before PCR.
This homogeneous assay is not restricted to a particular format, making it possible to envisage different high-throughput engineering strategies, such as DNA microarrays with many kinds of single base extension primers immobilized on one chip to type many SNPs at one time. The current microarray-based methods require washing before fluorescence measurement, but our method does not require washing because the dyes are excited by SYBR Green I by FRET. Our method uses single-color excitation, while the current microarray-based methods use dual-color excitation.
We believe that these advantages will allow our method to contribute greatly to pharmacogenetic studies.
ACKNOWLEDGEMENTS
We are grateful to Dr Y. Takatsu and Dr M. Inazuka for advice and constructive criticism of the manuscript.
REFERENCES
Sachidanandam,R., Weissman,D., Schmidt,S.C., Kakol,J.M., Stein,L.D., Marth,G., Sherry,S., Mullikin,J.C., Mortimore,B.J., Willey,D.L. et al. ( (2001) ) A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature, , 409, , 928–933.
Kwok,P.Y. ( (2001) ) Methods for genotyping single nucleotide polymorphisms. Annu. Rev. Genomics Hum. Genet., , 2, , 235–258.
Kirk,B.W., Feinsod,M., Favis,R., Kliman,R.M. and Barany,F. ( (2002) ) Single nucleotide polymorphism seeking long term association with complex disease. Nucleic Acids Res., , 30, , 3295–3311.
Syvanen,A.C. ( (2001) ) Accessing genetic variation: genotyping single nucleotide polymorphisms. Nature Rev. Genet., , 2, , 930–942.
Tong,A.K. and Ju,J. ( (2002) ) Single nucleotide polymorphism detection by combinatorial fluorescence energy transfer tags and biotinylated dideoxynucleotides. Nucleic Acids Res., , 30, , e19.
Syvanen,A.C. ( (1999) ) From gels to chips: ‘minisequencing’ primer extension for analysis of point mutations and single nucleotide polymorphisms. Hum. Mutat., , 13, , 1–10.
Pastinen,T., Kurg,A., Metspalu,A., Peltonen,L. and Syvanen,A.C. ( (1997) ) Minisequencing: a specific tool for DNA analysis and diagnostics on oligonucleotide arrays. Genome Res., , 7, , 606–614.
Pastinen,T., Partanen,J. and Syvanen,A.C. ( (1996) ) Multiplex, fluorescent, solid-phase minisequencing for efficient screening of DNA sequence variation. Clin. Chem., , 42, , 1391–1397.
Tully,G., Sullivan,K.M., Nixon,P., Stones,R.E. and Gill,P. ( (1996) ) Rapid detection of mitochondrial sequence polymorphisms using multiplex solid-phase fluorescent minisequencing. Genomics, , 34, , 107–113.
Chen,X., Zehnbauer,B., Gnirke,A. and Kwok,P.Y. ( (1997) ) Fluorescence energy transfer detection as a homogeneous DNA diagnostic method. Proc. Natl Acad. Sci. USA, , 94, , 10756–10761.
Hsu,T.M., Chen,X., Duan,S., Miller,R.D. and Kwok,P.Y. ( (2001) ) Universal SNP genotyping assay with fluorescence polarization detection. Biotechniques, , 31, , 560, 562, 564–568, passim.
Haff,L.A. and Smirnov,I.P. ( (1997) ) Multiplex genotyping of PCR products with MassTag-labeled primers. Nucleic Acids Res., , 25, , 3749–3750.
Nikiforov,T.T., Rendle,R.B., Goelet,P., Rogers,Y.H., Kotewicz,M.L., Anderson,S., Trainor,G.L. and Knapp,M.R. ( (1994) ) Genetic bit analysis: a solid phase method for typing single nucleotide polymorphisms. Nucleic Acids Res., , 22, , 4167–4175.
Raitio,M., Lindroos,K., Laukkanen,M., Pastinen,T., Sistonen,P., Sajantila,A. and Syvanen,A.C. ( (2001) ) Y-chromosomal SNPs in Finno-Ugric-speaking populations analyzed by minisequencing on microarrays. Genome Res., , 11, , 471–482.
Hirschhorn,J.N., Sklar,P., Lindblad-Toh,K., Lim,Y.M., Ruiz-Gutierrez,M., Bolk,S., Langhorst,B., Schaffner,S., Winchester,E. and Lander,E.S. ( (2000) ) SBE-TAGS: an array-based method for efficient single-nucleotide polymorphism genotyping. Proc. Natl Acad. Sci. USA, , 97, , 12164–12169.
Livak,K.J., Marmaro,J. and Todd,J.A. ( (1995) ) Towards fully automated genome-wide polymorphism screening. Nature Genet., , 9, , 341–342.
Piatek,A.S., Tyagi,S., Pol,A.C., Telenti,A., Miller,L.P., Kramer,F.R. and Alland,D. ( (1998) ) Molecular beacon sequence analysis for detecting drug resistance in Mycobacterium tuberculosis. Nat. Biotechnol., , 16, , 359–363.
Tyagi,S., Bratu,D.P. and Kramer,F.R. ( (1998) ) Multicolor molecular beacons for allele discrimination. Nature Biotechnol., , 16, , 49–53.
Howell,W.M., Jobs,M. and Brookes,A.J. ( (2002) ) iFRET: an improved fluorescence system for DNA-melting analysis. Genome Res., , 12, , 1401–1407.
Schutz,E., von Ahsen,N. and Oellerich,M. ( (2000) ) Genotyping of eight thiopurine methyltransferase mutations: three-color multiplexing, ‘two-color/shared’ anchor, and fluorescence-quenching hybridization probe assays based on thermodynamic nearest-neighbor probe design. Clin. Chem., , 46, , 1728–1737.
Takatsu,K., Yokomaku,T., Kurata,S. and Kanagawa,T. ( (2004) ) A new approach to SNP genotyping with fluorescently labeled mononucleotides. Nucleic Acids Res., , 32, , e60.
Vitzthum,F., Geiger,G., Bisswanger,H., Brunner,H. and Bernhagen,J. ( (1999) ) A quantitative fluorescence-based microplate assay for the determination of double-stranded DNA using SYBR Green I and a standard ultraviolet transilluminator gel imaging system. Anal. Biochem., , 276, , 59–64.
Thomasson,H.R., Edenberg,H.J., Crabb,D.W., Mai,X.L., Jerome,R.E., Li,T.K., Wang,S.P., Lin,Y.T., Lu,R.B. and Yin,S.J. ( (1991) ) Alcohol and aldehyde dehydrogenase genotypes and alcoholism in Chinese men. Am. J. Hum. Genet., , 48, , 677–681.
Lohmueller,K.E., Pearce,C.L., Pike,M., Lander,E.S. and Hirschhorn,J.N. ( (2003) ) Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nature Genet., , 33, , 177–182.
Siffert,W., Rosskopf,D., Siffert,G., Busch,S., Moritz,A., Erbel,R., Sharma,A.M., Ritz,E., Wichmann,H.E., Jakobs,K.H. et al. ( (1998) ) Association of a human G-protein beta3 subunit variant with hypertension. Nature Genet., , 18, , 45–48.
de Morais,S.M., Wilkinson,G.R., Blaisdell,J., Nakamura,K., Meyer,U.A. and Goldstein,J.A. ( (1994) ) The major genetic defect responsible for the polymorphism of S-mephenytoin metabolism in humans. J. Biol. Chem., , 269, , 15419–15422.
Levine,A.J., Momand,J. and Finlay,C.A. ( (1991) ) The p53 tumour suppressor gene. Nature, , 351, , 453–456.
Toguchida,J., Yamaguchi,T., Dayton,S.H., Beauchamp,R.L., Herrera,G.E., Ishizaki,K., Yamamuro,T., Meyers,P.A., Little,J.B., Sasaki,M.S. et al. ( (1992) ) Prevalence and spectrum of germline mutations of the p53 gene among patients with sarcoma. N. Engl. J. Med., , 326, , 1301–1308.
Kanagawa,T. ( (2003) ) Bias and artifacts in multitemplate polymerase chain reactions (PCR). J. Biosci. Bioeng., , 96, , 317–323.
Polz,M.F. and Cavanaugh,C.M. ( (1998) ) Bias in template-to-product ratios in multitemplate PCR. Appl. Environ. Microbiol., , 64, , 3724–3730.
Ritzler,M., Perschil,I. and Altwegg,M. ( (1999) ) Influence of residual uracil-DNA glycosylase activity on the electrophoretic migration of dUTP-containing PCR products. J. Microbiol. Methods, , 35, , 73–76.(Kyoko Takatsu, Toyokazu Yokomaku, Shinya)
* To whom correspondence should be addressed. Tel: +81 29 861 6026; Fax: +81 29 861 6400; Email: kanagawa-taka@aist.go.jp
ABSTRACT
Fluorescence resonance energy transfer (FRET) is a simple procedure for detecting specific DNA sequences, and is therefore used in many fields. However, the cost is relatively high, because FRET-based methods usually require fluorescent probes. We have designed a cost-effective way of using FRET, and developed a novel approach for the genotyping of single nucleotide polymorphisms (SNPs) and allele frequency estimation. The key feature of this method is that it uses a DNA-binding fluorogenic molecule, SYBR Green I, as an energy donor for FRET. In this method, single base extension is performed with dideoxynucleotides labeled with an orange dye and a red dye in the presence of SYBR Green I. The dyes incorporated into the extended products accept energy from SYBR Green I and emit fluorescence. We have validated the method with ten SNPs, which were successfully discriminated by end-point measurements of orange and red fluorescence intensity in a microplate fluorescence reader. Using a mixture of homozygous samples, we also confirmed the potential of this method for estimation of allele frequency. Application of this strategy to large-scale studies will reduce the time and cost of genotyping a vast number of SNPs.
INTRODUCTION
The completion of the Human Genome Project has set the stage for the screening of genetic mutations to identify disease-related genes on a genome-wide scale. Single nucleotide polymorphisms (SNPs), which occur approximately every 1.9 kb in the human genome (1), are attractive targets for genetic screening and pharmacogenetic studies. There are several different techniques and approaches for SNP genotyping, and each has different advantages and disadvantages (2). However, a precise, cost-effective, high-throughput method is still required for large-scale SNP genotyping involving analysis of large numbers of SNPs across many individuals. A method for allele frequency estimation is also required for efficient and economical identification of disease-associated SNPs without genotyping all individuals.
The current methods for SNP genotyping can be broadly classified into two categories: enzyme-based methods and hybridization-based methods (3,4). Enzyme-based methods use the inherent catalytic specificity of enzymes for their substrates. Among them, single base extension (SBE)-based methods have proved particularly attractive for their robustness, allowing specific genotyping of most SNPs under the same reaction conditions (4), simplicity of primer design (5), accuracy (6,7) and adaptability to various detection formats, including gel and capillary electrophoresis (8,9), solution-phase homogeneous assay (10,11), MALDI-TOF mass-based spectrometry (12), and solid-phase microtiter plates (13) or microarrays (14,15). These features are suitable for high-throughput genotyping. However, SBE-based methods usually require complicated experimental procedures, including size separation of the product or removal of surplus substrates that limit the potential for automation.
Hybridization-based methods depend on single base differences in hybridization stability between a short oligonucleotide probe and the target DNA. Therefore, the probes have to be carefully designed, and the hybridization conditions have to be optimized for each SNP. In general, hybridization-based methods possess low assay fidelity and numerous sequence limitations (4). However, they are attractive because of the simple procedures involved. Several rapid genotyping methods have been developed, such as TaqMan or 5' nuclease assay (16), Molecular Beacon (17,18), iFRET (19) and Hybridization Probe Assay (20). In these methods, two dye molecules that act as an energy donor and acceptor for fluorescence resonance energy transfer (FRET) are used, and the fluorescence emission is measured for SNP discrimination. FRET is a distance-dependent interaction between two dye molecules, and allows allele discrimination in a single procedure without a separation or washing step. However, FRET-based methods are costly, because they require fluorescently labeled oligonucleotide probes that are tailor-made and expensive.
Previously, we reported a cost-effective method for SNP genotyping based on FRET (21). The key feature of this method is that it uses fluorescently labeled mononucleotides as FRET reagents instead of the fluorescent probes that are used by current methods. This approach avoids the time and cost of synthesis of fluorescent probes. In this method, fluorescently labeled mononucleotides are inserted into the strand by the allele-specific primer extension and emit FRET signals. However, this method has three limitations: allele-specific primers have to be carefully designed for selective allele primer extension. Reaction conditions have to be optimized separately for each SNP; consequently, there is no single set of reaction conditions that is optimal for genotyping all SNPs, and the method requires two separate tubes for each pair of allele-specific primers. These features are disadvantageous for high-throughput analysis of large numbers of SNPs.
To overcome these limitations, we have developed a novel SBE-based method that uses FRET in a cost-effective way to identify SNPs. Here, we describe this novel method, which uses a DNA-binding fluorogenic molecule, SYBR Green I, as the energy donor in FRET to excite fluorescently labeled dideoxynucleotides bound to the primer by an SBE reaction, and discriminates SNPs by measuring fluorescence after the SBE reaction. The same detection principle using FRET between SYBR Green I and a fluorescent dye , and the same combination of SBE and FRET (10), have already been reported, but these reported methods require a fluorescent probe or primer, which increases the cost. We combined SBE and FRET in a cost-effective way, retaining most of the advantages of both technologies. We validated our method with ten SNPs, and then investigated the potential of our method for estimation of allele frequency.
MATERIALS AND METHODS
Chemicals
We obtained ddNTPs labeled with Cy5 from Amersham Biosciences, ddNTPs labeled with Rox from Perkin-Elmer, and SYBR Green I from Molecular Probes. We purchased ThermoSequenase DNA polymerase from Amersham Biosciences, Accuprime Supermix II from Invitrogen and shrimp alkaline phosphatase from Roche. We purchased E.coli exonuclease I and lambda exonuclease from New England Biolabs.
Oligonucleotides
All the oligonucleotides were synthesized and purified by Espec Oligo Service. The genotyping primers and PCR primers used in this study are described in Tables 1 and 2, respectively.
Table 1. Genotyping primers used in this study
Table 2. PCR primers used in this study
Real sample preparation by PCR
Genomic DNA was extracted with ISOHAIR (Nippon Gene) from black hair samples of 18 volunteers according to the manufacturer's manual. PCR was performed with 20 ng of genomic DNA template, 200 nM of each primer shown in Table 2 and Accuprime Supermix II in a final reaction volume of 25 μl in an I-Cycler (Bio-Rad Laboratories). The reaction mixture was held at 94°C for 2 min, followed by 37 cycles of denaturation at 94°C for 30 s, annealing at 54°C (CYP2C19) or 60°C (other primers) for 20 s and extension at 68°C for 25 s, and then kept at 68°C for 5 min. The quantity of genomic DNA extracted from the hair samples was not large enough for PCR amplification of all SNP sites shown in Table 1 for all 18 extracts, and therefore four SNP sites were amplified from fewer than 18 extracts.
Pretreatment of PCR product for SBE reaction
The PCR products were treated with 10 units of exonuclease I for 30 min at 37°C to degrade the primers left over from the PCR, then the exonuclease I was inactivated at 90°C for 10 min. The products were treated with 10 units of lambda exonuclease for 30 min at 37°C to generate single-stranded DNA, and were then treated with 2 units of shrimp alkaline phosphatase for 30 min at 37°C to degrade the dNTPs, followed by 10 min at 90°C for enzyme inactivation. Four microliters of the treated PCR product was used as the template for the SBE reaction.
SBE reaction
The SBE reaction mixture contained 50 mM Tris–HCl (pH 9.0), 50 mM KCl, 5 mM MgCl2, 5 mM NaCl, 200 nM each ddNTP and 2.5 U Thermosequenase in a total volume of 50 μl. The two ddNTPs corresponding to the two substitutions of the SNP site were labeled with Rox and Cy5. The two other ddNTPs were unlabeled. Thermal cycling was performed in an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems) under the following conditions: 94°C for 2 min, followed by 45 (genotyping) or 35 (allele frequency estimation) cycles of 93°C for 20 s and 50°C for 60 s. A no-template control (NTC) contained water instead of the template.
Fluorescence measurement
The fluorescence intensity after the SBE reaction was measured with a Spectra Max Gemini Microplate Spectrofluorometer (Molecular Devices) using an excitation wavelength of 495 nm at room temperature. For genotype assignment by end-point readings, emission wavelength/cutoff filter combinations of 605/570 nm (orange) and 670/630 nm (red) were used. A 550/530 nm filter was also used for fluorescence compensation for allele frequency estimation.
Signal ratio calculation
The end-point fluorescence intensity at 605 nm (I605) was divided by the intensity at 667 nm (I667) after correction for the fluorescence intensity of NTC (I605,NTC and I667,NTC), as follows:
Allele frequency estimation
The homozygous samples for each allele of TAP2 and SSPN gene polymorphisms were amplified by PCR, and purified with a QIAquick PCR purification kit (Qiagen) to remove the PCR primers and dNTPs left over from the PCR. Quantification was carried out using an Agilent 2100 Bioanalyzer (Agilent Technologies). The quantified products were then mixed in different proportions (100:0, 95:5, 90:10, 80:20, 50:50, 20:80, 10:90, 5:95 and 0:100) to provide pools with different allele frequencies. The mixed samples were then treated with lambda exonuclease as described above. The SBE reaction was performed with 10 ng of these mixed samples. After the SBE reaction, I550, I605 and I667 were measured using a fluorescence plate reader as described above, and the allele frequency was calculated as follows. The reaction mixture contained SYBR Green I, Rox and Cy5. From their fluorescence emission pattern, we considered I605 to be the sum of the fluorescence from SYBR Green I (I605(SYBR)) and Rox (I605(Rox)), and I667 to be the sum of SYBR Green I and Rox and Cy5 (I667(Cy5)). We considered I550 to arise only from SYBR Green I in all samples, and therefore I605(SYBR) and I667(SYBR) were calculated from the intensities of NTC as follows:
We calculated I605(Rox) and I667(Cy5) as follows:
where Ihomo was the intensity of the homozygous sample with the Rox-labeled nucleotide in the SBE reaction, and I605(Rox),homo and I667(Rox),homo were calculated as follows:
Then, the allele frequency f was calculated from I605(Rox) and I667(Cy5) after they were normalized by the fluorescence of the 50:50 heterozygous sample (I605(Rox),hetero and I667(Cy5),hetero) as follows:
RESULTS AND DISCUSSION
Strategy
A scheme of the method is shown in Figure 1. As an example, we describe the method to identify an SNP involving a C/T substitution. The amplified genomic DNA containing the polymorphic site is incubated with a primer (designed to anneal immediately next to the polymorphic site) in the presence of DNA polymerase, SYBR Green I, ddCTP labeled with orange dye (Rox; max, 605 nm), and ddTTP labeled with red dye (Cy5; max, 667 nm). The primer binds the complementary site and is extended with a single Rox-ddCTP or Cy5-ddTTP. SYBR Green I binds DNAs and emits green fluorescence on excitation with blue light (495 nm). SBE products containing Rox-ddCTP and Cy5-ddTTP can be excited by the green light emitted by SYBR Green I, and emit orange and red fluorescence, respectively, which are measured for SNP discrimination after the SBE reaction.
Figure 1. Schematic presentation of the new method. The SNP shown here as an example is a C/T substitution. Genotyping primers can be extended by a single base that is complementary to the nucleotide at the SNP site, and the corresponding fluorescence is emitted from the dye by FRET.
Effect of thermal cycling on the SBE reaction
In our experiments, the SBE reaction was performed with thermal cycling, which we found enhanced the signal fluorescence. This effect was examined with a DNA fragment of the ALDH2 gene, homozygous for C/C, which was prepared from genomic DNA by PCR. The SBE reaction was carried out with Rox-ddCTP and thermal cycling for 45 cycles, and the real-time fluorescence emission spectra were recorded (Figure 2). There was no peak in the spectrum after cycle 1, but the peak appeared after cycle 5, and the peak height increased as the cycle number increased. Since SYBR Green I bound to single-stranded DNA has been reported to emit fluorescence, although the intensity is one-tenth of that after binding to double-stranded DNA (22), such an increase in the peak height may be due to FRET between Rox in the extended primer and SYBR Green I attached to the primer. Another possible mechanism of this increase is that the extended primer itself forms a double-stranded structure in some regions, and the SYBR Green I attached there transfers energy to Rox in the primer. In subsequent experiments, thermal cycling for 35–45 cycles was performed in SBE reactions, and enhanced the signal fluorescence.
Figure 2. Effect of thermal cycling on the SBE reaction. Thermal cycling, with denaturation at 94°C for 20 s and annealing/extension at 50°C for 60 s, and the measurement of real-time fluorescence emission spectra (550–660 nm) in the annealing phase were performed on an ABI PRISM 7900HT Sequence Detection System.
Validation of the method
We validated the method with a DNA fragment of the ALDH2 gene containing a polymorphic site (C/C, C/T and T/T). The DNA fragment was subjected to the SBE reaction with Rox-ddCTP, Cy5-ddTTP, and unlabeled ddATP and ddGTP. After the reaction, the fluorescence emission spectra were measured by excitation of SYBR Green I (Figure 3). A peak in the orange emission band (610 nm) was observed with homozygous C/C, whereas a peak in the red emission band (670 nm) was observed with homozygous T/T. Two peaks in the red and orange emission bands were observed with heterozygous C/T. These results indicate that the use of two different fluorescently labeled dideoxynucleotides allows the detection of both SNP alleles in a single tube, and the three genotypes can be clearly distinguished from one another by measuring the intensity of orange and red fluorescence after the reaction.
Figure 3. Validation of the method with the three known genotypes of the ALDH2 gene (C/C, C/T, T/T). After the SBE reaction, the fluorescence emission spectra (580–700 nm) were measured with a microplate fluorescence reader. The NTC spectrum was subtracted from all spectra.
SNP genotyping
The new method was then used to type ten different SNPs containing all possible substitutions, i.e. C/T, G/A, C/A, G/T, A/T and G/C (Table 1). After the SBE reaction, the fluorescence intensities at 605 and 667 nm were measured by fluorescence plate reader and used to generate points on an XY scatter plot (Figure 4). For each SNP, homozygous and heterozygous SNP genotypes were discriminated as clusters on the graph, and the negative control was separated from them. The genotypes were then distinguished by comparing their signal ratio calculated as described in Materials and Methods. The signal ratios were 3.31 to 5.59 and –0.64 to 0.02 for the homozygous samples, and 0.87 to 1.55 for the heterozygous samples (Table 3). Thus, the three genotypes were clearly distinguishable. The SNPs were also analyzed in parallel by a conventional PCR–RFLP assay or an allele-specific primer extension method (21) to verify the results of our method.
Figure 4. Genotyping results for ten SNPs. A scatter plot of the end-point 605 nm and 667 nm fluorescence intensities is shown. The fluorescence intensities of NTC were subtracted from sample fluorescence readings.
Table 3. The signal ratio for 10 SNPs
We also measured the fluorescence intensities of all samples with a real-time PCR machine (ABI PRISM 7900) after the SBE reaction. The maximum wavelength measurable by this machine was 660 nm, so we could not calculate the signal ratio using I667. Therefore, we compared (I605 – I580) and (I660 – I645), and we could distinguish the genotypes (data not shown). In principle, real-time PCR machines are suitable for our method if the detection wavelength is suitable.
Some SNPs have three or four possible bases at the same position. In principle, our method is applicable for distinguishing those SNPs using three or four different dyes in one tube. We tried Joe (max, 550 nm)-labeled ddCTP and Cy5.5 (max, 695 nm)-labeled ddATP with synthetic oligonucleotide templates, and observed the fluorescence emission peaks corresponding to these dyes after SBE reaction with the corresponding templates (data not shown). Therefore, the four possible bases would be distinguishable by using Joe, Rox, Cy5 and Cy5.5, or other appropriate combinations of fluorescent dyes.
Allele frequency estimation
Allele frequency estimation is an efficient and economical method of identifying SNPs associated with disease without genotyping all individuals. We investigated the potential of our method for estimation of allele frequency. We mixed homozygous samples of each allele for the TAP2 and SSPN gene polymorphisms, and measured fluorescence intensities after the SBE reaction. The estimated allele frequencies were linearly related to the expected ratios, with correlation coefficients of more than 0.997 (Figure 5), indicating that our method is useful for the estimation of allele frequency.
Figure 5. Allele frequency calculation from analysis of PCR product pools for the TAP2 and SSPN gene polymorphisms. The experiments were performed with three replicates. Predicted allele frequencies are based on the amount of PCR product from each homozygous donor used to construct the pool.
In actual allele frequency estimation, PCR is performed after mixing of DNA samples. There are many reports on PCR bias (29), which may change the ratio of genotypes. Polz and Cavanaugh (30) examined the bias in PCR using a 1:1 to 1:20 mixture of two templates possessing six different nucleotides in the middle of the template. The initial ratio was maintained. Small differences in DNA sequences such as SNPs will not cause bias in PCR amplification. Therefore, we will be able to estimate allele frequency after DNA amplification by PCR.
Advantages of the method
We have demonstrated a novel, cost-effective and simple method for SNP analysis. The method has several major advantages.
The method uses SYBR Green I and fluorescently labeled ddNTPs for the FRET reagent instead of fluorescently labeled oligonucleotides. This can markedly reduce the cost.
Our method, based on the SBE reaction, is robust, allowing specific genotyping of most SNPs under the same reaction conditions, as well as simple primer design. Therefore, the effort required for assay design and optimization is minimal. Indeed, in our experiments, all genotyping primers worked well under the standard assay conditions with no requirements for optimization. This is a considerable advantage when compared with hybridization-based methods such as TaqMan and Molecular Beacon assays, which require optimization of the sequences of probes and hybridization conditions. Optimization of probe sequences is often performed by trial and error using expensive, fluorescently labeled oligonucleotides, and this process is costly and time-consuming. Our method is robust, and therefore very useful when a large number of SNPs—particularly new SNP sites—need to be analyzed.
Genotypes can be determined by end-point fluorescence measurement with a simple optical system. This simple procedure can significantly increase throughput.
All reactions, including PCR, enzyme treatment and SBE, can be performed simply in the same vessel by the addition of solutions and incubation, and the fluorescence of the resulting product can be measured directly in the same vessel. Hence, automation is facilitated, allowing the use of a robotic workstation for high throughput. However, our method needs post-PCR processing by opening the tube, and therefore the total assay time and the risk of carryover contaminations are greater than in the hybridization-based methods, which are closed-tube methods. The risk of carryover contaminants in our method will be reduced by substituting dTTP by dUTP in PCR and treating the reaction mixture with uracil-DNA glycosylase (31) to cut the contaminants before PCR.
This homogeneous assay is not restricted to a particular format, making it possible to envisage different high-throughput engineering strategies, such as DNA microarrays with many kinds of single base extension primers immobilized on one chip to type many SNPs at one time. The current microarray-based methods require washing before fluorescence measurement, but our method does not require washing because the dyes are excited by SYBR Green I by FRET. Our method uses single-color excitation, while the current microarray-based methods use dual-color excitation.
We believe that these advantages will allow our method to contribute greatly to pharmacogenetic studies.
ACKNOWLEDGEMENTS
We are grateful to Dr Y. Takatsu and Dr M. Inazuka for advice and constructive criticism of the manuscript.
REFERENCES
Sachidanandam,R., Weissman,D., Schmidt,S.C., Kakol,J.M., Stein,L.D., Marth,G., Sherry,S., Mullikin,J.C., Mortimore,B.J., Willey,D.L. et al. ( (2001) ) A map of human genome sequence variation containing 1.42 million single nucleotide polymorphisms. Nature, , 409, , 928–933.
Kwok,P.Y. ( (2001) ) Methods for genotyping single nucleotide polymorphisms. Annu. Rev. Genomics Hum. Genet., , 2, , 235–258.
Kirk,B.W., Feinsod,M., Favis,R., Kliman,R.M. and Barany,F. ( (2002) ) Single nucleotide polymorphism seeking long term association with complex disease. Nucleic Acids Res., , 30, , 3295–3311.
Syvanen,A.C. ( (2001) ) Accessing genetic variation: genotyping single nucleotide polymorphisms. Nature Rev. Genet., , 2, , 930–942.
Tong,A.K. and Ju,J. ( (2002) ) Single nucleotide polymorphism detection by combinatorial fluorescence energy transfer tags and biotinylated dideoxynucleotides. Nucleic Acids Res., , 30, , e19.
Syvanen,A.C. ( (1999) ) From gels to chips: ‘minisequencing’ primer extension for analysis of point mutations and single nucleotide polymorphisms. Hum. Mutat., , 13, , 1–10.
Pastinen,T., Kurg,A., Metspalu,A., Peltonen,L. and Syvanen,A.C. ( (1997) ) Minisequencing: a specific tool for DNA analysis and diagnostics on oligonucleotide arrays. Genome Res., , 7, , 606–614.
Pastinen,T., Partanen,J. and Syvanen,A.C. ( (1996) ) Multiplex, fluorescent, solid-phase minisequencing for efficient screening of DNA sequence variation. Clin. Chem., , 42, , 1391–1397.
Tully,G., Sullivan,K.M., Nixon,P., Stones,R.E. and Gill,P. ( (1996) ) Rapid detection of mitochondrial sequence polymorphisms using multiplex solid-phase fluorescent minisequencing. Genomics, , 34, , 107–113.
Chen,X., Zehnbauer,B., Gnirke,A. and Kwok,P.Y. ( (1997) ) Fluorescence energy transfer detection as a homogeneous DNA diagnostic method. Proc. Natl Acad. Sci. USA, , 94, , 10756–10761.
Hsu,T.M., Chen,X., Duan,S., Miller,R.D. and Kwok,P.Y. ( (2001) ) Universal SNP genotyping assay with fluorescence polarization detection. Biotechniques, , 31, , 560, 562, 564–568, passim.
Haff,L.A. and Smirnov,I.P. ( (1997) ) Multiplex genotyping of PCR products with MassTag-labeled primers. Nucleic Acids Res., , 25, , 3749–3750.
Nikiforov,T.T., Rendle,R.B., Goelet,P., Rogers,Y.H., Kotewicz,M.L., Anderson,S., Trainor,G.L. and Knapp,M.R. ( (1994) ) Genetic bit analysis: a solid phase method for typing single nucleotide polymorphisms. Nucleic Acids Res., , 22, , 4167–4175.
Raitio,M., Lindroos,K., Laukkanen,M., Pastinen,T., Sistonen,P., Sajantila,A. and Syvanen,A.C. ( (2001) ) Y-chromosomal SNPs in Finno-Ugric-speaking populations analyzed by minisequencing on microarrays. Genome Res., , 11, , 471–482.
Hirschhorn,J.N., Sklar,P., Lindblad-Toh,K., Lim,Y.M., Ruiz-Gutierrez,M., Bolk,S., Langhorst,B., Schaffner,S., Winchester,E. and Lander,E.S. ( (2000) ) SBE-TAGS: an array-based method for efficient single-nucleotide polymorphism genotyping. Proc. Natl Acad. Sci. USA, , 97, , 12164–12169.
Livak,K.J., Marmaro,J. and Todd,J.A. ( (1995) ) Towards fully automated genome-wide polymorphism screening. Nature Genet., , 9, , 341–342.
Piatek,A.S., Tyagi,S., Pol,A.C., Telenti,A., Miller,L.P., Kramer,F.R. and Alland,D. ( (1998) ) Molecular beacon sequence analysis for detecting drug resistance in Mycobacterium tuberculosis. Nat. Biotechnol., , 16, , 359–363.
Tyagi,S., Bratu,D.P. and Kramer,F.R. ( (1998) ) Multicolor molecular beacons for allele discrimination. Nature Biotechnol., , 16, , 49–53.
Howell,W.M., Jobs,M. and Brookes,A.J. ( (2002) ) iFRET: an improved fluorescence system for DNA-melting analysis. Genome Res., , 12, , 1401–1407.
Schutz,E., von Ahsen,N. and Oellerich,M. ( (2000) ) Genotyping of eight thiopurine methyltransferase mutations: three-color multiplexing, ‘two-color/shared’ anchor, and fluorescence-quenching hybridization probe assays based on thermodynamic nearest-neighbor probe design. Clin. Chem., , 46, , 1728–1737.
Takatsu,K., Yokomaku,T., Kurata,S. and Kanagawa,T. ( (2004) ) A new approach to SNP genotyping with fluorescently labeled mononucleotides. Nucleic Acids Res., , 32, , e60.
Vitzthum,F., Geiger,G., Bisswanger,H., Brunner,H. and Bernhagen,J. ( (1999) ) A quantitative fluorescence-based microplate assay for the determination of double-stranded DNA using SYBR Green I and a standard ultraviolet transilluminator gel imaging system. Anal. Biochem., , 276, , 59–64.
Thomasson,H.R., Edenberg,H.J., Crabb,D.W., Mai,X.L., Jerome,R.E., Li,T.K., Wang,S.P., Lin,Y.T., Lu,R.B. and Yin,S.J. ( (1991) ) Alcohol and aldehyde dehydrogenase genotypes and alcoholism in Chinese men. Am. J. Hum. Genet., , 48, , 677–681.
Lohmueller,K.E., Pearce,C.L., Pike,M., Lander,E.S. and Hirschhorn,J.N. ( (2003) ) Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nature Genet., , 33, , 177–182.
Siffert,W., Rosskopf,D., Siffert,G., Busch,S., Moritz,A., Erbel,R., Sharma,A.M., Ritz,E., Wichmann,H.E., Jakobs,K.H. et al. ( (1998) ) Association of a human G-protein beta3 subunit variant with hypertension. Nature Genet., , 18, , 45–48.
de Morais,S.M., Wilkinson,G.R., Blaisdell,J., Nakamura,K., Meyer,U.A. and Goldstein,J.A. ( (1994) ) The major genetic defect responsible for the polymorphism of S-mephenytoin metabolism in humans. J. Biol. Chem., , 269, , 15419–15422.
Levine,A.J., Momand,J. and Finlay,C.A. ( (1991) ) The p53 tumour suppressor gene. Nature, , 351, , 453–456.
Toguchida,J., Yamaguchi,T., Dayton,S.H., Beauchamp,R.L., Herrera,G.E., Ishizaki,K., Yamamuro,T., Meyers,P.A., Little,J.B., Sasaki,M.S. et al. ( (1992) ) Prevalence and spectrum of germline mutations of the p53 gene among patients with sarcoma. N. Engl. J. Med., , 326, , 1301–1308.
Kanagawa,T. ( (2003) ) Bias and artifacts in multitemplate polymerase chain reactions (PCR). J. Biosci. Bioeng., , 96, , 317–323.
Polz,M.F. and Cavanaugh,C.M. ( (1998) ) Bias in template-to-product ratios in multitemplate PCR. Appl. Environ. Microbiol., , 64, , 3724–3730.
Ritzler,M., Perschil,I. and Altwegg,M. ( (1999) ) Influence of residual uracil-DNA glycosylase activity on the electrophoretic migration of dUTP-containing PCR products. J. Microbiol. Methods, , 35, , 73–76.(Kyoko Takatsu, Toyokazu Yokomaku, Shinya)