Anal. Chem. 2010, 82, 7684–7690
Combination of DNA Ligase Reaction and Gold Nanoparticle-Quenched Fluorescent Oligonucleotides: A Simple and Efficient Approach for Fluorescent Assaying of Single-Nucleotide Polymorphisms Hao Wang,†,‡ Jishan Li,† Yongxiang Wang,†,‡ Jiangyu Jin,†,‡ Ronghua Yang,*,† Kemin Wang,*,† and Weihong Tan† State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, College of Biology, Hunan University, Changsha, 410082, China, and Beijing National Laboratory for Molecular Sciences, College of Chemistry and Molecular Engineering, Peking University, Beijing, 100871, China A new fluorescent sensing approach for detection of single-nucleotide polymorphisms (SNPs) is proposed based on the ligase reaction and gold nanoparticle (AuNPs)-quenched fluorescent oligonucleotides. The design exploits the strong fluorescence quenching of AuNPs for organic dyes and the difference in noncovalent interactions of the nanoparticles with single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA), where ssDNA can be adsorbed onto the surface of AuNPs while dsDNA cannot be. In the assay, two half primer DNA probes, one being labeled with a dye and the other being phosphorylated, were first incubated with a target DNA template. In the presence of DNA ligase, the two captured ssDNAs are linked for the perfectly matched DNA target to form a stable duplex, but the duplex could not be formed by the single-base mismatched DNA template. After addition of AuNPs, the fluorescence of dye-tagged DNA probe will be efficiently quenched unless the perfectly matched DNA target is present. To demonstrate the feasibility of this design, the performance of SNP detection using two different DNA ligases, T4 DNA ligase and Escherichia coli DNA ligase, were investigated. In the case of T4 DNA ligase, the signal enhancement of the dyetagged DNA for perfectly matched DNA target is 4.6-fold higher than that for the single-base mismatched DNA. While in the presence of E. coli DNA ligase, the value raises to be 30.2, suggesting excellent capability for SNP discrimination. Though genetic variation is the intrinsic factor for evolution,1 many of the variations such as single-nucleotide polymorphisms (SNPs), which are common among different genotypes of one * To whom correspondence should be addressed. E-mail:
[email protected] (R.Y.);
[email protected] (K.W.). Fax: +86-731-88822523. † Hunan University. ‡ Peking University. (1) Hawks, J.; Wang, E. T.; Cochran, G. M.; Harpending, H. C.; Moyzis, R. K. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 20753–20758.
7684
Analytical Chemistry, Vol. 82, No. 18, September 15, 2010
species, can result in terrible diseases2 and cancer.3 Since the completion of human genome draft sequence, more than 100 million of SNPs of various species have been submitted into the SNP database of NIH.4 Highly sensitive identification of these SNPs is significant for disease prognosis and treatment with genetic predisposition.5 In the past decades, various different protocols have been advanced for SNPs detection.6 One design type, for example, using allele-specific DNA microarray7 or melting temperature analysis8 of ordinary DNA probes to produce the differences of thermodynamic parameters9 and other properties,10 could be used to confirm the existence of SNPs. The other strategy to discriminate SNP is the uses of designed DNA probes with special structure11,12 or unnatural nucleobases,13 such as triple-stem molecular beacons.14 In addition, the nucleic acid enzymes have been applied for detecting SNPs. Coupled with DNA probes, MutS protein (DNA mismatch binding protein),15-17 nucleases,18,19 polymerases,20,21 and ligases22 have been successfully used for SNP identification with high efficiency. Especially, (2) Strausberg, R. L.; Buetow, K. H.; Emmert-Buck, M. R.; Klausner, R. D. Trends Genet. 2000, 16, 103–106. (3) Imyanitov, E. N. Hum. Genet. 2009, 125, 239–246. (4) http://www.ncbi.nlm.nih.gov/projects/SNP. (5) Savas, S.; Liu, G. Hum. Mutat. 2009, 30, 1369–1377. (6) Cottingham, K. Anal. Chem. 2004, 76, 179 A181 A. (7) Gerion, D.; Chen, F.; Kannan, B.; Fu, A.; Parak, W. J.; Chen, D. J.; Majumdar, A.; Alivisatos, A. P. Anal. Chem. 2003, 75, 4766–4772. (8) Crews, N.; Wittwer, C. T.; Montgomery, J.; Pryor, R.; Gale, B. Anal. Chem. 2009, 81, 2053–2058. (9) Russom, A.; Haasl, S.; Brookes, A. J.; Andersson, H.; Stemme, G. Anal. Chem. 2006, 78, 2220–2225. (10) Wei, F.; Chen, C.; Zhai, L.; Zhang, N.; Zhao, X. S. J. Am. Chem. Soc. 2005, 127, 5306–5307. (11) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303–308. (12) Xiao, Y.; Lou, X.; Uzawa, T.; Plakos, K. J. I.; Plaxco, K. W.; Soh, H. T. J. Am. Chem. Soc. 2009, 131, 15311–15316. (13) Okamoto, A.; Kanatani, K.; Saito, I. J. Am. Chem. Soc. 2004, 126, 4820– 4827. (14) Xiao, Y.; Plakos, K. J. I.; Lou, X.; White, R. J.; Qian, J.; Plaxco, K. W.; Soh, H. T. Angew. Chem., Int. Ed. 2009, 48, 4354–4358. (15) Stanisławska-Sachadyn, A.; Sachadyn, P. Acta Biochim. Pol. 2005, 52, 575– 583. (16) Sun, H. B.; Yokota, H. Anal. Chem. 2000, 72, 3138–3141. (17) Gong, H.; Zhong, T.; Gao, L.; Li, X.; Bi, L.; Kraatz, H.-B. Anal. Chem. 2009, 81, 8639–8643. (18) Chen, Y. T.; Hsu, C. L.; Hou, S. Y. Anal. Biochem. 2008, 375, 299–305. 10.1021/ac101503t 2010 American Chemical Society Published on Web 08/20/2010
the ligation reaction is superior to the two formers with respect to specificity as the high discrimination ability of ligases to basemismatched sequences.23-25 Although these approaches have made great contributions for SNP identification, limitations, such as insufficient discrimination capability for SNP, time-consuming, and relatively high cost, still exist. Therefore, it should be desirable to develop new methods to solve these problems. There currently has been interest in integration of nanotechnology with biology and chemistry to develop new analytical tools for life science and biotechnology. Recently, to get more robust, efficient, and sensitive SNPs analysis, DNA ligase has been combined with versatile metal nanoparticles to build various nanoassembly for colorimetric,26-28 surface plasmon resonance (SPR),29 and surface enhanced Raman scattering (SERS)30-based SNPs assays. Although they are highly selective and sensitive, covalent modification of mercapto group-labeled DNA probes to the surfaces of nanoparticles is a laborious and time-consuming procedure. More complicated steps, such as silver stain28 and washing,29 were also involved, which might bring additional influence to the detection results. Therefore, it is of importance to eliminate the troublesome process of covalent modification to develop a more convenient and sensitive SNPs assay technique which has both the advantages of DNA ligase and nanoparticles. It was discovered that there is a strong and reversible noncovalent interaction between gold nanoparticles (AuNPs) and nucleic acid molecules that depends on the structures of the DNA sequence.30 Single-stranded DNA (ssDNA) or RNA with free bases are looser to coordinate with gold atoms, while the more rigid double-stranded DNA (dsDNA) or quadruplex nucleic acids have less interaction with AuNPs as their bases form hydrogen bonds.31 This phenomenon inspires the development of label-free colorimetric methods for the kinds of molecules based on the SPR absorption of AuNPs.32,33 At the same time, with the quantum effect and high surface/volume ratio, the noble metal nanoparticles exhibited increased photochemical activity compared with the bulk metals. The fluorescence of most dyes can be quenched efficiently (19) (20) (21) (22) (23)
(24)
(25) (26) (27) (28) (29) (30) (31) (32) (33)
Li, K.; Liu, B. Anal. Chem. 2009, 81, 4099–4105. Duan, X.; Li, Z.; He, F.; Wang, S. J. Am. Chem. Soc. 2007, 129, 4154–4155. Liu, G.; Lin, Y. J. Am. Chem. Soc. 2007, 129, 10394–10401. Landegren, U.; Kaiser, R.; Sanders, J.; Hood, L. Science 1988, 241, 1077– 1080. Conze, T.; Shetye, A.; Tanaka, Y.; Gu, J. J.; Larsson, C.; Goransson, J.; Tavoosidana, G.; Soderberg, O.; Nilsson, M.; Landegren, U. Annu. Rev. Anal. Chem. 2009, 2, 215–239. Shi, C.; Eshleman, S. H.; Jones, D.; Fukushima, N.; Hua, L.; Parker, A. R.; Yeo, C. J.; Hruban, R. H.; Goggins, M. G.; Eshleman, J. R. Nat. Methods 2004, 1, 141–147. Toubanaki, D. K.; Christopoulos, T. K.; Ioannou, P. C.; Flordellis, C. S. Anal. Chem. 2008, 81, 218–224. Li, J.; Chu, X.; Liu, Y.; Jiang, J. H.; He, Z.; Zhang, Z.; Shen, G.; Yu, R. Q. Nucleic Acids Res. 2005, 33, e168. Li, J.; Jiang, J. H.; Xu, X. M.; Chu, X.; Jiang, C.; Shen, G.; Yu, R. Q. Analyst 2008, 133, 939–945. Xue, X.; Xu, W.; Wang, F.; Liu, X. J. Am. Chem. Soc. 2009, 131, 11668– 11669. Li, Y. A.; Wark, A. W.; Lee, H. J.; Corn, R. M. Anal. Chem. 2006, 78, 3158– 3164. Huh, Y. S.; Lowe, A. J.; Strickland, A. D.; Batt, C. A.; Erickson, D. J. Am. Chem. Soc. 2009, 131, 2208–2213. Wang, H.; Yang, R. H.; Yang, L.; Tan, W. H. ACS Nano 2009, 3, 2451– 2460. Li, H.; Rothberg, L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 14036–14039. Zhao, W.; Brook, M. A.; Li, Y. ChemBioChem 2008, 9, 2363–2371.
when the dye and the AuNPs are held in close proximity34 because of energy or electron transfer.35 On the basis of the fluorescence quenching effect of AuNPs, simple and sensitive fluorescent sensing approaches using single dye-labeled DNA have been developed for the detection of DNA hybridization,36 thrombin,37 and Hg2+ ions.38 Although the difference of the noncovalent interaction of ssDNA and dsDNA with AuNPs will offer a potential approach for highly sensitive SNPs detection, the main drawback based on the single base-pair mismatch types is the low discrimination capability for SNPs. Thus exponential amplification39,40 of the target sequence or the considerably high temperature26,36 was necessary for the pristine DNA probes to recognize the right nucleic acid molecules. Protein recognition of SNPs based on the oligonucleotide ligation reaction offers high specificity because it requires two recognition events.22,40 In oligonucleotide ligation reaction-based genotyping, DNA ligase catalyzes the formation of a phosphodiester bond between two oligonucleotides hybridized to adjacent positions of the interrogated sequence only if there is perfect complementarity with the target sequence at the junction. We present herein the combination of this high selectivity of the oligonucleotide ligation reaction with the fluorescence quenching of AuNPs to explore a more efficient and applicable protocol for SNPs detection. To demonstrate the feasibility of this protocol, two different DNA ligases, T4 DNA ligase22 and Escherichia coli DNA ligase,41 were investigated. With the help of DNA ligase, the new approach shows good SNPs discriminability and high sensitivity without needing expensive equipment. In addition, the noncovalent interaction between AuNPs and oligonucleotides makes the assay easy to handle, avoiding laborious probe design and modifications. EXPERIMENTAL SECTION Materials. HAuCl4 · 4H2O (AR) was supplied by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). T4 DNA ligase and E. coli DNA ligase with 10× ligation reaction buffer were obtained from Takara Biotechnology Co., Ltd. (Dalian, China). The 10× E. coli DNA ligation buffer for ligation reaction contains 300 mM Tris-HCl (pH 8.0), 40 mM MgCl2, 100 mM (NH4)2SO4, 12 mM EDTA, and 1 mM NAD. For T4 DNA ligase, the 10× ligation buffer includes 660 mM Tris-HCl (pH 7.5), 66 mM MgCl2, 100 mM DTT, and 1 mM ATP. All oligonucleotides (Table 1) were synthesized by Shanghai Sangon Biological Engineering Technology and Services Co., Ltd. (Shanghai, China) and used as received by dissolving in deionized water (18.3 MΩ cm) produced from a Millpore water purification system. All other chemicals (AR) were commercially available and used without further purification. (34) Ghosh, S. K.; Pal, T. Chem. Rev. 2007, 107, 4797–4862. (35) Fan, C.; Wang, S.; Hong, J. W.; Bazan, G. C.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 6297–6301. (36) Li, H.; Rothberg, L. J. Anal. Chem. 2004, 76, 5414–5417. (37) Wang, W.; Chen, C.; Qian, M.; Zhao, X. S. Anal. Biochem. 2008, 373, 213– 219. (38) Wang, H.; Wang, Y. X.; Jin, J. Y.; Yang, R. H. Anal. Chem. 2008, 80, 9021– 9028. (39) Shi, M. M. Clin. Chem. 2001, 47, 164–172. (40) Toubanaki, D. K.; Christopoulos, T. K.; Ioannou, P. C.; Flordellis, C. S. Anal. Chem. 2009, 81, 218–224. (41) Zhang, S. B.; Wu, Z. S.; Shen, G. L.; Yu, R. Q. Biosens. Bioelectron. 2009, 24, 3201–3207.
Analytical Chemistry, Vol. 82, No. 18, September 15, 2010
7685
Table 1. Oligonucleotides Sequences Used in This Worka name
sequences
P1 P2 P3 P4 T1 (M-DNA) T2 (W-DNA) T3
5′-FAM-AGTTGGAGCTGA -3′ 5′-p-TGGCGTAGGCAA-3′ 5′-Py-AGTTGGAGCTGA-3′ 5′-TAMRA-AGTTGGAGCTGA-3′ 5′- TTGCCTACGCCATCAGCTCCAACT -3′ 5′- TTGCCTACGCCACCAGCTCCAACT-3′ 5′-TCAGCTCCAACT-3′
Scheme 1. Schematic Representation for SNPs Detection Based on DNA Ligase Reaction and AuNPs-Quenched Fluorescent Oligonucleotides
a The bold base in T2 indicates the mismatched base. FAM in P1 is carboxyl fluorescein. Py in P3 is pyrene, and TAMRA in P4 is tetramethylrhodamine. The p in P2 represents phosphate at 5′ end.
P1 is a 12-mer carboxyl fluorescein(FAM)-labeled DNA probe at its 5′-terminus, and P2 is also a 12-mer DNA probe phosphorylated at its 5′-terminus. T1 is a part sequence of the K-ras oncogene and perfectly complementary to P1 and P2, in which one point mutation occurred from the wild type C to T. The wild type DNA sequence T2 is the contrast target of T1 for SNP detection. The bold bases denote the SNP position. To test the fluorescence quenching efficiency of AuNPs for different dyes, P3 and P4 was designed, which have the same sequences of P1 but were labeled with pyrene (Py) and tetramethylrhodamine (TAMRA), respectively, to the 5′-terminus. The 13 nm-AuNPs were produced from gold precursors by citrate reduction.42 The synthesized nanoparticles were characterized and quantified by absorptivity of a UV-vis absorption spectrum at 520 nm (ε ) 2.7 × 108 M-1 cm-1),43 which was recorded on a Hitachi U-4100 UV-vis spectrophotometer (Kyoto, Japan). General Methods. In a typical experiment, ligation reaction was first conducted. For T4 DNA ligase based reactions, 4 µL of P1 (2 µM), 4 µL of P2 (2 µM), and 4 µL of T1 or T2 were mixed with 1 µL of ligase (1 U/µL) and 1 µL of 10× ligation reaction buffer. For E. coli DNA ligase, 1 µL of BSA (0.05%) was added. After reaction for 0.5 h at room temperature (25 °C), 4.5 µL of the reaction solution was diluted with water. A volume of 80 µL of prepared AuNPs solution was added, and the final volume of the solution was 280 µL. After incubation for 0.5 h at 30 °C, the above solution was centrifuged at 12 000g for 3 min to remove the deposited nanoparticles, and the fluorescence of the supernatant solutions was detected by a Hitachi F-2500 fluorescence spectrophotometer (Kyoto, Japan) with a scan rate at 300 nm/ min. The excitation wavelength was at 480 nm, and the photomultiplier tube (PMT) voltage was 400 V. The slits for excitation and emission were set at 10 nm/10 nm. RESULTS AND DISCUSSION Principle of Operation. With dependence on the structure of nucleic acids, DNA has a different propensity to adsorb onto the surface of AuNPs.32,44 This noncovalent interaction originates from the coordination of nucleobases to gold atoms of nanoparticle (42) Grabar, K. C.; Freeman, R. G.; Hommer, M. B.; Natan, M. J. Anal. Chem. 1995, 67, 735–743. (43) Zhao, W.; Chiuman, W.; Brook, M. A.; Li, Y. ChemBioChem 2007, 8, 727– 731. (44) Gearheart, L. A.; Ploehn, H. J.; Murphy, C. J. J. Phys. Chem. B 2001, 105, 12609–12615.
7686
Analytical Chemistry, Vol. 82, No. 18, September 15, 2010
and the electrostatic force.45 The random wirelike structures facilitate ssDNA adsorbing onto the surface of nanoparticles more easily which result in the coordination of free bases to AuNPs. However, because of the hybridized bases, dsDNA and other nucleic acids with complex structures are more rigid to cause weak adsorption onto the cambered surface of AuNPs.46 This is the fundamentality of the present approach for SNPs detection. To improve the performance and sensitivity, the ligation reaction was employed. Scheme 1 shows the operation principle for SNP detection based on the formation of a DNA duplex with P1 and P2 and the target oligonucleotide strand T1 or T2. As shown in Scheme 1A, using the one-base mismatched target DNA (T2) as the ligation template, the two short probes of P1 and P2 cannot be linked by DNA ligase to form stable duplexes under a certain temperature, for example, at 30 °C. After addition of AuNPs, the fluorescence of P1 is totally quenched as the separated ssDNA probe is adsorbed to the AuNPs. When the ligation reaction catalyzed by DNA ligase is carried out in the presence of perfectly matched DNA target (T1), the stable dsDNA is formed at 30 °C. This ssDNA cannot be adsorbed onto the surface of nanoparticles. After removal of the deposited nanoparticles by centrifugation (at 12 000g for 3 min), a relatively stronger fluorescence signal from P1 was realized and compared to that in the presence of T2 (Scheme 1B). We chosen 30 °C as the operation temperature because at 30 °C the duplex between nonlinked short probes and mismatched target T2 could not be formed stably, while the formed duplex between the linked probes and perfectly matched target T1 was still stable (see below). Quenching Efficiency of AuNPs for Dye-Tagged ssDNA. AuNPs were confirmed to be superquenchers of organic dyes with the long-range nanoscale energy transfer property, which in combination with the unique DNA/nanoparticles interaction, forms the basis of a convenient and versatile strategy for fluorescent DNA analysis.47-49 In the present design, because the SNPs detection sensitivity is dependent on the fluorescence quenching efficiency of AuNPs, it is important to first estimate the quenching capability of AuNPs for organic dyes. The fluores(45) Zhao, W.; Lee, T. M. H.; Leung, S. S. Y.; Hsing, I. M. Langmuir 2007, 23, 7143–7147. (46) Li, H. X.; Rothberg, L. J. J. Am. Chem. Soc. 2004, 126, 10958–10961. (47) Maxwell, D. J.; Taylor, J. R.; Nie, S. J. Am. Chem. Soc. 2002, 124, 9606– 9612. (48) Song, S.; Liang, Z.; Zhang, J.; Wang, L.; Li, G.; Fan, C. Angew. Chem., Int. Ed. 2009, 48, 8670–8674. (49) Kim, J. H.; Estabrook, R. A.; Braun, G.; Lee, B. R.; Reich, N. O. Chem. Commun. 2007, 4342–4344.
Figure 2. Fluorescence quenching efficiency of AuNPs for dyetagged ssDNA probes (100 nM) in the Tris-HCl buffer solution. F0 and F denoted the fluorescence emission intensity of the probe in the absence and the presence of AuNPs, respectively. For recording the fluorescence emission intensity, the excitation wavelength of each probe was tuned to the respective maximum of the excitation of the dye.
Figure 1. Effect of AuNPs on the fluorescence emission of FAM, pyrene, and TAMRA and their ss-DNA complexes P1, P3, and P4 in 20 mM Tris-HCl buffer (pH 7.5): (A-C) fluorescence emission spectra of FAM, pyrene, and TAMRA in the absence (continuous lines) and the presence (dotted lines) of AuNPs (7.0 nM) and (D-F) fluorescence emission spectra of P1, P3, and P4 in the absence (continuous lines) and the presence (dotted lines) of AuNPs. The concentrations of all probes were 100 nM, and the excitation wavelength of each probe was tuned to the respective maximal excitation of the dye.
cence quenching efficiency of AuNPs was evaluated via fluorescent measurements of dye-tagged ssDNA in the presence of AuNPs. For each dye tested, the spectrofluorometer’s excitation and emission wavelength was tuned to the respective maximum of the excitation and emission of the dye. We found that the fluorescence intensities of fluorophores, such as FAM, pyrene, and TAMRA, were all depressed in the presence of AuNPs (Figure 1, left). The observed quenching phenomenon largely originated from the free fluctuations of fluorescent molecules and intercollisions between the fluorophore and nanoparticles, which leads to surface-energy-transfer. Interestingly, while both AuNPs and DNA are negatively charged, the fluorescence of the dye-tagged ssDNA probes, P1, P3, or P4 (Table 1) was also rapidly quenched by AuNPs as well (Figure 1, right), obviously indicating the strong interaction between oligonucleotide and the nanoparticles. Figure 2 shows the fluorescence quenching efficiency (F0/F) of the three dye-tagged ssDNA probes as functions of the AuNPs concentration. It is worth noting that the F0/F value of P1 or P3 was obvious higher than that of P4 in the AuNPs concentration range. We suggest that higher quenching efficiency of P3 over P4 results from the stronger interaction between pyrene and
Figure 3. Fluorescence emission spectra of P1 (100 nM, λex ) 480 nm) at different experimental conditions: (a) P1 in Tris-HCl buffer, (b) P1 + 5.0 nM AuNPs, and (c) P1 + T3 (100 nM) + 5.0 nM AuNPs.
AuNPs than that of TAMRA and AuNPs due to the planar structure of pyrene, while that of P1 over P3 or P4 could be a result of absorption of the FAM emission by the nanoparticles. In the system of P1/AuNPs, AuNPs show strong absorption at 520 nm, while the maximum emission wavelength of FAM is located at 518 nm. This main 518 nm emission band of the FAM overlaps nicely with the absorption band of AuNPs. Thus, the effective emission intensity of FAM would be more decreased if the two materials coexist in a sensory device.38 Because of the highest quenching efficiency of the FAM-tagged ssDNA, P1 was used as a fluorescent probe to perform SNPs detection in the present approach. To examine the effect of AuNPs on the fluorescence of dyetagged dsDNA, P1 was first hybridized with its perfect cDNA (T3). Figure 3 shows the fluorescence emission spectra of P1 and the duplex of P1 and T3 in the presence of 5.0 nM AuNPs, and significant fluorescence emission enhancement of the duplex is observed. The fluorescence intensity of P1 at 518 nm in the presence of 1 equiv of T3 is 8.4-fold higher than that without T3. This fluorescence enhancement is the result of the formation of Analytical Chemistry, Vol. 82, No. 18, September 15, 2010
7687
Figure 4. Effect of AuNPs concentrations on the discrimination ability of P1 and P2 for T1 and T2. The concentrations of the DNA sequences were 9.0 nM. The experiments were carried out as described in the Experimental Section. Inset: the change of R as a function of the AuNPs concentrations, where R is defined as the ratio of the fluorescence intensity of perfectly matched target T1 to that of one base-mismatched target T2.
dsDNA, which hampers energy transfer between FAM and AuNPs. This difference of AuNPs quenching efficiency for dyetagged ssDNA and dye-tagged dsDNA constitutes the basis for fluorescent detection of SNPs proposed in this paper. Optimization of the Variables of the Measuring Systems. The concentration of AuNPs would influence the result of SNP detection. In Figure 4, the quenching effect of different concentrations of AuNPs toward the ligation products for T1 and T2 were exhibited, respectively. It can be concluded from Figure 4 that, at lower concentration of AuNPs, the free DNA probes could not be adsorbed completely within the desired time, resulting in a nonsignificant signal difference for perfectly matched or mismatched targets. When excessive nanoparticles were present which increased the probability of dsDNA to collide with the nanoparticles, the fluorescent signal was also reduced accompanied by the increase of detection error. To get better discrimination of the SNPs with appropriate fluorescence intensity, 5.0 nM AuNPs were optimized for interaction with the oligonucleotide in the reaction solutions. The ratio of nanoparticles to labeled DNA probe was about 1:3, much lower than that of using mercaptan compounds for covalently modification.50 The concentration of salt concerns both the hybridization of nucleic acids and the interactions of nanoparticles with oligonucleotides. Usually, nucleic acids are negatively charged in neutral buffer solutions. With an increase in the concentration of salt, the electrostatic repulsion force will be screened. That is to say, with the increase of the salt concentration, the hybridization force will increase and the duplex formed between short probes and the target DNA will be more stable, which causes the fluorescence intensity to be increased as less adsorption of nucleic acids to the nanoparticle surface. At the same time, the synthesized citrate-capped nanoparticles can be separated in solutions because of the repulsion force coming from the anionic groups. However, the electrostatic balance would be disturbed in high salt solution and the dispersed nanoparticles would aggregate (50) Demers, L. M.; Mirkin, C. A.; Mucic, R. C.; Reynolds, R. A.; Letsinger, R. L.; Elghanian, R.; Viswanadham, G. Anal. Chem. 2000, 72, 5535–5541.
7688
Analytical Chemistry, Vol. 82, No. 18, September 15, 2010
Figure 5. Effect of salt concentration on the SNP assay. Before addition of AuNPs, the reaction solution contains different concentrations of NaCl. The final concentration of AuNPs is 5.0 nM, and the other conditions are same as that in Figure 4.
rapidly.32 So, the insufficient touching with nanoparticles would also bring down the quenching effect, causing false positive results. As shown in Figure 5, when higher concentrations of salts were introduced into the reaction solution, the fluorescence signal ratio (R) for T1 and T2 were reduced to nearly 1:1. It indicates that, at a higher concentration of salt, the formed duplexes between the short probes and mismatched target T2 were still stable under the same conditions. So, to get good performance for one base mutation detection, no additional salt was needed when the reaction solution was diluted with water. Temperature is also one of the most important parameters in nucleic acid hybridization. In the design, the two short 12-mer DNA probes will be linked with the help of DNA ligase in the presence of the right DNA template (T1), which greatly increases the melting temperature (Tm) of the duplex formed with target DNA. On the contrary, for mismatched target T2, the two short probes could not be linked by the DNA ligase, so the melting temperature of the duplex formed between the probes and T2 is still low. That is to say, to get satisfactory SNP discrimination, the assay temperature should be fixed at a temperature higher than the Tm for mismatched target (T2) but lower than that of the newly formed 24-base pairs dsDNA for perfectly matched target T1, which is the primary foundation of the AuNP quenching-based fluorescence assay for SNP. In Figure 6, the fluorescence signal ratios (R) at different operating temperatures were shown. One can conclude from Figure 6 that the discrimination capability of this approach for SNP was improved with the increasing of the operating temperature until 30 °C, while rapidly reduced when the temperature was higher 30 °C, for example 35 °C. The reason is that, at a temperature of 30 °C, the duplex between nonlinked short probes and mismatched target T2 could not be formed stably, so that the separated probes were adsorbed onto the AuNPs surface. At the same temperature, the formed duplex between the linked probes and perfectly matched target was still stable resulting in a low adsorption of the FAM-labeled probe to AuNPs. When the temperature was higher than 30 °C, the duplexes formed by P1, P2, and T1 would be partly transformed into ssDNA, resulting in lower sensitivity. With a further increase in operating temperature, for example >35 °C, since the duplexes for cDNA or mismatched DNA could not be formed,51,52
Figure 6. Effect of incubation temperature on the performance of the SNP assay. During the experiment, AuNPs was added into the reaction solution under different temperatures (the final concentration of AuNPs was 5.0 nM), and then the fluorescence was tested. Other conditions are the same as that in Figure 4.
adsorption of all separated ssDNA to AuNPs results in no significant difference in the P1 fluorescence. In our experiment, the SNP detection assay was performed at 30 °C. Performance of the SNP Detection. We implemented this AuNPs fluorescence quenching approach to detect and quantify
DNA targets. It can be seen from Figure 7A that P1 is highly fluorescent in the mixture of P1, P2, and T4 DNA ligase (curve a), suggesting the ligase hardly influences the P1 fluorescence. When AuNPs were introduced into the mixture, the P1 fluorescence was almost quenched (curve b) because the dye-labeled ssDNA absorbed on the nanoparticles surface. On the contrary, if the perfectly matched T1 was first introduced into the solution of P1, P2, and T4 DNA ligase, obvious P1 fluorescence emission was observed upon addition of the same amount of AuNPs (curve c), clearly indicating that in the presence of the right template DNA (T1), the DNA ligase linked the two half primers and formed a stable duplex, so that reduced the interaction of P1 with AuNPs. The addition of single-base mismatched target T2 reduced the fluorescence intensity to 22% in comparison with the fluorescence intensity caused by T1 (curve d). The SNP discrimination ability is higher than those of molecular beacons, for which single-base mismatches result in a 50-90% the signal caused by the perfectly matched DNA target.53,54 This is the advantage of the proposed approach over the traditional DNA hybridization method. Next, different amounts of perfectly matched target T1 or onebase mismatched target T2 were analyzed by this proposed method. It was observed from Figure 7B that the fluorescence intensity of P1 at 520 nm was increased with the increasing of T1 concentration. This is because the ligated products increased as
Figure 7. The performances of the SNP assay using FAM-tagged ssDNA, AuNPs, and DNA ligases: (A) fluorescence emission spectra (λex ) 480 nm) of the reaction mixtures of P1 (9.0 nM) and P2 (9.0 nM) at different experimental conditions. (a) P1, P2 and T4 DNA ligase in Tris-HCl buffer; (b) P1, P2 and T4 DNA ligase in Tris-HCl buffer + 5.0 nM AuNPs; (c) P1, P2 and T4 DNA ligase in Tris-HCl buffer + 9.2 nM T2 + 5.0 nM AuNPs; (d) P1, P2 and T4 DNA ligase in Tris-HCl buffer + 9.2 nM T1 + 5.0 nM AuNPs. (B) Fluorescence signal enhancement (F/F0) of P1 as functions of the concentration of T1 (b) or T2 ([) in the presence of T4 DNA ligase. (C) Fluorescence signal enhancement (F/F0) of P1 as functions of the concentrations of T1 (b) or T2 ([) in the presence of E. coli DNA ligase. Where F0 and F are the P1 fluorescence intensity (λex/λem ) 480 nm/520 nm) in the mixture of P1, P2, AuNPs, and DNA ligase in the absence and the presence of DNA target, respectively. The operating temperature is 30 °C, and the measurements were carried out as described in the Experimental Section. Analytical Chemistry, Vol. 82, No. 18, September 15, 2010
7689
more perfectly matched template DNA was present in the reactions, which hybridized and formed dsDNA, accompanied with a low adsorption of dye-labeled DNA to AuNPs. There is a linear relationship between the fluorescence intensity and the concentrations of T1 in the range of 1.0-10.0 nM with the detection limit of 0.3 nM. However, for the one-base mismatched target T2, a relatively lower fluorescence signal was also observed after the addition of AuNPs, indicating that a few ligated products were still generated even for the mismatched target as the mistaken action of ATP-depended DNA ligase. Although the resulted unwanted products would hybridize and form dsDNA with T2, giving a fluorescence signal, the maximum signal ratio for T1 to T2 was reached about 4.6 when a 9.0 nM target concentration was used. In ligation-based SNP detection approaches, the capability to discriminate one-base mismatch from cDNA originated from the fidelity of DNA ligase. To illustrate the generality of this method, we used a NAD-depended ligase, E. coli DNA ligase, to detect T1 and T2. It was reported that the NAD-depended ligases were more sensitive to mismatched bases and performed better than the ATPdepended T4 DNA ligase.55,56 Figure 7C shows the fluorescence signal enhancement, F/F0, of P1 as a function of the concentration of T1 or T2 in the presence of E. coli DNA ligase, where F0 and F are the fluorescence intensities of P1 in the mixture of P1, P2, AuNPs, and DNA ligase in the absence and the presence of DNA target. For the perfectly matched target T1, the fluorescence signal dynamic increase with the addition of T1 in the concentration range of 1.1-9.0 nM, and when the concentration of T1 reached 9.0 nM, the maximum of fluorescence intensity appeared, with a further increase in the T1 concentration as a result of a slight decrease in the fluorescence. For one-base mismatched target T2, the fluorescence signal was almost quenched when the concentration of T2 was less than 9.0 nM. In Figure 7C, the value of F0/F was estimated to be 60.7 in the presence of 9.0 nM T1, while the value dropped
down to be 2.1 upon the addition of 9.0 nM T2. The variance by a factor of 60.7/2.1 ) 30.3 was achieved, which is higher than that by using T4 DNA ligase, indicating that E. coli DNA ligase has the higher ligation fidelity than T4 DNA ligase.
(51) Bonnet, G.; Tyagi, S.; Libchaber, A.; Kramer, F. R. J. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6171–6176. (52) Darby, R. A. J.; Sollogoub, M.; Mckeen, C.; Brown, L.; Ristano, A.; Brown, N.; Barton, C.; Brown, T.; Fox, K. R. Nucleic Acids Res. 2002, 30, e39. (53) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303–308. (54) Yang, C. J.; Medley, C. D.; Tan, W. H. Curr. Pharm. Biotechnol. 2005, 6, 445–452. (55) Tomkinson, A. E.; Vijayakumar, S.; Pascal, J. M.; Ellenberger, T. Chem. Rev. 2006, 106, 687–699. (56) Liu, L.; Tang, Z.; Wang, K.; Tan, W.; Li, J.; Guo, Q.; Meng, X.; Ma, C. Analyst 2005, 130, 350–357. (57) Li, J.; Yan, H.; Wang, K.; Tan, W.; Zhou, X. Anal. Chem. 2007, 79, 1050– 1056.
ACKNOWLEDGMENT Financial support from the National Natural Science Foundation of China (Grant 20775005) and the National Grand Program on Key Infectious Disease of China (Grant 2009ZX10004-312) is highly acknowledged. J. Li also thanks Hunan University for the start-up funds (Grant 521105668).
7690
Analytical Chemistry, Vol. 82, No. 18, September 15, 2010
CONCLUSIONS In conclusion, a new approach for SNPs detection is proposed employing the DNA ligase reaction and AuNPsquenched fluorescent oligonucleotides. The design is based on the difference in the formation of dsDNA by a perfectly matched DNA template and single-base mismatched DNA template in the presence of DNA ligase and that in electrostatic affinity between ssDNA and dsDNA with AuNPs. Because of the high discriminability of DNA ligase and the high sensitivity of fluorometry, the proposed approach exhibited a good performance to efficiently differentiate the perfectly matched DNA target from the single-base mismatched DNA target. Although a number of approaches based on the ligase reaction and nanoparticles have been developed for the SNPs assay, this approach possesses three molecular engineering advantages. First, it requires no covalent modification of the AuNPs but shows high quenching efficiency by the nanoparticles, thus, the application demonstrates superior sensitivity. Second, the experimental procedure is more convenient than similar approaches, and it does not need a complicated instrument and higher operating temperature. Finally, and most important, this design strategy is flexible, which can be applied to other nucleic acids enzyme related diagnostics, such as the detection of DNA methylation by using different DNA enzymes.57 These features establish the universality and simplicity of the approach and could, therefore, provide the groundwork for the design of other nanodevices for biosensing applications. Currently, intensive research using new oligonucleotides for probing other biomolecular interactions is being conducted in our laboratory, and the results will be reported in due course.
Received for review June 7, 2010. Accepted August 10, 2010. AC101503T
