Open Bridge-Structured Gold Nanoparticle Array for Label-Free DNA

Oct 7, 2008 - nm Au particles, the response was read by an electrical readout system. Even in a simple measuring method, we obtained a rapid response ...
0 downloads 0 Views 964KB Size
Anal. Chem. 2008, 80, 8071–8075

Open Bridge-Structured Gold Nanoparticle Array for Label-Free DNA Detection Shiho Tokonami, Hiroshi Shiigi,* and Tsutomu Nagaoka Frontier Science Innovation Center, Osaka Prefecture University, 1-2 Gakuen-cho, Sakai 599-8570, Osaka, Japan We focused on changes in the electrical property of the open bridge-structured gold nanoparticles array consisting of 46-nm parent and 12-nm son gold nanoparticles by hybridization and applied it for a simple electrical DNA detection. Since a target DNA of a 24-mer oligonucleotide was added to the probe DNA modified 12-nm Au nanoparticles, which was arranged on the gap between the 46nm Au particles, the response was read by an electrical readout system. Even in a simple measuring method, we obtained a rapid response to the cDNA with a high S/N ratio of 30 over a wide concentration range and a detection limit of 5.0 fmol. Moreover, the array discriminated 1-base mismatches, regardless of their location in the DNA sequence, which enabled us to detect single-nucleotide polymorphism, which is one of the important diagnoses, without any polymerase chain reaction amplification, sophisticated instrumentation, or fluorescent labeling through an easy-to-handle electrical readout system. The detection of DNA hybridization is important in the diagnosis of genetic diseases, DNA mapping, and forensic identification. Fluorescent-based DNA chips have been well-known as conventional large-scale gene analysis tools. However, these chips are currently expensive since their manufacturing requires sophisticated instruments such as hybridization equipments and fluorescent scanners, which has prevented their clinical use and has confined their utilization to research fields. In recent years, chemists and biologists have shown a considerable interest in the optical and electrical properties of metal nanoparticles and have carried out studies on their application to DNA detection by utilizing their unique properties. For the novel detection of DNA, some studies employing the distancedependent properties of aggregated nanoparticles were carried out. Wang et al. demonstrated the use of different inorganic nanocrystal tracers for a multitarget electrochemical detection of DNA.1 In addition, an electrical DNA detection system has been reported by Mirkin et al., in which Au particles were captured in between a micrometer-gapped electrode by the hybridization of DNA strands.2 Several recent studies have suggested that π overlapping between adjacent base pairs makes double-strand (ds) DNA conductive through electron-transfer mechanisms such as electron * To whom correspondence should be addressed. E-mail: [email protected]. Fax: 81-72-254-9875. (1) Wang, J.; Liu, G. D.; Merkoci, A. J. Am. Chem. Soc. 2003, 125, 3214. (2) Park, S. J.; Taton, T. A.; Mirkin, C. A. Science 2002, 295, 1503. 10.1021/ac801088u CCC: $40.75  2008 American Chemical Society Published on Web 10/07/2008

or hole hopping.3 However, none of these studies has directly measured the change in the electrical properties of DNA caused by hybridization and applied it to DNA detection. The precise control of the electrode gap for nanometer lengths of DNA is crucial for the detection of DNA by a simple method. Therefore, it is extremely important to develop a nanosized electrode device to enable us to recognize imperceptible substances. Such electrodes are usually manufactured using photolithography techniques. Although it is theoretically possible to fabricate electrodes with a gap of less than 30 nm, the resolution of the current lithographic techniques are limited to the order of 45 nm; hence, it is difficult to construct the required electrodes with good reproducibility by conventional fabrication techniques. To address this problem, many different approaches have been recently made to revamp illumination sources such as UV, DUV, and EUV; however, electrodes with gap sizes of the order of 1 nm could not be achieved. Moreover, the practical applicability of these methods is very limited due to the expensive photolithography machines required and the numerous steps involved. Hence, a process that aligns nanosized molecules based on the self-assembly phenomenon, utilizing the construction mechanism in molecules, has gained importance as one of the most promising techniques for the construction of such a nanometer-sized gap.4,5 For the past 12 years, many researchers have witnessed the most profound transition of emerging developments in nanotechnology and focused on the architectural implications of organic, inorganic, bionanoparticles and their combination. Ever-increasing reports on developments in the fabrication and characterization of the nanoarchitectures are being reported every year. Mirkin’s DNA detection system that uses DNA-capped gold nanoparticles (AuNPs) has increased the sensitivity of detection by several orders of magnitude since the light scattered from one AuNP is equivalent to the light emitted over 105 fluorophores. This is the one of the most representative developments that has been put (3) (a) Kasumov, A. Yu.; Kociak, M.; Gue´ron, S.; Reulet, B.; Volkov, V. T.; Klinov, D. V.; Bouchiat, H. Science 2001, 291–280. (b) Finkpr, H.-W.; Scho ¨nenberger, C. Nature 1999, 398, 407. (c) Porath, D.; Bezyadin, A.; de Vries, S.; Dekker, C. Nature 2000, 403, 635. (d) Arkin, M. R.; Stemp, E. D. A.; Holmin, R. E.; Barton, J. K.; Hormann, A.; Olson, E. J. C.; Barbara, P. F. Science 1996, 273, 475. (4) (a) Tokonami, S.; Iwamoto, M.; Hashiba, K.; Shiigi, H.; Nagaoka, T. Solid State Ionics 2006, 177, 2317. (b) Shiigi, H.; Yamamoto, Y.; Yakabe, H.; Tokonami, S.; Nagaoka, T. Chem. Commun. 2003, 1038. (5) Shiigi, H.; Yamamoto, Y.; Yoshi, N.; Nakao, H.; Nagaoka, T. Chem. Commun. 2006, 4288.

Analytical Chemistry, Vol. 80, No. 21, November 1, 2008

8071

Table 1. Response to Various Sequences sample full match 1-base mismatch 2-base mismatch 24-base mismatch full match 1-base mismatch 2-base mismatch 24-base mismatch a

12-mer probe A/12-mer probe B 5′-SH-poly(T)12-3′/5′-poly(T)12-HS-3′

5′-SH-TCT CAA CTC gTA-3′/5′-ATg CTC AAC TCT-HS-3′

24-mer target DNAa

∆R/Ωb

5′-AAA AAA AAA AAA AAA AAA AAA AAA-3′ 5′-AAA AAA AAA AAA TAA AAA AAA AAA-3′ 5′-AAA AAA TAA AAA AAA AAT AAA AAA-3′ 5′-TTT TTT TTT TTT TTT TTT TTT TTT-3′ 5′-AgA gTT gAg CAT TAC gAg TTg AgA-3′ 5′-AgA gTT gAg CAA TAC gAg TTg AgA-3′ 5′-AgA gTT CAg CAT TAC gAC TTg AgA-3′ 5′-TCT CAA CTC gTA gTA ATg CTC AAC TCT-3′

111 56 20 0 124 51 36 0

Target DNA of 100 µM/5 µL was applied on the probe-modified film. b The film resistance change caused by hybridization.

into practical use.6 DNA microarrays based on fluorescence readout systems have been recognized as powerful tools for the high-throughput analysis of thousands of genes and are aimed at being used in many biomedical and pharmaceutical applications such as expression profiling and simultaneous expression level monitoring.7 However, the current fluorophore-probe-based systems have been limited to research laboratories due to several drawbacks such as overlapped spectral features of the fluorophores, heterolytic photobleaching rates in multiplexed analyses, and low-cost efficiencies in both instrumentation and running cost. Several alternative approaches have been reported, which include nanoparticle aggregation,6,8 electrochemical detection,9 and conductivity detection of metallized DNA.2,10 In this regard, we have developed a novel label-free technique based on the change in the resistance of a AuNP array due to a change in an open bridge structured by hybridization. The resistance change by hybridization can be directly detected with a resolution that is sufficiently high for the detection of SNPs. EXPERIMENTAL SECTION Materials. All the chemicals used in this study were of reagent grade. Chloroauric acid (HAuCl4), trisodium citrate, 1,10-decanedithiol, tris(hydroxymethyl)aminomethane, and NaCl were purchased from Wako Pure Chemical Industries, Ltd. EDTA · 4Na were purchased from Dojindo Corp. Artificially synthesized oligonucleotides, including 5′-hexachlorofluorescein phosphoramidite (HEX)-capped DNA and 3′- and 5′-terminally thiolated DNA were obtained from Nissinbo Industries, Inc. and were purified by high-performance liquid chromatography. The used sequences are represented in Table 1. Milli-Q grade (>18 MΩ) water with ultraviolet sterilization was used throughout the experiment. Preparation of Gold Nanoparticles. AuNPs with an average diameter of 12 nm were prepared by the citrate reduction of HAuCl4. An aqueous solution of HAuCl4 (1 mM, 500 mL) was brought to a rolling boil with stirring. The rapid addition of 50 (6) (a) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhuff, J. J. Nature 1996, 382, 607. (b) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078. (c) Taton, T. A.; Mirkin, C. A.; Letsinger, R. L. Science 2000, 289, 1757. (7) (a) Lockhart, D. J.; Winzeler, E. A. Nature 2000, 405, 827. (b) Niemeyer, C. M.; Blohm, D. Angew. Chem., Int. Ed. 1999, 38, 2865. (8) Sato, K.; Hosokawa, K.; Maeda, M. Nucleic Acids Res. 2005, 33, e4. (9) (a) Takenaka, S.; Yamashita, K.; Takagi, M.; Uto, Y.; Kondo, H. Anal. Chem. 2000, 72, 1334. (b) Inouye, M.; Ikeda, R.; Takase, M.; Tsuri, T.; Chiba, J. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 11606. (c) Liu, T.; Barton, J. K. J. Am. Chem. Soc. 2005, 127, 10160. (10) (a) Fan, Y.; Chen, X.; Tung, C.; Kong, J.; Gao, Z. J. Am. Chem. Soc. 2007, 129, 5437. (b) Fan, Y.; Chen, X.; Kong, J.; Tung, C.; Gao, Z. Angew. Chem., Int. Ed. 2007, 46, 2051.

8072

Analytical Chemistry, Vol. 80, No. 21, November 1, 2008

mL of 38.8 mM sodium citrate to the vortex of the solution resulted in a color change from pale yellow to deep red. The boiling was continued for 15 min; then, the solution was removed from the heating mantle, and stirring was continued for an additional 15 min. The solution was then allowed to cool to room temperature, producing an Au dispersion of 3.7 × 10-2 g L-1. A resulting solution was characterized by an absorption maximum at 520 nm. Transmission electron microscopy (TEM) indicated an average particle size of 12 nm (σ ) 1.6 nm).6,11 AuNPs with a mean diameter of ∼50 nm were prepared by the conventional procedure:4,1210 mL of 3.0% citric acid as a reducer was added into 200 mL of 0.03% aqueous chloroauric acid, and the mixture was stirred at 353 K for 20 min to produce a Au dispersion of 0.14 g L-1. The derived Au particles were stored in a glass bottle maintained at 278 K. The resulting solution has an absorption maximum at 530 nm and an average particle size of 46 nm (σ ) 25 nm). Modification of Probe DNA on Au Nanoparticles. Modification of oligonucleotide to AuNPs: 3.61 µM 5′-terminally thiolated oligonucleotide (5′-SH-poly(T)12-3′ and 5′-HS-TCT CAA CTC gTA-3′) was incubated with 100 µL of a 12-nm AuNP solution at 323 K for 16 h. The solution was added to 10 mM phosphate buffer (pH 7.0) containing 0.1 M NaCl (100 µL); the resulting mixture was maintained at 323 K for 40 h, followed by centrifugation for 25 min at 14 000 rpm at ∼278 K in order to remove any unreacted oligonucleotide. After the removal of the supernatant, the AuNPs were washed with 12.5 mM phosphate buffer solution containing 125 µM NaCl (100 µL). After another centrifugation under the same conditions, the precipitate was dispersed into 8 mM phosphate buffer containing 280 µM NaCl (100 µL). The modification of the other probe DNA (5′-poly(T)12-SH-3′ and 5′TCT CAA CTC gTA-SH-3′) to AuNPs was carried out in the same manner. The 12-nm AuNP modified by the 5′-terminally thiolated DNA is denoted as probe A, while the AuNP modified by the 3′terminally thiolated DNA was denoted as probe B.6,11 The surface density of the DNA on the probe was fluorometrically evaluated to be 16 pmol cm-2 by using poly(T)12 derivatized with HEX.13-15 (11) (a) Grabar, K. C.; Freeman, R. G.; Hommer, M. B.; Natan, M. J. Anal. Chem. 1995, 67, 735. (b) Grabar, K. C.; Brown, K. R.; Keating, C. D.; Stranick, S. J.; Tang, S.-L.; Natan, M. J. Anal. Chem. 1997, 69, 471. (12) Turkevich, J.; Garton, G.; Sevenson, P. C. J. Colloid Sci. 1954, 9, 26. (13) Demers, L. M.; Mirkin, C. A.; Mucic, R. C.; Reynolds, R. A.; Letsinger, R. L.; Elghanian, R.; Viswanadham, G. Anal. Chem. 2000, 72, 5535. (14) (a) Georgiadis, R.; Peterlinz, K. P.; Peterson, A. W. J. Am. Chem. Soc. 2000, 122, 3166. (b) Peterson, A. W.; Wolf, L. K.; Georgiadis, R. J. Am. Chem. Soc. 2002, 124, 14601. (15) Tokonami, S.; Shiigi, H.; Nagaoka, T. Electroanalysis 2008, 20, 355.

Figure 1. (A) Illustration of the complete procedure of the preparation of open bridge-structured nanoparticle array on the microelectrode and (B) SEM image of DNA-capped 12-nm AuNP probes modified 46-nm parent AuNP.

Preparation of Au Nanoparticle Array. Probes A and B were anchored to a 46-nm parent particle, which was used as a scaffold for fixing the probes, as shown in Figure 1A. The array was prepared on an interdigitated microelectrode having one pair of 65 fingers of Pt electrodes with a 5-µm anode-cathode spacing formed on a quartz glass substrate (3.6 mm2, NTT-AT). The electrode (Figure 1Aa) was immersed into a 5 mM ethanolic 1,10decanedithiol solution for 30 min (b) and then submerged in the 46-nm Au dispersion for 30 min at room temperature (c).15–17 A gap between adjacent 46-nm AuNPs fixed on the microelectrode due to less immersion times in the thiol binder solution, and the dispersion was larger than that of the previous report (∼1.3 nm). To anchor the DNA-modified probe particles to the scaffold, a 1:1 colloidal mixture of probes A and B was applied onto the microelectrode after immersion into the ethanolic 1,10-decanedithi(16) Shiigi, H.; Tokonami, S.; Yakabe, H.; Nagaoka, T. J. Am. Chem. Soc. 2005, 127, 3280. (17) Tokonami, S.; Shiigi, H.; Nagaoka, T. J. Electrochem. Soc. 2008, 155, J105.

Figure 2. (A) Sensorgram for the addition of a 500-pmol 24-mer cDNA, poly(A)24. The arrow indicates the sample spike. Schematic illustration of the conducting path way formed between a pair of microelectrodes (a) before and (b) after hybridization. (B) Dependence of the sensor response on the target DNA concentration. Target sequence, 3′-poly(A)24-5′; probe A, 5′-SH-poly(T)12-3′; and probe B, 5′-poly(T)12-HS-3′. The array resistance was ∼320 Ω.

ol solution (d). The prepared array was allowed to stand for 2 h in a cage regulated at 298 K and was then washed with distilled water before use (e). We have an emphasis on the given gap between adjacent AuNPs has freedom degrees of probes as shown in Figure 1A, and it was different from that the locked nanospace by thiol binder on the previous nanoparticles array.15–17 The surface of the resulting AuNP array was characterized using a FE-SEM (S-4700, Hitachi) at an applied of voltage 10 kV. The SEM image of the array for several widths revealed that most of the probe particles bind to the hemisphere of the parent surface as shown in Figure 1B (average, 16 probes/parent). Procedures of DNA Detection. A 2-µL TE buffer solution (pH 7.4; ionic strength, 1.0) was spread over the sensor array to Analytical Chemistry, Vol. 80, No. 21, November 1, 2008

8073

measure the base resistance. After reaching a steady state, a 5-µL aliquot of TE buffer containing the target single-stranded DNA was dropped on the array. The array resistance was measured by the two-probe technique using a digital multimeter (34970A, Agilent) operating at a constant-current mode of 1 mA. All the measurements were made in the cage regulated at 298 K.15–17 To investigate the dependence of the target DNA concentration on the sensor response, the sensor electrode was immersed into water at 353 K for 3 min to remove the target DNA by the denaturation of dsDNA for further determination. Every measurement for different concentrations was carried out in the same manner described above. All the experiments were repeated at least 3 times, and the resistance values reported for the DNA sensing had an accuracy of ±1 mΩ. RESULTS AND DISCUSSION Figure 2A shows the time course of the array resistance upon applying the target 24-mer DNA, which carries two complementary sections for probes A and B (in this case, the target is poly(A)24, and both the probes are poly(T)12). As a precondition, a 2-µL TE buffer solution was spread over the array. After reaching steady state, a 5-µL aliquot of the target DNA (500 pmol) dissolved in the TE buffer was applied on the array. The resistance immediately decreased and became constant in 60 s. The ∆R value, defined as the difference in the resistance before and after the hybridization, was 100 mΩ with an S/N ratio of >30. The sensor showed response over a wide concentration range with a detection limit of 5.0 fmol as shown in Figure 2B. Electrical measurements were carried out with the same and different arrays, having the same electrical resistance (∼320 Ω), and repeated at least 3 times, respectively. According to a previous report, AuNPs separated by a 1.3-nm gap could be used for recognizing the hybridization event; however, the concentration range was limited (detection limit of ∼25 pmol) because not every each probe could hybridize with a target DNA strand in the gap between the particles.15–17 Thus, they were not completely responsible for the formation of the conducting path on the surfaces of respective AuNPs. On the contrary, opening bridge hybridization occurred certainly between the adjacent particles; therefore, the conductivity change can be explained not only by the molecular conductivity of DNA3,15–18 but also by the shrinkage of the structural distance of the 12-nm probes in the gap of the 46-nm parent particles due to the hybridization (Figure 3). Since binder decanedithiol (∼1.3 nm) was used for the modification of son AuNPs (12 nm) to the parent AuNPs (46 nm), the son AuNPs, which were arrowed in the inset of Figure 3a, were anchored on the parent, and at the same time, flexible. When one side of target DNA (24-mer) is hybridized with the probe DNA (12-mer), for example, probe A, the hybridization of the other half of target DNA with the other probe DNA (probe B) leads us to decrease of electrical resistance (Figure 3b). The distance between the arrowed sons, shown in the inset of Figure 3b, was decreased compared to that before the hybridization. However, a larger distance among the son AuNPs, which is impossible to hybridize the other half of target (18) (a) Hihath, J.; Xu, B.; Zhang, P.; Tao, N. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 16979. (b) Xu, B.; Zhang, P.; Li, X.; Tao, N. Nano Lett. 2004, 4, 1105. (c) Wierzbinski, E.; Arndt, J.; Hammond, W.; Slowinski, K. Langmuir 2006, 22, 2426.

8074

Analytical Chemistry, Vol. 80, No. 21, November 1, 2008

Figure 3. Conceptual illustrations of the array consisting of the DNAcapped 12-nm AuNP probes, A and B, immobilized on the parent 46-nm AuNP with dithiol: Probe configuration (a) before and (b) after the addition of 24-mer target. The inset is a TEM image of the AuNP array revealing around 46-nm particles.

DNA and the other probe DNA, cannot form the conducting pathway at the gap. These suggest the efficient formation of the conductive path, causing the detection limit to increase by three powers of ten over that of the previous single-AuNP array system.15,16 Moreover, the response to target DNA is deeply committed to the percolation theory. The AuNPs film, consisting of parent and son AuNPs, had an electrical resistance of ∼320 Ω, which was much lower than the percolation threshold level, which led us a high reproducibility (see the inset of Figure 2A).17 To verify the effectiveness of this system for the identification of DNA mismatches, we carried out experiments using the targeted DNA along with complementary, 1-bp, 2-bp, and 24-bp mismatched (fully mismatched) DNA sequences (Table 1). The results for the poly(T)-poly(A) hybridization, represented in Figure 4 (square), indicate that the response was the highest for the cDNA and decreased with an increase in the number of mismatched bases. Finally, ∆R hardly changed at the fully mis-

±5 mΩ, same sample run (n) ) 3 times) in a given series of experiments, and the S/N remained constant. The response was nonlinear with respect to the number of mismatches, and a clear difference (over 50 mΩ) found between the complementary and the 1-bp mismatch suggests that this system can be efficiently used for the SNPs diagnosis. A similar trend for the mismatch detection for randomly sequenced fragments (closed circle) suggests that this method can be used for practical DNA identification.

Figure 4. Sensor response vs the number of mismatched base pairs in the target DNA. Square and circles represent the results obtained using probe A, 5′-SH-poly(T)12-3′; probe B, 5′-poly(T)12-HS-3′ and probe A, 5′-SH-TCT CAA CTC gTA-3′; probe B, 5′-ATg CTC AAC TCT-HS-3′, respectively. The measurement was performed in series from the complementary to mismatch (closed square) and from the mismatch to complementary (open square). The array resistance was ∼320 Ω.

matched sequence (∆R < 10 mΩ). Furthermore, the response to each target sequence was almost constant, even if the series of measurements were performed from the complementary or fully mismatched sequence. In the series of experiments repeated with hybridizing and denaturing, the measured response to the complementary strand was stable (104 and 111 mΩ for the first and final runs, respectively), and no morphological change was observed in the SEM images of the array surfaces after such a treatment. The accuracy of the resistance was ±4.2% (less than

CONCLUSIONS We fabricated an open bridge-structured electrode containing a Au nanoparticle-DNA-Au nanoparticle repeated sequence, in which the gap between the particles is precisely controlled using the probe sequence. By combining this electrode with the conductive DNA, we could measure very small changes in the electrical property of DNA with a high S/N (>30) and a detection limit of 5.0 fmol; a remarkable difference was observed between the 1-bp mismatched DNA and the cDNA. We can amplify the detection limit of the 1-base mismatch in the DNA sequence using the above-mentioned system. It can be emphasized that the Au particle film studied here is characterized as a highly selective device for the detection of DNA polymorphism, which is one of the most important rapid and sensitive diagnostic techniques, without involving any fluorescent labeling or sophisticated instrumentation. ACKNOWLEDGMENT This study was supported by the Industrial Technology Research Grant Program in 2005 from the NEDO of Japan. We gratefully acknowledge the financial support obtained from the Japan Society for Grant-in-Aid for Young Scientists (A) (18680038).

Received for review May 29, 2008. Accepted September 5, 2008. AC801088U

Analytical Chemistry, Vol. 80, No. 21, November 1, 2008

8075