Anal Bioanal Chem (2008) 391:405–413 DOI 10.1007/s00216-008-1988-x
ORIGINAL PAPER
Microarray-based detection of Korean-specific BRCA1 mutations Cheulhee Jung & Seong-Chun Yim & Dae-Yeon Cho & Ho Nam Chang & Hyun Gyu Park
Received: 3 January 2008 / Revised: 14 February 2008 / Accepted: 18 February 2008 / Published online: 28 March 2008 # Springer-Verlag 2008
Abstract A reliable multiplex assay procedure to detect human genetic mutations in the breast cancer susceptibility gene BRCA1 using zip-code microarrays and single base extension (SBE) reactions is described. Multiplex PCR amplification was performed to amplify the genomic regions containing the mutation sites. The PCR products were then employed as templates in subsequent multiplex SBE reactions using bifunctional primers carrying a unique complementary zip sequence in addition to a mutation-sitespecific sequence. The SBE primers, terminating one base before their mutation sites, were extended by a single base at a mutation site with a corresponding biotin-labeled ddNTP. Hybridization of the SBE products to zip-code microarrays was followed by staining with streptavidin– Cy3, leading to successful genotyping of several selected BRCA1 mutation sites with wild-type and heterozygote mutant samples from breast cancer patients. This work has led to the development of a reliable DNA microarray-based system for the diagnosis of human genetic mutations. Keywords DNA chip . BRCA . Mutation detection . Zip-code microarray . Single base extension
Cheulhee Jung and Seong-Chun Yim contributed equally to this work. C. Jung : S.-C. Yim : H. N. Chang : H. G. Park (*) Department of Chemical and Biomolecular Engineering, KAIST, 373–1 Guseong-dong, Yuseong-gu, Daejeon 305–701, Republic of Korea e-mail:
[email protected] D.-Y. Cho Research Director Clinical Research Institute Labgenomics Co. Ltd., Doo San Engineering Center, Seongbok Dong, Yong-In, 449–795 Gyeonggi, Republic of Korea
Introduction Mutations in the BRCA1 gene are associated with the development of breast and ovarian cancers. These mutations appear to be responsible for approximately 2–5% of all breast cancer cases in the general population [1–4]. Therefore, the BRCA1 gene has been extensively screened for mutations with BRCA2, and a database of the spectrum of mutations has been established [5, 6]. Up to now, more than 1,500 distinct variations of BRCA1 have been registered in the BIC database. Although the incidence is lower than in Western populations, breast cancer is also the most common cancer in Korean women. Owing to the rapid westernization of the Korean life style, the incidence of this disease and associated death rates have markedly increased. BRCA1 mutations are considered to be involved in a large portion of all familial breast cancers [7]. Since the prognosis for breast cancer survival is heavily dependent on the stage at which the disease is diagnosed, strategies for the reliable detection of the mutations are becoming increasingly important. Recently, array-based methods have attracted much attention as alternatives to conventional diagnostic methods for human genetic mutations owing to their unique advantages, including multiplex capability, rapid diagnostics, and low cost [8–13]. Array-based mutation detection methods typically rely on fluorescence detection of hybridization of PCRamplified products to allele-specific oligonucleotide capture probes immobilized on arrays. These probes can contain sequences that are complementary to both those of wild-type and mutant genes. Usually, the fluorescence intensity of each capture probe is intended to reveal which sequence is perfectly complementary to the PCR-amplified sequence, leading to correct genotyping at the mutation sites. By using this strategy, our group successfully identified the genotypes
406
for several mutation sites of disease-related human genes, including BRCA1 [14] and HNF-1α [15]. However, these direct hybridization approaches, in which PCR products containing mutation sites are directly hybridized to the capture probes affixed to arrays, have nonuniform success rates and some important limitations. One significant problem is the difficulty of reliably genotyping mutation sites especially for sequences that have localized regions of high GC or AT content [16]. Also, allele-specific direct hybridization methods require the use of different sets of capture probe arrays for each set of genetic mutations. Thus, time-consuming and technically demanding immobilization of capture probes is required and hybridization with PCR-amplified products must be optimized for each allele-specific microarray [17]. Recently, a zip-code microarray [18–20] strategy has been developed that successfully resolves these problems. This strategy usually employs single base extension (SBE) [21] or ligation detection method [22–24]. In particular, SBE by DNA polymerase can distinguish differences due to a single nucleotide with high accuracy and in a manner that is compatible with multiplex detection of mutations [18–20, 25]. With the zip-code microarray approach, mutation sites can be extended by a single base, using chimeric primers with 3′ complementarity to the specific mutation site and 5′ complementarity to the respective zip-code sequence on the microarray. By hybridizing the resulting SBE product on the zip-code microarray, one can achieve genotyping in a highly multiplexed fashion. Recently, we also demonstrated that the approach can be successfully used for the diagnosis of HNF1α mutations [21]. With the aim of expanding methods for the diagnosis of clinically important human genetic mutations, we have carried out studies which have led to the development of a reliable method for parallel detection of several selected Korean-specific mutations in the BRCA1 gene [5, 7]. By using a wild-type gene from an apparently healthy woman and mutant samples from breast cancer patients, we correctly genotyped several major Korean-specific mutations in BRCA1 exon 11 by employing a multiplex SBE reaction on a zip-code microarray. Especially significant is the fact that a novel approach was developed for the detection of mutations whose genotyping could not be achieved with a simple SBE reaction using one chimeric SBE primer. The results of this effort are described herein.
Materials and methods DNA isolation and nucleotide sequencing Genomic DNAwas isolated from whole blood of an apparently healthy subject and two patients suffering from breast cancer
Anal Bioanal Chem (2008) 391:405–413
by using a genomic DNA extraction kit (Qiagen, Hilden, Germany). After PCR amplification, direct sequencing was performed on these samples, using a BigDye terminator cycle sequencing kit (Applied Biosystems, Foster City, CA, USA) and the instructions provided by the manufacturer, in order to confirm their sequence information on the mutation sites. Oligonuleotides All oligonucleotides were synthesized and then purified by sodium dodecyl sulfate polyacrylamide gel electrophoresis by Bioneer (Seoul, Korea). The zip-code oligonucleotide probes used in the fabrication of zip-code microarrays were 12 bases long and 5′ aminohexyl-terminated (Table 1). 5′ Cy3-conjugated oligonucleotides were used as synthetic target probes. All nucleotides were quantified by their absorbances at 260nm on the basis of the calculated molar extinction coefficient and their identities were confirmed by matrix-assisted laser desorption/ionization time-of-flight mass-spectrometric analysis. Zip-code microarray manufacturing The aminated zip-code probes (Table 1) were diluted in 0.1 M carbonate buffer (pH 9.5) with 10% (v/v) dimethyl sulfoxide to make final concentrations of 100μM. The probes were spotted on aldehyde-activated glass slides (CEL Associates, Houston, TX, USA) using a microspotter (VersArray ChipWriter™ compact system, Bio-Rad Laboratories, USA), according to the manufacturer’s instructions, and allowed to stand overnight in a humid chamber. To verify reproducibility, four spots were printed for each oligonucleotide on the same glass slide. Then, the glass was treated according to the same procedures as described earlier [14]. PCR amplification Amplification of specific genomic regions containing the seven Korean-specific mutations (Table 2) was achieved by Table 1 Zip-code microarray sequences used in this study Zip no.
Sequence (5′–3′)
Zip1 Zip2 Zip3 Zip4 Zip5 Zip6 Zip7 Zip8 Zip9
Linker-TGCGGGTAATCG Linker-ATCGTGCGACCT Linker-GGTAATCGACCT Linker-ATCGGGTATGCG Linker-TGCGACCTATCG Linker-CAGCATCGTGCG Linker-GACCCAGCATCG Linker-ACCTGACCATCG Linker-GACCATCGACCT
Linker 6-aminohexyl linker
Anal Bioanal Chem (2008) 391:405–413
407
Table 2 Korean-specific mutations in BRCA1 exon 11 Region
Exon Exon Exon Exon Exon Exon Exon
11 11 11 11 11 11 11
Nucleotide change
Position
Wild type
Mutation type
C
T Deletion A Deletion A T T Insertion A T
G G C
1731 1942 2545 3033 3459 3746 4014
five PCR reactions (Table 3). Each PCR amplification was carried out using a PerkinElmer 9200 thermocycler (PerkinElmer, Norwalk, CT, USA) in a 50μl solution containing 100 ng of genomic DNA, 0.5μM of each primer, 1× PCR reaction buffer [10mM tris(hydroxymethyl)aminomethane hydrochloride (Tris-HCl), 40mM KCl, 1.5mM MgCl2], 0.2mM dNTPs, 0.16mM AA-dUTP, and 0.5 U Taq DNA polymerase (Bioneer, Seoul, Korea). PCR was programmed for 5min at 96 °C, followed by 35 cycles of 30 s at 96 °C, 30 s at 55 °C, 1min at 72 °C, and 5min at 72 °C. After amplification, the PCR products were purified with a QIAquick PCR purification kit (Qiagen) and amplification was confirmed by agarose-gel electrophoresis. Multiplex SBE reaction SBE reactions were carried out in 20μl solutions containing 8μl of PCR product, 100nM of each SBE primer (Table 4), 2 U of Thermosequenase (Amersham Bioscience, Uppsala, Sweden), 26mM Tris-HCl, 6.5mM MgCl2, 25μM specific biotin-labeled ddNTP and 100μM concentration of the other three ddNTPs (PerkinElmer, Norwalk, CT, USA). For
example, for GG allele detection, reaction of 25μM biotinlabeled ddGTP and 100μM ddCTP, ddATP, and ddUTP was performed. SBE reactions were carried out in a thermocycler (Applied Biosystems) with an initial polymerase activation at 96 °C for 5min, followed by 40 cycles at 96 °C for 30 s, 50 °C for 30 s, and 60 °C for 2min. The products of the SBE reactions were purified by using a nucleotide removal kit (Qiagen). The singleplex SBE reaction was performed in a total volume of 20μl containing 1μl of PCR product, 100nM SBE primer, 2 U Thermosequenase (Amersham Bioscience, Uppsala, Sweden), 26mM Tris-HCl, 6.5mM MgCl2, 2μM specific biotinlabeled ddNTP and 10μM concentration of each of the other three ddNTPs (PerkinElmer, Norwalk, CT, USA) Hybridization on zip-code array The SBE reaction products were denatured at 95 °C for 5min and cooled on ice for 5min, then allowed to hybridize on the zip-code array with 30μl of hybridization buffer containing 0.01% Triton X-100 and 1× SSPE (0.15 M NaCl, 10mM NaH2PO4, 1mM EDTA) at 37 °C for 12 h. After hybridization, the array was rinsed with washing buffer containing 0.005% Triton X-100 and 6× SSPE (0.9 M NaCl, 60mM NaH2PO4, 6mM EDTA, pH 7.4) for 10min at room temperature and then stained with 30μl staining solution containing 3μg/ml of streptavidin–Cy3 (SigmaAldrich, USA) and 0.01% Trion X-100 in 1× SSPE. Scanning and data analysis The microarrays were scanned with an Axon GenePix 4000B fluorescence reader (Axon, Union City, CA, USA). The laser power and photomultiplier tube were set at 100
Table 3 Primer sets for the preparation of PCR products Position
Amplicona
Primer sequence (5′–3′)
1731
1
1942
1
2545
2
3033
3
3459
4
3746
4
4014
5
Forward: GCAGATTTGGCAGTTCAAAAGACTC Reverse: GCATGAATATGCCTGGTAGAAGACTTCC Forward: GCAGATTTGGCAGTTCAAAAGACTC Reverse: GCATGAATATGCCTGGTAGAAGACTTCC Forward: GGCACAGCAGAAACCTACAACTCAT Reverse: TTGGAACAACCATGAATTAGTCCCT Forward: TGCCAAATGTAGTATCAAAGGAGGC Reverse: ATAGCATTCAATTTTGGCCCTCTGT Forward: TGGAAGTAATTGTAAGCATCCTGAAATAAAAA Reverse: GGGAAGCTCTTCATCCTCACTAGATAA Forward: CTGGAAGTAATTGTAAGCATCCTGAAATAAAAA Reverse: GGGAAGCTCTTCATCCTCACTAGATAA Forward: CAGGAACATCACCTTAGTGAGGAAAC Reverse: TGCTGCTTCACCTAAGTTTGAATCC
a
Amplicon 1 includes two mutation sites, 1731 and 1942, and amplicon 4 includes two mutation sites, 3459 and 3746.
408
Anal Bioanal Chem (2008) 391:405–413
Table 4 Single base extension primer sequences for the detection of mutations in BRCA1 Position
Name
Zip code
Primer sequencea (5′–3′)
1731 1942 2545 2545 3033 3459 3746 3746 4014
SBE01 SBE02 SBE03–1 SBE03–2 SBE04 SBE05 SBE06–1 SBE06–2 SBE07
1 2 3 4 5 6 7 8 9
CGATTACCCGCATCCTGAAATGATAAATCAGGGAACTAAC AGGTCGCACGATAATATCCACAATTCAAAAGCACCTAAAA AGGTCGATTACCTGTGAGTCAGTGTGCAGCATTTG CGCATACCCGATCATGAATTAGTCCCTTGGGGTTT CGATAGGTCGCAGAAACTGGACTCATTACTCCAAATAAACAT CGCACGATGCTGAGCATCCTGAAATAAAAAAGCAAGAATAT CGATGCTGGGTCCGAAGAGGGGCCAAGAAATT CGATGGTCAGGTCGAAGAGGGGCCAAGAAATTA AGGTCGATGGTCCAGTTCAGAAGGTTAAGTGACGTGA
a
The sequence complementary to the DNA zip-code sequence is underlined.
and 600, respectively. The 16-bit TIFF images obtained were analyzed with GenePix Pro 3.0 (Axon) provided with the instrument.
Results and discussion Genotyping strategy In Fig. 1 is shown the overall plan of our studies to determine genotypes using SBE reaction on zip-code Fig. 1 Mutation detection using single base extension (SBE) reactions on a zip-code microarrays
microarrays. Gene-specific primers were used in PCR amplification of a genomic region containing mutations. The amplification product was used as a template in subsequent SBE reactions. The SBE reactions were carried out using chimeric primers containing both a sequence covering the range up to one base before the mutation site and a sequence complementary to the corresponding unique zip-code sequence. The SBE primers are designed to be complementary to the antisense strand of double-stranded PCR products. The SBE reactions involve SBEs of the SBE primers at specific mutation points, using DNA polymerase
Anal Bioanal Chem (2008) 391:405–413
409
Optimization of the SBE reaction
in the presence of biotin-labeled ddNTP and three other ddNTPs in unlabeled forms. The resulting SBE products were hybridized to zip-code microarrays. The hybridized zip-code microarrays were then stained with streptavidin– Cy3 in order to obtain fluorescence images. As observed in our previous study [21], addition of the three unlabeled ddNTPs markedly reduces nonspecific incorporation of biotin-labeled ddNTPs and, as a result, signals are generated only from locations where specific incorporation of correct biotin-labeled ddNTP takes place (Fig. 2). In this plan, four separate reactions are used for genotyping, one with each of the four different biotinlabeled ddNTPs and three other ddNTPs in their unlabeled form. Although complicated for single genotyping, four separate reactions are sufficient for the multiple genotyping even though the number of mutation sites is increased. The only requirements for the multiple genotyping are that four SBE reactions be conducted in a multiplexed fashion using SBE primers for all the mutation sites and that the microarrays contain a correspondingly increased number of zip-code sequences. Zip-code DNA microarrays include unique 12-base oligonucleotides that have similar melting temperatues. Each of the zip-code sequences were composed of three tetramers [18] such that each tetramer has a sequence which differs from the others by at least two-base oligonucleotides (Table 1). All zip-code oligonucleotides carried an aminohexyl group at their 5′ ends, and were immobilized in defined positions on aldehyde-activated glass slides. These zip-code oligonucleotides were hybridized with their corresponding SBE primers containing complementary sequences. To examine specificity, the zip-code microarrays were first hybridized with fluorescently labeled synthetic oligonucleotides complementary to each specific zip-code probe. The synthetic oligonucleotides were 20-mers and conjugated with Cy3 dye at their 5′ ends. Even among the closely related zip-code oligonucleotides, cross-hybridization was not detected with a signal-to-noise ratio over 100. Fig. 2 Elimination of nonspecific SBE reactions by adding the three unlabeled ddNTPs. Here, the 2545 site was tested for the singleplex SBE reaction without unlabeled ddNTPs (a) and with unlabeled ddNTPs (b)
To determine the optimum conditions, SBE reactions were carried out at 2,545 mutation sites using SBE 03–1 primer and PCR product amplified from wild-type homozygous (AA) genomic DNA as a template. The effects of different amounts of PCR product, SBE primer, biotin-labeled ddNTP, and Thermosequenase on the final signal intensity for the zip-code microarray were determined. As shown in Fig. 3, similar trends were observed for the three components (PCR product, biotin-labeled ddNTP and Thermosequenase). At first, the hybridization signal was observed to increase in a manner proportional to the amounts of the three components. However, no significant additional increases were detected when the three components were increased further, indicating that saturation takes place. Interestingly, the effect of SBE primer shows a different pattern from that of the other three components. In this case, the signal intensity rapidly increases when up to 10nM SBE primer is added, but no significant change takes place in the concentration range from 10 to 100nM. A further increase of the concentration of SBE primer over 100nM results in a decrease of the signal intensity. This result can be ascribed to the fact that the excess SBE primers compete, in their original nonextended form, with the single base extended primers for the complementary zip-code sites. This leads to decreased signal intensities. Therefore, in this strategy the concentration of SBE primer should be carefully chosen to provide the optimum results. On the basis of the findings displayed in Fig. 3, the optimum conditions for SBE product formation involve the use of PCR product, SBE primer, biotin-labeled ddNTP, and Thermosequenase in respective quantities of 40 ng, 100 nM, 2 μM, and 2 U. Detection of mutations in BRCA1 by using SBE reactions and zip-code microarrays To demonstrate the potential advantages of the new strategy for accurate multiplex genotyping, seven mutation sites
2545* site
:
agtgtgcagcatttgaaaaccccaagggac
ddNTP-biotin
ddNTP-biotin and unlabeled ddNTPs
ddATP-biotin
ddCTP-biotin
ddATP-biotin
ddCTP-biotin
ddGTP-biotin
ddUTP-biotin
ddGTP-biotin
ddUTP-biotin
a
b
410
Anal Bioanal Chem (2008) 391:405–413 18000
bB
18000
Hybridization intensity
Hybridization intensity
20000
aA
16000 14000 12000 10000 8000 6000 4000 2000
16000 14000 12000 10000 8000 6000 4000 2000
0
0 0
10
20
30
40
50
60
70
80
0
50
PCR product (ng) 18000
cC
14000 12000 10000 8000 6000 4000 2000 0
150
200
6
8
dD
16000
Hybridization intensity
16000
100
SBE primer (nM)
18000
Hybridization intensity
Fig. 3 Optimization of SBE reactions. The hybridization intensities on the y-axis represent mean values of the fluorescence for each spot. Several experiments were preformed to find the optimum conditions for the SBE reaction at various concentrations of PCR products (a), SBE primers (b), biotin-labeled ddNTPs (c), and Thermosequenase units (d). Four independent experiments were performed and the results were averaged (each chip with a four-replicate set of probes). The experimental variations are indicated with error bars
14000 12000 10000 8000 6000 4000 2000
0
2
4
6
8
10
0
0
2
were selected from main Korean-specific BRCA1 mutations [26, 27]. Of the seven mutations selected, four were single base substitutions, two were single base deletions, and one was a single base insertion (Table 2). In the strategy we developed, one SBE primer is usually sufficient to determine the genotype at the mutation site for
4
Thermosequenase (units)
ddNTP-biotin (µM)
a single base substitution type. For some insertion or deletion mutations, however, the genotyping cannot be achieved only by using a single SBE primer. This situation is encountered especially when the same mononucleotide repeat sequences occur next to the mutation site. In those cases, the sequence of the mutant sample will remain the
a 2545 homozygote wild type sample First SBE primer (A/A) ddATP
dddCTP
ddGTP
ddUTP
Second SBE primer (T/T) ddATP
ddCTP
ddGTP
ddUTP
b 3746 homozygote wild type sample First SBE primer (A/A) ddATP
ddCTP
ddGTP
ddUTP
Second SBE primer (G/G) ddATP
ddCTP
ddGTP
ddUTP
c
d
1731 heterozygote mutant sample (C/T)
3033 heterozygote mutant sample (G/T)
ddATP
dddCTP
ddGTP
ddUTP
Fig. 4 Singleplex genotyping of the BRCA1 mutation sites using SBE reactions on zip-code microarrays with homozygote wild-type samples at the 2545 (a) and 3746 (b) positions and heterozygote single point mutation samples at the 1731 (c) and 3033 (d) positions. The images
ddATP
dddCTP
ddGTP
ddUTP
represent the fluorescence signals obtained from four different SBE reactions, each of which involved a different biotin-labeled ddNTP. The ddNTP mentioned above the image indicates that it was obtained from the reaction involving that biotin-labeled ddNTP
Anal Bioanal Chem (2008) 391:405–413
411
same as that of the mutation site within a wild-type sample regardless of the presence of deletions or insertions. Of the seven mutations probed in this study, two mutations at the 2545 and 3746 sites are examples of this situation. Since insertions or deletions are also frequently found in human genetic mutations, we executed a strategy to cover those types of mutations by employing a second SBE primer. The mutant sample with the mutation at the 3746 site has a single base A inserted next to the mutation base which is also A and the same ddATP will be extended at the mutation site regardless of whether or not the mutation is present when only a single SBE primer (SBE 06–1) is used in the SBE reaction. Therefore, we designed a second SBE primer (SBE 06–2) covering up to the mutation sequence, which can be single base extended at the point next to the mutation site. The SBE reaction with the second primer should generate a fluorescence signal only from ddGTP for the wild-type sample, while signals should be obtained from both ddATP and ddGTP for heterozygote mutant samples. By using this protocol, we readily accomplished genotyping at the 3746 mutation site on the basis of the
a
difference that results from the SBE reaction with the second primer. This type of second primer application is not applicable for genotyping the mutation at the 2545 site because the base sequences next to the mutation site are the same for both wild-type and mutant samples. To overcome this limitation, we designed an antisense SBE second primer, whose sequence is complementary to that of the sense strand of the sample DNA. With a wild-type sample, a signal should be generated only from the SBE reaction with ddUTP using the second primer, while both reactions with ddUTP and ddCTP should result in signals for the heterozygote mutant sample. By using this strategy, we successfully achieved genotyping at the 2545 and 3746 sites for wild-type samples (Fig. 4a,b). The genotyping for these mutation sites can be executed using only the second SBE primers because SBE reactions with the first primers always create the same signals for both types of samples. However, it is desirable to include SBE reactions with the first primer in order to confirm the reliability of this strategy.
ddATP
ddCTP
ddGTP
ddUTP
Wild-type sample (Homozygote) Zip 1 SBE01 (CC)
Zip 2 SBE02 (AA)
Zip 3 SBE03-1 (AA)
Zip 4 SBE03-2 (TT)
Zip 6 SBE05 (GG)
Zip 7 SBE06-1 (AA)
Zip 8 SBE06-2 (GG)
Zip 9 SBE07 (CC)
Zip 5 SBE04 (GG)
b
ddATP
ddCTP
1731 mutant sample (Heterozygote) Zip 1 SBE01 (CT)
Zip 2 SBE02 (AA)
Zip 3 SBE03-1 (AA)
Zip 4 SBE03-2 (TT)
Zip 6 SBE05 (GG)
Zip 7 SBE06-1 (AA)
Zip 8 SBE06-2 (GG)
Zip 9 SBE07 (CC)
Zip 5 SBE04 (GG)
c
ddGTP
ddUTP
ddATP
ddCTP
ddGTP
ddUTP
3033 mutant sample (Heterozygote) Zip 1 SBE01 (CC)
Zip 2 SBE02 (AA)
Zip 3 SBE03-1 (AA)
Zip 4 SBE03-2 (TT)
Zip 6 SBE05 (GG)
Zip 7 SBE06-1 (AA)
Zip 8 SBE06-2 (GG)
Zip 9 SBE07 (CC)
Zip 5 SBE04 (GT)
Fig. 5 Multiplex diagnosis of the BRCA1 mutations using SBE reactions on zip-code microarrays with one wild-type sample (a) and two mutant samples with mutations at the 1731 (b) and 3033 (c) positions. The zip-code microarrays are represented on the left; each
position in the 2×2 grid corresponds to an individual zip-code sequence and corresponding BRCA1 mutations. The images on the right represent the fluorescence signals obtained from four different multiplex SBE reaction products
412
Prior to performing multiplex SBE genotyping for multiple mutation sites in BRCA1, we first examined the specificity of our method for single point mutations at 1731 and 3033 with heterozygote mutant samples. As shown in Fig. 4c and d, signals were correctly obtained from both reactions with the corresponding two different ddNTPs, indicating that the samples are heterozygote mutants. Even though the two signal intensities were not exactly same for SBE reactions of each ddNTP, owing to differences in incorporation efficiencies, the two signals are sufficiently strong to enable correct genotyping with heterozygote samples. This is evidenced by the comparison of single signals with homozygote samples (Fig. 4a,b). For all the images displayed in Fig. 4, nonspecific signals were not detected, so the signal-to-noise ratios were greater than 20. Multiple genotyping using multiplex SBE reactions with zip-code arrays We also performed multiplex SBE reactions for the diagnosis of seven selected Korean-specific mutations in BRCA1 with one wild-type and two heterozygote mutant samples in a blind-study manner. The genomic DNA from the wild-type sample was isolated from the blood of an apparently healthy woman. Direct sequencing confirmed that the DNA had wild genotypes for all mutation sites. Two mutant DNA samples were isolated from breast cancer patients. Direct sequencing confirmed that these samples had BRCA1 mutations at the 1731 position and the 3033 position, respectively (Table 2), and wild-type alleles at all the other mutation sites. The seven mutation sites were divided into five different PCRs, which gave amplification products that were then pooled for use as templates for subsequent multiplex SBE reactions. Nine chimeric primers, corresponding to the nine zip-code sequences, were added to a single tube. The five mutations can be covered by five primers, but two mutation sites (2545, 3746) require additional second primers as discussed above. Four separate multiplex SBE reactions were then performed with each sample, one for each biotin-labeled ddNTP. Signals were correctly generated only at the zip positions corresponding to the homozygote allele (Fig. 5a) with the wild-type homozygote sample. The same procedure was repeated for the two mutant samples. Since the mutant sample with the mutation at the 1731 site has a CT heterozygote allele at zip1, two signals were generated at the zip1 position from reactions with both ddCTP and ddUTP (Fig. 5b). Also, with the mutant sample with the mutation at the 3033 site having a GT heterozygote allele at zip5, the reaction with ddUTP also generated an additional signal at the zip5 position along with the signal from ddGTP (Fig. 5c). These results show that a high fidelity of genotype discrimination for
Anal Bioanal Chem (2008) 391:405–413
multiple mutation sites can be accomplished using the new strategy developed in this study.
Conclusions In summary, we have developed a reliable multiple genotyping strategy that utilizes multiplex SBE reactions and zip-code microarrays. In addition, we have demonstrated that the technique can be successfully used as a clinical diagnostic tool for the detection of major Korean-specific BRCA mutations. As opposed to other conventional genotyping methods, the new method does not employ complicated or time-consuming experimental procedures and, as a result, it can be easily integrated into a lab-on-a-chip system to achieve a total assay system. The entire process for the final analysis takes less than 24 h. Furthermore, a high-throughput capability can be achieved without having a limitation due to the number of mutation sites since this is handled by increasing the number of chimeric primers for the multiplex SBE reactions and the corresponding zip-code sequences on the microarray. Acknowledgements This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD, KRF-2006–331-D00114) and by the Brain Korea 21 (BK21) program and the Center for Ultramicrochemical Process Systems sponsored by KOSEF.
References 1. Lenoir GM, Lynch H, Watson P, Conway T, Lynch J, Narod S, Feunteun J (1991) Lancet 338:82–83 2. Easton DF, Bishop DT, Ford D, Crockford GP (1993) Am J Hum Genet 52:678–701 3. Rowell S, Newman B, Boyd J, King MC (1994) Am J Hum Genet 55:678–701 4. Szabo CI, King MC (1997) Am J Hum Genet 60:1013–1020 5. Seo JH, Cho DY, Ahn SH, Yoon KS, Kang CS, Cho HM, Lee HS, Choe JJ, Choi CW, Kim BS, Shin SW, Kim YH, Kim JS, Son GS, Lee JB, Koo BH (2004) Hum Mutat 24:350 6. National Human Genome Research Institute (2007) Breast cancer information core. http://research.nhgri.nih.gov/projects/bic/ 7. Ahn SH, Hwang UK, Kwak BS, Yoon HS, Ku BK, Kang HJ, Kim JS, Ko BK, Ko CD, Yoon KS, Cho DY, Kim JS, Son BH (2004) J Korean Med Sci 19:269–274 8. Southern EM (1996) Trends Genet 93:13–16 9. Wang DG, Fan JB, Siao CJ, Berno A, Young P, Sapolsky RJ, Ghandour G, Perkins N, Winchester E, Spencer J (1998) Science 280:1077–1082 10. Favis R, Day JP, Gerry NP, Phelan C, Narod S, Barany F (2000) Nat Biotechnol 18:561–564 11. Goto S, Takahashi A, Kamasango K, Matubara K (2002) Anal Biochem 307:25–32 12. Tejedor D, Castillo S, Mozas P, Jimenez E, Lopez M, Tejedor MT, Artieda M, Alonso R, Mata P, Simon L, Martinez A, Pocovi M, Spanish FH Group (2005) Clin Chem 51:1137–1144 13. Park HG, Song JY, Park KH, Kim MH (2006) Chem Eng Sci 61:954–965
Anal Bioanal Chem (2008) 391:405–413 14. Yim SC, Park HG, Chang HN, Cho DY (2005) Anal Biochem 337:332–337 15. Park HG, Ham HO, Kim KH, Huh N (2005) Biosens Bioelectron 21:637–644 16. Hacia JG (1999) Nature Genet 21:42–47 17. Relogio A, Schwagner C, Richter A, Ansorge W, Valcarcel S (2002) Nucleic Acids Res 30:e51 18. Gerry NP, Witowski NE, Day J, Hammer RP, Barany G, Baraney F (1999) J Mol Biol 292:251–262 19. Fan JB, Chen X, Halushka MK, Berno A, Huang X, Ryder T, Lipshutz RJ, Lockhart DJ, Chakravarti A (2000) Genome Res 10:853–860 20. Hirschhorn JN, Skalr P, Lindblad-Toh K, Lim YM, Ruiz-Gutierrez M, Bolk S, Langhorst B, Schaffner S, Winchester E, Lander ES (2000) Proc Natl Acad Sci USA 97:12164–12169
413 21. Song JY, Park HG, Jung SO, Park J (2005) Nucleic Acids Res 33: e19 22. Barany F (1991) Proc Natl Acad Sci USA 88:189–193 23. Favis R, Day JP, Gerry NP, Phelan C, Narod S, Barany F (2000) Nat Biotechnol 18:561–564 24. Hashimoto M, Hupert ML, Murphy MC, Soper SA, Cheng YW, Barany F (2005) Anal Chem 77:3243–3255 25. Pastinen T, Raitio M, Lindroos K, Tainola P, Peltonen L, Syvänen A-C (2000) Genome Res 10:1031–1042 26. Kang HC, Kim IJ, Park JH, Kwon HJ, Won YJ, Heo SC, Lee SY, Kim KH, Shin Y, Noh DY, Yang DH, Choe KJ, Lee BH, Kang SB, Park JG (2002) Hum Mutat 20:235 27. Ahn SH, Son BH, Yoon KS, Noh DY, Han W, Kim SW, Lee ES, Park HL, Hong YJ, Choi JJ, Moon SY, Kim MJ, Kim KH, Kwak BS, Cho DY (2007) Cancer Lett 245:9095
