Molecular Beacon-Style Hybridization Assay for ... - ACS Publications

Molecular Beacon-Style Hybridization Assay for. Quantitative Analysis of Surface Invasive Cleavage. Reactions. Matthew R. Lockett, Michael R. Shortree...
0 downloads 0 Views 253KB Size
Download PDF
Anal. Chem. 2007, 79, 6031-6036

Molecular Beacon-Style Hybridization Assay for Quantitative Analysis of Surface Invasive Cleavage Reactions Matthew R. Lockett, Michael R. Shortreed, and Lloyd M. Smith*

Department of Chemistry, University of Wisconsin, 1101 University Avenue, Madison, Wisconsin 53706

A hybridization-based FRET format for the scoring of SNPs in surface invasive cleavage reactions is described. In early versions of the surface invasive cleavage reaction, dual-labeled oligonucleotides, containing both a quencher moiety and a fluorophore, were attached to the substrate. The invasive cleavage reaction cleaved the DNA strand between the two, resulting in an increase in fluorescence signal due to the separation of the quencher from the fluorophore. A limitation of this assay format was the relatively low quenching efficiency of 84% obtained, as well as the complexity of synthesis for these dual-labeled probes. In the assay format presented here, singly labeled oligonucleotides are employed, with the quencher and fluorophore placed on separate complementary oligonucleotides. The surface-bound probe is terminated at the 5′ end with the quencher and the complement is terminated at its 3′ end with a fluorophore, such that upon hybridization the two are positioned directly across from one another in the duplex. Quenching efficiency in this “molecular beacon” format is increased to 88%, much closer to the 91% level that has been reported for molecular beacon assays. A second benefit of the approach described here is that the portion of probe oligonucleotide that is removed by the enzyme is shorter, thus increasing the rate of probe cleavage. The improved quenching efficiency and increased probe cleavage rate result in a lower detection limit for the assay. A theoretical model of the FRET process occurring on the surfaces was used to relate the observed surface fluorescence intensity to the progress of the invasive cleavage reaction. Single nucleotide polymorphisms (SNPs) serve as excellent genetic markers because of their long-term stability and high frequency of occurrence.1 They are often used for the identification and characterization of gene mutations that are involved in biologically significant functions. High-throughput, multiplexed SNP scoring is employed in many assays where linkage is sought between SNPs and disease-associated genes. Surface-based DNA assays have a demonstrated ability for multiplex SNP scoring in complex biological mixtures,2 and the development of microarray

technologies has made it possible to perform as many as 780 000 oligonucleotide assays on a single surface.3,4 Analytical platforms requiring no DNA amplification would eliminate many complexities and possible sources of error arising from target amplification procedures such as PCR. The invasive cleavage reaction is a viable SNP scoring option and has a demonstrated ability for accurate SNP scoring on unamplified genomic DNA both in solution5,6 and on surfaces.7-9 The invasive cleavage reaction relies on formation of a secondary structure from the hybridization of a downstream primary probe and an upstream invader oligonucleotide with a single-stranded DNA target. When the upstream invader and downstream probe oligonucleotides have an overlap of one or more nucleotides, an unhybridized flap is produced (Figure 1) and becomes the substrate for a flap endonuclease (FEN) enzyme. The FEN enzyme cleaves the 5′ flap from the probe oligonucleotide after its first paired base.10 When the probe and target oligonucleotides are complementary to one another at the point of overlap, the flap is cleaved at a rate much greater than if the probe and target oligonucleotides are noncomplementary to one another at the point of overlap. Cycling occurs when the target molecule dissociates from the cleaved probe and then binds to other probe molecules leading to addition cleavage events. The structure-dependent cleavage forms the basis for a highly sensitive and selective SNP scoring assay.11 Two strategies for performing the invasive cleavage reaction in a DNA array format have been developed.7-9 In the first, the probe oligonucleotide is attached to the surface and the invader oligonucleotide is in a buffered solution with the target and enzyme. In the second, the invader and probe oligonucleotides are coimmobilized on the surface and a buffered solution contain(3) (4) (5) (6) (7)

(8) (9) (10)

* To whom correspondence should be addressed. [email protected]. (1) Stephens, J. C.; et al. Science 2001, 293 (5529), 489-493. (2) Hoheisel, J. D. Nat. Rev. Genet. 2006, 7 (3), 200-210. 10.1021/ac070424c CCC: $37.00 Published on Web 06/27/2007

E-mail:

© 2007 American Chemical Society

(11)

Singh-Gasson, S.; et al. Nat. Biotechnol. 1999, 17 (10), 974-978. Warren, C. L.; et al. Proc. Natl. Acad. Sci. U.S.A. 2006, 103 (4), 867-872. Hall, J. G.; et al. Proc. Natl. Acad. Sci. U.S.A. 2000, 97 (15), 8272-8277. Lyamichev, V.; et al. Nat. Biotechnol. 1999, 17 (3), 292-296. Lu, M. C.; Hall, J. G.; Shortreed, M. R.; Wang, L.; Berggren, W. T.; Stevens, P. W.; Kelso, D. M.; Lyamichev, V.; Neri, B.; Skinner, J. L.; Smith, L. M. J. Am. Chem. Soc. 2002, 124 (27), 7924-7931. Chen, Y.; Shortreed, M. R.; Olivier, M.; Smith, L. M. Anal. Chem. 2005, 77 (8), 2400-2405. Chen, Y.; Shortreed, M. R.; Peelen, D.; Lu, M. Smith, L. M. J. Am. Chem. Soc. 2004, 126 (10), 3016-3017. Lyamichev, V.; Brow, M. A. D.; Dahlberg, J. E. Science 1993, 260 (5109), 778-783. Lyamichev, V. I.; Kaiser, M. W.; Lyamicheva, N. E.; Vologodshii, A. V.; Hall, J. G.; Na, W.-P.; Allawi, H. T.; Neri, B. P. Biochemistry 2000, 39 (31), 9523-9532.

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007 6031

Figure 1. Schematic diagram of the secondary structure needed for the invasive cleavage reaction. The 3′ end of the upstream invader oligonucleotide is designed to overlap at least one base into the downstream duplex formed between the primary probe and DNA target oligonucleotides. The unpaired region on the 5′ end of the primary probe, flap structure, along with an immediate downstream paired nucleotide can then be removed by the FEN enzyme. Absolute complementarity between the primary probe and target oligonucleotides at the position of overlap is required for efficient enzymatic cleavage.

ing the target and enzyme is added to initiate the reaction. The second format lowers the possibility of cross-reactivity and eliminates the need to synthesize many different invader oligonucleotides. In early versions of the surface invasive cleavage reaction, a fluorophore and quencher moiety were attached to the probe oligonucleotide during synthesis.7 A quencher (dabcyl) moiety was placed at the free 5′ end of the probe and a fluorophore (fluorescein) was placed about 3-5 nt away, flanking the point of cleavage. In the uncleaved state, the proximity of the quencher to the fluorophore enabled relatively efficient fluorescence quenching. Upon cleavage, the fluorophore and quencher were spatially separated and an increase in the fluorescence was observed (Figure 2a). The 3-5 nt spacing between the fluorophore and the quencher was selected to maximize energy-transfer efficiency in the dual-labeled probe. However, this configuration does not provide quenching efficiencies as large as those obtained for molecular beacons wherein the fluorophore and quencher are positioned directly adjacent to one another, similar to the approach described here. Another significant parameter in the assay design is the flap length, as longer flaps decrease the cleavage rate.12 In the present work, these issues are addressed by placing the quencher and fluorophore moieties on separate oligonucleotide strands. Surface-bound probe oligonucleotides are synthesized with a terminal 5′-dabcyl quencher moiety. A complementary oligonucleotide containing a 3′-fluorescein is also synthesized and used to create a FRET pair upon hybridization. In this scenario, the fluorophore and quencher are positioned directly adjacent to one another analogous to the molecular beacon system,13 increasing the quenching efficiency. Prior to the invasive cleavage, the fluorophore-containing complements are hybridized to the surfacebound probe oligonucleotides, and a background fluorescence signal is recorded with a fluorescence scanner. The complement is stripped from the surface by treatment with urea. A buffered solution containing the enzyme, target, and invader oligonucleotides is added to the surface and the invasive cleavage reaction ensues, causing the quencher moiety to be cleaved from the surface where the probe and target oligonucleotides are complementary to one another at the SNP position. The surface is then washed and rehybridized with the fluorescent complement, (12) Kaiser, M. W.; et al. J. Biol. Chem. 1999, 274 (30), 21387-21394. (13) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14 (3), 303-308.

6032

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

following which the fluorescence signal is obtained with a fluorescent scanner. Comparison of the before and after images reveals the features that contained probe oligonucleotides cleaved in the reaction (Figure 2b). In SNP scoring assays, although the final result for the genotype at each SNP location is a qualitative yes/no answer, quantitative data are needed to make such determinations statistically robust. In this report, we present a simplified method for quantification of the signal from surface invasive cleavage reactions used for SNP scoring. The approach provides increased sensitivity over earlier versions of the assay. The change in observed fluorescence signal is related to the efficiency of the surface invasive cleavage using a previously developed theoretical model7 of quenching for surface-bound fluorophores and quenchers. EXPERIMENTAL SECTION Oligonucleotides. All oligonucleotides used in these experiments were synthesized by Integrated DNA Technologies (Coralville, IA) using standard phosphoramidite chemistries and were based upon the W1282X (TfC) mutation in the CFTR gene (Table 1). We note that W1282X is formally a mutation, rather than a SNP, because SNPs are defined as polymorphisms for which the minor allele frequency is at least 1%.14 However, as a single nucleotide variation of substantial medical consequence, it was chosen here as a model system. Surface-bound probe oligonucleotides were synthesized with a free 3′-thiol (C3 S-S) moiety separated by 10 18-atom hexaethylene glycol spacer (spacer 18) moieties. Thiol modifications allow for the covalent attachment of the probe oligonucleotides to the surface through a maleimide linker while the spacers provide enough distance between the surface and the oligonucleotides to ensure optimal hybridization.15 Two versions of the T- and C-allele oligonucleotides were synthesized: a 5′-dabcyl-containing oligonucleotide (dabcyl-T-allele and dabcyl-C-allele) and a control oligonucleotide containing a 5′-OH (control-T-allele and controlC-allele). Complementary oligonucleotides to the T- and C-alleles were synthesized with a 3′,6-carboxyfluorescein moiety. All 3′-thiol-modified oligonucleotides were deprotected and purified using previously reported methods.16 Each oligonucleotide was deprotected for 30 min in 20 µL of 100 mM dithiothreitol solution (100 mM triethanolamine (TEA), pH 7.0) and purified by reverse-phase binary gradient elution HPLC (SCL-10ADVP Shimadzu, Columbia, MD). Purified oligonucleotides were stored dry, under nitrogen, until needed. Upon use, each oligonucleotide was reconstituted in 100 mM TEA buffer, pH 7.0. All oligonucleotide concentrations were determined by absorption measurements at 260 nm (HP8453 UVVIS; Santa Clara, CA). Surface Preparation. Thiol-modified oligonucleotides were coupled to a gold thin film via chemical modifications described in detail elsewhere.17 A self-assembled monolayer (SAM) of the alkanethiol, 11-amino-1-undecanethiol (Dojindo, Gaithersburg, MD), was formed on a gold-coated glass substrate (Evaporated Metal Films, Ithaca, NY). The SAM free amine groups were then (14) Brookes, A. J. Gene 1999, 234 (2), 177-186. (15) Guo, Z.; et al. Nucleic Acids Res. 1994, 22 (24), 5456-5465. (16) Brockman, J. M.; Frutos, A. G.; Corn, R. M. J. Am. Chem. Soc. 1999, 121 (35), 8044-8051. (17) Smith, E. A.; Wanat, M. J.; Cheng, Y.; Barreira, S. V. P.; Frutos, A. G.; Corn, R. M. Langmuir 2001, 17 (8), 2502-2507.

Figure 2. (a) Schematic of the surface invasive cleavage reaction FRET readout system, as reported in previous publications.7-9 Probe oligonucleotides containing an internal fluorescein and a terminal dabcyl quencher moiety are coupled to the surface. Fluorescence intensity values are recorded before the reaction to determine the quenched signal. The invasive cleavage structure is formed, the reaction is carried out, and the surface is washed. An increase in fluorescence intensity indicates cleavage of the dabcyl-containing flap. (b) Schematic of the surface invasive cleavage reaction utilizing the hybridization-based FRET readout system described in the present work. Probe oligonucleotides containing 5′-dabcyl quencher moieties are coupled to the surface. Complementary oligonucleotides containing a 3′-fluorescein are then hybridized to the probe oligonucleotides, and an initial fluorescence measurement (quenched) is taken. The oligonucleotides are dehybridized, the invasive cleavage structure is formed, and the reaction is carried out. The surface is washed, and the probe oligonucleotides are again hybridized with their fluorescent complements. An increase in fluorescence intensity indicates that the target molecule contained the oligonucleotide of interest and that enzymatic cleavage removed the nucleotide flap containing the dabcyl quencher group.

reacted with a heterobifunctional linker, sulfosuccinimidyl 4-(Nmaleimidomethyl) cyclohexane-1-carboxylate (Pierce, Milwaukee, WI), creating a maleimide-terminated surface. Each probe oligonucleotide was reduced and reconstituted to a final concentration of 0.8 mM (100 mM TEA buffer, pH 7.0), coupled to the maleimide surface by depositing 0.2-µL droplets and allowed to react for 12 h in a humid chamber. Excess and nonspecifically bound DNA was removed by incubating the surfaces in 1× SSPE (10 mM NaH2PO4, 150 mM NaCl, 1 mM EDTA, pH 7.4) for 30 min at 37 °C. The surfaces were further treated with an 8 M urea solution for 30 min at room temperature followed by thorough rinsing with water. Invasive Cleavage Reaction. Before the surface invasive cleavage reaction, the stability of the oligonucleotide-modified surface was tested through a series of two hybridization/ dehybridization cycles. Probe oligonucleotides were hybridized by applying 40 µL of a 0.5 µM solution (1×SSPE) of the fluorescently labeled complement for 30 min in a humid chamber. Excess complement was removed by the incubation of the surface in 1×SSPE buffer for 10 min at 37 °C, followed by washing with deionized water. The fluorescence intensities were measured with a GeneTAC UC4X4 scanner (Genomic Solutions, Ann Arbor, MI). Surfaces were incubated in an 8 M urea solution at room temperature for 30 min to dehybridize the probe oligonucleotides, thoroughly rinsed with water, and rescanned to confirm there was no complement remaining.

The invasive cleavage reaction conditions used in these experiments were optimized in previous work.7 To each surface 200 µL of the invasive cleavage reaction solution was added followed by incubation in a humid chamber at 58.5 °C for 4 h. The reaction solution contained 5 nmol of the upstream invader, between 0 and 1 pmol of the target oligonucleotide, buffer (20 mM MOPS at pH 7.5 and 7.5 mM MgCl2), and Afu FEN 1 enzyme (cleavase X, 5 ng, provided by Third Wave Technologies, Madison, WI). After the reaction, the surfaces were thoroughly rinsed, the probe oligonucleotides were hybridized with their fluorescent complements, and the fluorescence intensities were determined. Changes in the fluorescence intensity of the hybridized surface before and after the invasive cleavage reaction were determined by comparing the absolute fluorescence intensities, corrected to remove any nonspecific, background fluorescence. Changes in fluorescence intensity of the targeted probe (dabcyl-T- and controlT-allele) oligonucleotide were used to determine the percent cleavage. Measurement Variation. The reproducibility of the measurement of fluorescence intensity from labeled complement hybridized to surface-bound probe molecules was determined as follows. A substrate was prepared as described above containing features with immobilized control-T-allele and control-C-allele probe oligonucleotides. This substrate was then treated with fluorescent complement for 30 min, rinsed briefly with 1× SSPE, and Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

6033

Table 1. Oligonucleotides (5′ f 3′) Based upon the W1282X (T_C) Mutation in the CFTR Genea

a Each probe oligonucleotide contains 10 hexaethylene glycol 18-atom spacer moieties and a 3′ thiol, allowing for proper coupling to the surface. Complement oligonucleotides were modified with a 3′ fluorescein moiety, allowing for hybridization detection by an increased fluorescence signal. Both target and upstream invader oligonucleotides were synthesized without modification. The bases at the polymorphic site are T or C (underlined). In both reaction sets, the 3′ terminal nucleotide of the upstream invader oligonucleotide overlaps (or invades) the first base pair of the downstream probe-target duplex (A-T for the T-allele target and G-C for the C-allele target). The 5′-nuclease specifically cleaves the probe oligonucleotides, releasing the 5′-dabcyl quencher moiety.

incubated in 37 °C 1× SSPE for 30 min, followed by the measurement of the average fluorescence signal from each feature with a fluorescence scanner. The bound complements were eluted from the surface by soaking the substrate in 8 M urea for 30 min followed by a brief rinse with DI water. The process of hybridization, rinsing, fluorescence measurement, and elution was repeated seven more times for a total of eight measurement cycles. Invasive Cleavage Reaction Simulation. A fraction of the 5′-dabcyl groups are removed from the surface-bound probe oligonucleotides during the course of the invasive cleavage reaction. This result was simulated by preparing surfaces with features containing mixtures of probe oligonucleotides terminated either in 5′-dabcyl or in 5′-OH. These surfaces provide calibration standards for the quantitative determination of cleavage and for measuring the quenching efficiency (see Results and Discussion below). A series of 0.8 µM mixtures containing the 5′-OH (controlT-allele) and 5′-dabcyl (dabcyl-T-allele) probe oligonucleotides were prepared and coupled to the surface. The ratios of 5′-OH to 5′-dabcyl were 5:0, 4:1, 3:2, 1:1, 2:3, 1:4, and 0:5; corresponding to 0, 20, 40, 50, 60, 80, and 100% probe cleavage, respectively. A second series of mixtures was prepared in 2% cleavage increments to provide finer resolution, from 0 to 20 and 80 to 100%. Each surface was prepared as described above in triplicate with 0.2-µL spots. Fluorescence intensity measurements after hybridization were made with the GeneTAC UC4X4 scanner. RESULTS AND DISCUSSION Measurement Variation. The average fluorescence intensity measured following hybridization of fluorescently tagged complement to surface-bound probe oligonucleotides was both reproducible and stable. Eight cycles of hybridization, rinsing, fluorescence measurement, and elution resulted in an average signal of 6034

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

13750 ( 370 counts/pixel, with a relative standard deviation of 1.9%. The signal intensity decreased by 2.1% over the course of the experiment (0.26%/hybridization cycle), which is attributed to a minor loss of probe oligonucleotides from the surface. Quantitative SNP Scoring. To demonstrate the results obtained after the surface invasive cleavage reaction, a series of sample surfaces were prepared with features containing mole fractions (x) ranging from 0 to 1 of a 5′-OH probe oligonucleotide mixed with a 5′-dabcyl probe oligonucleotide. The 5′-OH probe oligonucleotide corresponds to the cleaved product of the surface invasive cleavage reaction, and the 5′-dabcyl probe oligonucleotide corresponds to the uncleaved probe. Each surface was hybridized with its fluorescently tagged complement, and the fluorescence intensity of each feature was measured. These measured fluorescence intensities were divided by the average fluorescence intensity obtained from features with a 5′-OH probe oligonucleotide mole fraction of 0, to yield the normalized fluorescence intensities I(x). The normalized fluorescence intensities vary from 1 (IFQ ) I(0), corresponding to the completely dabcyl-modified surface), to ∼7.5 (IF ) I(1), corresponding to the surface completely lacking dabcyl groups). These measured fluorescence values are plotted in Figure 4 and vary nonlinearly with mole fraction in a manner similar to that observed previously.7 In previous work, a model was developed to understand the nonlinear relationship between the observed surface fluorescence intensities and the mole fraction of cleaved probe. In this model, the oligonucleotides are assumed to form a hexagonal closepacked monolayer on the gold surface with an intermolecular spacing of 50 Å and a density of 5 × 1012 molecules/cm2. Under these conditions, the relationship between the normalized fluorescence intensity and the mole fraction (x) can be expressed in

Figure 3. Observed fluorescence intensity changes as a function of probe cleavage fraction. A series of samples were prepared to represent different stages in the surface invasive cleavage reaction (see Experimental Section). The data points shown are averages from four separate experiments; error bars correspond to the standard deviation of the measured fluorescence intensities. (a) 20% increments of probe cleavage between 0 and 100%; (b) 2% increments of probe cleavage between 0 and 20%; (c) 2% increments of probe cleavage between 80 and 100%. These data were fit (dotted line) using the relationship developed below: I(x) ) xIF/(1 + a(1 - x)) + (1 - x)IFQ comparing the probe cleavage fraction (x) to the measured IF and IFQ fluorescence intensities with an adjustable (a) parameter.

Figure 4. Fluorescence images of the invasive cleavage reaction results targeting the mutant type (T-) allele of the W1282X mutation in the CFTR gene. (a) Schematic of the surface used; containing the dabcyl-T-allele (a), control-T-allele (b), and dabcyl-C-allele (c), and the controlC-allele probe oligonucleotides outlined in Table 1. (b, c) Fluorescence intensity observed before and after the invasive cleavage reaction, using 10.0 fmol of the target. Oligonucleotides were hybridized with their fluorescently labeled complement and their fluorescence intensities measured. (d) Difference image of the changes in fluorescence intensity resulting from the specific cleavage of the mutant type (T-) allele.

terms of IF, IFQ, and an adjustable parameter (a): I(x) ) xIF/(1 + a(1 - x)) + (1 - x)IFQ. The adjustable parameter is related to the Fo¨rster distance (Ro) and the intermolecular spacing (l) of the individual oligonucleotides, a ) 6(Ro/l)6. The above equation was used to fit the data shown in Figure 3, yielding a value of a ) 0.513. This corresponds to a Fo¨rster distance of Ro ) 32 Å, within the usual range for Fo¨rster radii (10-100 Å).18 Figure 3 shows that there is excellent agreement between the fluorescence data obtained and the model. A quantitative determination of the degree of cleavage that occurs in the surface invasive cleavage reaction is important for setting the thresholds necessary for assignment of SNP genotype in a statistically meaningful fashion. The above model makes this possible, in that it allows determination of the degree of probe cleavage based on the fluorescence intensity increases observed subsequent to hybridization with the fluorescently tagged complement. The degree of probe cleavage was determined using the

above model for five different invasive cleavage reactions using varying amounts of DNA target (0, 1, 10, 100, and 1000 fmol; Figure 5). The fluorescence intensity for the invasive cleavage reaction with 1 fmol (50 pM) of target yielded a 2.91-fold fluorescence intensity increase, corresponding to 16% probe cleavage. Increases in the amount of target to 10, 100, and 1000 fmol yielded 43.3, 71.5, and 94.7% cleavage, respectively. Detection Limit. The limit of detection for this assay is defined as the mole fraction of cleaved probe molecules producing a normalized fluorescence intensity, upon hybridization with fluorescently labeled complement, equal to the background fluorescence IFQ plus 3σ of the noise.19 The value of IFQ is equal to 1 by definition (see above), and the value of σ was found to have a value of 0.08 (0.11% RSD), taken as the standard deviation of the background fluorescence in three distinct 0.28-mm2 regions (the size of a feature made by spotting 0.2 µL of probe solution) of the sample where no DNA features were present. The detection limit

(18) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic/Plenum Publishers: New York, 1999/

(19) Thomsen, V.; Schatzlein, D.; Mercuro, D. Spectroscopy 2003, 18 (12), 112114.

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

6035

Figure 5. Increase in fluorescence signal as a function of synthetic target present during the surface invasive cleavage reaction. Signal increase corresponds to the number of probe oligonucleotides cleaved during the surface invasive cleavage reaction. Wild type (C-allele) and mutant type (T-allele) probe oligonucleotides were coupled onto each surface to demonstrate the selectivity of the flap endonuclease enzyme this assay. Calculations of the percentage of wild type probe cleaved were done using the calibration curves developed in Figure 4. The error bars correspond to the standard deviation of the fluorescent signal observed for each surface/probe concentration.

thus has a value of 1.24, which corresponds to 6% probe cleavage on the surface (from Figure 3b). A linear interpolation from the data in Figure 5 shows that 6% probe cleavage would result from 0.4 fmol of target, which is approximately a 2.5-fold lower detection limit than reported for a previous FRET-based detection format.7,8,20 The lower detection limit of this assay format as compared with the dual-labeled probe format7,20 is attributed to the following: an increase in cycling brought on by shortening the length of the cleavable flap;6,12 an increase in energy-transfer efficiency (see below); significantly lower variability in the measured fluorescence; and lower nonspecific cleavage. In the current system, 10.0 fmol of target resulted in 5.4% probe cleavage/h (0.12 cleavages per target per hour) while the dual-labeled system achieved 3.2% probe cleavage/h (0.07 cleavages per target per hour) using the same probe concentration.7 The signal increase due to nonspecific cleavage of the nontargeted C-allele probe oligonucleotide following the invasive cleavage reaction with 10.0 fmol of T-allele target was 3.21%. Comparison with the previous, dual-labeled probe approach is difficult as in that work a decrease in signal of 10% was observed, due to loss of probe from the surface. Additionally, the relative standard deviation of measured fluorescence for the current approach was 4.3% compared to the earlier dual-labeled probe approach that had a relative standard deviation of 15.2%. Energy-Transfer Efficiency. The improved detection limit of this assay is due to the efficient energy transfer between the fluorophore and quencher moieties. The energy-transfer efficiency can be determined using the normalized fluorescence intensity (IF and IFQ) values obtained in Figure 3 using the following (20) Lu, M. C.; et al. Biopolymers 2004, 73 (5), 606-613. (21) Tyagi, S.; Bratu, D. P.; Kramer, F. R. Nat. Biotechnol. 1998, 16 (1), 4953.

6036

Analytical Chemistry, Vol. 79, No. 15, August 1, 2007

relationship: E ) 1 - (IFQ/IF).21 Hybridization of the singly labeled, complementary oligonucleotides allow the quencher (dabcyl) and fluorophore (fluorescein) moieties to be located directly across the duplex, yielding an E ) 0.88 value. This is higher than previously reported values (E ) 0.84) for the duallabeled probe format, which required a 3-5 nt spacing between the fluorescein and dabcyl quencher to prevent unwanted enzymatic recognition and cleavage.7 Both formats have a lower energy-transfer efficiency than those found in solution, E ) 0.91.18 CONCLUSION In the present work, an alternative, hybridization-based FRET format for the scoring of SNPs in the surface invasive cleavage reaction (Figure 2b) is described. This assay format provides several advantages over the previously described dual-labeled probe format. The probe oligonucleotides are simpler, as only a single dye is utilized. The proximity of the fluorophore and quencher provides increased quenching efficiency, which lowers background. The decreased length of the flap enables increased cycling of target. The assay detection limits are decreased due to these effects. A theoretical model of the FRET process occurring on the surfaces was used to relate the observed surface fluorescence intensity to the progress of the invasive cleavage reaction. ACKNOWLEDGMENT We thank Third Wave Technologies (TWT) for providing the FEN enzyme used in these experiments. L.M.S. has a financial interest in TWT. This work was supported by NIH Grants R01HG02298 and R01HG003275. Received for May 5, 2007. AC070424C

review

March

1,

2007.

Accepted