Biosensors and Bioelectronics 54 (2014) 72–77
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A simple and sensitive immunoassay for the determination of human chorionic gonadotropin by graphene-based chemiluminescence resonance energy transfer Jiuqian Lei a,b, Tao Jing a,b,n, Tingting Zhou a,b, Yusun Zhou a,b, Wei Wu a,b, Surong Mei a,b, Yikai Zhou a,b a Key Laboratory of Environment and Health, Ministry of Education & Ministry of Environmental Protection, Tongji Medical College, Huazhong University of Science and Technology, #13 Hangkong Road, Wuhan, Hubei 430030, China b State Key Laboratory of Environmental Health (Incubating), Tongji Medical College, Huazhong University of Science and Technology, #13 Hangkong Road, Wuhan, Hubei 430030, China
art ic l e i nf o
a b s t r a c t
Article history: Received 18 July 2013 Received in revised form 16 October 2013 Accepted 21 October 2013 Available online 31 October 2013
In this study, we report a strategy of chemiluminescence resonance energy transfer (CRET) using graphene as an efficient long-range energy acceptor. Magnetic nanoparticles were also used in CRET for simple magnetic separation and immobilization of horseradish peroxidase (HRP)-labeled anti-HCG antibody. In the design of CRET system, the sandwich-type immunocomplex was formed between human chorionic gonadotropin (HCG, antigen) and two different antibodies bridged the magnetic nanoparticles and graphene (acceptors), which led to the occurrence of CRET from chemiluminescence light source to graphene. After optimizing the experimental conditions, the quenching of chemiluminescence signal depended linearly on the concentration of HCG in the range of 0.1 mIU mL 1–10 mIU mL 1 and the detection limit was 0.06 mIU mL 1. The proposed method was successfully applied for the determination of HCG levels in saliva and serum samples, and the results were in good agreement with the plate ELISA with colorimetric detection. It could also be developed for detection of other antigen–antibody immune complexes by using the corresponding antigens and respective antibodies. & 2013 Elsevier B.V. All rights reserved.
Keywords: Chemiluminescence resonance energy transfer Human chorionic gonadotropin Graphene Magnetic separation Immunoassay
1. Introduction Determination of tumor markers is a powerful technology to allow accurate diagnosis of cancer diseases and monitor the effectiveness of drug and surgery treatments (Qian et al., 2013). Nowadays, saliva has been considered as an alternative matrix for assessing a variety of tumor markers. Compared with blood collection, saliva collection is non-invasive and less stressful without the limitation of time and frequency (Matias et al., 2012; Lamy and Mau, 2012). However, due to the trace levels of tumor markers in saliva samples, it has become a great challenge to seek a simple, sensitive and selective method for determination of trace tumor markers. Immunoassays have been considered as an effective technology to determine trace tumor markers (Wani and Darwish, 2012). Because of highly selectivity of antibody–antigen immunocomplex, sensitive signal response has become a crucial step in the determination of
n Corresponding author at: School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, #13 Hangkong Road, Wuhan, Hubei 430030, China Tel.: þ 86 278 355 2611; fax: þ 86 278 365 7765. E-mail address: [email protected] (T. Jing).
0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2013.10.033
trace target molecules using immunoassay. Resonance energy transfer is known as extremely sensitive to separate distance between donor and acceptor, in which a luminescent donor transfers energy to a fluorescent or nonfluorescent acceptor via nonradiative dipole– dipole interaction, when brought in proximity by a bioaffinity reaction (Qin et al., 2012). Nowadays, fluorescence resonance energy transfer (FRET) is emerging as a useful tool in many fields, such as activation of receptors (Dereli-Korkut et al., 2013), cellular imaging (Lin and Hoppe, 2013), drug release (Lai et al., 2013) and trace detection (Liu et al., 2013). In contrast to FRET, chemiluminescence resonance energy transfer (CRET) is occurred by the oxidation of a luminescent substrate, which is beneficial for reducing the nonspecific signals caused by external light excitation. Therefore, CRET has become an attractive light-measuring scheme in bioassays. More importantly, CRET could be used for homogeneous analysis of complex samples without separation (Huang and Ren, 2012). Unfortunately, little study has been reported so far on CRET. One of the major problems is to identify an effective chemiluminescence (CL) donor or reaction. The most widely used CL reaction in CRET is the luminol-H2O2 system, catalyzed by horseradish peroxidase (HRP) (Chen and Li, 2013; Dong et al., 2013). Recently, a number of nanomaterials have been explored for the
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
enhancement of CL efficiency. Magnetic nanoparticles (MNPs) have many superior characteristics such as small size, high surface-to-volume ratio, fast and effective binding to biomolecules and high magnetic susceptibility. With regards to the application of magnetic nanoparticles in CRET detection, there are two main aspects to consider: (1) The magnetic nanoparticles can effectively conjugate HRP-labeled antibody on their surface and then increase the amounts of HRP in CL reaction; (2) The superparamagnetic properties of magnetic nanoparticles provide a simple magnetic separation to attain interference-free measurement for real detection. Another problem of CRET is poor energy-transfer efficiency and limited number of energy acceptors. Currently, small-molecule fluorophores, widely used as energy acceptors, have small Stokes shifts (Bi et al., 2009; Qin et al., 2012), which result in poor spectral separation of the acceptor emission from the donor emission. With the development of nanotechnology, carbon materials have attracted much attention in various fields (Morales-Narvaez and Merkoci, 2012; Chou et al., 2012; Dreyer et al., 2010). Especially, graphene had large Stokes shifts and long-range energy transfer efficiency (Lee et al., 2012), which is predicted to be an excellent acceptor in the CRET-based studies. Human chorionic gonadotropin (HCG), secreted by the trophoblastic cells of the placenta chorionic vesicle, is a kind of peptide hormone with a molecular mass of about 37 kDa (Yang et al., 2010). HCG level in saliva and serum samples has been used as an important marker to diagnose germ cell tumors, trophoblastic cancer, choriocarcinoma and many diseases of pregnancy (Vartiainen et al., 2002; Birken et al., 2001). Therefore, determination of HCG in saliva and serum samples plays an important role in clinical diagnosis, treatment and prognosis. Herein, we report a simple and sensitive immunoassay for the determination of HCG by using graphene-based CRET. In this protocol, HRP-labeled anti-HCG antibody–conjugated MNPs were incubated with a limited amount of HCG and anti-HCG antibody– conjugated graphene to form a sandwich-type immunocomplex. The close proximity led to the CRET phenomenon between CL light source and graphene as an efficient long-range energy acceptor. This immunoassay could be used for the determination of HCG in saliva and serum samples, which provided a promising potential in clinical diagnosis, treatment and prognosis.
73
were measured with a multifunction microplate reader (TECAN Infinite 200, San Jose, CA). 2.2. Preparation of anti-HCG antibody–conjugated graphene Anti-HCG antibody was immobilized on the surface of graphene sheet according to the literature (Lee et al., 2012) with a slight modification. Briefly, graphene (0.1 mg mL 1) was mixed with a 200 μL of aqueous solution containing 2 mg mL 1 EDC and 4 mg mL 1 NHS. The mixture was then incubated for 20 min at 25 1C in a shaking water bath. Anti-HCG antibody (1 mg mL 1) diluted with phosphate buffer (10 mmol L 1, pH 7.5) was added and the resulting suspensions were shaken at room temperature for 6 h. After washing with distilled water, remaining NHS-active sites of grapheme sheet were blocked with 2.0% BSA in a phosphate buffer (10 mmol L 1) for 3 h. The mixture was centrifuged at 15,000 rpm for 30 min at 4 1C to remove any unbound biomolecules. 2.3. Preparation of HRP-labeled antibody–conjugated MNPs HRP-labeled antibody–conjugated MNPs were synthesized based on the literature (Lai et al., 2010). Carboxyl-modified magnetic nanoparticles (4 mg) were washed three times with MES buffer and then suspended to a final volume of 250 μL in the same buffer solution. The HRP-labeled anti-HCG antibody– conjugated MNPs were prepared by adding 0.3 mL HRP-labeled anti-HCG antibody (1 μg mL 1), 0.5 mL EDC (2 mg mL 1) into Fe3O4 nanoparticles suspension, followed by stirring at room temperature for 12 h. Subsequently, the resulting mixture were separated magnetically and then resuspended in 10 mmol L 1 PBS solution (pH ¼ 7.5, containing 2.0% BSA). 2.4. Characterization The surface morphology of the immune complex was evaluated by atomic force microscopy analyses (DI NanoScope IV AFM, Veeco Co. Ltd., USA). Electrochemical measurements were performed on a CH 660C Electrochemical Workstation (Shanghai Chenhua Co. Ltd., China) with a conventional three-electrode system. The working electrode is a glassy carbon electrode, the reference electrode is a saturated calomel electrode (SCE), and the counter electrode is a platinum wire.
2. Materials and methods 2.5. Determination of HCG by using graphene-based CRET 2.1. Reagents and apparatus HRP-labeled anti-HCG antibody, anti-HCG antibody and HCG were obtained from Shanghai Linc-Bio Science Co. LTD. (Shanghai, China). Graphene sheet was obtained from JCNANO (Nanjing, China). Carboxyl-modified Fe3O4 nanoparticles and magnetic separation rack were purchased from BaseLine Chrom Tech Research Centre (Tianjin, China). (N-Morpholino)-ethanesulfonic acid monohydrate (MES), bovine serum albumin (BSA), 1-ethyl-3-(3′-dimethylaminopropyl)-carbodiimide hydrochloride (EDC), luminol and N-hydroxysuccinimide (NHS) were purchased from Sigma-Aldrich (St. Louis, MO). Hydrogen peroxide (H2O2), dopamine (DA), ascorbic acid (AA), uric acid (UA), glucose and lysozyme (Lys) were provided by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All other chemicals used in this work were of analytical grade. Doubly distilled water was used throughout the experiments. Human serum and saliva samples were kindly provided by a team of volunteers and then stored at 20 1C until analysis. ELISA Kit (HCG) was purchased from Diagnostic Products (DPC, USA). CL spectra
Schematic illustration of a graphene-based CRET platform is shown in Fig. 1. A 100 μL of HRP-labeled antibody–conjugated MNPs suspension (4 mg mL 1) was firstly mixed with 100 μL of human HCG standard solution or sample solution in 96-well plate and then incubated for 30 min at 37 1C. The mixture was separated by magnetic separation rack and then incubated with 200 μL of anti-HCG antibody–conjugated graphene suspension (0.1 mg mL 1) for 2.5 h at 37 1C to form a sandwich-type immunocomplex. After magnetic separation and gentle washing with distilled water, the resultant mixture was resuspended in 120 μL of PBS solution (10 mmol L 1, pH¼ 9.0). Subsequently, 50 μL H2O2 (3.5 mmol L 1) was added into the mixture and then 30 μL luminol solution (10 mmol L 1) was introduced by using an automatic injection of TECAN Infinite 200 (Tecan, San Jose, CA). The decreased chemiluminescence intensity caused by CRET was represented as ΔI¼I1 I0. Here, I1 and I0 were the chemiluminescence intensities of the system without and with adding the antibody–conjugated graphene, respectively. The chemiluminescence intensity was recorded after 30 s by using TECAN Infinite 200 microplate reader (Tecan, San
74
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
Fig. 1. The schematic illustration of the determination of HCG by using graphene-based CRET immunoassay.
Jose, CA). The standard curve method was used in quantification of trace amount of HCG in saliva and serum samples. 2.6. Verification test Ten serum samples from a team of volunteers were analyzed by using the proposed method and the plate ELISA with colorimetric detection to verify the practical perspective of graphene-based CRET detection. According to the guideline of ELISA Kit, the procedure was as following: 50 μL of sample was incubated with anti-HCG antibody adsorbed on the ELISA plate and then HRP-conjugate reagent was added to from a sandwich-type immunocomplex. After washing five times, substrates were orderly added into the micro-well. Subsequently, the mixture was incubated for 15 min and then the reaction was terminated by the addition of a sulfuric acid solution. The optical density (OD) at 450 nm was measured using a microplate reader (TECAN Infinite 200; San Jose, CA).
3. Results and discussion 3.1. Characterization of the antibody–conjugated functional materials The morphologies and microstructures of anti-HCG antibody– conjugated graphene (A), HRP-labeled antibody–conjugated MNPs (B) and the sandwich-type immunocomplex (C) were studied by means of AFM (Fig. 2). It is shown that antibodies were immobilized on the surface of graphene sheets (about 6 mg g 1), which induced the height increase of the complex (antibody–conjugated graphene) to about 2 nm. Then we found that the HRP-labeled antibody– conjugated MNPs were uniform cobblestone-like shape with particle diameter of 300–400 nm. Finally, antibody–conjugated graphene, antigen (HCG), and HRP-labeled antibody–conjugated MNPs were incubated in a phosphate buffer (10 mmol L 1). After magnetic separation and gentle washing with distilled water, a large number of magnetic particles were bound on the surface of graphene sheet by the antigen–antibody interactions, which would lead to the CRET phenomenon. Electric impedance spectroscopy (EIS) is a valuable and convenient tool to monitor the interface properties of surface-modified electrodes. Fig. S1 (Supporting Information) represents the EIS of variously modified electrodes after each step in 10 mmol L 1 PBS
solution (pH 7.4) containing 5.0 mmol L 1 Fe(CN)64 /3 and 0.1 mol L 1 KCl. It was observed that MNPs and graphene modified electrode exhibited very small semicircles. Compare with these electrodes, the electron-transfer resistance of antibody–conjugated MNPs (graphene) modified electrode increased, which suggested that antibody had been successfully immobilized on the surface of functional materials. Furthermore, because of poor electrical conductors of biological molecules, the formation of sandwich-type immunocomplex could result in the significant increasing of electron-transfer resistance. It was shown that the sandwich-type immunocomplex could be formed and the close proximity led to the CRET phenomenon between CL light source and grapheme. 3.2. Optimization of the parameters affecting CRET signals In a sandwich immunoreaction, selective recognition for target compound was resulted from the amount of HRP-labeled anti-HCG antibody immobilized on the MNPs. Therefore, different amounts of HRP-labeled anti-HCG antibody were used to prepare antibody– conjugated MNPs. Subsequently, this mixture was allowed to react with a HCG standard solution (100 IU mL 1) and residual HCG in supernatant solution after magnetic separation was determined by the plate ELISA with colorimetric detection. The ΔA value could be used to optimize the amount of HRP-labeled anti-HCG antibody, which was represented as ΔA¼ A0 A1. Here, A0 was the absorbance of HCG standard solution (100 IU mL 1) and A1 was the absorbance of residual HCG in supernatant solution. It was demonstrated that the amount of HRP-labeled anti-HCG antibody (1.0 μg mL 1) immobilized on MNPs showed a maximum at 0.3 mL, with a significant decrease at lower and higher volume of antibody (Fig. S2A, Supporting Information). Furthermore, anti-HCG antibody modified on the surface of graphene was used to form a sandwich-type immunocomplex and then cause CRET phenomenon. The results indicated that the amount of immunocomplex increased with the increasing concentration of anti-HCG antibody from 0.2 μg mL 1 to 1.0 μg mL 1 and further increasing the amount of antibody did not produce any significant improvement (Fig. S2B, Supporting Information). Thus, the optimized concentration of anti-HCG antibody (200 μL) was fixed at 1.0 μg mL 1. The reaction of the sandwich-type immunocomplex was greatly influenced by the incubation time. Therefore, 10 mIU mL 1 of HCG standard solution was prepared and then HRP-labeled antibody– conjugated MNPs and antibody–conjugated graphene were added to
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
75
Fig. 2. AFM images of (A) anti-HCG antibody–conjugated graphene, (B) HRP-labeled antibody–conjugated MNPs and (C) the sandwich-type immunocomplex.
the sample solution, respectively. In this case, the incubation time of the one-step sandwich reaction (antigen (HCG)-antibody–conjugated MNPs) varied from 0 to 60 min. It was found that the amount of adsorbed HCG increased sharply with the increasing of incubation time and then almost trended to the maximum at 30 min, which indicated that the immuno-binding reached the saturation (Fig. S2C, Supporting Information). For the incubation time between the antibody–conjugated graphene and antigen (HCG)-antibody–conjugated MNPs, the ΔI value increased and leveled off at 2.5 h with the increasing of incubation time (Fig. S2D, Supporting Information). Hence, 2.5 h of incubation time for the second immunoreaction was selected for the following experiments. Finally, chemiluminescence played an important role in graphene-based CRET. Thus, the parameters affecting CL signals were further optimized, such as the pH value of PBS solution, the volume of luminol and the concentration of H2O2. The results indicated that 10 mmol L 1 PBS solution (pH 9.0) was selected as a CL buffer solution, which could ensure the stability of immune complexes and maximize the sensitivity of CL detection. As the chemiluminescent substrates, luminol and H2O2 strongly influenced the luminescence intensity in CRET. It was shown that the ΔI values reached their maximum value when the volume of luminol (10 mmol L 1) was 30 μL. For the concentration of H2O2, the ΔI values were increased with the increasing of H2O2 and the CL signal reached the highest at 3.5 mmol L 1H2O2 (Fig. S3, Supporting Information).
spectral separation of the acceptor emission from the donor emission and then low energy-transfer efficiency. In order to investigate the performance of graphene as an energy acceptor in CRET, FITC-labeled anti-HCG antibody (obtained from Shanghai Linc-Bio Science Co. LTD.) or label-free anti-HCG antibody– conjugated graphene were incubated with antigen (HCG) standard solution and HRP-labeled anti-HCG antibody–conjugated MNPs, respectively (Fig. 3). The amount of FITC-labeled antibody was optimized and the results indicated that the ΔI values reached their maximum value when the concentration of FITC-labeled antibody (100 μL) was 7 μg mL 1. It was shown that a gradual quenching of CL intensity was observed with the increasing amount of antibody–conjugated graphene, which demonstrated that graphene could be used as an efficient long-range energy acceptor in CRET. For the immunoassay using FITC as an energy acceptor, the linear relationship between HCG concentration and CL quenching was changed from 0.5 mIU mL 1 to 5 mIU mL 1, which was narrower than that of the proposed immunoassay using graphene as an acceptor (0.1 mIU mL 1 to 10 mIU mL 1). The possible reason is that graphene exhibits a high planar surface, which enables an increased number of acceptor molecules. In fact, the fraction of photons that are transferred to the acceptor is generally enhanced by increasing the number of acceptor molecules (Morales-Narvaez and Merkoci, 2012). Thus, graphene can be proposed as a highly efficient long-range acceptor in CRET.
3.3. The effect of graphene as an acceptor in CRET
3.4. Interferences study
CRET involves the non-radiative energy transfer from a chemiluminescent donor to a suitable acceptor molecule. Fluorescein isothiocyanate (FITC) has small Stokes shifts (Huang and Ren, 2012; Morales-Narvaez and Merkoci, 2012), which result in poor
To apply the proposed method to determine HCG in human saliva and serum samples, components of blank sample (health male, without HCG) and effects of several foreign species (the concentration was equal to the upper limit of normal reference
76
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
3.5. Method validation
Fig. 3. The calibration curve obtained by the immunoassay using FITC and graphene as acceptors, respectively. HRP-labeled antibody–conjugated MNPs (4 mg mL 1) were incubated with HCG and antibody–conjugated graphene (Red line) or FITC-labeled antibody (Blue line) to form a sandwich-type immunocomplex, respectively. The decreased CL intensity caused by CRET was represented as ΔI¼ I1 I0. Here, I1 and I0 were the CL intensities of the system without and with adding the antibody–conjugated graphene (or FITC-labeled antibody), respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
We developed a simple and sensitive immunoassay for the determination of HCG using graphene-based CRET. Under the optimal conditions, a series of immunoassays was prepared for the determination of different concentrations of HCG (C, mIU mL 1). The change of CL intensity was proportion to the concentration of HCG in the range of 0.1 mIU mL 1 to 10 mIU mL 1 (equal to 0.02–2 ng mL 1), however, the linear range of ELISA kit (Diagnostic Products, USA) was 0.7–15 mIU mL 1, which showed that the proposed method was more sensitive. The equation of the calibration curve was ΔI¼ 23.8þ244.7C (mIU mL 1), with a correlation coefficient of 0.9991. The limit of detection (LOD) at 3s (s ¼S0/S; S0, standard deviation of blank sample; S, the slope of the calibration curve) was 0.06 mIU mL 1, which was lower than those of the reported electrochemical immunoassay (0.3 mIU mL 1 (Wang et al., 2010), 0.15 mIU mL 1 (Li et al., 2011), 12 mIU mL 1 (Tao et al., 2011)). Precision was evaluated by measuring intra-day and inter-day relative standard deviations (RSDs). It was performed by analyzing spiked serum samples five times in one day or over five days at three different concentrations of 0.5, 1 and 5 mIU mL 1. The intraday RSD ranged from 3.8% to 8.7% and the inter-day RSD ranged from 4.2% to 9.5% in 5 days. It was demonstrated that the proposed method could be used for the determination of HCG in complex samples. In addition, in order to investigate the practical perspective of the proposed method, the amounts of HCG in ten human serum samples were determined by the proposed method (X) and the plate ELISA with colorimetric detection (Y), respectively. Those results were statistically compared by the paired t-test. It was shown that the two methods had a good consistency, with a regression equation of Y ¼1.02 Xþ2.11 and the correlation coefficient was 0.9983 (Fig. 5). Compared with the plate ELISA with colorimetric detection, the proposed immunoassay is simple, because of the homogeneous analysis based on the CRET and magnetic separation. These results illustrated that CRET immunoassay exhibited significant promise as a reliable technique for the detection of HCG in human serum and saliva samples.
3.6. Determination of HCG in saliva and serum samples Fig. 4. Specificity of the graphene-based CRET immunoassay. The concentration of HCG was 1 ng mL 1. Interfering compounds: dopamine (DA, 122 ng mL 1), ascorbic acid (AA, 13.2 ng mL 1), uric acid (UA, 50 μg mL 1), glucose (1.1 mg mL 1), bovine serum albumin (BSA, 55 mg mL 1) and lysozyme (Lys, 30 μg mL 1).
range) were investigated by analyzing a HCG standard solution (5 mIU mL 1, equal to 1 ng mL 1), to which increasing amounts of interfering species had been added. Compared with background buffer solution, there were almost no change in the luminescence intensity for blank serum (RSD ¼4.2%) and blank saliva sample (RSD ¼2.9%), which demonstrated that sample matrix could not affect the determination of HCG or enhance background noise. As shown in Fig. 4, the ΔI values of interfering species (DA, AA, UA, glucose and BSA) were not higher than 10% of the ΔI value of HCG and that value of lysozyme was 32.0%, when the concentrations of interfering species were 122 ng mL 1 (DA), 13.2 ng mL 1 (AA), 50 μg mL 1 (UA), 1.1 mg mL 1 (glucose), 55 mg mL 1 (BSA) and 30 μg mL 1 (Lys), respectively. These phenomena showed that the proposed strategy had good selectivity in determination of HCG and could distinguish HCG from complex sample matrix. In fact, the excellent selectivity for CRET immunoassay is attributed to the high specificity of the antigen–antibody immune response and the magnetic separation for purification.
HCG level in serum samples has been used as an important marker to diagnose the early pregnancy and many diseases related to seminal system. Recently, saliva has been considered as an alternative matrix
Fig. 5. Correlation between the results measured by the proposed method and the plate ELISA with colorimetric detection.
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
77
Table 1 Determination of HCG in human saliva and serum samples by using graphene-based CRET immunoassay. Sample
Found in sample (mIU mL 1)
RSD (%, n¼ 3)
Added (mIU mL 1)
Total found (mIU mL 1)
Recovery (%, n ¼3)
Found in original sample (mIU mL 1)
Saliva 1
0.3
5.9
0.8
3.7
Saliva 3
0.5
4.3
Serum 1
3.9
7.9
Serum 2
5.4
6.3
Serum 3
4.6
5.1
0.9 1.4 3.4 1.2 1.9 4.0 1.1 1.6 3.6 4.5 5.1 6.8 6.2 6.2 8.5 4.9 5.9 7.8
112.5 107.7 103.0 92.3 105.6 105.3 110.0 106.7 102.9 102.3 104.1 98.6 105.1 96.9 101.2 96.1 105.4 102.6
0.3
Saliva 2
0.5 1 3 0.5 1 3 0.5 1 3 0.5 1 3 0.5 1 3 0.5 1 3
for biochemical parameter monitoring. Thus, the proposed graphenebased CRET immunoassay was used for the determination of HCG in serum and saliva samples. The results indicated that spiked recoveries were changed from 92.3% to 112.5% and RSD was lower than 7.9% (Table 1). It was also noted that HCG level could be detected in the saliva samples and the concentrations were found from 0.3 mIU mL 1 to 0.8 mIU mL 1, which was only about 1% of the concentrations of HCG in serum samples. Therefore, it is necessary to develop a simple and sensitive analytical method for determination of trace HCG in saliva samples. Furthermore, further studies are being undertaken by employing the proposed method to investigate the relationship of HCG in saliva and serum samples and develop a novel and simple clinical monitoring technology. In a word, a simple and sensitive graphene-based CRET immunoassay is developed for the determination of trace HCG in saliva and serum samples, which will exhibit widely practical perspective in clinical diagnosis, treatment and prognosis. 4. Conclusions In summary, a simple and sensitive immunoassay was developed for the determination of trace HCG in saliva and serum samples by using graphene-based CRET. After formation of a sandwich-type immunocomplex, the close proximity led to a CRET phenomenon between CL light source and graphene as a highly efficient longrange acceptor. Moreover, the proposed method had good selectivity for the determination of HCG, due to the interaction between antigen and two different antibodies. Under the optimized condition, the linear range was changed from 0.1 mIU mL 1 to 10 mIU mL 1 and the limit of detection was 0.06 mIU mL 1, which was lower than those of the plate ELISA with colorimetric detection and the reported electrochemical immunoassay. The proposed method could be successfully applied for determination of HCG levels in serum and saliva samples, which would exhibit widely practical perspective on the clinical diagnosis, treatment and prognosis. Acknowledgments This work was supported by the National Nature Science Foundation of China (Grant no. 21207044), Hubei Provincial Population and
0.8
0.5
38.9
54.3
45.6
Family Planning Commission (Grant no. JS-2012005), the National Basic research Grant (973) of China (Grant no. 2009CB320301), the Program for New Century Excellent Talents in University (Grant no. NCET-11-0178) and the Fundamental Research Funds for the Central Universities (HUST: 2012QN239).
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.2013.10.033. References Bi, S., Yan, Y.M., Yang, X.Y., Zhang, S.S., 2009. Chem. Eur. J. 15 (18), 4704–4709. Birken, S., Kovalevskaya, G., O’Connor, J., 2001. Arch. Med. Res. 32 (6), 635–643. Chen, C., Li, B.X., 2013. J. Mater. Chem. B 1 (19), 2476–2481. Chou, S.S., De, M., Luo, J.Y., Rotello, V.M., Huang, J.X., Dravid, V.P., 2012. J. Am. Chem. Soc. 134 (40), 16725–16733. Dereli-Korkut, Z., Gandhok, H., Zeng, L.G., Waqas, S., Jiang, X.J., Wang, S.H., 2013. Biotechnol. Bioeng. 110 (5), 1396–1404. Dong, S.C., Liu, F., Lu, C., 2013. Anal. Chem. 85 (6), 3363–3368. Dreyer, D.R., Park, S., Bielawski, C.W., Ruoff, R.S., 2010. Chem. Soc. Rev. 39 (1), 228–240. Huang, X.Y., Ren, J.C., 2012. Trac-Trend. Anal. Chem. 40, 77–89. Lai, J.P., Shah, B.P., Garfunkel, E., Lee, K.B., 2013. ACS Nano 7 (3), 2741–2750. Lai, Y.C., Yin, W.W., Liu, J.T., Xi, R.M., Zhan, J.H., 2010. Nanoscale Res. Lett. 5 (2), 302–307. Lamy, E., Mau, M., 2012. J. Proteom. 75 (14), 4251–4258. Lee, J.S., Joung, H.A., Kim, M.G., Park, C.B., 2012. ACS Nano 6 (4), 2978–2983. Li, R., Wu, D., Li, H., Xu, C.X., Wang, H., Zhao, Y.F., Cai, Y.Y., Wei, Q., Du, B., 2011. Anal. Biochem. 414 (2), 196–201. Lin, J., Hoppe, A.D., 2013. Microsc. Microanal. 19 (2), 350–359. Liu, B.X., Tan, H.L., Chen, Y., 2013. Anal. Chim. Acta 761, 178–185. Matias, I., Gatta-Cherifi, B., Tabarin, A., Clark, S., Leste-Lasserre, T., Marsicano, G., Piazza, P.V., Cota, D., 2012. PLoS One 7 (7), e423997, http://dx.doi.org/10.1371/ journal.pone.0042399. Morales-Narvaez, E., Merkoci, A., 2012. Adv. Mater. 24 (25), 3298–3308. Qian, J., Wang, C.Q., Pan, X.H., Liu, S.Q., 2013. Anal. Chim. Acta 763, 43–49. Qin, G.X., Zhao, S.L., Huang, Y., Jiang, J., Ye, F.G., 2012. Anal. Chem. 84 (6), 2708–2712. Tao, M.L., Li, X.F., Wu, Z.S., Wang, M., Hua, M., Yang, Y.H., 2011. Clin. Chim. Acta 412 (7–8), 550–555. Vartiainen, J., Alfthan, H., Lehtovirta, P., Stenman, U.H., 2002. BJOG-Int. J. Obstet. Gy 109 (5), 589–590. Wang, J.F., Yuan, R., Chai, Y.Q., Cao, S.R., Guan, S., Fu, P., Min, L.G., 2010. Biochem. Eng. J. 51 (3), 95–101. Wani, T.A., Darwish, I.A., 2012. New J. Chem. 36 (4), 1114–1120. Yang, H.C., Yuan, R., Chai, Y.Q., Zhuo, Y., Su, H.L., 2010. J. Chem. Technol. Biot 85 (4), 577–582.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
A simple and sensitive immunoassay for the determination of human chorionic gonadotropin by graphene-based chemiluminescence resonance energy transfer Jiuqian Lei a,b, Tao Jing a,b,n, Tingting Zhou a,b, Yusun Zhou a,b, Wei Wu a,b, Surong Mei a,b, Yikai Zhou a,b a Key Laboratory of Environment and Health, Ministry of Education & Ministry of Environmental Protection, Tongji Medical College, Huazhong University of Science and Technology, #13 Hangkong Road, Wuhan, Hubei 430030, China b State Key Laboratory of Environmental Health (Incubating), Tongji Medical College, Huazhong University of Science and Technology, #13 Hangkong Road, Wuhan, Hubei 430030, China
art ic l e i nf o
a b s t r a c t
Article history: Received 18 July 2013 Received in revised form 16 October 2013 Accepted 21 October 2013 Available online 31 October 2013
In this study, we report a strategy of chemiluminescence resonance energy transfer (CRET) using graphene as an efficient long-range energy acceptor. Magnetic nanoparticles were also used in CRET for simple magnetic separation and immobilization of horseradish peroxidase (HRP)-labeled anti-HCG antibody. In the design of CRET system, the sandwich-type immunocomplex was formed between human chorionic gonadotropin (HCG, antigen) and two different antibodies bridged the magnetic nanoparticles and graphene (acceptors), which led to the occurrence of CRET from chemiluminescence light source to graphene. After optimizing the experimental conditions, the quenching of chemiluminescence signal depended linearly on the concentration of HCG in the range of 0.1 mIU mL 1–10 mIU mL 1 and the detection limit was 0.06 mIU mL 1. The proposed method was successfully applied for the determination of HCG levels in saliva and serum samples, and the results were in good agreement with the plate ELISA with colorimetric detection. It could also be developed for detection of other antigen–antibody immune complexes by using the corresponding antigens and respective antibodies. & 2013 Elsevier B.V. All rights reserved.
Keywords: Chemiluminescence resonance energy transfer Human chorionic gonadotropin Graphene Magnetic separation Immunoassay
1. Introduction Determination of tumor markers is a powerful technology to allow accurate diagnosis of cancer diseases and monitor the effectiveness of drug and surgery treatments (Qian et al., 2013). Nowadays, saliva has been considered as an alternative matrix for assessing a variety of tumor markers. Compared with blood collection, saliva collection is non-invasive and less stressful without the limitation of time and frequency (Matias et al., 2012; Lamy and Mau, 2012). However, due to the trace levels of tumor markers in saliva samples, it has become a great challenge to seek a simple, sensitive and selective method for determination of trace tumor markers. Immunoassays have been considered as an effective technology to determine trace tumor markers (Wani and Darwish, 2012). Because of highly selectivity of antibody–antigen immunocomplex, sensitive signal response has become a crucial step in the determination of
n Corresponding author at: School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, #13 Hangkong Road, Wuhan, Hubei 430030, China Tel.: þ 86 278 355 2611; fax: þ 86 278 365 7765. E-mail address: [email protected] (T. Jing).
0956-5663/$ - see front matter & 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bios.2013.10.033
trace target molecules using immunoassay. Resonance energy transfer is known as extremely sensitive to separate distance between donor and acceptor, in which a luminescent donor transfers energy to a fluorescent or nonfluorescent acceptor via nonradiative dipole– dipole interaction, when brought in proximity by a bioaffinity reaction (Qin et al., 2012). Nowadays, fluorescence resonance energy transfer (FRET) is emerging as a useful tool in many fields, such as activation of receptors (Dereli-Korkut et al., 2013), cellular imaging (Lin and Hoppe, 2013), drug release (Lai et al., 2013) and trace detection (Liu et al., 2013). In contrast to FRET, chemiluminescence resonance energy transfer (CRET) is occurred by the oxidation of a luminescent substrate, which is beneficial for reducing the nonspecific signals caused by external light excitation. Therefore, CRET has become an attractive light-measuring scheme in bioassays. More importantly, CRET could be used for homogeneous analysis of complex samples without separation (Huang and Ren, 2012). Unfortunately, little study has been reported so far on CRET. One of the major problems is to identify an effective chemiluminescence (CL) donor or reaction. The most widely used CL reaction in CRET is the luminol-H2O2 system, catalyzed by horseradish peroxidase (HRP) (Chen and Li, 2013; Dong et al., 2013). Recently, a number of nanomaterials have been explored for the
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
enhancement of CL efficiency. Magnetic nanoparticles (MNPs) have many superior characteristics such as small size, high surface-to-volume ratio, fast and effective binding to biomolecules and high magnetic susceptibility. With regards to the application of magnetic nanoparticles in CRET detection, there are two main aspects to consider: (1) The magnetic nanoparticles can effectively conjugate HRP-labeled antibody on their surface and then increase the amounts of HRP in CL reaction; (2) The superparamagnetic properties of magnetic nanoparticles provide a simple magnetic separation to attain interference-free measurement for real detection. Another problem of CRET is poor energy-transfer efficiency and limited number of energy acceptors. Currently, small-molecule fluorophores, widely used as energy acceptors, have small Stokes shifts (Bi et al., 2009; Qin et al., 2012), which result in poor spectral separation of the acceptor emission from the donor emission. With the development of nanotechnology, carbon materials have attracted much attention in various fields (Morales-Narvaez and Merkoci, 2012; Chou et al., 2012; Dreyer et al., 2010). Especially, graphene had large Stokes shifts and long-range energy transfer efficiency (Lee et al., 2012), which is predicted to be an excellent acceptor in the CRET-based studies. Human chorionic gonadotropin (HCG), secreted by the trophoblastic cells of the placenta chorionic vesicle, is a kind of peptide hormone with a molecular mass of about 37 kDa (Yang et al., 2010). HCG level in saliva and serum samples has been used as an important marker to diagnose germ cell tumors, trophoblastic cancer, choriocarcinoma and many diseases of pregnancy (Vartiainen et al., 2002; Birken et al., 2001). Therefore, determination of HCG in saliva and serum samples plays an important role in clinical diagnosis, treatment and prognosis. Herein, we report a simple and sensitive immunoassay for the determination of HCG by using graphene-based CRET. In this protocol, HRP-labeled anti-HCG antibody–conjugated MNPs were incubated with a limited amount of HCG and anti-HCG antibody– conjugated graphene to form a sandwich-type immunocomplex. The close proximity led to the CRET phenomenon between CL light source and graphene as an efficient long-range energy acceptor. This immunoassay could be used for the determination of HCG in saliva and serum samples, which provided a promising potential in clinical diagnosis, treatment and prognosis.
73
were measured with a multifunction microplate reader (TECAN Infinite 200, San Jose, CA). 2.2. Preparation of anti-HCG antibody–conjugated graphene Anti-HCG antibody was immobilized on the surface of graphene sheet according to the literature (Lee et al., 2012) with a slight modification. Briefly, graphene (0.1 mg mL 1) was mixed with a 200 μL of aqueous solution containing 2 mg mL 1 EDC and 4 mg mL 1 NHS. The mixture was then incubated for 20 min at 25 1C in a shaking water bath. Anti-HCG antibody (1 mg mL 1) diluted with phosphate buffer (10 mmol L 1, pH 7.5) was added and the resulting suspensions were shaken at room temperature for 6 h. After washing with distilled water, remaining NHS-active sites of grapheme sheet were blocked with 2.0% BSA in a phosphate buffer (10 mmol L 1) for 3 h. The mixture was centrifuged at 15,000 rpm for 30 min at 4 1C to remove any unbound biomolecules. 2.3. Preparation of HRP-labeled antibody–conjugated MNPs HRP-labeled antibody–conjugated MNPs were synthesized based on the literature (Lai et al., 2010). Carboxyl-modified magnetic nanoparticles (4 mg) were washed three times with MES buffer and then suspended to a final volume of 250 μL in the same buffer solution. The HRP-labeled anti-HCG antibody– conjugated MNPs were prepared by adding 0.3 mL HRP-labeled anti-HCG antibody (1 μg mL 1), 0.5 mL EDC (2 mg mL 1) into Fe3O4 nanoparticles suspension, followed by stirring at room temperature for 12 h. Subsequently, the resulting mixture were separated magnetically and then resuspended in 10 mmol L 1 PBS solution (pH ¼ 7.5, containing 2.0% BSA). 2.4. Characterization The surface morphology of the immune complex was evaluated by atomic force microscopy analyses (DI NanoScope IV AFM, Veeco Co. Ltd., USA). Electrochemical measurements were performed on a CH 660C Electrochemical Workstation (Shanghai Chenhua Co. Ltd., China) with a conventional three-electrode system. The working electrode is a glassy carbon electrode, the reference electrode is a saturated calomel electrode (SCE), and the counter electrode is a platinum wire.
2. Materials and methods 2.5. Determination of HCG by using graphene-based CRET 2.1. Reagents and apparatus HRP-labeled anti-HCG antibody, anti-HCG antibody and HCG were obtained from Shanghai Linc-Bio Science Co. LTD. (Shanghai, China). Graphene sheet was obtained from JCNANO (Nanjing, China). Carboxyl-modified Fe3O4 nanoparticles and magnetic separation rack were purchased from BaseLine Chrom Tech Research Centre (Tianjin, China). (N-Morpholino)-ethanesulfonic acid monohydrate (MES), bovine serum albumin (BSA), 1-ethyl-3-(3′-dimethylaminopropyl)-carbodiimide hydrochloride (EDC), luminol and N-hydroxysuccinimide (NHS) were purchased from Sigma-Aldrich (St. Louis, MO). Hydrogen peroxide (H2O2), dopamine (DA), ascorbic acid (AA), uric acid (UA), glucose and lysozyme (Lys) were provided by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). All other chemicals used in this work were of analytical grade. Doubly distilled water was used throughout the experiments. Human serum and saliva samples were kindly provided by a team of volunteers and then stored at 20 1C until analysis. ELISA Kit (HCG) was purchased from Diagnostic Products (DPC, USA). CL spectra
Schematic illustration of a graphene-based CRET platform is shown in Fig. 1. A 100 μL of HRP-labeled antibody–conjugated MNPs suspension (4 mg mL 1) was firstly mixed with 100 μL of human HCG standard solution or sample solution in 96-well plate and then incubated for 30 min at 37 1C. The mixture was separated by magnetic separation rack and then incubated with 200 μL of anti-HCG antibody–conjugated graphene suspension (0.1 mg mL 1) for 2.5 h at 37 1C to form a sandwich-type immunocomplex. After magnetic separation and gentle washing with distilled water, the resultant mixture was resuspended in 120 μL of PBS solution (10 mmol L 1, pH¼ 9.0). Subsequently, 50 μL H2O2 (3.5 mmol L 1) was added into the mixture and then 30 μL luminol solution (10 mmol L 1) was introduced by using an automatic injection of TECAN Infinite 200 (Tecan, San Jose, CA). The decreased chemiluminescence intensity caused by CRET was represented as ΔI¼I1 I0. Here, I1 and I0 were the chemiluminescence intensities of the system without and with adding the antibody–conjugated graphene, respectively. The chemiluminescence intensity was recorded after 30 s by using TECAN Infinite 200 microplate reader (Tecan, San
74
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
Fig. 1. The schematic illustration of the determination of HCG by using graphene-based CRET immunoassay.
Jose, CA). The standard curve method was used in quantification of trace amount of HCG in saliva and serum samples. 2.6. Verification test Ten serum samples from a team of volunteers were analyzed by using the proposed method and the plate ELISA with colorimetric detection to verify the practical perspective of graphene-based CRET detection. According to the guideline of ELISA Kit, the procedure was as following: 50 μL of sample was incubated with anti-HCG antibody adsorbed on the ELISA plate and then HRP-conjugate reagent was added to from a sandwich-type immunocomplex. After washing five times, substrates were orderly added into the micro-well. Subsequently, the mixture was incubated for 15 min and then the reaction was terminated by the addition of a sulfuric acid solution. The optical density (OD) at 450 nm was measured using a microplate reader (TECAN Infinite 200; San Jose, CA).
3. Results and discussion 3.1. Characterization of the antibody–conjugated functional materials The morphologies and microstructures of anti-HCG antibody– conjugated graphene (A), HRP-labeled antibody–conjugated MNPs (B) and the sandwich-type immunocomplex (C) were studied by means of AFM (Fig. 2). It is shown that antibodies were immobilized on the surface of graphene sheets (about 6 mg g 1), which induced the height increase of the complex (antibody–conjugated graphene) to about 2 nm. Then we found that the HRP-labeled antibody– conjugated MNPs were uniform cobblestone-like shape with particle diameter of 300–400 nm. Finally, antibody–conjugated graphene, antigen (HCG), and HRP-labeled antibody–conjugated MNPs were incubated in a phosphate buffer (10 mmol L 1). After magnetic separation and gentle washing with distilled water, a large number of magnetic particles were bound on the surface of graphene sheet by the antigen–antibody interactions, which would lead to the CRET phenomenon. Electric impedance spectroscopy (EIS) is a valuable and convenient tool to monitor the interface properties of surface-modified electrodes. Fig. S1 (Supporting Information) represents the EIS of variously modified electrodes after each step in 10 mmol L 1 PBS
solution (pH 7.4) containing 5.0 mmol L 1 Fe(CN)64 /3 and 0.1 mol L 1 KCl. It was observed that MNPs and graphene modified electrode exhibited very small semicircles. Compare with these electrodes, the electron-transfer resistance of antibody–conjugated MNPs (graphene) modified electrode increased, which suggested that antibody had been successfully immobilized on the surface of functional materials. Furthermore, because of poor electrical conductors of biological molecules, the formation of sandwich-type immunocomplex could result in the significant increasing of electron-transfer resistance. It was shown that the sandwich-type immunocomplex could be formed and the close proximity led to the CRET phenomenon between CL light source and grapheme. 3.2. Optimization of the parameters affecting CRET signals In a sandwich immunoreaction, selective recognition for target compound was resulted from the amount of HRP-labeled anti-HCG antibody immobilized on the MNPs. Therefore, different amounts of HRP-labeled anti-HCG antibody were used to prepare antibody– conjugated MNPs. Subsequently, this mixture was allowed to react with a HCG standard solution (100 IU mL 1) and residual HCG in supernatant solution after magnetic separation was determined by the plate ELISA with colorimetric detection. The ΔA value could be used to optimize the amount of HRP-labeled anti-HCG antibody, which was represented as ΔA¼ A0 A1. Here, A0 was the absorbance of HCG standard solution (100 IU mL 1) and A1 was the absorbance of residual HCG in supernatant solution. It was demonstrated that the amount of HRP-labeled anti-HCG antibody (1.0 μg mL 1) immobilized on MNPs showed a maximum at 0.3 mL, with a significant decrease at lower and higher volume of antibody (Fig. S2A, Supporting Information). Furthermore, anti-HCG antibody modified on the surface of graphene was used to form a sandwich-type immunocomplex and then cause CRET phenomenon. The results indicated that the amount of immunocomplex increased with the increasing concentration of anti-HCG antibody from 0.2 μg mL 1 to 1.0 μg mL 1 and further increasing the amount of antibody did not produce any significant improvement (Fig. S2B, Supporting Information). Thus, the optimized concentration of anti-HCG antibody (200 μL) was fixed at 1.0 μg mL 1. The reaction of the sandwich-type immunocomplex was greatly influenced by the incubation time. Therefore, 10 mIU mL 1 of HCG standard solution was prepared and then HRP-labeled antibody– conjugated MNPs and antibody–conjugated graphene were added to
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
75
Fig. 2. AFM images of (A) anti-HCG antibody–conjugated graphene, (B) HRP-labeled antibody–conjugated MNPs and (C) the sandwich-type immunocomplex.
the sample solution, respectively. In this case, the incubation time of the one-step sandwich reaction (antigen (HCG)-antibody–conjugated MNPs) varied from 0 to 60 min. It was found that the amount of adsorbed HCG increased sharply with the increasing of incubation time and then almost trended to the maximum at 30 min, which indicated that the immuno-binding reached the saturation (Fig. S2C, Supporting Information). For the incubation time between the antibody–conjugated graphene and antigen (HCG)-antibody–conjugated MNPs, the ΔI value increased and leveled off at 2.5 h with the increasing of incubation time (Fig. S2D, Supporting Information). Hence, 2.5 h of incubation time for the second immunoreaction was selected for the following experiments. Finally, chemiluminescence played an important role in graphene-based CRET. Thus, the parameters affecting CL signals were further optimized, such as the pH value of PBS solution, the volume of luminol and the concentration of H2O2. The results indicated that 10 mmol L 1 PBS solution (pH 9.0) was selected as a CL buffer solution, which could ensure the stability of immune complexes and maximize the sensitivity of CL detection. As the chemiluminescent substrates, luminol and H2O2 strongly influenced the luminescence intensity in CRET. It was shown that the ΔI values reached their maximum value when the volume of luminol (10 mmol L 1) was 30 μL. For the concentration of H2O2, the ΔI values were increased with the increasing of H2O2 and the CL signal reached the highest at 3.5 mmol L 1H2O2 (Fig. S3, Supporting Information).
spectral separation of the acceptor emission from the donor emission and then low energy-transfer efficiency. In order to investigate the performance of graphene as an energy acceptor in CRET, FITC-labeled anti-HCG antibody (obtained from Shanghai Linc-Bio Science Co. LTD.) or label-free anti-HCG antibody– conjugated graphene were incubated with antigen (HCG) standard solution and HRP-labeled anti-HCG antibody–conjugated MNPs, respectively (Fig. 3). The amount of FITC-labeled antibody was optimized and the results indicated that the ΔI values reached their maximum value when the concentration of FITC-labeled antibody (100 μL) was 7 μg mL 1. It was shown that a gradual quenching of CL intensity was observed with the increasing amount of antibody–conjugated graphene, which demonstrated that graphene could be used as an efficient long-range energy acceptor in CRET. For the immunoassay using FITC as an energy acceptor, the linear relationship between HCG concentration and CL quenching was changed from 0.5 mIU mL 1 to 5 mIU mL 1, which was narrower than that of the proposed immunoassay using graphene as an acceptor (0.1 mIU mL 1 to 10 mIU mL 1). The possible reason is that graphene exhibits a high planar surface, which enables an increased number of acceptor molecules. In fact, the fraction of photons that are transferred to the acceptor is generally enhanced by increasing the number of acceptor molecules (Morales-Narvaez and Merkoci, 2012). Thus, graphene can be proposed as a highly efficient long-range acceptor in CRET.
3.3. The effect of graphene as an acceptor in CRET
3.4. Interferences study
CRET involves the non-radiative energy transfer from a chemiluminescent donor to a suitable acceptor molecule. Fluorescein isothiocyanate (FITC) has small Stokes shifts (Huang and Ren, 2012; Morales-Narvaez and Merkoci, 2012), which result in poor
To apply the proposed method to determine HCG in human saliva and serum samples, components of blank sample (health male, without HCG) and effects of several foreign species (the concentration was equal to the upper limit of normal reference
76
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
3.5. Method validation
Fig. 3. The calibration curve obtained by the immunoassay using FITC and graphene as acceptors, respectively. HRP-labeled antibody–conjugated MNPs (4 mg mL 1) were incubated with HCG and antibody–conjugated graphene (Red line) or FITC-labeled antibody (Blue line) to form a sandwich-type immunocomplex, respectively. The decreased CL intensity caused by CRET was represented as ΔI¼ I1 I0. Here, I1 and I0 were the CL intensities of the system without and with adding the antibody–conjugated graphene (or FITC-labeled antibody), respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
We developed a simple and sensitive immunoassay for the determination of HCG using graphene-based CRET. Under the optimal conditions, a series of immunoassays was prepared for the determination of different concentrations of HCG (C, mIU mL 1). The change of CL intensity was proportion to the concentration of HCG in the range of 0.1 mIU mL 1 to 10 mIU mL 1 (equal to 0.02–2 ng mL 1), however, the linear range of ELISA kit (Diagnostic Products, USA) was 0.7–15 mIU mL 1, which showed that the proposed method was more sensitive. The equation of the calibration curve was ΔI¼ 23.8þ244.7C (mIU mL 1), with a correlation coefficient of 0.9991. The limit of detection (LOD) at 3s (s ¼S0/S; S0, standard deviation of blank sample; S, the slope of the calibration curve) was 0.06 mIU mL 1, which was lower than those of the reported electrochemical immunoassay (0.3 mIU mL 1 (Wang et al., 2010), 0.15 mIU mL 1 (Li et al., 2011), 12 mIU mL 1 (Tao et al., 2011)). Precision was evaluated by measuring intra-day and inter-day relative standard deviations (RSDs). It was performed by analyzing spiked serum samples five times in one day or over five days at three different concentrations of 0.5, 1 and 5 mIU mL 1. The intraday RSD ranged from 3.8% to 8.7% and the inter-day RSD ranged from 4.2% to 9.5% in 5 days. It was demonstrated that the proposed method could be used for the determination of HCG in complex samples. In addition, in order to investigate the practical perspective of the proposed method, the amounts of HCG in ten human serum samples were determined by the proposed method (X) and the plate ELISA with colorimetric detection (Y), respectively. Those results were statistically compared by the paired t-test. It was shown that the two methods had a good consistency, with a regression equation of Y ¼1.02 Xþ2.11 and the correlation coefficient was 0.9983 (Fig. 5). Compared with the plate ELISA with colorimetric detection, the proposed immunoassay is simple, because of the homogeneous analysis based on the CRET and magnetic separation. These results illustrated that CRET immunoassay exhibited significant promise as a reliable technique for the detection of HCG in human serum and saliva samples.
3.6. Determination of HCG in saliva and serum samples Fig. 4. Specificity of the graphene-based CRET immunoassay. The concentration of HCG was 1 ng mL 1. Interfering compounds: dopamine (DA, 122 ng mL 1), ascorbic acid (AA, 13.2 ng mL 1), uric acid (UA, 50 μg mL 1), glucose (1.1 mg mL 1), bovine serum albumin (BSA, 55 mg mL 1) and lysozyme (Lys, 30 μg mL 1).
range) were investigated by analyzing a HCG standard solution (5 mIU mL 1, equal to 1 ng mL 1), to which increasing amounts of interfering species had been added. Compared with background buffer solution, there were almost no change in the luminescence intensity for blank serum (RSD ¼4.2%) and blank saliva sample (RSD ¼2.9%), which demonstrated that sample matrix could not affect the determination of HCG or enhance background noise. As shown in Fig. 4, the ΔI values of interfering species (DA, AA, UA, glucose and BSA) were not higher than 10% of the ΔI value of HCG and that value of lysozyme was 32.0%, when the concentrations of interfering species were 122 ng mL 1 (DA), 13.2 ng mL 1 (AA), 50 μg mL 1 (UA), 1.1 mg mL 1 (glucose), 55 mg mL 1 (BSA) and 30 μg mL 1 (Lys), respectively. These phenomena showed that the proposed strategy had good selectivity in determination of HCG and could distinguish HCG from complex sample matrix. In fact, the excellent selectivity for CRET immunoassay is attributed to the high specificity of the antigen–antibody immune response and the magnetic separation for purification.
HCG level in serum samples has been used as an important marker to diagnose the early pregnancy and many diseases related to seminal system. Recently, saliva has been considered as an alternative matrix
Fig. 5. Correlation between the results measured by the proposed method and the plate ELISA with colorimetric detection.
J. Lei et al. / Biosensors and Bioelectronics 54 (2014) 72–77
77
Table 1 Determination of HCG in human saliva and serum samples by using graphene-based CRET immunoassay. Sample
Found in sample (mIU mL 1)
RSD (%, n¼ 3)
Added (mIU mL 1)
Total found (mIU mL 1)
Recovery (%, n ¼3)
Found in original sample (mIU mL 1)
Saliva 1
0.3
5.9
0.8
3.7
Saliva 3
0.5
4.3
Serum 1
3.9
7.9
Serum 2
5.4
6.3
Serum 3
4.6
5.1
0.9 1.4 3.4 1.2 1.9 4.0 1.1 1.6 3.6 4.5 5.1 6.8 6.2 6.2 8.5 4.9 5.9 7.8
112.5 107.7 103.0 92.3 105.6 105.3 110.0 106.7 102.9 102.3 104.1 98.6 105.1 96.9 101.2 96.1 105.4 102.6
0.3
Saliva 2
0.5 1 3 0.5 1 3 0.5 1 3 0.5 1 3 0.5 1 3 0.5 1 3
for biochemical parameter monitoring. Thus, the proposed graphenebased CRET immunoassay was used for the determination of HCG in serum and saliva samples. The results indicated that spiked recoveries were changed from 92.3% to 112.5% and RSD was lower than 7.9% (Table 1). It was also noted that HCG level could be detected in the saliva samples and the concentrations were found from 0.3 mIU mL 1 to 0.8 mIU mL 1, which was only about 1% of the concentrations of HCG in serum samples. Therefore, it is necessary to develop a simple and sensitive analytical method for determination of trace HCG in saliva samples. Furthermore, further studies are being undertaken by employing the proposed method to investigate the relationship of HCG in saliva and serum samples and develop a novel and simple clinical monitoring technology. In a word, a simple and sensitive graphene-based CRET immunoassay is developed for the determination of trace HCG in saliva and serum samples, which will exhibit widely practical perspective in clinical diagnosis, treatment and prognosis. 4. Conclusions In summary, a simple and sensitive immunoassay was developed for the determination of trace HCG in saliva and serum samples by using graphene-based CRET. After formation of a sandwich-type immunocomplex, the close proximity led to a CRET phenomenon between CL light source and graphene as a highly efficient longrange acceptor. Moreover, the proposed method had good selectivity for the determination of HCG, due to the interaction between antigen and two different antibodies. Under the optimized condition, the linear range was changed from 0.1 mIU mL 1 to 10 mIU mL 1 and the limit of detection was 0.06 mIU mL 1, which was lower than those of the plate ELISA with colorimetric detection and the reported electrochemical immunoassay. The proposed method could be successfully applied for determination of HCG levels in serum and saliva samples, which would exhibit widely practical perspective on the clinical diagnosis, treatment and prognosis. Acknowledgments This work was supported by the National Nature Science Foundation of China (Grant no. 21207044), Hubei Provincial Population and
0.8
0.5
38.9
54.3
45.6
Family Planning Commission (Grant no. JS-2012005), the National Basic research Grant (973) of China (Grant no. 2009CB320301), the Program for New Century Excellent Talents in University (Grant no. NCET-11-0178) and the Fundamental Research Funds for the Central Universities (HUST: 2012QN239).
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.2013.10.033. References Bi, S., Yan, Y.M., Yang, X.Y., Zhang, S.S., 2009. Chem. Eur. J. 15 (18), 4704–4709. Birken, S., Kovalevskaya, G., O’Connor, J., 2001. Arch. Med. Res. 32 (6), 635–643. Chen, C., Li, B.X., 2013. J. Mater. Chem. B 1 (19), 2476–2481. Chou, S.S., De, M., Luo, J.Y., Rotello, V.M., Huang, J.X., Dravid, V.P., 2012. J. Am. Chem. Soc. 134 (40), 16725–16733. Dereli-Korkut, Z., Gandhok, H., Zeng, L.G., Waqas, S., Jiang, X.J., Wang, S.H., 2013. Biotechnol. Bioeng. 110 (5), 1396–1404. Dong, S.C., Liu, F., Lu, C., 2013. Anal. Chem. 85 (6), 3363–3368. Dreyer, D.R., Park, S., Bielawski, C.W., Ruoff, R.S., 2010. Chem. Soc. Rev. 39 (1), 228–240. Huang, X.Y., Ren, J.C., 2012. Trac-Trend. Anal. Chem. 40, 77–89. Lai, J.P., Shah, B.P., Garfunkel, E., Lee, K.B., 2013. ACS Nano 7 (3), 2741–2750. Lai, Y.C., Yin, W.W., Liu, J.T., Xi, R.M., Zhan, J.H., 2010. Nanoscale Res. Lett. 5 (2), 302–307. Lamy, E., Mau, M., 2012. J. Proteom. 75 (14), 4251–4258. Lee, J.S., Joung, H.A., Kim, M.G., Park, C.B., 2012. ACS Nano 6 (4), 2978–2983. Li, R., Wu, D., Li, H., Xu, C.X., Wang, H., Zhao, Y.F., Cai, Y.Y., Wei, Q., Du, B., 2011. Anal. Biochem. 414 (2), 196–201. Lin, J., Hoppe, A.D., 2013. Microsc. Microanal. 19 (2), 350–359. Liu, B.X., Tan, H.L., Chen, Y., 2013. Anal. Chim. Acta 761, 178–185. Matias, I., Gatta-Cherifi, B., Tabarin, A., Clark, S., Leste-Lasserre, T., Marsicano, G., Piazza, P.V., Cota, D., 2012. PLoS One 7 (7), e423997, http://dx.doi.org/10.1371/ journal.pone.0042399. Morales-Narvaez, E., Merkoci, A., 2012. Adv. Mater. 24 (25), 3298–3308. Qian, J., Wang, C.Q., Pan, X.H., Liu, S.Q., 2013. Anal. Chim. Acta 763, 43–49. Qin, G.X., Zhao, S.L., Huang, Y., Jiang, J., Ye, F.G., 2012. Anal. Chem. 84 (6), 2708–2712. Tao, M.L., Li, X.F., Wu, Z.S., Wang, M., Hua, M., Yang, Y.H., 2011. Clin. Chim. Acta 412 (7–8), 550–555. Vartiainen, J., Alfthan, H., Lehtovirta, P., Stenman, U.H., 2002. BJOG-Int. J. Obstet. Gy 109 (5), 589–590. Wang, J.F., Yuan, R., Chai, Y.Q., Cao, S.R., Guan, S., Fu, P., Min, L.G., 2010. Biochem. Eng. J. 51 (3), 95–101. Wani, T.A., Darwish, I.A., 2012. New J. Chem. 36 (4), 1114–1120. Yang, H.C., Yuan, R., Chai, Y.Q., Zhuo, Y., Su, H.L., 2010. J. Chem. Technol. Biot 85 (4), 577–582.
