Biosensors and Bioelectronics 70 (2015) 161–166
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Triple tumor markers assay based on carbon–gold nanocomposite Teng Xu, Na Liu, Jing Yuan, Zhanfang Ma n Department of Chemistry, Capital Normal University, Beijing 100048, China
art ic l e i nf o
a b s t r a c t
Article history: Received 30 December 2014 Received in revised form 15 March 2015 Accepted 16 March 2015 Available online 17 March 2015
A sandwich-format electrochemical immunosensor for simultaneous determination of three cancer biomarkers using the carbon–gold nanocomposite (CGN) as immunoprobes was introduced. The CGN were fabricated through a simple microwave-assisted carbonization of glucose and deposition of gold nanoparticles (AuNPs). This nanocomposite showed great adsorption ability to the redox probes such as some organic dyes and metal ions, due to the abundant reactive oxygen functional groups on its surface. The AuNPs decorated on the nanocomposite provided extra binding sites for the three antibodies carcinoembryonic antigen (CEA), prostate specific antigen (PSA), α-fetoprotein (AFP), respectively. The ionic liquid reduced graphene oxide was combined with poly(sodium-p-styrenesulfonate) as substrate to attach the three different antibodies through electrostatic adsorption. Three separate signals can be detected directly in a single run through square wave voltammetry. Under optimized conditions, the electrochemical immunosensor exhibited good sensitivity and selectivity for the simultaneous determination of CEA, PSA and AFP with linear ranges of 0.01–100 ng mL 1. The detection limit for CEA, PSA and AFP is 2.7, 4.8 and 3.1 pg mL 1, respectively. This method was applied for the analysis of CEA, PSA and AFP levels in clinical serum samples, and the results were in good agreement with those of enzyme linked immunosorbent assay (ELISA). This approach gives a promising simple and sensitive immunoassay strategy for the identification and validation of specific early cancer. & 2015 Elsevier B.V. All rights reserved.
Keywords: Tumor marker Cancer Multiplexed electrochemical detection Electrochemical immunosensor Carbon–gold nanocomposite
1. Introduction For the cancer patients, the early detection of cancer can contribute to the early diagnosis and treatment of malignancies and lead to a higher chance of survival. In recent years, protein biomarkers have begun to act an increasingly important role in the management of patients with malignancy. The quantified detection of protein biomarkers such as tumor markers in serum thus holds enormous promise, both in detecting the disease early and in subsequently tracking disease progression in response to therapy. This increased the demand of a more sensitive and fast detection of tumor markers. In addition, because of the inherent complexity of cancers, the different type cancers could commonly altered in different series of tumor markers. It was agreed that the simultaneous multiplexed immunoassay of a panel of targets can provide a more accurate and reliable diagnosis. Electrochemical immunoassay with the advantages of innate high sensitivity and simplicity has become a powerful analytical tool for the specific and sensitive detection of clinical samples (Liu and Ju, 2008). The traditional electrochemical immunoassay for one or two analytes has become common and lots of works have n
Corresponding author. E-mail address: [email protected] (Z. Ma).
http://dx.doi.org/10.1016/j.bios.2015.03.036 0956-5663/& 2015 Elsevier B.V. All rights reserved.
been done (Cao et al., 2013; Chen and Ma, 2014; Gao et al., 2015; Jia et al., 2014; Lian et al., 2013; Liu and Ma, 2013; Liu et al., 2014; Wang et al., 2014). However, many cancers were associated with multiple tumor markers. Such as the liver cancer and ovarian cancer, they were all specific to three tumor markers (Li et al., 2012; Petru et al., 1990). It is often a time-consuming and complicated work for diagnostic cancer screening. Therefore, the simultaneous multiplexed analyzes that could cover more tumor markers in a single run is urgently needed, for the advantages of shortened analysis time, decreased sample and simplified analytical procedure volume (Wu et al., 2007). But the simultaneous multiplexed analyses always face the difficulties of fabricating immunoprobes with different redox activity, particularly for the analyses covered more than two biomarkers in a single run. The voltammetric peaks of the redox probes need to be distinguishable and separate from each other. And in order to prevent the interference from blood serum substance, such as dopamine and ascorbic acid, the potential should be under 0.1 V (Chen et al., 2010; Wang et al., 2002). To solve this issue, it will be of great significance to develop new materials which can generally adsorb three or more redox probes and meanwhile possess abundant activity sites to bind antibody. Recently, lots of attentions have been paid to the carbon nanomaterials such as carbonaceous materials due to their large surface area, excellent biocompatibility, and fascinating
162
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
electrocatalytic activity (Gao et al., 2013; Hu et al., 2008; Liu et al., 2008; Makowski et al., 2008; Rezan et al., 2009; Xu et al., 2011). Carbon nanosphere obtained through hydrothermal carbonization usually got abundant reactive oxygen functional groups, showed good adsorption capacity to redox probes such as some organic dyes and metal ions. In addition, its intrinsic properties can be finely tuned by changing parameters such as bulk structure, diameter and chemical composition. These make the carbon nanosphere ideal material in the application of immunoprobes and biosensor platform for the simultaneous multiplexed assay. However, the carbon materials commonly face the problems of complicated synthesis process, hydrophobicity and poor dispersion (Tang et al., 2009). The traditional hydrothermal carbonization methods of carbohydrates could hardly control the size of the colloidal spheres, and will cost more than 12 h (Hu et al., 2010). In this work, a carbon–gold nanocomposite (CGN) was introduced. The carbon sphere was obtained by the microwave-assisted carbonization of glucose in several minutes. The growth mechanism of the carbon spheres was confirmed to the LaMer model (LaMer, 1952; Sun and Li, 2004). The hydrothermal conditions will lead to aromatization and carbonization of glucose. Firstly, the glucose was dehydrated into furan-like molecules, such as furfural aldehyde or 5-(hydroxymethyl)-2-furaldehyde. Subsequently a cross-linking was induced by the intermolecular dehydration. The resulting nuclei grew uniformly and isotropically by diffusion of solutes toward the particle surfaces. The immobilization of AuNPs was achieved through the in-situ formation, by dispersion the carbon nanosphere in the presence of HAuCl4. With the aid of the second step microwave irritation, the AuNPs could be deposited on the obtained carbon sphere at last. The as-prepared CGN has abundant reactive oxygen functional groups, which are easy to adsorb different electrochemical redox probes. In addition, antibody can be easily fixed on the AuNPs of CGN. Herein, the as-prepared CGNs were used to fabricate three electrochemical immunoprobes through adsorbing thionin (Thi), 2,3-diaminophenazine (DAP) and Cd2 þ , and fixing three kind of antibodies, respectively. The obtained three new electrochemical immunoprobes exhibited electrochemical windows can effectively avoid the interference from other substances in serum and were applied for simultaneous detection of three tumor markers carcinoembryonic antigen (CEA), prostate specific antigen (PSA), and α-fetoprotein (AFP). The results were well consistent with those of ELISA.
2. Experimental section 2.1. Reagents and Materials CEA, PSA, AFP was purchased from Biosynthesis Biotechnology Company (Beijing, China). Monoclonal anti-CEA, anti-PSA and antiAFP capture antibodies, monoclonal anti-CEA, anti-PSA and antiAFP labeled antibodies were purchased from Linc-Bio Company (Shanghai, China). The hydrogen tetrachloroaurate hydrate (HAuCl4 xH2O, 99.9%), acetic acid (AA), uric acid (UA), sodium citrate, glucose, thionine acetate, DAP and poly(sodium-p-styrenesulfonate) (PSS) were purchased from Alfa Aesar China (Tianjin, China). Ionic liquid (e.g., 1-aminopropyl-3-methylimidazolium chloride) was purchased from Shanghai Chengjie Chemical Co. Ltd. (Shanghai, China). Graphene oxide was purchased from JCNANO (Nanjing, China). NaH2PO4, Na2HPO4, Cd(NO3)2 potassium ferricyanide (K3[Fe(CN)6]), potassium ferrocyanide (K4[Fe(CN)6]), K2CO3 and bovine serum albumin (BSA) were purchased from Beijing Chemical Reagents Company (Beijing, China). Clinical serum samples were obtained from Hospital of Capital Normal University, China. The ELISA date of serum samples were obtained from Deyi clinical testing center, Beijing. Ultrapure water
(resistivity 4 18 MΩ cm) was used throughout the experiment and all the reagents were of analytical grade and used as received. 2.2. Apparatus During all experiment procedures, the water used was purified through Olst ultrapure K8 apparatus (Olst, Ltd., resistivity418 MΩ cm). The synthesis of CGN was conducted by microwave reaction instrument (CEM, American). Zeta potential analysis was performed on Nano ZS Zetasizer (Malvern Instruments Co. British). All the electrochemical experiments were carried out on CHI1140 electrochemical workstation (Chenhua Instruments Co., Shanghai, China). Three-electrode system consisting of an Ag/AgCl electrode (saturated KCl) as the reference electrode, a platinum wire as the auxiliary electrode and a glassy carbon electrode (GCE) (4 mm diameter) as the working electrode was used in experiment. 2.3. Preparation of CGN In previous works, the carbon–gold structure materials were synthesized using different method (Cui et al., 2008; Jerzy et al., 2012). Here we introduced a more simple strategy. Briefly, 4 mL (wt. 1%) glucose aqueous solution was mixed with 200 μL (wt. 10%) sodium citrate aqueous solution. The mixture was reacted in microwave reaction instrument (250 W) at 160 °C for 10 min and then cooled down to the room temperature. As programmed, the microwave reactor automatically controlled the reaction temperatures. The color of mixture changed from pale to dark brown, indicating the carbonization of carbohydrates was happened. Then resulting mixture was centrifuged and washed with water several times, then dispersed in 4 mL ultrapure water. A sonication of 45 min was carried on until the turbid mixture was transformed into a homogeneous solution. The mixture was mixed with 100 μL (wt. 1%) HAuCl4 aqueous solution and reacted in microwave reaction instrument (250 W) at 100 °C for 15 min, then cooled down to the room temperature. Subsequently, centrifuged and washed with water several times, redispersed in 4 mL ultrapure water. The CGN was obtained. 2.4. Preparation of the Ionic liquid reduced graphene oxide Graphene possessed excellent electron-transfer ability and exceptional thermal stability (Kang et al., 2009; Shan et al., 2009; Tang et al., 2011). Ionic liquid reduced graphene oxide (IL-rGO) was synthesized according to our previous work. Briefly, 50 mg ionic liquid was added to 50 mL dispersion of graphene oxide in water (0.5 mg mL 1). 50 mg KOH was added. A sonication was carried on until the mixture was transformed from turbid to a homogeneous solution. Then, the solution was vigorously stirred under 80 °C for 24 h. The obtained mixture was centrifuged and washed with ethanol and water repeatedly for six times, then dispersed in ultrapure water (0.5 mg mL 1) for further use. 2.5. Preparation of immunoprobes Three equal parts of 4 mL CGN were mixed with 100 μL 20 mM Thi, DAP and Cd(NO3)2, respectively. Then were stirred 5 h under room temperature. After centrifuged and washed with water for several times, three obtained bioconjugates CGN-Thi, CGN-DAP and CGN-Cd2 þ were re-dispersed in 4 mL ultrapure water. Under gently stirring, three labeled antibodies of CEA, PSA and AFP (200 μL, 1 mg mL 1) were added to the CGN dispersion, respectively, and stirring at room temperature for 2 h, bioconjugates of CGN-Thi-anti-CEA, CGN-DAP-anti-PSA and CGN-Cd2 þ -anti-AFP were obtained. After centrifugation and re-dispersed in 2 mL
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
163
ultrapure water, in order to block remaining active sites and eliminate the non-specific adsorption, 200 μL 2% BSA solutions were added and allowed to react for 2 h. The resulting nanocomposite was then collected by centrifugation at 8000 rpm for 10 min and washed three times with ultrapure water, subsequently, re-dispersed in 2 mL ultrapure water and stored at 4 °C for further use. 2.6. Fabrication of the immunosensor The GCE decorated with capture anti-CEA, capture anti-PSA and capture anti -AFP was prepared through the following procedures: GCE (4 mm in diameter) was polished with 1.0, 0.3 and 0.05 μm alumina powder separately, and rinsed thoroughly with ultrapure water after each polishing step. A successive sonication was carried on in doubly distilled water and ethanol for 5 min and dried in nitrogen, 10 μL IL-rGO solution was cast on to the electrode carefully and dried at 30 °C for 45 min. Then a thin graphene membrane was formed. 20 μL PSS (2 g L 1, in 6 mM NaCl solution) was drop on the top of the membrane and react for another 45 min under room temperature. After washing with ultrapure water and phosphate buffer (PB) (0.1 M, pH 7.3) three times to remove the loosely binding chemicals, the electrode was dried with nitrogen. Then the electrode were incubated with a mixture of 200 μg mL 1 capture anti-CEA, capture anti-PSA and capture capture anti-AFP solution at 4 °C overnight. The modified immunosensor was incubated in a solution of BSA (w/w, 1%) for about 1 h at 37 °C in order to block possible remaining active sites The immunosensor obtained was stored at 4 °C before use. 2.7. Electrochemical determination of CEA, PSA and AFP The immunosensor was incubated with the mixture of CEA, PSA and AFP solutions with different concentrations for 35 min at 37 °C. Then it was incubated with the mixture of 1:1:1 CGN-Thianti-CEA, CGN-DAP-anti-PSA and CGN-Cd2 þ -anti-AFP bioconjugates for 45 min at 37 °C. A sandwich-type immunoassay was performed. After each step, the electrodes were washed with 0.1 M PBS (pH 7.3). Subsequently, square wave voltammetry (SWV) from 1.2 V to 0.2 V (vs. SCE) with pulse amplitude of 50 mV and a pulse width of 50 ms was performed. The electrochemical response was recorded for simultaneous measurement of CEA, PSA and AFP in 0.1 M PBS (pH 6.5).
3. Results and discussion 3.1. Characterization of the immunoprobes As shown in Fig. 1A, the carbon nanosphere showed uniform morphology and size distribution, the average diameter was about 300 nm. The surface is rough. After a microwave-assisted reduction reaction, the AuNPs was loaded onto the surface of CGN. As shown in Fig. 1B, the AuNPs were evenly distributed. The antibody could be immobilized onto the CGN through the interaction of antibody and AuNPs. The high resolution TEM images of the obtained CGN were shown in Fig. 1C. The distribution of AuNPs on the carbon nanosphere surface was more clearly illustrated. The AuNPs were uniformly distributed on the carbon sphere surface. The Fig. 1D showed the crystalline structure of the AuNPs on the surface, the lattice space of which is 0.235 nm, confirmed to the (111) facets. The thickness of gold for CGN could be inferred from the Fig. 1S. The reactive oxygen groups on the surface were characterized through X-ray photoelectron spectroscopy (XPS) as shown in Fig. 2. The Fig. 2A gave the survey spectra of the nanocomposite, the peaks of C1s and O1s could be found, indicating
Fig. 1. TEM images of carbon nanosphere (A), CGN (B); high resolution TEM image of CGN (C), crystalline structure of CGN (D).
that the glucose was not completely carbonized and some oxygen groups were remained. The Fig. 2B gave the high resolution curve of the C1s. The three peaks at 284.6, 286.1, and 288 eV corresponding to the groups of C–C, C–O, C ¼O, respectively. These reactive oxygen groups could contribute the adsorption of redox probes. And Fig. 2S gave the XPS of immunoprobes including CGNThi-anti-CEA, CGN-DAP-anti-PSA and CGN–Cd2 þ -anti-AFP, respectively. 3.2. Principles of the multiplexed immunoassay For the preparation of electrochemical immunoprobes, CGNThi-anti-CEA, CGN-DAP-anti-PSA and CGN-Cd2 þ -anti-AFP bioconjugates were firstly synthesized. The carbonization of glucose and the reduction of AuNPs were happened under the microwaveassisted reaction. Three redox probes Thi, DAP and Cd2 þ were immobilized through the reactive oxygen groups of CGN. After the antibodies were directly immobilized on AuNPs of CGN, three electrochemical immunoprobes were obtained as illustrated in Scheme 1. For the modification of GCE, the GCE was modified directly with positively charged IL-rGO through casting method. Then GCE was immersed in PSS aqueous solution. The negatively charged PSS was adsorbed on the GCE with abundant sulfonic acid groups. The antibody was usually positively charged since its isoelectric point was nearly pH 9.0. In this way, the capture antibodies of CEA, AFP and PSA were immobilized on GCE. After the immunosensor was successively incubated with the mixture of antigen solutions and the mixture of 1:1:1 diluted CGN-Thi-anti-CEA, CGN-DAPanti-PSA and CGN-Cd2 þ -anti-AFP bioconjugates, a sandwich-type immunoassay was formed because of the highly biospecific recognition interactions between antigens and the corresponding antibodies. The current responses were recorded through SWV. The distinct voltammetric peaks of Thi, DAP, Cd2 þ were directly proportional to the concentrations of corresponding antigens. According to the linear relationship obtained, the CEA, AFP and PSA in serum samples could be determined quantitatively.
164
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
Fig. 2. XPS survey spectra of the prepared CGN (A) and high resolution XPS of C1s peak (B).
Fig. 3S gave the Nyquist diagrams of the immunosensor. Secondly, in order to monitor the stability of the biosensor platforms, a CV with 20 cycles of the GCE-IL-rGO-PSS was deployed as shown in the Fig. 3B. There was no obvious decrease of the voltammetric current during the whole procedures. This indicated that the biosensor platforms had an excellent stability. This was because of the high appetency between the ionic liquid reduced graphene and glass carbon electrode. 3.4. Optimization of immunosensor response
Scheme 1. Preparation and schematic illustration of the electrochemical immunoassay protocol .
3.3. Electrochemical characterization of the immunosensor The IL-rGO with superior conductivity was used as biosensor platforms for immobilization of proteins. In order to monitor the fabrication process of the immunosensor, the cyclic voltammograms (CV) measurements were performed in 0.1 M PBS containing 5 mM [Fe(CN)6]4 /3 (pH 7.0) in Fig. 3A. The curve a represented the distinct redox peaks of the bare GCE. After the GCE was modified with IL-rGO, the peak current of curve b increased sharply. The reason might be the excellent ability of electron transfer of the IL-rGO. Then the electrode was incubated with PSS, the peak current of curve c decreased, this might be the PSS, which could increase the resistance between the solution and the electrode surface as an organic membrane. After the electrode was incubated in mixture of capture anti-CEA, anti-AFP and anti-PSA, the peak currents of curve c decreased. This was attributed to the insulating layer of proteins hindered the electron transfer on the interface. To further analyze the fabrication of electrode platform,
The pH value of detection solution has great influence on both the antibodies activity and the immunosensor electrochemical performance. In order to choose an optimal pH, the mixture of CEA, PSA and AFP (10 ng mL 1) solutions was detected in PBS detection solution of different pH value, and the response of SWV was recorded, as shown in the Fig. 4S. The voltammetric peak currents initially increased and then decreased after pH 6.5. This may because of the electrical property of the antibody would be influenced greatly by the pH near its isoelectric point, the binding between the antibody and immunosensor could be weakened. In order to get a well sensitivity for three signal tags reduction peak, PBS of pH 6.5 was chosen as the detection solution for simultaneous determination of CEA, PSA and AFP. The incubation time was another important influence factor since it would take time for the antigen to reach antibody and form immunocomplexes. The incubation time was investigated in PBS of pH 6.5 with the mixture of CEA, PSA and AFP (10 ng mL 1) solutions. The peak current increased with the increasing of incubation time, for the reason that it would take time to reach the maximum formation of the sandwich immunocomplexes. After 35 min, the formation of sandwich-type immunoassay would be saturated and the peak current began to get flat. Therefore, we select 35 min as the optimal incubation time for the immunoassay.
Fig. 3. CV responses of the different modified electrodes in 0.1 M PBS containing 5 mM [Fe(CN)6]4 /3 (pH 7.0) (A), bare GCE (a), electrode treated successively with IL-rGO (b), PSS (c), mixture of capture antibodies (d); CV with 20 circles of the GCE-IL-rGO-PSS (B).
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
3.5. Evaluation of repeatability, stability, cross-reactivity, and specificity In order to illustrate the repeatability of the immunosensor, the mixture of CEA, PSA and AFP (10 ng mL 1) solutions was measured five times using five freshly prepared GCE. The relative standard deviation was 5.1% for CEA, 6.5% for PSA and 3.6% for AFP. The result demonstrated that repeatability of the proposed immunsensor was acceptable for CEA, PSA and AFP. Then the immunoprobes as prepared were store at 4 °C for two weeks. The electric signal dropped less than 10% compared with the freshly prepared at the end of fourteenth day, indicating the stability of proposed immunoprobes was acceptable. Another two control tests were carried out to evaluate the cross-reactivity of the multiplexed electrochemical immunosensor: (1) single antigen was assayed using the prepared immunosensor and probes; (2) CEA, PSA and AFP were simultaneously assayed using the prepared immunosensor and probes. The 50 ng mL 1 of CEA and AFP were assayed as examples for the test (1) and (2). The electrochemical responses of the control tests were displayed in Fig. 5S. The corresponding target analyte showed a higher current response, meanwhile, another one showed an extremely low current shift. The results indicated that the interference with each other was low, and the cross-reactivity was negligible. In order to simulate the interferences from other blood constituent in serum samples and evaluate the specificity of the immunosensor, the specificity of the immuosensor was also checked and IgG, BSA, AA, glucose and UA were chosen as possible interfering agents. 5 ng mL 1 CEA, AFP and PSA were mixed with 100 ng mL 1 of IgG, BSA, AA, glucose or UA, respectively. The response of immunosensor was measured and shown in Fig. 6S. Compared with the current response obtained from the mixture solutions of CEA, PSA and AFP (10 ng mL 1), the variation in current caused by the interference substances was less than 5%, indicating that the immunosensor possesses good specificity for CEA, AFP and PSA.
165
3.6. Analytical performance of the multiplexed immunoassay The performance of the proposed immunosensor was evaluated using different concentrations of CEA, AFP and PSA mixture under the optimal conditions. As shown in Fig. 4, the voltammetric peaks near 0.05 V, 0.35 V, and 0.65 V comes from the Thi, DAP, and Cd2 þ , respectively, which represent the existence of CEA, PSA and AFP, respectively. The SWV peak currents of the Thi, DPA and Cd2 þ increased with the increment of concentrations of CEA, AFP and PSA in the sample solution. In the ranges of 0.05–100 ng mL 1 CEA, AFP and PSA, the calibration plots displayed a good linear relationship between the voltammetric peaks currents and the concentration of analytes. The correlation coefficient of AFP, CEA and PSA was 0.997, 0.995 and 0.997 separately. The detection limit was determined at a signal-to-noise ratio of 3 for CEA, PSA and AFP is 2.7, 4.8 and 3.1 pg mL 1, respectively. The sensitivity and linear range was acceptable compared with the enzymatic reaction enhanced immunoassay (0.01 ng mL 1). The result indicated that the multiplexed electrochemical immunoassay possessed an acceptable wide linear range and low limit of detection.
3.7. Analysis of clinical serum samples In order to evaluate the reliability of the proposed multiplexed immunosensor, three clinical serum samples containing CEA, AFP and PSA was assayed. The obtained results were compared with those obtained by ELISA (Table 1). The relative errors between the two methods were less than 10%, showing an acceptable accuracy. The proposed immunosensor could have the potential application in clinical diagnostics for simultaneous determination of CEA, AFP and PSA. The comparison between proposed biosensor and some recently published works is listed in Table 1S.
Fig. 4. SWV responses of the proposed immunosensor after incubation with different concentrations of CEA, PSA and AFP; Calibration curves of the multiplexed immunoassay toward CEA (A), PSA (B) and APF (C) in PBS (pH 6.5).
166
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
Table 1 Assay results of clinical serum samples using the proposed and reference methods. No. ELISA (ng mL 1)
1 2 3 4 5 6 7 8
Proposed immunosensor (ng mL 1)
Relative error (%)
CEA
PSA
AFP
CEA
PSA
AFP
CEA
PSA
AFP
2.86 0.56 4.88 1.33 2.03 1.21 1.50 0.85
0.25 0.22 0.21 0.29 0.61 0.21 0.21 0.42
1.02 1.05 0.26 0.78 0.51 0.37 0.19 2.31
2.92 0.53 4.55 1.40 1.87 1.15 1.61 0.88
0.26 0.20 0.20 0.27 0.64 0.22 0.20 0.41
1.07 1.12 0.25 0.80 0.48 0.35 0.20 2.35
2.10 5.36 6.76 5.26 7.88 4.96 7.33 3.53
4.00 9.09 4.76 6.90 4.92 4.76 4.76 2.38
4.90 6.67 3.85 2.56 5.88 5.41 5.26 1.73
4. Conclusions In this work, a triple tumor markers assay based on the carbon– gold nanocomposite by microwave-assisted method was successfully developed. The distinct voltammetric peaks of Thi, DAP and Cd2 þ which had a close relationship with each corresponding antigen can be detected through SWV in a single run. Moreover, the unique adsorption properties of the immunoprobes were studied and could be extended to make other electrochemical immunosensors. Present work provided a promising way in developing simultaneous multiplexed analyses of a panel of targets.
Acknowledgements This research was financed by the grants from National Natural Science Foundation of China (21273153), Beijing Natural Science Foundation (2132008), and Research Base Construction Projects of Beijing Municipal Education Commission.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.03.036.
References Cao, X., Wang, N., Jia, S., Guo, L., Li, K., 2013. Biosens. Bioelectron. 39, 226–230. Chen, D.X., Wang, Q., Jin, J., Wu, P., Wang, H., Yu, S.Q., Zhang, H., Cai, C.X., 2010. Anal. Chem. 82, 2448–2455. Chen, X., Ma, Z.F., 2014. Biosens. Bioelectron. 55, 343–349. Cui, R.J., Liu, C., Shen, J.M., Gao, D., Zhu, J.J., Chen, H.Y., 2008. Adv. Funct. Mater. 18, 2197–2204. Gao, J., Guo, Z.K., Su, F.J., Gao, L., Pang, X.H., Cao, W., Du, B., Wei, Q., 2015. Biosens. Bioelectron. 63, 465–471. Gao, Q., Han, J.M., Ma, Z.F., 2013. Biosens. Bioelectron. 49, 323–328. Hu, B., Wang, K., Wu, L., Yu, S.H., Antonietti, M., Titirici, M.M., 2010. Adv. Mater. 22, 813–828. Hu, Y.S., Demir-Cakan, R., Titirici, M.M., Müller, J.O., Schlögl, R., Antonietti, M., Maier, J., 2008. Angew. Chem. Int. Ed. 47, 1645–1649. Jerzy, C., Dominik, J., Katarzyna, A., Michal, M., Mietek, J., 2012. Chem. Commun. 48, 3972–3974. Jia, X.L., Liu, Z.M., Liu, N., Ma, Z.F., 2014. Biosens. Bioelectron. 53, 160–166. Kang, X.H., Wang, J., Wu, H., Aksay, I.A., Liu, J., Lin, Y.H., 2009. Biosens. Bioelectron. 25, 901–905. LaMer, V.K., 1952. Ind. Eng. Chem. 44, 1270. Li, Y., Zhong, Z.Y., Chai, Y.Q., Song, Z.J., Zhuo, Y., Sun, H.L., Liu, S.M., Wang, D., Yuan, R., 2012. Chem. Commun. 48, 537–539. Lian, W.J., Liu, S., Yu, J.G., Li, J., Cui, M., Xu, W., Huang, J.D., 2013. Biosens. Bioelectron. 44, 70–76. Liu, N., Liu, Z.M., Han, H.L., Ma, Z.F., 2014. J. Mater. Chem. B 2, 3292–3298. Liu, X., Ju, H.X., 2008. Anal. Chem. 80, 5377–5382. Liu, Z., Liu, Q., Huang, Y., Ma, Y., Yin, S., Zhang, X., Sun, W., Chen, Y., 2008. Adv. Mater. 20, 3924–3930. Liu, Z.M., Ma, Z.F., 2013. Biosens. Bioelectron. 46, 1–7. Makowski, P., Cakan, R.D., Antonietti, M., Goettmann, F., Titirici, M.M., 2008. Chem. Commun., 999–1001. Petru, E., Sevin, B.U., Averette, H.E., Koechli, O.R., Perras, J.P., Hilsenbeck, S., 1990. Gynecol. Oncol., 181–186. Rezan, D.C., Niki, B., Markus, A., Maria-Magdalena, T., 2009. Chem. Mater. 21, 484–490. Shan, C.S., Yang, H.F., Han, D.X., Zhang, Q.X., Ivaska, A., Niu, L., 2009. Langmuir 25, 12030–12033. Sun, X.M., Li, Y.D., 2004. Angew. Chem. 116, 607–611. Tang, L.H., Wang, Y., Li, Y.M., Feng, H.B., Lu, J., Li, J.H., 2009. Adv. Funct. Mater. 19, 2782–2789. Tang, L.H., Wang, Y., Liu, Y., Li, J.H., 2011. ACS Nano 5, 3817–3822. Wang, Z.H., Liu, J., Liang, Q.L., Wang, Y.M., Luo, G.A., 2002. Analyst 127, 653–658. Wang, Z.F., Chen, X., Ma, Z.F., 2014. Biosens. Bioelectron. 61, 562–568. Wu, J., Fu, Z.F., Yan, F., Ju, H.X., 2007. Trends Anal. Chem. 26, 679–688. Xu, S.J., Liu, Y., Wang, T.H., Li, J.H., 2011. Anal. Chem. 83, 3817–3823.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
Triple tumor markers assay based on carbon–gold nanocomposite Teng Xu, Na Liu, Jing Yuan, Zhanfang Ma n Department of Chemistry, Capital Normal University, Beijing 100048, China
art ic l e i nf o
a b s t r a c t
Article history: Received 30 December 2014 Received in revised form 15 March 2015 Accepted 16 March 2015 Available online 17 March 2015
A sandwich-format electrochemical immunosensor for simultaneous determination of three cancer biomarkers using the carbon–gold nanocomposite (CGN) as immunoprobes was introduced. The CGN were fabricated through a simple microwave-assisted carbonization of glucose and deposition of gold nanoparticles (AuNPs). This nanocomposite showed great adsorption ability to the redox probes such as some organic dyes and metal ions, due to the abundant reactive oxygen functional groups on its surface. The AuNPs decorated on the nanocomposite provided extra binding sites for the three antibodies carcinoembryonic antigen (CEA), prostate specific antigen (PSA), α-fetoprotein (AFP), respectively. The ionic liquid reduced graphene oxide was combined with poly(sodium-p-styrenesulfonate) as substrate to attach the three different antibodies through electrostatic adsorption. Three separate signals can be detected directly in a single run through square wave voltammetry. Under optimized conditions, the electrochemical immunosensor exhibited good sensitivity and selectivity for the simultaneous determination of CEA, PSA and AFP with linear ranges of 0.01–100 ng mL 1. The detection limit for CEA, PSA and AFP is 2.7, 4.8 and 3.1 pg mL 1, respectively. This method was applied for the analysis of CEA, PSA and AFP levels in clinical serum samples, and the results were in good agreement with those of enzyme linked immunosorbent assay (ELISA). This approach gives a promising simple and sensitive immunoassay strategy for the identification and validation of specific early cancer. & 2015 Elsevier B.V. All rights reserved.
Keywords: Tumor marker Cancer Multiplexed electrochemical detection Electrochemical immunosensor Carbon–gold nanocomposite
1. Introduction For the cancer patients, the early detection of cancer can contribute to the early diagnosis and treatment of malignancies and lead to a higher chance of survival. In recent years, protein biomarkers have begun to act an increasingly important role in the management of patients with malignancy. The quantified detection of protein biomarkers such as tumor markers in serum thus holds enormous promise, both in detecting the disease early and in subsequently tracking disease progression in response to therapy. This increased the demand of a more sensitive and fast detection of tumor markers. In addition, because of the inherent complexity of cancers, the different type cancers could commonly altered in different series of tumor markers. It was agreed that the simultaneous multiplexed immunoassay of a panel of targets can provide a more accurate and reliable diagnosis. Electrochemical immunoassay with the advantages of innate high sensitivity and simplicity has become a powerful analytical tool for the specific and sensitive detection of clinical samples (Liu and Ju, 2008). The traditional electrochemical immunoassay for one or two analytes has become common and lots of works have n
Corresponding author. E-mail address: [email protected] (Z. Ma).
http://dx.doi.org/10.1016/j.bios.2015.03.036 0956-5663/& 2015 Elsevier B.V. All rights reserved.
been done (Cao et al., 2013; Chen and Ma, 2014; Gao et al., 2015; Jia et al., 2014; Lian et al., 2013; Liu and Ma, 2013; Liu et al., 2014; Wang et al., 2014). However, many cancers were associated with multiple tumor markers. Such as the liver cancer and ovarian cancer, they were all specific to three tumor markers (Li et al., 2012; Petru et al., 1990). It is often a time-consuming and complicated work for diagnostic cancer screening. Therefore, the simultaneous multiplexed analyzes that could cover more tumor markers in a single run is urgently needed, for the advantages of shortened analysis time, decreased sample and simplified analytical procedure volume (Wu et al., 2007). But the simultaneous multiplexed analyses always face the difficulties of fabricating immunoprobes with different redox activity, particularly for the analyses covered more than two biomarkers in a single run. The voltammetric peaks of the redox probes need to be distinguishable and separate from each other. And in order to prevent the interference from blood serum substance, such as dopamine and ascorbic acid, the potential should be under 0.1 V (Chen et al., 2010; Wang et al., 2002). To solve this issue, it will be of great significance to develop new materials which can generally adsorb three or more redox probes and meanwhile possess abundant activity sites to bind antibody. Recently, lots of attentions have been paid to the carbon nanomaterials such as carbonaceous materials due to their large surface area, excellent biocompatibility, and fascinating
162
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
electrocatalytic activity (Gao et al., 2013; Hu et al., 2008; Liu et al., 2008; Makowski et al., 2008; Rezan et al., 2009; Xu et al., 2011). Carbon nanosphere obtained through hydrothermal carbonization usually got abundant reactive oxygen functional groups, showed good adsorption capacity to redox probes such as some organic dyes and metal ions. In addition, its intrinsic properties can be finely tuned by changing parameters such as bulk structure, diameter and chemical composition. These make the carbon nanosphere ideal material in the application of immunoprobes and biosensor platform for the simultaneous multiplexed assay. However, the carbon materials commonly face the problems of complicated synthesis process, hydrophobicity and poor dispersion (Tang et al., 2009). The traditional hydrothermal carbonization methods of carbohydrates could hardly control the size of the colloidal spheres, and will cost more than 12 h (Hu et al., 2010). In this work, a carbon–gold nanocomposite (CGN) was introduced. The carbon sphere was obtained by the microwave-assisted carbonization of glucose in several minutes. The growth mechanism of the carbon spheres was confirmed to the LaMer model (LaMer, 1952; Sun and Li, 2004). The hydrothermal conditions will lead to aromatization and carbonization of glucose. Firstly, the glucose was dehydrated into furan-like molecules, such as furfural aldehyde or 5-(hydroxymethyl)-2-furaldehyde. Subsequently a cross-linking was induced by the intermolecular dehydration. The resulting nuclei grew uniformly and isotropically by diffusion of solutes toward the particle surfaces. The immobilization of AuNPs was achieved through the in-situ formation, by dispersion the carbon nanosphere in the presence of HAuCl4. With the aid of the second step microwave irritation, the AuNPs could be deposited on the obtained carbon sphere at last. The as-prepared CGN has abundant reactive oxygen functional groups, which are easy to adsorb different electrochemical redox probes. In addition, antibody can be easily fixed on the AuNPs of CGN. Herein, the as-prepared CGNs were used to fabricate three electrochemical immunoprobes through adsorbing thionin (Thi), 2,3-diaminophenazine (DAP) and Cd2 þ , and fixing three kind of antibodies, respectively. The obtained three new electrochemical immunoprobes exhibited electrochemical windows can effectively avoid the interference from other substances in serum and were applied for simultaneous detection of three tumor markers carcinoembryonic antigen (CEA), prostate specific antigen (PSA), and α-fetoprotein (AFP). The results were well consistent with those of ELISA.
2. Experimental section 2.1. Reagents and Materials CEA, PSA, AFP was purchased from Biosynthesis Biotechnology Company (Beijing, China). Monoclonal anti-CEA, anti-PSA and antiAFP capture antibodies, monoclonal anti-CEA, anti-PSA and antiAFP labeled antibodies were purchased from Linc-Bio Company (Shanghai, China). The hydrogen tetrachloroaurate hydrate (HAuCl4 xH2O, 99.9%), acetic acid (AA), uric acid (UA), sodium citrate, glucose, thionine acetate, DAP and poly(sodium-p-styrenesulfonate) (PSS) were purchased from Alfa Aesar China (Tianjin, China). Ionic liquid (e.g., 1-aminopropyl-3-methylimidazolium chloride) was purchased from Shanghai Chengjie Chemical Co. Ltd. (Shanghai, China). Graphene oxide was purchased from JCNANO (Nanjing, China). NaH2PO4, Na2HPO4, Cd(NO3)2 potassium ferricyanide (K3[Fe(CN)6]), potassium ferrocyanide (K4[Fe(CN)6]), K2CO3 and bovine serum albumin (BSA) were purchased from Beijing Chemical Reagents Company (Beijing, China). Clinical serum samples were obtained from Hospital of Capital Normal University, China. The ELISA date of serum samples were obtained from Deyi clinical testing center, Beijing. Ultrapure water
(resistivity 4 18 MΩ cm) was used throughout the experiment and all the reagents were of analytical grade and used as received. 2.2. Apparatus During all experiment procedures, the water used was purified through Olst ultrapure K8 apparatus (Olst, Ltd., resistivity418 MΩ cm). The synthesis of CGN was conducted by microwave reaction instrument (CEM, American). Zeta potential analysis was performed on Nano ZS Zetasizer (Malvern Instruments Co. British). All the electrochemical experiments were carried out on CHI1140 electrochemical workstation (Chenhua Instruments Co., Shanghai, China). Three-electrode system consisting of an Ag/AgCl electrode (saturated KCl) as the reference electrode, a platinum wire as the auxiliary electrode and a glassy carbon electrode (GCE) (4 mm diameter) as the working electrode was used in experiment. 2.3. Preparation of CGN In previous works, the carbon–gold structure materials were synthesized using different method (Cui et al., 2008; Jerzy et al., 2012). Here we introduced a more simple strategy. Briefly, 4 mL (wt. 1%) glucose aqueous solution was mixed with 200 μL (wt. 10%) sodium citrate aqueous solution. The mixture was reacted in microwave reaction instrument (250 W) at 160 °C for 10 min and then cooled down to the room temperature. As programmed, the microwave reactor automatically controlled the reaction temperatures. The color of mixture changed from pale to dark brown, indicating the carbonization of carbohydrates was happened. Then resulting mixture was centrifuged and washed with water several times, then dispersed in 4 mL ultrapure water. A sonication of 45 min was carried on until the turbid mixture was transformed into a homogeneous solution. The mixture was mixed with 100 μL (wt. 1%) HAuCl4 aqueous solution and reacted in microwave reaction instrument (250 W) at 100 °C for 15 min, then cooled down to the room temperature. Subsequently, centrifuged and washed with water several times, redispersed in 4 mL ultrapure water. The CGN was obtained. 2.4. Preparation of the Ionic liquid reduced graphene oxide Graphene possessed excellent electron-transfer ability and exceptional thermal stability (Kang et al., 2009; Shan et al., 2009; Tang et al., 2011). Ionic liquid reduced graphene oxide (IL-rGO) was synthesized according to our previous work. Briefly, 50 mg ionic liquid was added to 50 mL dispersion of graphene oxide in water (0.5 mg mL 1). 50 mg KOH was added. A sonication was carried on until the mixture was transformed from turbid to a homogeneous solution. Then, the solution was vigorously stirred under 80 °C for 24 h. The obtained mixture was centrifuged and washed with ethanol and water repeatedly for six times, then dispersed in ultrapure water (0.5 mg mL 1) for further use. 2.5. Preparation of immunoprobes Three equal parts of 4 mL CGN were mixed with 100 μL 20 mM Thi, DAP and Cd(NO3)2, respectively. Then were stirred 5 h under room temperature. After centrifuged and washed with water for several times, three obtained bioconjugates CGN-Thi, CGN-DAP and CGN-Cd2 þ were re-dispersed in 4 mL ultrapure water. Under gently stirring, three labeled antibodies of CEA, PSA and AFP (200 μL, 1 mg mL 1) were added to the CGN dispersion, respectively, and stirring at room temperature for 2 h, bioconjugates of CGN-Thi-anti-CEA, CGN-DAP-anti-PSA and CGN-Cd2 þ -anti-AFP were obtained. After centrifugation and re-dispersed in 2 mL
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
163
ultrapure water, in order to block remaining active sites and eliminate the non-specific adsorption, 200 μL 2% BSA solutions were added and allowed to react for 2 h. The resulting nanocomposite was then collected by centrifugation at 8000 rpm for 10 min and washed three times with ultrapure water, subsequently, re-dispersed in 2 mL ultrapure water and stored at 4 °C for further use. 2.6. Fabrication of the immunosensor The GCE decorated with capture anti-CEA, capture anti-PSA and capture anti -AFP was prepared through the following procedures: GCE (4 mm in diameter) was polished with 1.0, 0.3 and 0.05 μm alumina powder separately, and rinsed thoroughly with ultrapure water after each polishing step. A successive sonication was carried on in doubly distilled water and ethanol for 5 min and dried in nitrogen, 10 μL IL-rGO solution was cast on to the electrode carefully and dried at 30 °C for 45 min. Then a thin graphene membrane was formed. 20 μL PSS (2 g L 1, in 6 mM NaCl solution) was drop on the top of the membrane and react for another 45 min under room temperature. After washing with ultrapure water and phosphate buffer (PB) (0.1 M, pH 7.3) three times to remove the loosely binding chemicals, the electrode was dried with nitrogen. Then the electrode were incubated with a mixture of 200 μg mL 1 capture anti-CEA, capture anti-PSA and capture capture anti-AFP solution at 4 °C overnight. The modified immunosensor was incubated in a solution of BSA (w/w, 1%) for about 1 h at 37 °C in order to block possible remaining active sites The immunosensor obtained was stored at 4 °C before use. 2.7. Electrochemical determination of CEA, PSA and AFP The immunosensor was incubated with the mixture of CEA, PSA and AFP solutions with different concentrations for 35 min at 37 °C. Then it was incubated with the mixture of 1:1:1 CGN-Thianti-CEA, CGN-DAP-anti-PSA and CGN-Cd2 þ -anti-AFP bioconjugates for 45 min at 37 °C. A sandwich-type immunoassay was performed. After each step, the electrodes were washed with 0.1 M PBS (pH 7.3). Subsequently, square wave voltammetry (SWV) from 1.2 V to 0.2 V (vs. SCE) with pulse amplitude of 50 mV and a pulse width of 50 ms was performed. The electrochemical response was recorded for simultaneous measurement of CEA, PSA and AFP in 0.1 M PBS (pH 6.5).
3. Results and discussion 3.1. Characterization of the immunoprobes As shown in Fig. 1A, the carbon nanosphere showed uniform morphology and size distribution, the average diameter was about 300 nm. The surface is rough. After a microwave-assisted reduction reaction, the AuNPs was loaded onto the surface of CGN. As shown in Fig. 1B, the AuNPs were evenly distributed. The antibody could be immobilized onto the CGN through the interaction of antibody and AuNPs. The high resolution TEM images of the obtained CGN were shown in Fig. 1C. The distribution of AuNPs on the carbon nanosphere surface was more clearly illustrated. The AuNPs were uniformly distributed on the carbon sphere surface. The Fig. 1D showed the crystalline structure of the AuNPs on the surface, the lattice space of which is 0.235 nm, confirmed to the (111) facets. The thickness of gold for CGN could be inferred from the Fig. 1S. The reactive oxygen groups on the surface were characterized through X-ray photoelectron spectroscopy (XPS) as shown in Fig. 2. The Fig. 2A gave the survey spectra of the nanocomposite, the peaks of C1s and O1s could be found, indicating
Fig. 1. TEM images of carbon nanosphere (A), CGN (B); high resolution TEM image of CGN (C), crystalline structure of CGN (D).
that the glucose was not completely carbonized and some oxygen groups were remained. The Fig. 2B gave the high resolution curve of the C1s. The three peaks at 284.6, 286.1, and 288 eV corresponding to the groups of C–C, C–O, C ¼O, respectively. These reactive oxygen groups could contribute the adsorption of redox probes. And Fig. 2S gave the XPS of immunoprobes including CGNThi-anti-CEA, CGN-DAP-anti-PSA and CGN–Cd2 þ -anti-AFP, respectively. 3.2. Principles of the multiplexed immunoassay For the preparation of electrochemical immunoprobes, CGNThi-anti-CEA, CGN-DAP-anti-PSA and CGN-Cd2 þ -anti-AFP bioconjugates were firstly synthesized. The carbonization of glucose and the reduction of AuNPs were happened under the microwaveassisted reaction. Three redox probes Thi, DAP and Cd2 þ were immobilized through the reactive oxygen groups of CGN. After the antibodies were directly immobilized on AuNPs of CGN, three electrochemical immunoprobes were obtained as illustrated in Scheme 1. For the modification of GCE, the GCE was modified directly with positively charged IL-rGO through casting method. Then GCE was immersed in PSS aqueous solution. The negatively charged PSS was adsorbed on the GCE with abundant sulfonic acid groups. The antibody was usually positively charged since its isoelectric point was nearly pH 9.0. In this way, the capture antibodies of CEA, AFP and PSA were immobilized on GCE. After the immunosensor was successively incubated with the mixture of antigen solutions and the mixture of 1:1:1 diluted CGN-Thi-anti-CEA, CGN-DAPanti-PSA and CGN-Cd2 þ -anti-AFP bioconjugates, a sandwich-type immunoassay was formed because of the highly biospecific recognition interactions between antigens and the corresponding antibodies. The current responses were recorded through SWV. The distinct voltammetric peaks of Thi, DAP, Cd2 þ were directly proportional to the concentrations of corresponding antigens. According to the linear relationship obtained, the CEA, AFP and PSA in serum samples could be determined quantitatively.
164
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
Fig. 2. XPS survey spectra of the prepared CGN (A) and high resolution XPS of C1s peak (B).
Fig. 3S gave the Nyquist diagrams of the immunosensor. Secondly, in order to monitor the stability of the biosensor platforms, a CV with 20 cycles of the GCE-IL-rGO-PSS was deployed as shown in the Fig. 3B. There was no obvious decrease of the voltammetric current during the whole procedures. This indicated that the biosensor platforms had an excellent stability. This was because of the high appetency between the ionic liquid reduced graphene and glass carbon electrode. 3.4. Optimization of immunosensor response
Scheme 1. Preparation and schematic illustration of the electrochemical immunoassay protocol .
3.3. Electrochemical characterization of the immunosensor The IL-rGO with superior conductivity was used as biosensor platforms for immobilization of proteins. In order to monitor the fabrication process of the immunosensor, the cyclic voltammograms (CV) measurements were performed in 0.1 M PBS containing 5 mM [Fe(CN)6]4 /3 (pH 7.0) in Fig. 3A. The curve a represented the distinct redox peaks of the bare GCE. After the GCE was modified with IL-rGO, the peak current of curve b increased sharply. The reason might be the excellent ability of electron transfer of the IL-rGO. Then the electrode was incubated with PSS, the peak current of curve c decreased, this might be the PSS, which could increase the resistance between the solution and the electrode surface as an organic membrane. After the electrode was incubated in mixture of capture anti-CEA, anti-AFP and anti-PSA, the peak currents of curve c decreased. This was attributed to the insulating layer of proteins hindered the electron transfer on the interface. To further analyze the fabrication of electrode platform,
The pH value of detection solution has great influence on both the antibodies activity and the immunosensor electrochemical performance. In order to choose an optimal pH, the mixture of CEA, PSA and AFP (10 ng mL 1) solutions was detected in PBS detection solution of different pH value, and the response of SWV was recorded, as shown in the Fig. 4S. The voltammetric peak currents initially increased and then decreased after pH 6.5. This may because of the electrical property of the antibody would be influenced greatly by the pH near its isoelectric point, the binding between the antibody and immunosensor could be weakened. In order to get a well sensitivity for three signal tags reduction peak, PBS of pH 6.5 was chosen as the detection solution for simultaneous determination of CEA, PSA and AFP. The incubation time was another important influence factor since it would take time for the antigen to reach antibody and form immunocomplexes. The incubation time was investigated in PBS of pH 6.5 with the mixture of CEA, PSA and AFP (10 ng mL 1) solutions. The peak current increased with the increasing of incubation time, for the reason that it would take time to reach the maximum formation of the sandwich immunocomplexes. After 35 min, the formation of sandwich-type immunoassay would be saturated and the peak current began to get flat. Therefore, we select 35 min as the optimal incubation time for the immunoassay.
Fig. 3. CV responses of the different modified electrodes in 0.1 M PBS containing 5 mM [Fe(CN)6]4 /3 (pH 7.0) (A), bare GCE (a), electrode treated successively with IL-rGO (b), PSS (c), mixture of capture antibodies (d); CV with 20 circles of the GCE-IL-rGO-PSS (B).
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
3.5. Evaluation of repeatability, stability, cross-reactivity, and specificity In order to illustrate the repeatability of the immunosensor, the mixture of CEA, PSA and AFP (10 ng mL 1) solutions was measured five times using five freshly prepared GCE. The relative standard deviation was 5.1% for CEA, 6.5% for PSA and 3.6% for AFP. The result demonstrated that repeatability of the proposed immunsensor was acceptable for CEA, PSA and AFP. Then the immunoprobes as prepared were store at 4 °C for two weeks. The electric signal dropped less than 10% compared with the freshly prepared at the end of fourteenth day, indicating the stability of proposed immunoprobes was acceptable. Another two control tests were carried out to evaluate the cross-reactivity of the multiplexed electrochemical immunosensor: (1) single antigen was assayed using the prepared immunosensor and probes; (2) CEA, PSA and AFP were simultaneously assayed using the prepared immunosensor and probes. The 50 ng mL 1 of CEA and AFP were assayed as examples for the test (1) and (2). The electrochemical responses of the control tests were displayed in Fig. 5S. The corresponding target analyte showed a higher current response, meanwhile, another one showed an extremely low current shift. The results indicated that the interference with each other was low, and the cross-reactivity was negligible. In order to simulate the interferences from other blood constituent in serum samples and evaluate the specificity of the immunosensor, the specificity of the immuosensor was also checked and IgG, BSA, AA, glucose and UA were chosen as possible interfering agents. 5 ng mL 1 CEA, AFP and PSA were mixed with 100 ng mL 1 of IgG, BSA, AA, glucose or UA, respectively. The response of immunosensor was measured and shown in Fig. 6S. Compared with the current response obtained from the mixture solutions of CEA, PSA and AFP (10 ng mL 1), the variation in current caused by the interference substances was less than 5%, indicating that the immunosensor possesses good specificity for CEA, AFP and PSA.
165
3.6. Analytical performance of the multiplexed immunoassay The performance of the proposed immunosensor was evaluated using different concentrations of CEA, AFP and PSA mixture under the optimal conditions. As shown in Fig. 4, the voltammetric peaks near 0.05 V, 0.35 V, and 0.65 V comes from the Thi, DAP, and Cd2 þ , respectively, which represent the existence of CEA, PSA and AFP, respectively. The SWV peak currents of the Thi, DPA and Cd2 þ increased with the increment of concentrations of CEA, AFP and PSA in the sample solution. In the ranges of 0.05–100 ng mL 1 CEA, AFP and PSA, the calibration plots displayed a good linear relationship between the voltammetric peaks currents and the concentration of analytes. The correlation coefficient of AFP, CEA and PSA was 0.997, 0.995 and 0.997 separately. The detection limit was determined at a signal-to-noise ratio of 3 for CEA, PSA and AFP is 2.7, 4.8 and 3.1 pg mL 1, respectively. The sensitivity and linear range was acceptable compared with the enzymatic reaction enhanced immunoassay (0.01 ng mL 1). The result indicated that the multiplexed electrochemical immunoassay possessed an acceptable wide linear range and low limit of detection.
3.7. Analysis of clinical serum samples In order to evaluate the reliability of the proposed multiplexed immunosensor, three clinical serum samples containing CEA, AFP and PSA was assayed. The obtained results were compared with those obtained by ELISA (Table 1). The relative errors between the two methods were less than 10%, showing an acceptable accuracy. The proposed immunosensor could have the potential application in clinical diagnostics for simultaneous determination of CEA, AFP and PSA. The comparison between proposed biosensor and some recently published works is listed in Table 1S.
Fig. 4. SWV responses of the proposed immunosensor after incubation with different concentrations of CEA, PSA and AFP; Calibration curves of the multiplexed immunoassay toward CEA (A), PSA (B) and APF (C) in PBS (pH 6.5).
166
T. Xu et al. / Biosensors and Bioelectronics 70 (2015) 161–166
Table 1 Assay results of clinical serum samples using the proposed and reference methods. No. ELISA (ng mL 1)
1 2 3 4 5 6 7 8
Proposed immunosensor (ng mL 1)
Relative error (%)
CEA
PSA
AFP
CEA
PSA
AFP
CEA
PSA
AFP
2.86 0.56 4.88 1.33 2.03 1.21 1.50 0.85
0.25 0.22 0.21 0.29 0.61 0.21 0.21 0.42
1.02 1.05 0.26 0.78 0.51 0.37 0.19 2.31
2.92 0.53 4.55 1.40 1.87 1.15 1.61 0.88
0.26 0.20 0.20 0.27 0.64 0.22 0.20 0.41
1.07 1.12 0.25 0.80 0.48 0.35 0.20 2.35
2.10 5.36 6.76 5.26 7.88 4.96 7.33 3.53
4.00 9.09 4.76 6.90 4.92 4.76 4.76 2.38
4.90 6.67 3.85 2.56 5.88 5.41 5.26 1.73
4. Conclusions In this work, a triple tumor markers assay based on the carbon– gold nanocomposite by microwave-assisted method was successfully developed. The distinct voltammetric peaks of Thi, DAP and Cd2 þ which had a close relationship with each corresponding antigen can be detected through SWV in a single run. Moreover, the unique adsorption properties of the immunoprobes were studied and could be extended to make other electrochemical immunosensors. Present work provided a promising way in developing simultaneous multiplexed analyses of a panel of targets.
Acknowledgements This research was financed by the grants from National Natural Science Foundation of China (21273153), Beijing Natural Science Foundation (2132008), and Research Base Construction Projects of Beijing Municipal Education Commission.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.03.036.
References Cao, X., Wang, N., Jia, S., Guo, L., Li, K., 2013. Biosens. Bioelectron. 39, 226–230. Chen, D.X., Wang, Q., Jin, J., Wu, P., Wang, H., Yu, S.Q., Zhang, H., Cai, C.X., 2010. Anal. Chem. 82, 2448–2455. Chen, X., Ma, Z.F., 2014. Biosens. Bioelectron. 55, 343–349. Cui, R.J., Liu, C., Shen, J.M., Gao, D., Zhu, J.J., Chen, H.Y., 2008. Adv. Funct. Mater. 18, 2197–2204. Gao, J., Guo, Z.K., Su, F.J., Gao, L., Pang, X.H., Cao, W., Du, B., Wei, Q., 2015. Biosens. Bioelectron. 63, 465–471. Gao, Q., Han, J.M., Ma, Z.F., 2013. Biosens. Bioelectron. 49, 323–328. Hu, B., Wang, K., Wu, L., Yu, S.H., Antonietti, M., Titirici, M.M., 2010. Adv. Mater. 22, 813–828. Hu, Y.S., Demir-Cakan, R., Titirici, M.M., Müller, J.O., Schlögl, R., Antonietti, M., Maier, J., 2008. Angew. Chem. Int. Ed. 47, 1645–1649. Jerzy, C., Dominik, J., Katarzyna, A., Michal, M., Mietek, J., 2012. Chem. Commun. 48, 3972–3974. Jia, X.L., Liu, Z.M., Liu, N., Ma, Z.F., 2014. Biosens. Bioelectron. 53, 160–166. Kang, X.H., Wang, J., Wu, H., Aksay, I.A., Liu, J., Lin, Y.H., 2009. Biosens. Bioelectron. 25, 901–905. LaMer, V.K., 1952. Ind. Eng. Chem. 44, 1270. Li, Y., Zhong, Z.Y., Chai, Y.Q., Song, Z.J., Zhuo, Y., Sun, H.L., Liu, S.M., Wang, D., Yuan, R., 2012. Chem. Commun. 48, 537–539. Lian, W.J., Liu, S., Yu, J.G., Li, J., Cui, M., Xu, W., Huang, J.D., 2013. Biosens. Bioelectron. 44, 70–76. Liu, N., Liu, Z.M., Han, H.L., Ma, Z.F., 2014. J. Mater. Chem. B 2, 3292–3298. Liu, X., Ju, H.X., 2008. Anal. Chem. 80, 5377–5382. Liu, Z., Liu, Q., Huang, Y., Ma, Y., Yin, S., Zhang, X., Sun, W., Chen, Y., 2008. Adv. Mater. 20, 3924–3930. Liu, Z.M., Ma, Z.F., 2013. Biosens. Bioelectron. 46, 1–7. Makowski, P., Cakan, R.D., Antonietti, M., Goettmann, F., Titirici, M.M., 2008. Chem. Commun., 999–1001. Petru, E., Sevin, B.U., Averette, H.E., Koechli, O.R., Perras, J.P., Hilsenbeck, S., 1990. Gynecol. Oncol., 181–186. Rezan, D.C., Niki, B., Markus, A., Maria-Magdalena, T., 2009. Chem. Mater. 21, 484–490. Shan, C.S., Yang, H.F., Han, D.X., Zhang, Q.X., Ivaska, A., Niu, L., 2009. Langmuir 25, 12030–12033. Sun, X.M., Li, Y.D., 2004. Angew. Chem. 116, 607–611. Tang, L.H., Wang, Y., Li, Y.M., Feng, H.B., Lu, J., Li, J.H., 2009. Adv. Funct. Mater. 19, 2782–2789. Tang, L.H., Wang, Y., Liu, Y., Li, J.H., 2011. ACS Nano 5, 3817–3822. Wang, Z.H., Liu, J., Liang, Q.L., Wang, Y.M., Luo, G.A., 2002. Analyst 127, 653–658. Wang, Z.F., Chen, X., Ma, Z.F., 2014. Biosens. Bioelectron. 61, 562–568. Wu, J., Fu, Z.F., Yan, F., Ju, H.X., 2007. Trends Anal. Chem. 26, 679–688. Xu, S.J., Liu, Y., Wang, T.H., Li, J.H., 2011. Anal. Chem. 83, 3817–3823.
