Impedance Sensing of DNA Binding Drugs using Gold Substrates Modified with Gold Nanoparticles
BACKGROUND OF THE INVENTION
(01) According to the present knowledge DNA-drug interactions can be analyzed by SPR, NMR or other techniques. However, both of these methods require expensive equipment and complex procedures for sample preparations. Also, DNA probes have been developed for attachment to electrode surface.
(01) There are problems regarding the system, some drugs cannot be dissolved in aqueous solution. If organic solvent is used to dissolve the drugs, the organic solvent will denature the monolayer of DNA probe, and the DNA probe will lose its binding and selectivity for the drug. It is also problematical to control the non-specific binding of the drugs to the DNA probe e.g. by physical absorption. Non-specific binding to DNA will decrease the binding efficiency of the drugs, therefore decreasing the detection limits.
(01) Considerable efforts have also been devoted to manipulate electrode surfaces with nanomaterials to enhance their performance.1 In the rational design of new drugs and drug screening, the interaction of anticancer drugs with DNA has been studied by several techniques,2 including electrochemical characterization for the interracial reaction of drugs with DNA modified surfaces.3
(02) Control of the attachment of nanoparticles onto a surface is a key issue for the fabrication process of biosensing chips. Several procedures have been proposed and applied to the immobilization of nanoparticles, such as electrolysis deposition and monolayer assistance embedding12.
(03) Electrochemical impedance spectroscopy (EIS) has been proven as one of the most powerful tools for investigation of interfacial reaction mechanisms.4 This technique offers several advantages over chronoamperometry and cyclic voltammetry because the effects of solution resistance, double-layer charging, and currents due to diffusion or to other processes occurring in the monolayer are observed more explicitly.5 The use of gold nanoparticle-based materials has recently emerged as a novel approach for biosensing because of their unique optical, catalytic and other properties.6
SUMMARY OF THE INVENTION
(04) According to the invention we exploit EIS to study the interaction of DNA specific sequence binding drugs with a duplex DNA monolayer immobilized on a gold substrate surface e.g. through a thiol linker to develop a novel biosensor and biosensing scheme. Gold nanoparticles are electrochemically deposited onto a gold substrate e.g. electrodes and then encoded with a self-assembly of thiolated DNA monolayers that is specific to the drug target. This method offers several advantages over chronoamperometry and cyclic voltammetry because the effects of solution resistance, double-layer charging, and currents due to diffusion or to other processes occurring in the monolayer are observed more explicitly. Furthermore, the electrochemical deposition of gold nanoparticles on flat gold substrate surfaces results in the binding of higher amounts of DNA probes, and in turn a significant improvement in the detection sensitivity for the assay of interfacial drugs. The modified substrates made by electrochemical deposition of gold nanoparticles has proven to be very pure and stable. When compared to the deposition of gold particles on gold substrates using other chemical reagents, the method also avoids the negative effects from the chemical compounds on DNA immobilization and drug-DNA interactions. A gold electrode surface modified with gold nanoparticles showed a significant
improvement in the detection sensitivity. Also the DNA-capped gold nanoparticles on electrodes act as selective sensing interfaces.
(05) Specifically, the density and particle size of DNA probes and hence the sensitivity is controlled by varying the experimental parameters of gold nanoparticles electrochemical deposition, such as the concentration of gold salt, applied electrochemical deposition potential, deposition time, etc. These parameters will control the size, the density, even the shape of gold particles, and further will control the orientation of DNA probes, since eventually the DNA probes will attach on the gold substrate through gold nanoparticles.
(06) It will be appreciated that if the particle size is smaller, the number of particles is higher, the density is high, if the particle size is bigger, the density is lower.
(07) The formation of the DNA monolayer on a substrate surface modified with gold nanoparticles, is in cylindrical or concentric orientation. Although there is no technique which can directly observe the DNA monolayer on nanoparticle surfaces, from AFM image, there is about 20-70 nm increase in the diameter size of nanoparticles due to the DNA monolayer formation i.e. from 20-60 nm to 40-130 nm.. It is postulated that the increase of size is similar to the physical length of the ten base pair DNA that was used in the experiments, and corresponds to the fact that DNA is upstanding on the nanoparticles surface. Hence, the DNA capped nanoparticles shoud be formed in concentric orientation.
(08) According to one aspect of the invention, a biosensor is provided for detecting a target drug having specific DNA binding capability, comprising a flat gold substrate, gold nanoparticles of controlled particle size electrochemically deposited on the substrate, and a duplex DNA monolayer specific for the target drug bound to the gold nanoparticles.
(09) According to another aspect of the invention, a method is provided for detecting a target drug having specific DNA binding capability, comprising
(a) providing a biosensor comprising a fiat gold substrate, gold nanoparticles of controlled particle size electrochemically deposited on the substrate, and a duplex DNA monolayer specific for the target drug bound to the gold nanoparticles, and
(b) contacting the biosensor with a sample containing the target drug.
(010) According to yet another aspect of the invention, a method is provided for the analysis of interfacial DNA-drug interaction between a target drug having specific DNA binding capability and a DNA monolayer specific for the target drug, comprising
(a) providing an electrochemical cell including as working electrode a flat gold substrate, gold nanoparticles of controlled particle size electrochemically deposited on the substrate, and a duplex DNA monolayer specific for the target drug bound to the gold nanoparticles,
(b) contacting the working electrode with a sample containing the target drug, and
(c) measuring the interfacial interaction by EIS.
BRIEF DESCRIPTION OF THE DRAWING
Figure 1a,b, are AFM topographic images of GC-rich ds-DNA immobilized on (a) a flat gold surface and (b) on gold nanoparticle modified gold substrates.
Figure 1c,d, are Nyquist plots (Zim vs Zre ) for Faradaic impedance measurements.
Figure 2 is a schematic illustration of a general equivalent circuit model used for impedance data analysis.
Figure 3 is a schematic illustration, comprising a DNA probe binding to drug molecules (a) without gold nanoparticles and (b) with gold nanoparticles.
Figure 4 is a graph of the CV of flat gold electrode in 0.1 M NaOH solution containing 0.1 M HAuCI4 at 20 mV.s"1. The impedance experiments were run at an applied bias potential of the E0 of 220 mV vs Ag/AgCI, with 5 mV (rms) sinusoidal excitation amplitude.
Figure 5 are graphs of the Linear sweep voltammetry of (a) bare gold surface and (b) gold nanoparticle deposited gold surface in 0.5 M H2SO4 solution with a scan rate of 20 mV.s"1.
Figure 6 (A) representative AFM image of Au surface obtained by electron evaporation of gold on silica sheets after cleaning by electrochemical treatment showing atomically flat gold terraces. Scan size 500 x 500 nm. (B) and (C): the AFM images of gold nanoparticle deposited Au surfaces at different deposition times, (B) for 3 s, (C) for 10 s. Scan size 5000 x 5000 nm.
Figure 7 are AFM images of electrodeposited Au nanoparticles on flat gold substrate at E =0.2 V vs Ag/AgCI. The initial potential step at t = O s was started from 0.80 V. Images a-d correspond to different periods of deposition time, (a) 3, (b) 10, (c) 30, and (d) 60 s.
Figure 8 is a graph illustrating the comparison between the changes of electron-transfer resistance due to the DNA - drug interactions of flat gold electrode surfaces (zone A) and of gold nanoparticles - modified electrode surfaces (zone B). Each data point is the average of three measurements.
Figure 9 is a graph illustrating the relationship between the monolayer electron-transfer resistance and the concentration of nogalamycin on (■) gold nanoparticle-deposited gold electrodes and (•) flat gold electrodes. Both of the electrodes were immobilized with the GC-rich DNA probes.
Figure 10 are graphs illustrating the alteration of the monolayer electron- transfer resistance for two DNA-specific binding drugs mithramycin (■) and netropsin (•) on gold nanoparticle-deposited electrodes modified by (a) GC- rich DNA and (b) AT-rich DNA. The concentration of each drug is 1.5μM.
DETAILED DESCRIPTION OF THE INVENTION
(011) To illustrate the concept of the invention, two classical minor groove binders, mithramycin (MRA), a G-C specific DNA binding anticancer drug, netropsin (NRP), an A-T specific DNA binding drug and an intercalator nogalamycin (NGL)7 were investigated. Electrochemical deposition of gold nanoparticles8 on gold electrodes was performed in 100 mM NaOH solution containing 100 mM HAuCI4.
(012) Details of the electrode pretreatment, the preparation of drug binding solution, the description of EIS analysis model, AFM and electrochemical analysis of gold nanoparticle deposited surfaces at different deposition conditions, is provided below in the section headed Supporting Information.
(013) The roughness factor of the resulting gold film was 1.48-1.65, derived from the reduction peak current of electrochemically generated surface oxides by linear sweep voltammetry. 9 The deposition time of 10 s was controlled to yield a submonolayered gold film consisting of gold nanoparticles (20-60 nm in diameter) with an average density of 1.59x107/cm2. The resulting gold substrates were functionalized with two types (GC- rich ds-DNA: oligo 1, 5'- SH-(CHa)6-GGGGATGGGG-S1; oligo 2, δ'-CCCCATCCCC-S'; AT-rich ds-
DNA, oligo 3, 5'- SH-(CH3)6-AAAAGCAAAA-3'; oligo 4, 5'-TTTTGCTTTT) of pre-hybridized duplex DNA through the Au-S bond, respectively. Incubation for 48 h was required to attain high density packed, well-oriented monolayers of DNA on gold substrates or on gold nanoparticle deposited gold substrates. The AFM micrograph of a flat gold surface modified by DNA displayed a uniform distribution of the thiolated DNA probes that appeared as dense packed dots on gold terraces (Figure 1 a). Figure 1b depicts a DNA modified surface that had been adhered with gold nanoparticles. DNA-tagged gold particles, ranging from 40-130 nm, were bigger than unmodified gold nanoparticles. The electrode surfaces were characterized over a wide frequency range (100 kHz to 100 mHz; £°'= 220 mV vs. Ag/AgCI). For precise modeling of the impedance of DNA SAM modified electrode systems, the spectra in the form of Nyquist plots were analyzed with a modified Randies circuit (Supporting Information), to include an additional component in parallel with the Faradaic impedance of the interface, RSAM, of the monolayer- dependent resistance.10
(014) Figure 1 a,b. are AFM topographic images of GC-rich ds-DNA immobilized on (a) flat gold surface; (b) gold nanoparticle modified gold substrates; Figure 1c,d. are Nyquist plots (Zj171 vs Zre) for the Faradaic impedance measurements in the presence of 4 mM [Fe(CN)6]374" in 0.1 M phosphate buffer (pH 7.0) containing 0.3 M NaCIO4 at (c) flat gold surfaces with DNA probe modification (d) gold nanoparticle deposited gold surface with DNA probe modification. The experimental data points are shown as, (•) bare electrodes, (A) after the GC-rich ds-DNA modified electrodes and (■) after incubation with 1 μM binding drug nogalamycin for 30 min. The solid lines are the fitted curves to the equivalent circuit Figure 2 using the ZSimpWin software. The standard Randies circuit was used for the electrodes before DNA monolayer modification.
(015) Figure 1 c shows the results of impedance spectroscopy on bare gold, bare gold modified with GC-rich DNA probe, and after immersion for 30 min
with 5 μl_ of 1 μM nogalamycin. Electrochemical deposition of gold nanoparticles on the naked gold surface results in the expected straight line (Figure 1d), a characteristic of a diffusion-limiting step. The data fitted by the conventional Randies circuit are presented as the inset of Figure 1d. The charge-transfer resistance Ret of the gold nanoparticle deposited surface is 3.45(± 0.3) Ω.cm"2, noticeably smaller than the naked surface, 13.5(± 0.2) Ω.cnrϊ2. On either a flat gold surface or the gold particle deposited surface, the charge repulsion between the negatively charged phosphate backbone of DNA on the gold surface and the [Fe(CN)6]374" probe was observed as enlarged semicircles in the Nyquist curve, corresponding to a reduced ability for electron transfer on the electrode surfaces (Figure 1c,d). The resistance of the monolayer RSAM was estimated as 4.6 (± 0.3)x102 and 4.1 (± 0.2)x102 Ω.cm"2 for the flat gold surface and the gold nanoparticle deposited surface, respectively. The impedance spectra recorded after intercalation of nogalamycin with the surface-immobilized ds-DNA showing the monolayer resistance was increased to 7.6 (± 0.5) x102 Ω.cm"2, i.e., ΔRSAM = 3.0 (± 0.5)x102 Ω.cm"2 (Table 1), indicating the intercalation of nogalamycin with DNA and insulation of the conductive support. This behavior might be attributed to the increase in the duplex length and unwinding of the helix upon drug interaction11 resulting in the increased charge transfer resistance of DNA on the surface. The change in the real part of the monolayer resistance (ΔRSAM ) of the DNA modified bare gold electrode was 3.0 (± 0.5)x102 Ω.cm"2 compared to 8.8 (± 0.3)x102 Ω.cm'2 determined for the gold nanoparticle deposited surface modified with DNA. Such a result indicated that the changes of the impedance signal were remarkably expanded using gold nanoparticle deposited surfaces. The modified electrode can detect nogalamycin as low as 5 nM, i.e., ~40-fold more sensitive than the flat gold surface (-200 nM). This substantial enhancement is probably attributed to a higher binding efficiency in the capturing step for drug molecules due to the DNA concentric orientation on gold nanoparticles. The use of gold
(016) nanoparticles also provides a large quantity of probe DNA to facilitate the binding kinetics between the drugs and DNA probes.
(017) Table 1. The detection sensitivity for different drugs using flat gold electrodes and gold nanoparticle deposited gold electrodes. ΔRSAM (X102 Ω.cm"2) is the change of the charge transfer resistance before and after drugs interaction.
(018) We have also investigated two DNA specific sequence binding drugs, mithramycin and netropsin, which are important antitumor antibiotics to inhibit DNA transcription and replication in vivo by interaction to template DNA in the minor groove binding mode with GC or AT base specificity.7 The resistance of the charge transfer on the DNA modified surface increases with increasing drug concentration from 15 nM to 1 μM and from 40 nM to 1 μM for
mithramycin and netropsin, respectively. Gold nanoparticle arrays deposited on gold electrodes gave ~20-30-fold higher detection limits, compared to the bare gold surface. The sequence selectivity of the two drugs for AT-rich DNA and GC-rich DNA was excellent, as evidenced by little discernible signal change from the control electrodes modified with probe DNA which sequences are not their base binding specificities. The observed changes in the Nyquist curves due to the drug interaction in minor groove-binding mode are significantly smaller than that in the intercalative binding mode. Such alterations in impedance behavior are conceivable since only minor distortions of the B-helix were observed due to the interaction of small minor groove- binding drugs.11
(019) To summarize, the electrochemical deposition of gold nanoparticles on flat gold electrode surfaces results in higher amounts of bound DNA probes, and in turn a significant improvement in the detection sensitivity for the assay of interfacial drugs.
(020) Figure 3 is a schematic illustration, comparing a DNA probe binding to drug molecules (A) without gold nanoparticles and (B) with gold nanoparticles. Arrays of deposited gold nanoparticles on gold substrates may be envisioned to offer a selective sensing interface for applications in drug monitoring and in the development of tools to study bimolecular interactions.
Supporting Information Chemical and Reagents
(021) All reagents were of analytical grade. MiIIi-Q water was used in all experiments.
(022) DNA synthesis. Oligonucleotides were synthesized using standard solid phase phosphoramidite chemistry (Wincoff, F.; et al. Nucleic Acids Res. 1995, 25, 2677-2684) which is building a molecule off of a solid support-CPG
(controlled pore glass). The synthesis was carried out on a fully automated Beckman 1000M DNA Synthesizer (Fullerton, CA) at the Plant Biotechnology Institute, National Research Council Canada (PBI-NRC, Saskatoon, Saskatchewan). The two columns (200 nm and 1000 nm) were obtained from Beckman. After synthesis, concentrated ammonium hydroxide was used to remove the DNA from the CPG. The 5'-terminal amine group in the deprotected oligonucleotide was further derivatized with dithothreitol (Glen Research, Sterling, VA) to produce 5'-dithiol terminated oligonucleotides (DeIa T. Nucleosides Nucleotides 1993, 12(9), 993-1005). The oligonucleotides were purified by two-step reversed-phase HPLC using a C18 column which is based on hydrophobicity interaction. An acetonitrile/triethylamine acetate (TEAA) (pH 7.4) solution was used as the purification solvent. Finally the synthesized oligonucleotides were characterized by UV and matrix-assisted laser desorption ionization time of flight mass spectroscopy (MALDI-TOFMS, xxxxxxxx).
(023) Drug solution preparation. Nogalamycin, mithramycin and netropsin were obtained from Sigma. Nogalamycin was dissolved in 75% DMSO (with sonication) and diluted into 100 mM phosphate buffer, pH 7.2. Mithramycin and netropsin were dissolved in water. With DMSO as a solvent, the final concentration did not exceed 2% in the assay. Concentrations of nogalamycin were determined spectroscopically. ε258 = 24,755 M'1 cm"1 in buffer (Merck Index). The concentration of mithramycin and netropsin was determined from E400 = 10,000 M"1 cm'1 (Sarker, M.; Chen, F-M. Biochemistry 1989, 28, 6651) and ε296 = 20,200 M'1 cm"1 (Sigma), respectively . Stock solutions were stored for short periods at -20 0C. The divalent metal Mg2+, has been reported is essential for strong association for drug-DNA to occur. Thus, the reaction solution for drug-DNA binding was prepared in a 20 mM phosphate buffer, pH 7.2, containing 5 mM Mg2+ at room temperature. All binding reactions with DNA were carried out in dark using a cap of the electrode.
Electrochemical Instrumentation and Measurements
(024) The impedance measurements are based on the charge transfer kinetics of the [Fe(CN)6]374" redox couple. Compared to bare gold surfaces, the immobilization of DNA and then the DNA-drug interaction on electrode surfaces altered the capacitance and the interfacial electron resistance, and thus diminished the charge transfer kinetics by reducing the active area of the electrode or by preventing the redox species from approaching the electrode.
Impedance Data Analysis
(025) The experimental data of impedance was analyzed by using the model equivalent circuit Rdiff(QRsAM(RctW)) as shown in Figure 2. More specifically, it comprises a solution resistor Rditf, a charge transfer resistance Rct, a surface capacitance Q, the impedance due to mass transfer of the redox species to the electrode described by the conventional Warburg and a combination of one parallel element RSAM, which gives information on the monolayer resistance. For precise modeling of the impedance of DNA-SAM modified electrode systems, the spectra in the form of Nyquist plots were analyzed with a modified Randies circuit, including an additional component in parallel with the Faradaic impedance of the interface, RSAM, of the monolayer-dependent resistance.10
(026) Figure 2 is a general equivalent circuit model used for impedance data analysis: W. E., working electrode; R.E., Ag/AgCI, 3M NaCI reference electrode; C. E., platinum counter electrode. The dotted box enclosed the traditional Randies circuit and was only used for the gold electrodes before DNA immobilization. The extra monolayer electron-transfer resistance RSAM of the modified circuit was used for DNA and the DNA-drug interaction system.
(027) Cyclic voltammetry, linear sweep voltammetry and AC impedance analysis were performed with an EG&G potentiostat (Perkin Elmer, formally EG/G, Princeton Applied Research, Model 6310, Oak Ridge, TN). All electrolyte solutions were purged for 20 min. in argon prior to the measurement, and a blanket of argon was maintained over the solutions during the measurement. A three-electrode cell, consisting of a modified gold electrode as working electrode, a Ag/AgCI/3 M NaCI reference electrode (Bioanalytical System, West Lafayette, IN), and a platinum wire auxiliary electrode, was used in a grounded Faraday cage. All potentials are poised vs Ag/AgCI/3 M NaCI. Electrochemical deposition of gold nanoparticles was carried out in 0.1 M NaOH solution containing 0.1 M HAuCI4 using chronoamperometry. The solution was thoroughly purged with argon before each experiment. The electrode potential is stepped from an initial potential Ei = 0.8 V vs. Ag/AgCI/3 M NaCI where no reaction occurs to a potential Er = 0.20 V and AuCI4 ' is reduced to Au nanocrystals (Figure 4). The roughness factor of the bare electrode was estimated by the reduction peak current (the charge obtained by integrating the reduction peak area) of an oxidized gold layer formed by a potential step from 0.0 to 1.8 V in 0.5 M H2SO4 for 5 s. After 10 s of deposition, the roughness factor of the gold nanopartile modified electrode was estimated to be 1.48-1.65 by dividing the theoretical value for the reduction of a layer of divalent oxygen from the surface (Oesch, U.; Janata, J. Electrochim. Acta 1983, 28, 1247). The reduction peaks for the gold nanoparticle deposited gold surface show two peaks in shoulder (Figure 5), corresponding to the different reduction abilities on the oxidized gold nanoparticles and on the oxidized flat gold surface.
(028) The data was measured and collected for 31 harmonic frequencies from 0.1 Hz to 100 kHz at 5 steps/decade, then was analyzed using ZSimpWin software (Princeton Applied Research). The experimental data of impedance was analyzed by the fitting procedure using the model equivalent circuit Rdiff(QRsAM(RctW)). It comprises a solution resistor Rdjff, a charge
transfer resistance Rct, a surface capacitance Q, the impedance due to mass transfer of the redox species to the electrode described by conventional Warburg and a combination of one parallel element RSAM, which give information on the resistance of monolayer.
Cleaning of Gold Electrodes
(029) Gold disk electrodes with a polypropylene body (CH Instruments, Austin, TX, 2.0 mm diameter, ca. 3.14 mm2 geometrical area, roughness coefficients between 1.1 and 1.3) were used for the electrochemical measurement. Before modification, the bulk gold electrodes were first lightly polished with 0.05 μm alumina (Buhler), and ultrasonicated for 10 min in ethanol. These electrodes were immersed in boiling 2 M KOH (in 30 % ethanol for 30 min followed by sonication in concentrated nitric acid for 30 min, then sonicated with MiIi-Q water. The gold electrodes were then quickly rinsed with water and immediately used for modification.
(030) Silicon supported gold films for AFM analysis were purchased from Platypus Technologies LLC (Madison, Wl). The topical thickness of Au was ca. 1000 A. Before each experiment the electrodes were cleaned by electrochemical treatment. Electrochemical treatment was performed in the cell described below, by cycling from a potential of - 0.1 to +1.25 V vs Ag/AgCI in 0.5 M H2SO4 solution until a stable gold oxidation peak at +1.1 V versus Ag/AgCI was obtained. The prepared surfaces were immediately used for modification and for AFM measurements.
DNA Modification on Electrode Surface
(031) Duplex DNA was hybridized in 20 mM Tris-CIO4) pH 7.2 by a mixture of the equal amount of the oligo and its complementary strand. The mixture of DNA solution was heated to 50 0C for 5 min and then cooled slowly to room
temperature overnight. The freshly cleaned electrode surfaces were immersed with a drop of 0.4 mM prehybridized ds-DNA in 0.1 M TrJs-CIO4 buffer containing 0.5 M NaCIO4 over 2 days. The modified electrodes were repeatedly rinsed with Tris-CIO4 buffer (20 mM, pH 7.2) for 1 min before measurement. AFM Measurements
(032) Atomic force microscopy (AFM) measurements were used a Nanoscope IV system ( Digital Instruments-Veeco, Santa Barbara, CA) operating in air in tapping mode (TM). Silicon AFM probe with a cantilever length of 125 mm and a drive frequency of 235-255 k Hz were employed for image measurement. The image scan speed was 0.50 Hz at 512 lines per scan. All height measurements and size distribution were obtained from the analysis software package with the instrument (Figure 6).
(033) Electrochemical deposition of gold nanoparticles on gold electrodes was performed in 0.5 M H2 SO4 solution containing 10 mM HAuCI4 using chronoamperometry at a constant potential. Cyclic voltammethc measurements for gold electrodes in HAuCI4 solution revealed that reduction of AuCI4 - occurs at a peak potential of +0.15 V. The electrode potential is therefore stepped from an initial potential E1 = 0.8 V, where no reaction occurs, to a overreduction potential Ex = 0.20 V, at which potential AuCI4 - is fully reduced to Au nanocrystals. The roughness factor of the bare electrode was estimated by the reduction peak current (the charge obtained by integrating the reduction peak area) of an oxidized gold layer formed by a potential step from 0.0 to 1.8 V in 0.5 M H2SO4 for 5 s .9 By varying the deposition time at a constant applied potential, gold nanoparticles with different sizes and densities were deposited onto gold electrodes.
Characterization of Nanoparticle Modified Substrates
(034) AFM was used to probe the surface morphology of the electrodes and characterize the shape, size and number density of the gold nanoparticles. Figure 7 compares AFM micrographs of the gold substrates electrochemical deposited by gold nanoparticles with different periods of deposition time. Evidently, the AFM micrographs shown in Figure 7b reveal uniform, cubic with spherical nanocrystals grown directly on the bare gold substrates. Counting analysis of gold surfaces following a 10 s deposition yields an average nanoparticle density of 1.59x107/cm2 with an average size of 20-40 nm. In contrast, gold nanoparticles deposited on the substrates with deposition time of 3 s (Figure 7a), have average particles size in the range of 10-30 nm with a lower number density of 6.5x105 /cm2. As the deposition time is increased to 30 s, a number of aggregated Au crystallites with micrometer-size are visible (Figure 7c). When the deposition time reaches 60 s, the number of the micrometer-size gold clusters increases with a concomitant decrease in the number density of nanometer-size gold particles. The big size gold clusters lead to a very rough electrode surface. The roughness factors increased from 1.33-1.42, 1.48-1.65, 1.61 -1.71 and 1.90, respectively, with increasing electrochemical deposition time from 3, 10, 30 and 60 s. The reduction peaks for the gold nanoparticle deposited gold surface show two peaks differing in the reduction potentials, corresponding to the different reduction abilities of the oxidized gold nanoparticles and the oxidized flat gold surface.δa
Functionalization of Gold Nanoparticle Modified Substrates with DNA.
(035) The resulting gold substrates were functionalized with the pre- hybridized duplex DNA through the Au-S bond. Incubation for 48 h was required to attain high density packed, well-oriented DNA layers on these gold substrates. The gold nanoparticle modified substrates were used as the sensing devices for electrochemical analysis of DNA binding drugs. The
sensing sensitivity of the DNA binding drugs can be controlled by controlling the size and density of gold nanoparticles via the time employed for the gold nanoparticle deposition. Of all the impedance studies using the modified electrodes with different density of gold particles, the electrodes modified by 10 s electrochemical deposition, which lead to better evenly dispersed gold particles with relatively uniform size (20-80 nm) are most effective in enhancing the sensitivity for DNA binding drug analysis. This might be attributed to the fact that evenly dispersed particles are capable of carrying a larger number of DNA probes and the orientated DNA layers are able to provide more efficient space for drug binding processes.
(036) Nogalamycin is an antitumor anthracycline with an intercalation-binding mode whereby the two end groups align one in each groove of the DNA duplex while the central drug chromophore is intercalated between adjacent base pairs. After intercalation of nogalamycin with the surface-immobilized ds- DNA, the monolayer resistance, which is the most significant change among all the elements, increased to 7.6 (± 0.5) x102 Ω cm2, i.e., ΔRSAM = 3.0 (± 0.5)x102 Ω cm2, indicating the intercalation of nogalamycin with DNA and insulation of the conductive support. This behavior might be attributed to the increase in the duplex length and unwinding of the helix upon drug interaction11 resulting in the increased charge transfer resistance of DNA on the surface. These effects are connected with changes of the DNA hydration shell and with release of the double helix-condensed cations. The alteration in the real part of the monolayer resistance (ΔRSAM) due to the drug interaction of DNA modified bare gold electrode was 3.0 (± 0.5)x102 Ω cm2 compared to 8.8 (± 0.3)x102 Ω cm2 determined for the gold nanoparticle deposited surface (Figure 8). The result indicated that the changes of the impedance signal were remarkably expanded using gold nanoparticle deposited electrodes. Figure 9 shows alteration of the monolayer resistance upon the treatment of the sensing interface with different concentrations of the analyte drug- nogalamycin. As the bulk concentration of nogalamycin increases, the
electron-transfer resistance exponentially increases on either the bare gold electrode or the gold nanoparticle modified gold electrode, implying that a higher content of the analyte is linked to the modified surface. When the drug concentration was higher than 600 (±100) nM, the monolayer electron transfer resistance was no longer sensitive to further increases in the nogalamycin concentration. The gold nanoparticle modified electrode can detect nogalamycin as low as 5 nM, i.e., ~ 40-fold more sensitive than the flat gold surface (~ 200 nM). This substantial enhancement is probably attributed to a higher binding efficiency in the capturing step for drug molecules due to the concentric orientation of the DNA monolayer on the gold nanoparticle interfaces. The use of gold nanoparticles also provides a large quantity of probe DNA to facilitate the binding kinetics between the drugs and DNA probes. The effect of the sequence of DNA on the interfacial binding of the intercalator nogalamycin by using the AT-rich DNA modified electrodes was also tested. Due to the non-specific sequence binding properties of the drug, the alteration of DNA sequence should not affect the binding behaviors of the drug, and indeed very similar electrochemical responses were observed compared to the GC-rich DNA modified electrodes.
Specificity of Impedance Analysis
(037) Specificities of the analysis system were also investigated by the use of two different drugs that bind to different specific sequences of the DNA probes. The strategy involves the use of the GC-rich DNA modified electrodes and/or AT-rich DNA modified electrodes to analyze DNA specific binding drugs. We have investigated two DNA specific binding drugs, mithramycin and netropsin, which are important antitumor antibiotics to inhibit DNA transcription and replication in vivo by interaction to template DNA in the minor groove binding mode with GC or AT base specificity.13 Mithramycin binds as a Mg2+-coordinated dimer within the wide and shallow minor groove of the (GC)4 tract. The positioning of the two oligosaccharide chains of mithramycin in the minor groove determines its GC sequence specificity.
Conversely, netropsin fits in the narrow and deep minor groove of the (AT)4 tract. Both the local geometry and the negative electrostatic potential in the minor groove favor the AT specificity of netrospin. Notwithstanding their opposite preferences, mithramycin and netropsin both bind selectively to the minor groove of DNA sequences through a combination of close van der Waals contact, hydrogen bonding, electrostatic attraction and hydrophobic interaction. Figure 8 shows the representative columns of the alteration of monolayer electron transfer resistance due to the drug-DNA interaction. After binding of mithramycin and netropsin on gold nanoparticle modified surfaces, increases in the monolayer electron transfer resistance (ΔRSAM) were 4.8 (±0.6)x102 and 5.5 (± 0.3) x102 Ω cm2, respectively. Such increases are significant compared to 1.9 (±0.2) x102 and 2.4 (±0.3) x102 Ω cm2, obtained with the bare electrodes (Figure 8). The observed changes in the Nyquist curves due to the drug interaction in minor groove-binding mode are significantly smaller than that in the intercalative binding mode. Such alterations in the impedance behavior are conceivable since only minor distortions of the B-helix were observed due to the interaction of small minor groove-binding drugs.11
(038) For quantitative comparison, we measured the electron transfer resistances as a function of the antibiotics concentration on selected specific binding DNA probe modified electrodes. The charge transfer resistance on the gold nanoparticle modified surfaces increases with increasing drug concentration from 15 nM to 1 μM and from 40 nM to 1 μM for mithramycin and netropsin (Figure 10), whereas the detection limits of mithramycin and netropsin using bare gold electrodes were 300 nM and 500 nM, respectively, indicating significantly enhanced sensitivity in the detection limit by using the gold nanoparticle deposited electrodes. Gold nanoparticle arrays deposited on gold electrodes gave ~15-20-fold lower detection limits, compared to the bare gold surface. In contrast, gold nanoparticle modifications have the same qualitative effect on the major groove binding as they do on the binding of
nogalamysin that binds with DNA in the intercalation mode. The sequence selectivity of the two drugs for AT-rich DNA and GC-rich DNA was excellent, as evidenced by little discernible resistance change from the control electrodes modified with probe DNA which sequences are not their base binding specificities (Figure 10).
Effect of Changes in Concentration:
(039) We also did control experiments using various concentration of HAuCI4 solution. The gold particle size was found to increase with increasing the concentration of HAuCI4. However, the number of enhancement folds for sensitivity decreased. For example, when the concentration was increased to 1 M, the whole electrode surface was covered with a fresh gold layer, and the sensitivity was only enhanced by 1 fold. 0.5 M HAuCI4 gave a 25 fold increase in sensitivity and 0.05 M HAuCI4 enhanced by 12 fold. We found at 0.1 M HAuCI4 condition, the average size of gold nanoparticles are 20-40nm, which will enhance sensitivity most (it should be mentioned, in this case, the DNA is 20 bps. For different length of DNA different size of particles may work better.
Effect of Changes in the Applied Potential:
(040) Changes in the applied potential will generate different size of gold nanoparticles. In our case, 0.2 V made gold nanoparticles of 20-40 nm size, and a 40 fold increase in sensitivity. 0.15 V applied potential gave gold nanoparticles of >50nm size, and only 10 fold increase in sensitivity. 0.25 V gave gold nanoparticles of 5-15 nm size and better response (25 fold). Thus, the 0.20 V was the best (40 fold).
REFERENCES
Lahav, M.; Gabai, R.; Shipway, A.N.; Willner, I. Chem. Comm. 1999, 19, 1937-1938. 1. (b) Hrapovic, S.; Liu, Y.; Male, K.B.; Luong, J. HT. Anal, Chem. 2004, 76, 1083-1088 (c) Wang, J.; Musameh, M.; Lin, Y.H. J. Am.Chem. Soc. 2003, 125, 2408-2409. (d) Zhao, J.G.; Odaly, J.P.; Henkens, R.W.; Stonehuerner, J.; Crumbliss, A. L. Biosens. Bioelectron. 1996, 11, 493- 502.
2. (a) Haq, I.; Ladbury, H. E.; Chowdhry, B. Z.; Jenkins, T.C.; Chaires, J.B. J. MoI. Biol. 1997, 271, 244-257. (b) Gavathiotis, E.; Sharman, G.J.; Searle, M.S. Nucleic Acids Res. 2000, 28, 728-735. (c) Colson, P.; Bailly, C; Houssier, C. Biophys. Chem. 1996, 58, 125-140. (d) Krautbauer, R.; Pope, LH.; Schrader, T.E.; Allen, S.; Gaub, H.E. FEBS Lett. 2002, 510, 154-158.
3. (a) Brabec, V. Electrochimica Acta 2000, 45, 2929-2932. (b) Erdem, A.; Ozsoz, M. Electroanal. 2002, 14, 965-974. (c) Hason S.; Dvorak J.; Jelen, F.; and Vetterl V., Crit. Rev. Anal. Chem. 2002, 32, 167-179. (d) Wang, J.; Ozsoz, M.; Cai, X.; Rivas, G.; Shiraishi, H.; Grant, D. H.; Chicharro, M.; Fernandes, J.; Palecek, E. Bioelect. Bioener. 1998, 45, 33-40.
4. (a) Yan, F.; Sadik, A. O. J. Am. Chem Soc. 2001 , 123, 11335-11340. (b) Patolsky, F.; Lichtenstein, A.; Willner, I. J. Am. Chem. Soc. 2001 , 123, 5194-5205. (c) Li, C.Z., Long, Y.T., Lee, J.S., Kraatz, H.B. Chem.Comm. 2004, 5, 574-575.
5. (a) Finklea, H. O. Electroanalytical Chemistry. Marcel Dekker Inc., New York. 1996. (b) Bard, A. J.; Faulkner, L. R. Electrochemical method: fundamentals and applications; 2nd ed.; John Wiley: New York, 2001.
6. (a) Niemeyer, CM. Angew. Chem. Int. Ed. 2001 , 40, 4128-4158. (b) Shipway, A.N.; Willner, I. Chem. Comm. 2001 , 20, 2035-2045. (c) Nam, J. M.; Thaxton, C.S.; Mirkin, CA. Science 2003, 301, 1884-1886. (d) Wang, M.J. ; Sun, C.Y.; Wang, LY.; Ji, X.H.; Bai, Y.B.; Li, T.J.; Li, J.H. J. PharmaceutBiomed. 2003, 33, 1117-1125. (e) Fritzsche, W.; Taton, TA Nanotechnology 2003, 14, R63-R64.
7. (a) Fox, K.R.; Waring, M.J. Biochemistry 1986, 25, 4349-4356. (b) Keniry, M.A.; Banvilie, D. L; Simmonds, P.M.; Shafer, R. J. MoI. Biol. 1993, 231, 753-767. (c) Coll, M.; Aymami, J.; van der Marel, G.A.; van Boom, J. H.; Rich, A.; Wang, A.H.J. Biochemistry 1989, 28, 310-320.
8. (a) Boxley, CJ.; White, H.S.; Lister, T.E.; Pinhero, P.J. J. Phys.Chem. /3 2003, 107, 451-458. (b) Liu, H.; Penner, R.M. J. Phys.Chem. B 2000, 104, 9131 -9139. (c) Hwang, B.J.; Santhanam, R.; Lin, Y.L Electroanal. 2003, 15, 1667-1676.
9. (a) Rand, D.A.J. ; Woods, R. J. Electroanal. Chem. 1971, 31, 29-38. (b) Finot, M.O.; Braybrook, G. D.; McDermott, M.T. J. Electroanal. Chem. 1999, 466, 234-241.
10. (a) Janek, R.P.; Fawcett, W. R. Langmuir 1998, 14, 3011-3018. (b) Li1 CZ.; Long, Y.T.; Lee, J. S.; Kraatz, H.-B. J. Phys.Chem. B 2003, 107, 2291 - 2296. (c) Long, Y.T.; Li, CZ.; Kraatz, H.-B.; Lee, J.S. Biophysics. J. 2003, 84, 3218-3225.
11. (a) Tessmer, I.; Baumann, CG.; Skinner, G. M.; Molloy, J. E.; Hoggett, J.G.; Tendler, S.J.B.; Allen, S. J. Mod. Optic. 2003, 50, 1627-1636. (b) Coury, J. E.; McFail-lsom, L.; Dean-Williams, L.; Bottomley, L.A. Proc. Nat. Acad. Sci. USA 1996, 93, 12283-12286. (c) Kopka, M.L.; Yoon, C; Goodsell, D.; Pjura, P.; Dickerson, R.E. Proc. Nat. Acad. Sci. USA 1985, 82, 1376-1380.
12. (a) Resch, R; Meltzer, S.; Valiant, T.; Hoffmann, H.; Koel, B.E.; Madhukar A.; Requicha, A.A.G.; Will, P. Langmuir 2001 , 17 5666-5670. (b) You, T.; Niwa, O.; Tomita, M.; Hirono, S. Anal. Chem. 2003, 75,2080-2085. (c) Sarkar, D.K.;Zhou, XJ.; Tannous, A.; Leung, K.T. J. Phys. Chem. S2003, 107, 2879-2881. (d) Liu, Y.; Keith, B. M.; Bouvrette, P.; Luong, J. H.T. Chem.Matθr. 2003, 15, 4172-4180.
13. Aich, P.; Dasgupta, D. Biochem. Biophys. Res. Commun. 1990, 173, 689-696. (b) Sarker, M.; Chen, F.-M. Biochemistry 1989, 28, 6651 -6657. (c) Keniry, M.A.; Banville, D. L.; Simmonds, P.M.; Shafer, R. J. MoI. Biol. 1993, 231 , 753-767. (d) Coll, M.; Aymami, J.; van der Marel, G.A.; van Boom, J. H.; Rich, A.; Wang, A.H. J. .Biochemistry 1989, 28, 310-320.