pyridine Hexafluorophosphate Ionic Liquid Functionalized Gold

Article pubs.acs.org/ac

4‑Amino-1-(3-mercapto-propyl)-pyridine Hexafluorophosphate Ionic Liquid Functionalized Gold Nanoparticles for IgG Immunosensing Enhancement Rui Li,† Kangbing Wu,‡ Changxian Liu,† Yin Huang,† Yanying Wang,† Huaifang Fang,† Huijuan Zhang,† and Chunya Li*,† †

Key Laboratory of Analytical Chemistry of the State Ethnic Affairs Commission, College of Chemistry and Materials Science, South-Central University for Nationalities, Wuhan 430074, China ‡ Key Laboratory for Large-Format Battery Materials and System, Ministry of Education, School of Chemistry and Chemical Engineering, Huazhong University of Science and Technology, Wuhan 430074, China S Supporting Information *

ABSTRACT: A novel ionic liquid, 4-amino-1-(3-mercapto-propyl)pyridine hexafluorophosphate (AMPPH), was successfully synthesized and characterized. Subsequently, AMPPH was used as a functional monomer to fabricate AMPPH-modified gold nanoparticles (AMPPH−AuNPs) via a one-pot synthesis method. The as-prepared AMPPH−AuNPs were confirmed with transmission electron microscopy and X-ray photoelectron spectroscopy. AMPPH−AuNPs were used to construct a biocompatible interface to immobilize rabbit anti-human IgG (anti-HIgG) onto a glassy carbon electrode (GCE) surface, followed by a cross-linking step with glutaraldehyde to fabricate an anti-HIgG−AMPPH−AuNPs/ GCE. The nonspecific binding sites were enclosed with bovine serum albumin (BSA) to develop an immunosensor for human IgG. Electrochemical impedance spectroscopy, cyclic voltammetry and differential pulse voltammetry were used to investigate the electrochemical properties of the developed immunosensor. The results indicate that AMPPH−AuNPs can improve the immunosensing performance. The current response of the immunosensor was found linearly related to human IgG concentration in the range of 0.1−5.0 ng mL−1 and 5.0−100.0 ng mL−1. The detection limit is estimated to be 0.08 ng mL−1 (S/N = 3). The obtained immunosensor was successfully applied to the analysis human IgG immunoglobulin in human serum, and the results were well consistent with ELISA method.

H

techniques, immunoassay, which based on the highly specific molecular recognition of antigens by antibodies, has attracted more and more attention in recent years because of its excellent selectivity, high sensitivity and low detection limit.8−10 Conventional immunoassay methods mainly consist of radial immunodiffusion, fluorescent immunoassays, (electro)chemiluminescent immunoassays and enzyme-linked immunoassays.11−17 Despite many significant results have been reported, there still exist some limitations such as the short shelf life of 125I-labeled antibody, the radiation hazards, the complicated wash procedure, and the requirements for a long analysis time, expensive and cumbersome instruments, and/or skillful operators.18 Thus, the application of immunoassay techniques in fast, online, or fully automated determination was necessitated. Consequently, there is also a demand of new

uman IgG is a kind of important immunoglobulin mainly in human serum. Serum immunoglobulin levels are determined routinely in clinical practice because they provide key information on the humoral immune status. Low immunoglobulin levels define some humoral immunodeficiencies.1 However, liver diseases, chronic inflammatory diseases, infections, and malignancies will cause high immunoglobulin levels.2 Human IgG also has significance in the immunodeficiency disease patients and lymphocytic leukemia patients.3,4 Thus, it is very important to generate reliable and precise analytical technology for measuring human IgG in medical diagnosis and biological research. For example, serum immunoglobulin and IgG subclass concentrations were determined in more than 500 healthy Turkish children.5 When children are older than 16 years, the IgG concentration is in the range of 8.30−18.20 mg mL−1 with geometric mean of 12.249 ± 2.802 mg mL−1, which is similar to levels in adults. Effects of sex and age (or alcohol consumption and smoking) on serum concentration of human serum immunoglobulin were also investigated.6,7 Among different types of analytical © XXXX American Chemical Society

Received: January 2, 2014 Accepted: May 6, 2014

A

dx.doi.org/10.1021/ac500024n | Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Beijing Biosynthesis Biotechnology Co., LTD (Beijing, China). Other chemicals were analytical grade and were used as received. Experiments were carried out at room temperature without other statements. FTIR spectra were conducted on Nicolet NEXUS-470 FTIR spectrometer (Thermo Nicolet, USA). UV−vis analysis was performed on PE Labmda Bio 35 (PerkinElmer, USA). NMR was carried out with Avance 400 MHz NMR Spectrometer (Bruker, Switzerland). Mass spectra were acquired using an Agilent 6520 ESI quadrupole time-of-flight mass spectrometer with a standard ESI source in the positive-ion mode (Agilent Technologies, Inc., USA). The samples were directly introduced into the ESI source by a syringe pump. The gas temperature was 300 °C, the flow of drying gas was 10 L min−1, and the nebulizer was at 30 psi. The voltage of the capillary was 3500 V and the fragmentor was 125 V. Transmission electron microscopy (TEM) images were obtained on a FEI Tecnai G2 20S-TWIN instrument (FEI Company, Netherlands) operating at an acceleration voltage of 200 kV. X-ray photoelectron spectroscopy (Thermo Electron Corp., USA) was used to analysis the composition of AMPPH−AuNPs. Electrochemical measurements were performed with CHI 660E electrochemical workstation (Chenhua Corp., Shanghai, China). A conventional three-electrode system was employed with a modified glassy carbon working electrode (3.0 mm in diameter), a Pt wire auxiliary electrode, and a saturated calomel reference electrode (SCE). Synthesis of AMPPH Ionic Liquid. 4-Aminopyridine was dissolved into acetonitrile, and was added dropwise to acetonitrile containing 1,3-dibromopropane within 1 h. Then, the resulting solution was stirred at 70 °C for 12 h under the protection of nitrogen. After it was cooled to room temperature, the mixture was filtered, and the solid products which obtained from the filtrate evaporated under vacuum were thoroughly washed with diethyl ether to give 4-amino-1-(3bromide-propyl)-pyridine bromide salt. Subsequently, 4-amino1-(3-bromide-propyl)-pyridine bromide salt was reacted with potassium thioacetate using acetonitrile as solvent. The mixture was separated by filtering. The organic phase was dried over MgSO4 and evaporated under vacuum to afford 4-amino-1-(3thioacetate-propyl)-pyridine bromide salt as a pale yellow oil. A solution containing the 4-amino-1-(3-thioacetate-propyl)-pyridine bromide was stirred at 65 °C for 0.5 h under the protection of nitrogen. Then, ammonium hexafluorophosphate solution was added dropwise to it with a pale yellow precipitate separating out. The mixture was stirred at room temperature for 2 h and successively separated by filtering. The crude solid products were thoroughly washed with diethyl ether and dried under vacuum to obtain 4-amino-1-(3-thioacetate-propyl)pyridine hexafluorophosphate ionic liquid. 2.2 mL of 1.0 mol L−1 NaOH solution was homogeneously dissolved in 10 mL degassed methanol, and was then dropwise added into 4amino-1-(3-thioacetate-propyl)-pyridine hexafluorophosphate ionic liquid (1.0 mmol) methanol solution at 0 °C under nitrogen. After being stirred for 3 h under nitrogen protection, 1 mL of 1.0 mol L−1 HCl solution was diluted with degassed methanol (10 mL) and added dropwise, and then the solvent was removed under vacuum. 1.0 mol L−1 HCl was added to the remaining residue and the mixture was stirred for 2 h and then separated by filtering. The crude solid products were first washed with a small amount of ultrapure water, filtered and dried. Subsequently, the solid product was thoroughly washed with diethyl ether, filtered and dried under vacuum to give the

schemes and strategies for improvement of the sensitivity and simplicity of clinical immunoassays. Compared with conventional immunoassay techniques, electrochemical immunosensors based on the antigen−antibody interaction have attracted intensive research interest.19−21 Via the coupling of highly specific recognition events to appropriate transducers, electrochemical immunosensors are considered as specific, simple, label-free and direct detection techniques.22 The most important procedure responsible for the sensing performance of immunosensor is the design and fabrication of the interface between the biocomponents and the transducer. Electrochemical immunosensors are normally constructed by immobilization of biomolecules onto the electrode surface using noncovalent and covalent binding, and polymer entrapment etc.23−26 Most antibodies and antigens are intrinsically unable to act as redox partners. Thus, electron mediators should also be co-immobilized onto the electrode interface to promote electron transfer and improve sensitivity and stability. Nanogold has been demonstrated to be one of the most promising nanomaterials for the fabrication of electrochemical immunosensors.27−29 Nanogold possesses unique properties, including easy preparation, good biocompatibility, excellent conductivity, and large surface area, leading to increase in functional density of biomolecules and facilitate electron exchange.30−32 Ionic liquids are a class of materials composed of cations and anions that exist in the liquid state around room temperature. They possess many unique properties, such as high ionic and electrical conductivity, high ionic mobility, excellent thermal and chemical stability, wide electrochemical potential window, and the ability to facilitate direct electron transfer reactions.33 In addition, ionic liquids also show good biocompatibility and promotion effect on the bioactivity and stability of biomacromolecules.34,35 The work presented here takes advantage of the properties of nanogold and ionic liquid, which were expected to enhance immobilization and bioactivity of the antibodies, thus to improve sensing properties of immunosensor. In this work, 4-amino-1-(3-mercapto-propyl)-pyridine hexafluorophosphate ionic liquid was developed as a functional monomer to fabricate gold nanoparticles. The chemisorption of thiol ionic liquid on gold nanoparticle surface provides a stable and conductive amino interface that can then be activated for covalent coupling of anti-HIgG. As a result, the stability of antiHIgG immobilized on the electrode surface will increase accordingly, which is a crucial element for human IgG immunosensor. Use of ionic liquid and nanogold was also expected to improve the pathway of electron transfer, and enhance the biocompatibility and bioactivity of the immobilized anti-HIgG. The immunosensor was also applied to analysis human IgG immunoglobulin in human serum, and the results were well consistent with ELISA method. All of these results indicated that ionic liquid functionalized gold nanoparticle is a suitable substrate for the biomolecules immobilization, and opens a quite effective way for immunosensor fabrication.

■

EXPERIMENTAL SECTION Apparatus and Reagents. 4-Aminopyridine, chloroauric acid, bovine serum albumin (BSA), ammonium hexafluorophosphate, and glutaraldehyde were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). 1,3Dibromopropane was obtained from Aladdin (Shanghai, China). Human IgG, rabbit anti-human IgG, prostate specific antigen (PSA), and α-fetoprotein (AFP) were bought from B



Article

Figure 1. Characterization of AMPPH (a) and AMPPH−AuNPs (b) with FTIR (A), UV−vis (B), and X-ray photoelectron (C) spectroscopy and transmission electron microscopy (D).

dried at 4 °C. Then the anti-HIgG−AMPPH−AuNPs composite film modified electrode was treated with glutaraldehyde vapor for 5 min to immobilize anti-HIgG onto AMPPH−AuNPs surface via a cross-linking reaction. After that, the modified electrode was incubated into bovine serum albumin solution to block all unspecific sites to produce a human IgG immunosensor which was denoted as BSA/antiHIgG−AMPPH−AuNPs/GCE. The obtained human IgG immunosensor was fully washed with ultrapure water to remove all chemicals which were physically absorbed. When being used for specific determination, BSA/anti-HIgG− AMPPH−AuNPs/GCE was incubated into a human IgG solution for 30 min, and was washed thoroughly with PBS. Then electrochemical response signals was recorded using K3[Fe(CN)6]/K4[Fe(CN)6] as electroactive probes. The fabrication procedures and the electrochemical reaction mechanism were presented in Supporting Information Scheme S2. Electrochemical Measurements. Electrochemical experiments were carried out in a mixture solution of 0.1 mol L−1 KCl and 0.01 mol L−1 PBS (pH 7.0) containing 5.0 × 10−3 mol L−1 K3Fe(CN)6/K4Fe(CN)6. Cyclic voltammograms of the immunosensors before and after being incubated into human IgG solution were recorded in the potential range from −0.2 to 0.6 V. Differential pulse voltammograms of the immunosensors in the potential range from −0.1 to 0.5 V were also recorded. The difference of the oxidation peak current before and after being interacted with human IgG was used for quantitative analysis. The pulse amplitude, pulse period, and pulse width of differential pulse voltammetry were 50 mV, 0.2 s, and 50 ms, respectively. ELISA Assay. Blood serum samples were supplied by the central hospital of Wuhan (Wuhan, China). Before being used for analysis, the blood serum samples were diluted with PBS at

desired white loose solid-product: 4-amino-1-(3-mercaptopropyl)-pyridine hexafluorophosphate (yield: 79.3%). The product was characterized with 1H-NMR, 13C-NMR, and HPLC-MS. The results are shown in Supporting Information Figures S1−S3. Some data were presented as follows: 1H NMR (D2O) 7.95 (d, 1H), 7.93 (d, 1H), 6.80 (d, 1H), 6.78 (d, 1H), 4.19 (t, 2H), 2.48 (t, 2H), 2.10 (quint, 2H); 13C NMR (DMSO) 159.09, 143.36, 109.97, 56.15, 20.72; m/z = 169.082. Synthesis of AMPPH Functionalized Gold Nanoparticles. Ten milliliters of 30.0 mmol L−1 4-amino-1-(3mercapto-propyl)-pyridine hexafluorophosphate solution was added to 10 mL of 10.0 mmol L−1 HAuCl4 solution in dropwise. The solution was mixed homogeneously by vigorous stirring. Then, 10 mL of 60.0 mmol L−1 fresh prepared NaBH4 solution was injected into the mixture with stirring. The color of the solution was transferred from yellow to reddish brown. The as-produced dispersion solution was centrifuged for 2 min at 8000 rpm to remove the surplus reagents. The residual solid materials were washed thoroughly with ultrapure water, and dried under vacuum conditions to give AMPPH−AuNPs. The synthetic routes for AMPPH ionic liquid and AMPPH−AuNPs are shown in Supporting Information Scheme S1. Fabrication of IgG Immunosensors. Prior to modification, glassy carbon electrode was polished with emery paper and alumina slurries followed by rinsing thoroughly with ultrapure water. The electrodes were successively ultrasonicated in nitric acid, ethanol and water, and then allowed to dry at room temperature. AMPPH−AuNPs were homogeneously dispersed into phosphate buffer solution (PBS) with ultrasonication to give solution at concentration of 1.0 mg mL−1. In subsequent, equivalent volume (4.0 μL) of AMPPH−AuNPs solution and rabbit anti-human IgG solution (0.5 μg mL−1) were thoroughly mixed. 8.0 μL of the mixed solution was drop coated onto the cleared glassy carbon electrode surface, and C



Article

Electrochemical impedance spectra of the AMPPH−AuNPs/ GCE (a), the unmodified GCE (b), anti-HIgG−AMPPH− AuNPs/GCE (c), BSA/anti-HIgG−AMPPH−AuNPs/GCE (d), and IgG/BSA/anti-HIgG−AMPPH−AuNPs/GCE (e) were conducted in a mixture solution containing 5.0 × 10−3 mol L−1 K3Fe(CN)6/ K4Fe(CN)6, 0.1 mol L−1 KCl and 0.01 mol L−1 PBS with frequency varied from 100 kHz to 0.1 Hz at a potential of 0.22 V. As shown in Figure 2, a semicircle at higher

the ratio of 1:1000000. Human IgG concentration of each sample was also measured using Olympus AU5421 automatic biochemical analyzer (Olympus Optical Co., Ltd., Japan). Enzyme-linked immunosorbent assays were performed using commercial kits (Leadman Group Co., Ltd., China) according to the manufacturer’s instructions.

■

RESULTS AND DISCUSSION FTIR spectra of AMPPH (a) and the AMPPH−AuNPs (b) were displayed in Figure 1A. In curve a, the characteristic peaks at 3393 and 3452 cm−1 are corresponding to the asymmetric stretching vibration and symmetric stretching vibration of primary amine. The bending vibration of N−H was observed at 1660 cm−1. The stretching vibration of CN (νCN at 1542 cm−1), CC (νCC at 1469 cm−1), C−N (νC−N at 1109 cm−1), C−C (νC−C at 1193 cm−1) and characteristic absorption (δring + νring + νC−NH2 at 838 cm−1) indicate the existence of pyridine ring.36,37 The characteristic peak at 2337 cm−1 is corresponding to S−H stretching vibration. This 2337 cm−1 peak disappeared in curve b, inferring the replacing of the S−H bond by new bond between AMPPH and gold nanoparticles. In addition, the characteristic absorptions of primary amine and pyridine are retained. The above results indicate that the AMPPH has been successfully modified onto the gold nanopartilces surface. The conjugation of AMPPH and gold nanoparticles was also confirmed by UV−vis spectroscopy in an aqueous solution. As shown in Figure 1B, UV−vis spectrum of AMPPH (curve a) showed a sharp absorption peaks at 270 nm, owing to the π → π* transition of pyridine heterocycle. In the case of AMPPH− AuNPs nanocomposites (curve b), in addition to the 270 nm sharp absorption, a broad peak at around 540 nm was known to the characteristic absorption of gold nanoparticles, in agreement with the color changing from bright yellow to reddish brown during conjugation. All results illustrated the interaction between AMPPH ionic liquid and gold nanoparticles to form AMPPH−AuNPs nanocomposite. We also applied X-ray photoelectron spectroscopy (XPS) to verify the ionic liquid and the followed conjugate. XPS spectra are given in Figure 1C. It was obvious to see from curve a that C1s levels at 284.6 eV, N1s level at 403.8 eV, S2p level at 163.2 eV, P2p level at 136.2 eV, and F1s level at 686.1 eV were existed in the AMPPH ionic liquid, meaning the successful synthesis of the target ionic liquid. Curve b shows the existence of elements of C, N, P, S, and F signals and a strong peak of Au4f at 83.3 eV. The results indicate the coexistence of Au and AMPPH in the AMPPH−AuNPs nanocomposite, meaning the successful functionalization of gold nanoparticles surfaces with AMPPH ionic liquid. Transmission electron microscopy (TEM) was utilized to investigate the morphology of the as-prepared AMPPH− AuNPs. AMPPH−AuNPs were dispersed into a mixture solution of ethanol and water (V/V, 1:1) with ultrasonic stirring to prepare a sample for TEM characterization. A typical TEM image of AMPPH−AuNPs is shown in Figure 1D. It was shown that all AMPPH−Au nanoparticles are monodispersed in solution. The diameters of the as-prepared AMPPH−AuNPs are found in the range from 2 to 4 nm. The average diameter is calculated to be 2.92 ± 0.54 nm. After being stored for several months, no obvious change in color and dispersed state of the solution was observed, meaning good stability of AMPPH− AuNPs.

Figure 2. Nyquist plots of AMPPH−AuNPs/GCE (a), the unmodified GCE (b), anti-HIgG−AMPPH−AuNPs/GCE (c), BSA/anti-HIgG− AMPPH−AuNPs/GCE (d) and IgG/BSA/anti-HIgG−AMPPH− AuNPs/GCE (e) in a mixture solution containing 5.0 × 10−3 mol L−1 K3Fe(CN)6/K4Fe(CN)6, 0.1 mol L−1 KCl and 0.01 mol L−1 PBS. Inset is the Randles equivalent circuit. Potential: 0.22 V.

frequencies and a linear curve at lower frequencies were obtained at each electrode. The Randles equivalent circuit, the inset in Figure 2, was further chosen to fit the impedance data obtained in the experiments using ZSIMPWIN300 software, where Rs is the uncompensated solution resistance, Q is the constant phase element, Ret is the charge transfer resistance, Zw is the Warburg impedance. From curve a to e, the fitted values of Ret for K3Fe(CN)6/K4Fe(CN)6 at the these electrodes are 136.8, 194.6, 447.4, 842.1, and 1045.6 ohm, respectively. It was obvious to see that Ret for the unmodified GCE (curve b) is larger than the electrode modified with AMPPH−AuNPs (curve a). The main reason can be attributed to the superior conductivity of ionic liquid and gold nanoparticles. In addition, the electrostatic attraction between the 4-amino-1-(3-mercaptopropyl)-pyridine cations and ferricyanide anions may also contribute to this phenomenon. The Ret value increased significantly after the immobilization of anti-HIgG (curve c). After all unspecific sites being blocked with bovine serum albumin, the electron transfer resistance was further increased (curve d), meaning the successful reaction between BSA and the remained aldehyde group. In the following step, the fabricated immunosensor was incubated into human IgG solution. Immunoreaction between human IgG and antiHIgG immobilized on the electrode surface will occur. The binding leads to the further increasing of electron transfer resistance of K3Fe(CN)6/K4Fe(CN)6 (curve e). Figure 3 presents the cyclic voltammograms of 5.0 × 10−3 mol L−1 K3Fe(CN)6/K4Fe(CN)6 at the AMPPH−AuNPs/ GCE (a), the unmodified GCE(b), anti-HIgG−AMPPH− AuNPs/GCE (c), BSA/anti-HIgG−AMPPH−AuNPs/GCE (d), and IgG/BSA/anti-HIgG−AMPPH−AuNPs/GCE (e). It obviously to see the redox peak current is higher for AMPPH− AuNPs/GCE, in compared with unmodified GCE. With the D



Article

Information Figure S5. It was obvious to see that the peak current response of the immunosensor based on AMPPH− AuNPs is higher than others, meaning an enhanced effect can be obtained using AMPPH−AuNPs as interface to immobilize rabbit anti-human IgG. Influence of the scan rate on the cyclic voltammetric behaviors of K3Fe(CN)6/K4Fe(CN)6 at the IgG/BSA/antiHIgG−AMPPH−AuNPs/GCE was presented in Supporting Information Figure S6 (A). With the scan rate increasing from 0.02 to 0.6 V s−1, the oxidation peak slightly shifts toward positive potential. Reversely, negative shift was found for the reduction peak. The corresponding of peak current related to scan rate was plotted in Supporting Information Figure S6 (B). The peak currents were linearly related to the square root of scan rate in the studied range, indicating that the redox reaction of K3Fe(CN)6/K4Fe(CN)6 was diffusion-controlled. We further investigate the performance dependence of the immunosensor on these factors to obtain the best sensitivity: amount of coated anti-HIgG−AMPPH−AuNPs, incubation time, the pH value of the used PBS and incubation temperature. The volume of anti-HIgG−AMPPH−AuNPs coated to the electrode surface will influence the film thickness and the binding sites for immuoreaction, thus affecting the response sensitivity toward human IgG. Using the peak current difference (ΔI) of the immunosensor toward K3Fe(CN)6/ K4Fe(CN)6 before and after being interacted with 50.0 ng mL−1 of human IgG as standard, the relationship between the sensitivity and the coated volume was evaluated. It can be seen from Figure 4A that the ΔI values increased significantly with the coated volume increasing from 4 to 8 μL. The enhancement

Figure 3. Cyclic voltammograms of AMPPH−AuNPs/GCE (a), the unmodified GCE (b), anti-HIgG−AMPPH−AuNPs/GCE (c), BSA/ anti-HIgG−AMPPH−AuNPs/GCE (d), and IgG/BSA/anti-HIgG− AMPPH−AuNPs/GCE (e) in a mixture solution containing 5.0 × 10−3 mol L−1 K3Fe(CN)6/K4Fe(CN)6, 0.1 mol L−1 KCl, and 0.01 mol L−1 PBS at scan rate of 0.1 V s−1.

immobilization, blocking, and incubation steps followed, the redox peak currents droped gradually. The phenomenon is well consistent with the result from electrochemical impedance spectroscopy. Cyclic voltammograms of 5.0 × 10−3 mol L−1 K3Fe(CN) 6/K4Fe(CN)6 at immunosensors, which were fabricated with AMPPH−AuNPs (a), AuNPs (b) and AMPPH ionic liquid (c), before (solid line) and after (dash line) being interacted with 50.0 ng mL−1 of human IgG are shown in Supporting Information Figure S4. The oxidation peak current differences were also presented in Supporting

Figure 4. Relationship between ΔI values and the coating volume of anti-HIgG−AMPPH−AuNPs (A), the incubation time (B), pH value (C), and the incubation temperature (D). Error bars represent standard deviation, n = 3. E



Article

Figure 5. Differential pulse voltammograms of the immunosensing system incubated in human IgG solution with different concentrations (A) and calibration curves for human IgG determination (B). Error bars represent standard deviation, n = 3.

linear relationship between the ΔI values and IgG antigen concentration was obtained in the range of 0.1−5.0 and 5.0− 100.0 ng mL−1, and the equations can be expressed as following: ΔI (μA) = 1.0114 c (ng mL−1) + 0.0088 (R = 0.9984) and ΔI (μA) = 0.1423 c (ng mL−1) + 5.093 (R = 0.9989).The detection limit is estimated to be 0.08 ng mL−1 (S/N = 3). In the high concentration range of IgG, the ΔI values would tend to be stable due to the saturation binding of IgG to the IgG antibodies immobilized on the electrode surface. The sensing performance of the developed immunosensor was compared with the previous reports. As shown in Supporting Information Table S1, it can be found that the immunosensor based on AMPPH−AuNPs exhibited a wider linear range and lower detection limit than some of them. In addition, the analytical characteristics of the immunosensor were also compared with some commercial kits developed for human IgG. The dynamic range of this method is roughly consistent with that of these commercial kits, shown in Supporting Information Table S2, meaning a suitable range for diagnostics. The normal human IgG level in human serum is higher than the detection limit of the developed method. To evaluate the IgG levels in patients, we can dilute the human serum sample accordingly. The selectivity of the developed immunosensor for the detection of human IgG was tested against the prostate specific antigen (PSA) and α-fetoprotein (AFP). Immunosensors were incubated into a mixture of human IgG and PSA (or AFP) in PBS for 30 min followed by the thorough washing with buffer solution. As shown in Supporting Information Table S3, the oxidation peak currents of 5.0 × 10−3 mol L−1 K3(FeCN)6/ K4(FeCN)6 at the BSA/anti-HIgG−AMPPH−AuNPs/GCE after being incubated in a mixture solution containing 50.0 ng mL−1 of human IgG and 50.0 ng mL−1 of PSA (or AFP) were similar to that obtained for only 50.0 ng mL−1 of human IgG. These results demonstrate that the immunosensor possesses a good selectivity in detecting human IgG against PSA and AFP in PBS. In order to verify the blocking effect of BSA on the unspecific sites, a human IgG immunosensor without being treated with BSA was fabricated. Sensing performances toward 50.0 ng mL−1 of human IgG, PSA and AFP were also investigated. Seen from Supporting Information Figure S7, the oxidation peak current obviously decreased whether human IgG or PSA (AFP) was used as model antigen to interact with this immunosensor. Although the ΔI value of human IgG is higher than others, both PSA and AFP almost lead to the equivalent decrease on the peak current. The facts mean PSA and AFP can be adsorbed onto the immunosensor surface

in the ΔI values should be ascribed to more binding sites supplied for immunoreaction and thus caused more human IgG accumulated to the electrode surface through incubation. Accordingly, the poor conductivity of human IgG leads to the dramatic decrease in current response of K3Fe(CN)6/ K4Fe(CN)6. Further increasing the coated volume from 8 to 12 μL, the ΔI values tend to be stable. For GCE in a diameter of 3 mm, the suitable amount is around 8 μL. The incubation time for the immunoreaction between 50.0 ng mL−1 of human IgG and BSA/anti-HIgG−AMPPH− AuNPs/GCE was chosen based on the effective decrease of peak current after the immunoreaction. It obviously seen from Figure 4B, upon the addition of IgG solution, the peak current difference (ΔI) increased within 30 min and then reached a plateau. Thus, the optimal incubation time for further investigation was chosen to be 30 min. The pH value of the buffer solution greatly affected the electrochemical behavior of immunosensor. In order to optimize the pH value, a series of PBS with the pH value ranged from 5.0 to 8.0 were studied. The immunosensing performances toward 50.0 ng mL−1 of human IgG were tested. As shown in Figure 4C, the ΔI value increases with the variation of pH from 5.0 to 7.0, and then decreases with the variation of pH from 7.0 to 8.0. The experimental results show that the maximum response appears at about pH 7.0 as it is close to the pH of body fluid. Thus, we then choose PBS with a pH 7.0 in the following tests. Incubation temperature can have strong effects on the immunoreaction and further influence the response sensitivity of the fabricated biosensor. As shown in Figure 4D, it was found that the ΔI values of 50.0 ng mL−1 of human IgG increases with the variation of incubation temperature from 20 to 35 °C. Nevertheless, a slight decrease in ΔI values was observed in the incubation temperature varied from 35 to 40 °C, meaning that a high temperature may damage the structure of proteins leading to the ineffective binding of IgG. Although 37 °C might be most suitable for immunoreaction, the practical operating incubation temperature for the immunosensor was chosen to be 35 °C. Differential pulse voltammetry was used to investigate the relationship between the ΔI values and human IgG concentration. Figure 5A shows the voltammograms of K3Fe(CN)6/K4Fe(CN)6 at the as-prepared immunosensor after being interacted with human IgG at different concentration. It was found that the peak current response of electrochemical probes decreased with the human IgG concentration increasing. As presented in Figure 5B, a good F



Article

Table 1. Results of Real Sample Assay and Recoveries (n = 5) this method

ELISA

recovery test

samples

c (mg L−1)

RSD (%, n = 5)

c (mg L−1)

added (mg L−1)

detected (mg L−1)

recovery (%)

RSD (%, n = 5)

1 2 3 4

13.19 15.72 18.95 20.90

2.63 3.08 3.32 3.17

13.7 16.4 18.9 20.4

5.0 5.0 5.0 5.0

17.92 20.53 24.15 26.15

94.6 96.3 104.0 105.0

2.47 1.51 2.28 2.03

might provide a promising potential for application in biological or clinical target analysis.

through nonspecific interaction. Thus, the selectivity of this immunosensor is poorer than that had been blocked with BSA. After being incubated into 50.0 ng mL−1 of human IgG solution for 30 min, the reproducibility of the immunosensor was investigated by measuring the oxidation peak currents of 5.0 × 10−3 mol L−1 K3(FeCN)6/K4(FeCN)6 with five immunosensors freshly prepared. A good reproducibility in the oxidation peak currents with the RSD of 3.45% was obtained. A RSD value of 3.04% (n = 5) was also obtained for the determination of 50.0 ng mL−1 human IgG with an immuosensor which was regenerated by stirring in 0.2 mol L−1 glycine hydrochloric acid buffer solution (pH 2.8) for approximately 6 min. These results demonstrated that the immunosensor fabrication procedure was reliable. Reproducible responses can be obtained with different immunosensors fabricated independently or regeneration. The storage stability of the immunosensor was also investigated by comparing with immunosensing of 50.0 ng m L−1 human IgG. After 3 weeks storage, about 89.5% of the original reaction activity retained, meaning good storage stability for the immunosensor at 4 °C. To evaluate the practical application of the developed immuosystem, the concentration of human IgG in human serum sample was determined. To verify the accuracy of the developed immuobisensor, a recovery test method and ELISA method were also employed to assay these human serum samples. All results were summarized in Table 1. From the listed results, it was clearly to see that good reproducibility is obtained for real samples analysis. For the designated concentration of human IgG which was spiked into human serum, the recoveries ranged from 94.6% to 105.0% mean a good accuracy for IgG analysis. T-test method was also used to compare the average values obtained by the immunosensor and ELISA method. The calculated t value was 2.01 (confidence coefficient 95%), which was less than the standard value (2.78). These data indicate no systematic error in measurements. From the comparative results of two methods, we can see they are in good agreement, indicating the as-prepared immunosensor is reliable and accurate for human IgG determination.

■

ASSOCIATED CONTENT

S Supporting Information *

Additional material as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: + 86 27 67842752. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge the financial supports from Natural Science Foundation of China (No. 21275166, 21375041, and 21275165), Research foundation of State General Administration of The People’s Republic of China for Quality Supervision and Inspection and Quarantine (No. 2013Qk286).

■

REFERENCES

(1) Buckley, R. H. Clin. Immunol. Immunopathol. 1986, 40, 13−24. (2) Dispenzieri, A.; Gertz, M. A.; Therneau, T. M.; Kyle, R. A. Mayo Clin. Proc. 2001, 76, 476−87. (3) Orange, J. S.; Hossny, E. M.; Weiler, C. R.; Ballow, M.; Berger, M.; Bonilla, F. A.; Buckley, R.; Chinen, J.; El-Gamal, Y.; Mazer, B. D.; Nelson, R. P.; Patel, D. D.; Secord, E.; Sorensen, R. U.; Wasserman, R. L.; Cunningham-Rundles, C. C. J. Allergy. Clin. Immunol. 2006, 117, S525−S553. (4) Svensson, T.; Höglund, M.; Cherif, H. Scand. J. Infect. Dis. 2013, 45, 537−542. (5) Aksu, G.; Genel, F.; Koturoglu, G.; Kurugol, Z.; Kutukculer, N. Turk. J. Pediatr. 2005, 47, 19−24. (6) Cassidy, J. T.; Nordby, G. L.; Dodge, H. J. J. Chronic Dis. 1974, 27, 507−516. (7) Gonzalez-Quintela, A.; Alende, R.; Gude, F.; Campos, J.; Rey, J.; Meijide, L.M.; Fernandez-Merino, C.; Vidal, C. Clin. Exp. Immunol. 2008, 151, 42−50. (8) Morin, I.; Askin, S. P.; Schaeffer, P. M. Analyst 2011, 136, 4815− 4821. (9) Wei, X. L.; Zhang, Y. L.; Jiang, Z. L. Spectrosc. Spect. Anal. 2009, 29, 1640−1642. (10) Mo, Z. H.; Huang, H. J.; Qian, J. Z.; Zhu, L. H. Chem. J. Chinese Univ. 2007, 28, 649−651. (11) Jafri, L.; Khan, A. H.; Siddiqui, A. A.; Mushtaq, S.; Iqbal, R.; Ghani, F.; Siddiqui, I. Clin. Biochem. 2011, 44, 864−868. (12) Bodle, J.; Verity, E. E.; Ong, C.; Vandenberg, K.; Shaw, R.; Barr, I. G.; Rockman, S. Influenza Other Respir. Viruses 2013, 7, 191−200. (13) Iizumi, Y.; Okazaki, T.; Ikehara, Y.; Ogura, M.; Fukata, S.; Yudasaka, M. ACS Appl. Mater. Interfaces 2013, 5, 7665−7670. (14) Beloglazova, N. V.; Shmelin, P. S.; Speranskaya, E. S.; Lucas, B.; Helmbrecht, C.; Knopp, D.; Niessner, R.; De Saeger, S.; Goryacheva, I. Y. Anal. Chem. 2013, 85, 7197−7204.

■

CONCLUSION To conclude, using 4-amino-1-(3-mercapto-propyl)-pyridine hexafluorophosphate ionic liquid as modifier, highly dispersed gold nanoparticles were synthesized and confirmed with XPS and TEM. Rabbit anti-human IgG was immobilized onto the nanointerface based on AMPPH−AuNPs, and used for human IgG immuosensing. The immunosensor presented low level of nonspecific reaction, which allowed a superior performance for human IgG assay in a complex sample such as human blood serum directly. The behavior of the immunosensor correlated excellently with classical ELISA methods in the capability to quantitatively determine human IgG contents from real serum samples. The investigation also suggests that AMPPH−AuNPs would be suitable to build a biocompatible platform for various types of biomaterials and easy to achieve biosensors, which G



Article

(15) Akhavan-Tafti, H.; Binger, D. G.; Blackwood, J. J.; Chen, Y.; Creager, R. S.; de Silva, R.; Eickholt, R. A.; Gaibor, J. E.; Handley, R. S.; Kapsner, K. P.; Lopac, S. K.; Mazelis, M. E.; McLernon, T. L.; Mendoza, J. D.; Odegaard, B. H.; Reddy, S. G.; Salvati, M.; Schoenfelner, B. A.; Shapir, N.; Shelly, K. R.; Todtleben, J. C.; Wang, G. P.; Xie, W. H. J. Am. Chem. Soc. 2013, 135, 4191−4194. (16) Liao, N.; Zhuo, Y.; Chai, Y. Q.; Xiang, Y.; Han, J.; Yuan, R. Biosens. Bioelectron. 2013, 45, 189−194. (17) Gan, S. D.; Patel, K. R. J. Invest. Dermatol. 2013, 133, E10−E12. (18) Yu, H.; Yan, F.; Dai, Z.; Ju, H. X. Anal. Biochem. 2004, 331, 98− 105. (19) Cai, Y. Y.; Li, H.; Du, B.; Yang, M. H.; Li, Y.; Wu, D.; Zhao, Y. F.; Dai, Y. X.; Wei, Q. Biomaterials 2011, 32, 2117−2123. (20) Chen, J.; Tang, J. H.; Yan, F.; Ju, H. X. Biomaterials 2006, 27, 2313−2321. (21) Elshafey, R.; Tavares, A. C.; Siaj, M.; Zourob, M. Biosens. Bioelectron. 2013, 50, 143−149. (22) Liang, W. B.; Yi, W. J.; Li, S. H.; Yuan, R.; Chen, A.; Chen, S.; Xiang, G. M.; Hu, C. M. Clin. Biochem. 2009, 42, 1524−1530. (23) Hays, H. C. W.; Millner, P. A.; Prodromidis, M. I. Sens. Actuators, B 2006, 114, 1064−1070. (24) Liu, Y. C.; Wang, C. M.; Hsiung, K. P.; Huang, C. J. Biosens. Bioelectron. 2003, 18, 937−942. (25) Batalla, P.; Bolivar, J. M.; Lopez-Gallego, F.; Guisan, J. M. J. Chromatogr. A 2012, 1262, 56−63. (26) Akter, F.; Mie, M.; Grimm, S.; Nygren, P. A.; Kobatake, E. Anal. Chem. 2012, 84, 5040−5046. (27) Ding, Y. J.; Liu, J.; Wang, H.; Shen, G. L.; Yu, R. Q. Biomaterials 2007, 28, 2147−2154. (28) Li, R. Y.; Xia, Q. F.; Li, Z. J.; Sun, X. L.; Liu, J. K. Biosens. Bioelectron. 2013, 44, 235−240. (29) Jeong, B.; Akter, R.; Han, O. H.; Rhee, C. K.; Rahman, M. A. Anal. Chem. 2013, 85, 1784−1791. (30) Ding, Y.; Chen, M. W.; Erlebacher, J. J. Am. Chem. Soc. 2004, 126, 6876−6877. (31) Li, F.; Feng, Y.; Dong, P. J.; Yang, L. M.; Tang, B. Biosens. Bioelectron. 2011, 26, 1947−1952. (32) Liu, S. F.; Wang, L.; Zhao, F. J. Electroanal. Chem. 2007, 602, 55−60. (33) Li, R.; Liu, C. X.; Ma, M.; Wang, Z. G.; Zhan, G. Q.; Li, B. H.; Wang, X.; Fang, H. F.; Zhang, H. J.; Li, C. Y. Electrochim. Acta 2013, 95, 71−79. (34) Sun, W.; Wang, X. Z.; Sun, X. H.; Deng, Y.; Liu, J.; Lei, B. X.; Sun, Z. F. Biosens. Bioelectron. 2013, 44, 146−151. (35) Cai, Y. Y.; Li, H.; Du, B.; Yang, M. H.; Li, Y.; Wu, D.; Zhao, Y. F.; Zhao, Y. F.; Dai, Y. X.; Wei, Q. Biomaterials 2011, 32, 2117−2123. (36) Büyükmurat, Y.; Akyüz, S. J. Mol. Struct. 2001, 545, 563−564. (37) Dojer, B.; Pevec, A.; Jagodič, M.; Kristl, M.; Drofenik, M. Inorg. Chim. Acta 2012, 383, 98−104.

H


pyridine Hexafluorophosphate Ionic Liquid Functionalized Gold

Recommend Documents