- Email: [email protected]
Piezoelectric immunosensor based on gold nanoparticles capped with mixed self-assembled monolayer for detection of carcinoembryonic antigen
Thin Solid Films 518 (2010) 5010–5013
Contents lists available at ScienceDirect
Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f
Piezoelectric immunosensor based on gold nanoparticles capped with mixed self-assembled monolayer for detection of carcinoembryonic antigen Guangyu Shen ⁎, Jilin Lu College of Chemistry and Chemical Engineering, Hunan University of Arts and Science, Changde 415400, Hunan, China
a r t i c l e
i n f o
Article history: Received 17 June 2009 Received in revised form 12 March 2010 Accepted 12 March 2010 Available online 27 March 2010 Keywords: Gold nanoparticles Self-assembled monolayers Piezoelectric immunosensor Carcinoembryonic antigen
a b s t r a c t It is very important for a piezoelectric immunosensor to increase specific binding and decrease nonspecific adsorption. This study presents the development of such a piezoelectric immunosensor for the detection of carcinoembryonic antigen. An AT-cut quartz crystal's Au electrode surface was first modified with homogenous self-assembled monolayer of cysteamine (CE). Gold nanoparticles capped with mixed selfassembled monolayer of CE and MH (6-mercapto-1-haxanol) were then attached to the CE monolayer via glutaraldehyde (GA). Antibodies were immobilized onto a mixed self-assembled monolayer of CE and MH with GA as a reactive intermediate too. The binding of target antigens onto the immobilized antibodies decreased the sensor's resonant frequency, and the frequency shift was correlated to the antigen concentration. The stepwise assembly of the immunosensor was characterized by means of cyclic voltammetry technique. This immunoassay was shown to be specific and sensitive, thus providing a viable alternative to carcinoembryonic antigen detection method. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Carcinoembryonic antigen (CEA) is an important tumor marker responsible for clinical diagnosis of colorectal, pancreatic, gastric, and cervical carcinomas. Various immunoassays, such as enzyme-linked immunosorbent assay, radioimmunoassay, and chemiluminescence immunoassay (CLIA) have been developed for detecting CEA. However, all these techniques are time-consuming, expensive and/or requirement for sample pretreatment and concentration. So it is still significant and necessary to explore some simple, sensitive, and low-cost methods for the detection of carcinoembryonic antigen. Piezoelectric quartz crystal (PQC) devices have been attracting more and more attention for years, for they are portable, simple, cost-effective, and suitable for real-time monitoring biospecific interactions between antigen and antibody with high sensitivity and selectivity. Antibody immobilization is a vital part in development of a PQC immunosensor. The immobilization process must preserve the biological activity of the antibody, a more efficient binding and as well as reduce the nonspecific adsorption simultaneously. Current immobilization methods are mainly based on silanized layer [1], polymer membrane [2], Langmuir–Blodgett film [3], Protein A [4,5] self-assembled monolayer (SAM) [6–8] and nano-materials [9]. Of them, the SAM technique offers one of the simplest ways to provide a reproducible, ultrathin and well-ordered layer, however, pure SAM
⁎ Corresponding author. Tel.: + 86 7362563892. E-mail address: [email protected] (G. Shen). 0040-6090/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2010.03.077
may decrease capture efficiency and specificity due to high density of terminal functional groups [10,11]. Recently, to overcome limitations associated with pure SAM, mixed SAMs resulting from the coadsorption of two different thiols (i.e., mixed SAMs) have been tried to promote protein adsorption as a result of multiple chemical functionalities on the surfaces and decrease steric hindrance around the functional tails [12]. For example, Perez-Luna and his fellows fabricated the mixed SAMs of biotin-terminated thiol and 11hydroxy-1-undecanethiol on a gold surface for improving specific binding and eliminating nonspecific adsorption of wild type streptavidin and streptavidin mutants [13]. Using 11-hydroxy-1-undecanethiol, Dubrovsky and his co-workers also increased specific binding and controlled the nonspecific adsorption of protein on the surface of the gold-coated silica gel [14]. Unfortunately, besides unavailability and high price of long-chain thiols, affinity interfaces made up of longchain thiols are of relatively high viscoelasticity, which limits PQC application to the precise mass detection of biological materials in a liquid phase, because the Sauerbrey equation is derived from the assumption that the attached mass should be rigidly and strongly connected to the resonator [15]. Furthermore, these mixed SAMs assembled onto plane electrodes cannot supply enough functional groups to binding the probe molecules. To cope with these problems, this paper focuses on antibody immobilization by way of mixed SAMs composed of short-chain thiols, which are assembled on gold nanoparticles (GNPs) with threedimensional structure. In this strategy, cysteamine (CE) and 6mercapto-1-haxanol (MH) were co-assembled on the surface of nanoparticles, therefore resulting in mixed-SAM-capped GNPs. Then,
G. Shen, J. Lu / Thin Solid Films 518 (2010) 5010–5013
the mixed-SAM-capped GNPs were attached to gold electrode formally modified with cysteamine monolayer via glutaraldehyde (GA). MH functions as the “dilution reagent” to control the density of reactive groups on the surface, which increase the capture efficiency and minimize nonspecific adsorption. CE was utilized to introduce functional groups (–NH2), which can provide reaction sites for covalently bonding to glutaradehyde. Antibodies were then immobilized through Schiff base via glutaradehyde cross-linking. A specific binding occurs between the immobilized antibodies and antigens. The whole process can be succinctly shown in Fig. 1.
2. Experimental details
5011
The piezoelectric quartz crystal (AT-cut, 9 MHz, gold electrode) was obtained from Chenxing Radio Equipments (Beijing, China). In order to stabilize the frequency in the solution, one side of the PQC was sealed with an O-ring of silicone rubber covered by a plastic plate forming an air compartment isolated from aqueous solution. The analysis of immobilization quantity of thin film, antibodies, antigens and nonspecific binding was performed by mounting the prepared PQC in a laboratory-made cell in which 3 mL of PBS was added. A laboratory-made transistor–transistor logic-integrated circuit was designed to drive the PQC at its resonance frequency. The resonant frequency was monitored with a high frequency counter (Model FC 1250, Wellstar). According to Sauerbrey equation, the frequency decrease is related to the mass change of biomoleculars immobilized on the surface of PQC.
2.1. Chemicals 2.3. Synthesis of Au nanoparticles Cysteamine, 6-mercapto-1-haxanol and glutaraldehyde were obtained from Sigma-Aldrich. HAuCl4·4H2O was purchased from Shanghai Chemical Reagents (Shanghai, China). Bovine serum albumin (BSA) and human complement C3 (C3) were bought from Beijing Dingguo Biological technology Company (Beijing, China). Anti-CEA antibody was purchased from Zhongshan Biotechnology Company (Beijing, China). Purified CEA antigen of human serum and serum of cancer patients were provided by Hunan Provincial Tumor Hospital. Phosphate-buffered saline solution (PBS, pH 7.0) was prepared using 0.01 M Na2HPO4 and 0.01 M KH2PO4. All other reagents were of analytical grade. Double distilled water was used throughout the experiments.
2.2. Apparatus Electrochemical analysis was measured with a CHI760b electrochemistry working station (Jiangsu Chenhua Instruments, China). Cyclic voltammetry (CV) experiments were performed in a conventional three-electrode cell including a Pt electrode as counter electrode, a saturated calomel electrode as reference electrode and a modified Au electrode as working electrode. All electrochemical measurements were performed in the presence of 10 mM K3[Fe (CN)6]/K4[Fe(CN)6] (1:1) mixture, as a redox probe.
Various sizes of gold nanoparticles (10, 16, 25, and 30 nm) were prepared according to reference [16]. The mean diameter of the Au nanoparticles was determined using transmission electron microscopy (figures not shown). 2.4. Preparation of mixed-SAM-capped GNPs We have studied a series of solution compositions for the preparation of mixed-SAM-capped GNP. In a typical reaction, 500 µL of CE (5 mM) and 100 µL MH (5 mM) were added into 5 mL of Au colloid. This process involved a co-adsorption of two different thiols. After 12 h, the reaction mixture was then purified by centrifuging and washing with ethanol and water for 3 times by repeating the resuspension and recentrifugation process to remove unbound CE and MH complexes. The procedure was shown in Fig. 1 A. The resulting precipitate was dispersed in distilled water again and stored at 4 °C. 2.5. Preparation of CEA immunosensor Prior to the modification and measurement, each of the piezoelectric quartz crystal was cleaned in fresh piranha solution (70% H2SO4, 30% H2O2, v/v) followed by rinsing with water. The pretreated crystals were immersed in a solution of 5 mM cysteamine for 4 h to form a SAM. After rinsing with ethanol and water, 30 µL of GA was added onto the surface of crystal and incubated for 1 h at 37 °C. After rinsing with water and drying, the CE/GA-modified crystals were further immersed in a solution of mixed-SAM-capped GNPs for 6 h at 4 °C. After rinsing with water and drying, 30 µL of GA was added onto the surface of crystal and incubated for 1 h at 37 °C again. After rinsing with water and drying, 30 µL of anti-CEA antibody solution diluted with PBS (v/v = 1/2) was added onto the surface of crystal and incubated for 30 min at 37 °C. The excess antibodies were removed by rinsing with PBS. Then, the crystals were dried in air, and finally the sensor was ready. Before use, the sensors were stored at 4 °C. 2.6. Measurement procedure
Fig. 1. Schematic illustration of the proposed immunosensor.
The prepared PQC immunosensor were inserted into the reaction cell containing 3.0 mL of buffer solution (PBS, pH 7.0). With gentle stirring, the value of frequency was recorded when it reached stabilization (F1). After the crystal was rinsed with water and dried in air, 30 µL of CEA antigen solution or serum of cancer patients with various concentrations was dropped onto the surface of PQC, incubated for 30 min at 37 °C and washed with PBS three times followed by being put into reaction cell. The value of frequency was recorded again when it reached stabilization (F2). The ΔF (ΔF = F2 − F1) responds to the frequency of immunoreaction and is correlated to the antigen concentration.
5012
G. Shen, J. Lu / Thin Solid Films 518 (2010) 5010–5013
Fig. 2. Cyclic voltammograms of the Au electrode in different stages: (a) bare electrode; (b) CE-modified electrode; (c) CE/nanoparticles capped with mixed-SAM-modified electrode. Supporting electrolyte is10 mM Fe (CN)4−/3− . Scan rate is 100 mV/s. The 6 experiments were performed under conditions described in Section 2.5.
Fig. 4. Effect of Au nanoparticle size on the frequency change of immunoreaction. The experiments were performed under conditions described in Section 2.5 except for size of Au nanoparticles. The concentration of CEA is 300 ng mL− 1. Frequency changes are averages of three experiments.
3. Results and discussion 3.3. The effect of nanoparticle size 3.1. Electrochemical characteristics of the electrode surface The cyclic voltammogram of ferricyanide is a valuable and convenient tool to monitor the barrier of the modified electrode. Fig. 2 shows cyclic voltammograms of Fe(CN)4−/3− at a bare electrode 6 (curve a), CE-modified electrode (curve b), and mixed-SAM-capped GNP/CE-modified electrode (curve c). As shown in the figure, stepwise modification on the electrode is accompanied by a decrease in the amperometric response of the electrode and an increase in the peak to peak separation between the cathodic and anodic waves of the redox probe. This is consistent with the enhanced electron-transfer barriers introduced upon assembly of these layers.
The effect of the size of the gold nanoparticles on frequency change of the immunosensor was studied. Fig. 4 shows frequency responses of the immunosensors with different sizes of gold nanoparticles within the same concentration of CEA under steady-state conditions in a phosphate buffer solution (pH 7.0). The immunosensor fabricated with 16-nm gold nanoparticles exhibited a larger response than those of the other sizes. Therefore, 16-nm gold nanoparticles were chosen as the immobilized matrix. 3.4. Comparison of various modification procedures
The molar ratio of CE and MH is a vital factor and affects the amount of anti-CEA antibody immobilized on electrode. The impact of CE/MH at different molar ratio on the immunoreaction is shown in Fig. 3. The conclusion that can be made from Fig. 3 is that the maximum frequency change occurs between a CE/MH molar ratio of 4 and 6. The CE/MH molar ratio of 5 was selected.
In this study, the immobilization of antibody was achieved via different procedures including CE/MH-capped GNPs (CE/MH-capped) method and CE-capped GNPs (CE-capped) method. In order to investigate whether the CE/MH-capped GNPs film was better than CEcapped GNPs film for fabricating immunosensor, the frequency response caused by specific binding and nonspecific binding was used to evaluate. As shown in Fig. 5, the frequency response resulting from the antibody–antigen binding on CE/MH-capped GNPs film was greater than that on CE-capped GNPs film, while for the immunosensor based
Fig. 3. Effect of molar ratio of CE/MH on the frequency change of immunoreaction. The experiments were performed under conditions described in Section 2.5 except for molar ratio of CE/MH. The concentration of CEA is 300 ng mL− 1. Frequency changes are averages of three experiments.
Fig. 5. Comparison of specific binding and nonspecific binding of different antibody immobilization procedures including CE/MH-capped method and CE-capped method. The concentration of CEA, BSA and C3 is 300 ng mL− 1, 10 mg mL− 1 and 1 μg mL− 1, respectively. The experiments were performed under conditions described in Section 2.5. Frequency changes are averages of three experiments.
3.2. The effect of molar ratio of CE/MH
G. Shen, J. Lu / Thin Solid Films 518 (2010) 5010–5013
5013
Table 1 Comparison of CEA levels determined using two methods. Serum samples
1
2
3
4
5
6
Immunosensor (ng mL− 1) CLIA (ng mL− 1) Relative deviation (%)
81.4 93.2 − 12.6
128.7 117.5 8.9
206.9 193.8 6.7
267.7 28,103 − 4.8
331.2 318.9 3.8
429.5 403.1 6.5
between the results given by the two methods, that is, the developed immunoassay may provide a feasible alternative tool for determining CEA in human serum in clinical diagnosis. 3.8. The reproducibility and regeneration of immunosensor
Fig. 6. Effect of anti-CEA antibody concentration on frequency change of immunoreaction. The experiments were performed under conditions described in Section 2.5 except for dilution ratio of antibodies. The concentration of CEA is 300 ng mL− 1. Frequency changes are averages of three experiments.
on CE/MH-capped GNPs the frequency response to bovine serum albumin (BSA, 10 mg mL− 1) was lower than that for the immunosensor based on CE-capped GNPs, and similar cases took place to human complement C3 (1 μg mL− 1). It could be attributed to the lower density of reactive groups on surface of CE/MH-capped GNPs film, which decreased steric hindrance and increased the efficiency of specific binding.
To investigate the reproducibility of the PQC immunosensor, 300 ng mL− 1 CEA was determined repeatedly under the optimized experiment conditions. The average frequency change of three parallel experiments is 382 ± 23 Hz, and the relative standard deviation (R. S. D.) is 5.13%. These results indicate that the PQC immunosensor can offer a reliable result for the determination of CEA in human serum. The regeneration properties of the developed PQC probes were investigated by rinsing the used probes in 0.1 mol L− 1 glycine–hydrochloric acid buffer solution (pH 2.3) for 10 min followed by washing with distilled water several times to desorb the binding antibodies. Frequency changes revealed that the PQC immunosensor could be regenerated ten times without significant loss of detection sensitivity. The reason may be explained by the fact that antigen–antibody maybe denatured or desorbed from the electrode surface in the acidic solution. 4. Conclusions
3.5. The effect of anti-CEA antibody concentration The amount of anti-CEA antibody immobilized on electrode largely influences on the sensitivity of the immunosensor. In this study, it was controlled by dilution ratio of antibody solution with PBS solution (pH 7.0). The dilution ratios (Vanti-CEA/VPBS, 0.25, 0.33, 0.5 and 1) of antibody were examined and the results obtained are shown in Fig. 6. From Fig. 6, we can see that the frequency change increased with antiCEA antibody dilution ratio up to (the concentrations of anti-CEA increased gradually) 0.5, and then decreased gradually at higher dilution ratios. Increase in the response is most probably caused by an increase of antibody concentration attached to the surface when higher dilution ratios are used in the adsorption solutions, while decrease in the response probably results from steric hindrance of the immunoreaction due to even higher concentration attached to the surface. Therefore, an antibody dilution ratio of 0.5 was chosen for antibody immobilization in the present experiment.
In this paper, we have demonstrated that piezoelectric immunosensor based on gold nanoparticles capped with mixed self-assembled monolayer can increase the efficiency of antibody–antigen binding and minimize nonspecific adsorption. Antibodies are immobilized on the sensor surface by cross-linking with GA. It is unnecessary to block possible remaining active sites with BSA because of the use of mixed self-assembled monolayer. An additional advantage is that the sensor can be readily used to detect other proteins. Due to its simplicity, specificity and reliability, the proposed piezoelectric immunosensor is promising for clinical application of CEA determination. Acknowledgments This work was supported by Education Ministry Projects of Hunan Province (09C700) and the Natural Science Foundation of Hunan Province (07JJ3015).
3.6. Calibration curves
References
The representative calibration curve of CEA was obtained by dropping 30 µL of CEA solution at various concentrations to the antibody-coated crystals and subsequently incubating for 30 min at 37 °C. The frequency shift caused by immunoreaction and CEA concentration possessed a nearly linear relationship within 25–500 ng mL− 1. The calibration equation was y= 1.218x +16.376, in which y was the CEA concentration (ng mL− 1) and x represented the frequency shift (Hz).
[1] B. Konlg, M. Gratzel, Anal. Chem. 66 (1994) 341. [2] K. Nakanishi, H. Muguruma, I. Karube, Anal. Chem. 68 (1996) 1695. [3] S.T. Pathirana, J. Barbaree, B.A. Chin, M.G. Hartell, M.G. Neely, V. Vodyanoy, Biosens. Bioelectron. 15 (2000) 135. [4] H. Muramatsu, J.M. Dicks, E. Tamiya, I. Karube, Anal. Chem. 59 (1987) 2760. [5] S. Babacan, P. Pivarnik, S. Letcher, A.G. Rand, Biosens. Bioelectron. 15 (2000) 615. [6] I. Ben-Dov, I. Willner, E. Zisman, Anal. Chem. 69 (1997) 3506. [7] I. Park, N. Kim, Biosens. Bioelectron. 13 (1998) 1091. [8] Y.S. Fung, Y.Y. Wong, Anal. Chem. 73 (2001) 5302. [9] M.L. Mena, P. Yanez-Sedeno, J.M. Pingarion, Anal. Biochem. 336 (2005) 20. [10] R. Wildt, C. Mundy, B. Gorick, I. Tomlinson, Nat. Biotechnol. 1 (2000) 989. [11] R. Danczyk, B. Krieder, A. North, T. Webster, H. HogenEsch, A. Rundell, Biotechnol. Bioeng. 84 (2003) 215. [12] J. Lahiri, L. Isaacs, B. Grzybowski, J.D. Carbeck, G.M. Whitesides, Langmuir 15 (1999) 7186. [13] V.H. Perez-Luna, M.J. O'Brien, K.A. Opperman, P.D. Hampton, G.P. Lopez, L.A. Klumb, P.S. Stayton, J. Am. Chem. Soc. 121 (1999) 6469. [14] T.B. Dubrovsky, Z. Hou, P. Stroeve, N.L. Abbott, Anal. Chem. 71 (1999) 327. [15] Y. Zhang, V. Telyatnikov, M. Sathe, X.Q. Zeng, P.G. Wang, J. Am.Chem. Soc. 125 (2003) 9292. [16] A. Doron, E. Katz, I. Willner, Langmiur 11 (1995) 1313.
3.7. Clinical application Clinical application of the developed immunosensor was finally evaluated with real serum samples obtained from tumor patients. The CEA levels of six patient samples were detected by the proposed piezoelectric immunosensor and a CLIA method, respectively. The results and the relative deviations between the two methods are shown in Table 1. It shows that there is no significant difference
Contents lists available at ScienceDirect
Thin Solid Films j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / t s f
Piezoelectric immunosensor based on gold nanoparticles capped with mixed self-assembled monolayer for detection of carcinoembryonic antigen Guangyu Shen ⁎, Jilin Lu College of Chemistry and Chemical Engineering, Hunan University of Arts and Science, Changde 415400, Hunan, China
a r t i c l e
i n f o
Article history: Received 17 June 2009 Received in revised form 12 March 2010 Accepted 12 March 2010 Available online 27 March 2010 Keywords: Gold nanoparticles Self-assembled monolayers Piezoelectric immunosensor Carcinoembryonic antigen
a b s t r a c t It is very important for a piezoelectric immunosensor to increase specific binding and decrease nonspecific adsorption. This study presents the development of such a piezoelectric immunosensor for the detection of carcinoembryonic antigen. An AT-cut quartz crystal's Au electrode surface was first modified with homogenous self-assembled monolayer of cysteamine (CE). Gold nanoparticles capped with mixed selfassembled monolayer of CE and MH (6-mercapto-1-haxanol) were then attached to the CE monolayer via glutaraldehyde (GA). Antibodies were immobilized onto a mixed self-assembled monolayer of CE and MH with GA as a reactive intermediate too. The binding of target antigens onto the immobilized antibodies decreased the sensor's resonant frequency, and the frequency shift was correlated to the antigen concentration. The stepwise assembly of the immunosensor was characterized by means of cyclic voltammetry technique. This immunoassay was shown to be specific and sensitive, thus providing a viable alternative to carcinoembryonic antigen detection method. © 2010 Elsevier B.V. All rights reserved.
1. Introduction Carcinoembryonic antigen (CEA) is an important tumor marker responsible for clinical diagnosis of colorectal, pancreatic, gastric, and cervical carcinomas. Various immunoassays, such as enzyme-linked immunosorbent assay, radioimmunoassay, and chemiluminescence immunoassay (CLIA) have been developed for detecting CEA. However, all these techniques are time-consuming, expensive and/or requirement for sample pretreatment and concentration. So it is still significant and necessary to explore some simple, sensitive, and low-cost methods for the detection of carcinoembryonic antigen. Piezoelectric quartz crystal (PQC) devices have been attracting more and more attention for years, for they are portable, simple, cost-effective, and suitable for real-time monitoring biospecific interactions between antigen and antibody with high sensitivity and selectivity. Antibody immobilization is a vital part in development of a PQC immunosensor. The immobilization process must preserve the biological activity of the antibody, a more efficient binding and as well as reduce the nonspecific adsorption simultaneously. Current immobilization methods are mainly based on silanized layer [1], polymer membrane [2], Langmuir–Blodgett film [3], Protein A [4,5] self-assembled monolayer (SAM) [6–8] and nano-materials [9]. Of them, the SAM technique offers one of the simplest ways to provide a reproducible, ultrathin and well-ordered layer, however, pure SAM
⁎ Corresponding author. Tel.: + 86 7362563892. E-mail address: [email protected] (G. Shen). 0040-6090/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2010.03.077
may decrease capture efficiency and specificity due to high density of terminal functional groups [10,11]. Recently, to overcome limitations associated with pure SAM, mixed SAMs resulting from the coadsorption of two different thiols (i.e., mixed SAMs) have been tried to promote protein adsorption as a result of multiple chemical functionalities on the surfaces and decrease steric hindrance around the functional tails [12]. For example, Perez-Luna and his fellows fabricated the mixed SAMs of biotin-terminated thiol and 11hydroxy-1-undecanethiol on a gold surface for improving specific binding and eliminating nonspecific adsorption of wild type streptavidin and streptavidin mutants [13]. Using 11-hydroxy-1-undecanethiol, Dubrovsky and his co-workers also increased specific binding and controlled the nonspecific adsorption of protein on the surface of the gold-coated silica gel [14]. Unfortunately, besides unavailability and high price of long-chain thiols, affinity interfaces made up of longchain thiols are of relatively high viscoelasticity, which limits PQC application to the precise mass detection of biological materials in a liquid phase, because the Sauerbrey equation is derived from the assumption that the attached mass should be rigidly and strongly connected to the resonator [15]. Furthermore, these mixed SAMs assembled onto plane electrodes cannot supply enough functional groups to binding the probe molecules. To cope with these problems, this paper focuses on antibody immobilization by way of mixed SAMs composed of short-chain thiols, which are assembled on gold nanoparticles (GNPs) with threedimensional structure. In this strategy, cysteamine (CE) and 6mercapto-1-haxanol (MH) were co-assembled on the surface of nanoparticles, therefore resulting in mixed-SAM-capped GNPs. Then,
G. Shen, J. Lu / Thin Solid Films 518 (2010) 5010–5013
the mixed-SAM-capped GNPs were attached to gold electrode formally modified with cysteamine monolayer via glutaraldehyde (GA). MH functions as the “dilution reagent” to control the density of reactive groups on the surface, which increase the capture efficiency and minimize nonspecific adsorption. CE was utilized to introduce functional groups (–NH2), which can provide reaction sites for covalently bonding to glutaradehyde. Antibodies were then immobilized through Schiff base via glutaradehyde cross-linking. A specific binding occurs between the immobilized antibodies and antigens. The whole process can be succinctly shown in Fig. 1.
2. Experimental details
5011
The piezoelectric quartz crystal (AT-cut, 9 MHz, gold electrode) was obtained from Chenxing Radio Equipments (Beijing, China). In order to stabilize the frequency in the solution, one side of the PQC was sealed with an O-ring of silicone rubber covered by a plastic plate forming an air compartment isolated from aqueous solution. The analysis of immobilization quantity of thin film, antibodies, antigens and nonspecific binding was performed by mounting the prepared PQC in a laboratory-made cell in which 3 mL of PBS was added. A laboratory-made transistor–transistor logic-integrated circuit was designed to drive the PQC at its resonance frequency. The resonant frequency was monitored with a high frequency counter (Model FC 1250, Wellstar). According to Sauerbrey equation, the frequency decrease is related to the mass change of biomoleculars immobilized on the surface of PQC.
2.1. Chemicals 2.3. Synthesis of Au nanoparticles Cysteamine, 6-mercapto-1-haxanol and glutaraldehyde were obtained from Sigma-Aldrich. HAuCl4·4H2O was purchased from Shanghai Chemical Reagents (Shanghai, China). Bovine serum albumin (BSA) and human complement C3 (C3) were bought from Beijing Dingguo Biological technology Company (Beijing, China). Anti-CEA antibody was purchased from Zhongshan Biotechnology Company (Beijing, China). Purified CEA antigen of human serum and serum of cancer patients were provided by Hunan Provincial Tumor Hospital. Phosphate-buffered saline solution (PBS, pH 7.0) was prepared using 0.01 M Na2HPO4 and 0.01 M KH2PO4. All other reagents were of analytical grade. Double distilled water was used throughout the experiments.
2.2. Apparatus Electrochemical analysis was measured with a CHI760b electrochemistry working station (Jiangsu Chenhua Instruments, China). Cyclic voltammetry (CV) experiments were performed in a conventional three-electrode cell including a Pt electrode as counter electrode, a saturated calomel electrode as reference electrode and a modified Au electrode as working electrode. All electrochemical measurements were performed in the presence of 10 mM K3[Fe (CN)6]/K4[Fe(CN)6] (1:1) mixture, as a redox probe.
Various sizes of gold nanoparticles (10, 16, 25, and 30 nm) were prepared according to reference [16]. The mean diameter of the Au nanoparticles was determined using transmission electron microscopy (figures not shown). 2.4. Preparation of mixed-SAM-capped GNPs We have studied a series of solution compositions for the preparation of mixed-SAM-capped GNP. In a typical reaction, 500 µL of CE (5 mM) and 100 µL MH (5 mM) were added into 5 mL of Au colloid. This process involved a co-adsorption of two different thiols. After 12 h, the reaction mixture was then purified by centrifuging and washing with ethanol and water for 3 times by repeating the resuspension and recentrifugation process to remove unbound CE and MH complexes. The procedure was shown in Fig. 1 A. The resulting precipitate was dispersed in distilled water again and stored at 4 °C. 2.5. Preparation of CEA immunosensor Prior to the modification and measurement, each of the piezoelectric quartz crystal was cleaned in fresh piranha solution (70% H2SO4, 30% H2O2, v/v) followed by rinsing with water. The pretreated crystals were immersed in a solution of 5 mM cysteamine for 4 h to form a SAM. After rinsing with ethanol and water, 30 µL of GA was added onto the surface of crystal and incubated for 1 h at 37 °C. After rinsing with water and drying, the CE/GA-modified crystals were further immersed in a solution of mixed-SAM-capped GNPs for 6 h at 4 °C. After rinsing with water and drying, 30 µL of GA was added onto the surface of crystal and incubated for 1 h at 37 °C again. After rinsing with water and drying, 30 µL of anti-CEA antibody solution diluted with PBS (v/v = 1/2) was added onto the surface of crystal and incubated for 30 min at 37 °C. The excess antibodies were removed by rinsing with PBS. Then, the crystals were dried in air, and finally the sensor was ready. Before use, the sensors were stored at 4 °C. 2.6. Measurement procedure
Fig. 1. Schematic illustration of the proposed immunosensor.
The prepared PQC immunosensor were inserted into the reaction cell containing 3.0 mL of buffer solution (PBS, pH 7.0). With gentle stirring, the value of frequency was recorded when it reached stabilization (F1). After the crystal was rinsed with water and dried in air, 30 µL of CEA antigen solution or serum of cancer patients with various concentrations was dropped onto the surface of PQC, incubated for 30 min at 37 °C and washed with PBS three times followed by being put into reaction cell. The value of frequency was recorded again when it reached stabilization (F2). The ΔF (ΔF = F2 − F1) responds to the frequency of immunoreaction and is correlated to the antigen concentration.
5012
G. Shen, J. Lu / Thin Solid Films 518 (2010) 5010–5013
Fig. 2. Cyclic voltammograms of the Au electrode in different stages: (a) bare electrode; (b) CE-modified electrode; (c) CE/nanoparticles capped with mixed-SAM-modified electrode. Supporting electrolyte is10 mM Fe (CN)4−/3− . Scan rate is 100 mV/s. The 6 experiments were performed under conditions described in Section 2.5.
Fig. 4. Effect of Au nanoparticle size on the frequency change of immunoreaction. The experiments were performed under conditions described in Section 2.5 except for size of Au nanoparticles. The concentration of CEA is 300 ng mL− 1. Frequency changes are averages of three experiments.
3. Results and discussion 3.3. The effect of nanoparticle size 3.1. Electrochemical characteristics of the electrode surface The cyclic voltammogram of ferricyanide is a valuable and convenient tool to monitor the barrier of the modified electrode. Fig. 2 shows cyclic voltammograms of Fe(CN)4−/3− at a bare electrode 6 (curve a), CE-modified electrode (curve b), and mixed-SAM-capped GNP/CE-modified electrode (curve c). As shown in the figure, stepwise modification on the electrode is accompanied by a decrease in the amperometric response of the electrode and an increase in the peak to peak separation between the cathodic and anodic waves of the redox probe. This is consistent with the enhanced electron-transfer barriers introduced upon assembly of these layers.
The effect of the size of the gold nanoparticles on frequency change of the immunosensor was studied. Fig. 4 shows frequency responses of the immunosensors with different sizes of gold nanoparticles within the same concentration of CEA under steady-state conditions in a phosphate buffer solution (pH 7.0). The immunosensor fabricated with 16-nm gold nanoparticles exhibited a larger response than those of the other sizes. Therefore, 16-nm gold nanoparticles were chosen as the immobilized matrix. 3.4. Comparison of various modification procedures
The molar ratio of CE and MH is a vital factor and affects the amount of anti-CEA antibody immobilized on electrode. The impact of CE/MH at different molar ratio on the immunoreaction is shown in Fig. 3. The conclusion that can be made from Fig. 3 is that the maximum frequency change occurs between a CE/MH molar ratio of 4 and 6. The CE/MH molar ratio of 5 was selected.
In this study, the immobilization of antibody was achieved via different procedures including CE/MH-capped GNPs (CE/MH-capped) method and CE-capped GNPs (CE-capped) method. In order to investigate whether the CE/MH-capped GNPs film was better than CEcapped GNPs film for fabricating immunosensor, the frequency response caused by specific binding and nonspecific binding was used to evaluate. As shown in Fig. 5, the frequency response resulting from the antibody–antigen binding on CE/MH-capped GNPs film was greater than that on CE-capped GNPs film, while for the immunosensor based
Fig. 3. Effect of molar ratio of CE/MH on the frequency change of immunoreaction. The experiments were performed under conditions described in Section 2.5 except for molar ratio of CE/MH. The concentration of CEA is 300 ng mL− 1. Frequency changes are averages of three experiments.
Fig. 5. Comparison of specific binding and nonspecific binding of different antibody immobilization procedures including CE/MH-capped method and CE-capped method. The concentration of CEA, BSA and C3 is 300 ng mL− 1, 10 mg mL− 1 and 1 μg mL− 1, respectively. The experiments were performed under conditions described in Section 2.5. Frequency changes are averages of three experiments.
3.2. The effect of molar ratio of CE/MH
G. Shen, J. Lu / Thin Solid Films 518 (2010) 5010–5013
5013
Table 1 Comparison of CEA levels determined using two methods. Serum samples
1
2
3
4
5
6
Immunosensor (ng mL− 1) CLIA (ng mL− 1) Relative deviation (%)
81.4 93.2 − 12.6
128.7 117.5 8.9
206.9 193.8 6.7
267.7 28,103 − 4.8
331.2 318.9 3.8
429.5 403.1 6.5
between the results given by the two methods, that is, the developed immunoassay may provide a feasible alternative tool for determining CEA in human serum in clinical diagnosis. 3.8. The reproducibility and regeneration of immunosensor
Fig. 6. Effect of anti-CEA antibody concentration on frequency change of immunoreaction. The experiments were performed under conditions described in Section 2.5 except for dilution ratio of antibodies. The concentration of CEA is 300 ng mL− 1. Frequency changes are averages of three experiments.
on CE/MH-capped GNPs the frequency response to bovine serum albumin (BSA, 10 mg mL− 1) was lower than that for the immunosensor based on CE-capped GNPs, and similar cases took place to human complement C3 (1 μg mL− 1). It could be attributed to the lower density of reactive groups on surface of CE/MH-capped GNPs film, which decreased steric hindrance and increased the efficiency of specific binding.
To investigate the reproducibility of the PQC immunosensor, 300 ng mL− 1 CEA was determined repeatedly under the optimized experiment conditions. The average frequency change of three parallel experiments is 382 ± 23 Hz, and the relative standard deviation (R. S. D.) is 5.13%. These results indicate that the PQC immunosensor can offer a reliable result for the determination of CEA in human serum. The regeneration properties of the developed PQC probes were investigated by rinsing the used probes in 0.1 mol L− 1 glycine–hydrochloric acid buffer solution (pH 2.3) for 10 min followed by washing with distilled water several times to desorb the binding antibodies. Frequency changes revealed that the PQC immunosensor could be regenerated ten times without significant loss of detection sensitivity. The reason may be explained by the fact that antigen–antibody maybe denatured or desorbed from the electrode surface in the acidic solution. 4. Conclusions
3.5. The effect of anti-CEA antibody concentration The amount of anti-CEA antibody immobilized on electrode largely influences on the sensitivity of the immunosensor. In this study, it was controlled by dilution ratio of antibody solution with PBS solution (pH 7.0). The dilution ratios (Vanti-CEA/VPBS, 0.25, 0.33, 0.5 and 1) of antibody were examined and the results obtained are shown in Fig. 6. From Fig. 6, we can see that the frequency change increased with antiCEA antibody dilution ratio up to (the concentrations of anti-CEA increased gradually) 0.5, and then decreased gradually at higher dilution ratios. Increase in the response is most probably caused by an increase of antibody concentration attached to the surface when higher dilution ratios are used in the adsorption solutions, while decrease in the response probably results from steric hindrance of the immunoreaction due to even higher concentration attached to the surface. Therefore, an antibody dilution ratio of 0.5 was chosen for antibody immobilization in the present experiment.
In this paper, we have demonstrated that piezoelectric immunosensor based on gold nanoparticles capped with mixed self-assembled monolayer can increase the efficiency of antibody–antigen binding and minimize nonspecific adsorption. Antibodies are immobilized on the sensor surface by cross-linking with GA. It is unnecessary to block possible remaining active sites with BSA because of the use of mixed self-assembled monolayer. An additional advantage is that the sensor can be readily used to detect other proteins. Due to its simplicity, specificity and reliability, the proposed piezoelectric immunosensor is promising for clinical application of CEA determination. Acknowledgments This work was supported by Education Ministry Projects of Hunan Province (09C700) and the Natural Science Foundation of Hunan Province (07JJ3015).
3.6. Calibration curves
References
The representative calibration curve of CEA was obtained by dropping 30 µL of CEA solution at various concentrations to the antibody-coated crystals and subsequently incubating for 30 min at 37 °C. The frequency shift caused by immunoreaction and CEA concentration possessed a nearly linear relationship within 25–500 ng mL− 1. The calibration equation was y= 1.218x +16.376, in which y was the CEA concentration (ng mL− 1) and x represented the frequency shift (Hz).
[1] B. Konlg, M. Gratzel, Anal. Chem. 66 (1994) 341. [2] K. Nakanishi, H. Muguruma, I. Karube, Anal. Chem. 68 (1996) 1695. [3] S.T. Pathirana, J. Barbaree, B.A. Chin, M.G. Hartell, M.G. Neely, V. Vodyanoy, Biosens. Bioelectron. 15 (2000) 135. [4] H. Muramatsu, J.M. Dicks, E. Tamiya, I. Karube, Anal. Chem. 59 (1987) 2760. [5] S. Babacan, P. Pivarnik, S. Letcher, A.G. Rand, Biosens. Bioelectron. 15 (2000) 615. [6] I. Ben-Dov, I. Willner, E. Zisman, Anal. Chem. 69 (1997) 3506. [7] I. Park, N. Kim, Biosens. Bioelectron. 13 (1998) 1091. [8] Y.S. Fung, Y.Y. Wong, Anal. Chem. 73 (2001) 5302. [9] M.L. Mena, P. Yanez-Sedeno, J.M. Pingarion, Anal. Biochem. 336 (2005) 20. [10] R. Wildt, C. Mundy, B. Gorick, I. Tomlinson, Nat. Biotechnol. 1 (2000) 989. [11] R. Danczyk, B. Krieder, A. North, T. Webster, H. HogenEsch, A. Rundell, Biotechnol. Bioeng. 84 (2003) 215. [12] J. Lahiri, L. Isaacs, B. Grzybowski, J.D. Carbeck, G.M. Whitesides, Langmuir 15 (1999) 7186. [13] V.H. Perez-Luna, M.J. O'Brien, K.A. Opperman, P.D. Hampton, G.P. Lopez, L.A. Klumb, P.S. Stayton, J. Am. Chem. Soc. 121 (1999) 6469. [14] T.B. Dubrovsky, Z. Hou, P. Stroeve, N.L. Abbott, Anal. Chem. 71 (1999) 327. [15] Y. Zhang, V. Telyatnikov, M. Sathe, X.Q. Zeng, P.G. Wang, J. Am.Chem. Soc. 125 (2003) 9292. [16] A. Doron, E. Katz, I. Willner, Langmiur 11 (1995) 1313.
3.7. Clinical application Clinical application of the developed immunosensor was finally evaluated with real serum samples obtained from tumor patients. The CEA levels of six patient samples were detected by the proposed piezoelectric immunosensor and a CLIA method, respectively. The results and the relative deviations between the two methods are shown in Table 1. It shows that there is no significant difference