- Email: [email protected]
Electrochemical K-562 cells sensor based on origami paper device for point-of-care testing
Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Electrochemical K-562 cells sensor based on origami paper device for point-of-care testing Shenguang Ge a,b, Lina Zhang b, Yan Zhang a, Haiyun Liu a, Jiadong Huang a, Mei Yan a, Jinghua Yu a,n a Key Laboratory of Chemical Sensing & Analysis in Universities of Shandong, School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022, China b Shandong Provincial Key Laboratory of Preparation and Measurement of Building Materials, University of Jinan, Jinan 250022, China
art ic l e i nf o
a b s t r a c t
Article history: Received 10 February 2015 Received in revised form 2 May 2015 Accepted 4 May 2015
A low-cost, simple, portable and sensitive paper-based electrochemical sensor was established for the detection of K-562 cell in point-of-care testing. The hybrid material of 3D Au nanoparticles/graphene (3D Au NPs/GN) with high specific surface area and ionic liquid (IL) with widened electrochemical windows improved the good biocompatibility and high conductivity was modified on paper working electrode (PWE) by the classic assembly method and then employed as the sensing surface. IL could not only enhance the electron transfer ability but also provide sensing recognition interface for the conjugation of Con A with cells, with the cell capture efficiency and the sensitivity of biosensor strengthened simultaneously. Concanavalin A (Con A) immobilization matrix was used to capture cells. As proof-of-concept, the paper-based electrochemical sensor for the detection of K-562 cells was developed. With such sandwich-type assay format, K-562 cells as model cells were captured on the surface of Con A/IL/3D AuNPs@GN/PWE. Con A-labeled dendritic PdAg NPs were captured on the surface of K-562 cells. Such dendritic PdAg NPs worked as catalysts promoting the oxidation of thionine (TH) by H2O2 which was released from K-562 cells via the stimulation of phorbol 12-myristate-13-acetate (PMA). Therefore, the current signal response was dependent on the amount of PdAg NPs and the concentration of H2O2, the latter of which corresponded with the releasing amount from cells. So, the detection method of K-562 cell was also developed. Under optimized experimental conditions, 1.5 10 14 mol of H2O2 releasing from each cell was calculated. The linear range and the detection limit for K-562 cells were determined to be 1.0 103–5.0 106 cells/mL and 200 cells/mL, respectively. Such as-prepared sensor showed excellent analytical performance with good fabrication reproducibility, acceptable precision and satisfied accuracy, providing a novel protocol in point-of-care testing of cells. & 2015 Elsevier B.V. All rights reserved.
Keywords: Electrochemical sensor Paper analytical device Graphene Dendritic PdAg nanoparticles
1. Introduction Recently paper has been exploited as a novel platform for cytological tests due to its merits of low cost, simplicity, portability, disposability, great biodegradability and chemo/bio-compatibility. Paper-based cyto-devices showed their great potential for realizing in situ, simple and miniaturized testing in various areas including basic scientific advancement, clinical diagnostics, therapeutics and so forth. Up to date, paper-based analytical devices have been receiving considerable attentions to the development of point-of-care diagnostics [1–4]. Based on the distinct features of high sensitivity, wide dynamic concentration response n
Corresponding author. Tel.: þ 86 531 82767161; fax: þ86 531 82765969. E-mail address: [email protected] (J. Yu).
range, simple instrument and ease of operation, electrochemical (EC) techniques have significantly become a promising method in microfluidic paper analytical devices (μ-PADs) [5–7]. Our group has developed a series of μ-PADs for the sensitive detection of small molecules [8], proteins [9], and DNA [10] based on EC techniques. It has become increasingly evident that reactive oxygen species (ROS) participated in signal transduction process in a concentration dependent manner [11–13]. H2O2, as one of the reactive oxygen species, can be extremely cytotoxic since its long lifetime allows it to penetrate into almost any cellular compartment, which potentially induces various adverse biological modifications such as peroxidation of cell membrane lipids, DNA bases, and backbone hydroxylation [14–16]. Consequently, quantitative detection of H2O2 in cells and measurements of its dynamic release process
http://dx.doi.org/10.1016/j.talanta.2015.05.008 0039-9140/& 2015 Elsevier B.V. All rights reserved.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
from living cells are essential to fully reveal its roles in cellular physiology, which can further provide reliable diagnosis of pathological conditions. It is the most common representative of ROS studied in cell environments because of its stability and penetration through cell membranes. A variety of techniques, including colorimetry [17], chemiluminescence [18], fluorescence [19,20], electrochemical methods [21,22] etc., have been developed for the determination of cellular H2O2. In this paper, a lowcost, simple, portable and sensitive paper-based electrochemical sensor was developed for the detection of K-562 cells (one of the most aggressive human chronic myelogenous leukemia cell lines) based on the release of H2O2 from cells on origami paper device. Nano-materials have significantly contributed to the enlargement of the sensing interface [23–25]. Herein, 3D-AuNPs/GN composites with large specific surface area and good electroconductivity were synthesized to modify the working electrode for the fabrication of sensing interface in the assay [26]. Meanwhile, IL possessed unique chemical and physical properties such as non-volatility, film-forming property, wide electrochemical windows, good biocompatibility, and high conductivity [27–29]. Correspondingly, increasing attention has been paid to the preparation of modified electrodes with IL and nanomaterial composite in hopes of the coalescence of their unique properties. The sensing platform in this paper was fabricated with the nanocomposites of IL and 3DAuNPs/GN on the paper working electrode for the determination of cells. Signal amplification is another key factor to improve the analytical performance of the sensor, which is often achieved by the label of more enzymes on nanoparticles to construct biological nanoprobes [30,31]. Unfortunately, the direct result was the easy lost of the bioactivity of enzymes. Nanoparticles, which are biologically inert, have been found to possess intrinsic enzyme mimetic activity similar to that of natural peroxidase, thus opening a new door for the application of nanomaterials in signal amplification [32,33]. In comparison with natural enzymes, nanomaterials are considerably more stable over a wide range of pH and temperature, and possess additional advantages of controllable preparation, mass yield, relatively low cost and tunable catalytic activities. In light of such advantages, a series of inorganic nanomaterials have emerged as peroxidase mimics and been applied in the fields of environmental chemistry and biomedicine fields [32,33]. Dendritic PdAg NPs, horseradish peroxidase (HRP)-like, can quickly catalyze the reaction of typical HRP substrates, suggesting its peroxidase-like feature. Herein, we reported a novel IL supported 3D Au NPs/GN/PWE electrochemical sensing platform on the paper device for the detection of H2O2 released from K-562 cells via the stimulation of PMA [34,35], with the aim of providing fundamental understanding and practical guidance on cellular detection, drug discovery, toxicology, disease diagnosis and pointof-care testing.
K-562 cell line was kindly provided from Qilu Hospital of Shandong University, Jinan, China. K-562 cells were cultured in RPMI 1640 medium (GIBOC) containing fetal calf serum (HyClone, Logan, UT), penicillin (100 μg/mL), and streptomycin (100 μg/mL) in a humidified 37 °C incubator with 5% CO2. The cells in exponential growth were collected and separated from the medium by centrifugation at 1000 rpm for 5 min and then washed thrice with a sterile pH 7.4 phosphate buffer saline (PBS). The sediment was resuspended in sterile PBS containing 1 mmol/L Ca2 þ and Mn2 þ to obtain a homogeneous cell suspension. Here, the divalent cations Ca2 þ and Mn2 þ were required for the activity of Con A binding to cell surface mannose. Origami electrochemical device was fabricated by a solid-wax printer (Xerox Phaser 8560N color printer). Transmission electron microscope (TEM) images were recorded on a JEOL 2010 TEM operating at an accelerating voltage of 120 kV. Scanning electron microscope (SEM) images were obtained via QUANTA FEG 250 thermal field emission SEM (FEI Co., USA). X-ray photoelectron spectra (XPS) were measured using Thermofisher ESCALB 250 X-ray photoelectron spectrometer. Electrochemical impedance spectroscopy (EIS) was carried out on an IM6x electrochemical station (Zahner, Germany). EC measurements were performed with a CHI 760 (Shanghai CH Instruments, China) system. The electrodes consisted of a screen-printed Ag/AgCl reference electrode and a carbon counter electrode on the paper auxiliary zone and a screen-printed carbon working electrode on the paper sample zones. 2.2. Preparation of 3D-Au NPs/GN composite Graphene oxide (GO) was prepared according to Hummer's method previously reported with a slight modification [26,36]. Firstly, graphite powder (0.50 g) was added into 60 mL of H3PO4/H2SO4 mixture solution (v/v¼1:9) and mixed under stirring. 3.0 g of KMnO4 was gradually added to the mixture with magnetic stirring at ambient temperature. Secondly, the mixture was refluxed and stirred at 50 °C for 12 h. After that, the resultant was cooled down to ambient temperature, poured into 200 mL ice-water containing 3 mL H2O2, and the unexploited graphite was removed by centrifugation. The GO obtained was dehydrated at 60 °C under vacuum drying. GO dispersion (0.5 mg/mL) was obtained via the sonication of GO in water for further use. To form 3D-Au NPs/GN composite, PEG was used as reducing agent in the synthetic process. The mixture, containing 10 mL of GO (0.5 mg/mL), 200 μL of HAuCl4 4H2O (1%, w/w) and 20 μL of PEG, was sonicated for 1 h, and then reacted at 180 °C for 12 h. After being cooled to room temperature, the resultant was washed three times with water. Finally, 3D-AuNPs/GN composite was attained through freeze drying process. 1.0 mg/mL of 3D-AuNPs/GN was prepared and kept for further application. 2.3. Synthesis of dendritic PdAg NPs/Con A
2. Experimental section 2.1. Reagents and apparatus Chlorauric acid (HAuCl4 3H2O), potassium(II) tetrachloro-palladate (K2PdCl4), silver nitrate (AgNO3), ascorbic acid (AA), hydrogen peroxide (H2O2), cetyltrimethyl-ammonium bromide (CTAB), thionine (TH), phorbol 12-myristate-13-acetate (PMA), concanavalin A (Con A), and 1-ethyl-3-methylimidazolium tetrafluoroborate ([BMIM]BF4, IL) were all purchased from aladdin and used as received. Milli-Q water (Z18 MΩ cm) was used for all the solution preparation. All glassware used in the following procedures was cleaned in a bath of a piranha solution (H2SO4/30% H2O2 ¼ 7:3, v/v) and subjected to boiling for 30 min.
Dendritic PdAg NPs were prepared according to the reported method with a minor modification [37]. 40 μL of 10 mmol/L PdCl42 and 40 μL of 10 mmol/L AgNO3 were mixed with 2 mL of 0.25 mmol/L CTAB aqueous solution, followed by addition of 20 μL of 10 mmol/L AA. The solution was then immediately stirred sharply and placed at 30 °C for 5 h with the color of dark brown attained finally, suggesting the formation of dendritic PdAg NPs. Purification was subsequently conducted two times by centrifugation at 12,000 rpm for 5 min. The dendritic PdAg NPs obtained were redispersed in 2.0 mL of water for further characterization and application. The conjugation of dendritic PdAg NPs with Con A was via the noncovalent bond between dendritic PdAg NPs and available amine groups of Con A. 1.0 mL of dendritic PdAg NPs solution was mixed with 1.0 mL of 10 μg/mL of Con A with the reaction conducted
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
reacting at ambient temperature under the stirring for 40 min. Then centrifugation was carried out at 12,000 rpm for 5 min to collect the PdAg NPs/Con A which was washed thrice with PBS and then redispersed in 1 mL of PBS and stored at 4 °C before use.
2.4. Biosensor fabrication The origami electrochemical device was fabricated in batches based on wax printing via a commercially available office printer according to our prior work [38,39]. It was designed on a computer by Adobe illustrator CS4 software and then printed on chromatographic paper. When such a paper was heated, the wax ink dissolved into the interior of the paper, forming hydrophilic channels that could steer fluids through the device. The unprinted area (paper auxiliary zone and paper sample zones) still maintained good hydrophilicity. Detailed parameters are shown in Fig. 1A. The three electrodes system consisted of a screen-printed Ag/AgCl reference electrode and carbon counter electrode on the auxiliary zone and a screen-printed carbon working electrode on the paper sample zone, respectively. After the printed paper was folded, such printed three screen-printed electrodes were connected via the filled solution, whcih could be used for further modifications. As shown in Fig. 1B, 10 μL of 1.0 mg/mL 3D-Au NPs/GN composite solution was first dropped into the sample zone, followed by 10 μL of 1.0 mg/mL Con A solution added and incubated at room temperature for 1 h. Then, 5 μL of 1% BSA (w/v) solution was dropped into the sample zone for 30 min to block the nonspecific binding sites. After a rinse with PBS buffer, 5 μL of 50 mg/mL [BMIM]BF4 ionic liquid solution was added and stayed for 1 h. Finally, Con A-modified IL/3D-Au NPs/GN/PWE was prepared after a thorough rinse with PBS and used for cell capture.
Fig. 1. Detailed parameters of origami electrochemical paper device (A). Paper sheets were firstly patterned in bulk using a wax printer. After baking, three electrodes were screen-printed on wax-patterned sheet. Reference electrode and counter electrode were printed on the auxiliary zone, while working electrode was printed on the sample zone. The prepared sheet was cut to rectangular paper. After modification, the rectangular paper was folded and integrated with a device-holder for electrochemical assay. Schematic representation of the proposed strategy for cell detection (B). (1) Immobilization of 3D-AuNPs/GN; (2) blocking and washing of BSA; (3) immobilization of ionic liquid and Con A; (4) capturing and washing of the cells; and (5) incubation with PdAg NPs-Con A, washing and triggering of the electrochemical reaction.
3
2.5. Electrochemical assay 10 μL of PBS containing 1 105 cells/mL K-562 cells was dropped into the sample zone, and incubated at 37 °C for 50 min. K-562 cells were captured on Con A/IL/3D-Au NPs/GN/PWE via the specific binding between Con A and cell surface mannose. Then, the sample zone was rinsed with incubation buffer solution to dispel the culture solution. The cell-captured sensor was blocked with 1% BSA for 30 min and rinsed thrice with pH 7.4 PBS for subsequent assay utilization. 10 μL of Con A@PdAg NPs was dropped into the sample zone and incubated at 37 °C for 50 min. Afterwards, the sample zone was washed with pH 7.4 PBS to remove nonsepcifically bonded Con A@PdAg NPs. PMA was then added to induce H2O2 generation with a more consistent and durable chemotactic response [8,35,40]. As the result of stimulation by PMA, respiratory burst occurred with H2O2 as the end product inside cells. The endogenous generation of H2O2 in the cells was evaluated and the flux of H2O2 released from the cells was measured. Due to H2O2 release from the cells, the electrochemical response increased suddenly. The differential pulse voltammetric (DPV) measurements of the oxidation process of TH were monitored for cell quantification under the catalysis of PdAg NPs, which were carried out from 0.3 V to 0.6 V with a pulse amplitude of 50 mV in 20 μL of 10 mmol/L PBS (pH 7.4) containing 10 mmol/L PMA and 2 mmol/L TH.
3. Results and discussion 3.1. Characterization of 3D-AuNPs/GN and dendritic PdAg NPs As shown in Fig. 2A and B, 3D-Au NPs/GN was obtained. With such 3D structures possessing a higher specific surface area, more Con A could be attached. Meanwhile, Au NPs were able to enhance electronic transmission rate and anchor the Con A. The average diameter of Au NPs on the 3D GN surface was about 100 nm. The corresponding energy dispersive spectroscopy (EDS) was shown in Fig. 2B (inset figure), the result of which evidenced the successful growth of AuNPs on 3D GN. After 3D Au NPs/GN was modified on the PWE, enhanced specific surface area, improved high conductivity and good biocompatibility features of the sensor were observed. The addition of Ag induced the formation of spherical PdAg NPs with a branched structure at the Ag/Pd ratio of 1/1. Representative TEM images of the as-prepared products were shown in Fig. 2C and D. It could be clearly seen that dendritic PdAg NPs were uniformly dispersed and highly branched subunits, having an average size of approximately 30 nm. Such open nanodendritic structure and small primary particle size were expected to exhibit interesting performance in catalysis. From the energy dispersive X-ray analysis shown in Fig. 2D (inset figure), the coexistence of silver and palladium in the dendritic PdAg NPs was demonstrated where Ag and Pd were homogeneously distributed over the entire particle. Surface morphology was especially significant for catalysts. XPS was performed to study the elemental composition of dendritic PdAg NPs surface. Fig. 3 showed the survey spectrum of PdAg NPs, which was dominated by the signals of Ag and Pd. Two wellresolved peaks at 331.68 eV and 339.13 eV could be assigned to Pd 3d5/2 and 3d3/2, respectively, confirming the formation of metallic Pd. On the other hand, the observation of the Ag 3d 5/2 and 3d 3/2 peaks at 368.04 eV and 373.59 eV, respectively, indicated that Ag was mainly present in its metallic state. 3.2. Electrochemical character of ionic liquid The aim of the utilization of ionic liquid was mainly to attain good performance of the sensor such as high conductivity, wide
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
4
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 2. SEM image of 3D-AuNPs/GN (A), magnification SEM image of 3D-AuNPs/GN (B) with EDS shown in the inset, TEM image of dendritic PdAg NPs (C), and SEM of dendritic PdAg NPs (D) with EDS shown in the inset.
Fig. 3. XPS spectra of Pd(A) and Ag of PdAg NPs(B).
electrochemical window, good stability and satisfied biocompatibility. IL and 3D-Au NPs/GN would firmly bind together and form stable solid composition, which could not be washed away or diluted during PBS rinsing required to wash the device at different steps. Also, based on results from control experiment, the signal response was equivalent before and after rinsing, indicating that IL was not washed away or diluted. Cyclic voltammograms (CV) of the modified electrode with the presence and absence of IL in 5.0 mmol/L [Fe(CN)6]3 /4 solution are shown in Fig. 4A. The Con A/3D-AuNPs/GN/PWE, a pair of weak redox peaks, was observed (curve a) indicating the weak electron transfer rate at the interface, which was mainly ascribed to the poor conductivity of Con A. With IL present (Con A/IL/3D-AuNPs/ GN/PWE), a well-defined and enhanced redox peak was obtained
(curve b), which could be attributed to the high ionic conductivity and synergetic electrocatalytic effect of IL. As expected, the utilization of IL promoted the electron transfer during the electrochemical process. 3.3. Comparison of catalytic efficiency of Pd/Ag NPs and HRP Signal amplification and noise reduction are critical for a sandwich-type electrochemical assay. To investigate the effect of HRP and PdAg NPs on the sensitivity of the assay, two different labeled probes were prepared which were HRP-bound Con A (HRP-Con A) and PdAg-conjugated Con A. The same batch of sensors with the reaction of the same cell concentration (considering the amount range from 10 to 1000 cells) was utilized, and
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
5
Fig. 4. (A) Cyclic voltammograms of Con A/3D-AuNPs/GN/PWE (a) and Con A/IL/3D-AuNPs/GN/PWE (b), in 0.1 mol/L KCl containing 5 mmol/L of [Fe(CN)6]3 /4 with the scan rate of 100 mV/s. (B) Cyclic voltammograms of the electrochemical sandwich-type sensors by using different labels utilized: (a) HRP and (b) PdAg NPs, in pH 7.4 PBS containing 1.0 10 9 mol/mL of H2O2 and 2 mmol/L of TH, with the error bars representing the 95% confidence interval of the mean for the y-axis currents.
Fig. 5. (A) Cyclic voltammograms obtained on Cell/Con A/IL/3D-Au NPs/GN/PWE in the absence (a) and presence (b) of PMA as well as PdAg/ConA/Cell/Con A/IL/3D-Au NPs/ GN/PWE in the absence (c) and presence (d) of PMA, cell: 1.0 105 cells/mL, TH: 2 mmol/L and scan rate: 100 mV/s. (B) Electrochemical impedance spectra of (a) 3D-Au NPs/ GN/PWE, (b) IL/3D-Au NPs/GN/GCE, (c) Con A/IL/3D-Au NPs/GN/PWE, (d) Cell/ConA/IL/3D-Au NPs/GN/PWE and (e) PdAg/Con A/Cell/ Con A/IL/3D-Au NPs/GN/PWE in 0.5 M KCl solution with 5 mM [Fe(CN)6]4 /3 .
then different labeled probes were utilized as signal amplification. The measurement of cell quantity was based on the amperometric change of sensor before and after the cell–Con A interaction. As shown in Fig. 4B, when PdAg NPs-Con A was used, higher signals were recorded, suggesting a more sensitive assay designed. On the contrary, the signal was lower with the HRP-Con A. The signal intensity of PdAg NPs was almost five-fold higher than that of HRP, indicating the good catalytic efficiency of PdAg NPs. The reason might be due to the high surface-to-volume ratio and good enzyme-like properties of PdAg NPs. 3.4. Electrochemical behavior for 3D-Au NPs/GN and dendritic PdAg NPs As shown in Fig. 5A, the cyclic voltammogram (CV) of IL supported 3D-Au NPs/GN composite modified PWE exhibited a pair of well-defined redox peaks at 0.11 V and 0.19 V, which corresponded to the electrochemical oxidation and reduction of TH respectively. PMA was used to stimulate the cells to release H2O2, which induced the current signal increased correspondingly. After Con A/ PdAg NPs was incubated for 40 min, the peak current significantly increased, indicating the occurrence of a typical peroxide-like catalytic process. The assay was a sandwich-type assay format using PdAg NPs as peroxidase-like catalyst for the detection of K-562 cells based on H2O2 releasing from the cells as enzyme substrates and TH as electron mediator. 10 μL of 1 105 cells/mL K-562 cell was captured on the Con A/IL/3D-AuNPs/GN/PWE. Subsequently, PdAg NPs/Con A was anchored on the cell surface by
the combination of Con A and cell surface mannose. After a very short period upon injection of 10 μL of PMA (10 mmol/L), the electrochemical response increased suddenly, suggesting that a large amount of H2O2 released from the cells. A stable current could be reached finally, illuminating the end of the event. With more cells captured by Con A/IL/3D-Au NPs/GN/PWE, more H2O2 was released from cells. Moreover, the more PdAg NPs were anchored on the cell surface, the higher electrochemical signal was obtained correspondingly. Consequently, the proposed method herein enhanced the sensitivity dramatically because of the dual signal amplification of both PdAg NPs catalysis and in situ release of H2O2. Electrochemical impedance spectroscopy (EIS) is a powerful and facile electrochemical technique to verify the modification of the electrodes step by step. As shown in Fig. 5B, Niquist plots comprised a semicircle part at higher frequency range and a straight linear part at lower frequency range. The diameter of the semicircle equaled to the electron transfer resistance (Ret) at the interface of the electrode. Due to the good electronic transfer ability, 3D-Au NPs/GN/PWE (curve b) exhibited low Ret. However, after IL was attached, accompanied with the increased conductivity, IL/3D-Au NPs/GN/PWE (curve a) exhibited lower Ret than that of 3D-AuNPs/GN/PWE (curve b). It was clear that the diameter of the semicircles increased successively with assembly of Con A (curve c), capture of cell (curve d), and incubation of nanoprobes (curve e) sequentially, indicating an enhancement of Ret step by step. As a result, the successful preparation of the biosensor could be concluded.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
Fig. 6. Effects of (A) pH, (B) the concentration of TH, (C) the concentration of PMA and (D) the incubation time of PdAg NPs. The error bars meant the standard deviations from parallel determination of five duplicates, representing the 95% confidence interval of the mean for the y-axis currents.
Fig. 7. The reproducibility of electrochemical sensor of intra-assay (A) and inter-assay (B) in pH 7.4 PBS containing 10 mmol/L of PMA and 2 mmol/L of TH.
3.5. Optimization of experimental conditions The current response was significantly influenced by the pH value of solution as well as the concentration of TH. Fig. 6A shows the effect of pH on the current response. The current intensity increased with the pH value rising from 6.8 to 7.6 and then decreased. Correspondingly, the optimal pH of 7.4 was selected. Also, the current response to the concentration of TH was also investigated with the results seen in Fig. 6B. The current signal showed an observable increase with the concentration of TH increasing up to 2 mmol/L, and reached a plateau afterwards. And thus, 2 mmol/L of TH was used in the following experiments. The effect of PMA dose was also evaluated. Fig. 6C depicts the current signal induced by different doses of PMA. The current signal was enhanced with the addition of PMA dose, demonstrating the
H2O2 concentration-dependency of PMA dose. Along with the gradual enlarged concentration of PMA, the current signal showed a rapid enhancement till the concentration of PMA reached up to 10 mmol/L, followed by a slow increase and then a plateau. Based on the above results, 10 mmol/L of PMA was chosen for the following experiments conducted. The incubation time was a vital parameter for the capture of target cell and the anchor of dendritic PdAg NPs. The dependence of incubation time for PdAg NPs/Con A on the current response was studied and optimized as shown in Fig. 6D. When PdAg NPs/ Con A was captured onto K-562 cell/Con A/ IL/3D AuNPs/GN/PWE, the current signal increased with an incubation time within 40 min and reached a steady value. Therefore, the optimum incubation time of biosensor was set at 40 min for the capture of target cell capture as well as the anchor of dendritic PdAg NPs.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
7
Fig. 8. The calibration curve for different concentrations of H2O2 in pH 7.4 PBS containing 2 mmol/L of TH after incubation with 1.0 105 cells/mL. The error bars represented the 95% confidence interval of the mean for the y-axis currents.
Fig. 9. DPV responses to different concentrations of K-562 cell (A) and calibration curve for K-562 immunoassay (B) in pH 7.4 PBS containing 10 mmol/L of PMA and 2 mmol/L of TH after incubation with 10 μL of 1.0 103, 7.0 103, 7.0 104, 3.5 105, 7.0 105, 1.4 106, 2.5 106, and 5.0 106 cells/mL (from 'a' to 'h'). The error bars represented the 95% confidence interval of the mean for the y-axis currents.
3.6. Reproducibility and stability The reproducibility of as-prepared electrochemical sensor was assessed by the relative standard derivation (RSD) of intra- and interassay. The intra-assay precision of the developed sensor (Fig. 7A) was studied by assay of four cell levels five times with the RSD attained to be 4.3%, 2.8%, 3.1% and 4.7% at 1 103, 1 104, 1 105, and 1 106 cells/mL, respectively. Similarly, the RSD of inter-assay on the five sensors (Fig. 7B) was 2.4%, 5.8%, 3.5% and 4.6% at 1 103, 1 104, 1 105, and 1 106 cells/mL, respectively. Such results evidenced good reproducibility of the biosensor designed. The storage stability of PdAg NPs at room temperature was 94.7% in the first month and then decreased by 8.4% in the next 2 months. For PdAg NPs/Con A, the storage stability retained 93.1% of its initial response after 1 week and then gradually decreased to 87% after 2 weeks in 4 °C refrigerator. 3.7. Analytical performance To evaluate the analytical reliability and application potential of the proposed method, 10 μL of 1 105 cells/mL K-562 cell was captured on the Con A/IL/3D-AuNPs/GN/PWE. In the absence of PMA, such system did not release H2O2 from the cells. With the addition of different amounts of H2O2, the signal was assayed by the DPV. The linear range for H2O2 varied from 1.0 10 11 mol/mL to 8.0 10 8 mol/mL (Fig. 8). The linear regression equation was expressed as I ¼13.19 log c þ12.88 with a correlation coefficient of R¼ 0.989. When PMA was present, it stimulated K-562 cells to release H2O2, which gave rise to a distinct electrochemical response. According to the linear range of H2O2 attained above, the
average amount of 1.5 10 14 mol for H2O2 released from each cell was obtained. Under optimal conditions, a weak DPV peak current was observed in the absence of cells when the Con A/IL/3D-Au NPs/GN/PWE only was filled with the electrolyte. When the Con A/IL/3D-Au NPs/GN/ PWE was incubated with the K-562 cells and PdAg NPs/Con A subsequently, a certain amount of PdAg NPs/Con A was absorbed on the Cell/Con A/IL/3D-Au NPs/GO/PWE via the affinity interaction between Con A and the mannose groups existing on the surface of the cell. which made the DPV peak current increase steeply. Such current increased along with the addition of K-562 cells' concentration, which was proportional to the logarithmic value of the cell concentration ranging from 1.0 103 to 5.0 106 cells/mL (Fig. 9). The linear regression equation was gained to be I¼12.24 log c(cells/mL)þ14.97 with the correlation coefficient of R¼ 0.989 (n¼5), where I stands for the peak current intensity and c represents the concentration of K-562 cells. And the detection limit for cell concentration was 200 cells/mL at 3s. Based on the above performance, electrochemical sensing was successfully introduced into a lab-on-paper device for a highly sensitive determination of H2O2 released from living cancer cells. Therefore, the fabricated sensor in this work could potentially be applied in clinical tests.
4. Conclusion In summary, a novel electrochemical sensing platform for the detection of K-562 cell was developed on the paper device. 3D-Au NPs/GN modified PWE based on ionic liquid support on the paper device was fabricated. 3D-Au NPs/GN owned large specific surface,
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
8
good biocompatibility and high electron transfer ability. The capability of IL combining with 3D-Au NPs/GN to form conductive composites made them highly attractive for the electrode modification due to its distinct advantages, such as good sensitivity, stability and electronic conductivity as well as wide electrochemical window. With a sandwich-type assay format, the dendritic PdAg NPs labeled Con A was anchored to the surface of the cells to catalyze the oxidation of TH by H2O2 which was released from the cells under the stimulation of PMA. Therefore, the current signal response was dependent on two factors, which were the amount of PdAg NPs on the surface of the cells and the concentration of H2O2 generated from the cells. The proposed method demonstrated high sensitivity. Such versatile electrochemical sensing platform developed herein was believed to be of great clinical value for both diagnosis and treatment.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China (21207048, 21277058, 21175058), National High-tech R&D Program (863 Program) (SQ2015AAJY1562), Technology Development Plan of Shandong Province, China (Grant no. 2014GGX103012) and Doctor Foundation of University of Jinan (XBS1428).
References [1] Y. Zhang, C.B. Zhou, J.F. Nie, S.W. Le, Q. Qin, F. Liu, Y.P. Li, J.P. Li, Anal. Chem. 86 (2014) 2005–2012. [2] C. Renault, M.J. Anderson, R.M. Crooks, J. Am. Chem. Soc. 136 (2014) 4616–4623. [3] A.K. Yetisen, M.S. Akram, C.R. Lowe, Lab Chip 13 (2013) 2210–2251. [4] E. Carrilho, A.W. Martinez, G.M. Whitesides, Anal. Chem. 81 (2009) 7091–7095. [5] M. Santhiago, J.B. Wydallis, L.T. Kubota, C.S. Henry, Anal. Chem. 85 (2013) 5233–5239. [6] A.C. Glavan, D.C. Christodouleas, B. Mosadegh, H.D. Yu, B.S. Smith, J. Lessing, M. T. Fernández-Abedul, G.M. Whitesides, Anal. Chem. 86 (2014) 11999–12007. [7] W. Dungchai, O. Chailapakul, C.S. Henry, Anal. Chem. 81 (2009) 5821–5826. [8] F. Liu, S.G. Ge, J.H. Yu, M. Yan, X.R. Song, Chem. Commun. 50 (2014) 10315–10318.
[9] S.G. Ge, L. Ge, M. Yan, X.R. Song, J.H. Yu, J.D. Huang, Chem. Commun. 48 (2012) 9397–9399. [10] J.J. Lu, S.G. Ge, L. Ge, M. Yan, J.H. Yu, Electrochim. Acta 80 (2012) 334–341. [11] S. Kumar, W.K. Rhim, D.K. Lim, J.M. Nam, ACS Nano 7 (2013) 2221–2230. [12] Y. You, W. Nam, Chem. Sci. 5 (2014) 4123–4135. [13] E.G. Ju, Z. Liu, Y.D. Du, Y. Tao, J.S. Ren, X.G. Qu, ACS Nano 8 (2014) 6014–6023. [14] H.J. Forman, M. Maiorino, F. Ursini, Biochemistry 49 (2010) 835–842. [15] S.R. Panikkanvalappil, M.A. Mahmoud, M.A. Mackey, M.A. El-Sayed, ACS Nano 7 (2013) 7524–7533. [16] K.U. Ingold, V.W. Bowry, R. Stocker, C. Walling, Proc. Natl. Acad. Sci. USA 90 (1993) 45–49. [17] Y.N. Xiao, Z.Q. Ye, G.L. Wang, J.L. Yuan, Inorg. Chem. 51 (2012) 2940–2946. [18] Y.D. Lee, C.K. Lim, A. Singh, J. Koh, J. Kim, I.C. Kwon, S. Kim, ACS Nano 6 (2012) 6759–6766. [19] M. Kumar, N. Kumar, V. Bhalla, P.R. Sharma, Y. Qurishi, Chem. Commun. 48 (2012) 4719–4721. [20] T.T. Chen, Y.H. Hu, Y. Cen, X. Chu, Y. Lu, J. Am. Chem. Soc. 135 (2013) 11595–11602. [21] M.M. Liu, S.J. He, W. Chen, Nanoscale 6 (2014) 11769–11776. [22] W. Chen, S. Cai, Q.Q. Ren, W. Wen, Y.D. Zhao, Analyst 137 (2012) 49–58. [23] W. Chen, J. Chen, Y.B. Feng, L. Hong, Q.Y. Chen, L.F. Wu, X.H. Lin, X.H. Xia, Analyst 137 (2012) 1706–1712. [24] J.D. Gaynor, A.S. Karakoti, T. Inerbaev, S. Sanghavi, P. Nachimuthu, V. Shutthanandan, S. Seal, S. Thevuthasan, J. Mater. Chem. B 1 (2013) 3443–3450. [25] P.C. Pandey, D.S. Chauhan, Analyst 137 (2012) 376–385. [26] G.Q. Sun, J.J. Lu, S.G. Ge, X.R. Song, J.H. Yu, M. Yan, J.D. Huang, Anal. Chim. Acta 775 (2013) 85–92. [27] K. Kwak, S.S. Kumar, K. Pyo, D. Lee, ACS Nano 8 (2014) 671–679. [28] N. Liu, Z.F. Ma, Biosens. Bioelectron. 51 (2014) 184–190. [29] Z.H. Wang, F. Li, J.F. Xia, L. Xia, F.F. Zhang, S. Bi, G.Y. Shi, Y.Z. Xia, J.Q. Liu, Y.H. Li, L.H. Xia, Biosens. Bioelectron. 61 (2014) 391–396. [30] X.Q. Liu, J.M. Zhang, S.H. Liu, Q.Y. Zhang, X.H. Liu, D.K.Y. Wong, Anal. Chem. 85 (2013) 4350–4356. [31] B. Jeong, R. Akter, O.H. Han, C.K. Rhee, M.A. Rahman, Anal. Chem. 85 (2013) 1784–1791. [32] S.E.F. Kleijn, S.C.S. Lai, M.T.M. Koper, P.R. Unwin, Angew. Chem. Int. Ed. 53 (2014) 3558–3586. [33] X. Cao, N. Wang, S. Jia, Y.H. Shao, Anal. Chem. 85 (2013) 5040–5046. [34] X.G. Li, Y. Liu, A.W. Zhu, Y.P. Luo, Z.F. Deng, Y. Tian, Anal. Chem. 82 (2010) 6512–6518. [35] P. Wu, Y.D. Qian, P. Du, H. Zhang, C.X. Cai, J. Mater. Chem. 22 (2012) 6402–6412. [36] W.S. Hummers Jr., R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339. [37] W.W. He, X.C. Wu, J.B. Liu, X.N. Hu, K. Zhang, S. Hou, W.Y. Zhou, S.S. Xie, Chem. Mater. 22 (2010) 2988–2994. [38] S.G. Ge, W.Y. Liu, L. Ge, M. Yan, J.X. Yan, J.D. Huang, J.H. Yu, Biosens. Bioelectron. 49 (2013) 111–117. [39] J.X. Yan, L. Ge, X.R. Song, M. Yan, S.G. Ge, J.H. Yu, Chem.: A Eur. J. 18 (2012) 4938–4945. [40] Y. Zhang, C.Y. Wu, X.J. Zhou, X.C. Wu, Y.Q. Yang, H.X. Wu, S.W. Guo, J.Y. Zhang, Nanoscale 5 (2013) 1816–1819.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Electrochemical K-562 cells sensor based on origami paper device for point-of-care testing Shenguang Ge a,b, Lina Zhang b, Yan Zhang a, Haiyun Liu a, Jiadong Huang a, Mei Yan a, Jinghua Yu a,n a Key Laboratory of Chemical Sensing & Analysis in Universities of Shandong, School of Chemistry and Chemical Engineering, University of Jinan, Jinan 250022, China b Shandong Provincial Key Laboratory of Preparation and Measurement of Building Materials, University of Jinan, Jinan 250022, China
art ic l e i nf o
a b s t r a c t
Article history: Received 10 February 2015 Received in revised form 2 May 2015 Accepted 4 May 2015
A low-cost, simple, portable and sensitive paper-based electrochemical sensor was established for the detection of K-562 cell in point-of-care testing. The hybrid material of 3D Au nanoparticles/graphene (3D Au NPs/GN) with high specific surface area and ionic liquid (IL) with widened electrochemical windows improved the good biocompatibility and high conductivity was modified on paper working electrode (PWE) by the classic assembly method and then employed as the sensing surface. IL could not only enhance the electron transfer ability but also provide sensing recognition interface for the conjugation of Con A with cells, with the cell capture efficiency and the sensitivity of biosensor strengthened simultaneously. Concanavalin A (Con A) immobilization matrix was used to capture cells. As proof-of-concept, the paper-based electrochemical sensor for the detection of K-562 cells was developed. With such sandwich-type assay format, K-562 cells as model cells were captured on the surface of Con A/IL/3D AuNPs@GN/PWE. Con A-labeled dendritic PdAg NPs were captured on the surface of K-562 cells. Such dendritic PdAg NPs worked as catalysts promoting the oxidation of thionine (TH) by H2O2 which was released from K-562 cells via the stimulation of phorbol 12-myristate-13-acetate (PMA). Therefore, the current signal response was dependent on the amount of PdAg NPs and the concentration of H2O2, the latter of which corresponded with the releasing amount from cells. So, the detection method of K-562 cell was also developed. Under optimized experimental conditions, 1.5 10 14 mol of H2O2 releasing from each cell was calculated. The linear range and the detection limit for K-562 cells were determined to be 1.0 103–5.0 106 cells/mL and 200 cells/mL, respectively. Such as-prepared sensor showed excellent analytical performance with good fabrication reproducibility, acceptable precision and satisfied accuracy, providing a novel protocol in point-of-care testing of cells. & 2015 Elsevier B.V. All rights reserved.
Keywords: Electrochemical sensor Paper analytical device Graphene Dendritic PdAg nanoparticles
1. Introduction Recently paper has been exploited as a novel platform for cytological tests due to its merits of low cost, simplicity, portability, disposability, great biodegradability and chemo/bio-compatibility. Paper-based cyto-devices showed their great potential for realizing in situ, simple and miniaturized testing in various areas including basic scientific advancement, clinical diagnostics, therapeutics and so forth. Up to date, paper-based analytical devices have been receiving considerable attentions to the development of point-of-care diagnostics [1–4]. Based on the distinct features of high sensitivity, wide dynamic concentration response n
Corresponding author. Tel.: þ 86 531 82767161; fax: þ86 531 82765969. E-mail address: [email protected] (J. Yu).
range, simple instrument and ease of operation, electrochemical (EC) techniques have significantly become a promising method in microfluidic paper analytical devices (μ-PADs) [5–7]. Our group has developed a series of μ-PADs for the sensitive detection of small molecules [8], proteins [9], and DNA [10] based on EC techniques. It has become increasingly evident that reactive oxygen species (ROS) participated in signal transduction process in a concentration dependent manner [11–13]. H2O2, as one of the reactive oxygen species, can be extremely cytotoxic since its long lifetime allows it to penetrate into almost any cellular compartment, which potentially induces various adverse biological modifications such as peroxidation of cell membrane lipids, DNA bases, and backbone hydroxylation [14–16]. Consequently, quantitative detection of H2O2 in cells and measurements of its dynamic release process
http://dx.doi.org/10.1016/j.talanta.2015.05.008 0039-9140/& 2015 Elsevier B.V. All rights reserved.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
from living cells are essential to fully reveal its roles in cellular physiology, which can further provide reliable diagnosis of pathological conditions. It is the most common representative of ROS studied in cell environments because of its stability and penetration through cell membranes. A variety of techniques, including colorimetry [17], chemiluminescence [18], fluorescence [19,20], electrochemical methods [21,22] etc., have been developed for the determination of cellular H2O2. In this paper, a lowcost, simple, portable and sensitive paper-based electrochemical sensor was developed for the detection of K-562 cells (one of the most aggressive human chronic myelogenous leukemia cell lines) based on the release of H2O2 from cells on origami paper device. Nano-materials have significantly contributed to the enlargement of the sensing interface [23–25]. Herein, 3D-AuNPs/GN composites with large specific surface area and good electroconductivity were synthesized to modify the working electrode for the fabrication of sensing interface in the assay [26]. Meanwhile, IL possessed unique chemical and physical properties such as non-volatility, film-forming property, wide electrochemical windows, good biocompatibility, and high conductivity [27–29]. Correspondingly, increasing attention has been paid to the preparation of modified electrodes with IL and nanomaterial composite in hopes of the coalescence of their unique properties. The sensing platform in this paper was fabricated with the nanocomposites of IL and 3DAuNPs/GN on the paper working electrode for the determination of cells. Signal amplification is another key factor to improve the analytical performance of the sensor, which is often achieved by the label of more enzymes on nanoparticles to construct biological nanoprobes [30,31]. Unfortunately, the direct result was the easy lost of the bioactivity of enzymes. Nanoparticles, which are biologically inert, have been found to possess intrinsic enzyme mimetic activity similar to that of natural peroxidase, thus opening a new door for the application of nanomaterials in signal amplification [32,33]. In comparison with natural enzymes, nanomaterials are considerably more stable over a wide range of pH and temperature, and possess additional advantages of controllable preparation, mass yield, relatively low cost and tunable catalytic activities. In light of such advantages, a series of inorganic nanomaterials have emerged as peroxidase mimics and been applied in the fields of environmental chemistry and biomedicine fields [32,33]. Dendritic PdAg NPs, horseradish peroxidase (HRP)-like, can quickly catalyze the reaction of typical HRP substrates, suggesting its peroxidase-like feature. Herein, we reported a novel IL supported 3D Au NPs/GN/PWE electrochemical sensing platform on the paper device for the detection of H2O2 released from K-562 cells via the stimulation of PMA [34,35], with the aim of providing fundamental understanding and practical guidance on cellular detection, drug discovery, toxicology, disease diagnosis and pointof-care testing.
K-562 cell line was kindly provided from Qilu Hospital of Shandong University, Jinan, China. K-562 cells were cultured in RPMI 1640 medium (GIBOC) containing fetal calf serum (HyClone, Logan, UT), penicillin (100 μg/mL), and streptomycin (100 μg/mL) in a humidified 37 °C incubator with 5% CO2. The cells in exponential growth were collected and separated from the medium by centrifugation at 1000 rpm for 5 min and then washed thrice with a sterile pH 7.4 phosphate buffer saline (PBS). The sediment was resuspended in sterile PBS containing 1 mmol/L Ca2 þ and Mn2 þ to obtain a homogeneous cell suspension. Here, the divalent cations Ca2 þ and Mn2 þ were required for the activity of Con A binding to cell surface mannose. Origami electrochemical device was fabricated by a solid-wax printer (Xerox Phaser 8560N color printer). Transmission electron microscope (TEM) images were recorded on a JEOL 2010 TEM operating at an accelerating voltage of 120 kV. Scanning electron microscope (SEM) images were obtained via QUANTA FEG 250 thermal field emission SEM (FEI Co., USA). X-ray photoelectron spectra (XPS) were measured using Thermofisher ESCALB 250 X-ray photoelectron spectrometer. Electrochemical impedance spectroscopy (EIS) was carried out on an IM6x electrochemical station (Zahner, Germany). EC measurements were performed with a CHI 760 (Shanghai CH Instruments, China) system. The electrodes consisted of a screen-printed Ag/AgCl reference electrode and a carbon counter electrode on the paper auxiliary zone and a screen-printed carbon working electrode on the paper sample zones. 2.2. Preparation of 3D-Au NPs/GN composite Graphene oxide (GO) was prepared according to Hummer's method previously reported with a slight modification [26,36]. Firstly, graphite powder (0.50 g) was added into 60 mL of H3PO4/H2SO4 mixture solution (v/v¼1:9) and mixed under stirring. 3.0 g of KMnO4 was gradually added to the mixture with magnetic stirring at ambient temperature. Secondly, the mixture was refluxed and stirred at 50 °C for 12 h. After that, the resultant was cooled down to ambient temperature, poured into 200 mL ice-water containing 3 mL H2O2, and the unexploited graphite was removed by centrifugation. The GO obtained was dehydrated at 60 °C under vacuum drying. GO dispersion (0.5 mg/mL) was obtained via the sonication of GO in water for further use. To form 3D-Au NPs/GN composite, PEG was used as reducing agent in the synthetic process. The mixture, containing 10 mL of GO (0.5 mg/mL), 200 μL of HAuCl4 4H2O (1%, w/w) and 20 μL of PEG, was sonicated for 1 h, and then reacted at 180 °C for 12 h. After being cooled to room temperature, the resultant was washed three times with water. Finally, 3D-AuNPs/GN composite was attained through freeze drying process. 1.0 mg/mL of 3D-AuNPs/GN was prepared and kept for further application. 2.3. Synthesis of dendritic PdAg NPs/Con A
2. Experimental section 2.1. Reagents and apparatus Chlorauric acid (HAuCl4 3H2O), potassium(II) tetrachloro-palladate (K2PdCl4), silver nitrate (AgNO3), ascorbic acid (AA), hydrogen peroxide (H2O2), cetyltrimethyl-ammonium bromide (CTAB), thionine (TH), phorbol 12-myristate-13-acetate (PMA), concanavalin A (Con A), and 1-ethyl-3-methylimidazolium tetrafluoroborate ([BMIM]BF4, IL) were all purchased from aladdin and used as received. Milli-Q water (Z18 MΩ cm) was used for all the solution preparation. All glassware used in the following procedures was cleaned in a bath of a piranha solution (H2SO4/30% H2O2 ¼ 7:3, v/v) and subjected to boiling for 30 min.
Dendritic PdAg NPs were prepared according to the reported method with a minor modification [37]. 40 μL of 10 mmol/L PdCl42 and 40 μL of 10 mmol/L AgNO3 were mixed with 2 mL of 0.25 mmol/L CTAB aqueous solution, followed by addition of 20 μL of 10 mmol/L AA. The solution was then immediately stirred sharply and placed at 30 °C for 5 h with the color of dark brown attained finally, suggesting the formation of dendritic PdAg NPs. Purification was subsequently conducted two times by centrifugation at 12,000 rpm for 5 min. The dendritic PdAg NPs obtained were redispersed in 2.0 mL of water for further characterization and application. The conjugation of dendritic PdAg NPs with Con A was via the noncovalent bond between dendritic PdAg NPs and available amine groups of Con A. 1.0 mL of dendritic PdAg NPs solution was mixed with 1.0 mL of 10 μg/mL of Con A with the reaction conducted
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
reacting at ambient temperature under the stirring for 40 min. Then centrifugation was carried out at 12,000 rpm for 5 min to collect the PdAg NPs/Con A which was washed thrice with PBS and then redispersed in 1 mL of PBS and stored at 4 °C before use.
2.4. Biosensor fabrication The origami electrochemical device was fabricated in batches based on wax printing via a commercially available office printer according to our prior work [38,39]. It was designed on a computer by Adobe illustrator CS4 software and then printed on chromatographic paper. When such a paper was heated, the wax ink dissolved into the interior of the paper, forming hydrophilic channels that could steer fluids through the device. The unprinted area (paper auxiliary zone and paper sample zones) still maintained good hydrophilicity. Detailed parameters are shown in Fig. 1A. The three electrodes system consisted of a screen-printed Ag/AgCl reference electrode and carbon counter electrode on the auxiliary zone and a screen-printed carbon working electrode on the paper sample zone, respectively. After the printed paper was folded, such printed three screen-printed electrodes were connected via the filled solution, whcih could be used for further modifications. As shown in Fig. 1B, 10 μL of 1.0 mg/mL 3D-Au NPs/GN composite solution was first dropped into the sample zone, followed by 10 μL of 1.0 mg/mL Con A solution added and incubated at room temperature for 1 h. Then, 5 μL of 1% BSA (w/v) solution was dropped into the sample zone for 30 min to block the nonspecific binding sites. After a rinse with PBS buffer, 5 μL of 50 mg/mL [BMIM]BF4 ionic liquid solution was added and stayed for 1 h. Finally, Con A-modified IL/3D-Au NPs/GN/PWE was prepared after a thorough rinse with PBS and used for cell capture.
Fig. 1. Detailed parameters of origami electrochemical paper device (A). Paper sheets were firstly patterned in bulk using a wax printer. After baking, three electrodes were screen-printed on wax-patterned sheet. Reference electrode and counter electrode were printed on the auxiliary zone, while working electrode was printed on the sample zone. The prepared sheet was cut to rectangular paper. After modification, the rectangular paper was folded and integrated with a device-holder for electrochemical assay. Schematic representation of the proposed strategy for cell detection (B). (1) Immobilization of 3D-AuNPs/GN; (2) blocking and washing of BSA; (3) immobilization of ionic liquid and Con A; (4) capturing and washing of the cells; and (5) incubation with PdAg NPs-Con A, washing and triggering of the electrochemical reaction.
3
2.5. Electrochemical assay 10 μL of PBS containing 1 105 cells/mL K-562 cells was dropped into the sample zone, and incubated at 37 °C for 50 min. K-562 cells were captured on Con A/IL/3D-Au NPs/GN/PWE via the specific binding between Con A and cell surface mannose. Then, the sample zone was rinsed with incubation buffer solution to dispel the culture solution. The cell-captured sensor was blocked with 1% BSA for 30 min and rinsed thrice with pH 7.4 PBS for subsequent assay utilization. 10 μL of Con A@PdAg NPs was dropped into the sample zone and incubated at 37 °C for 50 min. Afterwards, the sample zone was washed with pH 7.4 PBS to remove nonsepcifically bonded Con A@PdAg NPs. PMA was then added to induce H2O2 generation with a more consistent and durable chemotactic response [8,35,40]. As the result of stimulation by PMA, respiratory burst occurred with H2O2 as the end product inside cells. The endogenous generation of H2O2 in the cells was evaluated and the flux of H2O2 released from the cells was measured. Due to H2O2 release from the cells, the electrochemical response increased suddenly. The differential pulse voltammetric (DPV) measurements of the oxidation process of TH were monitored for cell quantification under the catalysis of PdAg NPs, which were carried out from 0.3 V to 0.6 V with a pulse amplitude of 50 mV in 20 μL of 10 mmol/L PBS (pH 7.4) containing 10 mmol/L PMA and 2 mmol/L TH.
3. Results and discussion 3.1. Characterization of 3D-AuNPs/GN and dendritic PdAg NPs As shown in Fig. 2A and B, 3D-Au NPs/GN was obtained. With such 3D structures possessing a higher specific surface area, more Con A could be attached. Meanwhile, Au NPs were able to enhance electronic transmission rate and anchor the Con A. The average diameter of Au NPs on the 3D GN surface was about 100 nm. The corresponding energy dispersive spectroscopy (EDS) was shown in Fig. 2B (inset figure), the result of which evidenced the successful growth of AuNPs on 3D GN. After 3D Au NPs/GN was modified on the PWE, enhanced specific surface area, improved high conductivity and good biocompatibility features of the sensor were observed. The addition of Ag induced the formation of spherical PdAg NPs with a branched structure at the Ag/Pd ratio of 1/1. Representative TEM images of the as-prepared products were shown in Fig. 2C and D. It could be clearly seen that dendritic PdAg NPs were uniformly dispersed and highly branched subunits, having an average size of approximately 30 nm. Such open nanodendritic structure and small primary particle size were expected to exhibit interesting performance in catalysis. From the energy dispersive X-ray analysis shown in Fig. 2D (inset figure), the coexistence of silver and palladium in the dendritic PdAg NPs was demonstrated where Ag and Pd were homogeneously distributed over the entire particle. Surface morphology was especially significant for catalysts. XPS was performed to study the elemental composition of dendritic PdAg NPs surface. Fig. 3 showed the survey spectrum of PdAg NPs, which was dominated by the signals of Ag and Pd. Two wellresolved peaks at 331.68 eV and 339.13 eV could be assigned to Pd 3d5/2 and 3d3/2, respectively, confirming the formation of metallic Pd. On the other hand, the observation of the Ag 3d 5/2 and 3d 3/2 peaks at 368.04 eV and 373.59 eV, respectively, indicated that Ag was mainly present in its metallic state. 3.2. Electrochemical character of ionic liquid The aim of the utilization of ionic liquid was mainly to attain good performance of the sensor such as high conductivity, wide
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
4
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 2. SEM image of 3D-AuNPs/GN (A), magnification SEM image of 3D-AuNPs/GN (B) with EDS shown in the inset, TEM image of dendritic PdAg NPs (C), and SEM of dendritic PdAg NPs (D) with EDS shown in the inset.
Fig. 3. XPS spectra of Pd(A) and Ag of PdAg NPs(B).
electrochemical window, good stability and satisfied biocompatibility. IL and 3D-Au NPs/GN would firmly bind together and form stable solid composition, which could not be washed away or diluted during PBS rinsing required to wash the device at different steps. Also, based on results from control experiment, the signal response was equivalent before and after rinsing, indicating that IL was not washed away or diluted. Cyclic voltammograms (CV) of the modified electrode with the presence and absence of IL in 5.0 mmol/L [Fe(CN)6]3 /4 solution are shown in Fig. 4A. The Con A/3D-AuNPs/GN/PWE, a pair of weak redox peaks, was observed (curve a) indicating the weak electron transfer rate at the interface, which was mainly ascribed to the poor conductivity of Con A. With IL present (Con A/IL/3D-AuNPs/ GN/PWE), a well-defined and enhanced redox peak was obtained
(curve b), which could be attributed to the high ionic conductivity and synergetic electrocatalytic effect of IL. As expected, the utilization of IL promoted the electron transfer during the electrochemical process. 3.3. Comparison of catalytic efficiency of Pd/Ag NPs and HRP Signal amplification and noise reduction are critical for a sandwich-type electrochemical assay. To investigate the effect of HRP and PdAg NPs on the sensitivity of the assay, two different labeled probes were prepared which were HRP-bound Con A (HRP-Con A) and PdAg-conjugated Con A. The same batch of sensors with the reaction of the same cell concentration (considering the amount range from 10 to 1000 cells) was utilized, and
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
5
Fig. 4. (A) Cyclic voltammograms of Con A/3D-AuNPs/GN/PWE (a) and Con A/IL/3D-AuNPs/GN/PWE (b), in 0.1 mol/L KCl containing 5 mmol/L of [Fe(CN)6]3 /4 with the scan rate of 100 mV/s. (B) Cyclic voltammograms of the electrochemical sandwich-type sensors by using different labels utilized: (a) HRP and (b) PdAg NPs, in pH 7.4 PBS containing 1.0 10 9 mol/mL of H2O2 and 2 mmol/L of TH, with the error bars representing the 95% confidence interval of the mean for the y-axis currents.
Fig. 5. (A) Cyclic voltammograms obtained on Cell/Con A/IL/3D-Au NPs/GN/PWE in the absence (a) and presence (b) of PMA as well as PdAg/ConA/Cell/Con A/IL/3D-Au NPs/ GN/PWE in the absence (c) and presence (d) of PMA, cell: 1.0 105 cells/mL, TH: 2 mmol/L and scan rate: 100 mV/s. (B) Electrochemical impedance spectra of (a) 3D-Au NPs/ GN/PWE, (b) IL/3D-Au NPs/GN/GCE, (c) Con A/IL/3D-Au NPs/GN/PWE, (d) Cell/ConA/IL/3D-Au NPs/GN/PWE and (e) PdAg/Con A/Cell/ Con A/IL/3D-Au NPs/GN/PWE in 0.5 M KCl solution with 5 mM [Fe(CN)6]4 /3 .
then different labeled probes were utilized as signal amplification. The measurement of cell quantity was based on the amperometric change of sensor before and after the cell–Con A interaction. As shown in Fig. 4B, when PdAg NPs-Con A was used, higher signals were recorded, suggesting a more sensitive assay designed. On the contrary, the signal was lower with the HRP-Con A. The signal intensity of PdAg NPs was almost five-fold higher than that of HRP, indicating the good catalytic efficiency of PdAg NPs. The reason might be due to the high surface-to-volume ratio and good enzyme-like properties of PdAg NPs. 3.4. Electrochemical behavior for 3D-Au NPs/GN and dendritic PdAg NPs As shown in Fig. 5A, the cyclic voltammogram (CV) of IL supported 3D-Au NPs/GN composite modified PWE exhibited a pair of well-defined redox peaks at 0.11 V and 0.19 V, which corresponded to the electrochemical oxidation and reduction of TH respectively. PMA was used to stimulate the cells to release H2O2, which induced the current signal increased correspondingly. After Con A/ PdAg NPs was incubated for 40 min, the peak current significantly increased, indicating the occurrence of a typical peroxide-like catalytic process. The assay was a sandwich-type assay format using PdAg NPs as peroxidase-like catalyst for the detection of K-562 cells based on H2O2 releasing from the cells as enzyme substrates and TH as electron mediator. 10 μL of 1 105 cells/mL K-562 cell was captured on the Con A/IL/3D-AuNPs/GN/PWE. Subsequently, PdAg NPs/Con A was anchored on the cell surface by
the combination of Con A and cell surface mannose. After a very short period upon injection of 10 μL of PMA (10 mmol/L), the electrochemical response increased suddenly, suggesting that a large amount of H2O2 released from the cells. A stable current could be reached finally, illuminating the end of the event. With more cells captured by Con A/IL/3D-Au NPs/GN/PWE, more H2O2 was released from cells. Moreover, the more PdAg NPs were anchored on the cell surface, the higher electrochemical signal was obtained correspondingly. Consequently, the proposed method herein enhanced the sensitivity dramatically because of the dual signal amplification of both PdAg NPs catalysis and in situ release of H2O2. Electrochemical impedance spectroscopy (EIS) is a powerful and facile electrochemical technique to verify the modification of the electrodes step by step. As shown in Fig. 5B, Niquist plots comprised a semicircle part at higher frequency range and a straight linear part at lower frequency range. The diameter of the semicircle equaled to the electron transfer resistance (Ret) at the interface of the electrode. Due to the good electronic transfer ability, 3D-Au NPs/GN/PWE (curve b) exhibited low Ret. However, after IL was attached, accompanied with the increased conductivity, IL/3D-Au NPs/GN/PWE (curve a) exhibited lower Ret than that of 3D-AuNPs/GN/PWE (curve b). It was clear that the diameter of the semicircles increased successively with assembly of Con A (curve c), capture of cell (curve d), and incubation of nanoprobes (curve e) sequentially, indicating an enhancement of Ret step by step. As a result, the successful preparation of the biosensor could be concluded.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
6
Fig. 6. Effects of (A) pH, (B) the concentration of TH, (C) the concentration of PMA and (D) the incubation time of PdAg NPs. The error bars meant the standard deviations from parallel determination of five duplicates, representing the 95% confidence interval of the mean for the y-axis currents.
Fig. 7. The reproducibility of electrochemical sensor of intra-assay (A) and inter-assay (B) in pH 7.4 PBS containing 10 mmol/L of PMA and 2 mmol/L of TH.
3.5. Optimization of experimental conditions The current response was significantly influenced by the pH value of solution as well as the concentration of TH. Fig. 6A shows the effect of pH on the current response. The current intensity increased with the pH value rising from 6.8 to 7.6 and then decreased. Correspondingly, the optimal pH of 7.4 was selected. Also, the current response to the concentration of TH was also investigated with the results seen in Fig. 6B. The current signal showed an observable increase with the concentration of TH increasing up to 2 mmol/L, and reached a plateau afterwards. And thus, 2 mmol/L of TH was used in the following experiments. The effect of PMA dose was also evaluated. Fig. 6C depicts the current signal induced by different doses of PMA. The current signal was enhanced with the addition of PMA dose, demonstrating the
H2O2 concentration-dependency of PMA dose. Along with the gradual enlarged concentration of PMA, the current signal showed a rapid enhancement till the concentration of PMA reached up to 10 mmol/L, followed by a slow increase and then a plateau. Based on the above results, 10 mmol/L of PMA was chosen for the following experiments conducted. The incubation time was a vital parameter for the capture of target cell and the anchor of dendritic PdAg NPs. The dependence of incubation time for PdAg NPs/Con A on the current response was studied and optimized as shown in Fig. 6D. When PdAg NPs/ Con A was captured onto K-562 cell/Con A/ IL/3D AuNPs/GN/PWE, the current signal increased with an incubation time within 40 min and reached a steady value. Therefore, the optimum incubation time of biosensor was set at 40 min for the capture of target cell capture as well as the anchor of dendritic PdAg NPs.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
7
Fig. 8. The calibration curve for different concentrations of H2O2 in pH 7.4 PBS containing 2 mmol/L of TH after incubation with 1.0 105 cells/mL. The error bars represented the 95% confidence interval of the mean for the y-axis currents.
Fig. 9. DPV responses to different concentrations of K-562 cell (A) and calibration curve for K-562 immunoassay (B) in pH 7.4 PBS containing 10 mmol/L of PMA and 2 mmol/L of TH after incubation with 10 μL of 1.0 103, 7.0 103, 7.0 104, 3.5 105, 7.0 105, 1.4 106, 2.5 106, and 5.0 106 cells/mL (from 'a' to 'h'). The error bars represented the 95% confidence interval of the mean for the y-axis currents.
3.6. Reproducibility and stability The reproducibility of as-prepared electrochemical sensor was assessed by the relative standard derivation (RSD) of intra- and interassay. The intra-assay precision of the developed sensor (Fig. 7A) was studied by assay of four cell levels five times with the RSD attained to be 4.3%, 2.8%, 3.1% and 4.7% at 1 103, 1 104, 1 105, and 1 106 cells/mL, respectively. Similarly, the RSD of inter-assay on the five sensors (Fig. 7B) was 2.4%, 5.8%, 3.5% and 4.6% at 1 103, 1 104, 1 105, and 1 106 cells/mL, respectively. Such results evidenced good reproducibility of the biosensor designed. The storage stability of PdAg NPs at room temperature was 94.7% in the first month and then decreased by 8.4% in the next 2 months. For PdAg NPs/Con A, the storage stability retained 93.1% of its initial response after 1 week and then gradually decreased to 87% after 2 weeks in 4 °C refrigerator. 3.7. Analytical performance To evaluate the analytical reliability and application potential of the proposed method, 10 μL of 1 105 cells/mL K-562 cell was captured on the Con A/IL/3D-AuNPs/GN/PWE. In the absence of PMA, such system did not release H2O2 from the cells. With the addition of different amounts of H2O2, the signal was assayed by the DPV. The linear range for H2O2 varied from 1.0 10 11 mol/mL to 8.0 10 8 mol/mL (Fig. 8). The linear regression equation was expressed as I ¼13.19 log c þ12.88 with a correlation coefficient of R¼ 0.989. When PMA was present, it stimulated K-562 cells to release H2O2, which gave rise to a distinct electrochemical response. According to the linear range of H2O2 attained above, the
average amount of 1.5 10 14 mol for H2O2 released from each cell was obtained. Under optimal conditions, a weak DPV peak current was observed in the absence of cells when the Con A/IL/3D-Au NPs/GN/PWE only was filled with the electrolyte. When the Con A/IL/3D-Au NPs/GN/ PWE was incubated with the K-562 cells and PdAg NPs/Con A subsequently, a certain amount of PdAg NPs/Con A was absorbed on the Cell/Con A/IL/3D-Au NPs/GO/PWE via the affinity interaction between Con A and the mannose groups existing on the surface of the cell. which made the DPV peak current increase steeply. Such current increased along with the addition of K-562 cells' concentration, which was proportional to the logarithmic value of the cell concentration ranging from 1.0 103 to 5.0 106 cells/mL (Fig. 9). The linear regression equation was gained to be I¼12.24 log c(cells/mL)þ14.97 with the correlation coefficient of R¼ 0.989 (n¼5), where I stands for the peak current intensity and c represents the concentration of K-562 cells. And the detection limit for cell concentration was 200 cells/mL at 3s. Based on the above performance, electrochemical sensing was successfully introduced into a lab-on-paper device for a highly sensitive determination of H2O2 released from living cancer cells. Therefore, the fabricated sensor in this work could potentially be applied in clinical tests.
4. Conclusion In summary, a novel electrochemical sensing platform for the detection of K-562 cell was developed on the paper device. 3D-Au NPs/GN modified PWE based on ionic liquid support on the paper device was fabricated. 3D-Au NPs/GN owned large specific surface,
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i
S. Ge et al. / Talanta ∎ (∎∎∎∎) ∎∎∎–∎∎∎
8
good biocompatibility and high electron transfer ability. The capability of IL combining with 3D-Au NPs/GN to form conductive composites made them highly attractive for the electrode modification due to its distinct advantages, such as good sensitivity, stability and electronic conductivity as well as wide electrochemical window. With a sandwich-type assay format, the dendritic PdAg NPs labeled Con A was anchored to the surface of the cells to catalyze the oxidation of TH by H2O2 which was released from the cells under the stimulation of PMA. Therefore, the current signal response was dependent on two factors, which were the amount of PdAg NPs on the surface of the cells and the concentration of H2O2 generated from the cells. The proposed method demonstrated high sensitivity. Such versatile electrochemical sensing platform developed herein was believed to be of great clinical value for both diagnosis and treatment.
Acknowledgments This work was financially supported by the National Natural Science Foundation of China (21207048, 21277058, 21175058), National High-tech R&D Program (863 Program) (SQ2015AAJY1562), Technology Development Plan of Shandong Province, China (Grant no. 2014GGX103012) and Doctor Foundation of University of Jinan (XBS1428).
References [1] Y. Zhang, C.B. Zhou, J.F. Nie, S.W. Le, Q. Qin, F. Liu, Y.P. Li, J.P. Li, Anal. Chem. 86 (2014) 2005–2012. [2] C. Renault, M.J. Anderson, R.M. Crooks, J. Am. Chem. Soc. 136 (2014) 4616–4623. [3] A.K. Yetisen, M.S. Akram, C.R. Lowe, Lab Chip 13 (2013) 2210–2251. [4] E. Carrilho, A.W. Martinez, G.M. Whitesides, Anal. Chem. 81 (2009) 7091–7095. [5] M. Santhiago, J.B. Wydallis, L.T. Kubota, C.S. Henry, Anal. Chem. 85 (2013) 5233–5239. [6] A.C. Glavan, D.C. Christodouleas, B. Mosadegh, H.D. Yu, B.S. Smith, J. Lessing, M. T. Fernández-Abedul, G.M. Whitesides, Anal. Chem. 86 (2014) 11999–12007. [7] W. Dungchai, O. Chailapakul, C.S. Henry, Anal. Chem. 81 (2009) 5821–5826. [8] F. Liu, S.G. Ge, J.H. Yu, M. Yan, X.R. Song, Chem. Commun. 50 (2014) 10315–10318.
[9] S.G. Ge, L. Ge, M. Yan, X.R. Song, J.H. Yu, J.D. Huang, Chem. Commun. 48 (2012) 9397–9399. [10] J.J. Lu, S.G. Ge, L. Ge, M. Yan, J.H. Yu, Electrochim. Acta 80 (2012) 334–341. [11] S. Kumar, W.K. Rhim, D.K. Lim, J.M. Nam, ACS Nano 7 (2013) 2221–2230. [12] Y. You, W. Nam, Chem. Sci. 5 (2014) 4123–4135. [13] E.G. Ju, Z. Liu, Y.D. Du, Y. Tao, J.S. Ren, X.G. Qu, ACS Nano 8 (2014) 6014–6023. [14] H.J. Forman, M. Maiorino, F. Ursini, Biochemistry 49 (2010) 835–842. [15] S.R. Panikkanvalappil, M.A. Mahmoud, M.A. Mackey, M.A. El-Sayed, ACS Nano 7 (2013) 7524–7533. [16] K.U. Ingold, V.W. Bowry, R. Stocker, C. Walling, Proc. Natl. Acad. Sci. USA 90 (1993) 45–49. [17] Y.N. Xiao, Z.Q. Ye, G.L. Wang, J.L. Yuan, Inorg. Chem. 51 (2012) 2940–2946. [18] Y.D. Lee, C.K. Lim, A. Singh, J. Koh, J. Kim, I.C. Kwon, S. Kim, ACS Nano 6 (2012) 6759–6766. [19] M. Kumar, N. Kumar, V. Bhalla, P.R. Sharma, Y. Qurishi, Chem. Commun. 48 (2012) 4719–4721. [20] T.T. Chen, Y.H. Hu, Y. Cen, X. Chu, Y. Lu, J. Am. Chem. Soc. 135 (2013) 11595–11602. [21] M.M. Liu, S.J. He, W. Chen, Nanoscale 6 (2014) 11769–11776. [22] W. Chen, S. Cai, Q.Q. Ren, W. Wen, Y.D. Zhao, Analyst 137 (2012) 49–58. [23] W. Chen, J. Chen, Y.B. Feng, L. Hong, Q.Y. Chen, L.F. Wu, X.H. Lin, X.H. Xia, Analyst 137 (2012) 1706–1712. [24] J.D. Gaynor, A.S. Karakoti, T. Inerbaev, S. Sanghavi, P. Nachimuthu, V. Shutthanandan, S. Seal, S. Thevuthasan, J. Mater. Chem. B 1 (2013) 3443–3450. [25] P.C. Pandey, D.S. Chauhan, Analyst 137 (2012) 376–385. [26] G.Q. Sun, J.J. Lu, S.G. Ge, X.R. Song, J.H. Yu, M. Yan, J.D. Huang, Anal. Chim. Acta 775 (2013) 85–92. [27] K. Kwak, S.S. Kumar, K. Pyo, D. Lee, ACS Nano 8 (2014) 671–679. [28] N. Liu, Z.F. Ma, Biosens. Bioelectron. 51 (2014) 184–190. [29] Z.H. Wang, F. Li, J.F. Xia, L. Xia, F.F. Zhang, S. Bi, G.Y. Shi, Y.Z. Xia, J.Q. Liu, Y.H. Li, L.H. Xia, Biosens. Bioelectron. 61 (2014) 391–396. [30] X.Q. Liu, J.M. Zhang, S.H. Liu, Q.Y. Zhang, X.H. Liu, D.K.Y. Wong, Anal. Chem. 85 (2013) 4350–4356. [31] B. Jeong, R. Akter, O.H. Han, C.K. Rhee, M.A. Rahman, Anal. Chem. 85 (2013) 1784–1791. [32] S.E.F. Kleijn, S.C.S. Lai, M.T.M. Koper, P.R. Unwin, Angew. Chem. Int. Ed. 53 (2014) 3558–3586. [33] X. Cao, N. Wang, S. Jia, Y.H. Shao, Anal. Chem. 85 (2013) 5040–5046. [34] X.G. Li, Y. Liu, A.W. Zhu, Y.P. Luo, Z.F. Deng, Y. Tian, Anal. Chem. 82 (2010) 6512–6518. [35] P. Wu, Y.D. Qian, P. Du, H. Zhang, C.X. Cai, J. Mater. Chem. 22 (2012) 6402–6412. [36] W.S. Hummers Jr., R.E. Offeman, J. Am. Chem. Soc. 80 (1958) 1339. [37] W.W. He, X.C. Wu, J.B. Liu, X.N. Hu, K. Zhang, S. Hou, W.Y. Zhou, S.S. Xie, Chem. Mater. 22 (2010) 2988–2994. [38] S.G. Ge, W.Y. Liu, L. Ge, M. Yan, J.X. Yan, J.D. Huang, J.H. Yu, Biosens. Bioelectron. 49 (2013) 111–117. [39] J.X. Yan, L. Ge, X.R. Song, M. Yan, S.G. Ge, J.H. Yu, Chem.: A Eur. J. 18 (2012) 4938–4945. [40] Y. Zhang, C.Y. Wu, X.J. Zhou, X.C. Wu, Y.Q. Yang, H.X. Wu, S.W. Guo, J.Y. Zhang, Nanoscale 5 (2013) 1816–1819.
Please cite this article as: S. Ge, et al., Talanta (2015), http://dx.doi.org/10.1016/j.talanta.2015.05.008i