Conformational Switching Immobilized Hairpin DNA Probes Following

Aug 31, 2011 - The sequence-specific detection of DNA hybridization has attracted considerable interest in numerous fields including molecular diagnos...
1 downloads 0 Views 2MB Size
ARTICLE pubs.acs.org/ac

Conformational Switching Immobilized Hairpin DNA Probes Following Subsequent Expanding of Gold Nanoparticles Enables Visual Detecting Sequence-specific DNA Yajing Niu, Yanjun Zhao, and Aiping Fan* Tianjin Key Laboratory for Modern Drug Delivery & High-Efficiency, School of Pharmaceutical Science and Technology, Tianjin University, Tianjin 300072, China ABSTRACT: A simple, rapid, and sensitive method for visual detection of sequence-specific DNA was developed using hairpin DNA as the recognition element and hydroxylamine-enlarged gold nanoparticles (Au-NPs) as the signal producing component. In the assay, we employed a hairpin DNA probe dually labeled with amine and biotin at the 50 - and 30 -end, respectively. The probe was coupled with reactive N-oxysuccinnimide in a DNAbind 96-well plate. Without the target DNA, the immobilized hairpin probe was in a “closed” state, which kept the streptavidin-gold off the biotin. The hybridization between the loop sequence and the target broke the short stem duplex upon approaching the target DNA. Consequently, biotin was forced away from the 96-well plate surface and available for conjugation with the streptavidin-gold. The hybridization could be detected visually after the HAuCl4NH2OH redox reaction catalyzed by the Au-NPs. Under the optimized conditions, the visual DNA sensor could detect as low as 100 amol of DNA targets with excellent differentiation ability and even a single-base mismatch.

T

he sequence-specific detection of DNA hybridization has attracted considerable interest in numerous fields including molecular diagnostics, environmental monitoring, and antibioterrorism.1,2 Wide-scale DNA hybridization tests necessitates the development of easy-to-use, inexpensive, and miniaturized analytical devices. The sandwich-type strategy was popularly employed, which involved a capture probe and a labeled reporter probe to recognize the target DNA.36 A range of reporting moieties including fluorescent dyes,7,8 radioactive species,9 and chemiluminescent10,11 and electrochemical compounds1214 have been designed previously to facilitate the sensitive analysis. However, these methods generally suffer from the low specificity and are incapable of differentiating single-nucleotide polymorphisms (SNPs).15 Molecular beacon (MB) comprising a hairpin-like DNA stemloop has been reported to exhibit high differentiation ability toward SNPs.1618 The high specificity arises from the conformational constraint of the stem-loop structure. However, the ability of traditional MBs in quantitative assay is limited mainly because residual fluorescence from the hairpin greatly reduces the detection sensitivity of MBs.19 To address this, several research groups are engaged in the modification of MBs.20,21 For example, Heeger et al. developed electrochemical DNA (E-DNA) sensors to quantify 17-mer target DNA at a concentration as low as 10 pM.22 Zhang et al. developed an electrogenerated chemiluminescence (ECL)-based DNA sensor by self-assembling the thiolated hairpin DNA that was tagged with ruthenium complex on the surface of a gold electrode.23 These sensors as a “signal-off” device could assist the sensitive detection of DNA targets without the use of exogenous reagents. Nevertheless, r 2011 American Chemical Society

“signal-off” sensors pose a major limitation in that they are highly susceptible to false-positive responses. Hence, for the detection of sequence-specific DNA, the “signal-on” sensors were developed, wherein the hairpin-like DNA probes were immobilized at one terminus and the affinity label was fixed at the other end. For example, Bredehorst et al. designed DNA stem-loop structured probes for enzymatic detection of nucleic acid targets.24 Among all of these studies, the enzyme was employed as the detecting molecule, which would induce poor reproducibility and relatively high background interference. Alternatively, nanomaterials have been widely used in chemical and biological sensing because of their unique optical and mechanical properties.25,26 Au-NPs are particularly attractive in a bioassay by virtue of their facile synthesis, large specific surface area, high chemical stability, and biocompatibility.27,28 Hence, Au-NPs coupled to nucleic acids have been widely used as active units to assay sequence-specific DNA.29 In order to improve the sensitivity and specificity, much work has been done in Au-NPs based biosensing.13,30 Our group developed a gold staining method by using hydroxylamine as the reductant for visual detecting sequence-specific DNA.31 The assay results can be observed visually or by using a microplate spectrophotometer, which greatly simplified the procedure and offered a great promise in a wide range of biomedical applications. Nevertheless, to the best of our knowledge, the determination of sequencespecific DNA using hairpin DNA as a recognition element and Received: July 7, 2011 Accepted: August 31, 2011 Published: August 31, 2011 7500

dx.doi.org/10.1021/ac201755x | Anal. Chem. 2011, 83, 7500–7506

Analytical Chemistry

ARTICLE

hydroxylamine-enlarged Au-NPs as visual signal amplification has not been reported yet. In this paper, we developed a “signal-on” sensor for visually detecting DNA hybridization. In this novel approach, we designed a hairpin structure DNA probe to improve selectivity and used the Au-NPs as an “amplifier” for visual detection. The strategy of DNA hybridization detection was shown (Scheme 1). The hairpin DNA probe modified with an amine at the 50 -end and a biotin at the 30 -end was coupled to the surface of a DNA-bind 96-well plate. In the absence of target DNA, the immobilized hairpin probe was in the “closed” state and the biotin was inactive since it was embedded on the plate surface via the blocking agent. Upon contact with the target DNA, the hairpin structure was destroyed and the biotin group was forced away from the surface of the 96-well plate, which made it available for conjugating with the streptavidin-gold. Finally, the specific DNA sequence determination was translated into Au-NP growth signal through a HAuCl4NH2OH redox reaction. The new visual DNA assay combines the inherent signal amplification of gold staining with discrimination against nonhybridized DNA without using complex instruments. The optimization and performance of the new “signalon” device in detecting sequence-specific DNA related to the hepatitis B virus (HBV) are investigated.

’ EXPERIMENTAL SECTION Reagents and Apparatus. All chemicals were of analytical grade and were used as received. Bovine serum albumin (BSA) was purchased from Dingguo Changsheng Biotechnology Co., Ltd. (Beijing, China). Au-NPs (5, 15, 40, and 80 nm, 0.01% HAuCl4, 0.75 A520 units/mL) were purchased from Hualan Chemical Technology Co., Ltd. (Shanghai, China). Lysine, leucine, hydroxylamine, chlorauric acid, and other chemical reagents were obtained from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). All solutions were prepared using deionized

Scheme 1. Scheme for the Visual Detection of Target DNA via Hybridization and Subsequent Expanding of StreptavidinGold Nanoparticles

water generated by a Millipore-XQ system. The oligonucleotide probes were acquired from Sangon Biological Engineering Technology & Services Co., Ltd. (Shanghai, China) and had the following sequences in Table 1. The hairpin DNA probe is modified with an amine group at the 50 -end and a biotin at the 30 -end, respectively. It has six complementary bases at its 50 - and 30 -ends, which results in the formation of the stem-loop structure at an appropriate ionic strength. A TECAN (A-5002) microplate spectrophotometer (M€annedorf, Switzerland) was used to record the absorption signals of the samples. A DNA-bind 96-well plate from Corning Inc. was used as the platform for probe immobilization, DNA hybridization, and spectrophotometric measurements. Preparation of the Streptavidin-Gold. Streptavidin-coated Au-NPs were prepared via slightly modifying the previously published method.32 The streptavidin (10% more than the minimum amount, which was determined using a flocculation test) was added to 1 mL of pH-adjusted 5, 15, 40, and 80 nm of colloidal gold suspension (pH 6.4, adjusted by 0.1 M K2CO3), respectively, followed by incubation at room temperature for 30 min. The conjugate was centrifuged for 60 min (25 000, 17 000, 12 000, and 6 000 rpm) and the red soft sediment was resuspended in the hybridization buffer solution (PBS consisted of 0.01 M phosphate-buffered saline, 1.37 M NaCl, 3 mM KCl, and 5 mM MgCl2). The final volume was one tenth of the value of the original Au-NPs. The addition of BSA with a final concentration of 2% allows storage of streptavidin-gold at 4 °C for several days. Procedure of Hairpin DNA-Based Sequence-Specific DNA Detection. In a typical experiment, the stem-loop oligonucleotide probes were diluted to 1 pmol in 50 μL of coupling buffer (0.05 M Na2HPO4NaH2PO4, pH 8.5), and the diluted stemloop oligonucleotide probes were divided into the wells of a 96well plate (50 μL per well). Following the incubation with gentle mixing at 37 °C for 60 min, the wells were washed in triplicate with washing buffer (7 mM Tris-HCl, pH 8.0, 0.01 M NaCl, 0.05% Tween 20) before adding the blocking solution (2% BSA in coupling buffer). A volume of 50 μL of a differing amount of cDNA samples in a hybridization buffer was added and incubated at 37 °C for 60 min. The wells were then washed in triplicate with washing buffer, and then the plate was placed in a refrigerator overnight. After the supernatant was discarded, 50 μL of streptavidin-gold was then added and incubated in the hybridization buffer containing 2% BSA at 4 °C for 20 min. The wells were rinsed four times with ice-cold HW buffer (hybridization buffer containing 0.05% Tween 20) to remove the unreacted streptavidin-gold. Au-NPs assembled on the surface of the 96-well plate were catalytically enlarged in the presence of 18 mM NH2OH and 3 mM HAuCl4 at room temperature for 20 min. After the hydroxylamine amplification, the signal could be observed visually and the data were recorded using a microplate spectrophotometer.

Table 1. Sequences of Oligonucleotides Used in This Study name

sequence

hairpin DNA probe

50 -NH2-GGCCGTCCTAATCTCCTCCCCCAACACGGCC-biotin-30

T1

50 -TGGGAGGAGTTGGGGGAGGAGATTAGGTTAAAGGT-30

T2

50 -TGGGAGGAGTTGGGGGACGAGATTAGGTTAAAGGT-30

T3 T4

50 -TGGGAGGAGTTGGCGGACGAGATTAGGTTAAAGGT-30 50 -TGGGAGGAGTTGGCGGACGAGAATAGGTTAAAGGT-30

T5

50 -TGGGAGGAGTTGCCGGACGAGAATAGGTTAAAGGT-30 7501

dx.doi.org/10.1021/ac201755x |Anal. Chem. 2011, 83, 7500–7506

Analytical Chemistry

Figure 1. Absorbance of the gold staining at different wavelengths with target DNA (Δ) and without target DNA (2) and the ratio of corresponding signal (signal/noise ratio, Δ). Experimental conditions: hairpin DNA, 1 pmol; target DNA, 25 fmol; streptavidin-gold (15 nm), 1:10; the reaction time between biotin and streptavidin-gold (TBS), 20 min; HAuCl4, 3 mM; NH2OH, 18 mM; the reduction time (TR), 20 min.

ARTICLE

Figure 3. Optimization of BSA concentration. Experimental conditions: different concentrations of BSA, other experimental conditions were the same as Figure 1.

Figure 4. Detection of hybridization of the probe with complementary target using different sizes of streptavidin-gold. Experimental conditions were the same as Figure 1.

Figure 2. Effect of block reagents on signal and background values. Experimental conditions: lysine, 10 mM; leucine, 10 mM; Tris, 10 mM; BSA, 2%; other experimental conditions were the same as Figure 1.

’ RESULTS AND DISCUSSION We developed a novel hairpin DNA probe dually labeled with amine and biotin at the 50 - and 30 -end, respectively, which could be directly coupled to the reactive N-oxysuccinnimide esters (referred to as the NOS group) surface of the DNA-bind 96-well plate and transduced to a visual signal by using streptavidin-gold nanoparticles. In the absence of target DNA, the immobilized hairpin probe was in a “closed” state, which shielded biotin from being approached by the streptavidin-gold (Scheme 1). In the presence of target DNA, the loop sequence was hybridized with target, which broke the relatively shorter stem duplex. Consequently, biotin was forced away from the surface of the 96-well plate and became available to the streptavidin-gold. Then, the Au-NPs which assembled on the surface of the 96-well plate were catalytically enlarged in the presence of HAuCl4 and NH2OH. The gold staining signal could be observed visually or by a microplate spectrophotometer. Optimization of Detection Wavelength. NH2OH was employed as the reducing agent for the gold deposition of Au-NPs in solution. While NH2OH is capable of reducing Au3+ to bulk metal, the reaction is dramatically accelerated by Au surfaces. As a result, no new particle nucleation occurs in solution, and all

added Au3+ goes into production of larger Au-NPs, which results in the enhancement of optical absorbance.31 Previously, we developed a homogeneous detection model for sensitive detection of sequence-specific DNA, in which the maximum absorption wavelength for the enlarged Au-NPs is about 560 nm.33 However, the hairpin DNA based assay is a heterogeneous detection system, in which the Au-NPs are finally immobilized on the surface of 96well plates. Parts of the Au-NPs surface face to the 96-well plate, which cannot react with the HAuCl4NH2OH solution. Hence, the size and shape of the enlarged Au-NPs for the heterogeneous detection system might be different from that of the homogeneous detection system, which strongly influence the optical properties of enlarged Au-NPs. Hence, we investigated the absorption behavior of the heterogeneous detection model by using a microplate reader. The detection wavelength ranged from 490 to 700 nm, and the absorbance of gold staining with or without target DNA were detected every 10 nm wavelength interval. The absorbance of gold staining with and without target DNA is illustrated in parts A and B of Figure 1, respectively. Irrespective of the presence of target DNA, the absorption value increased gradually from 490 to 700 nm. However, the signal ratio (referred to as the signal/noise ratio, Figure 1C, red triangle) increased at the beginning, leveled off between 630 and 660 nm, and then decreased slowly. Hence, 630 nm was selected as an analytical parameter for quantitative analysis of the particle growth throughout the assay. Selection of Blocking Reagent. It is essential to use the blocking reagent to block the remaining NOS group after the 7502

dx.doi.org/10.1021/ac201755x |Anal. Chem. 2011, 83, 7500–7506

Analytical Chemistry

Figure 5. Optimization of hairpin DNA amount: plot for hairpin DNA amount vs signal/noise ratio. Experimental conditions: different amounts of hairpin DNA, target DNA, 50 fmol; streptavidin-gold (15 nm), 1:6.7; TBS, 5 min; HAuCl4, 4 mM; NH2OH, 18 mM; TR, 20 min.

Figure 6. Optimization of streptavidin-gold concentration: plot for dilution coefficient of streptavidin-gold vs signal/noise ratio. Experimental conditions: different concentrations of streptavidin-gold (15 nm), hairpin DNA, 1 pmol; other experimental conditions were the same as Figure 5.

coupling of the hairpin DNA probe to the 96-well plate. The remaining NOS group may react with streptavidin-gold due to the amino-group of streptavidin, which may generate a high background. As shown in Figure 2, the background signal with no blocking reagent in coupling buffer was higher than those with blocking reagents. The blocking reagent has three purposes. One is to block the remaining NOS group of the 96-well plate after hairpin DNA probe coupling. The second job is to embed the biotin group at the 30 -end of the hairpin DNA probe and make it completely inactive upon the formation of the hairpin structure. The last role is to maintain the orientation of hairpin DNA probes on the surface of the 96-well plate, which results in a high DNA hybridization efficiency. Hence, the molecular size and concentration of blocking agent would affect the performance of the sensor. First, the effect of molecular size was investigated. Several kinds of blockers including macromolecule blocker (BSA) and small molecule blockers (Tris, lysine, leucine) were tested (Figure 2). Compared to small molecule blockers, a relative small and stable background was obtained using BSA as the blocking agent. We speculated that the small molecule blockers were too small to embed the biotin, which resulted in a partially active biotin and a corresponding relatively high background. Hence, BSA was considered to be an ideal blocking agent in the assay. Second, we studied the effect of BSA concentration on the absorbance of the tests by varying the BSA concentration in the coupling buffer (from 0.001% to 4%). It was found that the

ARTICLE

Figure 7. Optimization of TBS: plot for TBS vs signal/noise ratio. Experimental conditions: different TBS; hairpin DNA, 1 pmol; streptavidin-gold (15 nm), 1:10; other experimental conditions were the same as Figure 5.

Figure 8. Optimization of HAuCl4 concentration: plot for HAuCl4 concentration vs signal/noise ratio. Experimental conditions: hairpin DNA, 1 pmol; target DNA, 50 fmol; streptavidin-gold (15 nm), 1:10; TBS, 20 min; different concentrations of HAuCl4; NH2OH, 18 mM; TR, 20 min.

background decreased slowly with increasing BSA concentration and then plateaued. However, the signal of perfect-matched target DNA also decreased slowly with increasing BSA concentration. The signal/noise ratio increased with raising BSA concentration from 0.001% to 2%; further increasing the concentration of BSA led to the drop of the signal/noise ratio (Figure 3). Finally, 2% BSA was selected for the following experiments. Optimization of the Size of Streptavidin-Gold. The size of streptavidin-gold may affect the binding efficiency between biotin and streptavidin. Hence, we employed streptavidin-gold differing in size (5, 15, 40, and 80 nm) as detector molecules and tried to find out the optimum NP size for the assay (Figure 4). We found that both background and sample signals increased with decreasing particle size. We assumed that the Au-NPs with smaller sizes might show lower steric hindrance, therefore providing higher reaction efficiency between the biotin and streptavidin-gold. Hence, a higher signal was obtained by using a smaller particle. However, the background signal increased sharply by using 5 nm of streptavidin-gold. We speculated that the size of 5 nm of streptavidin-gold may be too small. They may travel into the gap of blocking reagent, which also leads to a partially reactive biotin and a corresponding relatively high background signal. It is evident that the best response signal was obtained by using 15 nm of streptavidin-gold. In terms of 40 and 80 nm streptavidin-gold, 7503

dx.doi.org/10.1021/ac201755x |Anal. Chem. 2011, 83, 7500–7506

Analytical Chemistry

ARTICLE

Figure 9. Optimization of NH2OH concentration: plot for NH2OH concentration vs signal/noise ratio. Experimental conditions: different concentrations of NH2OH; HAuCl4, 3 mM; other experimental conditions were the same as Figure 8.

Figure 11. Visual detection (A1 and A2) and calibration curve of absorbance at 630 nm (B) vs different amounts of complementary target DNA. Target DNA, A1 (from left to right) 50, 25, 10, 7.5, 5, 2.5, 1, 0 fmol; A2 (from left to right) 1, 0.1, 0.01, 0 fmol; other experimental conditions were the same as Figure 1.

Figure 10. Optimization of TR: plot for TR vs signal/noise ratio. Experimental conditions: HAuCl4, 3 mM; NH2OH, 18 mM; different reduction times; other experimental conditions were the same as Figure 8.

there is not much difference in response with the increase of the target amount. Therefore, 15 nm of streptavidin-gold was chosen for the following studies. Optimization of Other Assay Conditions. To obtain the best detection sensitivity and selectivity, several parameters have been systematically investigated and optimized, including the hairpin DNA amount, the concentration of streptavidin-gold, TBS, the concentration of HAuCl4 and NH2OH, and TR. Since the levels of both background and target signal depend on the number of immobilized hairpin DNA probes, the effect of different amounts of hairpin DNA probes on the assay performance was first analyzed (Figure 5). The signal/noise ratio increased when the amount of hairpin DNA probes was increased from 0.5 to 1 pmol and then decreased rapidly. This was because the background signal was greatly enhanced. Hence, 1 pmol of hairpin DNA probe was selected for further investigations. To obtain a maximum response using a minimal concentration of streptavidin-gold, we then optimized the concentration of streptavidin-gold by investigating the signal/noise ratio of the DNA assay with different dilution of stock streptavidin-gold (15 nm) (Figure 6). High signal/noise ratio could be obtained when a stock streptavidin-gold’s solution is diluted 10 times. It is 7.80 for 50 pmol of target DNA. Hence, a 10 times dilution of stock streptavidin-gold (15 nm) was used for the following studies.

Figure 12. Specificity of the DNA assay detecting different targets. T1, perfect matched DNA; T2, one-base mismatched DNA; T3, two-bases mismatched DNA; T4, three-bases mismatched DNA; T5, four-bases mismatched DNA. The target amounts of T1, T2, T3, T4, and T5 are 25 fmol. All data were normalized to the perfect matched target sample.

Another factor to affect the sensitivity of the test is the incubation time in the presence of streptavidin-gold (TBS). It was found that the background signal was maintained almost the same even when the incubation time ranged from 5 to 20 min. Such a small and stable background signal benefits from an appropriate blocking reagent, which embed the biotin group at the 30 -end of hairpin DNA and made it completely inactive when the hairpin structure formed. The signal/noise ratio increased with increasing incubation time within 20 min and then decreased gradually, due to that the background signal was slightly enhanced (Figure 7). We postulate that the longer reaction time (TBS) means streptavidin-gold has more chances to approach the shield biotin. Hence, extending the reaction time is not appropriate in our experiment. As a result, an incubation time of 20 min at 4 °C was selected for the following studies. NH2OH was employed as the reducing agent in the signal amplification step, where gold metal was catalytically deposited 7504

dx.doi.org/10.1021/ac201755x |Anal. Chem. 2011, 83, 7500–7506

Analytical Chemistry

ARTICLE

Table 2. Comparison of Sensitivity for Different DNA Assay Methods analytical method

label

detection limit

sensor mode

electrochemical detection

ferrocene

10 pM22

turn-off

electrochemical detection

methylene blue

400 fM34

turn-on

electrochemical detection

HRP

10 fM35

turn-on

electrochemical impedance spectroscopy

unlabeled

10 pM36

turn-on

colorimetric detection

AP

60 pM24

turn-on

colorimetric detection

Au-NPs

50 pM37

turn-on

colorimetric detection (this work)

Au-NPs

2 pM

turn-on

electrogenerated chemiluminescence fluorescent detection

ruthenium complex tetramethylrhodamine

90 pM38 10 nM39

turn-off turn-on

fluorescent detection

oxazine derivative MR121

100 fM40

turn-on

onto the surface of Au-NP labels. Hence, both TR and the concentration of NH2OH and HAuCl4 affected the quantity of gold atoms that deposited on the surface of gold labels. The influence of HAuCl4 concentration was first investigated (Figure 8). As the concentration of HAuCl4 increased, the signal/noise ratio increased at the beginning, indicating that more and more gold metal was catalytically deposited onto the surface of Au-NPs and leveled off between 3 and 4 mM. Hence, 3 mM of HAuCl4 was selected for the following experiments. The effect of NH2OH concentration on the performance of the assay is shown in Figure 9. The signal/noise ratio increased quickly when the concentration of NH2OH rose from 0.5 to 4.5 mM, then increased slowly, and maintained almost the same within the range of 1824 mM. Thus, 18 mM of NH2OH was selected for the following experiments. The influence of the reduction time was illustrated as shown in Figure 10. The signal/noise ratio increased when the reduction time increased from 5 to 20 min and then maintained almost the same in the range of 2025 min. After that, the signal/noise ratio decreased quickly due to that the background signal was greatly enhanced. Hence, the reduction time was set at 20 min. Assay Performance. The quantitative behavior was assessed by monitoring the dependence of the absorbance at 630 nm upon the amount of target DNA. Experiments were carried out by adding increasing amounts of target DNA to the DNA sensor to examine whether the absorbance change could be used for sequence-specific DNA quantification. Figure 11A1 and Figure 11A2 shows the colorific changes of the DNA sensor upon detecting different amounts of target DNA. The absorbance at 630 nm increased significantly with increasing target binding. A good linear relationship between the absorbance at 630 nm and the target amount was from 0.1 to 25 fmol for sensitive sequencespecific DNA quantification with a correlation coefficient of 0.9812 for the linear calibration curve shown as Figure.11B. Also, at an amount higher than 25 fmol, the absorbance at 630 nm was slightly increased and deviated from the calibration curve. The detection limit was experimentally found to be 100 amol (>3 SD), i.e., 2 pM (100 amol in the 50 μL sample). Further lowering of the detection limit could be achieved in connection with a dissolved colloidal gold-based chemiluminescent technique which is expected to improve the detection sensitivity 100fold. In addition, a series of seven repetitive measurements of 1, 7.5, and 25 fmol of target DNA was used for estimating the precision and yielded the relative standard deviation of 7.53%, 9.70%, and 7.99%, respectively. Specificity of DNA Hybridization. In order to evaluate the selectivity of this novel visual DNA sensor, we challenged the

sensor with a single-base mismatched oligonucleotide (T2), a two-bases mismatched oligonucleotide (T3), a three-bases mismatched oligonucleotide (T4), and a four-bases oligonucleotide (T5). As shown in Figure 12 (the signal was normalized with the signal of perfect matched DNA), the signal of T2, T3, T4, and T5 were calculated to be 33% ( 10%, 21% ( 5%, 2% ( 3%, and 5% ( 4%, respectively, suggesting that this novel DNA sensor is highly selective and may be used to differentiate even a singlebase mismatch. These results indicate that the developed assay could exhibit an excellent specificity to distinguish the single-base mismatched DNA target, which benefits from using the stemloop DNA probe as the sensing component.

’ CONCLUSIONS In this study, we developed a simple, rapid, and sensitive method for the visual detection of nucleic acid target. The hairpin DNA probe was used to selectively detect the target DNA that could induce a conformation change of the hairpin probe. The hydroxylamine-enlarged Au-NPs were employed to transduce such a conformation change to a visual signal. This novel DNA assay showed a high sensitivity compared to other DNA assay techniques (Table 2). Further improving the sensitivity is still desirable to meet high-end requirements. One of the straightforward approaches is releasing gold ions from the Au-NPs by using an oxidative HClNaClBr2 solution and detecting the released gold ions by using a sensitive chemiluminescent method, which is expected to improve the detection sensitivity 100-fold. Another advantage of this assay is the high selectivity and specificity. In addition, by taking advantage of the visualization read-out, this hairpin DNA probe-based DNA molecular recognition does not need complex instruments and can be extended for many other biological assays. ’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT We acknowledge the financial support from the Innovation Foundation of Tianjin University and National Natural Science Foundation of China (Grant 21105071). ’ REFERENCES (1) Debouck, C.; Goodfellow, P. N. Nat. Genet. 1999, 21, 48–50. 7505

dx.doi.org/10.1021/ac201755x |Anal. Chem. 2011, 83, 7500–7506

Analytical Chemistry (2) Heller, M. J. Annu. Rev. Biomed. Eng. 2002, 4, 129–153. (3) Patolsky, F.; Lichtenstein, A.; Willner, I. J. Am. Chem. Soc. 2001, 123, 5194–5205. (4) Zhang, Y. C.; Pothukuchy, A.; Shin, W.; Kim, Y.; Heller, A. Anal. Chem. 2004, 76, 4093–4097. (5) Cao, Z. J.; Li, Z. X.; Zhao, Y. J.; Song, Y. M.; Lu, J. Z. Anal. Chim. Acta 2006, 557, 152–158. (6) Dong, X. Y.; Mi, X. N.; Xu, J. J.; Chen, H. Y. Talanta 2011, 84, 531–537. (7) Rant, U.; Arinaga, K.; Fujita, S.; Yokoyama, N.; Abstreiter, G.; Tornow, M. Nano Lett. 2004, 4, 2441–2445. (8) Sanborn, M. E.; Connolly, B. K.; Gurunathan, K.; Levitus, M. J. Phys. Chem. B 2007, 111, 11064–11074. (9) Burnette, W. H. Anal. Biochem. 1981, 112, 195–203. (10) Cai, S.; Xin, L.; Lau, C. W.; Lu, J. Z. Analyst 2010, 135, 615–620. (11) Fan, A. P.; Cao, Z. J.; Li, H. A.; Kai, M.; Lu, J. Z. Anal. Sci. 2009, 25, 587–597. (12) Fojta, M.; Kostecka, P.; Trefulka, M. R.; Havran, L.; Palecek, E. Anal. Chem. 2007, 79, 1022–1029. (13) Park, S. J.; Taton, T. A.; Mirkin, C. A. Science 2002, 295, 1503–1506. (14) Wang, J.; Musameh, M.; Lin, Y. J. Am. Chem. Soc. 2003, 125, 2408–2409. (15) Zhang, Y.; Tang, Z.; Wang, J.; Hong, W.; Maham, A.; Lin, Y. Anal. Chem. 2010, 82, 6440–6446. (16) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303–308. (17) Tsourkas, A.; Behlke, M. A.; Rose, S. D.; Bao, G. Nucleic Acids Res. 2003, 31, 1319–1330. (18) Bonnet, G.; Tyagi, S.; Libchaber, A.; Kramer, F. R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 6171–6176. (19) Mao, X.; Xu, H.; Zeng, Q.; Zeng, L.; Liu, G. Chem. Commun. 2009, 3065–3067. (20) Miranda-Castro, R.; De-los-Santos-Alvarez, N.; Lobo-Castanon, M. J.; Miranda-Ordieres, A. J.; Tunon-Blanco, P. Electroanalysis 2009, 21, 2077–2090. (21) Algar, W. R.; Massey, M.; Krull, U. J. Trends Anal. Chem. 2009, 28, 292–306. (22) Fan, C. H.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9134–9137. (23) Zhang, J.; Qi, H.; Li, Y.; Yang, J.; Gao, Q.; Zhang, C. Anal. Chem. 2008, 80, 2888–2894. (24) Bockisch, B.; Grunwald, T.; Edzard, S.; Bredehorst, R. Nucleic Acids Res. 2005, 33, e101. (25) Liu, S. Q.; Tang, Z. Y. J. Mater. Chem. 2010, 20, 24–35. (26) Coto-Garcia, A. M.; Sotelo-Gonzalez, E.; Fernandez-Arguelles, M.; Pereiro, R.; Costa-Fernandez, J. M.; Sanz-Medel, A. Anal. Bioanal. Chem. 2011, 399, 29–42. (27) Niemeyer, C. M. Angew. Chem., Int. Ed. 2001, 40, 4128–4158. (28) Merkoci, A.; Aldavert, M.; Marin, S.; Alegret, S. Trends Anal. Chem. 2005, 24, 341–349. (29) Xu, K.; Huang, J. R.; Ye, Z. Z.; Ying, Y. B.; Li, Y. B. Sensors 2009, 9, 5534–5557. (30) Zhang, H.; Harpster, M. H.; Park, H. J.; Johnson, P. A. Anal. Chem. 2011, 83, 254–260. (31) Fan, A. P.; Lau, C. W.; Lu, J. Z. Analyst 2009, 134, 497–503. (32) Verkleij, A. J.; Leunissen, J. L. M. Immuno-Gold-Labeling in Cell Biology; CRC Press: Boca Raton, FL, 1989. (33) Fan, A.; Cai, S.; Cao, Z. J.; Lau, C.; Lu, J. Analyst 2010, 135, 1400–1405. (34) Xiao, Y.; Lubin, A. A.; Baker, B. R.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 16677–16680. (35) Liu, G.; Wan, Y.; Gau, V.; Zhang, J.; Wang, L. H.; Song, S. P.; Fan, C. H. J. Am. Chem. Soc. 2008, 130, 6820–6825. (36) Wang, Y.; Li, C.; Li, X. H.; Li, Y. F.; Kraatz, H.-B. Anal. Chem. 2008, 80, 2255–2260. (37) Mao, X.; Xu, H.; Zeng, Q.; Zeng, L. W.; Liu, G. D. Chem. Commun. 2009, 3065–3067. (38) Zhang, J.; Qi, H. L.; Li, Y.; Yang, J.; Gao, Q.; Zhang, C. X. Anal. Chem. 2008, 80, 2888–2894.

ARTICLE

(39) Du, H.; Strohsahl, C. M.; Camera, J.; Miller, B. L.; Krauss, T. D. J. Am. Chem. Soc. 2005, 127, 7932–7940. (40) Piestert, O.; Barsch, H.; Buschmann, V.; Heinlein, T.; Knemeyer, J. -P.; Weston, K. D.; Sauer, M. Nano Lett. 2003, 3, 979–982.

7506

dx.doi.org/10.1021/ac201755x |Anal. Chem. 2011, 83, 7500–7506