Label-Free Sensitive Electrogenerated Chemiluminescence

Aug 21, 2014 - Label-Free Sensitive Electrogenerated Chemiluminescence Aptasensing Based on Chitosan/Ru(bpy)32+/Silica Nanoparticles Modified Electrod...
1 downloads 10 Views 2MB Size
Download PDF
Article pubs.acs.org/ac

Label-Free Sensitive Electrogenerated Chemiluminescence Aptasensing Based on Chitosan/Ru(bpy)32+/Silica Nanoparticles Modified Electrode Jie Dang, Zhihui Guo, and Xingwang Zheng* Key Laboratory of Analytical Chemistry for Life Science of Shaanxi Province, School of Chemistry and Chemical Engineering, Shaanxi Normal University, Xi’an 710062, P.R. China S Supporting Information *

ABSTRACT: In this work, a label-free and sensitive electrogenerated chemiluminescence (ECL) aptasensing scheme for K+ was developed based on G-rich DNA aptamer and chitosan/Ru(bpy)32+/silica (CRuS) nanoparticles (NPs)-modified glass carbon electrode. This ECL aptasensing approach has benefited from the observation that the Grich DNA aptamer at the unfolded state showed more ECL enhancing signal at CRuS NPs-modified electrode than the binding state with K+, which folds into G-quadruplex structure. As such, the decreasing ECL signals could be used to detect K+. Compared to other aptasensing K+ approaches previously reported, the proposed ECL sensing scheme is a label-free aptasensing strategy, which eliminates the labeling, separation, and immobilization steps, and behaves in a simple, low-cost way. More importantly, because the proposed ECL sensing mechanism utilizes the nanosized ECL active CRuS NPs to sense the nanoscale conformation change from the aptamer binding to target, it is specific. In addition, due to the great conformation changes of the aptamer’s G-bases on CRuS NPs and the excellent ECL enhancing effect of guanine bases to the Ru(bpy)32+ ECL reaction, a 0.3 nM detection limit for K+ was achieved with the proposed ECL method. On the basis of these advantages, the proposed ECL aptasensing method was also successfully used to detect K+ in colorectal cancer cells. fluorescence, UV−visible absorbance, and chemiluminescence detection systems using fluorescent dyes,14 gold nanoparticles,15 and DNAzymes16 as signal reporters. As we know, electrogenerated chemiluminescence (ECL) methods have potential advantages over other analytical systems in that they are a low background signal, being easily controlled, with wide dynamic range, simple instrumentation, and a low detection limit, and several ECL aptasensors for proteins and small molecules have been reported by using tris(2,2-bipyridyl)ruthenium(II) and its derivatives, Ru-doped silica nanoparticles, as signal reporters.17−19 However, these ECL aptasensors still require labeling both ECL reporters and thiol groups on aptamers, and the disadvantages of the labelbased aptasensors (mentioned above) is not avoided. Thus, it is still necessary to fabricate a label-free and sensitive ECL aptasensing method. Herein, a label-free and sensitive ECL method for the aptasensing platform is first developed using K+-binding DNA aptamer as the model. In this ECL aptasensing platform, the nanoparticles (NPs), which are composed of silica, chitosan, and Ru(bpy)32+, were synthesized and, further, were selfassembled on the CNT/Nafion coating film electrode based on

A

ptamers are single-stranded DNA or RNA molecules, isolated from large pools of random-sequence oligonucleotides, which are capable of binding a wide range of chemical or biological entities with high affinity and specificity. As such, aptasensors, which use aptamers as biorecognition elements to fabricate chemical or biosensors, have appeared as promising devices for selective and high-efficiency detection of a variety of analytes such as small ions, proteins, and even the whole cells.1,2 In particular, compared to the conventional bioreceptors such as antibodies or enzymes, aptamers are much smaller in size, more chemically stable, and easily synthesized;3 thus, many aptamer-based sensors (aptasensors) based on electrochemical, chemiluminescence, fluorescence, colorimetric, and mass-sensitive transduction modes have been extensively investigated.4−10 Among these methods reported, some processes such as tag labeling on aptamers, target separation from the sample matrix, and aptamer immobilization onto solid substrates are always required. These steps not only are time-consuming and of high cost but also affect the binding affinities of aptamers to targets. For overcoming these limits, great efforts have been recently devoted to the development of label-free aptasensors,11 in which the whole measurements can be performed in a homogeneous solution, and the labeling, separation, and immobilization steps are effectively eliminated.12,13 Currently, most of these label-free aptasensors are mainly based on © 2014 American Chemical Society

Received: October 16, 2013 Accepted: August 21, 2014 Published: August 21, 2014 8943

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950

Analytical Chemistry

Article

Science and Technology Development Co., Ltd.), which was used to prepare the Tris-HCl buffer for aptamer binding to K+, was purified three times at 170 °C with sublimation under vacuum according to the previously reported method.23 The 100 mL polytetrafluoroethene volumetric flask was also used to prepare the K+ stock solution or other reagent solutions to avoid trace K+ permeate into solution from glass vessels. In addition to these, all other reagents were of analytical grade, and the aqueous solutions were prepared with the highly pure water (sterile Minipore water, 18.3 MΩ). Apparatus. Cyclic voltammetric experiments were performed by using a CHI420b electrochemical working station (Chenhua Inc., China). All experiments were carried out with a conventional three-electrode system in a 10 mL electrochemical cell. A glassy carbon electrode (GCE) was used as the basic working electrode, and a platinum sheet was used as the counter electrode. All the potentials quoted here were relative to a homemade Ag/AgCl (saturated KCl) reference electrode, which was designed to avoid the interference of the K+ from the reference electrode to K+ assay, and its detailed structure is shown in Figure S1 of the Supporting Information (SI). The ECL signal produced in the electrolytic cell was detected and recorded by a chemiluminescence analyzer (IFFD, Xi’an Remax Electronic Science Tech. Co. Ltd. China), and the photomultiplier tube (PMT) was operated in current mode. Unless noted otherwise, the PMT was biased at −800 V. Circular dichroism (CD) spectra of DNA were recorded at room temperature on a Jasco J-815 spectropolarimeter. A multiposition magnetic stirrer (IKA, Germany) and a high-speed centrifuge (5804 R, Eppendorf) were used to stir the synthesizing CRuS NPs and separate nanoparticles, respectively. The fluorescence spectrum was recorded with an F-7000 spectrofluorometer (Hitachi, Japan). A transmission electron microscopy (TEM) image of the nanoparticles was obtained by a JEM-2100 TEM (Hitachi, Japan). UV−visible adsorption spectra were recorded on a UV−vis spectrophotometer (TU1901, China). Synthesis of CRuS NPs. The W/O reverse microemulsion system was prepared first by mixing 1.77 mL of TX-100, 7.5 mL of cyclohexane, 1.8 mL of 1-hexanol, and 300 μL of water. Then 50 μL 0.01 M Ru(bpy)32+ aqueous solution and 100 μL 0.5% chitosan were added into the mixture. In the presence of 90 μL of TEOS, a polymerization reaction was initiated by adding 60 μL of NH4OH (25%−28%). The hydrolysis reaction was allowed to continue for 24 h. Acetone was then added to destroy the emulsion, followed by centrifuging and washing with ethanol and water. At last, the orange CRuS NPs were obtained. The synthesized CRuS NPs were characterized by spectrofluorometry and transmission electron microscope (TEM) for size and morphology. Preparation of CRuS NPs-CNT/Nafion-Modified Electrodes. A glassy carbon electrode (GCE) 4 mm in diameter was polished with 0.3 and 0.05 μm alumina, followed by successive sonication in distilled water and ethanol for 5 min, and than was dried in air. Afterward, 1.5 mg CNT were dispersed in 3.0 mL of 0.05% Nafion solution, and then the mixture of the CNT with Nafion were ultrasonicated in a water bath to homogeneously distribute the nanotubes. Following that, 10 μL of the mixture solution was drop-coated on the surface of the pretreated GCE and dried for about 30 min at room temperature to form a CNT/Nafion film-modified GCE. After being washed with distilled water, the modified electrode was inserted into 1000 μL CRuS NPs solution of 0.1 M NaAc−

the stronger electrostatic interactions between the positive charged chitosan embedded in the NPs and negative charged Nafion on the electrode. In this label-free sensing platform, ECL-active CRuS NPs are used as sensing indicators to report the conformational change of an aptamer upon recognition of K+. While the DNA aptamers with G-rich sequences are in their free state, they can be strongly adsorbed on the CRuS NPs surface based on the strong affinity of the chitosan to the DNA aptamer20 and can produce the stronger ECL signal due to the excellent ECL enhancing activity of the guanine bases in the aptamer to Ru(bpy)32+ inside the NPs.21,22 In contrast, while the K+ is present, the K+ can promote the conversion of a Grich DNA aptamer sequence from a loose random coil into a compact G-quadruplex via intramolecular hydrogen-bonding interactions. In this case, since most the guanine bases were folded into the interior of this G-quadruplex, the adsorbed state G-quadruplex on the chitosan/Ru(bpy)32+/silica (CRuS) nanoparticles (NPs) can only produce the weak ECL signal. On the basis of these findings, a sensitive label-free ECL method for K+ was developed. Compared to the previous aptasensing scheme, the proposed aptasensing platform owns the following advantages. First, the proposed ECL sensing method is simple and low cost because it is a label-free aptasensing strategy, which eliminates the labeling, separation, and immobilization steps. Second, because the proposed ECL sensing mechanism utilizes the nanosized ECL active CRuS NPs to sense the nanoscale conformation change from aptamer binding to the target, it offers the higher specificity to measure the conformation change of aptamer binding K+, which paved the basis of the higher specific feature of the proposed method. Third, due to the great conformation changes of aptamer’s Gbases on CRuS NPs and the excellent ECL enhancing effect of guanine bases to Ru(bpy)32+ ECL reaction, the proposed ECL aptasensing method presents the higher sensitivity for K+.

■

EXPERIMENTAL SECTION Materials and Reagents. Triton X-100, chitosan, tetraethylorthosilicate (TEOS, 99%), Tris-(2,2′-bipyridyl)dichlororuthenium(II) hexahydrate and Nafion 117 (∼5% in a mixture of lower aliphatic alcohols and water) were purchased from Sigma-Aldrich; n-hexanol was purchased from Tianjin Chemical Reagent Factory, cyclohexane and ammonium hydroxide (25−28 wt %), acetone, and ethanol were provided by Xian Chemical Reagent Factory. The potassium ion-specific oligonucleotides (aptamer, 5′-GGG TTA GGG TTA GGG TTA GGG-3′) were ordered from Shenggong Bioengineering Ltd. (Shanghai, China). Multiwalled carbon nanotubes (CNT) (0.05g) were dispersed in 60 mL of 2.2 M HNO3 for 20 h at room temperature with the aid of ultrasonic agitation (for 30 min), then washed with distilled water to neutrality, and dried in an oven at 37 °C. A 0.5% m/m chitosan stock solution was prepared by dissolving 250 mg of chitosan powder in 50 mL of 1.0% v/v acetic acid solution and stirred for 3 h at room temperature until complete dissolution occurred. The chitosan solution was stored under refrigeration at 4 °C when not in use. A 1.0 × 10−2 M stock solution of Ru(bpy)32+ was prepared by dissolving a suitable amount of Ru(bpy)32+ in redistilled water without further purification. Sodium phosphate buffer stock solution (0.1 M, PBS, pH 7.4) or acetate buffer (pH 5.0) was prepared by dissolving suitable amount of analytical grade Na2HPO4−NaH2PO4 or HAc−NaAc in highly pure water (sterile Minipore water, 18.3 MΩ). Especially, the tris(hydroxymethyl) aminomethane (Tris) (Shanghai Lanji 8944

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950

Analytical Chemistry

Article

exchange of reacting species. According to this particle formation mechanism in W/O, the smaller size of Ru(bpy)32+ dye-doped silica/chitosan NPs, shown in Figure 2B, can be explained due to the presence of chitosan, which not only carries hydrophilic groups such as the amino groups but also carries the hydrophobic carbon chain. While the chitosan polymer was added into the W/O system, the hydrophobic interaction between carbon chain in chitosan and hydrophobic chain in surfactant molecules could enhance the stability of the surfactant films and cause a lower deformability or a stronger attachment to droplets; interdroplet open water channels are less likely to allow reacting species to pass through. With fewer interdroplet dynamic exchanges, reacting species tend to retain and grow in their respective droplets, leading to the formation of more nuclei and smaller final particles. The TEM results indicate that the chitosan may be doped in these nanoparticles. As shown in A and B of Figure 2, the spectral behaviors of Ru(bpy)32+ in nanoparticles were also characterized through fluorescence and UV−vis-spectra measurements. The emission spectra of the pure Ru(bpy)32+ and CRuS NPs were measured in the aqueous solution (Figure 2B). Pure Ru(bpy)32+ shows maximal emission at 594 nm upon excitation at 458 nm in aqueous solution. The absorption spectra remains the same for the pure Ru(bpy)32+ and CRuS NPs in the aqueous solution (as shown in Figure 2A), However, the maximal emission wavelength of the CRuS NPs blue shifts by 14 nm compared with the pure Ru(bpy)32+ (Figure 2B). The shorter-wavelength shift in the fluorescence emission of Ru(bpy)32+ dye-doped silica-based nanoparticles, shown in Figure 2B, can be explained due to interaction between SiO− groups in CRuS NPs and Ru(bpy)32+. Because the SiO− groups have a negative charge and present stronger electrostatic adsorption ability to the positive charged Ru(bpy)32+ dye, the dye molecules mostly reside in the bulk of the silica nanomatrix. Hence, when these dye-doped nanoparticles are dispersed in water, they have less interaction with the surrounding water molecules and show fluorescence emission at a lower relative wavelength. The insets in Figure 2A and B respectively show the visible and fluorescence photography of the CRuS NPs in the aqueous solution, which indicate that the CRuS NPs present the good dispersing ability in the water. In addition, the zeta potential of CRuS NPs was also measured in acidic medium to explore whether the chitosan was present inside nanoparticles since the chitosan was a typical cationic polymer especially in acidic medium. Our results showed that compared to −20 mV of pure RuSiO2 NPs, the +25 mV zeta potential was observed for the CRuS NPs. All these results mentioned above indicate that the chitosan and Ru(bpy)32+ were successfully doped into the silica nanoparticles. The Self-Assembly of CRuS NPs on CNT/Nafion FilmCoated Electrode. The self-assembly behavior of the CRuS NPs on the CNT/Nafion film-coated GC electrode was investigated by cyclic voltammetry and ECL, respectively (as shown in Figure 3). First, the CV results showed that, compared to the bare GC electrode (curve a in Figure 3), the charging current at the CNT/Nafion-modified electrode (curve b in Figure 3) was obviously enhanced, which indicated that the CNT/Nafion film was successfully coated on the GC electrode. Second, the self-assembling procedure of CRuS NPs on CNT/Nafion film electrode was also studied by the CV technique. As shown in Figure 4, our results show that, while

HAc buffer (pH 5.0) and a light magnetic stirring procedure was applied to this CRuS NPs for 30 min. During this process, CRuS NPs were self-assembled onto the surface of the GC electrode. Of note, for preparing the electrodes with the good reproducibility, the pretreatment of the CRuS NPs was important. The original synthesized CRuS NPs solution was centrifuged at 3000 rpm for 10 min to separate the bigger particles from the solution; then, the retained solution was further centrifuged at 7000 rpm to get the resulting NPs. Thereafter, the electrode was washed carefully and thoroughly with pH 7.4 phosphate buffer solution to remove nonspecifically bound nanoparticles to minimize the background response. The resulting electrode was characterized via cyclic voltammograms and ECL measurement in 0.1 M phosphate buffer solution (PBS, pH 7.4). The ECL measurements were performed from 0 to 1.3 V with a scan rate of 100 mV· s−1. Aptamer Binding to K+. Prior to use, all the DNA aptamers were prepared in the binding reaction medium (50 mM Tris-HCl, pH 7.4) and heated to 90 °C for 5 min and allowed to cool slowly in water to room temperature. Then, 10 μL of 100 nM aptamer was added to 80 μL 0.1 M Tris−HCl buffer and incubated for another 30 min at 25 °C in the presence of 10 μL of different levels of concentration of standard K+ solution or K+ sample solution. ECL Measurements. Ten μL of the complex of the aptamer with different levels of concentration of standard K+ solution or K+ sample solution was placed on the surface of the CRuS NPs-CNT/Nafion-modified electrodes for about 5 min, and this modified electrode was further rinsed with highly pure water. Thereafter, the resulting electrode was inserted into the ECL cell, which only contained the 0.1 M PBS (pH 7.4), and the relating ECL signals were recorded when the electrode potential was scanned from 0 to 1.3 V at the scan rate of 100 mV· s−1. The negative voltage of the photomultiplier tube (PMT) was set at −800 V in the process of detection.

■

RESULTS AND DISCUSSION Characterization of Synthesized CRuS NPs. In this work, the CRuS NPs were synthesized by a typical W/O microemulsion route22 and were further investigated by TEM, fluorescence spectroscopy, and zeta potential techniques, respectively. As shown in Figure 1A, TEM showed that the

Figure 1. TEM images of CRuS NPs (A) and RuSiO2 NPs (B).

as-prepared nanoparticles presented the uniform size of about 50 nm. Of note, compared to the pure Ru-doped silica nanoparticles (RuSiO2 NPs) in Figure 1B, the size of the CRuS NPs was decreased, and there were some polymer sheets on the CRuS NPs. A previous work24 suggested that the size of colloidal particles formed in W/O microemulsions is directly influenced by two factors. One is the number of microemulsion droplets that host reacting species, and the other factor is the steric hindrance of surfactant films from the interdroplet dynamic 8945

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950

Analytical Chemistry

Article

Figure 2. UV−vis spectra (A) and fluorescence spectra (B) of Ru(bpy)32+ solution (black curve), CRuS NPs (red curve).

modified CNT-Nafion film-coated electrode was investigated by ECL intensity−potential curve. As shown in Figure 5, the

Figure 3. Cyclic voltammograms recorded in PBS (pH 7.4) at (a) the bare glassy carbon electrode, (b) the CNT/Nafion-modified electrode. Potential scan rate, 100 mV/s. Figure 5. ECL profiles of CRuS NPs-CNT/Nafion-modified electrode in 0.10 M PBS (pH 7.4) with a scan rate of 100 mV·s−1.

Figure 4. Cyclic voltammograms recorded of the CNT/Nafionmodified electrode immersed in the CRuS NPs solution for different time periods (c1−c5 represent 10, 20, 30, 40, 50 min, respectively; the potential of the scan rate was 100 mV/s).

peak ECL intensity occurred at about 1.1 V and showed the typical ECL profile of Ru(bpy)32+ ECL reaction, which indicated that the Ru(bpy)32+ inside NPs on modified electrode could effectively do its ECL reaction. Last, for obtaining the strong ECL intensity of the CRuS NPs on the modified electrode, the relationship between the ECL intensities of modified electrode and the self-assembly time was further studied since the more NPs on the electrode could produce stronger ECL signals. Our results (Figure 6) showed that, as CRuS NPs selfassembling time on the electrode increased from 0 to 20 min, the ECL intensities increased, and then a saturated platform appeared above 20 min. Thus, 30 min of self-assembled time

the CNT/Nafion electrode was inserted into the CRuS NPs in pH 5.0 HAc−NaAc buffer medium for different times, a new redox wave was observed and the oxidation peak and reduction peak occurred at 1.15 and 1.03 V, respectively, which was the typical redox wave of Ru(bpy)32+. In addition, the redox peak currents of Ru(bpy)32+ increase regularly (c1−c5) with increasing the electrode inserting time in pH 5.0 HAc−NaAc buffer medium. These results indicate that the CRuS NPs were really self-assembled on the CNT/Nafion film electrode and the electrostatic interaction between the positively charged CRuS NPs and the negatively charged Nafion film should be the driving force to assemble the CRuS NPs on the electrode.25 Furthermore, the good electrochemical performances of Ru(bpy)32+ inside CRuS NPs on the resulting electrode should be ascribed to the excellent electron channel effects of CNT on the electrode. Moreover, the ECL behavior of CRuS NPs-

Figure 6. Effect of the self-assembling time on the ECL intensity of CRuS NPs-CNT/Nafion-modified electrode at 100 mV·s−1 in 0.10 M PBS (pH 7.4). 8946

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950

Analytical Chemistry

Article

was used to prepare the ECL electrode. Of note, as mentioned above in Figure 4, the redox peak currents of Ru(bpy)32+ on CRuS NPs self-assembled electrode increased with the electrode inserting time from 10 to 50 min. These results showed that the ECL intensities did not increase with the increase of redox peak currents of Ru(bpy)32+ above 20 min. The possible reason can be explained due to both the chitosan in the CRuS NPs and the ECL properties of Ru(bpy)32+ in aqueous solution because the ECL signal of Ru(bpy)32+ in aqueous solution was mainly ascribed to the chemiluminescence reaction of electrogenerated Ru(bpy)33+ with hydroxyl radical.26 Thus, while more CRuS NPs was assembled on the electrode with increasing time, especially above 30 min, more chitosan was also presented on electrode surface. In this case, the chemical reaction of amino groups on chitosan with hydroxyl radical may partly consume ECL intermediators,27 which leads to the ECL signal being enhanced slowly. Scheme 1 may be the best graphic to illustrate the CRuS NPs’ self-assembling procedure on the electrode.

Figure 7. (A) ECL intensity and (B) cyclic voltammograms of the CRuS NPs-CNT/Nafion composite film electrode in PBS (pH 7.4) under continuous CV for 11 cycles with the scan rate of 100 mV·s−1.

the enhancing ECL signal of guanine bases decreased. Therefore, we explored the application of CRuS NPs-modified electrode to detect label-free aptamer binding K+ event as shown in Scheme 2. On the basis of this idea, the interaction between CRuS NPs and aptamer and the ECL aptasensing performances were studied.

Scheme 1. Representation of CNT/Nafion Film and CRuS NPs-Modified Electrode

Scheme 2. Representation of ECL Sensing the Binding Event of Aptamer to K+

To investigate the stability of the CRuS NPs-CNT/Nafion composite films on the electrode, the CRuS NPs-modified electrodes were placed in pH 7.4 buffers for 2 weeks, and the following measurements were performed in the PBS (pH 7.4). First, the composite coating films on the electrode did not come off during the test period, indicating that CRuS NPsCNT/Nafion composite films were well adhered to the GC electrode. Second, while the potential scanned from 0 to 1.3 V, the peak potentials of Ru(bpy)32+ at the test electrode were essentially unchanged for 11 times of continued potential scanning, and the oxidation peak current decreased less than 5% compared with the first scanning value. At the same time, the ECL responses of the test electrode showed that, under continuous potential scanning for 11 cycles, the relative standard deviation of ECL signals was less than 5%, suggesting good ECL reproducibility of the test electrode (as shown in the inset of Figure 7). These results indicated that the as-prepared ECL electrode presented the good ECL and electrochemical stability and can be used for the subsequent aptasensing. Label-Free ECL Sensing Scheme for Aptamers Binding to K+. The enhancing ECL effect of Ru(bpy)32+ by guanine bases,28 and the stronger interaction of the chitosan with the DNA aptamer,29 enable CRuS NPs to function as an indicator in designing a new label-free ECL aptasensing platform. When G-rich DNA aptamers are in their coil-like state, the strong binding affinity of CRuS NPs to aptamers is expected to allow the coil-like aptamer to be wrapped around CRuS NPs on the electrode, which enables the guanine bases on the aptamers to closely approach Ru(bpy)32+ on the surface of CRuS NPs to do their ECL reaction. In contrast, as G-rich DNA aptamers bind to K+, a rigid folded structure of aptamer−K+ complex, Gquadruplex, was formed, and it could prevent the G-bases in the aptamer from approaching the CRuS NPs surface. As a result,

To investigate the binding affinity between CRuS NPs and aptamers, the adsorption efficiency of CRuS NPs toward aptamers was studied by UV−vis adsorption spectra (as shown in Table S1 of SI). The results showed that the adsorption efficiency of CRuS NPs toward aptamers distributed in the range of 38%−46%, which indicated the strong binding affinity between CRuS NPs and aptamers. At the same time, the concentration of K+ had no obvious effect on the adsorption efficiency. As shown in Figure 8, our initial results showed that, compared to the blank ECL signal (curve a), the ECL signal (curve b) of the CRuS NPs-modified electrode not only increased by 12-fold, but also presented two ECL peaks after 10 nM DNA aptamers was dropped onto the electrode. This ECL signal-enhancing effect was attributed to the ECL enhancing effect of the guanine bases in aptamers to the ECL of Ru(bpy)32+ on the electrode. The occurrence of two ECL peaks should be related to the structural characteristics of the Ru(bpy)32+ inside CRuS NPs. Here, the Ru(bpy)32+ has two states, one located on the surface and the other existing inside CRuS NPs. Although the surface Ru(bpy)32+ molecules were easily oxidized electrochemically, they presented the relative 8947

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950

Analytical Chemistry

Article

Figure 8. ECL intensity−potential curves and cyclic voltammograms obtained from 0 to 1.3 V in 0.10 M PBS (pH 7.4) with a scan rate of 100 mV·s−1 at different electrodes. (a) CRuS NPs-modified electrode, (b) the modified electrode with 10 nM DNA aptamer, and (c) the modified electrode with 10 nM mixture of DNA aptamer and K+. The PMT was based at −800 V.

Figure 9. CD spectra of selected sequences of the aptamer in 50 mM Tris−HCl (pH 7.4) buffer without (black line) and with (red line) K+. [Aptamer] = 2 μM; [K+] = 2 μM.

lower luminescence emission yield due to the quenching of oxygen in solution. As for the Ru(bpy)32+ inside the silica phase of NPs, their electrochemical oxidation potential should be higher than that of the surface one due to the poor electric conduction ability of silica-based materials; however, they owned the higher luminescence yields, which led to the higher ECL enhancing signal from the guanine bases in the aptamer. Thus, two ECL peaks were formed. However, the ECL intensity of the same concentration aptamer was obviously decreased (curve c) after preadding the 10 nM K+ in aptamer solution, which implied that the formation of G-quadruplex of aptamer with K+ could greatly prevent G-bases on the aptamer from approaching CRuS NPs. As a result, the ECL signal, generated from the G-bases with Ru(bpy)32+ in nanoparticles, was greatly decreased. At the same time, the CV results (as shown in the inset of Figure 8) also showed that, after adding the K+ into aptamers, the electro-oxidation peak currents were obviously decreased; this result further demonstrated that the G-quadruplex of aptamer with K+ could prevent the G-bases on the aptamer from approaching CRuS NPs nanoparticles. To understand whether the G-quadruplex was formed in the presence of K+, we measured circular dichroism (CD) spectra because CD can be used to gain information about quadruplex structures of the DNA aptamer reacting with K+. The CD spectra were recorded between 220 and 320 nm at 500 nm· min−1. As can be seen from Figure 9, compared to the free DNA aptamer, the complex of aptamer with K+ has a nonconservative spectrum with a positive band and negative band centered at wavelengths 295 and 270 nm, respectively, which proved the formation of antiparallel G-quartets.30 These results obviously suggest that the addition of K+ to the DNA aptamer solution really results in the formation of G-quadruplex at the experimental conditions the same as that of ECL. Analytical Performance. Under the optimal conditions, the ECL analytical performances, such as K + sensing concentration range, and the selectivity of the proposed method for K+ were investigated, respectively. The typical ECL responsive curve of K+ in the proposed sensing platform is shown in Figure 10. As can be seen, the ECL intensities (from curve a to f) of the CRuS NPs-CNT/ Nafion-modified electrode decreased with increasing the amount of K+ in DNA aptamer binding medium. Under the optimal experimental conditions, while K+ concentration was

Figure 10. (A) ECL responses of the ECL electrode to 10 nM DNA aptamer in the presence of different concentration of K+: (a) 0 nM, (b) 10 nM, (c) 30 nM, (d) 50 nM, (e) 70 nM, (f) 90 nM in 0.10 M PBS (pH 7.4).

increased from 1.0 nM to 90 nM in the aptamer binding K+ reaction medium, the ECL intensities of CRuS NPs-CNT/ Nafion-modified electrode in the ECL reaction medium gradually decreased. In addition, the decreasing ECL intensities were linear with K+ concentration (as shown in the inset of Figure 10). The related regression equation of ECL signals to K+ at both 10 to 90 nM and 1.0 to 10 nM was ΔIECL = 1756.7 + 34.4CK+ (R = 0.9981) and ΔIECL = 912.5 + 527.3 CK+ (R = 0.9952), respectively, and a 0.3 nM (S/N = 3) detection limit for K+ was achieved. The specificity of this method was investigated by mixing 1.0 × 10−8 mol·L−1 K+ with 100-fold other metal ions such as Hg2+, Zn2+, Cu2+, Ca2+, Fe3+, Al3+, Na+, and Mg2+. The results showed that except for Na+, the other metal ions presented serious interference for ECL sensing K+. However, as shown in Figure 11, while 1.0 × 10−5 mol·L−1 EDTA was added into the aptamer binding K+ medium, compared to the ECL signal of K+ alone, other metal ions including Zn2+, Cu2+, Ca2+, Fe3+, Al3+, Na+, Mg2+, and Hg2+ did not interfere with the detection of K+. However, in this paper, the formation of G-quadruplex by aptamer and K+ occurred in Tris-HCl buffer solutions which have been purified. The ECL sensing K+ took place in PBS buffer and the coexisting K+ and Na+ in PBS did not interfere with ECL detection of K+ (as shown in Figure S2 in SI). Moreover, the repeatability and fabrication reproducibility of the proposed ECL sensor were determined and the results are shown in Figure S3 and Figure S4 of the SI. The repeatability of one electrode to determine 10.0 nM K+ was fairly good, and the 8948

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950

Analytical Chemistry

Article

proposed aptasensing platform has the ability to sensitively discriminate free state G-rich aptamer from its complex state (G-quadruplex). According to the proposed ECL sensing mechanism, this sensing platform may be extended to sensitively detect other targets (such as protein). In addition, as the ECL sensing nanoindicator, the design and selfassembling of RuSiO2 NPs on CNT/Nafion electrode play the key role for achieving this ECL aptasensing mechanism. Thus, we believe that the label-free ECL aptasensing method demonstrated in this work could open up a new way for nucleic acid studies and molecular sensing.

■

Figure 11. Dependence of relative ECL intensities on the various ions. Experimental conditions: K+ and DNA aptamer is 1.0 × 10−8 mol·L−1, respectively. Other metal ions are 1.0 × 10−6 mol·L−1 and EDTA is 1.0 × 10−5 mol·L−1.

Further experimental details and additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

relative standard deviation (RSD) was 1.3% for three successive assays. The electrode-to-electrode reproducibility was also estimated for 10 nM K+ at five different ECL sensors. This series yielded a RSD of 6.1%. The good reproducibility of ECL sensors may be explained by the fact that the self-assembled procedure of CRuS NPs is controllable and the CRuS NPs presented the good ECL performances toward the aptamer or the complex of aptamer with K+. Application in Analysis of SW480 Colorectal Cancer Cells. To evaluate the applicability and reliability of the proposed ECL sensing platform, two samples of SW480 colorectal cancer cells were examined. A sample of 1000 SW480 colorectal cancer cells was first dispersed into the 10 mL of pure water and further was disrupted by the ultrasound pretreating technique to release the K+ in the cells into solution. Thereafter, 10 μL sample of cells was diluted to 100 μL with pure water as the ECL detection sample and the relating ECL signals of the K+ in the sample were measured by the procedure similar to that described in the Experimental Section. The analytical results are shown in Table 1.

1 2

addition (nM)

found (nM)

0 5 0 10

69 74 68 78

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone/fax: 86-29-81530791. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are indebted to Dr. Y. L. Tang for her kind help and valuable suggestions. This work was financially supported by the projects from National Natural Science Foundation of China (No. 21375085), the Cultural Heritage Conversation Science and Technology Research Foundation (No. 20090106), and the Fundamental Research Funds for the Central Universities (GK201001007) and (GK201302018).

■

REFERENCES

(1) Liu, J. W.; Cao, Z. H.; Lu, Y. Chem. Rev. 2009, 109, 1948−1998. (2) Willner, I.; Zayats, M. Angew. Chem., Int. Ed. 2007, 46, 6408− 6418. (3) Mayer, G. Angew. Chem., Int. Ed. 2007, 48, 2672−2689. (4) Zelada-Guillen, G. A.; Bhosale, S. V.; Riu, J.; Rius, F. X. Anal. Chem. 2010, 82, 9254−9260. (5) Zhang, H. Q.; Li, F.; Dever, B.; Li, X. F.; Le, X. C. Chem. Rev. 2013, 113, 2812−2841. (6) Ouyang, X.; Yu, R.; Jin, J.; Li, J.; Yang, R.; Tan, W.; Yuan, J. Anal. Chem. 2010, 83, 782−789. (7) Zhang, S. H.; Yan, Y. M.; Bi, S. Anal. Chem. 2009, 81, 8695− 8701. (8) Liu, C. W.; Huang, C. C.; Chang, H. T. Anal. Chem. 2009, 81, 2383−2387. (9) Liu, G. D.; Mao, X.; Phillips, J. A.; Xu, H.; Tan, W. H.; Zeng, L. W. Anal. Chem. 2009, 81, 10013−10018. (10) So, H. M.; Won, K.; Kim, Y. H.; Kim, B.; Ryu, B. H.; Na, P. S.; Kim, H.; Lee, J. J. Am. Chem. Soc. 2005, 127, 11906−11907. (11) Hu, L. Z.; Bian, Z.; Li, H. J.; Han, S.; Yuan, Y. L.; Gao, L. X.; Xu, G. B. Anal. Chem. 2009, 81, 9807−9811. (12) Zuo, X. L.; Xiao, Y.; Plaxco, K. W. J. Am. Chem. Soc. 2009, 131, 6944−6945. (13) Ding, J. W.; Chen, Y.; Wang, X. W.; Qin, W. Anal. Chem. 2012, 84, 2055−2061. (14) Liu, X. Q.; Freeman, R.; Golub, E.; Willner, I. ACS Nano 2011, 5, 7648−7655. (15) Du, Y.; Chen, C.; Yin, J. Y.; Li, B. L.; Zhou, M.; Dong, S. J.; Wang, E. Anal. Chem. 2010, 82, 1556−1563. (16) Zhu, L.; Li, C.; Zhu, Z.; Liu, D.; Zou, Y.; Wang, C.; Fu, H.; Yang, C. Anal. Chem. 2012, 84, 8383−8390.

Table 1. Detection Results and Recoveries of K+ in SW480 Colorectal Cancer Cells Samples sample no.

ASSOCIATED CONTENT

S Supporting Information *

recovery % 94 ± 5 100 ± 6

The K+ concentrations in the samples were found to be 68 and 69 nM, respectively, which indicated that the proposed ECL method presented good reproducibility. To evaluate the accuracy ability, the added standard method was explored, and our results showed that the average recoveries of the proposed method were in the range of 94% ± 5% to 100% ± 6%, indicating good accuracy and acceptable precision. Thus, the present method is satisfactory for practical application in the detection of K+ for clinical diagnosis.

■

CONCLUSION In summary, a label-free and sensitive ECL aptasensing platform was developed by using CRuS NPs as the ECL sensing indicator and the native guanine bases in the aptamer as signaling molecules. Our results demonstrated that this 8949

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950

Analytical Chemistry

Article

(17) Yin, X. B.; Xin, Y. Y.; Zhao, Y. Anal. Chem. 2009, 81, 9299− 9305. (18) Hu, L. Z.; Bian, Z.; Li, H. J.; Han, S.; Yuan, Y.; Gao, L. X.; Xun, G. B. Anal. Chem. 2009, 81, 9807−9811. (19) Chen, L. F.; Cai, Q. H.; Luo, F.; Chen, X.; Zhu, X.; Qiu, B.; Lin, Z. Y.; Chen, G. N. Chem. Commun. 2010, 46, 7751−7753. (20) Hagan, K. A.; Meier, W. L.; Ferrance, J. P.; Landers, J. P. Anal. Chem. 2009, 81, 5249−5256. (21) Johnston, D. H.; Glasgow, K. C.; Thorp, H. H. J. Am. Chem. Soc. 1995, 177, 8933−8938. (22) Dennany, L.; Forster, R. J.; Rusling, J. F. J. Am. Chem. Soc. 2003, 125, 5213−5218. (23) Zhang, Z. L.; Liang, S. Q. Chin. Sci. Bull. 1982, 7, 411−412. (24) Chang, C. L.; Fogler, H. S. Langmuir 1997, 13, 3295−3307. (25) Zhang, M. G.; Smith, A.; Gorski, W. Anal. Chem. 2004, 76, 5045. (26) Lin, J. M.; Qu, F.; Yamada, M. Anal. Bioanal. Chem. 2002, 374, 1159. (27) Wan, A. J.; Xu, Q.; Sun, Y.; Li, H. L. J. Agric. Food Chem. 2013, 61, 6921−6928. (28) Wang, J.; Liu, G. D.; Engelhard, M. H.; Lin, Y. H. Anal. Chem. 2006, 78, 6974−6979. (29) Ma, P. L.; Lavertu, M.; Winnik, F. M.; Buschmann, M. D. Biomacromolecules 2009, 10, 1490−1499. (30) Kankia, B. I.; Marky, L. A. J. Am. Chem. Soc. 2001, 123, 10799− 10804.

8950

dx.doi.org/10.1021/ac5017584 | Anal. Chem. 2014, 86, 8943−8950