Bioconjugate Chem. 2008, 19, 1269–1273
1269
Minisequencing with Acyclonucleoside Triphosphates Tethered to Lanthanide(III) Chelates Pia Ollikka, Alice Ylikoski, Annukka Kaatrasalo, Heli Harvala, Harri Hakala, and Jari Hovinen* PerkinElmer Life and Analytical Sciences, Turku Site, POB 10, FI-20101 Turku, Finland. Received March 3, 2008; Revised Manuscript Received April 14, 2008
Four acyclic nucleoside triphosphates (derivatives of cytosine, thymine, 7-deazaadenine, and 7-deazaguanine) labeled with nonluminescent europium, terbium, dysprosium, and samarium chelates of 2,2′,2′′,2′′′-[[4-(4isothiocyanatophenyl)ethyl]pyridine-2,6-diyl]bis(methylenenitrilo)]tetrakis(acetic acid) were applied to minisequencing using two mutations (∆F 508 and 1717-1 G to A) of cystic fibrosis as a model system. When synthetic targets were used, all four alleles involved could be analyzed in a single reaction using four terminating substrates labeled with four different lanthanide(III) chelates and DELFIA technology for detection. Blood spot samples without DNA isolations were used for PCR amplification and genotyping the target mutations by minisequencing. The single- and dual-labeled minisequencing assays were robust, while the four-label assay still requires further optimization of the multiplexed PCR amplification.
INTRODUCTION Over 15 years ago, a method called minisequencing was introduced for the detection of point mutations in DNA (1, 2). The method involves hybridization of a single-stranded target DNA with a primer that ends immediately before the site of mutation, and elongation of the chain with a single-labeled dideoxyribonucleoside 5′-triphosphate (ddNTP). Parallel runs with each of the four possible labeled nucleotides enable identification of the mutated base. Originally, only radioactive detection was employed (2–6). Later, the detection has been predominantly based on fluorescence, where the ddNTPs are labeled with dyes and the identification of the single nucleotide polymorphisms (SNPs) is based on the signal intensity of each dye-specific emission wavelength (7–13). Nowadays, SNPs can be genotyped using multiplex, four-color fluorescent minisequencing in a microarray format (11, 12). However, application of fluorescence techniques in SNP analysis suffers from low detection sensitivity and problems associated with the properties of the labeled triphosphates to act as good substrates for DNA polymerases. Furthermore, the organic fluorophores as labels have several drawbacks, such as Raman and Rayleigh scattering, short Stokes’ shift, low water solubility, and concentration quenching. By contrast, the unique properties of lanthanide(III) chelates such as strong long-decay time luminescence make them ideal markers for numerous applications (14–16). Furthermore, the large Stokes’ shift and very sharp emission bands enable the simultaneous use of four lanthanides (i.e., Eu, Tb, Sm, and Dy) in the analysis. Time-resolved fluorimetric assays based on lanthanide(III) chelates have found applications in diagnostics, research, and high-throughput screening. The heterogeneous DELFIA technique is applied especially in assays requiring exceptional sensitivity, robustness, and multilabel approach. Up to now, this technology has not been applied for sequencing, although each of four DNA bases could be analyzed with the four different lanthanide(III) ions commonly used in DELFIA based assays. We report here the preparation of four different acyclonucleoside triphosphates labeled with different lanthanide(III) chelates. * Dr. Jari Hovinen. Tel +358 2 2678 513; fax +358 2 2678 380;
[email protected].
Also, their applicability in detection of mutations in a model system (cystic fibrosis mutations ∆F 508 and 1717-1 G to A) has been demonstrated.
EXPERIMENTAL PROCEDURES General. The PCR primers, minisequencing primers, and synthetic 5′-biotinylated targets for optimization of the minisequencing reactions (cystic fibrosis mutations ∆F 508 in exon 10 and 1717-1 G to A in exon 11 of CFTR gene) were purchased from Sigma-Genosys (HPLC purified). The sequences of synthetic targets and primers are shown in Table 1. HPLC purifications of the labeled acyclonucleoside triphosphates were performed using a Merck-Hitachi L-7100 instrument and a reversed-phase column (Termoquest Hypersil 4.6 µm). Mobile phase: Buffer A, 0.02 M triethylammonium acetate (pH 7.5); Buffer B, A in 50% (v/v) acetonitrile. Gradient: from 0 to 30 min from 10% to 45% B and from 30 to 35 min from 45% to 100% B. Flow rate was 1.0 mL min-1. ESI-TOF mass spectra were recorded on an Applied Biosystems Mariner instrument. For PCR, DNA polymerase (DyNAzyme II), buffer solution, and MgCl2 solution were purchased from Finnzymes, and dNTPs were from GE Healthcare. For minisequencing, DNA polymerase (AcycloPol), dilution buffer, unlabeled acycloterminators, streptavidin-coated microtiter plates, assay buffer, DELFIA Wash Solution, DELFIA Enhancement solution, DELFIA Enhancer, and lanthanide(III) chelates were from PerkinElmer Life and Analytical Sciences. Wallac DBS puncher, DELFIA Plateshake, Platewash, Plate Dispense, and Victor D Multilabel Counter were from PerkinElmer Life and Analytical Sciences. PCR and minisequencing reactions were performed on a GeneAmp 9700 Instrument (Applied Biosystems) and an IEMS incubator (Labsystems), respectively. Synthesis of Labeled Acyclonucleoside Triphosphates (acycloNTP-Lns). The acycloterminators tethered to propargylamino group (1.0 nmol) (17) and 2,2′,2′′,2′′′-[[4-(4-isothiocyanatophenyl)ethyl]pyridine-2,6-diyl]bis(methylenenitrilo)]tetrakis(acetic acid) lanthanide(III) (18) (3.0 nmol) were dissolved in sterile water (40 µL). Pyridine (90 µL) and triethylamine (2 µL) were added, and the reaction was allowed to proceed overnight at ambient temperature. The products were isolated by precipitation with acetone (5 mL). The precipitate was dried,
10.1021/bc800081n CCC: $40.75 2008 American Chemical Society Published on Web 05/28/2008
1270 Bioconjugate Chem., Vol. 19, No. 6, 2008
Ollikka et al.
Table 1. Sequences of Synthetic Targets and Primers CFTR mutation ∆F ∆F ∆F ∆F
508 508 508 508
∆F 508 1717-1 1717-1 1717-1 1717-1
G G G G
to to to to
A A A A
1717-1 G to A a ∆
oligonucleotide
wt/m
sequence in 5′ - 3′a
PCR primer_forward PCR primer_reverse sequencing primer synthetic target
both both both wild type
synthetic target
mutant
PCR primer_forward PCR primer_reverse sequencing primer synthetic target
both both both wild type
synthetic target
mutant
AAG CAC AGT GGA AGA ATT TC Biotin-CTC TTC TAG TTG GCA TGC T GCA CCA TTA AAG AAA ATA TCA T Biotin-ATA GGA AAC ACC AAA GAT GAT ATT TTC TTT AAT GGT GCC AGG Biotin-ATA GGA AAC ACC A∆AT GAT ATT TTC TTT AAT GGT GCC AGG GAG CAT ACT AAA AGT GAC TC Biotin-CAT GAA TGA CAT TTA CAG CAA TCT AAT TTT CTA TTT TTG GTA ATA Biotin-CTT GGA GAT GTC CTA TTA CCA AAA ATA GAA AAT TAG AGA GTC Biotin-CTT GGA GAT GTC TTA TTA CCA AAA ATA GAA AAT TAG AGA GTC
indicates the trinucleotide deletion. Sequence complement to the sequencing primer underlined.
redissolved in sterile water (1 mL), purified on HPLC, and characterized by ESI-TOF MS. The following acycloNTP-Lns were prepared: acycloGTP-Sm: ESI-TOF MS for C36H40N9O20P3SSm- (M-) calcd, 1195.05; obsd, 1195.10 acycloGTP-Eu: ESI-TOF MS for C36H40N9O20P3SEu- (M-) calcd, 1196.05; obsd, 1195.13 acycloGTP-Tb: ESI-TOF MS for C36H40N9O20P3STb- (M-) calcd, 1202.06; obsd, 1202.01 acycloATP-Tb: ESI-TOF MS for C36H40N9O19P3STb- (M-) calcd, 1186.06; obsd, 1186.12 acycloATP-Dy: ESI-TOF MS for C36H40N9O19P3SDy- (M-) calcd, 1191.07; obsd, 1191.11 acycloUTP-Eu: ESI-TOF MS for C34H38N7O21P3SEu- (M-) calcd, 1158.03; obsd, 1158.12 acycloUTP-Tb: ESI-TOF MS for C34H38N7O21P3STb- (M-) calcd, 1164.03; obsd, 1164.08 acycloCTP-Eu: ESI-TOF MS for C34H39N8O20P3SEu- (M-) calcd, 1157.04; obsd, 1157.07 acycloCTP-Tb: ESI-TOF MS for C34H39N8O20P3STb- (M-) calcd, 1163.05; obsd, 1162.91 PCR. For the analysis of the mutations, dried blood samples on a filter paper were used. From a dried blood spot sample, a 3.2 mm disk was punched directly into the PCR tube and 100 µL of master mix was added. The 100 µL amplification reactions contained 1× Dynazyme buffer, nucleotides (0.2 mM), MgCl2 (2.5 mM), primers, and 1 U of Dynazyme II Polymerase enzyme. For single- and dual-label minisequencing assays, the PCR mixtures contained either ∆F 508 or 1717-1 G to A primer pairs. For four-label minisequencing, the PCR mixture contained all four PCR primers. The primer concentrations for ∆F 508 amplifications were 0.2 µM, while the primer concentrations for 1717-1 G to A amplification were 0.1 µM. In the positive control reactions, the blood disk was substituted with 200-300 ng of purified DNA. The negative control reaction contained only the reaction mixture. The PCR program used was the following: preheating to 95 °C followed by 30 cycles of 95 °C for 1 min, 56 °C for 1 min, and 72 °C for 1 min. When the amplification was completed, the PCR products were cooled to 4 °C and stored at -20 °C. Both the mutations in exons 10 and 11 were amplified using the same reaction conditions. Minisequencing Reactions Using Synthetic Oligonucleotides. The minisequencing reaction was optimized using synthetic targets simulating the wild-type and mutated alleles of the exons 10 and 11. The biotinylated synthetic targets were allowed to bind to the streptavidin-coated wells (1011 target molecules per well in 50 µL of DELFIA Assay Buffer containing 1 M NaCl and 0.1% Tween 20) for 30 min with shaking at ambient temperature. Unbound targets were removed
using DELFIA Wash Solution and Platewash (3 washings). The minisequencing reaction conditions were optimized as the function of concentration of reaction components (primers, acyclonucleotides, and enzyme), reaction time, reaction volume and temperature, and the focus was on the signal levels and specificity of the reaction. The minisequencing reaction was allowed to proceed in sealed plates with shaking for 20 min at 50 °C. The reactions contained 1 pmol of sequencing primers, 1-5 pmol of acycloNTP-Lns, and 50 µL of reaction buffer. The optimized reaction buffer consisted of 10 mM Tris-HCl, pH 8.5, 50 mM NaCl, 50 µM EDTA, 7.5 mM MgCl2, 0.1 mg/ mL BSA, and 0.4 U/mL AcycloPol polymerase. When the reactions were completed, the wells were rinsed 6 times with DELFIA Wash Solution. Analyses of reactions were performed according to DELFIA protocol using Victor D Multilabel Counter. Shortly, lanthanide ions were released to solution using DELFIA Enhancement Solution (200 µL) and shaking for 25 min. In the case of Tb and Dy, additional incubation of 5 min after the addition of 50 µL DELFIA Enhancer was used. Luminescent chelates formed during this period of time were measured with Victor D Multilabel Counter. Minisequencing of PCR Products. The minisequencing reaction optimized with synthetic oligonucleotides was tested with PCR products. The capture of the PCR product was performed in a total volume of 50 µL as described above. A 25 µL aliquot of PCR product together with 25 µL of DELFIA Assay Buffer containing 1 M NaCl and 0.1% Tween 20 was transferred and allowed to bind onto streptavidin-coated microtiter plates, and after the washing the products bound to the wells were denatured by shaking for 5 min in the presence of NaOH (10 mM) and EDTA (300 µM). The unbound singlestranded DNA and the denaturation solution were removed by washing three times. The minisequencing reaction was performed in 50 µL as described above. The control reactions were the following: positive PCR control containing purified DNA, negative PCR control containing only the PCR reaction mixture, and a positive minisequencing control containing 1011 molecules of the synthetic target. The negative minisequencing control was performed in the reaction buffer in the absence of the enzyme.
RESULTS Synthesis of the Labeled Acyclonucleoside Triphosphates. The acycloterminators labeled with four different lanthanide(III) chelates were synthesized by allowing the acycloterminators tethered to propargylamino group to react with europium, terbium, samarium, and dysprosium chelates of the DELFIA DNA labeling chelate, 2,2′,2′′,2′′′-[[4-(4-isothiocyanatophenyl)ethyl]pyridine-2,6-diyl]bis(methylenenitrilo)]tetrakis(acetic acid). After precipitation from acetone, the crude products were
Minisequencing with Acyclonucleoside Triphosphates
Figure 1. Structures of the acycloterminators labeled with lanthanide(III) chelates. Ln is Eu, Tb, Sm, and Dy.
redissolved in water and purified on HPLC. The labeled triphosphates were analyzed on mass spectrometry, and the observed molecular weights were in accordance with the proposed structures (Figure 1). Accordingly, in all the cases the ligand structure is the same, and each of the four nucleobases can be identified by using the specific signal derived from europium, samarium, terbium, and dysprosium. Optimization of Minisequencing. The minisequencing reaction conditions were optimized using synthetic oligonucleotides
Bioconjugate Chem., Vol. 19, No. 6, 2008 1271
as the function of reaction components, temperature, and time. For ∆F 508 mutation, the best results were obtained using a 22-base primer; the use of longer primers caused significant cross reactivity, and the shorter sequences, in turn, resulted in decrease of the specific signal. For 1717-1 G to A mutation, the primer did not have a significant effect on the signal-tonoise ratio, and a 24-mer primer was chosen. With respect to both mutations, 1 pmol of primer per well was enough, and the use of higher amounts did not give better results. In single-label assays with synthetic oligonucleotides, one of the acyclonucleotides was labeled with europium (5 pmol per well) and the other three acyclonucleotides were unlabeled (5 pmol per well). In dual- and four-label assays, the suitability of each label to each nucleotide was investigated. In the duallabel assay for detection of ∆F 508 mutation, the best results were obtained using acycloCTP-Eu and acycloUTP-Tb, wherein acycloATP and acycloGTP were unlabeled, and when each of which was 1 pmol per well. For detection of 1717-1 G to A mutation, in turn, the best results were obtained using acycloGTP-Eu and acycloATP-Tb (1 pmol of each per well) and 5-fold excess of the unlabeled acycloCTP and acycloUTP. In four-label assays, where both of the mutations were analyzed from a single reaction, the following labels were chosen: acycloCTP-Eu (5 pmol), acycloUTP-Tb (1 pmol), acycloGTPSm (5 pmol), and acycloATP-Dy (5 pmol). In all cases, the best results were obtained at 50 °C using a 20 min reaction time. The most suitable reaction volume was 50 µL. Minisequencing of PCR Products. In the first set of experiments, PCR amplified products of 32 blood spots were analyzed using single-label assays. Only the mutant and wildtype alleles of ∆F 508 mutation were analyzed, each in their own reactions using acycloCTP-Eu specific to the wild typeallele, and acycloUTP-Eu specific to the mutant allele. The differentiation of wild-type, carrier, and mutant was based on the data obtained with the synthetic model developed above, and the result was calculated as a ratio of mutant acycloUTP signal to wild-type acycloCTP signal. Here, the UTP/CTP ratio for wild-type samples was below 1, for mutants over 5, and for carriers 1.5-2.5 (Figure 2). Accordingly, the single-label method allows us to detect all mutants, carriers, and wild-type samples from the blood spots. In dual-label assays, both the wild-type and mutant alleles of the same target area were analyzed from a single reaction. Genotyping of ∆F 508 and 1717-1 G to A mutations from 16 blood spot samples was performed using two dual-label assays.
Figure 2. Minisequencing of blood disk samples with a single-label assay. The single-label assays utilized acycloCTP-Eu for the detection of wild-type allele and acycloUTP-Eu for the detection of mutant allele. The UTP/CTP ratio for wild-type, mutant, and carrier samples were below 1, over 5, and 1.5-2.5, respectively.
1272 Bioconjugate Chem., Vol. 19, No. 6, 2008
Ollikka et al.
Figure 3. Minisequencing of blood disk samples with a dual-label assay. The dual-label assay for ∆F 508 utilized acycloCTP-Eu for the detection of wild-type allele and acycloUTP-Tb for the detection of mutant allele. The dual-label assay for 1717-1 G to A utilized acycloGTP-Eu for the detection of wild-type allele and acycloATP-Tb for the detection of mutant allele. The control samples were mutant (M)/wild-type (WT) and wild-type (WT)/mutant (M) with respect to ∆F 508/1717-1 G to A. For the ∆F 508 assay (gray bars), the UTP/CTP ratios for wild-type, mutant, and carrier samples were below 1, over 5, and 1.5-2.5, respectively. For the 1717-1 G to A assay (white bars), the ATP/GTP ratios for wild type was below 0.1, for mutant over 1, and that for carrier was not determined due to lack of carrier samples.
Figure 4. Minisequencing of synthetic DNA samples with a four-label assay. The four-label assay utilized acycloCTP-Eu for wild type ∆F 508, acycloUTP-Tb for mutant ∆F 508, acycloGTP-Sm for wild-type 1717-1 G to A, and acycloATP-Dy for mutant 1717-1 G to A. X-axis shows the target composition in respect to ∆F 508/1717-1 G to A. WT, C, and M denote wild-type, carrier, and mutant, respectively. Gray bars show the ∆F 508 (UTP/CTP ratio) and white bars the 1717-1 G to A (ATP/GTP ratio) results. The mutant/wild-type signal ratio for wild type, mutant, and carrier samples was below 0.1, over 5, and 0.5-1, respectively.
The acycloCTP-Eu/acycloUTP-Tb pair and acycloGTP-Eu/ acycloATP-Tb pair was chosen for ∆F 508 and 1717-1 G to A mutations, respectively. GTP and ATP were specific nucleotides for wild-type and mutated sequences of the 1717-1 G to A mutation. The UTP/CTP ratio for ∆F 508 samples was comparable to the single-label assay (Figure 3). The ATP/GTP ratio in the 1717-1 G to A wild-type samples was always below 0.05 (mean of 16 samples 0.015) (Figure 3). Accordingly, the dual-label minisequencing reaction easily differentiates the wildtype and mutated sequences. The four-label system was first tested with synthetic oligonucleotides. By mixing ∆F 508 and 1717-1 G to A target sequences, we could simulate the conditions in real patient samples. The concentration of oligonucleotide targets used (2 × 1011 molecules/well) was comparable to PCR product concentration used in earlier experiments. The ∆F 508 and 1717-1 G to A mutations were measured using acycloCTPEu and acycloUTP-Tb for wild-type and mutated ∆F 508, respectively, and acycloGTP-Sm and acycloATP-Dy for the wild-type and mutated 1717-1 G to A, respectively. The fourlabel assay clearly discriminates the four different targets in the
same reaction: UTP/CTP for wild-type, carrier, and mutated ∆F 508, and ATP/GTP for wild-type, carrier, and mutated 1717-1 G to A were clearly divided into three groups (Figure 4). For the four-label assay with synthetic targets, the mutant/wild-type signal ratio was below 0.1, 0.5-1, and over 5 for the wildtype, carrier, and mutant samples, respectively. Although the results with the synthetic targets were very promising, our fourlabel system could not discriminate PCR-amplified products from blood-spot samples.
DISCUSSION To study the applicability of acyclonucleoside triphosphates labeled with lanthanide(III) chelates to minisequencing, we selected the major mutation ∆F 508 (frequency 66%) as well as the mutation 1717-1 G to A (frequency 0.6%), which is among the eight most common mutations of CF in the world. ∆F 508 is a three base deletion in the exon 10 of CFTR gene codon 508, while in 1717-1 G to A there is a single transversion of G to A in the position before exon 11. The three base deletion in ∆F 508 causes the loss of phenylalanine in CFTR protein,
Minisequencing with Acyclonucleoside Triphosphates
and the transversion of G to A at position 1717 has an effect on breakdown of the mRNA produced by the gene. We knew in advance that out of the 32 samples 7 were ∆F 508 mutant homozygotes, 9 were ∆F 508 heterozygote carriers, and the genotypes of the remaining 16 samples were unknown. By using the single-label assay, we could easily determine the genotype for ∆F 508 mutation for all 32 samples. The results of the known samples were in accordance with the previous information. However, since this approach requires the determination of each allele in separate reactions, it is timeconsuming and requires significant amounts of rather expensive reagents. In the dual-label assays, we could determine both wild-type and mutated alleles of each mutation in a single reaction. Accordingly, for determination of ∆F 508 and 1717-1 G to A mutations, only two reactions were needed. The results of the ∆F 508 analysis were in accordance with the results obtained in the single-label assays. All the samples were analyzed as 1717-1 G to A wild types. Although we did not know in advance any of the genotypes of 1717-1 G to A mutations, it is not likely that any of the samples were 1717-1 G to A carriers or mutants due to the extremely low abundance of that mutation (0.6% of population). Although we did not have blood spot samples of 1717-1 G to A carriers or mutants, the results obtained with synthetic mutated sequences show that the system is sensitive enough to analyze 1717-1 G to A mutations. Due to lack of carrier samples, we did not estimate the carrier ranges for 1717-1 G to A. Although the four-label assay was sensitive enough to differentiate the synthetic targets, we were not able to perform the same assay with blood spot specimens. Since the minisequencing is working well with the synthetic targets, the problem with blood spot samples is, in all likelihood, derived from PCR amplification, which has been shown to be the bottleneck of minisequencing (1). For the four-label assay, we had to amplify the sequences of both the mutations in the same PCR reaction, i.e., there were two biotinylated and two nonbiotinylated primers in the same reaction. Accordingly, and in contrast to singleand dual-label assays, we had two different mutation sequences as well as two different pairs of PCR primers in a single well, which can bind to each other. Therefore, in order to use fourlabel assay for blood spot samples, optimization of multiplexPCR amplification is required. Alternatively, a software program enabling data extraction should be developed. The major goal of the present work was to develop a method based on minisequencing where we could determine the genotype of the patient two mutations in a single reaction using four different lanthanide-labeled nucleotides. Although this goal has not been reached yet, we have clearly demonstrated that the dual-label assay is well-suited for detection of ∆F 508 and 1717-1 G to A mutations involved in CF: the assay is simple, and the mutations can be determined directly from the products of PCR-amplified blood spot samples. The method is also suitable for detection of four alleles simultaneously.
LITERATURE CITED (1) For a review, see (a) Syva¨nen, A.-C. (1999) From gels to chips: “minisequencing” primer extension for analysis of point mutations and single nucleotide polymorphism. Hum. Mutat. 13, 1– 10.
Bioconjugate Chem., Vol. 19, No. 6, 2008 1273 (2) Syva¨nen, A.-C., Aalto-Seta¨la¨, K., Harju, L., Kontula, K., and So¨derlund, H. (1990) A primer-guided nucleotide incorporation assay in the genotyping of aspartylgucosaminuria in Finland. Genomics 12, 590–595. (3) Sokolov, B. P. (1990) Primer extension technique for the detection of single nucleotides in genomic DNA. Nucleic Acids Res. 18, 3671. (4) Jalanko, A., Kere, J., Savilahti, E., Swartz, M., Syva¨nen, A.C., Ranki, M., and So¨derlund, H. (1992) Screening for defined cystic fibrosis mutations by solid-phase minisequencing. Clin. Chem. 38, 39–43. (5) Ikonen, E., Manninen, T., Peltonen, L., and Syva¨nen, A.-C. (1992) Quantitative determination of rare mRNA species by PCR and solid-phase minisequencing. PCR Methods Amplification 1, 234–240. (6) Ihalainen, J., Siitari, H., Laine, S., Syva¨nen, A.-C., and Palotie, A. (1994) Towards automatic detection of point mutations: use of scintillating microplates in solid-phase minisequencing. Biotechniques 16, 938–943. (7) Chen, X., Zehnbauer, B., Gnirke, A., and Kwok, P. Y. (1997) Fluorescence energy transfer detection as a homogenous DNA diagnostic method. Proc. Natl. Acad. Sci. U.S.A. 94, 10756– 10761. (8) Shapero, M. H., Leuther, K. K., Nguyen, A., Scott, M., and Jones, K. W. (2001) SNP genotyping by multiplex solid-phase amplification and fluorescent minisequencing. Genome Res. 11, 1926–1934. (9) Raitio, M., Lindroos, K., Laukkanen, M., Pastinen, T., Sistonen, P., Sajantila, A., and Syva¨nen, A.-C. (2001) Y-Chromosomal SNPs in Finno-Ugric-speaking population analyzed by minisequencing on microarrays. Genome Res. 11, 471–482. (10) Brown, N. M., Bernacki, S., Pratt, V. M., and Stenzel, T. T. (2001) Fluorescent, multiplexed, automated, primer-extension assay for 3120 + 1 G)A and 1148 T mutations in cystic fibrosis. Clin. Chem. 47, 2053–2055. (11) Lindroos, K., Liljedahl, U., Raitio, M., and Syva¨nen, A.-C. (2001) Minisequencing on oligonucleotide microarrays: comparison of immobilisation chemistries. Nucleic Acids Res. 29, e69. (12) Lindroos, K., Sugurdsson, S., Johansson, K., Ro¨nnblom, L., and Syva¨nen, A.-C. (2002) Multiplexing SNP genotyping in pooled DNA samples by a four-color microarray system. Nucleic Acids Res. 30, e70. (13) Wang, W., Kham, S. K. Y., Yeo, G.-H., Quah, T.-C., and Chong, S. S. (2003) Multiplex minisequencing screen for common Southeast Asian and Indian β-thalassemia mutations. Clin. Chem. 49, 209–218. (14) Hemmila¨, I. (1991) Applications of Fluorescence in Immunoassays, Vol 117, Chemical Analysis: A Series of Monographs on Analytical Chemistry and Its Applications, John Wiley & Sons, Inc. New York. (15) Hemmila¨, I., Dakubu, S., Mukkala, V.-M., Siitari, H., and Lo¨vgren, T. (1984) Europium as a label in time-resolved immunofluorometric assays. Anal. Biochem. 137, 335–343. (16) Hemmila¨, I., and Mukkala, V.-M. (2001) Time-resolution in fluorometry technologies, labels, and applications. Crit. ReV. Clin. Lab. Sci. 38, 441–4519. (17) Hobbs, F. W., Jr, and Trainor, G. L. Alkynylamino-nucleosides, U.S. Patent 5,151,507. (18) Takalo, H., Mukkala, V.-M., Mikola, H., Liitti, P., and Hemmila¨, I. (1994) Synthesis of europium(III) chelates suitable for labeling of bioactive molecules. Bioconjugate Chem. 5, 278–282. BC800081N
