Detection of Oligonucleotide Hybridization on a Single Microparticle

Detection of Oligonucleotide Hybridization on a Single. Microparticle by Time-Resolved Fluorometry: Quantitation and. Optimization of a Sandwich Type ...
1 downloads 0 Views 115KB Size
Download PDF
316

Bioconjugate Chem. 1998, 9, 316−321

Detection of Oligonucleotide Hybridization on a Single Microparticle by Time-Resolved Fluorometry: Quantitation and Optimization of a Sandwich Type Assay Harri Hakala,* Esa Ma¨ki, and Harri Lo¨nnberg Department of Chemistry, University of Turku, FIN-20014 Turku, Finland. Received October 24, 1997; Revised Manuscript Received January 22, 1998

Uniformly sized (50 µm) porous glycidyl methacrylate/ethylene dimethacrylate particles (SINTEF) were used as the solid phase in a sandwich type mixed-phase hybridization assay based on timeresolved fluorescence detection on a single particle. These particles were coated with oligodeoxyribonucleotide probes by conventional phosphoramidite chain assembly. An oligodeoxyribonucleotide bearing a photoluminescent europium(III) chelate, {2,2′,2′′,2′′′-{{4′-{4′′′-[(4,6-dichloro-1,3,5-triazin-2yl)amino]phenyl}-2,2′:6′,2′′-terpyridine-6,6′′-diyl}bis(methylenenitrilo)}tetrakis(acetato)}europium(III), was hybridized to a complementary sequence of the target oligonucleotide, and the resulting duplex was further hybridized to the particle-bound probes. The latter binding was quantified by time-resolved measurement of the emission signal of a single particle. Kinetics of hybridization and the effect of the concentration of the target oligomer and the fluorescently tagged probe on the efficiency of hybridization were studied. The intensity of the emission signal was linearly related to the concentration of the target oligomer over a range of 5 orders of magnitude. The length of the complementary region between the target oligomer and the particle-bound probe was varied, and the effect of point mutations and deletions on the hybridization efficiency was determined in each case. The maximal selectivity was observed with 10-16-base pair complementary sequences, the optimal length depending on the oligonucleotide loading on the particle. Discrimination between the complete matches and point mismatches was unequivocal, a single point mutation and/or deletion decreasing the efficiency of hybridization by more than 2 orders of magnitude.

INTRODUCTION

Solid supports bearing immobilized oligodeoxyribonucleotide probes are a useful tool in analyzing the base sequence of nucleic acids (1-10). This kind of mixedphase assay combined with efficient amplification techniques, such as polymerase chain reaction (PCR; 11), has gained increasing popularity in clinical diagnostics. These assays are usually based on arrays of allele-specific oligonucleotides attached to a solid support, hybridization of fluorescently or radioactively labeled PCR-amplified sequences to the support-bound probes, and detection of the hybridized sequences by location of the label spots on the support (12-14). We have recently tried to evaluate the feasibility of an alternative approach based on the following principle (15). A mixture of microscopic particles, each of which bears a given allele-specific oligonucleotide and a reporter group defining the particle category, is used as the solid phase. After hybridization, individual particles are separately subjected to two parallel measurements: one identifies the particle category (reporter groups on the particle), and the other quantifies the fluorescently tagged oligonucleotide hybridized to the particle-bound probes. The fluorescent markers employed for labeling of oligonucleotides are photoluminescent lathanide chelates. Such markers have advantages over organic fluorophores, since their long-lived fluorescence allows the usage of the time* Send correspondence to Harri Hakala, Department of Chemistry, University of Turku, FIN-20014 Turku, Finland. Telephone: +358-2-333 8091. Fax: +358-2-333 6700. E-mail: [email protected].

resolved mode in the measurement, and this, in turn, allows efficient elimination of the prompt background fluorescence. Furthermore, the difference between the wavelength of excitation and emission is large, and the emission bands are narrow (16, 17). Lanthanide chelates have earlier been utilized in highly sensitive immunoassays (18-20) and also in studying nucleic acid hybridization (19-24). However, none of the previous hybridization assays is based on a single-particle approach similar to that used in this work. In other words, they do not form a basis for development of a multiparametric assay. We have previously described several methods for covalent immobilization of oligonucleotide probes to uniformly sized polymer particles (25) and shown that the same particles may also, after appropriate derivatization, be used as a solid support in automated oligonucleotide synthesis (26). In fact, the postsynthetic immobilization of purified oligonucleotide probes appears to offer no advantage over the more convenient direct chain assembly on the particles. With both types of particles, the intensity of the emission signal measured directly from the particle surface is proportional to the concentration of the fluorescently tagged oligomer in solution over a concentration range of 5 orders of magnitude. About 104 oligonucleotides hybridized to a single particle is sufficient for reliable quantitation. To show that the particles obtained by direct chain assembly also exhibit excellent selectivity of hybridization, although they bear a considerable amount of truncated sequences, we now report on quantitation of a sandwich type hybridization assay on these particles. The principle of the assay is described in Figure 1. The results show that

S1043-1802(97)00186-9 CCC: $15.00 © 1998 American Chemical Society Published on Web 04/16/1998

Detection of Oligonucleotide Hybridization

Bioconjugate Chem., Vol. 9, No. 3, 1998 317

Figure 1. Principle of the sandwich hybridization assay studied. Table 1. Target Oligodeoxyribonucleotidesa compd

sequence

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

5′-d(CTATATTCATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(CTATATTAATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(CTATATTCCTCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(CTATATTvATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(ATATTCATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(ATATTCCTCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(ATATTCvTCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(TATTCATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(TATTCAGCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(TATTCAvCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(ATTCATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(ATTCATAATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(ATTCAvCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(TCATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(TCATCCTAGGAAACACCAAAGATGATATTC)-3′ 5′-d(TCATCvTAGGAAACACCAAAGATGATATTC)-3′ 5′-d(CATCATAGGAAACACCAAAGATGATATTC)-3′ 5′-d(ATCATAGGAAACACCAAAGATGATATTC)-3′

a The sequence complementary to the particle-bound probes is indicated by bold letters and that complementary to the fluorescently tagged probe in italics. The site of possible point mutation or deletion is underlined or shown by an arrow, respectively.

proper adjustment of the length of the complementary region between the target oligonucleotide and the particlebound probes allows an unequivocal discrimination between a complete match and a point mismatch (mutation and/or deletion) over a wide concentration range. Under optimal conditions, a single mismatch reduces the efficiency of hybridization by more than 2 orders of magnitude. EXPERIMENTAL PROCEDURES

Oligonucleotide-Coated Particles (1). The particles were porous uniformly sized beads (L ) 50 µm) made of a copolymer of glycidyl methacrylate (40%) and ethylene dimethacrylate (60%) (37% matrix) and derivatized with bis(3-aminopropyl)amine, yielding a 1 mmol g-1 density of primary amino functions (SINTEF). They were derivatized for oligonucleotide synthesis as described previously (26), and the sequence 5′-d(CCTATGATGAATATAGTTT)-3′ was assembled with the normal phosphoramidite strategy. According to the DMTr cation response (27), the loadings of the full-length oligonucleotide sequence on three separate batches prepared (1) were 2, 5, and 10 µmol g-1. After conventional ammonolysis and washings, the particles were dried under reduced pressure, removed from the column, and washed with the buffer used in the hybridization assays.

Oligonucleotides and Their Fluorescent Conjugates. The unlabeled oligodeoxyribonucleotides used as targets in the sandwich assays were synthesized with the conventional phosphoramidite strategy and purified by ion exchange and RP-HPLC and desalted. Table 1 lists the oligomers studied. Two fluorescently tagged oligonucleotides were used in the hybridization assays: 5′-d(X5ATATCATCTTTGGTGT)-3′ (20) and 5′-(X20ATCATCTTTGGT)-3′ (21). Here X stands for a N4-(6-aminohexyl)-2′-deoxycytidine unit tethered to a photoluminescent europium(III) chelate, {2,2′,2′′,2′′′-{{4′-{4′′′-[(4,6-dichloro-1,3,5-triazin-2-yl)amino]phenyl}-2,2′:6′,2′′-terpyridine-6,6′′-diyl}bis(methylenenitrilo)}tetrakis(acetato)}europium(III) (28). These oligonucleotide conjugates were prepared as described previously (29). Probes 20 and 21 were observed to bear 5 and 17 labels per oligomer, respectively, when assayed according to the DELFIA protocol (18). In other words, the extent of labeling of the aliphatic amino functions was 1.0 with 20 and 0.85 with 21. Hybridization Assays. The hybridization assays were carried out in a Tris buffer (50 mmol L-1 at pH 7.5, 0.01% Tween 20 and 0.5 mol L-1 NaCl). Usually, 50 particles were incubated in 10 µL of the buffer containing a known amount of the target oligonucleotide (2-19) and the fluorescently labeled probe (20 or 21). The reaction

318 Bioconjugate Chem., Vol. 9, No. 3, 1998

time was typically 24 h. However, on examination of the kinetics of hybridization, 500 particles in a total volume of 100 µL was used, from which aliquots of 10 µL were withdrawn. In these experiments, the concentration of the target oligonucleotide (2) was 5 nmol L-1 and that of the labeled oligonucleotide (20) 17 nmol L-1. The particles were washed twice with 200 µL of glycine buffer (50 mM at pH 10), containing 20% propanol, and then transferred into a quartz capillary tube for measurement. However, in the experiments of kinetics of hybridization, particles were washed once with 300 µL. The fluorescence emission of each particle was determined separately on a time-resolved microfluorometer as described previously (30). A delay time of 296 µs was used, and the signal was collected for 300 µs, which allowed complete elimination of the prompt background fluorescence. Typically, 5-10 individual particles were measured, the standard deviation ranging from 5 to 10%. RESULTS

Quantitation of the Sandwich Type Hybridization. Particles 1 bearing the sequence 5′-d(CCTATGATGAATATAGTTT)-3′ (three batches with loadings of 2, 5, and 10 µmol g-1) were used as the solid support in all hybridization studies. The target oligonucleotides (219) were 28-36 nucleotides long. They contained at their 5′-terminus an 8-16-nucleotide sequence complementary to the particle-bound probe (indicated by bold letters in Table 1) and at their 3′-terminus a 16- or 12-nucleotide sequence (indicated in italics in Table 1) complementary to the fluorescently tagged oligomers 20 [5′-d(X5ATATCATCTTTGGTGT)-3′] or 21 [5′-d(X20ATCATCTTTGGT)3′], respectively. The labeled oligomer (20 or 21) was used in excess compared to the target (2-19) to ensure quantitative hybridization of the target to the labeled probe. With 20, which has a 16-nucleotide sequence complementary to the target, no excess is really needed. The emission signal measured was independent of the concentration of 20, as long as this concentration was at least equal to that of the target. With 21, which has only a 12-nucleotide sequence complementary to the target, a 15-fold excess was needed to obtain a maximal emission signal. For this reason, 20 was used as the fluorescent probe in most of the measurements. Unspecific adsorption of 20 or 21 to the particles was estimated by treating the particles with the labeled probe in the absence of the target oligomer. The unspecific binding represented in all the cases less than 1% of the total amount of the labeled probe. The resulting background signal was determined at each concentration of 20 (or 21) employed and subtracted from the signal obtained in the presence of the target. Throughout this paper, only these corrected signals referring to sequence-selective hybridization are given. To convert the microfluorometry signals to the number of fluorescently labeled oligonucleotides hybridized to the particle, a calibration line was determined. Two sets of hybridization reactions were carried out, in which the concentrations of the target (2) and fluorescent probe (20) were varied from 0.17 to 50 nM. After incubation for 20 h, one set of particles was subjected to measurement by microfluorometry, while from the other set, europium(III) ions were released in solution and determined according to the DELFIA protocol (18). The calibration line obtained (data not shown) was essentially similar to that described previously (25, 26) for the direct hybridization of a fluorescently tagged oligonucleotide to the same particles, and it was used throughout this work

Hakala et al.

Figure 2. Kinetics of formation of a sandwich hybrid on particles 1, having an oligonucleotide loading of 2 µmol g-1. The concentration of the target oligonucleotide (2) was 5 nmol L-1 and that of the fluorescently tagged probe (20) 17 nmol L-1. The particle density was 5 particles/µL and the temperature 25 °C.

to convert the microfluorometry signals to the number of fluorescently labeled oligonucleotide molecules. Figure 2 shows the hybridization kinetics observed with particles 1, having an oligonucleotide loading of 2 µmol g-1. The concentration of target (2) was 5 nmol L-1 (3 × 1011 molecules in 100 µL) and that of the fluorescently tagged oligonucleotide (20) 17 nmol L-1. The kinetics of this sandwich type hybridization are very similar to those observed previously (26) for direct hybridization of fluorescently labeled oligonucleotides to the same particles. Evidently, hybridization of 2 to 20 is, in the excess of the latter, quantitative and fast compared to the hybridization of the resulting duplex to the solid phase. The latter sandwich hybridization obeyes first-order kinetics. The curve indicated in Figure 2 represents the best least-squares fit obtained on the basis of this model. The kinetics were similar also when 21 was used as the labeled probe (data not shown). The dependence of the intensity of the emission signal on the concentration of the target oligomer was studied by keeping the concentration of the fluorescent probe (20) constant and varying that of the target (2): [20] ) 166 pmol L-1 when [2] ) 0.5-100 pmol L-1, and [20] ) 50 nmol L-1 when [2] ) 0.1-166 nmol L-1. The oligonucleotide loading of the particles was 10 µmol g-1. This experiment mimics the real assay of amplified oligomers, because then the amount of target is unknown and a constant higher concentration of the labeled probe must be used. The results obtained are shown in Figure 3. As seen, the linear range is wide, 5 orders of magnitude. However, the entire range cannot be utilized by using a single constant probe concentration, because at the lowest target concentrations the unspecific binding of the probe becomes too high compared to the specific hybridization signal. The excess of the fluorescent probe cannot be larger than 1000-fold. Although the intensity of the emission signal is linearly related to the concentration of the target oligomer in solution over a range of 5 orders of magnitude, the slope of this correlation line is not exactly unity but 1.05. At least apparently the efficiency of hybridization increases with the increasing concentration of the target. At the lowest concentration of 2, the efficiency [(amount of 2 hybridized)/(total amount of 2)] is only 16% but at the highest concentration 85%. In this respect, the sandwich hybridization seems to differ from the direct hybridization of the labeled probe to the particles. As shown previously (26), in the latter case, the hybridization efficiency remains constant over the entire concentration range of 5 orders of magnitude.

Detection of Oligonucleotide Hybridization

Bioconjugate Chem., Vol. 9, No. 3, 1998 319

A

B Figure 3. Intensity of the emission signal measured from a single microparticle plotted against the concentration of the target oligonucleotide (2). The concentration of the fluorescently tagged probe was 0.166 nmol L-1 when [2] < 0.1 nmol L-1 and 50 nmol L-1 at higher concentrations of 2.

Effect of the Length of the Complementary Region on the Hybridization Efficiency. To determine the minimal length of the complementary sequence between the target oligonucleotide and the particle-bound probe that allows hybridization with maximal efficiency, a set of target oligomers (2, 6, 9, 12, 15, 18, and 19) having sequences of 16, 14, 13, 12, 10, 9, and 8 bases complementary to the particle-bound oligonucleotide were hybridized to 1. Panels A and B of Figure 4 refer to oligonucleotide loadings of 10 and 2 µmol g-1 on the particles, respectively. With both types of particles, the intensity of the emission signal is increased linearly with the increasing concentration of target, as long as the concentration of the target oligomer remains lower than that of the labeled probe. As seen, the concentrationdependent signal levels are identical as long as the complementary region is 12 nucleotides or longer. With shorter sequences, the signal level gradually decreases. Similar results were obtained when target 2 was allowed to hybridize with particles bearing probes shorter than 1 (data not shown). Comparison of the curves in panels A and B of Figure 4 reveals that decreasing the length of the complementary sequence has a more dramatic effect on the efficiency of hybridization when the oligonucleotide loading on the particles is low. Table 2 contains the hybridization efficiencies observed at high target concentrations, i.e. at concentrations exceeding the one required to make the signal independent of the target concentration (17-170 nmol L-1). The data obtained with an oligonucleotide loading of 5 µmol g-1 are included in the same table. These results indicate that the minimum length of complementarity needed for maximal efficiency of hybridization is 10 nucleotides when the oligonucleotide loading on the particle is 5 µmol g-1 or higher and 12 nucleotides for particles where the oligonucleotide loading is 2 µmol g-1. Selectivity of Hybridization. To determine how sensitive the efficiency of hybridization is to the presence of a single mismatch within a complementary sequence of variable length, a set of target oligonucleotides containing a single point mutation or deletion was prepared and hybridized to particles 1. The length of the complementary region was varied from 10 to 16 nucleotides, and the mismatch was placed approximately in the center of this region to obtain a maximal discrimination between

Figure 4. Effect of the length of the complementary sequence between the target and the particle-bound probes (1) on the hybridization efficiency. The intensity of the emission signal measured from a single microparticle plotted against the concentration of the target (2, 12, 15, 18, and 19). The concentration of the fluorescently tagged probe was 17 nmol L-1. Panels A and B correspond to oligonucleotide loadings of 10 and 2 µmol g-1, respectively. Table 2. Hybridization Efficiencies of Particles 1 Having Oligonucleotide Loadings of 2, 5, or 10 µmol g-1 a efficiency (%) compd

length

2 µmol g-1

5 µmol g-1

10 µmol g-1

2 6 9 12 15 18 19

16 14 13 12 10 9 8

76 78 78 50 0.7 0.4 0.07

71 87 75 72 17 1.6 0.07

77 76 60 61 18 3 0.06

a The effect of the length of the complementary sequence between the target oligonucleotide and the particle-bound probe.

matches and mismatches (12, 31). The concentration of these targets (3-5, 7, 8, 10, 11, 13, 14, 16, and 17) was varied from 0.17 to 170 nmol L-1 in the hybridization assays, and that of the fluorescent probe (20) was 17 nmol L-1. Particles 1 with all three different oligonucleotide loadings (2, 5, and 10 µmol g-1) were used as a solid phase. Figure 5 shows an example of the results: the effect of a single point mutation (13) and deletion (14) on the concentration-dependent hybridization efficiency of a target having 12 bases complementary to the particle-bound oligomers (loading of 10 µmol g-1). As seen, both mismatches reduce the hybridization efficiency

320 Bioconjugate Chem., Vol. 9, No. 3, 1998

Hakala et al.

on the hybridization efficiency are rather similar to those of point mutations. DISCUSSION

Figure 5. Effect of a single point mutation or deletion on hybridization of a target oligonucleotide (12) having a 12nucleotide sequence complementary to the particle-bound probes of particles 1 (loading of 10 µmol g-1). The concentration of the fluorescently tagged probe was 17 nmol L-1. Table 3. Effect of a Single Point Mutation on the Hybridization Efficiency of Oligonucleotides Having a 10-16-Nucleotide Sequence Complementary to the Particle-Bound Probes on Particles 1a efficiency (%) compd

length

2 µmol g-1

5 µmol g-1

10 µmol g-1

3 4 7 10 13 16

16 16 14 13 12 10

4.6 5.6 0.9 0.5 0.3 0.9

47 28 1.4 1.3 0.08 2

73 63 4.2 3.3 0.1 1

a The results obtained with particles having oligonucleotide loadings of 2, 5, or 10 µmol g-1 are indicated. The hybridization efficiencies are expressed as a percentage of the value of the unmutated sequence.

Table 4. Effect of a Single Point Deletion on the Hybridization Efficiency of Oligonucleotides Having a 10-16-Nucleotide Sequence Complementary to the Particle-Bound Probes on Particles 1a efficiency (%) compd

length

2 µmol g-1

5 µmol g-1

10 µmol g-1

5 8 11 14 17

16 14 13 12 10

1.4 1.4 0.3 0.3 1.1

15 5.2 0.9 0.05 3

50 9.6 3.8 0.2 1.3

a The results obtained with particles having oligonucleotide loadings of 2, 5, or 10 µmol g-1 are indicated. The hybridization efficiencies are expressed as a percentage of the value of the parent sequence.

by more than 2 orders of magnitude. Table 3 summarizes the entire data for point mutations and Table 4 for point deletions. As seen, the oligonucleotide loading on particles slightly affects the discrimination of matches and mismatches. If the loading on particles is low (2 µmol g-1), a point mutation may unequivocally be distinguished from the unmutated sequence when the length of the complementary sequence falls in the range of 1014 bases, and rather clearly even when this sequence is 16 nucleotides long. With higher oligonucleotide loadings, the complementary sequence should not be longer than 13 bases, preferably 12 bases. There are no significant differences between the effects of mutations in C‚G or A‚T base pairs. The effects of point deletions

A sandwich type hybridization assay may be carried out on microparticles in two different manners. Either the mutation-specific probe is immobilized to the particles, as described in Figure 1, or the fluorescently tagged oligomer is mutation-specific. Of these two approaches, the former was applied in this study, since it is more convenient to coat the particles with various mutation-specific probes than to prepare a large selection of fluorescently tagged oligonucleotides, each having a different sequence. The results presented above clearly show that porous polymer particles bearing 3′-attached oligonucleotide probes (1) constitute a good solid support for sandwich hybridization assays. The dynamic range achieved with the single-particle technique described is exceptionally large; the intensity of the emission signal measured from a single particle is linearly related to the concentration of the target oligonucleotide over a concentration range of 5 orders of magnitude (Figure 3). The main reasons for the wide linear range include (i) the high capacity of porous particles, (ii) the almost negligible concentration quenching of the photoluminescent europium(III) chelate used as a fluorescent marker, and (iii) the efficient elimination of the prompt background fluorescence by the use of the time-resolved mode in the measurement. The sensitivity of the sandwich assay described is high. On using 5 particles/µL, the detection limit for the target oligonucleotide is less than 1.7 pmol L-1. In other words, about 3 × 104 probes (0.05 amol) hybridized to a single particle becomes detected (signal > 2 × standard deviation). In terms of concentration, the detection limit may still be decreased; decreasing the number of particles per volume increases the signal measured from a single particle, but the time required to reach the equilibrium is simultaneously increased. To some extent, the sensitivity may also be increased by increasing the number of lanthanide chelates per probe (21). Even as it is, the sensitivity is excellent compared to that of other reported assay systems, including those based on radioactive detection (32). The major shortcoming of the methodology introduced is relatively slow hybridization. Up to 20 h was needed to reach the equilibrium on using 5 particles/µL. Our previous results (26) on direct hybridization of a fluorescently tagged oligomer on the same particles, however, suggest that the reaction time may be reduced to less than 1/10 of the value by increasing the number of particles in a given volume of the sample. Unfortunately, this simultaneously decreases the sensitivity of detection, since the same amount of target oligonucleotides is then distributed among a larger number of particles. The selectivity of hybridization is good. However, a prerequisite for this is a proper adjustment of the length of the complementary region between the target oligonucleotide and the particle-bound probe. The minimum length required for maximal hybridization efficiency somewhat depends on the loading of the probes on particles, ranging from 10 (loading of 10 µmol g-1) to 12 (loading of 2 µmol g-1) nucleotides. When this minimal length of complementarity is used, a single point mutation or deletion near the center of the hybridizing sequence decreases the efficiency of hybridization to less than 1% of the original value. However, even complementarity 3 to 4 nucleotides longer gives selectivity that

Detection of Oligonucleotide Hybridization

is almost as good (Tables 3 and 4). Accordingly, the method appears not to be too sensitive to the length of the mutation-specific probes. ACKNOWLEDGMENT

The particles employed were a generous gift of Dr. Ruth Schmid (SINTEF). The photoluminescent marker was prepared in the Laboratory of Wallac Oy (Turku, Finland) by Dr. Veli-Matti Mukkala and Dr. Harri Takalo. We also thank Mr. Harri Salo, M.Sc., for his help in oligonucleotide synthesis. Financial support from the Academy of Finland, the Research Council for Natural Sciences and Technology, is gratefully acknowledged. LITERATURE CITED (1) Southern, E. M., Case-Green, S. C., Elder, J. K., Johnson, M., Mir, K. U., Wang, L., and Williams, J. C. (1994) Arrays of complementary oligonucleotides for analysing the hybridisation behaviour of nucleic acids. Nucleic Acids Res. 22, 1368-1373. (2) Maskos, U., and Southern, E. M. (1992) Oligonucleotide hybridisations on glass supports: a novel linker for oligonucleotide synthesis and hybridisation properties of oligonucleotides synthesised in situ. Nucleic Acids Res. 20, 16791684. (3) 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. (4) Peterlinz, K. A., and Georgiadis, R. M. (1997) Observation of hybridization and dehybridization of thiol-tethered DNA using two-color surface plasmon resonance spectroscopy. J. Am. Chem. Soc. 119, 3401-3402. (5) Gingeras, T. R., Kwoh, D. Y., and Davis, G. R. (1987) Hybridization properties of immobilized nucleic acids. Nucleic Acids Res. 15, 5373-5390. (6) Albretsen, C., Kalland, K.-H., Haukanes, B.-I., Håvarstein, L.-S., and Kleppe, K. (1990) Applications of magnetic beads with covalently attached oligonucleotides in hybridization: isolation and detection of specific measles virus mRNA from a crude cell lysate. Anal. Biochem. 189, 40-50. (7) Lund, V., Schmid, R., Rickwood, D., and Hornes, E. (1988) Assessment of methods for covalent binding of nucleic acids to magnetic beads, Dynabeads, and the characteristics of the bound nucleic acids in hybridization reactions. Nucleic Acids Res. 16, 10861-10880. (8) Stimpson, D. I., Hoijer, J. V., Hsieh, W. T., Jou, C., Gordon, J., Theriault, T., Gamble, R., and Baldeschwieler, J. D. (1995) Real-time detection of DNA hybridization and melting on oligonucleotide arrays by using optical wave guides. Proc. Natl. Acad. Sci. U.S.A. 92, 6379-6383. (9) Erout, M.-N., Troesch, A., Pichot, C., and Cros, P. (1996) Preparation of conjugates between oligonucleotides and Nvinylpyrrolidone/N-acryloxysuccinimide copolymers and applications in nucleic acid assays to improve sensitivity. Bioconjugate Chem. 7, 568-575. (10) O’Donnell-Maloney, M. J., Smith, C. L., and Cantor, C. R. (1996) The development of microfabricated arrays for DNA sequencing and analysis. TIBTECH 14, 401-407. (11) Saiki, R., Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K. B., and Ehrlich, H. A. (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487-491. (12) 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. U.S.A. 93, 4913-4918. (13) Guo, Z., Guilfoyle, R. A., Thiel, A. J., Wang, R., and Smith, L. M. (1994) Direct fluorescence analysis of genetic polymorphisms by hybridization with oligonucleotide arrays on glass supports. Nucleic Acids Res. 22, 5456-5465. (14) Lamture, J. B., Beattie, K. L., Burke, B. E., Eggers, M. D., Ehrlich, D. J., Fowler, R., Hollis, M. A., Kosicki, B. B., Reich,

Bioconjugate Chem., Vol. 9, No. 3, 1998 321 R. K., Smith, S. R., Varma, R. S., and Hogan, M. E. (1994) Direct detection of nucleic acid hybridization on the surface of a charge coupled device. Nucleic Acids Res. 22, 2121-2125. (15) Soini, E. (1991) Biospecific multianalyte assay method. U.S. Patent 5,028,545. (16) Georges, J. (1993) Lanthanide-sensitized luminescence and applications to the determination of organic analytes. Analyst 18, 1481-1486. (17) Soini, E., and Lo¨vgren, T. (1987) Time-resolved fluorescence of lanthanide probes and applications in biotechnology. CRC Crit. Rev. Anal. Chem. 18, 105-154. (18) 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. (19) Diamandis, E. P., and Christopoulos, T. K. (1990) Europium chelate labels in time-resolved fluorescence immunoassays and DNA hybridization assays. Anal. Chem. 62, 1149A-1157A. (20) Elbanowski, M., and Makowska, B. (1996) The lanthanides as luminescent probes in investigations of biochemical systems. J. Photochem. Photobiol., A 99, 85-92. (21) Kwiatkowski, M., Samiotaki, M., Lamminma¨ki, U., Mukkala, V.-M., and Landegren, U. (1994) Solid-phase synthesis of chelate-labelled oligonucleotides: application in triple-color ligase-mediated gene analysis. Nucleic Acids Res. 22, 26042611. (22) Iitia¨, A., Hogdall, E., Dahlen, P., Hurskainen, P., Vuust, J., and Siitari, H. (1992) Detection of mutation ∆F508 in the cystic fibrosos gene allele-specific PCR primers and timeresolved fluorometry. PCR Methods Appl. 2, 157-162. (23) Dahlen, P., Iitia¨, A., Mukkala, V.-M., Hurskainen, P., and Kwiatkowski, M. (1991) The use of europium (Eu3+) labelled primers in PCR amplification of specific target DNA. Mol. Cell. Probes 5, 143-149. (24) Bush, C. E., Di Michele, L. J., Peterson, W. R., Sherman, D. G., and Godsey, J. H. (1992) Solid-phase time-resolved fluorescence detection of human immunodeficiency virus polymerase chain reaction amplification products. Anal. Biochem. 202, 146-151. (25) Hakala, H., and Lo¨nnberg, H. (1997) Time-resolved fluorescence detection of oligonucleotide hybridization on a single microparticle: covalent immobilization of oligonucleotides and quantitation of a model system. Bioconjugate Chem. 8, 232237. (26) Hakala, H., Heinonen, P., Iitia¨, A., and Lo¨nnberg, H. (1997) Detection of oligonucleotide hybridization on a single microparticle by time-resolved fluorometry: hybridization assays on polymer particles obtained by direct solid phase assembly of the oligonucleotide probes. Bioconjugate Chem. 8, 378384. (27) Pon, R. T. (1993) in Methods in Molecular Biology (S. Agrawal, Ed.) Vol. 20, p 467, Humana Press, Totowa, NJ. (28) Mukkala, V.-M., Helenius, M., Hemmila¨, I., Kankare, J., and Takalo, H. (1993) Development of luminescent europium(III) and terbium(III) chelates of 2,2′:6′,2′′-terpyridine derivatives for protein labelling. Helv. Chim. Acta 76, 1361-1378. (29) Lo¨vgren, T., Iitia¨, A., Hurskainen, P., and Dahlen, P. (1995) Detection of lanthanide chelates by time-resolved fluorescence. In Blotting and Sequencing (L. Kricka, Ed.) pp 350358, Academic Press, San Diego, CA. (30) Lo¨vgren, T., Heinonen, P., Lehtinen, P., Hakala, H., Heinola, J., Harju, R., Takalo, H., Mukkala, V.-M., Schmid, R., Lo¨nnberg, H., Pettersson, K., and Iitia¨, A. (1997) Sensitive bioaffinity assays on individual microparticles using timeresolved fluorometry. Clin. Chem. 43, 1937-1943. (31) Pease, A. C., Solas, D., Sullivan, E. J., Cronin, M. T., Holmes, C. P., and Fodor, S. P. A. (1994) Light-generated oligonucleotide arrays for rapid DNA sequence analysis. Proc. Natl. Acad. Sci. U.S.A. 91, 5022-5026. (32) Urdea, M. S., Warner, B. D., Running, J. A., Stempien, M., Clyne, J., and Horn, T. (1988) A comparison of nonradioisotopic hybridization assay methods using fluorescent, chemiluminescent and enzyme labeled synthetic oligodeoxyribonucleotide probes. Nucleic Acids Res. 16, 4937-4956.

BC9701860