Author’s Accepted Manuscript Capillary electrophoresis-chemiluminescence detection for carcino-embryonic antigen based on aptamer/graphene oxide structure Zi-Ming Zhou, Zhe Feng, Jun Zhou, Bi-Yun Fang, Xiao-Xiao Qi, Zhi-Ya Ma, Bo Liu, Yuan-Di Zhao, Xue-Bin Hu www.elsevier.com/locate/bios
PII: DOI: Reference:
S0956-5663(14)00737-4 http://dx.doi.org/10.1016/j.bios.2014.09.050 BIOS7132
To appear in: Biosensors and Bioelectronic Received date: 16 June 2014 Revised date: 25 August 2014 Accepted date: 22 September 2014 Cite this article as: Zi-Ming Zhou, Zhe Feng, Jun Zhou, Bi-Yun Fang, Xiao-Xiao Qi, Zhi-Ya Ma, Bo Liu, Yuan-Di Zhao and Xue-Bin Hu, Capillary electrophoresis-chemiluminescence detection for carcino-embryonic antigen based on aptamer/graphene oxide structure, Biosensors and Bioelectronic, http://dx.doi.org/10.1016/j.bios.2014.09.050 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Capillary electrophoresis-chemiluminescence detection for carcino-embryonic antigen based on aptamer/graphene oxide structure Zi-Ming Zhou a+, Zhe Feng a+, Jun Zhou b, Bi-Yun Fang a, Xiao-Xiao Qi a, Zhi-Ya Ma a, Bo Liu a, Yuan-Di Zhao a*, Xue-Bin Hu c* a
Britton Chance Center for Biomedical Photonics at Wuhan National Laboratory for Optoelectronics – Hubei
Bioinformatics & Molecular Imaging Key Laboratory, Department of Biomedical Engineering, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, P. R. China b
Department of Clinical Laboratory, Wuhan Commercial Staff Hospital, Wuhan 430021, P. R. China.
c
Union Hospital, Tongji Medical College of Huazhong University of Science & Technology, Wuhan 430022, P. R.
China.
+ Z-M Zhou and Z. Feng contributed equally to this work. * To whom correspondence should be addressed.Fax: [email protected] (Y.D. Zhao), [email protected] (X.B. Hu)
+86-27-87792202.
Email:
Abstract A new strategy is proposed for determination of carcino-embryonic antigen (CEA) based on aptamer/graphene oxide (Apt/GO) by capillary electrophoresis-chemiluminescence (CE-CL) detection system. CEA aptamer conjugated with horseradish peroxidase (HRP) firstly mixes with GO, and the CL will be quenched because the stack of HRP-Apt on GO leads to chemiluminescence resonance energy transfer (CRET). When CEA exists, the specific combination of HRP-Apt and CEA can form HRP-Apt-CEA complex, which dissociates from GO. Then, the CL catalyzed by HRP-Apt-CEA complex can be detected without any CRET, and the content of CEA can be estimated by the CL intensity. It has been proved that the interference issue resulted from free 1
HRP-Apt is solved well by mixing GO firstly with HRP-Apt, which blocks the free HRP-Apt’s CL signal due to CL quenching effect of GO; and the interference resulted from GO to CL is also solved by CE, then the sensitivity and accuracy can be greatly improved. Results also showed that the CL intensity had a linear relationship with the concentration of CEA in the range from 0.0654 ~ 6.54 ng/mL, and the limit of detection was approximately 4.8 pg/mL (S/N=3). This proposed method with high specificity offers a new way for separation and determination of biomolecule, and has good potential in application of biochemistry and bioanalysis. Keywords: carcino-embryonic antigen; graphene oxide; aptamer; capillary electrophoresis; chemiluminescence 1. Introduction Carcino-embryonic antigen (CEA) is firstly proved to be a tumor-associated antigen in 1965 (Gold and Freeman, 1965). At present, as a tumor marker, CEA has an important role in the diagnosis and screening for colon cancer and other malignancies (Chan and Stanners, 2007). Recently, large number of clinical studies has shown that CEA level plays an important role in colorectal cancer surgery, and the elevated level of CEA indicates tumor recurrence (Su et al., 2012). Therefore, accurate determination of CEA has a very important significance in clinical medicine. At present, there are many methods for detection of CEA, like colorimetric immunoassay (Guo et al., 2013), fluorescent immunoassay (Yan et al., 2012), electrochemical immunoassay (Song et al., 2010; Zhang et at., 2007; Tang et. al., 2008; Tang et al., 2006), chemiluminescent immunoassay (Pei et al., 2010; Lin et al., 2004; Dungchai et al., 2007; Fu et al., 2007), and electrochemiluminescence immunoassay (Cheng et al., 2012; Cao et al., 2013). However, some of these methods are less sensitive; some need stable excitation source; some are easy interfered by external environment; 2
some are complex testing process, low sensitivity, and poor reproducibility. In a word, it is particularly important to develop new way for determination of CEA. Aptamers are the artificial single-stranded DNA or RNA sequences. Owing to their high affinity and thermostability, convenient synthesis and modification, and extremely high specificity with certain targets, aptamers have sparked tremendous interest in biomedical analysis (Iliuk et al., 2011). At present, aptamer that can specially recognize CEA has been screened out, and applied to analyze the content of CEA (Tabar and Smith, 2010; Lin et al., 2012). Recently, graphene oxide (GO) becomes a hot issue in aptasensor (Liu et al., 2012). It has been reported that single-stranded nucleic acid could adsorb strongly on GO by π-stacking interaction and dissociate from GO after specific combination with targets. What’s more, GO has good fluorescent quenching effect by effective fluorescence resonance energy transfer (FRET) between GO and fluorophores (Wang et al., 2012; Wang et al., 2010; Chen et al., 2012). Based on these theories, some groups had designed aptasensors by adsorption of aptamer conjugated with fluorophore on GO firstly, and then the fluorescence was quenched by FRET. In the presence of target, the FRET could be removed and then the fluorescence was restored. According to this light signal ‘off-on’ switch, it could realize the fluorescent detection of targets. For example, Chen’s group developed a high sensitive method based on aptamer/GO for triphosadenine and cocaine (Lu et al., 2010). In addition, it has reported that the quenching effect of GO is effective to chemiluminescence (CL) because of chemiluminescence resonance energy transfer (CRET) (Bi et al., 2012; Luo et al., 2012). For example, He and coworkers demonstrated a chemiluminescence biosensor for human immunodeficiency virus (HIV) oligonucleotide sequence based on GO and a horseradish peroxidase-mimicking DNAzyme (Luo et al., 2012). 3
Compared with fluorescence method, CL needs no external light source, which avoids the effects of stray light and instability of light source; therefore, CL has a high analytical sensitivity with simple equipment. CL detection based on luminol-H2O2-HRP system has been successfully applied in many field (Bi et al., 2012; Luo et al., 2012). However, CL is limited by less CL reaction system and poor selectivity, so it is frequently associated with separation technique to make up its shortage. Capillary electrophoresis (CE) is a powerful liquid phase separation technique, which has been widely applied in separation of biomolecule owing to its advantage in efficiency and sample consumption. It can simultaneously achieve efficient separation and sensitive detection of trace components in complex sample by combination of sensitive CL detection with effective CE separation (Zhang et al., 2008; Lara et al., 2010). For example, Zhang’s group designed an immunoassay method for determination of erythropoietin (EPO) by couple of CE and CL (Wang et al., 2009). In this paper, a new strategy was proposed for determination of CEA based on aptamer/GO (Apt/GO) by combining CE and CL detection system. Firstly, HRP-Apt was obtained by conjugation of CEA aptamer and HRP, and then incubated with GO. HRP-Apt adsorbed on GO to form HRP-Apt/GO and the CL catalyzed by HRP-Apt would be quenched in this situation due to CRET, so the interference issue by free HRP-Apt was solved. When CEA existed, the specific combination of CEA and aptamer formed HRP-Apt-CEA complex, it couldn’t bind to GO stably anymore, and dissociated from GO surface. Then, the CL catalyzed by HRP-Apt-CEA complex could be detected after separation of HRP-Apt-CEA complex and GO (or excess HRP-Apt/GO) by CE, and the interference resulted from GO to CL was also solved. Therefore, according to the light signal ‘off-on’ procedure, the content of CEA could be estimated by the CL intensity catalyzed by HPR-Apt-CEA 4
complex. It was indicated that the proposed method had high sensitivity and specificity, and achieved ng/mL accurate analysis of content of CEA in serums of patients. It was a meaningful technique for target molecule analysis based on GO and aptamer or molecule beacon in biosensing field. 2. Experimental section 2.1 Chemicals and Materials Sodium tetraborate (99%), boric acid, graphite (≥95%), potassium peroxodisulfate, phosphorus pentoxide, vitriol (98%), potassium permanganate, hydrogen peroxide (30%) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). 1-ethyl-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC·HCl, 98.5%), N-hydroxy succinimide sulfur generation (Sulfo-NHS), luminol, para-iodophenol (p-Ip), and horseradish peroxidase were purchased from Aladdin Chemistry (Shanghai) Co., Ltd. (Shanghai, China). Ultrafilter was supplied by Millipore (Bedford, MA, America). CEA was purchased from Meridian Life Science, Inc. (Memphis, America). Human immune globulin (IgG, 99%), human immunoglobulin (HSA, 99%) were obtained from Wuhan Boster Biological Technology Co., Ltd. (Wuhan, China). Thrombin was supplied by Sigma-Aldrich Fine Chemicals (St. Louis, MO, America). CEA aptamer, the base sequences as follow: 5’-NH2-(CH2)6- ATACCAGCTTATTCAATT-3’, was synthesized and purified by Sangon Biotech Co., Ltd. (Shanghai. China), and then diluted to 100 μM by PBS (0.01 M, pH 7.4). Fused-silica capillary (75 μm of inner diameter) was supplied by Yongnian Optical Fiber Factory (Hebei, China). All other materials and reagents were of analytical grade, and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Ultrapure water (≥18.2 MΩ) from a Milli-Q system (Millipore, Bedford, MA, America) was used for all solutions. The synthesis and characterization of GO was referred to our previous work (Zhou et al., 2014). 5
2.2 Apparatus The absorption spectra were measured by UV-vis spectrophotometer (UV-2550, Shimadzu, Japan). A chemiluminescent immunoassay system (LIASON, DiaSorin S.p.A, Italy) with standard CEA kit (DiaSorin S.p.A, Italy) was used to standardize the concentrations of CEA in patient’s sample. CE analyses with chemiluminescent detection were carried out on a home-built system, consisting of a high-voltage supply (0-30 kV) (Shanghai Nuclear Research Institute, China), DHL-B computer constant flow pump (Shanghai Huxi Analysis Instrument Factory Co., Ltd., China), a QE65000 fiber-optic spectrometer (Hamamatsu S7031-1006 back-thinned CCD) (Ocean Optics, U.S.A) with a SI 600/720 fiber (diameter of 80 mm, length of 2 m, Yangtze Optical Fiber and Cable Co., Ltd., China), a four-port plastic tube with inner diameter of 1.6 mm, and also a personal computer. All other instrumentations included Eppendorf 5415 D rotary concentration meter (Gene Co., Ltd., Germany), KQ5200 ultrasonic cleaner (Kunshan ultrasonic instrument Co., Ltd., China), and a Milli-Q system (Millipore, Bedford, MA, America). 2.3 Preparation of HRP Conjugated with CEA Aptamer (HRP-Apt) HRP (20 μL, 2 mg/mL) was put into EDC·HCl solution (30 μL, 5 mg/mL) followed by addition of Sulfo-NHS (20 μL, 1.5 mg/mL), and the solution was kept stirring for 0.5 h. After that, CEA aptamer (50 μL, 100 μM) solution was added and the mixture was whirled. Following the reaction for 2 h, the solution was centrifuged by ultrafilter to remove excess aptamer. Finally, the obtained residue was dissolved in PBS buffer for future experiments. 2.4 Preparation of HRP-Apt/GO for Detection of CEA HRP-Apt (5 μL) was added in GO solution (5 μL) and then whirled. After incubation for 4 h, the HRP-Apt/GO solution was obtained. Then, CEA samples were added respectively and kept 6
stirring for 4 h. the measurements for final solutions were performed by CE-CL at 25 ºC. 2.5 CE-CL Procedure CE-CL analyses were carried out on a home-built system. A capillary was fixed on the platform, and the end column was horizontally inserted in a four-port plastic tube; the fiber connected to the fiber optic spectrometer for sensing and transducing light signal was inserted in another port of the four-port plastic tube, as close as possible to the end column of the capillary for detecting CL signal. CL buffer (containing 10 mM luminol, 10 mM p-Ip, 20 mM hydrogen peroxide) was injected continuously in the four-port plastic tube by computer constant flow pump. The non-coating quartz capillary was 60 cm long with inner diameter of 75 μm. Before analysis, the capillary was rinsed for 20 min, successively by 1 M NaOH, ultrawater, 1 M HCl, ultrawater, electrophoresis buffer (Na2B4O7, pH 8.4). The electrophoresis buffer with different pH was obtained by adjusting 0.05 M Na2B4O7 and 0.2 M H3BO2. The anode at the capillary outlet end, was used for all separations, and finally, equilibrated with electrophoresis buffer for 20 min. The samples were hydrodynamically introduced to the capillary. In addition, the capillary was washed by electrophoresis buffer for 15 min in sequence to ensure the reproducibility.. 2.6 Detection of CEA in Patient’s Serums The serum samples of patient were come from Wuhan Commercials Staff Hospital, including 5 positive samples and 5 negative samples. All of the concentrations of CEA in samples had been standardized by a Chemiluminescent Immunoassay System (LIASON, DiaSorin S.p.A, Italy) with standard CEA kit (DiaSorin S.p.A, Italy). All positive samples and negative samples were diluted to proper solution and put into HRP-Apt/GO solution, respectively and standing for 4 h. The final solutions were analyzed by CE-CL, successively. 7
3. Results and discussion 3.1. CE-CL Analysis of HRP, HRP-Apt, and HRP-Apt-CEA. We studied the UV-vis absorption of CEA aptamer before and after conjugating with HRP (Fig. S1). It was found all of HRP had conjugated with aptamer in the present of coupling reagent of EDC and Sulfo-NHS. After obtaining the coupling product, CE-CL analyses for each sample were carried out on a home-built system (Fig. 1). Compared with blank sample (Fig. 1, a), a CL electrophoresis peak at about 496.2 ± 5.1 s was observed when 0.1 mg/mL HRP was introduced (Fig. 1, b); the CL electrophoresis peak was hardly shifted after aptamer mixed with HRP (Fig. 1, c), suggesting that aptamer wasn’t attached to HRP if they were just mixed simply. This result was also consistent with analysis of UV-vis spectrum (Fig. S1 (B)). When HRP-Apt was introduced to capillary, it was found that the CL electrophoresis peak catalyzed by HRP-Apt (the concentration of HRP was also 0.1 mg/mL) was shifted to 461 ± 6.7 s (Fig. 1, d), probably because of the decrease of the charge-mass ratio; after adding 0.514 μg/mL CEA, two CL electrophoresis peaks with serious overlap were observed (Fig. 1, e), one was consistent with HRP-Apt (Fig. 1, d), and the other was shifted to 445 ± 8.3 s, in advance about 16 s. It was conjectured that after combination with CEA, the aptamer liked a chain bridge that bound HRP and CEA together, forming HRP-Apt-CEA complex, a double-protein complex structure, and the new peak should be related to the HRP-Apt-CEA complex. These two electrophoresis peaks were considerable overlap, indicating the poor separation efficiency for HRP-Apt and HRP-Apt-CEA complex in this condition. Therefore, some experiments had been done to improve the separation efficiency of HRP-Apt and HRP-Apt-CEA complex by optimizing the experimental conditions.
8
3.2. CE-CL Analysis of CEA by Optimizing the Experimental Conditions CL is a chemical kinetic process with the signal occurring, rising and decaying as the reaction starts, proceeds and finishes. Therefore, it presents a serious overlap and interference of optical signal due to relaxation phenomenon, especially when it is coupled with separation technologies, such as CE (Dadoo et al., 1994). We designed the end-column detection of CE-CL system and decrease the influence of relaxation phenomenon by continuously injecting CL buffer. The injection rate of CL buffer was studied (Fig. S2). It was found CL detection was achieved completely and relaxation phenomenon was effective decreased by continuously injecting fresh CL buffer, when the injection rate was 0.01 mL/min. Therefore, 0.01 mL/min was applied to subsequent experiments. In generally, the pH of electrophoretic buffer and the separation voltage have influence on the separation efficiency in CE experiments. Firstly, the influence of pH of the electrophoretic buffer was studied (Fig. S3). It could be seen the overlap of CL electrophoresis peaks was still serious, although the separation efficiency for HRP-Apt and HRP-Apt-CEA complex was improved in a certain extent when pH was 8.4 (Fig. S3, b). Secondly, the separation efficiency for HRP-Apt and HRP-Apt-CEA had been small improved by increasing the separation voltage other than migration time decreasing (Fig. S4). It might be due to the similar property of HRP-Apt and HRP-Apt-CEA complex, and this is a common issue in protein separation by CE (Wang et al., 2009; Shen et al., 2010; Nilsson et al., 2011). For example, Zhang’s group studied the detection of EPO by CE-CL and found a serious overlap electrophoresis peaks coursed by freely HRP-EPO and immune-complex (Wang et al., 2009). Therefore, 8.4 for the pH of electrophoretic buffer and 17 kV for separation voltage were used in subsequent experiments. 9
3.3. Design Strategy for Detection of CEA Based on Aptamer/GO by CE-CL According to the above results, it could be seen that HRP-Apt and HRP-Apt-CEA complex could not be completely separated, even though optimized the experimental conditions by CE-CL. In order to realize the separation of HRP-Apt and HRP-Apt-CEA complex, a new strategy was proposed for analysis of content of CEA based on the couple of GO and HRP-Apt by CE-CL detection (Fig. 2). In the absence of CEA, HRP-Apt was stacked on GO surface and formed HRP-Apt/GO, and the CL catalyzed by HRP-Apt would be not observed in this situation because GO could quench the CL by CRET. When CEA existed, the specific combination of HRP-Apt and CEA formed HRP-Apt-CEA complex, it couldn’t bind to GO stably anymore, and dissociated from GO. The CL catalyzed by HRP-Apt-CEA complex could be detected due to CRET being removed; after separating by CE, the interference resulted from GO to CL was also solved. Therefore, according to the light signal ‘off-on’ procedure, CEA could be estimated by the CL intensity catalyzed by HPR-Apt-CEA complex, and the separation of GO and HRP-Apt-CEA complex by CE achieved great improvement of CL detection sensitivity.
3.4. CE-CL Analysis of Combination of GO and HRP-Apt Reproducibility On the basis of the above optimized conditions, the feasibility of the strategy was further verified. According to the above results, the CL electrophoresis peaks catalyzed by HRP-Apt and HRP-Apt-CEA complex were 506 ± 6.5 s᧨468 ± 7.4 s, respectively, in the optimized conditions (Fig. 3, a), but the overlap of CL electrophoresis peaks was still serious. However, when GO was added, it could be seen the CL electrophoresis peak catalyzed by HRP-Apt-CEA complex was hardly decreased and shifted (466 ± 6.7 s), and the CL electrophoresis peak catalyzed by HRP-Apt (497 ± 10
7.1 s) decreased obviously (Fig. 3, b). It was indicated that the CL catalyzed by HRP-Apt was quenched by effective CRET due to HRP-Apt attaching on GO, the CL catalyzed by HRP-Apt-CEA complex didn’t decrease because HRP-Apt-CEA complex couldn’t bind to GO stably and CRET did not occur. When the concentration of GO increased to 20 μg/mL, it could be seen the CL electrophoresis peak catalyzed by HRP-Apt disappeared and the CL electrophoresis peak catalyzed by HRP-Apt-CEA was still not changed (Fig. 3, c), suggesting HRP-Apt completely stacked on GO. The same phenomenon was observed by increasing the concentration of GO sequentially (Fig. 3, d and e). It was worth noting that there was a weak CL peak at 310 ± 5.9 s after addition of GO, and it might be caused by HRP-Apt/GO, which meant the CL catalyzed by HRP-Apt hadn’t been quenched completely by GO. But, it was certain that the CL catalyzed by HRP-Apt-CEA complex would not be interfered by this residual CL due to good separation effect between HRP-Apt/GO and HRP-Apt-CEA complex by CE. In a word, the feasibility of the strategy was verified. In order to further optimize the addition of GO, the CL intensity catalyzed by HRP-Apt was studied by adding different concentration of GO in the absence of CEA (Fig. S5). It could be seen that the CL electrophoresis peaks catalyzed by HRP-Apt disappeared when the concentration of GO increased to 25 μg/mL, indicating the CL electrophoresis peak catalyzed by HRP-Apt was totally quenched by GO. Therefore, 25 μg/mL of GO was applied in the subsequent experiments.
3.5. CE-CL Analysis of Combination of CEA and HRP-Apt/GO Fig. 4A showed the electropherograms for the CL intensity catalyzed by HRP-Apt-CEA complex with different concentration of CEA. It was found the CL intensity catalyzed by HRP-Apt-CEA complex increased rapidly with the increase of concentration of CEA. Fig. 4B exhibited the 11
relationship of the peak area of the CL electrophoresis peak catalyzed by HRP-Apt-CEA and the concentration of CEA. It could be seen that the CL intensity increased rapidly below 6.54 ng/mL, and slowed down when the concentration of CEA was above 6.54 ng/mL. The relationship was linear in the range from 0.0654 ~ 6.54 ng/mL (inset in Fig. 4B), and the concentration of CEA could be estimated by the equation, Log [I] = 1.633 + 0.830 × Log [c] (R2 = 0.992), where [I] was the CL intensity of the CL electrophoresis peak area after integration, and [c] was the concentration of CEA. The limit of detection was 4.8 pg/mL based on a signal-to-noise ratio of 3. Table S1 listed some common methods for the detection of the content of CEA. Relative to the limit of detection, the proposed method had a lower limit of detection than most of others.
3.6. Specificity. Three kinds of proteins, IgG, HSA, and thrombin were selected as interferent under optimized conditions. The four proteins were incubated with the HRP-Apt/GO solution for 4 h, respectively, and then the four solutions were measured by CE-CL. It was found the interferential proteins did not restore CL intensity as much as CEA did, and CEA could be clearly distinguished from other proteins though the concentration of CEA was three orders of magnitude lower than those interferential proteins (Fig. 5). Such results indicated that the proposed method had a high specificity for CEA.
3.7. Analysis of Content of CEA in Patient Serum. The proposed method was applied to analyze the content of CEA in patient’s serum. The serum samples of patient were come from Wuhan Commercials Staff Hospital, which included five positive samples and five negative samples. The concentrations of CEA in samples had been standardized by 12
LIAISON chemiluminescent immunoassay system with standard CEA kit. Table 1 shows the detection results of the 10 samples by LIAISON chemiluminescent immunoassay system and the proposed method. The samples in the Table 1 had to be diluted into the linear range before analyzing. After CE-CL experiments, the diluted concentration of samples could be calculated by the linear equation. And then, multiplied by the diluted factor, the actual concentration of samples could be estimated. It was found the results detected by our method were very close to the calibration values by LIAISON chemiluminescence immunoassay system, indicating that the proposed method was good at quantitative analysis of content of CEA in practical sample.
4. Conclusions In conclusion, a new strategy is proposed for determination of CEA based on Apt/GO by couple of CE-CL detection system. The interference issue is solved well by mixing GO firstly with HRP-Apt due to CL quenching effect of GO, which blocked the free HRP-Apt signal, and the interference resulted from GO to CL is also solved by CE, then the sensitivity and accuracy could be greatly improved and achieved ng/mL accurate analysis of content of CEA in serums of patients. This proposed method with high specificity offered a new way for separation and determination of biomolecule and had good potential in application of biochemistry and bioanalysis.
Acknowledgements This work was supported by the National Key Technology R&D Program of China (2012BAI23B02), National Natural Science Foundation of China (Grant No. 81271616, 81471697), the Foundation for Innovative Research Groups of the NNSFC (Grant No. 61121004), the Fundamental Research Funds 13
for the Central Universities (Hust, 2014YGYL022, 2013TS085, 2014QN116), Scientific Research Fund of Union Hospital. We also thank the facility support of the Center for Nanoscale Characterization and Devices, Wuhan National Laboratory for Optoelectronics (WNLO) of HUST and Analytical and Testing Center (HUST) for the help of measurement.
References Bi, S., Zhao, T.T., Luo, B.Y., 2012. Chem. Comm., 48, 106-108. Cao, X., Wang, N., Jia, S., Guo, L., Li, K., 2013. Biosens. Bioelectron., 39, 226-230. Chan, C.H.F., Stanners, C.P., 2007. Curr. Oncol., 14, 70-73. Chen, L., Zhang, X.W., Zhou, G.H., Xiang, X., Li, X.H., Zheng, Z.H., He, Z.K., Wang, H. Z., 2012. Anal. Chem., 84, 3200-3207. Cheng, Y.F., Yuan, R., Chai, Y.Q., Niu, H., Cao, Y.L., Liu, H.J., Yuan, Y.L., 2012. Clin. Chim. Acta, 745, 137-142. Dungchai, W., Siangproh, W., Lin, J.M., Chailapakul, O., Lin, S., Ying, X.T., 2007. Anal. Bioanal. Chem., 387, 1965-1971. Fu, Z.F.,Yang, Z.J., Tang, J.H., Liu, H., Yan, F., Ju, H.X., 2007. Anal. Chem., 79, 7376-7382. Gold, P., Freedman, S.O., 1965. J. Exp. Med., 122, 467-481. Guo, X.D., Huang, Q., Lin, Y.N., 2013. Anal. Lett., 46, 2040-2047. Iliuk, A.B., Hu, L.H., Tao, W.A., 2011. Anal. Chem., 83, 4440-4452. Lara, F.J., García-Campańa, A.M., Velasco, A. I., 2010. Electrophoresis, 31, 1998-2027. Lin, J.H., Yan, F., Ju, H.X., 2004. Clin. Chim. Acta, 341, 109-115. Lin, Z.Y., Zhang, G.Y., Yang, W.Q., Qiu, B., Chen, G.N., 2012. Chem. Comm., 48, 9918-9920. Liu, Y.X., Dong, X.C., Chen, P., 2012 Chem. Soc. Rev., 41, 2283-2307. Lu, C.H., Li, J., Lin, M.H., Wang, Y.W., Yang, H.H., Chen, X., Chen, G.N., 2010 Angew. Chem., Int. Ed., 49, 8454-8457. Luo, M., Chen, X., Zhou, G.H., Xiang, X., Chen, L., Ji, X.H., He, Z.K., 2012. Chem. Comm., 48, 1126-1128. Nilsson, C., Birnbaum, S., Nilsson, S., 2011. Electrophoresis, 32, 1141-1147. Pei, X.P., Chen, B.A., Li, L., Gao, F., Jiang, Z., 2010. Analyst, 135, 177-181. Shen, R., Guo, L., Zhang, Z.Y., Meng, Q.W., Xie, J.W., 2010. J. Chromatogr. A, 1217, 5635-5641. Song, Z.J., Yuan, R., Chai, Y.Q., Zhuo, Y., Jiang, W., Su, H.L., Che, X., Li, J.J., 2010. Chem. Commun., 46, 6750-6752. Su, B.B., Shi, H., Wan, J., 2012. World J. Gastroentero., 18, 2121-2126. Tabar, G.R.H., Smith, L.C., 2010. World Appl. Sci. J., 8, 16-21. Tang, D.P., Yuan, R., Chai, Y.Q., 2006. J. Phys. Chem. B, 110, 11640-11646. Tang, D.P., Yuan, R., Chai, Y. Q., 2008. Anal. Chem., 80, 1582-1588. Wang, L., Zhu, J.B., Han, L., Jin, L.H., Zhu, C.Z., Wang, E.K., Dong, S.J., 2012. ACS Nano, 6, 6659-6666. 14
Wang, W.J., Zhang, S.H., Liu, C.H., Lu, L.Z., Wang, S.D., Zhang, X.R., 2009. Electrophoresis, 30, 3092-3098. Wang, Y., Li, Z.H., Hu, D.H., Lin, C.T., Li, J.H., Lin, Y., 2010. J. Am. Chem. Soc., 132, 9274-9276. Yan, M., Ge, S.G., Gao, W.Q., Chu, C.C., Yu, J.H., Song, X.R., 2012. Analyst, 137, 2834-2839. Zhang, B., Zhang, X., Yan, H.H., Xu, S.J., Tang, D.H., Fu, W.L., 2007. Biosens. Bioelectron., 23, 19-25. Zhang, H.Q., Li, X.F., Le, X.C., 2008. J. Am. Chem. Soc., 130, 34-35. Zhou, Z.M., Zhou, J., Chen, J., Yu, R.N., Zhang, M.Z., Song, J.T., Zhao, Y.D., 2014. Biosens. Bioelectron. 59, 397-403.
Figture capitons: Fig. 1. Electropherograms for blank (a); HRP (b); HRP mixed with aptamer simply (c); HRP conjugated with aptamer by the couple reagents (d); HRP-Apt incubated with 0.514 μg/mL CEA (e). The separation voltage was 17 kV, the pH of electrophoresis buffer was 8.7, and the injection rate of buffer was 0.01 mL/min. The concentrations of HRP and HRA-Apt were 0.1 mg/mL, and 10 mM, 10mM, 20 mM for luminol, p-Ip, H2O2, respectively. Each experiment had been done for 3 times. Fig. 2. Schematic depiction of detection of CEA based on aptamer/GO by CE-CL. Fig. 3. Electropherograms for HRP-Apt-CEA incubated with different concentration of GO. a) 0᧨b) 10, c) 20, d) 25, e) 30 μg/mL. The pH of electrophoretic buffer was 8.4. The other conditions were the same as Fig. 1 and each experiment had been done for 3 times. Fig. 4. Electropherograms for the CL of HRP-Apt/GO restoring with different concentration of CEA (A) and the relationship of CL intensity and the concentration of CEA (B). The concentration of CEA was 0, 0.0654᧨0.129, 0.257, 0.645, 1.29, 2.57, 6.45, 12.9, 25.7 and 129 ng/mL from a to k. The inset of A showed the amplifying diagram of the CL electrophoresis peaks of curve a and b; the inset of B showed the logarithm linear relationship between the CL intensity and the concentration of CEA. Error bar indicated the standard deviation (n=3). The concentration of GO was 25 μg/mL and the pH of electrophoretic buffer was 8.4. The other conditions were the same as Fig. 1 and each experiment had been done for 3 times. Error bar indicated the standard deviation.
15
Fig. 5. Specificity of the proposed method. The concentration of CEA was 2.57 ng/mL, and IgG, HSA, thrombin were 10 μg/mL, 10 mg/mL and 10 μg/mL, respectively. The other conditions were the same as Fig. 4 and the error bar indicated the standard deviation.
Table 1. The Detecting Results of CEA in Patient’s Serum a Number Positive Sample b
Results c
Number Negative sample b
Results c
1
130.8
128.3 ± 4.1
6
1.60
1.61 ± 0.04
2
122.1
119.6 ± 6.4
7
1.37
1.33 ± 0.09
3
17.1
17.3 ± 0.66
8
1.79
1.74 ± 0.07
4
339.4
340.3± 8.1
9
3.07
3.13 ± 0.07
5
74.2
74.3 ± 3.9
10
1.59
1.59 ± 0.05
a
The concentration unit of Table 1 was ng/mL. The calibration value for the samples by LIAISON chemiluminescence immunoassay system. c The results detected by our method, and each sample had been tested for three times. b
Highlights z z z z
Carcino-embryonic antigen is detected by capillary electrophoresis-chemiluminescence. The interference issue is solved well by mixing GO firstly with HRP-Apt. Good sensitivity and specificity has been obtained. The proposed method is good at quantitative analysis of CEA in practical sample.
16
Figure 1
Figure 2
Figure 3
Figure 4-A
Figure 4-B
Figure 5
PII: DOI: Reference:
S0956-5663(14)00737-4 http://dx.doi.org/10.1016/j.bios.2014.09.050 BIOS7132
To appear in: Biosensors and Bioelectronic Received date: 16 June 2014 Revised date: 25 August 2014 Accepted date: 22 September 2014 Cite this article as: Zi-Ming Zhou, Zhe Feng, Jun Zhou, Bi-Yun Fang, Xiao-Xiao Qi, Zhi-Ya Ma, Bo Liu, Yuan-Di Zhao and Xue-Bin Hu, Capillary electrophoresis-chemiluminescence detection for carcino-embryonic antigen based on aptamer/graphene oxide structure, Biosensors and Bioelectronic, http://dx.doi.org/10.1016/j.bios.2014.09.050 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Capillary electrophoresis-chemiluminescence detection for carcino-embryonic antigen based on aptamer/graphene oxide structure Zi-Ming Zhou a+, Zhe Feng a+, Jun Zhou b, Bi-Yun Fang a, Xiao-Xiao Qi a, Zhi-Ya Ma a, Bo Liu a, Yuan-Di Zhao a*, Xue-Bin Hu c* a
Britton Chance Center for Biomedical Photonics at Wuhan National Laboratory for Optoelectronics – Hubei
Bioinformatics & Molecular Imaging Key Laboratory, Department of Biomedical Engineering, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, P. R. China b
Department of Clinical Laboratory, Wuhan Commercial Staff Hospital, Wuhan 430021, P. R. China.
c
Union Hospital, Tongji Medical College of Huazhong University of Science & Technology, Wuhan 430022, P. R.
China.
+ Z-M Zhou and Z. Feng contributed equally to this work. * To whom correspondence should be addressed.Fax: [email protected] (Y.D. Zhao), [email protected] (X.B. Hu)
+86-27-87792202.
Email:
Abstract A new strategy is proposed for determination of carcino-embryonic antigen (CEA) based on aptamer/graphene oxide (Apt/GO) by capillary electrophoresis-chemiluminescence (CE-CL) detection system. CEA aptamer conjugated with horseradish peroxidase (HRP) firstly mixes with GO, and the CL will be quenched because the stack of HRP-Apt on GO leads to chemiluminescence resonance energy transfer (CRET). When CEA exists, the specific combination of HRP-Apt and CEA can form HRP-Apt-CEA complex, which dissociates from GO. Then, the CL catalyzed by HRP-Apt-CEA complex can be detected without any CRET, and the content of CEA can be estimated by the CL intensity. It has been proved that the interference issue resulted from free 1
HRP-Apt is solved well by mixing GO firstly with HRP-Apt, which blocks the free HRP-Apt’s CL signal due to CL quenching effect of GO; and the interference resulted from GO to CL is also solved by CE, then the sensitivity and accuracy can be greatly improved. Results also showed that the CL intensity had a linear relationship with the concentration of CEA in the range from 0.0654 ~ 6.54 ng/mL, and the limit of detection was approximately 4.8 pg/mL (S/N=3). This proposed method with high specificity offers a new way for separation and determination of biomolecule, and has good potential in application of biochemistry and bioanalysis. Keywords: carcino-embryonic antigen; graphene oxide; aptamer; capillary electrophoresis; chemiluminescence 1. Introduction Carcino-embryonic antigen (CEA) is firstly proved to be a tumor-associated antigen in 1965 (Gold and Freeman, 1965). At present, as a tumor marker, CEA has an important role in the diagnosis and screening for colon cancer and other malignancies (Chan and Stanners, 2007). Recently, large number of clinical studies has shown that CEA level plays an important role in colorectal cancer surgery, and the elevated level of CEA indicates tumor recurrence (Su et al., 2012). Therefore, accurate determination of CEA has a very important significance in clinical medicine. At present, there are many methods for detection of CEA, like colorimetric immunoassay (Guo et al., 2013), fluorescent immunoassay (Yan et al., 2012), electrochemical immunoassay (Song et al., 2010; Zhang et at., 2007; Tang et. al., 2008; Tang et al., 2006), chemiluminescent immunoassay (Pei et al., 2010; Lin et al., 2004; Dungchai et al., 2007; Fu et al., 2007), and electrochemiluminescence immunoassay (Cheng et al., 2012; Cao et al., 2013). However, some of these methods are less sensitive; some need stable excitation source; some are easy interfered by external environment; 2
some are complex testing process, low sensitivity, and poor reproducibility. In a word, it is particularly important to develop new way for determination of CEA. Aptamers are the artificial single-stranded DNA or RNA sequences. Owing to their high affinity and thermostability, convenient synthesis and modification, and extremely high specificity with certain targets, aptamers have sparked tremendous interest in biomedical analysis (Iliuk et al., 2011). At present, aptamer that can specially recognize CEA has been screened out, and applied to analyze the content of CEA (Tabar and Smith, 2010; Lin et al., 2012). Recently, graphene oxide (GO) becomes a hot issue in aptasensor (Liu et al., 2012). It has been reported that single-stranded nucleic acid could adsorb strongly on GO by π-stacking interaction and dissociate from GO after specific combination with targets. What’s more, GO has good fluorescent quenching effect by effective fluorescence resonance energy transfer (FRET) between GO and fluorophores (Wang et al., 2012; Wang et al., 2010; Chen et al., 2012). Based on these theories, some groups had designed aptasensors by adsorption of aptamer conjugated with fluorophore on GO firstly, and then the fluorescence was quenched by FRET. In the presence of target, the FRET could be removed and then the fluorescence was restored. According to this light signal ‘off-on’ switch, it could realize the fluorescent detection of targets. For example, Chen’s group developed a high sensitive method based on aptamer/GO for triphosadenine and cocaine (Lu et al., 2010). In addition, it has reported that the quenching effect of GO is effective to chemiluminescence (CL) because of chemiluminescence resonance energy transfer (CRET) (Bi et al., 2012; Luo et al., 2012). For example, He and coworkers demonstrated a chemiluminescence biosensor for human immunodeficiency virus (HIV) oligonucleotide sequence based on GO and a horseradish peroxidase-mimicking DNAzyme (Luo et al., 2012). 3
Compared with fluorescence method, CL needs no external light source, which avoids the effects of stray light and instability of light source; therefore, CL has a high analytical sensitivity with simple equipment. CL detection based on luminol-H2O2-HRP system has been successfully applied in many field (Bi et al., 2012; Luo et al., 2012). However, CL is limited by less CL reaction system and poor selectivity, so it is frequently associated with separation technique to make up its shortage. Capillary electrophoresis (CE) is a powerful liquid phase separation technique, which has been widely applied in separation of biomolecule owing to its advantage in efficiency and sample consumption. It can simultaneously achieve efficient separation and sensitive detection of trace components in complex sample by combination of sensitive CL detection with effective CE separation (Zhang et al., 2008; Lara et al., 2010). For example, Zhang’s group designed an immunoassay method for determination of erythropoietin (EPO) by couple of CE and CL (Wang et al., 2009). In this paper, a new strategy was proposed for determination of CEA based on aptamer/GO (Apt/GO) by combining CE and CL detection system. Firstly, HRP-Apt was obtained by conjugation of CEA aptamer and HRP, and then incubated with GO. HRP-Apt adsorbed on GO to form HRP-Apt/GO and the CL catalyzed by HRP-Apt would be quenched in this situation due to CRET, so the interference issue by free HRP-Apt was solved. When CEA existed, the specific combination of CEA and aptamer formed HRP-Apt-CEA complex, it couldn’t bind to GO stably anymore, and dissociated from GO surface. Then, the CL catalyzed by HRP-Apt-CEA complex could be detected after separation of HRP-Apt-CEA complex and GO (or excess HRP-Apt/GO) by CE, and the interference resulted from GO to CL was also solved. Therefore, according to the light signal ‘off-on’ procedure, the content of CEA could be estimated by the CL intensity catalyzed by HPR-Apt-CEA 4
complex. It was indicated that the proposed method had high sensitivity and specificity, and achieved ng/mL accurate analysis of content of CEA in serums of patients. It was a meaningful technique for target molecule analysis based on GO and aptamer or molecule beacon in biosensing field. 2. Experimental section 2.1 Chemicals and Materials Sodium tetraborate (99%), boric acid, graphite (≥95%), potassium peroxodisulfate, phosphorus pentoxide, vitriol (98%), potassium permanganate, hydrogen peroxide (30%) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). 1-ethyl-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC·HCl, 98.5%), N-hydroxy succinimide sulfur generation (Sulfo-NHS), luminol, para-iodophenol (p-Ip), and horseradish peroxidase were purchased from Aladdin Chemistry (Shanghai) Co., Ltd. (Shanghai, China). Ultrafilter was supplied by Millipore (Bedford, MA, America). CEA was purchased from Meridian Life Science, Inc. (Memphis, America). Human immune globulin (IgG, 99%), human immunoglobulin (HSA, 99%) were obtained from Wuhan Boster Biological Technology Co., Ltd. (Wuhan, China). Thrombin was supplied by Sigma-Aldrich Fine Chemicals (St. Louis, MO, America). CEA aptamer, the base sequences as follow: 5’-NH2-(CH2)6- ATACCAGCTTATTCAATT-3’, was synthesized and purified by Sangon Biotech Co., Ltd. (Shanghai. China), and then diluted to 100 μM by PBS (0.01 M, pH 7.4). Fused-silica capillary (75 μm of inner diameter) was supplied by Yongnian Optical Fiber Factory (Hebei, China). All other materials and reagents were of analytical grade, and purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Ultrapure water (≥18.2 MΩ) from a Milli-Q system (Millipore, Bedford, MA, America) was used for all solutions. The synthesis and characterization of GO was referred to our previous work (Zhou et al., 2014). 5
2.2 Apparatus The absorption spectra were measured by UV-vis spectrophotometer (UV-2550, Shimadzu, Japan). A chemiluminescent immunoassay system (LIASON, DiaSorin S.p.A, Italy) with standard CEA kit (DiaSorin S.p.A, Italy) was used to standardize the concentrations of CEA in patient’s sample. CE analyses with chemiluminescent detection were carried out on a home-built system, consisting of a high-voltage supply (0-30 kV) (Shanghai Nuclear Research Institute, China), DHL-B computer constant flow pump (Shanghai Huxi Analysis Instrument Factory Co., Ltd., China), a QE65000 fiber-optic spectrometer (Hamamatsu S7031-1006 back-thinned CCD) (Ocean Optics, U.S.A) with a SI 600/720 fiber (diameter of 80 mm, length of 2 m, Yangtze Optical Fiber and Cable Co., Ltd., China), a four-port plastic tube with inner diameter of 1.6 mm, and also a personal computer. All other instrumentations included Eppendorf 5415 D rotary concentration meter (Gene Co., Ltd., Germany), KQ5200 ultrasonic cleaner (Kunshan ultrasonic instrument Co., Ltd., China), and a Milli-Q system (Millipore, Bedford, MA, America). 2.3 Preparation of HRP Conjugated with CEA Aptamer (HRP-Apt) HRP (20 μL, 2 mg/mL) was put into EDC·HCl solution (30 μL, 5 mg/mL) followed by addition of Sulfo-NHS (20 μL, 1.5 mg/mL), and the solution was kept stirring for 0.5 h. After that, CEA aptamer (50 μL, 100 μM) solution was added and the mixture was whirled. Following the reaction for 2 h, the solution was centrifuged by ultrafilter to remove excess aptamer. Finally, the obtained residue was dissolved in PBS buffer for future experiments. 2.4 Preparation of HRP-Apt/GO for Detection of CEA HRP-Apt (5 μL) was added in GO solution (5 μL) and then whirled. After incubation for 4 h, the HRP-Apt/GO solution was obtained. Then, CEA samples were added respectively and kept 6
stirring for 4 h. the measurements for final solutions were performed by CE-CL at 25 ºC. 2.5 CE-CL Procedure CE-CL analyses were carried out on a home-built system. A capillary was fixed on the platform, and the end column was horizontally inserted in a four-port plastic tube; the fiber connected to the fiber optic spectrometer for sensing and transducing light signal was inserted in another port of the four-port plastic tube, as close as possible to the end column of the capillary for detecting CL signal. CL buffer (containing 10 mM luminol, 10 mM p-Ip, 20 mM hydrogen peroxide) was injected continuously in the four-port plastic tube by computer constant flow pump. The non-coating quartz capillary was 60 cm long with inner diameter of 75 μm. Before analysis, the capillary was rinsed for 20 min, successively by 1 M NaOH, ultrawater, 1 M HCl, ultrawater, electrophoresis buffer (Na2B4O7, pH 8.4). The electrophoresis buffer with different pH was obtained by adjusting 0.05 M Na2B4O7 and 0.2 M H3BO2. The anode at the capillary outlet end, was used for all separations, and finally, equilibrated with electrophoresis buffer for 20 min. The samples were hydrodynamically introduced to the capillary. In addition, the capillary was washed by electrophoresis buffer for 15 min in sequence to ensure the reproducibility.. 2.6 Detection of CEA in Patient’s Serums The serum samples of patient were come from Wuhan Commercials Staff Hospital, including 5 positive samples and 5 negative samples. All of the concentrations of CEA in samples had been standardized by a Chemiluminescent Immunoassay System (LIASON, DiaSorin S.p.A, Italy) with standard CEA kit (DiaSorin S.p.A, Italy). All positive samples and negative samples were diluted to proper solution and put into HRP-Apt/GO solution, respectively and standing for 4 h. The final solutions were analyzed by CE-CL, successively. 7
3. Results and discussion 3.1. CE-CL Analysis of HRP, HRP-Apt, and HRP-Apt-CEA. We studied the UV-vis absorption of CEA aptamer before and after conjugating with HRP (Fig. S1). It was found all of HRP had conjugated with aptamer in the present of coupling reagent of EDC and Sulfo-NHS. After obtaining the coupling product, CE-CL analyses for each sample were carried out on a home-built system (Fig. 1). Compared with blank sample (Fig. 1, a), a CL electrophoresis peak at about 496.2 ± 5.1 s was observed when 0.1 mg/mL HRP was introduced (Fig. 1, b); the CL electrophoresis peak was hardly shifted after aptamer mixed with HRP (Fig. 1, c), suggesting that aptamer wasn’t attached to HRP if they were just mixed simply. This result was also consistent with analysis of UV-vis spectrum (Fig. S1 (B)). When HRP-Apt was introduced to capillary, it was found that the CL electrophoresis peak catalyzed by HRP-Apt (the concentration of HRP was also 0.1 mg/mL) was shifted to 461 ± 6.7 s (Fig. 1, d), probably because of the decrease of the charge-mass ratio; after adding 0.514 μg/mL CEA, two CL electrophoresis peaks with serious overlap were observed (Fig. 1, e), one was consistent with HRP-Apt (Fig. 1, d), and the other was shifted to 445 ± 8.3 s, in advance about 16 s. It was conjectured that after combination with CEA, the aptamer liked a chain bridge that bound HRP and CEA together, forming HRP-Apt-CEA complex, a double-protein complex structure, and the new peak should be related to the HRP-Apt-CEA complex. These two electrophoresis peaks were considerable overlap, indicating the poor separation efficiency for HRP-Apt and HRP-Apt-CEA complex in this condition. Therefore, some experiments had been done to improve the separation efficiency of HRP-Apt and HRP-Apt-CEA complex by optimizing the experimental conditions.
8
3.2. CE-CL Analysis of CEA by Optimizing the Experimental Conditions CL is a chemical kinetic process with the signal occurring, rising and decaying as the reaction starts, proceeds and finishes. Therefore, it presents a serious overlap and interference of optical signal due to relaxation phenomenon, especially when it is coupled with separation technologies, such as CE (Dadoo et al., 1994). We designed the end-column detection of CE-CL system and decrease the influence of relaxation phenomenon by continuously injecting CL buffer. The injection rate of CL buffer was studied (Fig. S2). It was found CL detection was achieved completely and relaxation phenomenon was effective decreased by continuously injecting fresh CL buffer, when the injection rate was 0.01 mL/min. Therefore, 0.01 mL/min was applied to subsequent experiments. In generally, the pH of electrophoretic buffer and the separation voltage have influence on the separation efficiency in CE experiments. Firstly, the influence of pH of the electrophoretic buffer was studied (Fig. S3). It could be seen the overlap of CL electrophoresis peaks was still serious, although the separation efficiency for HRP-Apt and HRP-Apt-CEA complex was improved in a certain extent when pH was 8.4 (Fig. S3, b). Secondly, the separation efficiency for HRP-Apt and HRP-Apt-CEA had been small improved by increasing the separation voltage other than migration time decreasing (Fig. S4). It might be due to the similar property of HRP-Apt and HRP-Apt-CEA complex, and this is a common issue in protein separation by CE (Wang et al., 2009; Shen et al., 2010; Nilsson et al., 2011). For example, Zhang’s group studied the detection of EPO by CE-CL and found a serious overlap electrophoresis peaks coursed by freely HRP-EPO and immune-complex (Wang et al., 2009). Therefore, 8.4 for the pH of electrophoretic buffer and 17 kV for separation voltage were used in subsequent experiments. 9
3.3. Design Strategy for Detection of CEA Based on Aptamer/GO by CE-CL According to the above results, it could be seen that HRP-Apt and HRP-Apt-CEA complex could not be completely separated, even though optimized the experimental conditions by CE-CL. In order to realize the separation of HRP-Apt and HRP-Apt-CEA complex, a new strategy was proposed for analysis of content of CEA based on the couple of GO and HRP-Apt by CE-CL detection (Fig. 2). In the absence of CEA, HRP-Apt was stacked on GO surface and formed HRP-Apt/GO, and the CL catalyzed by HRP-Apt would be not observed in this situation because GO could quench the CL by CRET. When CEA existed, the specific combination of HRP-Apt and CEA formed HRP-Apt-CEA complex, it couldn’t bind to GO stably anymore, and dissociated from GO. The CL catalyzed by HRP-Apt-CEA complex could be detected due to CRET being removed; after separating by CE, the interference resulted from GO to CL was also solved. Therefore, according to the light signal ‘off-on’ procedure, CEA could be estimated by the CL intensity catalyzed by HPR-Apt-CEA complex, and the separation of GO and HRP-Apt-CEA complex by CE achieved great improvement of CL detection sensitivity.
3.4. CE-CL Analysis of Combination of GO and HRP-Apt Reproducibility On the basis of the above optimized conditions, the feasibility of the strategy was further verified. According to the above results, the CL electrophoresis peaks catalyzed by HRP-Apt and HRP-Apt-CEA complex were 506 ± 6.5 s᧨468 ± 7.4 s, respectively, in the optimized conditions (Fig. 3, a), but the overlap of CL electrophoresis peaks was still serious. However, when GO was added, it could be seen the CL electrophoresis peak catalyzed by HRP-Apt-CEA complex was hardly decreased and shifted (466 ± 6.7 s), and the CL electrophoresis peak catalyzed by HRP-Apt (497 ± 10
7.1 s) decreased obviously (Fig. 3, b). It was indicated that the CL catalyzed by HRP-Apt was quenched by effective CRET due to HRP-Apt attaching on GO, the CL catalyzed by HRP-Apt-CEA complex didn’t decrease because HRP-Apt-CEA complex couldn’t bind to GO stably and CRET did not occur. When the concentration of GO increased to 20 μg/mL, it could be seen the CL electrophoresis peak catalyzed by HRP-Apt disappeared and the CL electrophoresis peak catalyzed by HRP-Apt-CEA was still not changed (Fig. 3, c), suggesting HRP-Apt completely stacked on GO. The same phenomenon was observed by increasing the concentration of GO sequentially (Fig. 3, d and e). It was worth noting that there was a weak CL peak at 310 ± 5.9 s after addition of GO, and it might be caused by HRP-Apt/GO, which meant the CL catalyzed by HRP-Apt hadn’t been quenched completely by GO. But, it was certain that the CL catalyzed by HRP-Apt-CEA complex would not be interfered by this residual CL due to good separation effect between HRP-Apt/GO and HRP-Apt-CEA complex by CE. In a word, the feasibility of the strategy was verified. In order to further optimize the addition of GO, the CL intensity catalyzed by HRP-Apt was studied by adding different concentration of GO in the absence of CEA (Fig. S5). It could be seen that the CL electrophoresis peaks catalyzed by HRP-Apt disappeared when the concentration of GO increased to 25 μg/mL, indicating the CL electrophoresis peak catalyzed by HRP-Apt was totally quenched by GO. Therefore, 25 μg/mL of GO was applied in the subsequent experiments.
3.5. CE-CL Analysis of Combination of CEA and HRP-Apt/GO Fig. 4A showed the electropherograms for the CL intensity catalyzed by HRP-Apt-CEA complex with different concentration of CEA. It was found the CL intensity catalyzed by HRP-Apt-CEA complex increased rapidly with the increase of concentration of CEA. Fig. 4B exhibited the 11
relationship of the peak area of the CL electrophoresis peak catalyzed by HRP-Apt-CEA and the concentration of CEA. It could be seen that the CL intensity increased rapidly below 6.54 ng/mL, and slowed down when the concentration of CEA was above 6.54 ng/mL. The relationship was linear in the range from 0.0654 ~ 6.54 ng/mL (inset in Fig. 4B), and the concentration of CEA could be estimated by the equation, Log [I] = 1.633 + 0.830 × Log [c] (R2 = 0.992), where [I] was the CL intensity of the CL electrophoresis peak area after integration, and [c] was the concentration of CEA. The limit of detection was 4.8 pg/mL based on a signal-to-noise ratio of 3. Table S1 listed some common methods for the detection of the content of CEA. Relative to the limit of detection, the proposed method had a lower limit of detection than most of others.
3.6. Specificity. Three kinds of proteins, IgG, HSA, and thrombin were selected as interferent under optimized conditions. The four proteins were incubated with the HRP-Apt/GO solution for 4 h, respectively, and then the four solutions were measured by CE-CL. It was found the interferential proteins did not restore CL intensity as much as CEA did, and CEA could be clearly distinguished from other proteins though the concentration of CEA was three orders of magnitude lower than those interferential proteins (Fig. 5). Such results indicated that the proposed method had a high specificity for CEA.
3.7. Analysis of Content of CEA in Patient Serum. The proposed method was applied to analyze the content of CEA in patient’s serum. The serum samples of patient were come from Wuhan Commercials Staff Hospital, which included five positive samples and five negative samples. The concentrations of CEA in samples had been standardized by 12
LIAISON chemiluminescent immunoassay system with standard CEA kit. Table 1 shows the detection results of the 10 samples by LIAISON chemiluminescent immunoassay system and the proposed method. The samples in the Table 1 had to be diluted into the linear range before analyzing. After CE-CL experiments, the diluted concentration of samples could be calculated by the linear equation. And then, multiplied by the diluted factor, the actual concentration of samples could be estimated. It was found the results detected by our method were very close to the calibration values by LIAISON chemiluminescence immunoassay system, indicating that the proposed method was good at quantitative analysis of content of CEA in practical sample.
4. Conclusions In conclusion, a new strategy is proposed for determination of CEA based on Apt/GO by couple of CE-CL detection system. The interference issue is solved well by mixing GO firstly with HRP-Apt due to CL quenching effect of GO, which blocked the free HRP-Apt signal, and the interference resulted from GO to CL is also solved by CE, then the sensitivity and accuracy could be greatly improved and achieved ng/mL accurate analysis of content of CEA in serums of patients. This proposed method with high specificity offered a new way for separation and determination of biomolecule and had good potential in application of biochemistry and bioanalysis.
Acknowledgements This work was supported by the National Key Technology R&D Program of China (2012BAI23B02), National Natural Science Foundation of China (Grant No. 81271616, 81471697), the Foundation for Innovative Research Groups of the NNSFC (Grant No. 61121004), the Fundamental Research Funds 13
for the Central Universities (Hust, 2014YGYL022, 2013TS085, 2014QN116), Scientific Research Fund of Union Hospital. We also thank the facility support of the Center for Nanoscale Characterization and Devices, Wuhan National Laboratory for Optoelectronics (WNLO) of HUST and Analytical and Testing Center (HUST) for the help of measurement.
References Bi, S., Zhao, T.T., Luo, B.Y., 2012. Chem. Comm., 48, 106-108. Cao, X., Wang, N., Jia, S., Guo, L., Li, K., 2013. Biosens. Bioelectron., 39, 226-230. Chan, C.H.F., Stanners, C.P., 2007. Curr. Oncol., 14, 70-73. Chen, L., Zhang, X.W., Zhou, G.H., Xiang, X., Li, X.H., Zheng, Z.H., He, Z.K., Wang, H. Z., 2012. Anal. Chem., 84, 3200-3207. Cheng, Y.F., Yuan, R., Chai, Y.Q., Niu, H., Cao, Y.L., Liu, H.J., Yuan, Y.L., 2012. Clin. Chim. Acta, 745, 137-142. Dungchai, W., Siangproh, W., Lin, J.M., Chailapakul, O., Lin, S., Ying, X.T., 2007. Anal. Bioanal. Chem., 387, 1965-1971. Fu, Z.F.,Yang, Z.J., Tang, J.H., Liu, H., Yan, F., Ju, H.X., 2007. Anal. Chem., 79, 7376-7382. Gold, P., Freedman, S.O., 1965. J. Exp. Med., 122, 467-481. Guo, X.D., Huang, Q., Lin, Y.N., 2013. Anal. Lett., 46, 2040-2047. Iliuk, A.B., Hu, L.H., Tao, W.A., 2011. Anal. Chem., 83, 4440-4452. Lara, F.J., García-Campańa, A.M., Velasco, A. I., 2010. Electrophoresis, 31, 1998-2027. Lin, J.H., Yan, F., Ju, H.X., 2004. Clin. Chim. Acta, 341, 109-115. Lin, Z.Y., Zhang, G.Y., Yang, W.Q., Qiu, B., Chen, G.N., 2012. Chem. Comm., 48, 9918-9920. Liu, Y.X., Dong, X.C., Chen, P., 2012 Chem. Soc. Rev., 41, 2283-2307. Lu, C.H., Li, J., Lin, M.H., Wang, Y.W., Yang, H.H., Chen, X., Chen, G.N., 2010 Angew. Chem., Int. Ed., 49, 8454-8457. Luo, M., Chen, X., Zhou, G.H., Xiang, X., Chen, L., Ji, X.H., He, Z.K., 2012. Chem. Comm., 48, 1126-1128. Nilsson, C., Birnbaum, S., Nilsson, S., 2011. Electrophoresis, 32, 1141-1147. Pei, X.P., Chen, B.A., Li, L., Gao, F., Jiang, Z., 2010. Analyst, 135, 177-181. Shen, R., Guo, L., Zhang, Z.Y., Meng, Q.W., Xie, J.W., 2010. J. Chromatogr. A, 1217, 5635-5641. Song, Z.J., Yuan, R., Chai, Y.Q., Zhuo, Y., Jiang, W., Su, H.L., Che, X., Li, J.J., 2010. Chem. Commun., 46, 6750-6752. Su, B.B., Shi, H., Wan, J., 2012. World J. Gastroentero., 18, 2121-2126. Tabar, G.R.H., Smith, L.C., 2010. World Appl. Sci. J., 8, 16-21. Tang, D.P., Yuan, R., Chai, Y.Q., 2006. J. Phys. Chem. B, 110, 11640-11646. Tang, D.P., Yuan, R., Chai, Y. Q., 2008. Anal. Chem., 80, 1582-1588. Wang, L., Zhu, J.B., Han, L., Jin, L.H., Zhu, C.Z., Wang, E.K., Dong, S.J., 2012. ACS Nano, 6, 6659-6666. 14
Wang, W.J., Zhang, S.H., Liu, C.H., Lu, L.Z., Wang, S.D., Zhang, X.R., 2009. Electrophoresis, 30, 3092-3098. Wang, Y., Li, Z.H., Hu, D.H., Lin, C.T., Li, J.H., Lin, Y., 2010. J. Am. Chem. Soc., 132, 9274-9276. Yan, M., Ge, S.G., Gao, W.Q., Chu, C.C., Yu, J.H., Song, X.R., 2012. Analyst, 137, 2834-2839. Zhang, B., Zhang, X., Yan, H.H., Xu, S.J., Tang, D.H., Fu, W.L., 2007. Biosens. Bioelectron., 23, 19-25. Zhang, H.Q., Li, X.F., Le, X.C., 2008. J. Am. Chem. Soc., 130, 34-35. Zhou, Z.M., Zhou, J., Chen, J., Yu, R.N., Zhang, M.Z., Song, J.T., Zhao, Y.D., 2014. Biosens. Bioelectron. 59, 397-403.
Figture capitons: Fig. 1. Electropherograms for blank (a); HRP (b); HRP mixed with aptamer simply (c); HRP conjugated with aptamer by the couple reagents (d); HRP-Apt incubated with 0.514 μg/mL CEA (e). The separation voltage was 17 kV, the pH of electrophoresis buffer was 8.7, and the injection rate of buffer was 0.01 mL/min. The concentrations of HRP and HRA-Apt were 0.1 mg/mL, and 10 mM, 10mM, 20 mM for luminol, p-Ip, H2O2, respectively. Each experiment had been done for 3 times. Fig. 2. Schematic depiction of detection of CEA based on aptamer/GO by CE-CL. Fig. 3. Electropherograms for HRP-Apt-CEA incubated with different concentration of GO. a) 0᧨b) 10, c) 20, d) 25, e) 30 μg/mL. The pH of electrophoretic buffer was 8.4. The other conditions were the same as Fig. 1 and each experiment had been done for 3 times. Fig. 4. Electropherograms for the CL of HRP-Apt/GO restoring with different concentration of CEA (A) and the relationship of CL intensity and the concentration of CEA (B). The concentration of CEA was 0, 0.0654᧨0.129, 0.257, 0.645, 1.29, 2.57, 6.45, 12.9, 25.7 and 129 ng/mL from a to k. The inset of A showed the amplifying diagram of the CL electrophoresis peaks of curve a and b; the inset of B showed the logarithm linear relationship between the CL intensity and the concentration of CEA. Error bar indicated the standard deviation (n=3). The concentration of GO was 25 μg/mL and the pH of electrophoretic buffer was 8.4. The other conditions were the same as Fig. 1 and each experiment had been done for 3 times. Error bar indicated the standard deviation.
15
Fig. 5. Specificity of the proposed method. The concentration of CEA was 2.57 ng/mL, and IgG, HSA, thrombin were 10 μg/mL, 10 mg/mL and 10 μg/mL, respectively. The other conditions were the same as Fig. 4 and the error bar indicated the standard deviation.
Table 1. The Detecting Results of CEA in Patient’s Serum a Number Positive Sample b
Results c
Number Negative sample b
Results c
1
130.8
128.3 ± 4.1
6
1.60
1.61 ± 0.04
2
122.1
119.6 ± 6.4
7
1.37
1.33 ± 0.09
3
17.1
17.3 ± 0.66
8
1.79
1.74 ± 0.07
4
339.4
340.3± 8.1
9
3.07
3.13 ± 0.07
5
74.2
74.3 ± 3.9
10
1.59
1.59 ± 0.05
a
The concentration unit of Table 1 was ng/mL. The calibration value for the samples by LIAISON chemiluminescence immunoassay system. c The results detected by our method, and each sample had been tested for three times. b
Highlights z z z z
Carcino-embryonic antigen is detected by capillary electrophoresis-chemiluminescence. The interference issue is solved well by mixing GO firstly with HRP-Apt. Good sensitivity and specificity has been obtained. The proposed method is good at quantitative analysis of CEA in practical sample.
16
Figure 1
Figure 2
Figure 3
Figure 4-A
Figure 4-B
Figure 5
