Modulating Molecular Level Space Proximity - American Chemical

Apr 13, 2010 - Modulating Molecular Level Space Proximity: A. Simple and Efficient Strategy to Design Structured. DNA Probes. Jing Zheng,† Jishan Li...
1 downloads 0 Views 998KB Size
Download PDF
Anal. Chem. 2010, 82, 3914–3921

Modulating Molecular Level Space Proximity: A Simple and Efficient Strategy to Design Structured DNA Probes Jing Zheng,† Jishan Li,† Xiaoxia Gao,† Jianyu Jin,†,‡ Kemin Wang,† Weihong Tan,† and Ronghua Yang*,† State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha, 410082, China, and Beijing National Laboratory for Molecular Sciences, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China To construct efficient oligonucleotide probes, specific nucleic acid is designed as a conformationally constrained form based on the formation of a Watson-Crick-based duplex. However, instability of Watson-Crick hydrogen bonds in a complex biological environment usually leads to high background signal from the probe itself and false positive signal caused by nonspecific binding. To solve this problem, we propose a way to restrict the labeleddyes in a hydrophobic cavity of cyclodextrin. This bounding,whichactslikeextrabasepairstoformtheWatson-Crick duplex, achieves variation of level of space proximity of the two labels and thus the degree of conformational constraint. To demonstrate the feasibility of the design, a stem-containing oligonucleotide probe (P1) for DNA hybridization assay and a stemless one (P2) for protein detection were examined as models. Both oligonucleotides were doubly labeled with pyrene at the 5′- and 3′ends, respectively. It is the cyclodextrin/pyrene inclusion interaction that allows modulating the degree of conformational constraints of P1 and P2 and thus their background signals and selectivity. Under the optimal conditions, the ratio of signal-to-background of P1/γ-CD induced by 1.0 equiv target DNA is near 174, which is 4-fold higher than that in the absence of γ-CD. In addition, the usage of γ-CD shifts the melting temperature of P1 from 57 to 68 °C, which is reasonable for improving targetbinding selectivity. This approach is simple in design, avoiding any variation of the stem’s length and sequences. Furthermore, the strategy is generalizable which is suited for not only the stem-containing probe but also the linear probe with comparable sensitivity and selectivity to conventional structured DNA probes. Since the discovery of the double helix structure of DNA,1 the Watson-Crick type of hydrogen bonds, combined with electrostatic force, π-stacking, and hydrophobic forces, has made it * To whom correspondence should be addressed. E-mail: [email protected]. Fax: +86-731-8882 2523. † Hunan University. ‡ Peking University. (1) Watson, J. D.; Crick, F. H. C. A. Nature 1953, 171, 737–738.

3914

Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

possible to design various functional oilgonucleotide probes.2-7 To achieve desirable sensitivity and selectivity, an oligonucleotide was initially designed as a conformationally constrained structure through the formation of Watson-Crick base pairing. This explains the attention given to such well-known Taqmans,8 molecular beacons (MBs),9,10 aptamers,11 and MB-aptamers.12-14 However, even though the presence of a conformationally constrained structure is the advantage for recognizing their targets with higher specificity than linear probes,15-17 it is far from functionally ideal. For instance, high levels of stem sequencedependent background signal and false positive signals that result from endogenous nuclease degradation and nonspecific binding are commonly observed for these probes due to instability of the stem in a complex biological environment.18-20 Focusing on these drawbacks of the Watson-Crick base paired-stem, researchers have advanced a few new strategies to modulate the stem stability. One design type, for example, is (2) Liu, J.; Cao, Z. H.; Lu, Y. Chem. Rev. 2009, 109, 1948–1998. (3) Drummond, T. G.; Hill, M. G.; Barton, J. K. Nat. Biotechnol. 2003, 21, 1192–1199. (4) Bratu, D. P. Methods Mol. Biol. 2006, 319, 1–14. (5) Wang, H.; Yang, R. H.; Yang, L.; Tan, W. H. ACS Nano 2009, 3, 2451– 2460. (6) Matayoshi, E. D.; Wang, G. T.; Krafft, G. A.; Erickson, J. Science 1990, 247, 954–958. (7) Ali, M. M.; Li, Y. F. Angew. Chem., Int. Ed. 2009, 48, 3512–3515. (8) Holland, P. M.; Abramson, R. D.; Watson, R.; Gelfand, D. H. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 7276–7280. (9) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303–308. (10) Yang, C. J.; Medley, C. D.; Tan, W. H. Curr. Pharm. Biotechnol. 2005, 6, 445–452. (11) Jhaveri, S. D.; Kirby, R.; Conrad, R.; Maglott, E. J.; Bowser, M.; Kennedy, R. T.; Glick, G.; Ellington, A. D. J. Am. Chem. Soc. 2000, 122, 2469–2473. (12) Tang, Z. W.; Mallikaratchy, P.; Yang, R. H.; Kim, Y. M.; Zhu, Z.; Wang, H.; Tan, W. H. J. Am. Chem. Soc. 2008, 130, 11268–11269. (13) Nutiu, R.; Mei, S.; Liu, Z. J.; Li, Y. F. Pure Appl. Chem. 2004, 76, 1547– 1561. (14) Nutiu, R.; Li, Y. F. Methods 2005, 37, 16–25. (15) Bonnet, G.; Tyagi, S.; Libchaber, A.; Kramer, F. R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6171–6176. (16) Tsourkas, A.; Behlke, M. A.; Rose, S. D.; Bao, G. Nucleic Acids Res. 2003, 31, 1319–1330. (17) Bonnet, G.; Krichevsky, O.; Libchaber, A. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 8602–8606. (18) Fisher, T. L.; Terhorst, T.; Cao, X.; Wagner, R. W. Nucleic Acids Res. 1993, 21, 3857–3865. (19) Leonetti, J. P.; Mechti, N.; Degols, G.; Gagnor, C.; Lebleu, B. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 2702–2706. (20) Uchiyama, H.; Hirano, K.; Kashiwasake-Jibu, M.; Taira, K. J. Biol. Chem. 1996, 271, 380–384. 10.1021/ac1004713  2010 American Chemical Society Published on Web 04/13/2010

replacements of the original DNA Watson-Crick base pairs with homo-DNA,21,22 locked nucleic acid (LNA),23-25 peptide nucleic acid (PNA),26-28 as well as other nuclease-resistant nucleic acids,29,30 which make these probes especially useful under stringent experimental conditions. Because of the molecule design, however, all these probes required some variation of the stem length or sequence. Such modulation adds steps, including complex resynthesis of several different probes or the use of nonnatural DNA bases, which limit their applications as common biosensors. Another strategy is the use of a metal-dependent stem based on either the G-quadruplex system or metal-DNA base pairs.31-33 These alternatives do provide opportunities to optimize target recognition from both thermodynamic and kinetic points of view. At the same time, however, these DNA analogues are challenged by the toxicity of the heavy metal ion and the difficulty of addressing metal ion concentration in applications for in vivo detection. Besides these strategies, a novel kind of triplex hairpinshaped probe has been recently designed by Seitz et al. through introducing a DNA competitor in the stem portion.34 The DNA competitor hybridizes with the stem sequence to form a triplexstem but allows dehybridization when the probe binds to its target molecule. Although the approach requires several stem-forming oligonucleotides and careful optimization to avoid conformational changes, it offers a general means for more precisely modulating stem stability. In assessing the merits and limitations of these approaches, we sought to develop an alternative method that could conveniently modulate the degree of conformational constraint but did not require any variation of the stem sequences or length. To accomplish this, we considered (1) whether the degree of the stem stability of a DNA probe could be governed by not only the stem sequence and length but also the molecular level distance of the two terminal-labels and, if so, (2) how the variation of space proximity of the two labeled dyes modulates the degree of conformational constraint and thus the thermodynamic and kinetic features of the probe. The present work was performed with the aim of investigating these two questions. We reasoned that the stem formation of a DNA probe could be the combinative results of the Watson-Crick

base pairs and the distance of the labeled-dyes35-37 and that, as such, bounding the labeled dyes in a hydrophobic cavity might increase the degree of conformational constraint by decreasing the molecular level space of the dyes. Here, we describe an examination of this hypothesis by employing cyclodextrins (CDs) inclusion interaction. To demonstrate the feasibility of this design, a bis-pyrene-labeled stemcontaining probe (P1) and a stemless probe (P2), human R-thrombin (Tmb) binding aptamer, were selected as models. Pyrene is a simple hydrocarbon aromatic molecule, which has the advantage of versatility for various chemical modifications.38-40 Bis-pyrenelabeled oligonucleotides have been widely used to probe DNA duplex formation and RNA folding by monitoring the monomer and excimer emission fluctuations.41-46 In MBs and MB-aptamers, pyrenes have been labeled at opposite terminals to facilitate an excimer-monomer switching mechanism that is controlled by the bound and unbound conformations of the oligonucleotide.47-51 At the same time, extensive studies have documented that the truncated, corn-shaped hydrophobic cavity of CDs can selectively bind diverse organic molecules to form a versatile supramolecular assembly.52-55 The inclusion interactions of γ-cyclodextrin (γ-CD) and pyrene have been extensively investigated, proving the 1:2 and/or 2:2 host-guest complexes.56-58 In the present article, we aim to tune the molecular level space proximity of the labeled pyrenes by bounding the two pyrene molecules in the γ-CD cavity

(21) Eschenmoser, A.; Dobler, M. Helv. Chim. Acta 1992, 75, 218–259. (22) Grey-Desbiolles, D.; Ahn, D-.R.; Leumann, C. J. Nucleic Acids Res. 2005, 33, e77. (23) Koshkin, A. A.; Nielsen, P.; Meldgaard, M.; Rajwanshi, V. K.; Singh, S. K.; Wengel, J. J. Am. Chem. Soc. 1998, 120, 13252–13253. (24) Wang, L.; Yang, C. Y.; Medley, C. D.; Benner, S. A.; Tan, W. H. J. Am. Chem. Soc. 2005, 127, 15664–15665. (25) Hrdlicka, P. J.; Babu, B. R.; Sφ´ rensen, M. D.; Harrit, N.; Wengle, J. J. Am. Chem. Soc. 2005, 127, 13293–13299. (26) Seitz, O. Angew. Chem., Int. Ed. 2000, 39, 3249–3252. (27) Kuhn, H.; Demidov, V. V.; Gildea, B. D.; Fiandaca, J. M.; Coull, J. C.; FrankKamenetskii, M. D. Antisense Nucleic Acid Drug Dev. 2001, 11, 265–270. (28) Kuhn, H.; Demidov, V. V.; Coull, J. C.; Fiandaca, J. M.; Gildea, B. D.; FrankKamenetskii, M. D. J. Am. Chem. Soc. 2002, 124, 1097–1103. (29) Braasch, D. A.; Corey, D. R. Chem. Biol. 2001, 8, 1–7. (30) Tsourkas, A.; Behlke, M.; Bao, G. Nucleic Acids Res. 2002, 30, 5168–5174. (31) Bourdoncle, A.; Este´vez Torres, A.; Gosse, C.; Lacroix, L.; Vekhoff, P.; Le Saux, T.; Jullien, L.; Mergny, J.-L. J. Am. Chem. Soc. 2006, 128, 11094– 11105. (32) Lin, Y.-W.; Ho, H.-T.; Huang, C.-C.; Chang, H.-T. Nucleic Acids Res. 2008, 36, e123. (33) Yang, R. H.; Jin, J. Y.; Long, L. P.; Wang, Y. X.; Wang, H.; Tan, W. H. Chem. Commun. 2009, 322–324. (34) Grossmann, T. N.; Ro ¨glin, L.; Seitz, O. Angew. Chem., Int. Ed. 2007, 46, 5223–5225.

(46)

(35) (36) (37) (38) (39) (40) (41) (42) (43) (44) (45)

(47) (48) (49) (50)

(51) (52) (53) (54) (55) (56) (57) (58)

Wilson, J. N.; Kool, E. K. Org. Biomol. Chem. 2006, 4, 4265–4274. Chaudhuri, N. C.; Kool, E. K. J. Am. Chem. Soc. 1995, 117, 10434–10442. Matray, T. J.; Kool, E. K. J. Am. Chem. Soc. 1998, 120, 6191–6192. Birks, J. B. Photophysics of Aromatic Molecules. In Wiley Monographs in Chemical Physics; Wiley: New York, 1979; p 704. Nohta, H.; Satozono, H.; Koiso, K.; Yoshida, H.; Ishida, J.; Yamaguchi, M. Anal. Chem. 2000, 72, 4199–4204. Snare, M. J.; Thistlethwaite, P. J.; Ghiggino, K. P. J. Am. Chem. Soc. 1983, 105, 3328–3332. Pairs, P. L.; Langenhan, J. M.; Kool, E. T. Nucleic Acids Res. 1998, 26, 3789–3793. Kostenko, E.; Dobrikov, M.; Pyshnyi, D.; Petyuk, V.; Komarova, N.; Vlassov, V.; Zenkova, M. Nucleic Acids Res. 2001, 29, 3611–3620. Lewis, F. D.; Zhang, Y.; Letsinger, R. L. J. Am. Chem. Soc. 1997, 119, 5451–5452. Yamana, K.; Iwai, T.; Ohtani, Y.; Sato, S.; Nakamura, M.; Nakano, H. Bioconjugate Chem. 2002, 13, 1266–1273. Yamana, K.; Fukunaga, Y.; Ohtani, Y.; Sato, S.; Nakamura, M.; Kim, W. J.; Akaike, T.; Maruyama, A. Chem. Commun. 2005, 2509–2511. Martı´, A. A.; Li, X. X.; Jockusch, S.; Li, Z. M.; Raveendra, B.; Kalachikov, S.; Russo, J. J.; Morozova, I.; Puthanveettil, S. V.; Ju, J. Y.; Turro, N. J. Nucleic Acids Res. 2006, 34, 3161–3168. Oh, K. J.; Cash, K. J.; Plaxco, K. W. J. Am. Chem. Soc. 2006, 128, 14018– 14019. Fujimoto, K.; Shimizu, H.; Inouye, M. J. Org. Chem. 2004, 69, 3271–3275. Yang, C. J.; Jockusch, S.; Vicens, M.; Turro, N. J.; Tan, W. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 17278–17283. Conlon, P.; Yang, C. Y.; Wu, Y. R.; Chen, Y.; Martinez, K.; Kim, Y. M.; Stevens, N.; Marti, A. A.; Jockusch, S.; Turro, N. J.; Tan, W. H. J. Am. Chem. Soc. 2008, 130, 336–342. Chen, Y.; Yang, C. Y.; Wu, Y. R.; Conlon, P.; Kim, Y. M.; Lin, H.; Tan, W. H. ChemBioChem. 2008, 9, 355–359. Saejtli, J. Cyclodextrins and Their Inclusion Complexes; Akademia Kiado: Budapest, Hungary, 1982. Szejtli, J. Chem. Rev. 1998, 98, 1743–1753. Liu, Y.; Yu, L.; Chen, Y.; Zhao, Y. L.; Yang, H. J. Am. Chem. Soc. 2007, 129, 10656–10657. Ihara, T.; Uemura, A.; Futamura, A.; Shimizu, M.; Baba, N.; Nishizawa, S.; Teramae, N.; Jyo, A. J. Am. Chem. Soc. 2009, 131, 1386–1387. Yorozu, T.; Hoshlno, M.; Imamura, M. J. Phys. Chem. 1982, 86, 4426– 4429. Hamai, S. J. Phys. Chem. 1989, 93, 6527–6529. Dyck, A. S. M.; Kisiel, U.; Bohne, C. J. Phys. Chem. B 2003, 107, 11652– 11659.

Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

3915

Scheme 1. Schematic Representation of the Cyclodextrin-Bounded Excimer-Monomer Switching Probes for Detections of Target DNA (A) and Protein (B)

(Scheme 1). It is the γ-CD/pyrene bounding interaction that allows tuning the stem stability of the probes, thereby achieving higher target-binding selectivity and sensitivity than conventionally structured DNA probes. EXPERIMENTAL SECTION Materials and Apparatus. All oligonucleotides were synthesized by Shanghai Sangon Biological Engineering Technology and Services Co., Ltd. (Shanghai China). All sequences were dissolved in highly pure water (sterile Minipore water, 18.3 MΩ) as stock solutions, the solution concentrations were estimated by UV absorption using published sequence-dependent absorption coefficients.59 R-CD, β-CD, and γ-CD stock solutions (100 mM) were prepared by dissolving the desired amount of the materials in doubly distilled water. The P1/CDs inclusion complex was formed by adding a few microliters of a stock solution of CDs to 2.0 mL of 100 nM P1 with a quartz cell and standing, at room temperature, for 10 min; the addition was limited to 100 µL so that the volume change was insignificant. All other reagents were of analytical reagent grade and were purchased from Sigma (Switzerland). All work solutions were prepared with Tris-HCl buffer (pH 7.2, 2.5 mM Mg2+ and 100 mM K+). (59) Cantor, C. R.; Warshaw, M. M.; Shapiro, H. Biopolymers 1970, 9, 1059– 1077.

3916

Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

UV-vis absorption spectra were recorded in 1 cm path length quartz cuvettes on a Hitachi U-4100 UV/vis spectrophotometer (Kyoto, Japan). The steady-state fluorescence emission spectra were obtained on a Hitachi F-7000 fluorescence spectrofluorometer (Kyoto, Japan). Fluorescence emission spectra were collected using a bandwidth of 5 nm and 0.2 × 1 cm2 quartz cuvettes containing 500 µL of solution. Time-resolved emission was measured on a Photon Technology Intl. spectrofluorometer time-correlated single-photon counter at 20.0 ± 0.1 °C. The pH was measured by a model 868 pH meter (Orion). DNA Hybridization Kinetic and Thermal Profiles. To study the kinetics and time-dependence of the interactions of P1 with γ-CD and subsequent the target T1, the excimer fluorescence intensity of P1 at 486 nm was recorded at 20 °C. The excimer fluorescence of 500 µL of P1 (100 nM) was monitored for a few minutes. Then, an excess of γ-CD stock solution was added to the probe buffer and the excimer fluorescence was measured. After confirmation that there was no change of fluorescence with time, target T1 was added and the level of fluorescence was recorded with time. For thermodynamic and temperature-dependent studies, the monomer and excimer fluorescence emission of solutions of P1, in the presence or the absence of γ-CD, was measured as a function of temperature. The temperature was controlled by

Table 1. Designs of Probe and Target Oligonucleotides type

sequence

P1 P2 T1 T2 T3

5′-pyrene-CGTAGGCCTGACTTCTATGCCCACCTACG-pyrene-3′ 5′-pyrene-GGTTGGTGTGGTTGG-pyrene-3′ 5′-TGGGCATAGAAGTCAGG-3′ 5′-TGGGCATACAAGTCAGG-3′ 5′-ACGCGTATGATCACTCC-3′

PolyScience 9112 refrigerating/heating circulators at every 5 °C from 15 to 85 °C. RESULTS AND DISCUSSION Probe Choices and Inclusion Interaction of P1 with CDs. In this work, two probe sequences were selected: the stemstructured one hybridizes with a DNA target and the stemless one binds to Tmb. Pyrene was used as the fluorophore. Most studies have indicated that a 15-25 bases loop together with a 5-7 mer stem will provide an appropriate balance for the formation of a hairpin structure. For P1, we chose a 17 base loop and a 6 complementary base sequence. The oilgonucleotide was labeled with pyrene at the 3′- and 5′-ends, respectively (Table 1). The stem sequence of P1 was designed to participate in hairpin formation but not target hybridization. The target DNA molecules of T1, T2, and T3 were 17 bases long and were either perfectly complementary to the bases in the loop or contained one or more mismatches with the loop. As for P2, the oligonucleotide was 15 bases long. However, it did not have the complementary stem-forming bases on either end and thus could not form a hairpin structure. Figure 1A shows the absorption spectra of P1 in Tris-HCl buffer solution. The maximum absorptions of the free P1 are located at 334, 319, and 306 nm, which are blue-shifted compared to those of pyrene in aqueous solution.56 In the presence of γ-CD in aqueous solution of P1, the CD induces a slight increase in oscillator strength of the absorption spectrum of the MB, while the addition of β-CD substantially alters the shape of the spectrum. Specifically, the spectrum becomes narrow, and the absorption edge shifts by ∼5 nm toward the longer wavelength. No significant alteration of the probe’s absorption spectrum was observed in the presence of R-CD. The alterations of absorption spectra of P1 by β-CD or γ-CD preliminarily indicate that the inclusion interaction occurs between the CDs and P1 in the ground state. Figure 1B shows the fluorescence emission spectra of P1 in the buffer solution in the absence and the presence of CDs at room temperature. In the absence of CDs, P1 exhibits weak monomer fluorescence of pyrene with a peak at 378 and 397 nm, respectively, and a structureless band with an emission maximum center at 486 nm, typicality of pyrene excimer emission.56 The monomer-to-excimer intensity ratio, F378 nm/F486 nm, is about 0.27, where F378 nm and F486 nm are the pyrene fluorescence intensities at 378 and 486 nm, respectively. This fluorescence feature is related to the hairpin-structure of P1. A bis-pyrene-labeled beacon is free in solution without target binding. The pyrene moieties are brought together by the probe’s stem portion, allowing the formation of an excimer. When β-CD was added to the solution of P1, the excimer fluorescence intensity slightly decreases but the monomer fluorescence intensity significantly increases (F378 nm/F486 nm ) 0.53). This result indicates that in

the presence of β-CD, the single pyrene moiety of P1 encapsulates in the β-CD cavity. In marked contrast, upon addition of γ-CD to the solution of P1, the spectrum displays much more significant enhancement of the excimer fluorescence emission but a slight increase of the monomer emission with a ratio of monomer-to-excimer intensity of 0.11. This fluorescence emission change of P1 induced by γ-CD suggests more space proximity of the two pyrene molecules by housing a part of the dimer within the γ-CD cavity (internal diameter ) 8.5 Å52) (Figure S1 in the Supporting Information), as shown in Scheme 1A. Subsequently, fluorescence decay measurements further manifest the interaction between γ-CD and P1. Since very weak monomer fluorescence is observed in aqueous solution, whether in the presence or in the absence of γ-CD, fluorescence decay curves for P1 and its γ-CD inclusion compound were recorded at 486 nm. In the absence of γ-CD, the decay time of the excimer of P1 was 44.7 ns and the excited state was almost monoexponential decay. In the presence of γ-CD, two components of excimer fluorescence are resolved from the decay data. The fluorescence lifetime of 76 ns is evidently prolonged over that of the P1, and this value is also longer than the values obtained in other solvents (Figure S2 in the Supporting Information). Tunable Stem-Stability of P1 by γ-CD. To study the effect of γ-CD on the stem stability of P1, melting temperature (Tm) measurement was first conducted. The correlation between the Tm value and the fluorescence emission of P1 was examined in the temperature range from 15 to 85 °C. As the temperature of a solution containing P1 was slowly raised, the excimer emission (486 nm) decreased but the monomer emission (378 nm) increased. Figure 2 shows the monomer to excimer fluorescence intensity ratio of P1 as a function of temperature. The results show that at lower temperatures, the free P1 is in a closed state and emits weak monomer fluorescence. However, at higher temperatures, the helical order of the stem gives a way to a random-coil configuration, separating the two pyrene molecules and restoring a higher degree of the monomer fluorescence (Figure 2, trace a). The Tm value, defined as the temperature at which the monomer fluorescence of P1 reaches 50% of its maximum value, was estimated to be 57 °C. This experiment was repeated in the presence of γ-CD. The Tm value of P1/γCD was determined to be 68 °C (Figure 2, trace b), which is 11 °C higher than that of P1 under the same conditions (Figure S3 in the Supporting Information). The increase in the Tm value clearly indicates the higher stem stability of P1 induced by γ-CD due to bounding the two pyrenes in the CD cavity. It was reported that the formation of the hairpin-structure of a DNA probe is related to the presence of metal ions (Mg2+ or K+ ions).60,61 However, we envisioned that the space proximity of the two pyrenes in P1 by γ-CD could facilitate the formation of a hairpin-structure without the addition of metal ions (Figure S4 in the Supporting Information). Therefore, we recorded the fluorescence emission spectra of P1 in the absence and the presence of γ-CD, which were obtained at room temperature in Tris-HCl buffer solution without any metal ions. In the absence of γ-CD, P1 displays slight pyrene excimer fluorescence (F486 nm/F378 nm ) (60) Tan, Z. J.; Chen, S. J. Biophys. J. 2008, 95, 738–752. (61) Kuznetsov, S. V.; Ren, C. C.; Woodson, S. A.; Ansar, A. Nucleic Acids Res. 2008, 36, 1098–1112.

Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

3917

Figure 1. Electronic absorption (A) and fluorescence emission (B) spectra of aqueous P1 solution (100 nM) in the absence (s) and the presence of β-CD ( · · · ) or γ-CD ( · - · - · ). [CDs] ) 2.0 mM, λex ) 344 nm.

Figure 2. Monomer and excimer fluorescence intensity changes of P1 (100 nM, λex ) 344 nm) as a function of temperature in the absence (a) and the presence of 8.0 mM γ-CD (b). Conditions: 2.5 mM MgCl2 in 20 mM Tris-HCl buffer.

2.81) due to few hairpin structures of the probe. The addition of γ-CD to the buffer solution containing the same concentration of P1 significantly increases excimer emission (F486 nm/ F378 nm ) 12.1). These observations demonstrate that the workability of a DNA probe can be attributed to space proximity of the labels, which essentially acts as a lock for closing the stemless constructs. Finally, we evaluated the rate of stem opening by comparing the DNA hybridization kinetics of P1 in the absence and the presence of γ-CD (Figure S5 in the Supporting Information). The excimer fluorescence intensity was monitored relative to time in the presence of complementary target T1 in Tris-HCl buffer solution. As shown in Figure 3, the complex of P1 with γ-CD was formed fast at room temperature and achieved equilibrium within 3 min with the maximum excimer intensity. Introduction of T1 to the solution of P1 and γ-CD results in a decrease in the excimer fluorescence as a result of the formation of duplex and separation of the two pyrenes. The time curve exhibits a rapid reduction of fluorescence intensity in the first 5 min followed by a slower decrease over the next 5-10 min, which is a little longer than that of P1 (Figure 3, inset). The slow hybridization kinetics could be explained by the high energy barrier created by the necessity of separation of both the stembase pairs and the pyrene dimer that was bounded in the γ-CD 3918

Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

Figure 3. Real-time fluorescence records of P1 (100 nM) at 486 nm upon addition of γ-CD (8 mM) and subsequent target T1 (50 nM). For the measurement, we distinguish four steps: (1) the cuvette was filled with 500 µL of Tris-HCl buffer solution, (2) P1 was introduced in the cuvette, (3) γ-CD was added, and (4) T1 was added. The transition between each regime is marked with an arrow. Inset: realtime fluorescence record of P1 (100 nM) at 486 nm upon addition of 50 nM T1.

cavity. The phenomenon is similar to the slow binding kinetics observed for MBs with a long stem. Comparison of Target Binding Sensitivity and Specificity. As noted above, the modulation of stem-stability of P1 by γ-CD implies that P1/γ-CD could be an alternative approach for DNA hybridization assay. We thus successively evaluated the work ability of P1 and P1/γ-CD according to sensitivity and selectivity of the detection. Parts A and B of Figure 4 show the fluorescence responses of P1 and P1/γ-CD, respectively, to different concentrations of T1. In the absence of target DNA, P1/γ-CD displays strong pyrenes excimer emission with a maximum centered band around 486 nm. Upon addition of T1 (0.1-1.2 equiv) to the solution, a significant decrease in the excimer fluorescence and increase in the monomer intensity are observed with a clear isoemissive point at around 425 nm. From Figure 4B, one can realize that P1 also exhibits monomer and excimer fluorescence responses toward T1. However, there is a rather large difference in the signal-to-background ratio (S/B) in the two cases because of

Figure 4. Fluorescence emission spectra of P1/γ-CD (A) and P1 (B) in Tris-HCl buffer solution in the presence of different concentrations of T1. The arrows indicate the signal changes as increases in the T1 concentration (0, 0.2, 2.0, 5.0, 8.0, 10, 20, 50, 80, and 100 nM). [P1] ) 100 nM, [γ-CD] ) 8.0 mM. λex ) 344 nm.

the different background signals. The S/B was determined by using equation62

S/B )

(F378 nm /F486 nm)hybrid (F378 nm /F486 nm)free

(1)

where the terms of (F378 nm/F486 nm)free and (F378 nm/F486 nm)hybrid take into account the ratios of monomer fluorescence intensity to the excimer fluorescence intensity of pyrene without the target and the probe-target hybrid, respectively. The (F378 nm/ F486 nm)free value of P1 was estimated to be 0.14 in the absence of T1, which is consistent with the stem-closed form of P1. Upon the addition of T1, hybridization between the probe sequence and the target diminishes the dimer form of pyrene, leading to an increase in the value of F378 nm/F486 nm. The (F378 nm/F486 nm)hybrid of P1 was approximated to be 8.85 when 1 equiv of T1 was added and the corresponding S/B is 63.2. However, in the case of P1/γ-CD, the S/B went up to 172.4 under the same conditions, which is significantly higher than that of P1. This comparison clearly demonstrated that the P1/γ-CD assembly greatly improves the S/B ratio and, thus, the analytical sensitivity of the probe. To quantitatively evaluate the spectral changes of the two approaches at 378 and 486 nm, the S/Bs of P1 and P1/γ-CD at the two wavelengths on the target concentration were obtained. As shown in Figure 5, the S/B could be clearly followed up at a sufficiently low concentration of 0.5 nM T1 in the case of P1/γCD, which is around 20-fold lower than that of P1. The detection limit of P1/γ-CD, defined as 3 times as the standard derivation in blank solution, is 0.1 nM. Moreover, by reducing the probe concentration, we observed that a lower detection limit could be obtained. The results suggest that P1/γ-CD is potentially appropriate for highly sensitive quantification of nucleic acid content. It has been demonstrated that hairpin-structured probes have high specificity with which to discriminate against non-complementary DNA sequences and single-base mismatches. We expect that the enhancement of stem stability of P1 by γ-CD will make P1/γ-CD outperform conventional MBs with its higher power to discriminate between the perfect target and the mismatches. (62) Marras, S. A. E.; Kramer, F. R.; Tyagi, S. Methods Mol. Biol. 2003, 212, 111–128.

Figure 5. S/B of P1 (a) and P1/γ-CD (b) as a function of the target concentrations. The measuring conditions as shown in Figure 4. Fluorescence was monitored at 486 and 378 nm with an excitation wavelength of 344 nm.

Figure 6. Monomer and excimer fluorescence changes of P1 and P1/γ-CD induced by different targets (100 nM) under the same conditions. The response toward T1 is used as the standard (F378 nm/F486 nm ) 100).

Figure 6 compares the changes of F378 nm/F486 nm of P1 and P1/ γ-CD in the presence of 1 equiv of T1, T2, and T3, respectively. The value for the perfectly complementary T1 was used as the standard (F378 nm/F486 nm ) 100). Both T1 and T2 increase the F378 nm/F486 nm relative to that without a target, and that of P1 by single-base mismatched T2 is 74% compared to perfectly matched T1. For P1/γ-CD, T2 reduced the F378 nm/F486 nm to be 61% compared to that generated by T1. The results reveal that the Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

3919

Figure 7. (A) Fluorescence emission spectra of P2 (λex ) 344 nm) under different experimental conditions: (a) 100 nM P2 in Tris-HCl buffer; (b) 100 nM P2 in Tris-HCl buffer + 0.1 µM Tmb; (c) 100 nM P2 in Tris-HCl buffer + 5.0 mM γ-CD; and (d) 100 nM P2 in Tris-HCl buffer + 5.0 mM γ-CD + 0.1 µM Tmb. (B) Monomer and excimer fluorescence changes of P2/γ-CD induced by different proteins (0.5 µM, y-axis markers) under the same conditions.

single nucleotide polymorphism (SNP) detection ability of P1/ γ-CD is higher than P1. This improvement of DNA binding specificity is due to an increase of thermostability of the hairpin structure through bounding the two pyrenes in the hydrophobic cavity of γ-CD. Since our purpose was to compare the selectivity of P1 and its γ-CD complex, the experiment was not carried out under optimized stringent conditions to achieve the best SNP performance. Because of the flexibility of the γ-CDmodulated thermostability, one can expect that the SNP detection capability could be conveniently tuned by varying the concentration of γ-CD. To further characterize the target binding specificity, the ability of the two probes to resist nonspecific protein binding was tested using single-stranded binding protein (SSB). This is necessary because the traditionally used MBs are subject to nuclease degradation and nonspecific binding by DNA or RNA binding proteins, which can produce false positive signals. In the assay, if the probe is digested by the protein, the pyrene’s monomer fluorescence will increase at the expense of excimer fluorescence because the two pyrene molecules will separate from each other. Although both P1 and P1/γ-CD display obvious monomer fluorescence enhancement by 1-fold excess of SSB, the response of P1/γ-CD toward SSB is significantly smaller than that of P1. As shown in Figure 6, the ratio of F378 nm/F486 nm of P1/γ-CD was estimated to be 13.8 in comparison with T1. This value is obviously lower than that of P1, in which the ratio of F378 nm/ F486 nm is approximated to be 62.1 under the same conditions, suggesting that the usage of γ-CD could improve the ability to resist nonspecific binding, makes this probe well suited to monitor a substrate in the complex biological milieu containing nonspecific binding substrates. Protein Assay Using P2/γ-CD. Our studies also demonstrated that the bounding interaction of γ-CD is efficient not only for the stem-forming DNA probe but also for stemless probe (Figure S6 in the Supporting Information). Figure 7A shows the fluorescence emission spectra of P2 at different conditions. Spectrum a was measured in Tris-HCl in the absence of γ-CD, where P2 shows not only moderate monomer fluorescence emission but also excimer fluorescence of pyrene, indicating that in aqueous solution P2 exists in a random state. Addition of 5-fold excess of Tmb to the solution induces a slight enhancement of the monomer fluorescence but no obvious excimer emission 3920

Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

change could be observed (spectrum b). The value of S/B is approximated to be 1.3. In the presence of γ-CD, the excimer fluorescence emission by pyrene is ∼7-fold that of the free P2 (spectrum c), suggesting the formation of the pyrene dimer of P2 due to two pyrene molecules are included in the CD cavity. Spectrum d of Figure 7A shows the fluorescence emission spectrum of P2/γ-CD in the presence of 5-fold excess of Tmb. Binding of the aptamer with Tmb leads to obvious decreases in the excimer fluorescence of pyrene concomitant with the increase of the monomer emission. In Figure 7A, the ratio of F378 nm/F486 nm of P2/γ-CD was estimated to be 0.2 in the absence of Tmb, while the value went up to 0.73 upon the addition. The S/B ) 3.7 is higher than the free P2 under the same conditions. The limit of Tmb detection, based on 3 times the signal-to-noise level, was estimated to be ∼15 nM, which is comparable with that of the regular dye-quencher pair-labeled aptamers.63,14 These results indicate that the formation of the conformationally constrained form makes P2 efficient in probing Tmb binding. To understand the response behaviors of P2/γ-CD toward different proteins, the fluorescence intensity changes of P2/γ-CD by different proteins (Tmb, IgG, HSA, BSA, and Cyt C) were studied. No obvious excimer or monomer fluorescence change could be observed by HSA, BSA, and cytochrome C (Cyt C) compared with the sample containing 100 nM Tmb. However, when tested with IgG at the concentration of 100 nM, P2 produced a slight decrease of the excimer fluorescence and an increase in the monomer to excimer intensity ratio (Figure 7B). We reasoned that the excimer and monomer fluorescence change of P2 induced by the protein is due to variation of the polarity of the solution. CONCLUSIONS In summary, the conformationally constrained DNA probes of either stem-containing oligonucleotide or stemless oligonucleotide could be easy constructed by bounding the oligonucleotided-labels into a hydrophobic cavity. With regard to the traditionally used stem-containing probes, the central hallmark of the design is the feasibility to conveniently tune the stem-stability and thus kinetic and thermodynamic features of the probes. For traditional DNA probe design, the variation of stem-stability can be achieved by (63) Jhaveri, S. D.; Kirby, R.; Conrad, R.; Maglott, E. J.; Bowser, M.; Kennedy, R. T.; Glick, G.; Ellington, A. D. J. Am. Chem. Soc. 2000, 122, 2469–2473.

variation of the stem length and sequences, which requires synthesizing different probes. Whereas the inclusion interaction of γ-CD with the stem-labels offers unprecedented, easy, and predictable new degrees of freedom to tune stem-stability, which allows for a very facile introduction of further functionalities of DNA probes such as not only increases the signal-to-background ratio but also improves the target binding specificity. Although we used γ-CD and pyrene in this design, its applicability is universal by selecting not only different types of the stem-labels and cyclodextrins but also the probe sequences to be applied to other types of molecular probes. These features establish the simplicity, effectivity, and universality of the platform and could, therefore, provide the groundwork for the design of functional nucleic acid probes for biosensing applications.

ACKNOWLEDGMENT The work was supported by the National Natural Foundation of China (Grant 20775005) and the National Grand Program on Key Infectious Disease (Grant 2009ZX10004-312). SUPPORTING INFORMATION AVAILABLE Cyclodextrin concentration effects, fluorescence delay curves, and Mg2+ and γ-CD concentration effects. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review February 21, 2010. Accepted April 1, 2010. AC1004713

Analytical Chemistry, Vol. 82, No. 9, May 1, 2010

3921