Carbon Nanotubes as a Low Background Signal Platform for a

Sep 20, 2010 - Although holding the advantages of both an aptamer and a molecular beacon (MB), a molecular aptamer beacon (MAB) needs complicated and ...
4 downloads 11 Views 1MB Size
Download PDF
Anal. Chem. 2010, 82, 8432–8437

Carbon Nanotubes as a Low Background Signal Platform for a Molecular Aptamer Beacon on the Basis of Long-Range Resonance Energy Transfer Shu Jun Zhen,† Li Qiang Chen,‡ Sai Jin Xiao,† Yuan Fang Li,† Ping Ping Hu,‡ Lei Zhan,§ Li Peng,‡ Er Qun Song,§ and Cheng Zhi Huang*,†,§ Education Ministry Key Laboratory on Luminescence and Real-Time Analysis, College of Chemistry and Chemical Engineering, College of Life Science, and College of Pharmaceutical Science, Southwest University, Chongqing 400715, China Although holding the advantages of both an aptamer and a molecular beacon (MB), a molecular aptamer beacon (MAB) needs complicated and expensive modifications at both of its ends and usually has a high background signal because of the low energy transfer efficiency between the donor and the acceptor. To overcome these shortcomings, in this study, we develop a long-range resonance energy transfer (LrRET) system by separating the donor from the acceptor, wherein only one end of the MAB is fluorescently labeled and acts as the energy donor and multiwalled carbon nanotubes (MWCNTs) are introduced as the energy acceptor. To test the feasibility of the newly designed MAB system, adenosine triphosphate (ATP) has been employed as a proof-of-concept target. It is found that the fluorescence of the designed MAB is completely quenched by MWCNTs, supplying a very low background signal. Then the quenched fluorescence is recovered significantly with the addition of ATP, so that ATP can be detected in the range of 0.8-80 µM with a limit of detection of 0.5 µM (3σ). Compared with the conventional fluorescence resonance energy transfer, the efficiency of LrRET between the dye and MWCNTs is much higher. Since only one end of the MAB needs the modification, the present strategy is simple and costeffective. Furthermore, the use of MWCNTs can greatly reduce the fluorescence background of the MAB and supply a high sensitivity, showing its generality for detection of a variety of targets. A molecular beacon (MB), a hairpin-shaped oligonucleotide whose 5′- and 3′-ends are respectively labeled with the fluorophore (donor) and the quencher (acceptor), is a type of fluorescence probe for real-time detection of nucleic acids. Generally, the MB forms a stem-loop structure, which can place the fluorophore and the quencher close to each other, resulting in fluorescence quenching owing to the occurrence of Fo¨rster resonance energy transfer (or fluorescence resonance energy transfer, FRET). When * To whom correspondence should be addressed. Phone: 86-23-68254659. Fax: 86-23-68866796. E-mail: [email protected]. † Education Ministry Key Laboratory on Luminescence and Real-Time Analysis, College of Chemistry and Chemical Engineering. ‡ College of Life Science. § College of Pharmaceutical Science.

8432

Analytical Chemistry, Vol. 82, No. 20, October 15, 2010

the MB binds to its target and forms a relatively rigid probe-target hybrid, the fluorophore dissociates from the quencher and thereby its fluorescence is restored. Therefore, this unique property makes the MB extremely useful in molecular biochemistry and clinical tests.1-4 Recently, to develop alternative MBs that depend on target molecules other than nucleic acids, a new type of probe, a molecular aptamer beacon (MAB), has been designed which holds the advantages of the high binding specificity and generality of an aptamer and the good signal transduction capability of an MB. The first example of the designation of an MAB was reported by Yamamoto et al.5 In their work, the RNA aptamer of the Tat protein of HIV was split into two subunits. One was designed to form a hairpin structure whose 5′- and 3′-terminals were modified with a fluorophore and a quencher, respectively, and the other was a nonstructured oligomer. In the presence of the Tat protein, the two oligomers undergo a conformational change to form a duplex, and thus, the fluorescence is enhanced. Since then, a variety of MABs have been developed to detect targets such as thrombin,6-10 platelet-derived growth factor (PDGF),11 erythropoietin,12 and so on.13 Although these MAB strategies have their distinct advantages, they have been confined by limitations such as (1) a high background signal owing to the low efficiency of FRET between the conventional fluorophore and quencher, reducing the sensitivity of the MABs, (2) expensive and difficult labeling similar to that of MBs since it is compulsory to label both (1) Wang, K.; Tang, Z.; Yang, C. J.; Kim, Y.; Fang, X.; Li, W.; Wu, Y.; Medley, C. D.; Cao, Z.; Li, J.; Colon, P.; Lin, H.; Tan, W. Angew. Chem., Int. Ed. 2009, 48, 856–870. (2) Broude, N. E. Trends Biotechnol. 2002, 20, 249–256. (3) Tan, W.; Wang, K.; Drake, T. J. Curr. Opin. Chem. Biol. 2004, 8, 547–553. (4) Marras, S. A. E.; Tyagi, S.; Kramer, F. R. Clin. Chim. Acta 2006, 363, 48–60. (5) Yamamoto, R.; Kumar, P. K. R. Genes Cells 2000, 5, 389–396. (6) Li, J. J.; Perlette, J.; Fang, X.; Kelley, S.; Tan, W. Proc. SPIEsInt. Soc. Opt. Eng. 2000, 3926, 27–33. (7) Hamaguchi, N.; Ellington, A.; Stanton, M. Anal. Biochem. 2001, 294, 126– 131. (8) Li, J. J.; Fang, X.; Tan, W. Biochem. Biophys. Res. Commun. 2002, 292, 31–40. (9) Hall, B.; Cater, S.; Levy, M.; Ellington, A. D. Biotechnol. Bioeng. 2009, 103, 1049–1059. (10) Heyduk, E.; Heyduk, T. Anal. Chem. 2005, 77, 1147–1156. (11) Vicens, M. C.; Sen, A.; Vanderlaan, A.; Drake, T. J.; Tan, W. ChemBioChem 2005, 6, 900–907. (12) Zhang, Z.; Guo, L.; Tang, J.; Guo, X.; Xie, J. Talanta 2009, 80, 985–990. (13) Goulko, A. A.; Li, F.; Le, X. C. Trends Anal. Chem. 2009, 28, 878–892. 10.1021/ac100709s  2010 American Chemical Society Published on Web 09/20/2010

fluorophores and quenchers at both ends of the MABs, (3) variations of the qualities of the labeling and the purification of the probes which strongly affect the fluorescence enhancement when reacting with the targets, and (4) easy digestion by endogenous nuclease when applied in vivo. In our opinion, these collective problems of MABs might be avoided by introducing carbon nanotubes (CNTs) since it is wellknown that CNTs can interact with soft single-stranded DNA (ssDNA) nonconvalently by means of π-π stacking interactions between nucleotide bases and the carbon nanotube sidewalls but hardly interact with rigid double-stranded DNA (dsDNA).14-17 Moreover, CNTs can act as excellent “nanoquenchers” for the fluorophore,18 and they can protect DNA against nuclease digestion and enhance the self-delivery capability and the intracellular stability of DNA compared with free DNA probes.19 On the other hand, as stated by Sebastian et al.,20,21 the traditional FRET occurs through dipole-dipole interaction when the donor and the acceptor are two dye molecules, which can be described by the Fo¨rster equation with the R-6 dependence of the rate of energy transfer on the relative distance (R) between the donor and the acceptor. However, the rate of FRET deviates from the usual R-6 dependence when either the donor or the acceptor or both are extended systems with delocalized charge densities.21 When the acceptor is a carbon nanotube, for example, the rate of energy transfer between the dye and the carbon nanotube has an R-5 dependence. It is obvious that the efficiency of the energy transfer of the R-5 dependence is higher than that of the R-6 dependence, and thus in our opinion, acting as an energy acceptor, CNTs might supply much higher analytical sensitivity and reliability. Therefore, in this paper, we propose a new strategy to design an MAB with only one end labeled by a fluorophore acting as the energy donor and multiwalled carbon nanotubes (MWCNTs) acting as the energy acceptor, by taking ATP, an important small molecule in many biological processes,22-26 as a target molecule, and develop a sensitive detection method of ATP. Compared with the common MAB, our designed MAB has a low background signal and is simple and cost-effective. In addition, our designed mechanism is general and may be useful in other MAB-based target detections. (14) Zheng, M.; Jagota, A.; Semke, E. D.; Diner, B. A.; Mclean, R. S.; Lustig, S. R.; Richardson, R. E.; Tassi, N. G. Nat. Mater. 2003, 2, 338–342. (15) Zheng, M.; Jagota, A.; Strano, M. S.; Santos, A. P.; Barone, P.; Chou, S. G.; Diner, B. A.; Dresselhaus, M. S.; Mclean, R. S.; Onoa, G. B.; Samsonidze, G. G.; Semke, E. D.; Usrey, M.; Walls, D. J. Science 2003, 302, 1545– 1548. (16) Chen, R. J.; Zhang, Y. J. Phys. Chem. B 2006, 110, 54–57. (17) Zhang, L.; Huang, C. Z.; Li, Y. F.; Xiao, S. J.; Xie, J. P. J. Phys. Chem. B 2008, 112, 7120–7122. (18) Yang, R.; Jin, J.; Chen, Y.; Shao, N.; Kang, H.; Xiao, Z.; Tang, Z.; Wu, Y.; Zhu, Z.; Tan, W. J. Am. Chem. Soc. 2008, 130, 8351–8358. (19) Wu, Y.; Phillips, J. A.; Liu, H.; Yang, R.; Tan, W. ACS Nano 2008, 2, 2023– 2028. (20) Swathi, R. S.; Sebastian, K. L. J. Chem. Phys. 2008, 129, 054703. (21) Swathi, R. S.; Sebastian, K. L. J. Chem. Sci. 2009, 121, 777–787. (22) Khakh, B. S.; North, R. A. Nature 2006, 442, 527–532. (23) Eguchi, Y.; Shimizu, S.; Tsujimoto, Y. Cancer Res. 1997, 57, 1835–1840. (24) Miyoshi, N.; Oubrahim, H.; Chock, P. B.; Stadtman, E. R. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 1727–1731. (25) Sperla´gh, B.; Kittel, A´.; Lajtha, A.; Vizi, E. S. Neuroscience 1995, 66, 915– 920. (26) Finger, T. E.; Danilova, V.; Barrows, J.; Bartel, D. L.; Vigers, A. J.; Stone, L.; Hellekant, G.; Kinnamon, S. C. Science 2005, 310, 1495–1499.

EXPERIMENTAL SECTION Apparatus. Fluorescence spectra were measured with an F-2500 fluorescence spectrophotometer (Hitachi, Tokyo, Japan). A vortex mixer, QL-901 (Haimen, China), was employed to blend the solution, and a high-speed TGL-16 M centrifuge (Hunan, China) was used for centrifugation of the solution. Materials. Two oligomers, Apt1 and Apt2, were designed by splitting the sequence of the ATP aptamer into two portions. The designed Apt1 has a hairpin structure with a sequence of 5′-TAMRACCC CTA CCT GGG GGA GTA TAT AAG GGG-3′, which has the specific loop sequence for ATP binding, while Apt2 has a sequence of 5′- TAT AGC GGA GGA AGG T-3′, which has not been labeled with any dye molecule. A standard doubly labeled molecular aptamer beacon (Apt3) with a sequence of 5′-TAMRACCC CTA CCT GGG GGA GTA TAT AAG GGG-DABCYL-3′ was designed to compare the signal-to-background ratio with that of our designed aptamer beacon in the presence of MWCNTs. A control sequence (Pn) of 5′-TAMRA-CCC CTT TTT TTT TTT TTT TTT TAG GGG-3′ was employed to confirm the specificity of our strategy. All of the above sequences were synthesized by Beijing SBS Genetech Co., Ltd. (Beijing, China). Adenosine 5′-triphosphate disodium salt (ATP), uridine 5′-triphosphate trisodium salt hydrate (UTP), cytidine 5′-triphosphate disodium salt hydrate (CTP), and guanosine 5′-triphosphate disodium salt hydrate (GTP) were commercially available from Sigma (St. Louis, MO). All of the above reagents were prepared by dissolving their commercial products in doubly distilled water (18.2 MΩ). Tris-HCl buffer solution (pH 7.2) was used to control the acidity of the reaction solution. Pretreatment of MWCNTs. The commercial MWCNTs were purified and oxidized according to the literature27 with some modifications. Briefly, 50 mg of MWCNTs was refluxed for two days in 100 mL of 2.0 mol/L HNO3. The solution was kept overnight, and the clear solution above the suspension was removed. The remaining suspension was centrifuged (14000 rpm, 30 min), and the precipitates were oxidized by 16 mL of 1:3 HNO3/H2SO4 in an ultrasonic bath for 2 h. The suspension was diluted 10-fold by water and kept overnight. After removal of the clear solution over the precipitates, the remaining solution was filtered through a 0.45 µm filtration membrane and further washed with water to neutral pH. The resulting precipitates were kept in an oven at 45 °C for 5 h to remove the water. The oxidized MWCNTs were suspended by 100 mL of water, and the final concentration of the MWCNTs was 0.476 mg/mL. ATP Detection. A certain concentration of ATP was added to the mixture of 47.6 µg/mL MWCNTs, 0.1 µM Apt1, and 0.1 µM Apt2 at pH 7.2. The mixture was kept at room temperature for 30 min and centrifuged at 6000 rpm for 5 min to remove the possible aggregations of MWCNTs. The supernatant was transferred for measurements of fluorescence on the F-2500 fluorescence spectrophotometer with an excitation wavelength of 553 nm. RESULTS AND DISCUSSION Design Strategy. As shown in Scheme 1A, the literaturereported MAB undergoes a conformational change from a hairpin (27) He, P.; Bayachou, M. Langmuir 2005, 21, 6086–6092.

Analytical Chemistry, Vol. 82, No. 20, October 15, 2010

8433

Scheme 1. Comparison of the Doubly Labeled MAB (A) and Singly Labeled MAB Using MWCNTs as the Nanoquencher (B) for the Detection of the Targeta

a In the case of part A, the aptamer of the target is split into two subunits: one is designed to form a hairpin structure with a fluorophore and a quencher at both of its ends, and the other is a nonstructured oligomer. Only in the presence of the target can the fluorescence of the MAB be turned on. In the case of part B, the MAB only needs to be labeled on one end. The fluorescence of the MAB is quenched by MWCNTs, which can be recovered after the reaction between the MAB and the target.

structure to a duplex in the presence of the target, giving enhanced fluorescence emission. Although it holds the advantages of both an MB and an aptamer, it has to be labeled with both the fluorophore and the quencher, which is complicated and expensive. Moreover, it has strong background signals because of the low FRET efficiency between the common fluorophore and quencher. Different from the above strategy in our present study, only one oligomer needs to be labeled with a fluorophore, and MWCNTs are used as nanoquenchers in the mode of long-range resonance energy transfer (LrRET).20,21 In the absence of target, both oligomers are adsorbed onto the sidewall of the MWCNTs, and the fluorescence of the dye-labeled oligomer is quenched by the MWCNTs. In the presence of target, however, the two subunits can hybridize with each other to form a duplex, leaving the sidewall of the MWCNTs, and thus, the fluorescence is turned on (shown in Scheme 1B). This strategy is simple and costeffective because the designed MAB needs to be labeled only on one end. Most importantly, it might have a very low fluorescence 8434

Analytical Chemistry, Vol. 82, No. 20, October 15, 2010

background because of the high FRET efficiency between the MWCNTs and the fluorophore. ATP-Dependent MAB Fluorescence Change. To test our strategy, the anti-ATP aptamer, which has the structure of two stacked G-quartets and two short double-helix stems,28 is split into two subunits. A stem sequence is added to one subunit to change it into a hairpin structure (Apt1) whose 5′-end is modified with TAMRA, and the other subunit (Apt2) is a nonstructured oligomer. To demonstrate the feasibility of our strategy, we first investigated the fluorescence features of the designed MAB that was assembled by Apt1, Apt2, and MWCNTs with the addition of ATP. A series of ATP solutions with final concentrations from 0.8 to 800 µM were added to the mixture of Apt1/Ap2/MWCNTs. As shown in Figure 1, when the excitation wavelength was kept at 553 nm, there was no fluorescence emission for Apt1/Apt2/ MWCNTs in the absence of ATP, and only green Rayleigh (28) Huizenga, D. E.; Szostak, J. W. Biochemistry 1995, 34, 656–665.

Figure 1. Recovered fluorescence emission with the addition of ATP. The inset pictures are photographed directly from the side window of the F-2500 spectrofluorometer when excited at 553 nm, showing the turn-on fluorescence emission gets stronger with increasing ATP. The green emission of photograph 1 is the Rayleigh scattering signals. Concentrations: ATP, 1-13, 0, 0.8, 4, 8, 12, 20, 40, 60, 80, 100, 120, 200, and 400 µM; Apt1, 0.1 µM; Apt2, 0.1 µM; MWCNTs, 47.6 µg/mL.

scattering light was visually observed from the side window of the fluorescence spectrophotometer (photograph 1 in the inset of Figure 1). With increasing ATP concentration, however, a sequential increase of the fluorescence emission centered at 575 nm could be measured (Figure 1) and could also be visualized directly through the digital pictures (inset in Figure 1). The redyellow color of the emission became more obvious with increasing ATP concentration, indicating the quenched fluorescence was restored gradually. These results confirm that the designed MAB can be successfully applied for ATP detection. It was assumed that both Apt1 and Apt2 were adsorbed on the sidewall of the MWCNTs by π-π stacking interactions without ATP, leading to the fluorescence of TAMRA on Apt1 quenched by the MWCNTs through the electron or energy transfer from TAMRA to the MWCNTs.29-31 In the presence of ATP, the split Apt1 and Apt2 may self-assemble to form a duplex structure32 and the stoichiometry is 2:1 (Figure S1 in the Supporting Information). As a result, the Apt1/Apt2 hybrid is far from the MWCNTs, and thus, the fluorescence of TAMRA is turned on. To confirm this mechanism, the MWCNTs were removed by means of centrifugal ultrafiltration (molecular weight cutoff 100 000), and the mixture trapped in the ultrafiltration device after ultrafiltration was diluted to the same volume before ultrafiltration. It was found that the MWCNTs complex trapped in the ultrafiltration device was nonfluorescent, but the product in the collection tube was highly fluorescent (Figure S2 in the Supporting Information). This result clearly indicated that the binding interaction of Apt1/Apt2 with ATP occurred in solution rather than on the surface of the MWCNTs. MWCNTs as a Low-Fluorescence Background Platform. Considering that CNTs can effectively quench the fluorescence of the fluorophore, acting as excellent nanoquenchers,18 herein (29) Nakayama-Ratchford, N.; Bangsaruntip, S.; Sun, X.; Welsher, K.; Dai, H. J. Am. Chem. Soc. 2007, 129, 2448–2449. (30) Boul, P. J.; Cho, D.-G.; Rahman, G. M. A.; Marquez, M.; Ou, Z.; Kadish, K. M.; Guldi, D. M.; Sessler, J. L. J. Am. Chem. Soc. 2007, 129, 5683– 5687. (31) Li, H.; Zhou, B.; Lin, Y.; Gu, L.; Wang, W.; Fernando, K. A. S.; Kumar, S.; Allard, L. F.; Sun, Y.-P. J. Am. Chem. Soc. 2004, 126, 1014–1015. (32) Stojanovic, M. N.; Prada, P. d.; Landry, D. W. J. Am. Chem. Soc. 2000, 122, 11547–11548.

Figure 2. MWCNTs as a low fluorescence background platform for the turn-on fluorescence signals with the addition of ATP. The inset shows the turn-on ratio (F/F0) at different concentrations of MWCNTs. Key: dark gray columns, 40 µM ATP; light gray columns, without ATP. Concentrations: Apt1, 0.1 µM; Apt2, 0.1 µM. λex ) 553 nm, and λem ) 575 nm. All data were collected from three measurements, and the error bars indicate the standard deviation.

we introduce MWCNTs to reduce the background of the MAB. Figure 2 can identify our strategy. In the absence of MWCNTs, although the fluorescence of Apt1 was turned on owing to the opening of the MAB structure with the addition of ATP to the mixture of Apt1 and Apt2, the high background fluorescence signal of TAMRA-modified Apt1 caused the fluorescence turn-on ratio (signal-to-background ratio, S/B, or F/F0 ) 1.01, where F0 and F are the fluorescence intensities of TAMRA at 575 nm in the absence and presence of ATP, respectively) to be relatively low. When MWCNTs were introduced, the background fluorescence signal was gradually reduced, and the fluorescence turn-on ratio increased with the addition of ATP (inset in Figure 2). Experiments showed that the fluorescence turn-on ratio could reach 9.99 when 47.6 µg/mL of MWCNTs was employed (Figure S3A in the Supporting Information) since the background fluorescence of TAMRA-modified Apt1 was quenched with a quenching efficiency of 99%. To further compare the signal-to-background ratio between our designed MAB and the traditional MAB, a standard doubly labeled molecular aptamer beacon (Apt3) with the same sequence as Apt1 has been designed. Under the same experimental conditions, the S/B was only 1.15 after the addition of ATP in the solution of Apt3/Apt2 (Figure S3B in the Supporting Information). These results clearly indicate that introduction of MWCNTs can significantly improve the sensitivity since they greatly reduce the fluorescence background and thus significantly enhance the S/B signal. It is well-known that the energy transfer is through the dipole-dipole interaction when the donor and the acceptor are two dye molecules, and the rate of this process can be evaluated by the following Fo¨rster equation:

k)

( )

1 R0 τ0 R

6

(1)

where R is the distance between the donor and the acceptor, τ0 is the lifetime of the donor in the absence of the acceptor, and R0 is the Fo ¨rster radius. Therefore, the rate of energy transfer between two dye molecules has an R-6 dependence. However, Analytical Chemistry, Vol. 82, No. 20, October 15, 2010

8435

when either the donor or the acceptor or both are extended systems with delocalized charge densities, as reported by Sebastian et al.,20,21 the rate of energy transfer deviates from the usual R-6 dependence. When the acceptor is a carbon nanotube, the rate follows another expression:21 k(pΩ) ) (33(µz()2 + 77(µF()2) 3m* 8192πp3ε2 R5 {mi}

∑

|

µeg2 m*(pΩ - εg) p2

-

mi2

|

a2 (2)

where pΩ is the transferred energy, R is the distance between the dye and carbon nanotube, m* is the effective mass, mi is the quantum number, ε refers to the permittivity of the medium separating the donor and the acceptor, εg is the band gap of the carbon nanotube, ± in µz± and µ±F denote states above (+) and below (-) the band gap, µz± and µ±F are the transition dipolars of the ith lattice site of the carbon nanotube, and µeg is the transition dipolar of the donor. Thus, the rate of energy transfer between the dye and carbon nanotube has an R-5 dependence. The common FRET usually occurs over a distance of about 10-100 Å. In contrast, according to Sebastian’s conclusion,21 it can occur at a larger distance between the dye and carbon nanotube, which is the so-called LrRET. That is to say, the energy transfer between the dye and carbon nanotube is more efficient than that between two dye molecules. Therefore, MWCNTs can provide a low background in our experiment. Splitting Chemistry of the ATP Aptamer. As stated above, we designed Apt1 and Apt2 by splitting the aptamer of ATP. Therefore, it is compulsory to identify the rationality of our design. Apt1 was designed with an extension of five bases at both of its ends to form the hairpin structure of MAB, and the loop of the designed MAB is the key portion of the anti-ATP aptamer. In the presence of ATP, the loop of Apt1 can assemble with Apt2 and the stem helix opens to form a duplex structure. Therefore, the molar ratio of Apt2 to Apt1 decides the interaction efficiency of ATP with Apt1 and Apt2, the amount of unreacted Apt1, and the restoration of the fluorescence of TAMRA. Experiments showed that the fluorescence of TAMRA was increased with increasing cApt2/cApt1 ratio, accompanied by an increased background signal (Figure 3). It was found that the ideal ratio of cApt2/cApt1 was 1.0 since it could induce the highest F/F0 ratio (inset in Figure 3). This phenomenon could be understood easily. On one hand, there might be some unspecific hybridization between Apt1 and Apt2 if the ratio of cApt2/cApt1 is higher than 1.0, leading to the higher background fluorescence signal. If the ratio of cApt2/cApt1 is lower than 1.0, on the other hand, there will not be enough Apt2 to hybridize with Apt1 after the addition of ATP, leading to the lower fluorescence restoration. Therefore, only when the amounts of both Apt1 and Apt2 are equal can they assemble perfectly in the presence of ATP, resulting in the biggest fluorescence enhancement with the lowest background signal. High Specificity and High Sensitivity of the ATP Detection. For the specificity study, the fluorescence changes of TAMRA brought by UTP, CTP, and GTP were compared with that brought 8436

Analytical Chemistry, Vol. 82, No. 20, October 15, 2010

Figure 3. Binding ratio of Apt2 and Apt1. The inset shows the turnon fluorescence signals with the addition of ATP at different ratios of Apt2 and Apt1. Key: light gray columns, without ATP; dark gray columns, 40 µM ATP. Concentrations: MWCNTs, 47.6 µg/mL; Apt1, 0.1 µM. λex ) 553 nm, and λem ) 575 nm. All data were collected from three measurements, and the error bars indicate the standard deviation.

Figure 4. Turn-on fluorescence (F/F0) of Apt1/Apt2/MWCNTs with the addition of 40 µM ATP, UTP, CTP, or GTP. Concentrations: Apt1, 0.1 µM; Apt2, 0.1 µM; MWCNTs, 47.6 µg/mL. λex ) 553 nm, and λem ) 575 nm. All data were collected from three measurements, and the error bars indicate the standard deviation.

by ATP considering the only difference between them was the base attached to ribose. The experimental results indicated that only ATP caused a dramatic fluorescence enhancement (F/F0 ) 13). The other three ATP analogues with the same concentrations of ATP failed to cause significant fluorescence enhancement (F/F0 ) 1.9 for UTP, 2.5 for CTP, and 1.3 for GTP) (Figure 4), indicating that UTP, CTP, and GTP could not interact with Apt1 and Apt2 to open the stem helix of Apt1. Thus, the interaction between ATP and Apt1/Apt2 was selective in the presence of MWCNTs, which was due to the inherent specificity of the aptamer toward ATP. To further verify the mechanism of this detection method, a control DNA sequence (Pn) was introduced. Pn also had the hairpin structure with the same number of bases as that of Apt1, but the loop sequence was changed from the portion of the antiATP aptamer to 17 thymines. At the same conditions as the above experiment, nonsignificant fluorescence changes were found after addition of ATP with final concentrations from 0.8 to 800 µM to

In recent years, aptamers have been widely employed in the detection of ATP.32,34,35 However, most of these methods are based on the turn-off mode in principle, which is usually unfavorable for its relatively low sensitivity and poor anti-interference ability compared with those of the turn-on assays. On the other hand, aptamer-based fluorescent turn-on strategies for ATP detection by switching structures from a DNA/DNA duplex to a DNA/target complex were reported,36-38 which, however, need the expensive and complex double labeling. Therefore, our designed MAB is obviously more suitable for ATP detection due to its cost-effective, simple, turn-on, selective, and sensitive properties.

Figure 5. Fluorescence intensity changes (F/F0) of Apt1/Apt2/ MWCNTs (triangles) and Pn/Apt2/MWCNTs (circles) at increasing ATP concentrations from 0.8 to 800 µM. The inset is the linear relationship between F/F0 and the concentration of ATP. The linear regression equation (cATP, µM) was F/F0 ) 2.6 + 0.3cATP for ATP over the range 0.8-80 µM with a correlation coefficient of 0.9938. The limit of detection was 0.5 µM (3σ). Concentrations: Apt1, 0.1 µM; Apt2, 0.1 µM; MWCNTs, 47.6 µg/mL. λex ) 553 nm, and λem ) 575 nm. All data were collected from three measurements, and the error bars indicate the standard deviation.

the solution of Pn/Apt2/MWCNTs (Figure 5). This control experiment supports our argument that the fluorescence enhancement resulted from the specific recognition between ATP and Apt1/Apt2. Apt1 and Apt2 can assemble together by ATP to form the duplex structure. However, when Apt1 is replaced by Pn, the specific recognition disappears, and the fluorescence of TAMRA on Pn is still quenched by MWCNTs in the presence of ATP. To determine the sensitivity of the method for ATP detection, different amounts of ATP were introduced to the complex of Apt1/ Apt2/MWCNTs. The typical response curve (Figure 5) showed that the fluorescence intensity increased with increasing amount of ATP until a plateau was reached. There was a good linear relationship between the fluorescence enhancement (F/F0) and ATP concentration over the range of 0.8-80 µM with a correlation coefficient of 0.9938 (inset in Figure 5). On the basis of the measurement of the standard deviation of the blank solution of Apt1/Apt2/MWCNTs in the absence of ATP, the limit of detection (LOD; 3σ) was determined to be 0.5 µM, which was comparable to that of the other reported aptamer-based ATP detection method.33 Li, N.; Ho, C.-M. J. Am. Chem. Soc. 2008, 130, 2380–2381. Wang, J.; Jiang, Y.; Zhou, C.; Fang, X. Anal. Chem. 2005, 77, 3542–3546. Merino, E. J.; Weeks, K. M. J. Am. Chem. Soc. 2003, 125, 12370–12371. Nutiu, R.; Li, Y. J. Am. Chem. Soc. 2003, 125, 4771–4778. Su, S.; Nutiu, R.; Filipe, C. D. M.; Li, Y.; Pelton, R. Langmuir 2007, 23, 1300–1302. (38) Tang, Z.; Mallikaratchy, P.; Yang, R.; Kim, Y.; Zhu, Z.; Wang, H.; Tan, W. J. Am. Chem. Soc. 2008, 130, 11268–11269. (33) (34) (35) (36) (37)

CONCLUSION In conclusion, we have developed a new MAB system by using MWCNTs as a quenching reagent in this study, where ATP has been selected as a model target. The turn-on mechanism is based on the conformational change of the MAB, which is derived from the anti-ATP aptamer, from a hairpin structure to a duplex structure upon ATP binding. As a result, the TAMRA-labeled oligomer moves away from the MWCNTs, and the fluorescence of MAB is recovered. Through the LrRET between TAMRA and MWCNTs, the background fluorescence signal is reduced and the S/B is improved greatly. This method shows a high sensitivity and selectivity for ATP detection. Thus, our method has great potential in biochemical studies and clinical diagnosis. Compared with the conventional MAB, our present strategy possesses at least two advantages. First, only one end of the MAB needs to be labeled in the newly designed system, which is simple and cost-effective. Second, the background signal is significantly reduced and the signal-to-background ratio is greatly enhanced, owing to the high energy transfer efficiency between the MWCNTs and the fluorescent label. Given that a similar MAB system can be constructed with a variety of other aptamers, it is expected that the newly designed strategy has universal applications not only for small molecule detection but also for large molecule (e.g., proteins) detection. ACKNOWLEDGMENT We are grateful for the financial support of the National Natural Science Foundation of China (Grant Nos. 20775061, 21035005, and 90813019). SUPPORTING INFORMATION AVAILABLE Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review March 18, 2010. Accepted September 6, 2010. AC100709S

Analytical Chemistry, Vol. 82, No. 20, October 15, 2010

8437