ARTICLE pubs.acs.org/ac
Label-Free Detection of DNA-Binding Proteins Based on Microfluidic Solid-State Molecular Beacon Sensor Jun Wang,*,† Daisuke Onoshima,†,‡ Michihiko Aki,§ Yukihiro Okamoto,†,‡ Noritada Kaji,†,‡ Manabu Tokeshi,†,‡ and Yoshinobu Baba*,†,‡,^ †
Department of Applied Chemistry, Graduate School of Engineering, and ‡FIRST Research Center for Innovative Nanobiodevice, Nagoya University, Nagoya, Japan § Fujitsu Laboratories Ltd., Tokyo, Japan ^ Health Research Center, National Institute of Advanced Industrial Science and Technology (AIST), Takamatsu, Japan ABSTRACT: A solid-state molecular beacon using a gold support as a fluorescence quencher is combined with a polydimethylsiloxane (PDMS) microfluidic channel to construct an optical sensor for detecting singlestranded DNA binding protein (SSBP) and histone protein. The singlestranded DNA-Cy3 probe or double-stranded DNA-Cy3 probe immobilized on the gold surface is prepared for the detection of SSBP or histone, respectively. Due to the different quenching ability of gold to the immobilized single-stranded DNA-Cy3 probe and the immobilized double-stranded DNA-Cy3 probe, low fluorescence intensity of the attached single-stranded DNA-Cy3 is obtained in SSBP detection, whereas high fluorescence intensity of the attached double-stranded DNA-Cy3 is obtained in histone detection. The amounts of SSBP in sample solutions are determined from the degree of fluorescence recovery of the immobilized single-stranded DNA-Cy3 probe, whereas that of histone in sample solutions is determined from the degree of fluorescence quenching of the immobilized double-stranded DNA-Cy3 probe. Using this approach, label-free detection of target proteins at nanomolar concentrations is achieved in a convenient, general, continuous flow format. Our approach has high potential for the highly sensitive label-free detection of various proteins based on bindinginduced conformation changes of immobilized DNA probes.
I
n the last two decades, huge effort has been made in academic and industrial laboratories to develop biosensors for label-free detection of analytes. Biosensor technologies hold great promise for the healthcare market, food industry, and environmental monitoring. Biosensors have already been commercialized for some special applications, such as blood glucose sensing. A variety of optical, acoustic, and electrochemical approaches for biosensor applications have been reported.18 Among them, optical sensors based on molecular beacons have attracted significant interest because molecular beacons are considered one of the most promising technologies.912 Molecular beacons are single-stranded oligonucleotides with a hairpin structure in which the 50 - and 30 - ends are self-complementary, bringing a fluorophore and a quencher into close proximity. Fluorescence is restored when the molecular beacon binds to a complementary nucleic acid target. A solution-based molecular beacon technique has been widely used in diagnosing genetic diseases because it showed nice reproducibility as well as a good linear relationship between the target concentration and fluorescence intensity. Recently, a surface-based molecular beacon technique has been developed for cDNA target detection. Several research groups have used a gold substrate, instead of organic dye molecules, as a more efficient quencher of the fluorescence from the donor dyes combined with a surface-immobilized hairpin stem-loop singlestranded DNA (ssDNA).1315 This surface-based molecular r 2011 American Chemical Society
beacon technique has great potential to develop new types of sensors and microarrays because it does not necessitate washing steps. This kind of surface-based molecular beacon using a gold support as fluorescence quencher is potentially suitable for microchip-based optical detection. The combination of microfluidic technology and the solid-state molecular beacon technique may have the capability to perform in situ, continuous, real-time monitoring of specific analytes in aqueous samples, and that is important for a broad range of applications from medical diagnostics to environmental monitoring. However, to the best of our knowledge, protein detection based on changes in fluorescence intensity of the attached ssDNA or double-stranded DNA (dsDNA) in a microfluidic solid state sensor has not been reported. In this work, we report the development of a microfluidic, solid-state molecular beacon sensor to achieve continuous, realtime, and label-free monitoring of histone protein and singlestranded binding protein (SSBP). Such proteins are abundant and essential in cells: they interact with DNA to organize its packing, to regulate transcription, and to carry out replication and repair. One clinically important subset of these naturally occurring proteins is the transcription factors that have been Received: January 27, 2011 Accepted: March 29, 2011 Published: April 05, 2011 3528
dx.doi.org/10.1021/ac200236r | Anal. Chem. 2011, 83, 3528–3532
Analytical Chemistry
ARTICLE
proven to be useful for diagnostic applications.16,17 The interaction of DNA-binding proteins with our ssDNA or dsDNA immobilized on the gold support can cause changes in fluorescence intensity, and the intensity changes can be used for the quantification of this kind of protein. Here, we also proved SSBP detection based on fluorescence changes using an immobilized ssDNA probe with no hairpin structure. Unlike the traditional molecular beacon probe requiring hairpin structure design, our microfluidic solid state molecular beacon sensor does not require hair pin design and offers highly sensitive detection of SSBP.
’ EXPERIMENTAL SECTION Reagents and Materials. We have employed the following probe DNA sequences. ssDNA probe: 50 -HS(CH2)6GCC GGC CAC AGC CAA TCA GCA GCG CGG ACC CCT CCC CAG GGC GGA GCT GAC GGC C-Cy3-30 . dsDNA probe is prepared by hybridization reaction using this ssDNA probe and its cDNA (cDNA). The ssDNA and cDNA were obtained from Tsukuba Oligo Service Corporation (Tsukuba, Japan). 6Mercaptohexanol and histone were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO) and used without further purification. Escherichia coli SSBP was obtained from Promega Corp. (Madison, WI). All water used in the preparation of buffers and rinse had a resistivity of 18.2 MΩ cm and was produced by a Milli-Q purification system (Millipore, Billerica, MA). Gold substrate was provided by Fujitsu Laboratories Ltd. (Kanagawa, Japan). Gold lines of 200 μm width were prepared on 3-in. single crystalline sapphire wafers by subsequentially depositing thin layers of Ti (10 nm), Pt (40 nm), and Au (200 nm) films using standard optical lithography and metallization techniques. Fabrication of a Microfluidic Chip. A silicon wafer with positive surface reliefs was prepared by a conventional photolithography method and used as a mold for the preparation of a polydimethylsiloxane (PDMS) microfluidic channel. The straight PDMS microfluidic channel (width, 200 μm; height, 5 μm) was prepared by a PDMS kit; PDMS base material and curing agent were mixed in 10:1 ratio, stirred vigorously for 15 min, and then degassed for 1 h under dynamic vacuum to remove all air bubbles. The clear solution was poured onto the positive mold, and then the mold was heated at 75 C for 1 h. After cooling, the PDMS microfluidic chip was pealed off from the mold. Two holes (2 mm diameter) were punched directly above the end of the PDMS microfluidic channel for sample transfer by a syringe pump. Immobilization of DNA Probe on Au Surface. Before DNA adsorption, the Au surface was cleaned by piranha solution (H2SO4/H2O2 = 7:3) and exposed to HNO3 (60%) for 15 min, followed by a final rinse with deionized water. Preparation of the ssDNA-Cy3 layer on the Au surface was done by the adsorption of ssDNA-Cy3 from aqueous solution (1 μM ssDNA-Cy3 in 10 mM Tris buffer containing 50 mM NaCl, pH 7.4) in a small PDMS reservoir (50 μL) on the gold surface. The ssDNA probe was immobilized on the gold surface by the SAu bond. After washing with distilled water, the incubation of mercaptohexanol molecules (1 mM mercaptohexanol in 10 mM Tris buffer containing 50 mM NaCl; pH 7.4) on the gold surface was executed for 1 h to improve the ssDNA layer structure and passivate the remaining exposed Au surface.18 Preparation of the dsDNA-Cy3 layer proceeded in two steps. First, a relatively dense layer of ssDNA-Cy3 was formed on the Au surface in the same way as described in the ssDNA layer preparation. cDNA solution (1 μM cDNA in10 mM Tris buffer
Figure 1. Schematic diagrams of the target detection principle and fluorescence images before and after introduction of targets in this solid state molecular beacon sensor: (a, e) cDNA target, (b, f) SSBP target, and (c, g) histone target. The binding of the targets to the surfaceimmobilized ssDNA-Cy3 or dsDNA-Cy3 can cause the distance changes between gold surface and Cy3 dye, which influences fluorescence energy transfer efficiency between the gold and Cy3.
containing 200 mM NaCl; pH 7.4) was introduced for 1 h to form the dsDNA layer structure on the Au surface. Determination by Microscopy. After the immobilization of ssDNA-Cy3 probe on the Au surface, the PDMS microfluidic channel was put on the Au surface to construct the sensor, and a cDNA or protein flow was introduced by a syringe pump to react with the ssDNA-Cy3 probe on the gold surface. All cDNA or protein solution was prepared in 10 mM Tris buffer (containing 200 mM NaCl, pH7.4). Fluorescence images of ssDNA-Cy3 probe immobilized on the gold in the microfluidic channel were obtained with a Nikon confocal microscope equipped with a silicon intensified target (SIT) (excitation wavelength = 532 nm, NA = 0.45, 20 objective lens). Fluorescence images were collected through the same objective; passed through a dichroic beam splitter and a band-pass filter (575 nm); and finally, collected by SIT. Under laser illumination, the protein was introduced by a syringe pump at a desired flow rate, and the fluorescence images were recorded with time. The fluorescence recovery and quenching data were obtained from the SIT images by the Adobe premiere software and Cosmos software.
’ RESULTS AND DISSCUSSION Principle of the Microfluidic Sensor for SSBP or Histone Detection. Detection of SSBP and histone is based on the low
fluorescence intensity of ssDNA-Cy3 and high fluorescence intensity of dsDNA-Cy3 immobilized on respective Au surfaces. The difference in fluorescence intensity between the immobilized ssDNA-Cy3 and dsDNA-Cy3 is proved by the hybridization reaction between cDNA and the ssDNA-Cy3 immobilized on the Au surface.19 A schematic diagram describing this fluorescence variation is shown in Figure 1a. By immobilization of ssDNA-Cy3 probe on the Au surface, the Cy3 at the 30 -end of the ssDNA molecule may be brought into close proximity to the Au surface because of the flexibility of the ssDNA molecule, and its fluorescence will be quenched by the Au though nonradiative energy transfer.2023 Upon hybridization to the cDNA target, the ssDNA-Cy3 will have high fluorescence intensity due to the rigid rodlike structure of dsDNA. The resulting dsDNA-Cy3 may stand up perpendicular to the gold surface, rendering Cy3 separated from the Au and, hence, minimizing the resonance energy transfer between Cy3 and Au. For this reason, this kind of ssDNA-Cy3 without the hairpin structure may also be as effective as the previously reported optical sensors using 3529
dx.doi.org/10.1021/ac200236r |Anal. Chem. 2011, 83, 3528–3532
Analytical Chemistry
ARTICLE
Figure 2. (a) Curve of fluorescence intensity with time from the optical, microfluidic sensor after the introduction of SSBP. Low fluorescence intensity that is obtained by the immobilized ssDNA-Cy3 on the gold surface appears (0 min) before introducing SSBP, fluorescence intensity increases with time, and different fluorescence intensity (60 min) results from different concentrations of SSBP. (b) Doseresponse curve of SSBP whose concentrations decreased consecutively from the highest concentration of 5000 ng/mL to the lowest concentration of 10 ng/mL. The curve is hyperbolic, with a minimum detectable concentration of 20 ng/mL, which is comparable to or even better than previously reported values. (c) Fluorescence recovery effect upon the introduction of different proteins for this SSBP sensor: no significant signal is observed in this SSBP sensor, even in the presence of high concentrations of BSA or histone.
ssDNA probe with the complicated hairpin stemloop structure design. The background fluorescence intensities used for calculating the amounts of SSBP and histone are obtained using only the immobilized ssDNA-Cy3 and the dsDNA-Cy3, respectively. After binding with the target protein, the changes in the fluorescence signal indicate the presence of the target. Figure 1b and c show a fluorescence-intensity-change-based detection principle for SSBP and histone. Upon SSBP binding, fluorescence is recovered because of the binding interaction between SSBP and the immobilized ssDNA-Cy3; the increased distance between Cy3 and the Au surface caused the blocking of fluorescence resonance energy transfer between Cy3 and Au. Upon histone binding, compression of the dsDNA occurs, and that enhances the nonradiative energy transfer between Cy3 and the Au surface. Thus, fluorescence of dsDNA-Cy3 is quenched upon binding with histone. This effect increases with decreasing distance between Cy3 and the Au surface. The changes in the fluorescence intensity signal by the introduction of cDNA, SSBP, and histone can be observed from microscopy images. Figure 1d, e, and f shows the fluorescence images obtained before and after the target binding. These fluorescent images of the DNA probes on the Au surface in the microfluidic channel change with the introduction of different target samples: fluorescence recovery is observed upon the introduction of cDNA or SSBP, which binds to the ssDNA-Cy3 probe. After the introduction of cDNA or SSBP to the ssDNA-Cy3 immobilized on the Au, fluorescence images change from black to bright white, and that is an indication of fluorescence recovery (Figures 1d and 1e). Fluorescence quenching is observed upon the introduction of histone, which binds to the dsDNA-Cy3 probe. When introducing histone, fluorescence image changes from bright white to gray (fluorescence quenching) (Figure 1f). The changes in fluorescence intensity before and after the introduction of three kinds of targets are clear, and fluorescence intensity in the images can be calculated by using the commercial software. Detection of SSBP by the Microfluidic Sensor. Microfluidic sensors with high sensitivity are useful analytical tools for detection of biologically important substances. A microfluidic solid state sensor is ideally suited for the continuous monitoring of a flow of
analytes. As a model for protein detection, we applied this sensor to detect the SSBP flow based on the binding interaction with the immobilized ssDNA. It is known that SSBP binds with ssDNA in multiple modes, and binding modes may change with the concentration of salt.16 The binding of immobilized ssDNA with SSBP in a high salt concentration (200 mM) is used in this study to confirm whether the binding can cause fluorescence recovery due to separation of Cy3 from the Au surface. The curve of the fluorescence signal recovery with time was obtained upon the introduction of SSBP flow into this microfluidic sensor. Figure 2a shows typical curves of fluorescence intensity change with the time (060 min) upon introduction of SSBP at different concentrations. Compared with the background fluorescence intensity, the fluorescence intensity obtained after the introduction of SSBP flow is much higher because of conformation change-induced fluorescence recovery. A constant difference in fluorescence intensity is reached at 60 min for all the different SSBP concentrations, and thus, the fluorescence intensity curve can be used for quantitative determination of SSBP concentration. We use the ratio of Fpost-SSBP (fluorescence intensity at 60 min after the introduction of SSBP) to Fpre-SSBP (fluorescence intensity before the introduction of SSBP) to obtain the doseresponse curve of SSBP. Figure 2b shows the doseresponse curve of SSBP, which is hyperbolic, as expected, for single-site saturable binding. From the curve, a minimum detectable concentration of 20 ng/mL was determined; this value is comparable to or even better than previously reported values.24 Analysis of the figure indicates that we can obtain nearly 7-fold fluorescence intensity enhancement by the introduction of SSBP protein. The signal enhancement can be influenced by the immobilization density of ssDNA probe on the Au surface. The desired probe density is optimized by controlling the exposure time and concentration of probe molecules in the solution. The probe density on the Au surface increased with increasing incubation concentration of ssDNA probe. Lower concentration of ssDNACy3 probe molecules gives the lower surface density and poorer fluorescence recovery after binding reaction. However, incubation of a higher concentration of ssDNA probe can cause saturated surface density, which does not lead to a better chip 3530
dx.doi.org/10.1021/ac200236r |Anal. Chem. 2011, 83, 3528–3532
Analytical Chemistry
ARTICLE
Figure 3. (a) Curves of time-dependent fluorescence intensity change obtained after the introduction of different concentrations of histone to the dsDNA-Cy3 probe immobilized on the Au surface. The immobilized dsDNA-Cy3 caused high fluorescence intensity because of the rigidity of the dsDNA. The time-dependent fluorescence intensity decrease after the introduction of histone can be recorded by microscopy and calculated with software. (b) Doseresponse curve of histone whose concentrations decreased consecutively from the highest concentration of 170 μg/mL to the lowest concentration of 0.017 μg/mL. The curve is a bilinear shape with a dissociation constant of 13 ( 0.5 μg/mL, comparable to previous experimental result. (c) Fluorescence quenching effect by different proteins in histone sensor. No significant signal quenching is observed in this SSBP sensor, even in the presence of a high concentration of BSA, SSBP, or lysozyme.
performance (data not shown). Here, we obtained the optimized surface density by using 1 μM ssDNA-Cy3 according to the following procedure. The optimized surface density of ssDNA probe on the gold surface is prepared by the following protocol. First, the incubation of ssDNA solution (1 μM ssDNA-Cy3 in 10 mM Tris buffer containing 50 mM NaCl, pH 7.4) on the gold surface for 1 h was used to obtain the surface coverage of the ssDNA probe layer. After washing with distilled water, 1 mM mercaptohexanol (prepared in 10 mM Tris buffer containing 50 mM NaCl, pH 7.4) was added onto the gold surface for 1 h to provide spacer molecules between the adsorbed ssDNA probe molecules. Here, mercaptohexanol was used as the blocking molecule, since its short carbon chain does not interfere with the hybridization reactions of the surface-bound ssDNA end. Use of the mixed monolayer to control surface coverage was critical. We think the presence of mercaptohexanol probably reduces number of nonspecific adsorption of ssDNA probe on the Au surface.18,19 Thus, the Au surface with immobilized ssDNA-Cy3 and passivated by mercaptohexanol is accessible for specific binding with SSBP and is able to discriminate target molecules, as reported by others.25 The sensor based on the immobilized ssDNA is specific for the SSBP. We checked the specificity of this SSBP sensor by introducing other proteins to the gold surface. We could not detect any significant fluorescence enhancement upon introduction of other proteins, such as bovine serum albumin (BSA) and histone (Figure 2c), although the ability for this sensor to discriminate cDNA and SSBP is limited. The signal enhancement is based on the binding-induced conformation change of the immobilized ssDNA-Cy3 probe (and not simply adsorption of target to the Au surface), and that is similar to the protein binding-induced conformation change of the aptamer. Therefore, it can be potentially employed in aptamer-based protein detection. Detection of Histone by the Microfluidic Sensor. We fabricated a sensor for histone using the attached dsDNA-Cy3 probe on the Au surface in the PDMS microfluidic channel. Histone is a strongly alkaline protein that can wind and condense the long dsDNA chromatin into the chromosomes. This process allows a very large amount of DNA to be fitted into the small nucleus of a cell. The condensation of attached dsDNA on the Au
surface by histone causes the fluorescence quenching effect with the decreased distance of Cy3 and the gold surface. Figure 3a shows a typical curve of fluorescence intensity versus time for the introduction of various concentrations of histone. The fluorescence decrease occurs by introducing histone samples at various concentrations into the sensor. Compared with the blank fluorescence intensity of dsDNA on the gold surface in Figure 3a, the limit of detection, calculated by summing the background fluorescence intensity and three times the standard deviation, was found to be 0.17 μg/mL. This value is comparable to other molecular beacon methods for protein detection.26 Fluorescence quenching in the histone binding process is obtained by the introduction of histone, and the quenching effect differs with the histone concentrations. This can be used for the quantitative determination of histone. We used the ratio of Fposthistone (fluorescence intensity at 20 min after the introduction of histone) to Fprehistone (fluorescence intensity before the introduction of histone) as the fluorescence quenching percent to obtain the doseresponse curve of histone. Figure 3b shows the doseresponse curve of histone, which exhibits a bilinear shape, presumably arising because of the different binding modes of the dsDNA-Cy3, depending on histone concentration. This graph, Figure 3b, also shows that fluorescence quenching percent increases to the maximum and then decreases with the increase in the histone concentration. At a low concentration of histone (less than 50 μg/mL), the number of dsDNA molecules is larger than the number of histone molecules, and that can cause the binding of histone molecules to ds-DNA-Cy3 molecules in a 1:1 ratio. Therefore, the proximity of Au and Cy3 is achieved, and that caused the stronger fluorescence quenching. Introducing a high concentration of histone (170 μg/mL) decreased the fluorescence quenching percent, probably due to the binding of multiple histone molecules to one dsDNA-Cy3 molecule. Because histone acts as a spool around which dsDNA-Cy3 winds, compared with single histone binding to one dsDNA-Cy3 molecule, multiple histone binding to one ds-DNA-Cy3 molecule increases the distance between the Cy3 and Au, and that minimizes the energy transfer between the Cy3 and Au. The dissociation constant between histone and the immobilized dsDNA is about 13 ( 0.5 μg/mL 3531
dx.doi.org/10.1021/ac200236r |Anal. Chem. 2011, 83, 3528–3532
Analytical Chemistry (565 ( 20 nM), which is larger than the previously reported value of 168 ( 142 nM for histonelong-chain dsDNA (146 bp).27 The difference is probably due to the different experimental designs or the employment of different lengths of dsDNA. It is well-known that the binding strength of histone to dsDNA decreases with decreasing dsDNA length.28 Our result about the dissociation constant obtained using short-chain dsDNA (55 bp) agrees well with this trend. The sensor based on the immobilized dsDNA-Cy3 probe is specific to histone. For example, we could not detect any significant fluorescence quenching when introducing BSA or SSBP protein (Figure 3c). The fluorescence quenching is based on the binding-induced conformation change of the immobilized dsDNA-Cy3 probe (not on simple adsorption of target to the Au surface), and that is similar to the protein binding-induced conformation change of the aptamer. Therefore, our sensor can be potentially employed in conformation change-based detection.
’ CONCLUSIONS We have demonstrated an optical, microfluidic solid state molecular beacon sensor for the label-free detection of DNA binding proteins. We verified the detection of protein targets on the basis of fluorescence quenching and recovery, which were caused by the binding interaction between the target and the ssDNA-Cy3 or dsDNA-Cy3 probe immobilized on the gold surface. The quantity of SSBP was determined on the basis of strong fluorescence recovery, whereas that of histone was determined on the basis of strong fluorescence quenching. This approach, based on the fluorescence intensity difference between the ssDNA and ds-DNA immobilized on the gold surface, allows continuous detection of analytes in a flow with high sensitivity. The approach permits achievement of a label-free detection of target proteins in a nanomolar concentration level in a convenient, general, and continuous flow format. The strategy used for this sensor may be generalized for the detection of other biological molecules and especially serve as a technical platform for DNA or protein detection.
ARTICLE
(7) Patolsky, F.; Lichtenstein, A.; Willner, I. J. Am. Chem. Soc. 2001, 123, 5194–5205. (8) Boon, E. M.; Ceres, D. M.; Drummond, T. G.; Hill, M. G.; Barton, J. K. Nat. Biotechnol. 2000, 18, 1096–1100. (9) Broude, N. E. Trends Biotechnol. 2002, 20, 249–256. (10) Dai, X.; Greizerstein, M. B.; Nadas-Chinni, K.; Rothman-Denes, L. B. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 2174–2179. (11) Heyduk, T.; Heyduk, E. Nat. Biotechnol. 2002, 20, 171–176. (12) Tyagi, S.; Kramer, F. R. Nat. Biotechnol. 1996, 14, 303–308. (13) Fang, X.; Liu, X.; Schuster, S.; Tan, W. J. Am. Chem. Soc. 1999, 121, 2921–2922. (14) Du, H.; Disney, M. D.; Miller, B. L.; Krauss, T. D. J. Am. Chem. Soc. 2003, 125, 4012–4013. (15) Fan, C.; Plaxco, K. W.; Heeger, A. J. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 9134–9137. (16) Roy, R.; Kozlov, A. G.; Lohman, T. M.; Ha, T. J. Mol. Biol. 2007, 369, 1244–1257. (17) Kouzarides, T. Cell 2007, 128 (4), 693–705. (18) Rant, U.; Arinaga, K.; Tornow, M.; Kim, Y.; Netz, R.; Fujita, S.; Yokoyama, N.; Abstreiter, G. Biophys. J. 2006, 90 (10), 3666–3671. (19) Rant, U.; Arinaga, K.; Scherer, S.; Pringsheim, E.; Fujita, S.; Yokoyama, N.; Tornow, M.; Abstreiter, G. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 17364–17369. (20) Prigogine, I.; Rice, S. Adv. Chem. Phys. 1987, 37, 1. (21) Stoermer, R. L.; Keating, C. D. J. Am. Chem. Soc. 2006, 128, 13243–13254. (22) Neumann, T.; Johansson, M.-L.; Kambhampati, D.; Knoll, W. Adv. Funct. Mater. 2002, 12, 575–586. (23) Maxwell, D. J.; Taylor, J. R.; Nie, S. J. Am. Chem. Soc. 2002, 124, 9606–9612. (24) Ricci, F.; Bonham, A. J.; Mason, A. C.; Reich, N. O.; Plaxco, K. W. Anal. Chem. 2009, 81, 1608–1614. (25) Herne, T. M.; Tarlov, M. J. J. Am. Chem. Soc. 1997, 119, 8916–8920. (26) Hamaguchi, N.; Ellington, A.; Stanton, M. Anal. Biochem. 2001, 294, 126–131. (27) Mazurkiewicz, J.; Kepert, J. F.; Rippe, K. J. Biol. Chem. 2006, 281, 16462–16472. (28) Aviles, F. J.; Diez-Caballero, T.; Palau, J.; Albert, A. Biochimie 1978, 60, 445–451.
’ AUTHOR INFORMATION Corresponding Author
*Fax: 081-52-789-5499. (J.W.); 081-52-789-4666 (Y.B.). E-mail:
[email protected] (J.W.);
[email protected] (Y.B.).
’ ACKNOWLEDGMENT We thank Dr. Kenji Arinaga and Dr. Shozo Fujita for valuable discussions. ’ REFERENCES (1) Cao, Y. W. C.; Jin, R. C.; Mirkin, C. A. Science 2002, 297, 1536–1540. (2) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 10954–10957. (3) Peterson, A. W.; Wolf, L. K.; Georgiadis, R. M. J. Am. Chem. Soc. 2002, 124, 14601–14607. (4) Kelley, S. O.; Barton, J. K.; Jackson, N. M.; Hill, M. G. Bioconjugate Chem. 1997, 8, 31–37. (5) Cooper, M. A.; Dultsev, F. N.; Minson, T.; Ostanin, V. P.; Abell, C.; Klenerman, D. Nat. Biotechnol. 2001, 19, 833–837. (6) Endo, T.; Kerman, K.; Nagatani, N.; Takamura, Y.; Tamiya, E. Anal. Chem. 2005, 77 (21), 6976–6984. 3532
dx.doi.org/10.1021/ac200236r |Anal. Chem. 2011, 83, 3528–3532
