Direct Detection of DNA with an Electrocatalytic Threading Intercalator

Tania García-Mendiola , Tegra Barreiro Martínez , Félix Pariente , Jesús Molano , Encarnación Lorenzo. Electroanalysis 2014 26 (10.1002/elan.v26...
1 downloads 0 Views 247KB Size
Download PDF
Anal. Chem. 2005, 77, 126-134

Direct Detection of DNA with an Electrocatalytic Threading Intercalator Natalia C. Tansil, Hong Xie, Fang Xie, and Zhiqiang Gao*

Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, Singapore 138669, Republic of Singapore

Herein we report the synthesis, intercalating properties, and analytical applications of an imidazole-substituted naphthalene diimide, N,N′-bis(3-propylimidazole)-1,4,5,8naphthalene diimide (PIND), functionalized with electrocatalytic redox moieties. PIND was prepared in a singlestep reaction from the corresponding dianhydride. Attachment of the redox moieties to PIND relied upon ligand exchange with one of the liable chloride ligands of an Os(bpy)2Cl2 (bpy ) 2,2′-bipyridine) complex. The Os(bpy)2Cl2 complex was grafted onto PIND through coordinative bonds with the two imidazole groups at its termini, forming a PIND-[Os(bpy)2Cl]+ compound (PIND-Os). Gel electrophoretic studies revealed that PIND-Os binds more strongly to double-stranded DNA (ds-DNA) than its parent compound 1,4,5,8-naphthalene diimide. The naphthalene diimide group binds to ds-DNA in a “classical” threading intercalation mode, while the two Os(bpy)2Cl+ pendants interact with DNA via electrostatic interaction, reinforcing the intercalation by “locking up” the naphthalene diimide group in place. An electrochemical biosensor was fabricated using the redox-active and catalytic PIND-Os intercalator. An increase in sensitivity of 2500-fold over direct voltammetry was obtained in electrocatalytic amperometry, making this an interesting system for amperometric DNA sensing. Under optimized experimental conditions, the biosensor allowed the detection of a 50-mer target DNA in the range of 1.0-300 pM with a detection limit of 600 fM (1.5 amol, 23 fg). DNA-based biosensors have potential applications that range from genotyping to molecular diagnostics.1-4 The use of fluorescently labeled oligonucleotides in conjunction with surface modification techniques affords high-density DNA arrays for analyzing specific DNA sequences and gene expression. Although widely employed, these methods require labeling of the target DNA. Moreover, many of the fluorescence-based techniques only have sufficient sensitivities for the detection of DNA at subnanomolar levels.5-8 Electrochemical transduction methods have therefore been proposed for ultrasensitive detection of DNA hybridization * Corresponding author. Tel: +6824-7113. Fax: +6478-9084. E-mail: zqgao@ ibn.a-star.edu.sg. (1) Service, R. Science 1998, 282, 396-399. (2) Ramsey, G. Nat. Biotechnol. 1998, 16, 40-45. (3) Drummond, T. G.; Hill, M. G.; Barton, J. K. Nat. Biotechnol. 2003, 21, 11921196. (4) Christopoulos, T. K. Anal. Chem. 1999, 71, 425R-438R.

126 Analytical Chemistry, Vol. 77, No. 1, January 1, 2005

events.9-11 The inherent miniaturization of electrochemical biosensors and their compatibility with advanced semiconductor technologies promise to provide a simple, accurate, and inexpensive platform for nucleic acid assays. Both low- and high-density electrochemical DNA biosensor arrays have been successfully fabricated.12,13 Economy-wise, the cost of high-density electrochemical biosensor arrays must be substantially lowered in order to compete with optical microarrays. On the other hand, lowdensity electrochemical biosensor arrays are more attractive than optical biosensor arrays. The advantages of low-density electrochemical biosensor arrays are as follows: (i) more cost-effective than optical biosensor arrays; (ii) ultrasensitive when coupled with catalysis; (iii) rapid, direct, turbid, and light-absorbing-tolerant detection; and (iv) portable, robust, low-cost, and easy-to-handle electrical components suitable for field tests and home care use. Furthermore, electrochemical detection of DNA through an electron-transfer reaction offers the benefits of ultrasensitive DNA detection without prior labeling of the target DNA. Electrochemical biosensors for sequence-specific detection of a target DNA have been reported since 1990.14 A wide variety of approaches, including those based on gold nanoparticles,15 direct oxidation of guanine,16 electrogenerated chemiluminescence,17 and DNA intercalators,18-22 have been developed. The employment (5) Berre, V. L.; Trevisiol, E.; Dagkessamanskaia, A.; Sokol, S.; Caminade, A.; Majoral, J. P.; Meunier, B.; Francois, J. Nucleic Acids Res. 2003, 31, e88. (6) Dolan, P. L.; Wu, Y.; Ista, L. K.; Metzenberg, R. L.; Nelson, M. A.; Lopez, G. P. Nucleic Acids Res. 2001, 29, e107. (7) Zammatteo, N.; Jeanmart, L.; Hamels, S.; Courtois, S.; Louette, P.; Hevesi, L.; Remacle, J. Anal. Biochem. 2000, 280, 143-150. (8) Lipshutz, R. J.; Fodor, S. P. A.; Gingeras, T. R.; Lockhart, D. J. Nat. Genet. 1999, 21 (Suppl.), 20-24. (9) Palecek, E.; Jelen, F. Crit. Rev. Anal. Chem. 2002, 3, 261-270. (10) (a) Kelley, S. O.; Barton, J. K.; Jackson, N. M.; Hill, M. G. Bioconjugate Chem. 1997, 8, 31-37. (b) Kelly, S. O.; Boon, E. M.; Barton, J. K.; Jackson, N. M.; Hill, M. G. Nucleic Acids Res. 1999, 27, 4830-4837. (11) Wang, J. Nucleic Acids Res. 2000, 28, 3011-3016. (12) Nebling, E.; Grunwald, T.; Albers, J.; Schafer, P.; Hintsche, R. Anal. Chem. 2004, 76, 689-696. (13) Dill, K.; Montgomery, D. D.; Wang, W.; Tsai, J. C. Anal. Chim. Acta 2001, 444, 69-78. (14) Wilson, W. D. In Nucleic Acids in Chemistry and Biology; Blackburn, G. M., Gaic, M. J., Eds. Oxford University Press: New York, 1996; p 329. (15) (a) Park, S.; Taton, T. A.; Mirkin, C. A. Science 2002, 295, 1503-1506. (b) Taton, T. A.; Mirkin, C. A.; Letsinger, R. L. Science 2000, 289, 1757-1760. (16) (a) Wang, J.; Cai, X.; Rivas, G.; Shiraishi, H.; Farias, P. A. M.; Dontha, N. Anal. Chem. 1996, 68, 2629-2634. (b) Wang, J.; Bollo, S.; Paz, J. L. L.; Sahlin, E.; Mukherjee, B. Anal. Chem. 1999, 71, 1910-1913. (c) Wang, J.; Fernandes, J. R.; Kubota, L. T. Anal. Chem. 1998, 70, 3699-3702. (17) Miao, W.; Bard, A. J. Anal. Chem. 2003, 75, 5825-5834. (18) Hashimoto, K.; Ito, K.; Ishimori. Y. Anal. Chem. 1994, 66, 3830-3833. (19) Carter, M. T.; Rodriguez, M.; Bard, A. J. J. Am. Chem. Soc. 1989, 111, 8901-8911. 10.1021/ac0493469 CCC: $30.25

© 2005 American Chemical Society Published on Web 12/03/2004

of a redox-active DNA intercalator as a hybridization indicator avoids labeling of the target DNA as commonly used in conventional DNA detection techniques.18-22 However, most of these biosensors suffer from low signal/noise ratio problems since most DNA intercalators bind not only to double-stranded DNA (dsDNA) but also, although to a much less extent, to single-stranded DNA (ss-DNA) by electrostatic interaction.18-22 New intercalators, offering better discrimination between ss- and ds-DNA are being developed to achieve greater signal/noise ratios. Takenaka and co-workers have synthesized a ferrocene-grafted naphthalene diimide (ND) threading intercalator that binds to ds-DNA more selectively than the usual intercalators.22 Some interesting results were obtained, but many issues need to be rectified. To further enhance the sensitivity and lower the detection limit, chemical and biological amplification mechanisms were introduced.23-26 For example, Thorp et al. proposed an electrocatalytic scheme for the detection of DNA by employing a homogeneous electrocatalyst, Ru(bpy)32+/3+ or Os(bpy)32+/3+ complex.23 Two fundamental issues that need to be addressed in the development of catalytic DNA biosensors are the high background signal and the insufficient sensitivity of the assays. In previous studies where transitional metal complexes were used as homogeneous catalysts, the analytical signal was superimposed onto an intrinsically large background current due to direct oxidation of the catalyst and the catalytic oxidation of the oligonucleotide capture probe (CP) by the catalyst.23 While most of the catalytic oxidation current can be eliminated by replacing the oligonucleotide CP with a nucleic acid analogue,11,23 little can be done to reduce the direct oxidation of the catalyst. In enzyme-based DNA assays, the background currents, usually in nanoamperes range, are directly associated with non-DNA-related enzyme uptake such as nonspecific adsorption and electrostatic interaction. In a more recent report, it has been shown that the background current can be minimized at DNA biosensors having a bilayer configuration.26 Nonetheless, the development of highly sensitive DNA biosensors that are inexpensive to manufacture and simple to use would enable further adoption of microarray techniques in biomedical research and health care. The quest for DNA-specific binding agents has been fueled by the desire to modulate gene expression, to search for new antitumor drugs, and to develop molecular probes for DNA. Threading intercalators are the most important group of compounds that interact reversibly with ds-DNA. Some of them are valuable antitumor drugs currently used for the treatment of (20) Zeman, S. M.; Phillips, D. R.; Crothers, D. M. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 11561-11565. (21) Erkkila, K. E.; Odom, T. D.; Barton, J. K. Chem. Rev. 1999, 99, 27772795. (22) Takenaka, S.; Yamashita, K.; Takagi, M.; Uto. Y.; Kondo, H. Anal. Chem. 2000, 72, 1334-1341. (23) (a) Armistead, P. M.; Thorp, H. H. Anal. Chem. 2000, 72, 3764-3770. (b) Gore, M. R.; Szalai, V. A.; Ropp, P. A.; Yang, I. V.; Silverman, J. S.; Thorp, H. H. Anal. Chem. 2003, 75, 6586-6592. (c) Napier, M. E.; Loomis, C. R.; Sistare, M. F.; Kim, J.; Eckhardt, A. E.; Thorp, H. H. Bioconjugate Chem. 1997, 8, 906-913. (24) (a) Zhang, Y.; Kim, H.; Heller, A. Anal. Chem. 2003, 75, 3267-3269. (b) Dequaire, M.; Heller, A. Anal. Chem. 2002, 74, 4370-4377. (25) Patolsky, F.; Katz, E.; Willer, I. Angew. Chem., Int. Ed. 2002, 41, 33983402. (26) (a) Xie, H.; Zhang, C.; Gao, Z. Anal. Chem. 2004, 76, 1611-1617. (b) Xie, H.; Yu, Y. H.; Xie, F.; Lao, Y. Z.; Gao. Z. Anal. Chem. 2004, 76, 4023-4029. (c) Xie, H.; Yu, Y. H.; Mao, P. L.; Gao, Z. Nucleic Acids Res. 2004, 32, e15.

ovarian and breast cancers.28,29 They share common structural features such as the presence of planar polyaromatic systems that bind in insertion between base pairs of ds-DNA. The moieties linked to the intercalating units also play an important role in the affinity and selectivity shown by these compounds.28 We report here the synthesis, characterization, and analytical applications of a novel electrocatalytic DNA threading intercalator. The intercalator used in this study is a ND derivative grafted with two redox moieties at its termini. Os(bpy)2Cl2 complex was chosen as the electrocatalytic moiety because of its low redox potential, high electron-transfer rate, and excellent catalytic properties toward the oxidation of ascorbic acid (AA).27 Experimental results suggested that the intercalator is highly selective to ds-DNA with an enhanced stability constant. In addition, the Os(bpy)2Cl+ pendants effectively catalyze the oxidation of AA at a potential as low as 0.16 V as compared to 0.85 V at a target DNA hybridized biosensor without the intercalator. The combination of the selective incorporation of the intercalator in ds-DNA and the highly efficient electrocatalysis provides a generic platform for ultrasensitive nonlabeling detection of DNA. EXPERIMENTAL SECTION Chemicals. 1-(3-Aminopropyl)imidazole (AI, 98%) and 1,4,5,8naphthalene tetracarboxylic dianhydride (>95%) were purchased from Sigma-Aldrich (St. Louis, MO). 2,2′-Bipyridine (bpy, 99%) was from Avocado Research Chemicals Ltd. (Leysham, Lancester, U.K.). Os(bpy)2Cl2 was synthesized from K2OsCl6 (99%, Stem Chemicals) according to the procedure described by Lay.30 All other reagents were obtained from Sigma-Aldrich and used without further purification. CP used in this work was custommade by Alpha-DNA (Montreal, Canada) and all other oligonucleotides were custom-made by 1st Base Pte Ltd. (Singapore) (Table 1). A 10 mM Tris-HCl/1.0 mM EDTA/0.10 M NaCl buffer solution (TE) was used as the hybridization buffer. A phosphatebuffered saline (PBS, pH 7.4), consisting of 0.15 M NaCl and 20 mM phosphate buffer, was used as the supporting electrolyte. Apparatus. Electrochemical experiments were carried out using a CH Instruments model 660A electrochemical workstation coupled with a low-current module (CH Instruments, Austin, TX). A conventional three-electrode system, consisting of a 3.0-mmdiameter gold working electrode, a nonleak miniature Ag/AgCl reference electrode (Cypress Systems, Lawrence, KS), and a platinum wire counter electrode, was used in all electrochemical measurements. To avoid the spreading of the sample droplet beyond the 3.0-mm-diameter working area, a patterned hydrophobic film was applied to the gold electrode after CP immobilization. All potentials reported in this work were referred to the Ag/ AgCl electrode. Ultraviolet-visible absorption spectra were recorded on an Agilent 8453 UV-visible spectrophotometer. Mass spectrometric experiments were performed with a Finnigan/MAT LCQ mass spectrometer (ThermoFinnigan, San Jose, CA). All spectra were recorded at room temperature unless otherwise noted. Synthesis of N,N′-Bis(3-propylimidazole)-1,4,5,8-naphthalene Diimide-[Os(bpy)2Cl]+ (PIND-Os). The synthesis (27) Fei, J.; Luo, L.; Hu, S.; Gao. Z. Electroanalysis 2004, 15, 319-323. (28) Brana, M. F.; Cacho, M.; Gradillas, A.; Pascual-Teresa, B.; Ramos, A. Curr. Pharm. Des. 2001, 17, 1745-1780. (29) Malonne, H.; Atassi, G. Anticancer Drugs 1997, 8, 811-22. (30) Lay, P. A.; Sargeson, A. M.; Taube, H. Inorg. Synth. 1986, 24, 291-296.

Analytical Chemistry, Vol. 77, No. 1, January 1, 2005

127

Table 1. Oligonucleotide Sequences Used in This Work capture probe target DNA complementary one-base mismatched two-base mismatched noncomplementary hairpin oligonucleotides for intercalation study

5′-GCCAGCGTTCAATCTGAGCCATGATCAAACTCTTCAAATGCCG ATTAGGC-(A)6-(CH2)6-SH 5′-GCCTAATCGGCATTTGAAGAGTTTGATCATGGCTCAGATTGAACGCTGGC 5′-GCCTAATCGGCATTTGAAGAGTTTGATCATGGCTCAGAATGAACGCTGGC 5′-GCCTAATCGCCATTTGAAGAGTTTGATCATGGCTCAGAATGAACGCTGGC 5′-AGGCCTTAAGATACGCTATTAAGCTACTAGTTGGCCTAAAGCTGATTCCA 5′-AATTT-CCCCC-AAATT 5′-AATAT-CCCCC-ATATT 5′-ATTTA-CCCCC-TAAAT

Scheme 1. Synthetic Route to PIND-Os Intercalator

of PIND-Os is outlined in Scheme 1. PIND was prepared based on a general procedure for the synthesis of ND.31,32 Briefly, to a magnetically stirred mixture of 3.0 mL of AI and 3.0 mL of tetrahydrofuran was slowly added 0.30 g of 1,4,5,8-naphthalene tetracarboxylic dianhydride. The rate of addition was controlled so that there was little clogging. The reaction mixture was refluxed for 24 h and then cooled to room temperature. Next, it was dispersed in 10 mL of acetone/water (3/1) mixture and poured into 500 mL of rapidly stirred anhydrous ether to precipitate the (31) Rademacher, A.; Maerkle, S.; Ianghals, H. Chem. Ber. 1982, 115, 29722976. (32) Katz, H. E.; Johnson, J.; Lovinger, A. J.; Li, W. J. Am. Chem. Soc. 2000, 122, 7787-7792.

128

Analytical Chemistry, Vol. 77, No. 1, January 1, 2005

compound. The precipitate was collected by suction filtration through a fine fritted funnel and washed briefly with ethanol. Purification was performed by crystallization from chloroform/ ethanol mixture (1/1 by volume), followed by drying overnight under vacuum at 40 °C to give 0.46 g of yellow crystals (yield 85%): 1H NMR (300 MHz CDCl3) δ 8.76 (4H), 7.54 (2H), 7.26 (2H), 4.27 (4H), 4.12(4H), 2.31 (4H), 1.83 (2H). PIND-Os was synthesized in a single-step ligand-exchange reaction. To a solution of Os(bpy)2Cl2 (0.32 g, 0.52 mmol) in 8.0 mL of freshly distilled ethylene glycol was added PIND (0.12 g, 0.25 mmol) in small portions over 10 min, and the resulting mixture was refluxed for 30-40 min. The completion of the ligandexchange reaction was monitored by cyclic voltammetry. The

Scheme 2. Schematic Illustration of DNA Assay Using PIND-Os Intercalator

Figure 1. Normalized cyclic voltammograms of Os(bpy)2Cl2 after (1) 0 and (2) 30 min of refluxing with PIND in ethylene glycol, and (3) purified PIND-Os.

purple reaction mixture was then poured slowly into 100 mL of rapidly stirred ethanol saturated with KCl. The precipitate was collected by suction filtration through a fine fritted funnel. The crude product was washed with PBS, dissolved in 3.0-5.0 mL of ethanol, and precipitated again from KCl-saturated ethanol. The precipitate was further purified by crystallization from ethanol giving the pure product with 78% yield. Cyclic voltammetric tests of the product showed a single pair of reversible redox peaks at a gold electrode with an E1/2 of 0.12 V in PBS. To ensure a complete double ligand exchange at the two imidazole termini of PIND, a slight excess of Os(bpy)2Cl2 (5-15%) was required. Immobilization of CP on a Gold Electrode. The preparation and pretreatment of gold electrodes were as previously described.26,33 Briefly, prior to CP adsorption, a gold electrode was exposed to oxygen plasma for 5-10 min and immediately immersed in absolute ethanol for 20 min to reduce the oxide layer. A CP monolayer was formed by immersing the gold electrode in a PBS solution of 100 µg/mL CP for 16-24 h. After adsorption, the electrode was copiously rinsed with PBS and soaked in PBS for 20 min, rinsed again, and blown dry with a stream of air. The surface density of CP, assessed electrochemically by the use of a cationic redox probe according to the procedure proposed by Tarlov,34 was found to be in the range of (1.13-1.30) × 10-11 mol/ cm2. To minimize non-DNA-related PIND-Os uptake and improve the quality and stability of the CP monolayer, a layer of 1-mercaptododecane (MD) was deposited on the CP-coated electrode by immersing it in an ethanolic solution of 2.0 mg/mL MD for 4-6 h. Unreacted MD molecules were rinsed off, and the electrode was washed by immersion in a stirred ethanol for 10 min followed by thorough rinsing with ethanol and water. The electrode was ready after air-drying. Hybridization and Detection. The hybridization of a target DNA and its electrochemical detection were carried out in three steps, as illustrated in Scheme 2. First, the CP-coated electrode was placed in a moisture-saturated environmental chamber maintained at 60 °C (low stringency, 27 °C below the salt-adjusted (33) Gao, Z.; Siow, K. S.; Chan, H. Synth. Met. 1995, 75, 5-10. (34) Steel, A. B.; Herne, T. T.; Tarlov, M. J. Anal. Chem. 1998, 70, 4670-4677.

melting temperature). A 2.5-µL aliquot of hybridization solution containing the target DNA was uniformly spread onto the electrode. It was then rinsed thoroughly with a blank hybridization solution at 60 °C after a 30-min hybridization period and incubated at 25 °C for 10 min with a 5.0-µL aliquot of 100 µg/mL PIND-Os in the hybridization solution. PIND-Os was attached to the hybridized target DNA via threading intercalation. It was then thoroughly rinsed with a NaCl-saturated phosphate buffer (pH 7.4) containing 10% ethanol. The AA electrooxidation current was measured amperometrically in vigorously stirred PBS containing 5.0 mM AA. At low DNA concentrations, smoothing was applied after each amperometric measurement to remove random noise and electromagnetic interference. RESULTS AND DISCUSSION Formation of PIND-Os. The formation of the redox-active PIND-Os intercalator can be conveniently monitored by cyclic voltammetry. During reflux in ethylene glycol, cyclic voltammetric tests were conducted every 5 min. Figure 1 shows two typical voltammograms obtained in the first 30 min. As shown in trace 1 in Figure 1, before addition of PIND to Os(bpy)2Cl2, one pair of reversible voltammetric peaks centered at -0.11 V was obtained, corresponding to the well-known redox process of Os(bpy)2Cl2. Upon adding PIND, a new pair of voltammetric peaks appeared at 0.12 V, indicating the formation of PIND-Os (Figure 1, trace 2). Both electron-transfer processes are clearly resolved and have all the characteristics of reversible processes, except the slightly larger peak-to-peak potential separations, which are mainly due to a higher iR drop of the reaction medium. The intensities of the voltammetric peaks at 0.12 V increased gradually with reaction time. Simultaneously, those at -0.11 V diminished gradually. Both of the redox pairs reached a steady state after 30-40 min of refluxing. The minute voltammetric peaks at -0.11 V are indicative of the excess amount of Os(bpy)2Cl2. After purification, voltammetric tests of the thus purified PIND-Os showed only one pair of voltammetric peaks, implying that the purification process is very effective (Figure 1, trace 3). UV-visible absorption spectra of the starting materials, an Os(bpy)2Cl(AI) model compound, and PIND-Os are depicted in Figure 2. UV-visible spectrum of PIND-Os (Figure 2, trace 3) is similar to that of the Ru(bpy)3-ND compound.35-37 It exhibits Analytical Chemistry, Vol. 77, No. 1, January 1, 2005

129

Figure 2. UV-visible absorption spectra of (1) 50 µM PIND, (2) 50 µM Os(bpy)2Cl2, (3) 50 µM PIND-Os, and (4) 50 µM Os(bpy)Cl(AI) in ethanol. For clarity, (4) was lifted up by 2.0 units.

an intense band in the UV region due to intraligand (IL)∂ f ∂*(bpy) transitions, followed by a broad band in the visible region (400-600 nm) due to spin-allowed Os(dπ) f bpy(π*) metal-toligand charge-transfer (MLCT) transition.35 The peaks at 361 and 380 nm are mainly due to π f π* transition in PIND with some contribution from underlying MLCT absorption. The absorption maximum of PIND-Os is red-shifted from 410 to 438 nm with respect to Os(bpy)2Cl2 (Figure 2, trace 2), and a new absorption maximum was observed at 532 nm (Figure 2, trace 3). The same changes were also observed in the spectrum of the model compound Os(bpy)2Cl(AI) as compared to Os(bpy)2Cl2 (Figure 2, trace 4). This is likely a direct consequence of the ligand exchange, which results in two types of MLCT transitions within the osmium complex: Os* f bpy, and Os* f AI. The imidazole groups of PIND are conjugated, resulting in a lower π* level for this ligand relative to the chloride of the complex. Moreover, the spectrum of PIND-Os is a composite of the absorption spectra from both the ND moiety and the Os(bpy)2Cl(AI) model compound (Figure 2, traces 1 and 4). A simple overlay of Os(bpy)2Cl(AI) and PIND (2Os(bpy)2Cl(AI) + PIND) generated a spectrum which is almost identical to that of PIND-Os (not shown), confirming the formation of PIND-Os. Although we concluded from the UV-visible spectrophotometric and electrochemical evidence that the coupling between PIND and Os(bpy)2Cl2 results in a coordinative linkage and both imidazole termini of the PIND are grafted with Os(bpy)2Cl+, a more direct proof of the formation of PIND-Os was necessary. Thus, we conducted a series of mass spectrometric tests on PIND-Os using electrospray ionization mass spectrometry (ESIMS). Predominant peaks were found at m/z 780 (68%), 483.3 (46%), and 242.3 (100%), corresponding to (PIND-Os)2+/2, (PIND + H+), and (PIND + 2H+)/2, respectively, which are in good agreement with the molecular weights of the desired compounds. Since monografted PIND was not observed in the ESI-MS spectrum, we can rule out any incomplete grafting of Os(bpy)2Cl2. (35) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.; von Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85-277. (36) Yen, S.; Gabbay, E. J.; Wilson, W. D. Biochemistry, 1982, 21, 2070-2076. (37) Dixon, D. W.; Thornton, N. B.; Steullet, V.; Netzel, T. Inorg. Chem. 1999, 38, 5526-5534.

130 Analytical Chemistry, Vol. 77, No. 1, January 1, 2005

Electrochemical Properties of PIND-Os. As illustrated in Figure 1, PIND-Os behaved exactly as anticipated for a highly reversible redox couple in solution. Little change was observed after numerous repetitive potential cycling between -0.30 and +0.70 V, revealing good stability of PIND-Os in solution. At slow scan rates of 4.0, and a constant hypochromism was observed for the ratio above 4.0, indicating that the binding of PIND-Os to dsDNA takes place by preferential intercalation of the ND. To have a quantitative estimation of the intercalating property, a competition experiment similar to that proposed by Boger,38 was designed using AT-rich short hairpin oligonucleotides (Table 1) to establish the binding constant. It has been demonstrated that these hairpin oligonucleotides form a 1:1 complex with threading intercalators. The basis of this methodology involves the use of two intercalators, one fluorescent and the other nonfluorescent. (38) Boger, D. L.; Fink, B. E.; Brunette, S. R.; Tse, W. C.; Hedrick, M. P. J. Am. Chem. Soc. 2001, 123, 5878-7891.

Figure 3. UV-visible absorption spectra of 15 µM PIND-Os as a function of increasing concentration of salmon sperm DNA (in base pair) of (1) 0, (2) 30, and (3) 60 µM. Inset: Enlarged UV-visible adsorption spectra of the intercalative binding region.

The fluorescent intercalator first saturates the ds-DNA. Then a second intercalator, in this case PIND-Os, is introduced into the system with gradual increase in concentration. We hypothesized that the two intercalators would bind to similar sites in the dsDNA. In the competition experiment, we are interested in monitoring the changes in fluorescence intensity during the displacement of ds-DNA-bound fluorescent molecules by PINDOs through an increasing concentration of PIND-Os in the system. A well-known threading intercalator, ethidium bromide (EB), was chosen as our fluorescent indicator. EB has been widely studied as an efficient DNA intercalator and is one of the most popular fluorescent intercalators used in DNA assays. It possesses relatively little sequence preference and displays a 25-fold fluorescence enhancement upon binding to ds-DNA, which provides sufficient sensitivity and good discrimination against free EB molecules in fluorescence measurement. In addition, the kinetics of EB intercalation is quite fast,39 which significantly shortens the time needed to reach equilibrium. Quantitative displacement of EB from the hairpin generates a titration curve from which the free ligand concentration may be determined. The binding constant Kd can then be established by Scatchard analysis.40 The following equation is used to determine the free ligand concentration employed to generate a Scatchard plot:38

[DNA](X - ∆Fx/∆Fsat) ) [free ligand]

(1)

where [DNA] is the total concentration of DNA, X molar equivalent of ligand versus DNA, ∆Fx change in fluorescence, ∆Fsat change in fluorescence at the point where DNA is saturated with ligand, and [free ligand] concentration of free ligand. A plot of ∆F/[free ligand] versus ∆Fx around the saturation region yields a linear portion of the Scatchard plot and provides -Kd as the slope of this plot (Figure 4B). To ensure that our approach is appropriate for this study, an increasing concentration (0-100 µΜ) of a well-studied nonfluo(39) Macgregor, R. B.; Clegg, R. M.; Jovin, T. M. Biochemistry 1987, 26, 40084016. (40) Scatchard, G. Ann. N. Y. Acad. Sci. 1949, 51, 660-672.

Figure 4. (A) Gel electrophoretic results of DNA binding property of PIND-Os. PIND-Os/EB ratio from lanes 1 to 5: 0/8, 1/8, 1/4, 3/8, and 1/2. The samples were run at 10 V/cm in 90 mM Tris-90 mM boric acid-2.0 mM EDTA buffer on a 2.5% agarose gel. (B) Scatchard plot for the titration of hairpin oligonucleotides/EB (2.0/3.0) mixture with PIND-Os.

rescent intercalator, the mother compound ND, was first added to the EB-saturated ds-DNA solution. Gel electrophoretic experiments showed that the fluorescence intensity of the EB intercalated with ds-DNA diminished gradually as the concentration of ND was increased. The binding constant Kd of 4.8 × 105, estimated from the experimental data, was in good agreement with the literature value.41 Subsequently, PIND-Os was studied with respect to its ability to compete against EB for binding to ds-DNA using the same approach. Different amounts of PIND-Os were mixed with the EB-saturated ds-DNA to examine its binding ability. Figure 4A shows representative gel electrophoretic data obtained from EB-saturated DNA solutions treated with solutions of PIND-Os of increasing concentrations. As shown in Figure 4A, PIND-Os exhibited a remarkable binding affinity toward dsDNA. Lanes 1-5 correspond to different ratios of PIND-Os/EB. The higher the ratio of PIND-Os/EB, the lower the fluorescence intensity (bottom row in Figure 4A). The lower fluorescence intensities of the ds-DNA obtained with the higher ratios of PIND-Os/EB (lanes 2-5) suggests that more PIND-Os molecules are bound to the ds-DNA and larger amounts of EB molecules are replaced. As depicted in lane 3, at a PIND-Os/EB molar ratio of as low as 1/4, more than 60% of the ds-DNA-bound EB was replaced, as evidenced by the diminished fluorescence intensity of intercalated EB and the increased fluorescence intensity of free EB (top row in Figure 4A), suggesting that PIND-Os is a much stronger DNA intercalator than EB. The (41) Murr, M. M.; Harting, M. T.; Guelev, V.; Ren, J.; Chaires, J. B.; Iverson, B. L. Bioorg. Med. Chem. 2001, 9, 1141-1446.

Analytical Chemistry, Vol. 77, No. 1, January 1, 2005

131

Figure 5. (A) Cyclic voltammograms of PIND-Os bound to (1) 1.0 µM noncomplementary target DNA and to (2) 10, (3) 200, and (4) 500 nM complementary target DNA hybridized biosensors. Intercalation was carried out in 100 µg/mL PIND-Os in TE. Supporting electrolyte PBS, potential scan rate 100 mV/s. (B) Cyclic voltammograms of 5.0 mM AA at a 1.0 µM complementary target DNA hybridized biosensor (1) before and (3) after PIND-Os incubation in 100 µg/mL PIND-Os in TE, and (2) a 1.0 µM complementary target DNA hybridized biosensor after PIND-Os incubation in blank PBS. Supporting electrolyte PBS; potential scan rate 100 mV/s.

binding constant Kd, estimated from the experimental data, was 2.0 × 107 (Figure 4B), corresponding to ∼40-fold enhancement over ND. A plausible explanation for the stability constant enhancement would be that after the ND group has intercalated with the ds-DNA, the two cationic Os(bpy)2Cl+ groups in PINDOs form ion pairs with phosphates on each side of the ds-DNA, making ND more tightly fixed between the base pairs of the dsDNA. In addition, a closer examination of the gel image showed that, accompanying the weakening of fluorescence intensity, there was a systematic change in DNA mobility. The higher the ratio of PIND-Os/EB, the higher the band appeared in the gel image and, in turn, the lower the mobility of the ds-DNA. The molecular mass of monocationic EB is 324 while that of dicationic PINDOs is 1560, as determined by mass spectrometry. Apparently, the lower mobility is caused by the bulky and dicationic nature of PIND-Os. Analytical Applications of PIND-Os in Ultrasensitive DNA Biosensors. DNA biosensors with redox-active moieties grafted ND as an electrochemical indicator have previously been reported.22 When hybridization occurs, ND selectively interacts with ds-DNA and gives a greatly enhanced analytical signal compared to the nonhybridized ss-DNA. The difference in voltammetric peak currents is used for quantitation purposes. In a similar way, PIND-Os was evaluated as a novel redox-active indicator for possible applications in ultrasensitive DNA sensing. In the first hybridization test, a complementary and a noncomplementary oligonucleotide were selected as our target DNAs. Upon hybridization, the complementary target DNA was selectively bound to CP and became fixed on the biosensor surface. On the other hand, little if any of the noncomplementary DNA was captured during hybridization; hence, minute voltammetric response of the biosensor was expected. Thorough rinsing with the hybridization buffer washed off most of the nonhybridization132

Analytical Chemistry, Vol. 77, No. 1, January 1, 2005

related DNA. PIND-Os was then brought to the biosensor surface during a subsequent incubation with a PIND-Os solution. Cyclic voltammograms for the biosensors after hybridization with the complementary and noncomplementary target DNAs are shown in Figure 5A. One pair of minute voltammetric peaks were observed at the redox potential of PIND-Os after hybridization to noncomplementary DNA (Figure 5A, trace 1), largely due to pure electrostatic interaction between PIND-Os and CP on the biosensor surface. As shown in traces 2-4 in Figure 5A, after hybridization with different amounts of the complementary target DNA, a slight positive shift in the redox potential was observed and the peak current increased by as much as 100-fold. It was found that extensive washing with the NaCl-saturated phosphate buffer removed most of the non-DNA-related PIND-Os uptake. These results clearly demonstrated that PIND-Os selectively interacts with ds-DNA and the PIND-Os-ds-DNA adduct has a very low dissociation rate, paving the way for the development of ultrasensitive DNA biosensors. Consequently, the usage of intercalated PIND-Os as the redox-active indicator for direct detection of DNA was evaluated. A detection limit of 1.5 nM (3.8 fmol) and a dynamic range up to 500 nM were obtained. The hybridization efficiency at the high end of the dynamic range was evaluated electrochemically using Tarlov’s method,34 taking 1.2 × 10-11 mol/cm2 (midrange of the estimated values) as the CP surface coverage on the 3.0-mm-diameter gold electrode. It was found that the hybridization efficiency is in the range of 1721%, corresponding to 14-17% of target DNA in the sample droplet, which is comparable to the values found in the literature.42,43 The number of PIND-Os molecules producing the observed current was estimated from the charge under the (42) Caruso, F.; Rodda, E.; Furlong, D. N.; Niikura, K.; Okahata, Y. Anal. Chem. 1997, 69, 2043-2049. (43) Satjapipat, M.; Sanedrin, R.; Zhou, F. Langmuir 2001, 17, 7637-7644.

Figure 6. (A) Amperometric responses of PIND-Os bound to (1) a 25 pM complementary target DNA, (2) a 25 pM one-base mismatched DNA, (3) a 25 pM two-base mismatched DNA, and (4) a 200 pM noncomplementary target DNA hybridized biosensor after PIND-Os incubation in 100 µg/mL PIND-Os in TE. Supporting electrolyte PBS, 5.0 mM AA, and poise potential 0.20 V. (B) Calibration plot. Conditions are as in (A).

oxidation current peak. Since two electrons are transferred per PIND-Os molecule, the observed current of 0.58 µA after hybridization to 500 nM of the complementary target DNA, resulted therefore from 2.6 pmol of active and intercalated PINDOs, representing a PIND-Os/DNA base pair ratio of ∼1/4. In other words, maximum PIND-Os loading was achieved throughout the dynamic range.36,37 This amount represents