- Email: [email protected]
Single-Nucleotide Polymorphism Analysis by Pyrosequencing
Analytical Biochemistry 280, 103–110 (2000) doi:10.1006/abio.2000.4493, available online at http://www.idealibrary.com on
Single-Nucleotide Polymorphism Analysis by Pyrosequencing Afshin Ahmadian, Baback Gharizadeh, Anna C. Gustafsson, Fredrik Sterky, Pål Nyre´n, Mathias Uhle´n, and Joakim Lundeberg 1 Department of Biotechnology, The Royal Institute of Technology (KTH), SE-100 44 Stockholm, Sweden
Received October 6, 1999
There is a growing demand for high-throughput methods for analysis of single-nucleotide polymorphic (SNP) positions. Here, we have evaluated a novel sequencing approach, pyrosequencing, for such purposes. Pyrosequencing is a sequencing-by-synthesis method in which a cascade of enzymatic reactions yields detectable light, which is proportional to incorporated nucleotides. One feature of typing SNPs with pyrosequencing is that each allelic variant will give a unique sequence compared to the two other variants. These variants can easily be distinguished by a pattern recognition software. The software displays the allelic alternatives and allows for direct comparison with the pyrosequencing raw data. For optimal determination of SNPs, various protocols of nucleotide dispensing order were investigated. Here, we demonstrate that typing of SNPs can efficiently be performed by pyrosequencing using an automated system for parallel analysis of 96 samples in approximately 5 min, suitable for large-scale screening and typing of SNPs. © 2000 Academic Press
Key Words: SNP; DNA sequencing; pyrosequencing.
Genetic variation is the basis for human diversity and plays an important role in human diseases. Methods to screen and map genetic variability have, for more than two decades, been based on restriction fragment length polymorphism and microsatellite markers. More recent efforts have focused on the most common type of human genetic variation, single-nucleotide polymorphisms (SNPs). 2 A position is referred to as a 1 To whom correspondence should be addressed: Fax: ⫹46 8 245452. E-mail: [email protected]. 2 Abbreviations used: SNP, single-nucleotide polymorphism; LOH, loss of heterozygosity; ASO, allele-specific oligonucleotide; PP i, pyrophosphate; ATP, adenosine triphosphate; APS, adenosine 5⬘-phosphosulfate.
0003-2697/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
SNP when it exists in at least two variants with a frequency of more than 1% for the least common alternative (1). SNPs are distributed across the human genome by an approximate average of 1 SNP per 1000 base pairs (1, 2). As for microsatellite markers, SNPs can be used in linkage studies for identifying disease genes, in clinical genetic testing, in forensics, and for determination of loss of heterozygosity (LOH). The properties that make SNP analysis preferable compared to microsatellites are that SNPs are more prevalent than microsatellites and that many SNPs are located within the genes, directly affecting the gene product (protein). As the number of identified SNPs increases, there will be an increasing demand for efficient methods to type and assess the biological impact of this kind of genetic variation. The golden standard method for SNP scoring has been conventional Sanger DNA sequencing. However, sequencing based on gel electrophoresis generates more information than necessary, is time-consuming, is laborious, and requires labeling. Significant efforts have been made to improve SNP analysis with alternative techniques. Many of these techniques use allele-specific oligonucleotide (ASO) hybridization to discriminate between allelic variants such as high-density microarray chips (3– 6), padlock probes (7), or allelic discrimination during PCR (8 –12). Minisequencing approaches are an attractive alternative because of the direct interrogation of the variable position by DNA polymerase extension using different means of detection of the extended product (i.e., fluorescent and radioactive dyes or by mass spectrometry) (13–16). Recently, an alternative technique for DNA sequencing, called pyrosequencing, was described (17). Using a four-enzyme mixture, this sequencing-by-synthesis method relies on the luminometric detection of pyrophosphate (PP i) released upon nucleotide incorporation. Here, we have investigated the possible use of pyrosequencing for SNP analysis. The technique has 103
104
AHMADIAN ET AL. TABLE 1
List of the Primers Used for PCR Amplification and Pyrosequencing a wiaf1764
wiaf797
wiaf41
p53 codon 72
outercod innercod outerrev innerrev seqrev outercod innercod outerrev innerrev seqcod outercod innercod outerrev innerrev seqrev outercod innercod outerrev innerrev seqrev
AGTGAAAACATTGAAAACACA TCCAATGTGTGAAAAATATATAC AATGTTTTCACTGTCATAAAG AGAACACATACGTTTTACCA CATTTGTTAAGCTTTT TGTTTTAAGTTGCAGAGATG GAAGCTGAAGTTCAATCTTT AAATGGAAGCTCTTCAATAC GCTCTTCAATACTTAAGGTA GTTTTCTGTTGTAAATGC ACAGCATGCTTAGTTCTCT CATTTACACCACACTGCAAT AGACTTAATCCCTTGAATTG TTGACAGAGAAATAACGGG GTAGTAAAACCTGGATTA ATGCTGTCCCCGGACGA TCCAGATGAAGCTCCCAG CAGGAGGGGGCTGGTG AGGGGCCGCCGGTGTA GCTGCTGGTGCAGGGGCCA
a Outer, primers used in the outer PCR; inner, primers used in the inner PCR; cod, upstream; rev, downstream. All innercod primers are biotinylated at the 5⬘-end. Seqrev, pyrosequencing primers used on the immobilized single-stranded DNA; seqcod, pyrosequencing primer used on the eluted single-stranded DNA.
been applied on four SNPs located on chromosomes 9q and 17p. MATERIALS AND METHODS
Sample, PCR, and Template Preparation Human genomic DNA was extracted from 24 unrelated individuals. An outer multiplex PCR was performed to amplify four SNPs located on chromosomes 9q and 17p. The outer multiplex PCR (94°C for 1 min, 50°C for 40 s, and 72°C for 2 min for 35 cycles) was followed by individual inner PCRs (94°C for 1 min, 50°C for 40 s, and 72°C for 1 min for 35 cycles), generating ⬃80-bp fragments for each SNP. SNPs on chromosome 9 (wiaf1764, wiaf797, and wiaf41) were taken from the Whitehead Institute/ MIT server (http//www.genome.wi.mit.edu/SNP/human/ maps/Chr9.ALL.html). The SNP located on chromosome 17 was amplified from exon 4 in the p53 gene (codon 72). Primers for outer and inner PCR are listed in Table 1. One of the inner primers in the respective set was biotinylated at the 5⬘-end to allow immobilization. The outer and inner amplification mixture consisted of 10 mM Tris–HCl (pH 8.3), 2 mM MgCl2, 50 mM KCl, 0.1% (v/v) Tween 20, 0.2 mM dNTPs, 0.1 M of each primer, and 1 unit of AmpliTaq DNA polymerase (Perkin-Elmer, Norwalk, CT) in a total volume of 50 l. Five microliters of total human DNA (1 ng/l), isolated from blood samples of normal individuals, was used as outer PCR template. Forty microliters of biotinylated inner PCR products was immobilized onto streptavidin-coated superparamag-
netic beads (Dynabeads M280; Dynal, Oslo, Norway). Single-stranded DNA was obtained by incubating the immobilized PCR product in 10 l of 0.1 M NaOH for 5 min. The supernatant containing the eluted strand was neutralized by adding 5 l of 0.2 M HCl. The immobilized strand was resuspended in 8 l of H 2O. The eluted strand was used to pyrosequence wiaf797 and the immobilized strand to pyrosequence wiaf1764, wiaf41, and codon 72. One microliter of annealing buffer (100 mM Tris–acetate, pH 7.75, 20 mM Mg–acetate) and 0.1 M of respective sequencing primers (Table 1) were added to the singlestranded templates in a total volume of 10 l. Hybridization was performed by incubation at 94°C for 20 s, 65°C for 2 min, and then cooling to room temperature. Conventional DNA Sequencing Exon 4 of the p53 gene was amplified as described earlier (18) and sequenced by cycle sequencing with dye-labeled terminators (BigDye Terminator, PerkinElmer). The data were analyzed on a ABI PRISM 377 XL DNA sequencer (PE Applied Biosystems, Foster City, CA). Pyrosequencing Pyrosequencing was performed at 25°C in a volume of 50 l on an automated pyrosequencer instrument (kindly supplied by Pyrosequencing AB, Uppsala, Sweden). Single-stranded DNA with annealed sequence primer (the substrate) was added to the pyrosequencing reaction mixture containing 10 U of exonucleasedeficient (exo ⫺) Klenow DNA polymerase (Amersham Pharmacia Biotech, Uppsala, Sweden), 40 mU of apyrase (Sigma Chemical Co., St. Louis, MO), 4 g of purified luciferase/ml (BioThema, Dalaro¨, Sweden), 15 mU of recombinant ATP sulfurylase (19), 0.1 M Tris– acetate (pH 7.75), 0.5 mM EDTA, 5 mM Mg–acetate, 0.1% (w/v) bovine serum albumin (BioThema), 1 mM dithiothreitol, 10 M adenosine 5⬘-phosphosulfate (APS), 0.4 mg of poly(vinylpyrrolidone)/ml (360,000), and 100 g of D-luciferin/ml (BioThema). Stepwise elongation of the primer strand upon sequential addition of the different deoxynucleoside triphosphates (Amersham Pharmacia Biotech) and degradation of excess nucleotides by apyrase were carried out simultaneously. RESULTS
Genomic DNA from 24 unrelated individuals was used for determination of four single-nucleotide polymorphism sites by pyrosequencing. Three of these SNPs correspond to variations in chromosome 9 (wiaf1764, wiaf797, wiaf41) and one corresponds to a coding SNP in the p53 tumor suppressor gene (codon 72). The principle of the pyrosequencing technique is illustrated in Fig. 1. Various protocols of nucleotide
SINGLE-NUCLEOTIDE POLYMORPHISM ANALYSIS BY PYROSEQUENCING
105
FIG. 1. Principle of pyrosequencing. The reaction mixture consists of a single-stranded DNA with a short annealed primer, DNA polymerase, ATP sulfurylase, luciferase, and apyrase. The four nucleotide bases are added to the mixture in a defined order, i.e., CGAT (the arrow above the nucleotides indicates cyclic addition). (a, b) If the added nucleotide forms a base pair, the DNA polymerase incorporates the nucleotide and pyrophosphate will consequently be released. The released pyrophosphate will then be converted to ATP by ATP sulfurylase. Luciferase uses the ATP to generate detectable light. The excess of the added nucleotide will be degraded by apyrase. (c) If the added nucleotide does not form a base pair to the DNA template, the polymerase will not incorporate it and no light will be produced. The nucleotide will rapidly be degraded by apyrase. (d) Pyrosequencing raw data obtained for the case discussed above (SNP on p53 codon 72, homozygous G).
dispensing order were investigated for pyrosequencing of these SNPs. In addition, conventional DNA sequencing, used for p53 codon 72 (exon 4), has been included for comparison. The possibility of detecting loss of heterozygosity by SNP analysis was also examined using a breast cancer tumor. Pyrosequencing at Codon 72: Cyclic Protocol Template DNA prepared from blood samples was used in an outer multiplex amplification and was followed by four separate amplifications of the individual SNPs using a biotinylated primer. The biotinylated inner PCR products were separately immobilized onto streptavidin-coated superparamagnetic beads. Singlestranded DNA was obtained by incubating the immobilized PCR product in NaOH. The immobilized strand was resuspended in H 2O and was used for pyrosequencing. The coding single-nucleotide polymorphism at codon 72 involves either a G or C residue (Fig. 2a),
corresponding to amino acid proline (CCC) or arginine (CGC). The results using a cyclic addition of nucleotides (CGAT) are shown in Fig. 2b for the three possible variants of the SNP. The sequencing primer was designed so that the first incorporated nucleotide was identical in all variants, functioning as an internal control. Successful incorporation is expected by addition of the first nucleotide (nucleotide C), set to correspond to 1 peak equivalent (incorporation in only one of two alleles corresponds to 0.5 peak equivalents). In the case of homozygous C (top panel, b), the internal calibration peak is followed by a peak indicating incorporation of six nucleotides when the nucleotide G is added to the reaction. The subsequent sequence shows the consensus sequence (one A, one G, one C, and one A). Note that no signal is generated when noncomplementary nucleotides are added. In the heterozygous situation (middle panel, b), after one C, a peak equivalent to 3.5 G appears (six incorporations on one allele
106
AHMADIAN ET AL.
FIG. 2. Pyrosequencing on the coding SNP on the p53 gene (codon 72). (a) Sequence for the SNP. The arrows show primer position. The underlined nucleotides indicate the SNP site (C/G) corresponding to amino acid proline (CCC) or arginine (CGC). (b) Pyrosequencing result using cyclic addition of nucleotides (CGAT). (c) Pyrosequencing result using sequential (SNP specific) addition of nucleotides (see text for detailed explanation for (b) and (c). (d) Sequencing data obtained by conventional DNA sequencing.
plus one incorporation on the other allele). Following this addition, extension on the separate alleles will be out of phase. When the SNP position is homozygous G (lower panel, b), the pyrosequence will follow the consensus sequence in phase as the complementary nucleotides are added. All three possible variants of the SNP in codon 72 were identified among the 24 individuals (G, 14 of 24; G/C, 9 of 24; and C, 1 of 24). Pyrosequencing at Codon 72: Sequential Protocol Because the sequence context of the SNP is known, it is possible to reduce the number of nucleotide additions by using a sequential (SNP specific) dispensing protocol. Figure 2c shows the sequential addition of nucleotides CGCGAG for the same three variants of the SNP at codon 72 (top, middle, and lower panels). As in the cyclic protocol, the first nucleotide addition (C) normalizes the peak values. The second nucleotide addition (G) yields the same pattern as for the cyclic additions: 6, 3.5, and 1 base peak equivalent incorporation for the homozygous C, heterozygous C/G, and homozygous G, respectively. However, with the sequential procedure, it is possible to elaborate with the dispensing order to achieve synchronized extensions after the variable po-
sition. Here, this was achieved by addition of nucleotide C to extend the lagging allele in the heterozygous case followed by addition of nucleotide G. Hereby the extensions will be synchronized in all cases after the actual SNP position. The two remaining additions of nucleotides A and G will therefore function as postcontrols. The results obtained using a conventional DNA sequencing procedure on the three variants of the SNP in codon 72 of the p53 gene are shown in Fig. 2d. The homozygous alternatives are clearly distinguished (top and lower panels), while the heterozygous state is more difficult to interpret (middle panel, d). Actually, the automatic base-call algorithm (ABI PRISM 377 XL) scores the middle panel as being homozygous C. Analysis of wiaf1764, wiaf797, and wiaf41 As shown above, the allelic variants of a SNP determined by pyrosequencing can easily be distinguished by pattern recognition using either cyclic or sequential nucleotide addition. By taking advantage of a simple pattern recognition program, the allelic alternatives can be displayed. The program presents predicted peaks for the individual SNPs. Comparison between raw data and predicted patterns was made manually.
SINGLE-NUCLEOTIDE POLYMORPHISM ANALYSIS BY PYROSEQUENCING
107
FIG. 3. Pyrosequencing on a SNP positioned on chromosome 9q (wiaf1764). (a) Sequence for the three allelic alternatives. (b) Predicted pyrosequencing pattern for the three variants of the SNP when a cyclic addition of nucleotides (GTCA) is performed. (c) Raw data obtained by pyrosequencing for the three variants. As it is shown, the raw data completely match the predicted pyrosequencing pattern.
The user can hereby test alternative nucleotide dispensing orders to choose the most informative alternative. This was evaluated on SNPs positioned on chromosome 9 (wiaf1764, wiaf797, and wiaf41). For wiaf1764 all three SNP variants were observed among the 24 individuals (A, 12 of 24; A/C, 8 of 24; and C, 4 of 24). Figure 3 displays the three allelic alternatives (Fig. 3a), predicted pyrosequencing pattern (Fig. 3b), and the three examples of pyrosequencing raw data (Fig. 3c) using the cyclic protocol (GTCA). The top and lower panels show the two homozygous alternatives (C and A) and the middle panel corresponds to the heterozygous variant (C/A). The similarities between raw data and the three predicted patterns allow for direct and reliable determination of the SNP. Occasionally, reagents will contain minor contaminations of pyrophosphate that will yield a low background signal that can be subtracted prior to analysis. The obtained data for wiaf797 and wiaf41 showed a low frequency of variability (Fig. 4). In the case of wiaf41 (Fig. 4a), two allelic variants were found (A, 22 of 24; A/T, 2 of 24; and T, 0 of 24). In the case of wiaf797, only one variant was found (Fig. 4b) among the 24 individuals (G, 0 of 24; G/T, 0 of 24; and T, 24 of 24).
Analysis of Loss of Heterozygosity Analysis of loss of heterozygosity (LOH) plays an important role in the process of characterizing tumor suppressor genes. In the case of informative (heterozygous) SNPs, these may be used for such investigations as a complement to the more informative microsatellites. From a breast cancer study (20) we selected a codon 72 heterozygous sample. This sample, which showed loss of one allele by distal microsatellite markers, was evaluated for loss of heterozygosity using the pyrosequencing protocol. Genomic DNA was amplified and codon 72 was determined by a sequential protocol (CGCGAG) as outlined above. Figure 5 shows the sequence surrounding the SNP (Fig. 5a), the predicted alternatives for allelic loss (Fig. 5b), and the obtained data from normal and cancer tissues (Fig. 5c). As expected, the patient was shown to be heterozygous by pyrosequencing in the normal tissue sample (middle panel). The obtained pyrosequencing data for the tumor sample (Fig. 5c) show a high pattern similarity with the predicted pattern for a sample with loss of haplotype G (top panel). The deviations between the predicted pattern and the raw data are due to the
108
AHMADIAN ET AL.
FIG. 4. Pyrosequencing on SNPs wiaf41 (a) and wiaf797 (b). Raw data (on right) are compared to the predicted patterns (on left).
heterogeneous nature of tumor samples, having admixed “contaminating” normal cells. In this case about 50% of the cells contain both alleles, which correlates well with the microsatellite results (data not shown). DISCUSSION
As the sequencing of the human genome becomes complete, characterization and scoring of genetic variations will be increasingly important to correlate phenotype and genotype differences. Here, we have investigated the possibility of typing single-base variations in DNA by using a recently developed sequencing technique, called pyrosequencing, that allows for rapid real-time determination of 20 –30 base pairs of a target sequence. At present, pyrosequencing is performed in an automated microtiter-based pyrosequencer instrument, which allows simultaneous analysis of 96 samples within 10 min. Each round of nucleotide dispensing takes approximately 1 min and thus offers a rapid way to determine the exact sequence of the SNPs, including adjacent positions as a control. A unique property with pyrosequencing in typing SNPs is that each allele combination (homozygous, heterozygous) will give a specific pattern compared to the two other variants (Figs. 2 and 3). This feature makes typing extremely accurate and easy. Simple manual comparison of predicted SNP patterns and the obtained raw data from the pyrosequencer can score a
SNP, especially as no editing is needed. Because specific patterns can readily be achieved for the individual SNPs, it will also be possible to automatically score the allelic status by a pattern recognition software with the basic concept shown in Figs. 3 and 4. However, this could be further developed into a color-coded design for the different allelic combinations for an even more rapid interpretation. The predicted pyrosequencing pattern depends on the order of nucleotide additions. Thus, the user has the possibility of changing the order of nucleotide additions if distinguishing of the patterns is not satisfactory. In all SNPs analyzed in this work, we have tested two different order of nucleotide additions, cyclic and sequential (Fig. 2). In both cases, SNP determination starts with analysis of the nucleotide(s) that precedes the investigated position. This step will function as a positive control of the amplification process as well as a calibration of the reaction conditions. The advantage of using cyclic addition of nucleotides (i.e., repeated addition of CGAT) is that it generates a complete pattern difference along the sequence of the three variants of the SNP. In contrast, the sequential nucleotide addition (i.e., SNP-specific dispensing order) generates differences in three peak positions and can then be designed so that the individual allele extensions are in phase. Hereby further nucleotide additions will give the consensus sequence of the target and can improve raw-data interpretation. Another advantage of using sequential nucleotide addition is that the pyrosequencing will be finished twice as fast compared to the cyclic protocol. For loss of heterozygosity (LOH) analysis it is also clear that pyrosequencing can be useful, although the analysis will vary, dependent on the purity of the tumor samples. In order to better distinguish between the allelic versions, especially if the sample represents a mixture of tumor and normal cells, the pattern recognition software needs to be adapted to display “in silico” mixtures of allelic variants. An important feature of pyrosequencing for typing SNPs is that the technique is very rapid, making it feasible for large-scale studies (one microtiter plate can be analyzed and processed in 5–7 min using the sequential protocol). Consequently, the limiting factor is not the SNP analysis but rather the template preparation. In order to score thousands of SNPs simultaneously, novel approaches are needed. In this study, we used a multiplex PCR with specific primers for amplification of four SNPs. However, multiplex PCR amplification of 46 SNPs has been reported by using constant sequence tags at the 5⬘-ends of each primer (1). In conclusion, this work shows that typing of SNPs can efficiently be performed by pyrosequencing by using an automated system for parallel processing and pattern recognition software.
SINGLE-NUCLEOTIDE POLYMORPHISM ANALYSIS BY PYROSEQUENCING
109
FIG. 5. Loss of heterozygosity (LOH) analysis by pyrosequencing. (a) Sequence for codon 72 SNP when it is heterozygous (middle panel), when the allele containing the G variant is deleted (upper panel), and when the allele containing the C variant is lost (lower panel). (b) Predicted pyrosequencing pattern for the three variants in (a) when a sequential addition of nucleotides (CGCGAG) is performed. (c) Pyrosequencing raw data for the normal heterozygous sample and the corresponding tumor tissue with loss of haplotype G in about 50% of the cells.
REFERENCES 1. Wang, D. G., Fan, J. B., Siao, C. J., Berno, A., Young, P., Sapolsky, R., Ghandour, G., Perkins, N., Winchester, E., Spencer, J., Kruglyak, L., Stein, L., Hsie, L., Topaloglou, T., Hubbell, E., Robinson, E., Mittmann, M., Morris, M. S., Shen, N., Kilburn, D., Rioux, J., Nusbaum, C., Rozen, S., Hudson, T. J., Lipshutz, R., Chee, M., and Lander, E. S. (1998) Large-scale identification, mapping, and genotyping of single-nucleotide polymorphisms in the human genome. Science 280, 1077–1082. 2. Landegren, U., Nilsson, M., and Kwok, P. Y. (1998) Reading bits of genetic information: Methods for single-nucleotide polymorphism analysis. Genome Res. 8, 769 –776. 3. Fodor, S. P., Read, J. L., Pirrung, M. C., Stryer, L., Lu, A. T., and Solas, D. (1991) Light-directed, spatially addressable parallel chemical synthesis. Science 251, 767–773. 4. Southern, E. M., Maskos, U., and Elder, J. K. (1992) Analyzing and comparing nucleic acid sequences by hybridization to arrays of oligonucleotides: Evaluation using experimental models. Genomics 13, 1008 –1017. 5. Yershov, G., Barsky, V., Belgovskiy, A., Kirillov, E., Kreindlin, E., Ivanov, I., Parinov, S., Guschin, D., Drobishev, A., Dubiley, S., and Mirzabekov, A. (1996) DNA analysis and diagnostics on oligonucleotide microchips. Proc. Natl. Acad. Sci. USA 93, 4913– 4918. 6. Shuber, A. P., Michalowsky, L. A., Nass, G. S., Skoletsky, J., Hire, L. M., Kotsopoulos, S. K., Phipps, M. F., Barberio, D. M., and Klinger, K. W. (1997) High throughput parallel analysis of hundreds of patient samples for more than 100 mutations in multiple disease genes. Hum. Mol. Genet. 6, 337–347.
7. Nilsson, M., Krejci, K., Koch, J., Kwiatkowski, M., Gustavsson, P., and Landegren, U. (1997) Padlock probes reveal single-nucleotide differences, parent of origin and in situ distribution of centromeric sequences in human chromosomes 13 and 21. Nat. Genet. 16, 252–255. 8. Lee, L. G., Connell, C. R., and Bloch, W. (1993) Allelic discrimination by nick-translation PCR with fluorogenic probes. Nucleic Acids Res. 21, 3761–3766. 9. Livak, K. J., Marmaro, J., and Todd, J. A. (1995) Towards fully automated genome-wide polymorphism screening. Nat. Genet. 9, 341–342. 10. Tyagi, S., and Kramer, F. R. (1996) Molecular beacons: Probes that fluoresce upon hybridization. Nat. Biotechnol. 14, 303–308. 11. Tyagi, S., Bratu, D. P., and Kramer, F. R. (1998) Multicolor molecular beacons for allele discrimination. Nat. Biotechnol. 16, 49 –53. 12. Howell, W. M., Jobs, M., Gyllensten, U., and Brookes, A. J. (1999) Dynamic allele-specific hybridization. A new method for scoring single nucleotide polymorphisms. Nat. Biotechnol. 17, 87– 88. 13. Chen, X., and Kwok, P. Y. (1997) Template-directed dye-terminator incorporation (TDI) assay: A homogeneous DNA diagnostic method based on fluorescence resonance energy transfer. Nucleic Acids Res. 25, 347–353. 14. Syva¨nen, A. C. (1999) From gels to chips: “Minisequencing” primer extension for analysis of point mutations and single nucleotide polymorphisms. Hum. Mutat. 13, 1–10. 15. Laken, S. J., Jackson, P. E., Kinzler, K. W., Vogelstein, B., Strickland, P. T., Groopman, J. D., and Friesen, M. D. (1998)
110
AHMADIAN ET AL.
Genotyping by mass spectrometric analysis of short DNA fragments. Nat. Biotechnol. 16, 1352–1356. 16. Ross, P., Hall, L., Smirnov, I., and Haff, L. (1998) High level multiplex genotyping by MALDI-TOF mass spectrometry. Nat. Biotechnol. 16, 1347–1351. 17. Ronaghi, M., Uhle´n, M., and Nyre´n, P. (1998) A sequencing method based on real-time pyrophosphate. Science 281, 363– 365. 18. Berg, C., Hedrum, A., Holmberg, A., Ponte´n, F., Uhle´n, M., and Lundeberg, J. (1995) Direct solid-phase sequence analysis of the human p53 gene by use of multiplex polymerase chain reaction
and alpha-thiotriphosphate nucleotides. Clin. Chem. 41, 1461– 1466. 19. Karamohamed, S., Nilsson, J., Nourizad, K., Ronaghi, M., Pettersson, B., and Nyre´n, P. (1999) Production, purification, and luminometric analysis of recombinant Saccharomyces cerevisiae MET3 adenosine triphosphate sulfurylase expressed in Escherichia coli. Protein Expr. Purif. 15, 381–388. 20. Williams, C., Norberg, T., Ahmadian, A., Ponten, F., Bergh, J., Ingana¨s, M., Lundeberg, J., and Uhle´n, M. (1998) Assessment of sequence-based p53 gene analysis in human breast cancer: Messenger RNA in comparison with genomic DNA targets. Clin. Chem. 44, 455– 462.
Single-Nucleotide Polymorphism Analysis by Pyrosequencing Afshin Ahmadian, Baback Gharizadeh, Anna C. Gustafsson, Fredrik Sterky, Pål Nyre´n, Mathias Uhle´n, and Joakim Lundeberg 1 Department of Biotechnology, The Royal Institute of Technology (KTH), SE-100 44 Stockholm, Sweden
Received October 6, 1999
There is a growing demand for high-throughput methods for analysis of single-nucleotide polymorphic (SNP) positions. Here, we have evaluated a novel sequencing approach, pyrosequencing, for such purposes. Pyrosequencing is a sequencing-by-synthesis method in which a cascade of enzymatic reactions yields detectable light, which is proportional to incorporated nucleotides. One feature of typing SNPs with pyrosequencing is that each allelic variant will give a unique sequence compared to the two other variants. These variants can easily be distinguished by a pattern recognition software. The software displays the allelic alternatives and allows for direct comparison with the pyrosequencing raw data. For optimal determination of SNPs, various protocols of nucleotide dispensing order were investigated. Here, we demonstrate that typing of SNPs can efficiently be performed by pyrosequencing using an automated system for parallel analysis of 96 samples in approximately 5 min, suitable for large-scale screening and typing of SNPs. © 2000 Academic Press
Key Words: SNP; DNA sequencing; pyrosequencing.
Genetic variation is the basis for human diversity and plays an important role in human diseases. Methods to screen and map genetic variability have, for more than two decades, been based on restriction fragment length polymorphism and microsatellite markers. More recent efforts have focused on the most common type of human genetic variation, single-nucleotide polymorphisms (SNPs). 2 A position is referred to as a 1 To whom correspondence should be addressed: Fax: ⫹46 8 245452. E-mail: [email protected]. 2 Abbreviations used: SNP, single-nucleotide polymorphism; LOH, loss of heterozygosity; ASO, allele-specific oligonucleotide; PP i, pyrophosphate; ATP, adenosine triphosphate; APS, adenosine 5⬘-phosphosulfate.
0003-2697/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.
SNP when it exists in at least two variants with a frequency of more than 1% for the least common alternative (1). SNPs are distributed across the human genome by an approximate average of 1 SNP per 1000 base pairs (1, 2). As for microsatellite markers, SNPs can be used in linkage studies for identifying disease genes, in clinical genetic testing, in forensics, and for determination of loss of heterozygosity (LOH). The properties that make SNP analysis preferable compared to microsatellites are that SNPs are more prevalent than microsatellites and that many SNPs are located within the genes, directly affecting the gene product (protein). As the number of identified SNPs increases, there will be an increasing demand for efficient methods to type and assess the biological impact of this kind of genetic variation. The golden standard method for SNP scoring has been conventional Sanger DNA sequencing. However, sequencing based on gel electrophoresis generates more information than necessary, is time-consuming, is laborious, and requires labeling. Significant efforts have been made to improve SNP analysis with alternative techniques. Many of these techniques use allele-specific oligonucleotide (ASO) hybridization to discriminate between allelic variants such as high-density microarray chips (3– 6), padlock probes (7), or allelic discrimination during PCR (8 –12). Minisequencing approaches are an attractive alternative because of the direct interrogation of the variable position by DNA polymerase extension using different means of detection of the extended product (i.e., fluorescent and radioactive dyes or by mass spectrometry) (13–16). Recently, an alternative technique for DNA sequencing, called pyrosequencing, was described (17). Using a four-enzyme mixture, this sequencing-by-synthesis method relies on the luminometric detection of pyrophosphate (PP i) released upon nucleotide incorporation. Here, we have investigated the possible use of pyrosequencing for SNP analysis. The technique has 103
104
AHMADIAN ET AL. TABLE 1
List of the Primers Used for PCR Amplification and Pyrosequencing a wiaf1764
wiaf797
wiaf41
p53 codon 72
outercod innercod outerrev innerrev seqrev outercod innercod outerrev innerrev seqcod outercod innercod outerrev innerrev seqrev outercod innercod outerrev innerrev seqrev
AGTGAAAACATTGAAAACACA TCCAATGTGTGAAAAATATATAC AATGTTTTCACTGTCATAAAG AGAACACATACGTTTTACCA CATTTGTTAAGCTTTT TGTTTTAAGTTGCAGAGATG GAAGCTGAAGTTCAATCTTT AAATGGAAGCTCTTCAATAC GCTCTTCAATACTTAAGGTA GTTTTCTGTTGTAAATGC ACAGCATGCTTAGTTCTCT CATTTACACCACACTGCAAT AGACTTAATCCCTTGAATTG TTGACAGAGAAATAACGGG GTAGTAAAACCTGGATTA ATGCTGTCCCCGGACGA TCCAGATGAAGCTCCCAG CAGGAGGGGGCTGGTG AGGGGCCGCCGGTGTA GCTGCTGGTGCAGGGGCCA
a Outer, primers used in the outer PCR; inner, primers used in the inner PCR; cod, upstream; rev, downstream. All innercod primers are biotinylated at the 5⬘-end. Seqrev, pyrosequencing primers used on the immobilized single-stranded DNA; seqcod, pyrosequencing primer used on the eluted single-stranded DNA.
been applied on four SNPs located on chromosomes 9q and 17p. MATERIALS AND METHODS
Sample, PCR, and Template Preparation Human genomic DNA was extracted from 24 unrelated individuals. An outer multiplex PCR was performed to amplify four SNPs located on chromosomes 9q and 17p. The outer multiplex PCR (94°C for 1 min, 50°C for 40 s, and 72°C for 2 min for 35 cycles) was followed by individual inner PCRs (94°C for 1 min, 50°C for 40 s, and 72°C for 1 min for 35 cycles), generating ⬃80-bp fragments for each SNP. SNPs on chromosome 9 (wiaf1764, wiaf797, and wiaf41) were taken from the Whitehead Institute/ MIT server (http//www.genome.wi.mit.edu/SNP/human/ maps/Chr9.ALL.html). The SNP located on chromosome 17 was amplified from exon 4 in the p53 gene (codon 72). Primers for outer and inner PCR are listed in Table 1. One of the inner primers in the respective set was biotinylated at the 5⬘-end to allow immobilization. The outer and inner amplification mixture consisted of 10 mM Tris–HCl (pH 8.3), 2 mM MgCl2, 50 mM KCl, 0.1% (v/v) Tween 20, 0.2 mM dNTPs, 0.1 M of each primer, and 1 unit of AmpliTaq DNA polymerase (Perkin-Elmer, Norwalk, CT) in a total volume of 50 l. Five microliters of total human DNA (1 ng/l), isolated from blood samples of normal individuals, was used as outer PCR template. Forty microliters of biotinylated inner PCR products was immobilized onto streptavidin-coated superparamag-
netic beads (Dynabeads M280; Dynal, Oslo, Norway). Single-stranded DNA was obtained by incubating the immobilized PCR product in 10 l of 0.1 M NaOH for 5 min. The supernatant containing the eluted strand was neutralized by adding 5 l of 0.2 M HCl. The immobilized strand was resuspended in 8 l of H 2O. The eluted strand was used to pyrosequence wiaf797 and the immobilized strand to pyrosequence wiaf1764, wiaf41, and codon 72. One microliter of annealing buffer (100 mM Tris–acetate, pH 7.75, 20 mM Mg–acetate) and 0.1 M of respective sequencing primers (Table 1) were added to the singlestranded templates in a total volume of 10 l. Hybridization was performed by incubation at 94°C for 20 s, 65°C for 2 min, and then cooling to room temperature. Conventional DNA Sequencing Exon 4 of the p53 gene was amplified as described earlier (18) and sequenced by cycle sequencing with dye-labeled terminators (BigDye Terminator, PerkinElmer). The data were analyzed on a ABI PRISM 377 XL DNA sequencer (PE Applied Biosystems, Foster City, CA). Pyrosequencing Pyrosequencing was performed at 25°C in a volume of 50 l on an automated pyrosequencer instrument (kindly supplied by Pyrosequencing AB, Uppsala, Sweden). Single-stranded DNA with annealed sequence primer (the substrate) was added to the pyrosequencing reaction mixture containing 10 U of exonucleasedeficient (exo ⫺) Klenow DNA polymerase (Amersham Pharmacia Biotech, Uppsala, Sweden), 40 mU of apyrase (Sigma Chemical Co., St. Louis, MO), 4 g of purified luciferase/ml (BioThema, Dalaro¨, Sweden), 15 mU of recombinant ATP sulfurylase (19), 0.1 M Tris– acetate (pH 7.75), 0.5 mM EDTA, 5 mM Mg–acetate, 0.1% (w/v) bovine serum albumin (BioThema), 1 mM dithiothreitol, 10 M adenosine 5⬘-phosphosulfate (APS), 0.4 mg of poly(vinylpyrrolidone)/ml (360,000), and 100 g of D-luciferin/ml (BioThema). Stepwise elongation of the primer strand upon sequential addition of the different deoxynucleoside triphosphates (Amersham Pharmacia Biotech) and degradation of excess nucleotides by apyrase were carried out simultaneously. RESULTS
Genomic DNA from 24 unrelated individuals was used for determination of four single-nucleotide polymorphism sites by pyrosequencing. Three of these SNPs correspond to variations in chromosome 9 (wiaf1764, wiaf797, wiaf41) and one corresponds to a coding SNP in the p53 tumor suppressor gene (codon 72). The principle of the pyrosequencing technique is illustrated in Fig. 1. Various protocols of nucleotide
SINGLE-NUCLEOTIDE POLYMORPHISM ANALYSIS BY PYROSEQUENCING
105
FIG. 1. Principle of pyrosequencing. The reaction mixture consists of a single-stranded DNA with a short annealed primer, DNA polymerase, ATP sulfurylase, luciferase, and apyrase. The four nucleotide bases are added to the mixture in a defined order, i.e., CGAT (the arrow above the nucleotides indicates cyclic addition). (a, b) If the added nucleotide forms a base pair, the DNA polymerase incorporates the nucleotide and pyrophosphate will consequently be released. The released pyrophosphate will then be converted to ATP by ATP sulfurylase. Luciferase uses the ATP to generate detectable light. The excess of the added nucleotide will be degraded by apyrase. (c) If the added nucleotide does not form a base pair to the DNA template, the polymerase will not incorporate it and no light will be produced. The nucleotide will rapidly be degraded by apyrase. (d) Pyrosequencing raw data obtained for the case discussed above (SNP on p53 codon 72, homozygous G).
dispensing order were investigated for pyrosequencing of these SNPs. In addition, conventional DNA sequencing, used for p53 codon 72 (exon 4), has been included for comparison. The possibility of detecting loss of heterozygosity by SNP analysis was also examined using a breast cancer tumor. Pyrosequencing at Codon 72: Cyclic Protocol Template DNA prepared from blood samples was used in an outer multiplex amplification and was followed by four separate amplifications of the individual SNPs using a biotinylated primer. The biotinylated inner PCR products were separately immobilized onto streptavidin-coated superparamagnetic beads. Singlestranded DNA was obtained by incubating the immobilized PCR product in NaOH. The immobilized strand was resuspended in H 2O and was used for pyrosequencing. The coding single-nucleotide polymorphism at codon 72 involves either a G or C residue (Fig. 2a),
corresponding to amino acid proline (CCC) or arginine (CGC). The results using a cyclic addition of nucleotides (CGAT) are shown in Fig. 2b for the three possible variants of the SNP. The sequencing primer was designed so that the first incorporated nucleotide was identical in all variants, functioning as an internal control. Successful incorporation is expected by addition of the first nucleotide (nucleotide C), set to correspond to 1 peak equivalent (incorporation in only one of two alleles corresponds to 0.5 peak equivalents). In the case of homozygous C (top panel, b), the internal calibration peak is followed by a peak indicating incorporation of six nucleotides when the nucleotide G is added to the reaction. The subsequent sequence shows the consensus sequence (one A, one G, one C, and one A). Note that no signal is generated when noncomplementary nucleotides are added. In the heterozygous situation (middle panel, b), after one C, a peak equivalent to 3.5 G appears (six incorporations on one allele
106
AHMADIAN ET AL.
FIG. 2. Pyrosequencing on the coding SNP on the p53 gene (codon 72). (a) Sequence for the SNP. The arrows show primer position. The underlined nucleotides indicate the SNP site (C/G) corresponding to amino acid proline (CCC) or arginine (CGC). (b) Pyrosequencing result using cyclic addition of nucleotides (CGAT). (c) Pyrosequencing result using sequential (SNP specific) addition of nucleotides (see text for detailed explanation for (b) and (c). (d) Sequencing data obtained by conventional DNA sequencing.
plus one incorporation on the other allele). Following this addition, extension on the separate alleles will be out of phase. When the SNP position is homozygous G (lower panel, b), the pyrosequence will follow the consensus sequence in phase as the complementary nucleotides are added. All three possible variants of the SNP in codon 72 were identified among the 24 individuals (G, 14 of 24; G/C, 9 of 24; and C, 1 of 24). Pyrosequencing at Codon 72: Sequential Protocol Because the sequence context of the SNP is known, it is possible to reduce the number of nucleotide additions by using a sequential (SNP specific) dispensing protocol. Figure 2c shows the sequential addition of nucleotides CGCGAG for the same three variants of the SNP at codon 72 (top, middle, and lower panels). As in the cyclic protocol, the first nucleotide addition (C) normalizes the peak values. The second nucleotide addition (G) yields the same pattern as for the cyclic additions: 6, 3.5, and 1 base peak equivalent incorporation for the homozygous C, heterozygous C/G, and homozygous G, respectively. However, with the sequential procedure, it is possible to elaborate with the dispensing order to achieve synchronized extensions after the variable po-
sition. Here, this was achieved by addition of nucleotide C to extend the lagging allele in the heterozygous case followed by addition of nucleotide G. Hereby the extensions will be synchronized in all cases after the actual SNP position. The two remaining additions of nucleotides A and G will therefore function as postcontrols. The results obtained using a conventional DNA sequencing procedure on the three variants of the SNP in codon 72 of the p53 gene are shown in Fig. 2d. The homozygous alternatives are clearly distinguished (top and lower panels), while the heterozygous state is more difficult to interpret (middle panel, d). Actually, the automatic base-call algorithm (ABI PRISM 377 XL) scores the middle panel as being homozygous C. Analysis of wiaf1764, wiaf797, and wiaf41 As shown above, the allelic variants of a SNP determined by pyrosequencing can easily be distinguished by pattern recognition using either cyclic or sequential nucleotide addition. By taking advantage of a simple pattern recognition program, the allelic alternatives can be displayed. The program presents predicted peaks for the individual SNPs. Comparison between raw data and predicted patterns was made manually.
SINGLE-NUCLEOTIDE POLYMORPHISM ANALYSIS BY PYROSEQUENCING
107
FIG. 3. Pyrosequencing on a SNP positioned on chromosome 9q (wiaf1764). (a) Sequence for the three allelic alternatives. (b) Predicted pyrosequencing pattern for the three variants of the SNP when a cyclic addition of nucleotides (GTCA) is performed. (c) Raw data obtained by pyrosequencing for the three variants. As it is shown, the raw data completely match the predicted pyrosequencing pattern.
The user can hereby test alternative nucleotide dispensing orders to choose the most informative alternative. This was evaluated on SNPs positioned on chromosome 9 (wiaf1764, wiaf797, and wiaf41). For wiaf1764 all three SNP variants were observed among the 24 individuals (A, 12 of 24; A/C, 8 of 24; and C, 4 of 24). Figure 3 displays the three allelic alternatives (Fig. 3a), predicted pyrosequencing pattern (Fig. 3b), and the three examples of pyrosequencing raw data (Fig. 3c) using the cyclic protocol (GTCA). The top and lower panels show the two homozygous alternatives (C and A) and the middle panel corresponds to the heterozygous variant (C/A). The similarities between raw data and the three predicted patterns allow for direct and reliable determination of the SNP. Occasionally, reagents will contain minor contaminations of pyrophosphate that will yield a low background signal that can be subtracted prior to analysis. The obtained data for wiaf797 and wiaf41 showed a low frequency of variability (Fig. 4). In the case of wiaf41 (Fig. 4a), two allelic variants were found (A, 22 of 24; A/T, 2 of 24; and T, 0 of 24). In the case of wiaf797, only one variant was found (Fig. 4b) among the 24 individuals (G, 0 of 24; G/T, 0 of 24; and T, 24 of 24).
Analysis of Loss of Heterozygosity Analysis of loss of heterozygosity (LOH) plays an important role in the process of characterizing tumor suppressor genes. In the case of informative (heterozygous) SNPs, these may be used for such investigations as a complement to the more informative microsatellites. From a breast cancer study (20) we selected a codon 72 heterozygous sample. This sample, which showed loss of one allele by distal microsatellite markers, was evaluated for loss of heterozygosity using the pyrosequencing protocol. Genomic DNA was amplified and codon 72 was determined by a sequential protocol (CGCGAG) as outlined above. Figure 5 shows the sequence surrounding the SNP (Fig. 5a), the predicted alternatives for allelic loss (Fig. 5b), and the obtained data from normal and cancer tissues (Fig. 5c). As expected, the patient was shown to be heterozygous by pyrosequencing in the normal tissue sample (middle panel). The obtained pyrosequencing data for the tumor sample (Fig. 5c) show a high pattern similarity with the predicted pattern for a sample with loss of haplotype G (top panel). The deviations between the predicted pattern and the raw data are due to the
108
AHMADIAN ET AL.
FIG. 4. Pyrosequencing on SNPs wiaf41 (a) and wiaf797 (b). Raw data (on right) are compared to the predicted patterns (on left).
heterogeneous nature of tumor samples, having admixed “contaminating” normal cells. In this case about 50% of the cells contain both alleles, which correlates well with the microsatellite results (data not shown). DISCUSSION
As the sequencing of the human genome becomes complete, characterization and scoring of genetic variations will be increasingly important to correlate phenotype and genotype differences. Here, we have investigated the possibility of typing single-base variations in DNA by using a recently developed sequencing technique, called pyrosequencing, that allows for rapid real-time determination of 20 –30 base pairs of a target sequence. At present, pyrosequencing is performed in an automated microtiter-based pyrosequencer instrument, which allows simultaneous analysis of 96 samples within 10 min. Each round of nucleotide dispensing takes approximately 1 min and thus offers a rapid way to determine the exact sequence of the SNPs, including adjacent positions as a control. A unique property with pyrosequencing in typing SNPs is that each allele combination (homozygous, heterozygous) will give a specific pattern compared to the two other variants (Figs. 2 and 3). This feature makes typing extremely accurate and easy. Simple manual comparison of predicted SNP patterns and the obtained raw data from the pyrosequencer can score a
SNP, especially as no editing is needed. Because specific patterns can readily be achieved for the individual SNPs, it will also be possible to automatically score the allelic status by a pattern recognition software with the basic concept shown in Figs. 3 and 4. However, this could be further developed into a color-coded design for the different allelic combinations for an even more rapid interpretation. The predicted pyrosequencing pattern depends on the order of nucleotide additions. Thus, the user has the possibility of changing the order of nucleotide additions if distinguishing of the patterns is not satisfactory. In all SNPs analyzed in this work, we have tested two different order of nucleotide additions, cyclic and sequential (Fig. 2). In both cases, SNP determination starts with analysis of the nucleotide(s) that precedes the investigated position. This step will function as a positive control of the amplification process as well as a calibration of the reaction conditions. The advantage of using cyclic addition of nucleotides (i.e., repeated addition of CGAT) is that it generates a complete pattern difference along the sequence of the three variants of the SNP. In contrast, the sequential nucleotide addition (i.e., SNP-specific dispensing order) generates differences in three peak positions and can then be designed so that the individual allele extensions are in phase. Hereby further nucleotide additions will give the consensus sequence of the target and can improve raw-data interpretation. Another advantage of using sequential nucleotide addition is that the pyrosequencing will be finished twice as fast compared to the cyclic protocol. For loss of heterozygosity (LOH) analysis it is also clear that pyrosequencing can be useful, although the analysis will vary, dependent on the purity of the tumor samples. In order to better distinguish between the allelic versions, especially if the sample represents a mixture of tumor and normal cells, the pattern recognition software needs to be adapted to display “in silico” mixtures of allelic variants. An important feature of pyrosequencing for typing SNPs is that the technique is very rapid, making it feasible for large-scale studies (one microtiter plate can be analyzed and processed in 5–7 min using the sequential protocol). Consequently, the limiting factor is not the SNP analysis but rather the template preparation. In order to score thousands of SNPs simultaneously, novel approaches are needed. In this study, we used a multiplex PCR with specific primers for amplification of four SNPs. However, multiplex PCR amplification of 46 SNPs has been reported by using constant sequence tags at the 5⬘-ends of each primer (1). In conclusion, this work shows that typing of SNPs can efficiently be performed by pyrosequencing by using an automated system for parallel processing and pattern recognition software.
SINGLE-NUCLEOTIDE POLYMORPHISM ANALYSIS BY PYROSEQUENCING
109
FIG. 5. Loss of heterozygosity (LOH) analysis by pyrosequencing. (a) Sequence for codon 72 SNP when it is heterozygous (middle panel), when the allele containing the G variant is deleted (upper panel), and when the allele containing the C variant is lost (lower panel). (b) Predicted pyrosequencing pattern for the three variants in (a) when a sequential addition of nucleotides (CGCGAG) is performed. (c) Pyrosequencing raw data for the normal heterozygous sample and the corresponding tumor tissue with loss of haplotype G in about 50% of the cells.
REFERENCES 1. Wang, D. G., Fan, J. B., Siao, C. J., Berno, A., Young, P., Sapolsky, R., Ghandour, G., Perkins, N., Winchester, E., Spencer, J., Kruglyak, L., Stein, L., Hsie, L., Topaloglou, T., Hubbell, E., Robinson, E., Mittmann, M., Morris, M. S., Shen, N., Kilburn, D., Rioux, J., Nusbaum, C., Rozen, S., Hudson, T. J., Lipshutz, R., Chee, M., and Lander, E. S. (1998) Large-scale identification, mapping, and genotyping of single-nucleotide polymorphisms in the human genome. Science 280, 1077–1082. 2. Landegren, U., Nilsson, M., and Kwok, P. Y. (1998) Reading bits of genetic information: Methods for single-nucleotide polymorphism analysis. Genome Res. 8, 769 –776. 3. Fodor, S. P., Read, J. L., Pirrung, M. C., Stryer, L., Lu, A. T., and Solas, D. (1991) Light-directed, spatially addressable parallel chemical synthesis. Science 251, 767–773. 4. Southern, E. M., Maskos, U., and Elder, J. K. (1992) Analyzing and comparing nucleic acid sequences by hybridization to arrays of oligonucleotides: Evaluation using experimental models. Genomics 13, 1008 –1017. 5. Yershov, G., Barsky, V., Belgovskiy, A., Kirillov, E., Kreindlin, E., Ivanov, I., Parinov, S., Guschin, D., Drobishev, A., Dubiley, S., and Mirzabekov, A. (1996) DNA analysis and diagnostics on oligonucleotide microchips. Proc. Natl. Acad. Sci. USA 93, 4913– 4918. 6. Shuber, A. P., Michalowsky, L. A., Nass, G. S., Skoletsky, J., Hire, L. M., Kotsopoulos, S. K., Phipps, M. F., Barberio, D. M., and Klinger, K. W. (1997) High throughput parallel analysis of hundreds of patient samples for more than 100 mutations in multiple disease genes. Hum. Mol. Genet. 6, 337–347.
7. Nilsson, M., Krejci, K., Koch, J., Kwiatkowski, M., Gustavsson, P., and Landegren, U. (1997) Padlock probes reveal single-nucleotide differences, parent of origin and in situ distribution of centromeric sequences in human chromosomes 13 and 21. Nat. Genet. 16, 252–255. 8. Lee, L. G., Connell, C. R., and Bloch, W. (1993) Allelic discrimination by nick-translation PCR with fluorogenic probes. Nucleic Acids Res. 21, 3761–3766. 9. Livak, K. J., Marmaro, J., and Todd, J. A. (1995) Towards fully automated genome-wide polymorphism screening. Nat. Genet. 9, 341–342. 10. Tyagi, S., and Kramer, F. R. (1996) Molecular beacons: Probes that fluoresce upon hybridization. Nat. Biotechnol. 14, 303–308. 11. Tyagi, S., Bratu, D. P., and Kramer, F. R. (1998) Multicolor molecular beacons for allele discrimination. Nat. Biotechnol. 16, 49 –53. 12. Howell, W. M., Jobs, M., Gyllensten, U., and Brookes, A. J. (1999) Dynamic allele-specific hybridization. A new method for scoring single nucleotide polymorphisms. Nat. Biotechnol. 17, 87– 88. 13. Chen, X., and Kwok, P. Y. (1997) Template-directed dye-terminator incorporation (TDI) assay: A homogeneous DNA diagnostic method based on fluorescence resonance energy transfer. Nucleic Acids Res. 25, 347–353. 14. Syva¨nen, A. C. (1999) From gels to chips: “Minisequencing” primer extension for analysis of point mutations and single nucleotide polymorphisms. Hum. Mutat. 13, 1–10. 15. Laken, S. J., Jackson, P. E., Kinzler, K. W., Vogelstein, B., Strickland, P. T., Groopman, J. D., and Friesen, M. D. (1998)
110
AHMADIAN ET AL.
Genotyping by mass spectrometric analysis of short DNA fragments. Nat. Biotechnol. 16, 1352–1356. 16. Ross, P., Hall, L., Smirnov, I., and Haff, L. (1998) High level multiplex genotyping by MALDI-TOF mass spectrometry. Nat. Biotechnol. 16, 1347–1351. 17. Ronaghi, M., Uhle´n, M., and Nyre´n, P. (1998) A sequencing method based on real-time pyrophosphate. Science 281, 363– 365. 18. Berg, C., Hedrum, A., Holmberg, A., Ponte´n, F., Uhle´n, M., and Lundeberg, J. (1995) Direct solid-phase sequence analysis of the human p53 gene by use of multiplex polymerase chain reaction
and alpha-thiotriphosphate nucleotides. Clin. Chem. 41, 1461– 1466. 19. Karamohamed, S., Nilsson, J., Nourizad, K., Ronaghi, M., Pettersson, B., and Nyre´n, P. (1999) Production, purification, and luminometric analysis of recombinant Saccharomyces cerevisiae MET3 adenosine triphosphate sulfurylase expressed in Escherichia coli. Protein Expr. Purif. 15, 381–388. 20. Williams, C., Norberg, T., Ahmadian, A., Ponten, F., Bergh, J., Ingana¨s, M., Lundeberg, J., and Uhle´n, M. (1998) Assessment of sequence-based p53 gene analysis in human breast cancer: Messenger RNA in comparison with genomic DNA targets. Clin. Chem. 44, 455– 462.