Biosensors and Bioelectronics 68 (2015) 487–493
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
HCR-stimulated formation of DNAzyme concatamers on gold nanoparticle for ultrasensitive impedimetric immunoassay Li Hou a, Xiaoping Wu a,n, Guonan Chen a, Huanghao Yang a, Minghua Lu b,nn, Dianping Tang a,n a Institute of Nanomedicine and Nanobiosensing, Key Laboratory of Analysis and Detection for Food Safety (MOE & Fujian Province), Department of Chemistry, Fuzhou University, Fuzhou 350116, PR China b Institute of Environmental and Analytical Science, School of Chemistry and Chemical Engineering, Henan University, Kaifeng 475004, Henan, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 November 2014 Received in revised form 13 January 2015 Accepted 19 January 2015 Available online 20 January 2015
A novel signal-amplified impedimetric immunosensing strategy was successfully developed for ultrasensitive detection of low-abundance proteins (carcinoembryonic antigen, CEA, used as a model) based on hybridization chain reaction (HCR)-stimulated formation of DNAzyme concatamers on nanogold particle accompanying enzyme-triggered biocatalytic precipitation. The assay was carried out on capture antibody-modified electrode by using gold nanoparticle heavily functionalized with initiator strand and detection antibody as the signal-transduction tag with a sandwich-type immunosensing format. In the presence of target CEA, the formed immunocomplex underwent an unbiased strand-displacement reaction by the initiator strand on the gold nanoparticle between two auxiliary single-stranded DNA with the hemin aptamer. Upon hemin introduction, numerous DNAzyme molecules were formed on the concatamers and nanogold particle, which could catalyze 4-chloro-1-naphthol to produce an insoluble precipitation on the electrode, thereby resulting in the amplification of impedimetric signal. Under the optimal conditions, the immuno-HCR assay exhibited good impedimetric responses for the detection of target CEA in the working range from 1.0 pg mL 1 to 20 ng mL 1 with a detection limit of 0.42 pg mL 1. In addition, the immuno-HCR assay was validated by measuring six human serum specimens for target CEA, receiving a highly matched correction between the obtained results by the immuno-HCR assay and the commercialized ELC-based immunoassay method. & 2015 Elsevier B.V. All rights reserved.
Keywords: Impedimetric immunosensor Biocatalytic precipitation Immuno-HCR assay DNAzyme concatamer Carcinoembryonic antigen
1. Introduction The detection of trace protein biomarkers poses a formidable challenge in clinical diagnosis, biomedical research, food quality control, and environmental analysis (Zangheri et al., 2015; Marisonneuve et al., 2015). Immunoassay based on the specific antigen–antibody recognition event is a valuable tool for this purpose (Zhang et al., 2012; Song et al., 2014). Recent research has shown that electrochemical immunoassays attract an increasing attention due to their high sensitivity and specificity, rapid detection, low cost and manpower requirement, and simple instrumentation (Kavosi et al., 2014). Routine approaches on the signal amplification during the measurement mainly consist of nano-signal amplification and enzyme-assisted signal amplification strategies (Liang et al., 2012; Yuan et al., 2012; Li et al., 2012; Martic et al., n
Corresponding authors. Fax: þ 86 591 2286 6135. Corresponding author. Fax: þ 86 378 3881 599. E-mail addresses: [email protected] (X. Wu), [email protected] (M. Lu), [email protected] (D. Tang). nn
http://dx.doi.org/10.1016/j.bios.2015.01.043 0956-5663/& 2015 Elsevier B.V. All rights reserved.
2012). Owing to the limitation of available sites on the antibody for the conjugation of bioactive enzyme, however, nanolabels and enzyme labels are co-employed usually for the construction of electrochemical immunosensing schemes to achieve a high sensitivity and a low detection limit (Alves et al., 2015; Fang et al., 2015). Theoretically, the covalent conjugation of the antibody with the enzyme decreases their intrinsic bioactivity to some extent (Yang et al., 2012). In contrast, the emergence of nanobiotechnology opens a new horizon for the development of enzyme-labeling strategies with the functionalized nanostructures. Typically, bioactive enzymes [e.g., horseradish peroxidase (HRP) and alkaline phosphatase (ALP)] are usually applied for the development of enzyme immunoassays (Zhu et al., 2013). In general, enzymes are globular proteins and range from just 62 amino acid residues in size to over 2500 residues (Smith, 1994). For such a large-sized enzyme protein, hence, the immobilized amount on one nanoparticle is relatively limited. Moreover, most enzymes are much larger than the substrates they act on, and only small portion of the enzyme (around 2–4 amino acids) is directly involved in catalysis. Aptamers are nucleic acid sequences that have been
488
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
selected for their ability to bind specific molecular targets, which can range from small organic molecules to large proteins (Catherine et al., 2014). Inspiringly, some G-quadruplex DNA aptamers have been found to strongly bind the hemin to form the DNAzymes with peroxidase-like activity toward H2O2-mediated oxidation (Yuan et al., 2012). Meanwhile, DNAzymes exhibit many excellent advantages, e.g., simple synthesis, good stability, easy labeling and design flexibility. Ge et al. designed a colorimetric biosensor for rapid detection of Cu2 þ by using hemin-G-quadruplex peroxidase-mimicking DNAzyme (Ge et al., 2014). Yang's group prepared a Pb2 þ -responsive hydrogel by using a DNAzyme and its substrate as cross-linker for sensitive colorimetric detection of Pb2 þ (Huang et al., 2014). Huang and co-workers developed biosensors in vitro selection of RNA-cleaving DNAzyme for ratiometric sensing lanthanides (Huang et al., 2014). Ju's group devised an array-based chemiluminescence imaging method for detection of protein targets by using proximity-dependent DNAzyme formation (Zong et al., 2014). Recently, we also found that the hemin-based DNAzyme could catalyze 4-chloro-1-naphthol (4-CN) to produce an insoluble product on the electrode (Hou et al., 2014). Nevertheless, the immobilized amount of DNAzyme on the gold–palladium hybrid nanostructure was limited. To improve this concern, our motivation in this work is to construct massive DNAzyme concatamers on single nanoparticle to enhance the catalytic efficiency toward H2O2-mediated oxidation. DNA concatamer, one of linear polymeric structures that arise by self-association of short DNA fragments through specific interaction, can be utilized for the formation of DNAzyme concatamer (Filippov et al., 2009). To the best of our knowledge, there is no report focusing on DNAzyme concatamers for the development of impedimetric immunoassay until now. Herein we combine the nanolabel (gold nanoparticle label used in this case due to its easy preparation and good biocompatibility with proteins) with DNAzyme concatamer to construct a novel impedimetric immunosensor for ultrasensitive determination of low-abundant proteins (carcinoembryonic antigen, CEA, as a model) in biological fluids by coupling with enzymatic biocatalytic precipitation (BCP)
technique (Scheme 1). BCP, as a versatile and effective method to form an insoluble product on the support through DNAzyme, can be used to enhance the interfacial electron-transfer feature (Zhao et al., 2012; Sheng et al., 2009; Tang et al., 2012a). Initially, DNA concatamers are triggered on the initiator strand-modified nanogold particle between two auxiliary single-stranded DNA after the formation of the sandwiched immunocomplex. Then DNAzyme can be generated on the DNA concatamer in the presence of hemin, thereby resulting in the formation of DNAzyme concatamer. Gold nanoparticle heavily functionalized with initiator strand and detection antibody is expected to enhance the amount of DNAzyme. Two auxiliary DNA strands (S1 and S2) are used to participate in the concatamer reaction. Upon addition of 4-chloro1-naphthol and H2O2, the concatenated DNAzyme catalyzes 4-chloro-1-naphthol to produce an insoluble benzo-4-chlorohexadienone product and precipitates on the electrode, thus hindering heavily the electron transfer of redox probe in the solution, which can be quantitatively monitored using electrochemical impedance spectroscopy. The impedimetric signal indirectly depends on the concentration of target CEA in the sample.
2. Experimental 2.1. Chemicals Monoclonal rabbit anti-human CEA antibody (designated as Ab1) and CEA standards were obtained from Biocell Biotechnol. Co., Ltd. (Zhengzhou, China). Polyclonal rabbit anti-human CEA antibody (designated as Ab2) was purchased from Bioss Biosynth. Biotechnol. Co., Ltd. (Beijing, China). Hemin was from Tokyo Chem. Inc. (Japan). β-cyclodextrin (CD), bovine serum albumin (BSA), HAuCl4 4H2O and 4-chloro-1-naphthol (4-CN) were achieved from Sinopharm Chem. Re. Co. (Shanghai, China). Oligonucleotides designed in this study were synthesized by Sangon Biotech. Co., Ltd. (Shanghai, China), and the sequences were as follows: Initiator strand (S0): 5′-SH-GTACTACAGCAGCTG-3′
Scheme 1. Schematic illustration of HCR-stimulated formation of DNAzyme concatamers on gold nanoparticle for ultrasensitive impedimetric immunoassay.
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
Strand I (S1): 5′-GGGTAGGGCGGGTTGGGTATCTCCTAATAGCAGCAGCTGCTGTAGTAC-3′ Strand II (S2): 5′-CTGCTATTAGGAGATGTACTACAGCAGCTG-3′ In the strand I, the hemin-binding aptamer is underlined. DNA stock solution was obtained by dissolving the corresponding oligonucleotide in 0.01 M phosphate-buffered saline (PBS, pH 7.4) solution. Each oligonucleotide was heated to 90 °C for 5 min, and slowly cooled down to room temperature (RT) before use. All other chemicals were of analytical grade. All solutions were prepared with deionized water obtained from a Milli-Q water purifying system (18.2 MΩ cm 1, Millipore). 2.2. Preparation of gold nanoparticle heavily functionalized with S0 and Ab2 Gold nanoparticle heavily functionalized with the initiator strand and detection antibody (designated as Ab2-AuNP-S0) was synthesized similar to our previous report (Zhang et al., 2012). Briefly, detection antibody (Ab2, 200 μL, 100 μg mL 1) was initially added to colloidal gold nanoparticles (AuNP with 16 nm in diameter, 1.0 mL, C[Au] ¼24 μM, pH 9.0–9.5) and incubated for 20 min at RT. Afterward, the initiator strand (S0, 1.0 OD) was injected into the resulting mixture and incubated overnight at 4 °C with gentle shaking on an end-over-end shaker (MS, IKA GmbH, Staufen, Germany) to make S0 and Ab2 adsorbed on the nanogold particle through the Au–S or Au–NH2 bond. On the following day, the resultant suspension was centrifuged (12,000g) for 10 min at 4 °C to remove the excess S0 and Ab2. Finally, the as-prepared Ab2-AuNP-S0 was dispersed into 1-mL PBS (0.01 M, pH 7.4) containing 1.0 wt % BSA for use. 2.3. Fabrication of impedimetric immunosensor The impedimetric immunosensor was prepared according to your previous report (Tang et al., 2012b). Briefly, a cleaned glassy carbon electrode (GCE, 3 mm in diameter) was scanned for five cycles in 0.1 M H2SO4 within the potential range from 0 to 1.5 V to make the –OH or –COOH formed on the electrode (Won et al., 2005). After being washed with distilled water, β-cyclodextrin aqueous solution (5 μL, 10 mg mL 1) was dropped onto the pretreated GCE, and dried for 2 h at RT to form a CD-modified surface (Jia et al., 2011). Following that, 10 μL of Ab1 in pH 7.4 PBS (100 μ g mL 1) was thrown on the upstanding electrode and reacted for 6 h at 4 °C to capture the antibody on the β-cyclodextrin based on the host–guest chemistry. Finally, the as-prepared immunosensor (i.e., Ab1-CD-GCE) was utilized for the detection of target CEA. 2.4. Electrochemical measurement In this work, all impedimetric measurements were carried out by an AutoLab electrochemical workstation (μAUTIII.FRA2.v, Eco Chemie, The Netherlands) with a three-electrode system containing a modified glassy carbon working electrode, platinum wire-counter electrode, and a saturated calomel electrode (SCE) reference electrode. Before measurement, a biocatalytic precipitation solution was prepared in 0.01 M pH 7.4 PBS including 1.0 mM 4-CN, 0.15 mM H2O2 and 2% (v/v) ethanol. The assay was implemented as follows: (i) dropping an incubation solution including CEA standards/samples (3 μL) and Ab2-AuNP-S0 (7 μL) onto a upstanding Ab1-CD-GCE and incubating for 20 min at RT to form a sandwiched immune complex due to the specific antigen– antibody reaction; (ii) immersing into the hybridization solution containing 0.5 μM S1 and 0.5 μM S2, and incubating for 60 min at RT to form DNA concatamer on the AuNP because of the stranddisplacement reaction; (iii) suspending the resultant electrode into 0.2 mM hemin solution and reacting for 40 min at RT to form
489
the DNAzyme concatamer; (iv) dipping the electrode into the asprepared precipitation solution and incubating for 20 min at RT (Note. The resulting electrode was washed with pH 7.4 PBS after each step). Finally, the impedimetric measurement was carried out in 0.01 M pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 and 0.1 M KCl (Frequency: 10 2–106 Hz; Potential: 220 mV; Alternating voltage: 10 mV). A Nyquist plot (Zre vs. Zim) was drown to analyze the impedimetric results. The shift in the resistance (ΔRet) was calculated relative to zero analyte. Analyses were always made in triplicate.
3. Results and discussion 3.1. Characterization of DNAzyme concatamer and Ab2-AuNP-S0 Scheme 1 gives the fabrication process of the impedimetric immunosensor. During the measurement, the successful construction of DNAzyme concatamers on the gold nanoparticle should be very crucial. Single-stranded S1 and S2 are initially used for the progression of DNA concatamers, and then introduction of hemin results in formation of DNAzyme concatamers. The initial 15 bases at the 3′ end of S1 are complementary to the initiator strand and the 3′ end of S2, while the 15 bases in the middle of S1 are complementary to the 5′ end of S2. The initial 18 bases at the 5′ end of S1 are the hemin-binding aptamer. In the presence of target CEA, the formed sandwich-type immunocomplex can trigger the strand-displacement reaction by the labeled initiator strand on the gold nanoparticle between two auxiliary DNA strands. Upon hemin introduction, numerous DNAzyme molecules were formed on the concatamers. In this way, each initiator strand on single gold nanoparticle propagates a HCR event between S1 and S2 to form a long DNAzyme concatamer. Upon addition of 4-chloro-1naphthol and H2O2, the concatenated DNAzyme can catalyze 4-chloro-1-naphthol to produce an insoluble product and precipitates on the electrode, thus resulting in the amplification of impedimetric signal. To investigate whether the designed initiator strand could trigger the progression of hybridization chain reaction between S1 and S2, the process was characterized by using gel electrophoresis (Fig. 1A). Lanes 1–3 represent the electrophoresis images of S0, S1 and S2 alone, respectively. The corresponding base number was almost the same as our design. Significantly, two similar spots with S0 and S2 could be appeared at the mixture containing 0.5 μM S0 and 0.5 μM S2 (lane 4), suggesting that strands S0 and S2 could not self-hybridize with each other. When 0.5 μM S0 was incubated with 0.5 μM S1, one spot at 60 nt was observed (lane 5), indicating that strand S0 could hybridize with strand S1. When 0.5 μM S0 was reacted with 0.5 μM S1 and 0.5 μM S2 in sequence, however, one strong spot was achieved (lane 6). Obviously, the base number almost summed to S0, S1 and S2. The strong spot mainly derived from the formed long-nicked DNA concatamer due to the hybridization chain reaction. Thus, DNA concatamer could be triggered through the guidance of S0 primer between S1 and S2. To further demonstrate the formation of Ab2-AuNP-S0, the UV– vis absorption spectroscopic characteristics of gold colloids before and after modification were investigated (Fig. 1B). Curve ‘a’ shows UV–vis absorption spectroscope of the newly prepare gold colloids, and a characteristic peak at 518 nm was achieved. After formation of Ab2-AuNP-S0, two absorption peaks at 262 and 521 nm were observed (curve ‘d’). Compared with Ab2 (260 nm, curve ‘b’), S0 (278 nm, curve ‘c’) and gold colloids (518 nm, curve ‘a’) alone, the shift in the absorption peak was mainly attributed to the strong interaction between S0/Ab2 and AuNP.
490
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
Fig. 1. (A) Gel electrophoresis (lane 1: 0.5 μM S0, lane 2: 0.5 μM S1, lane 3: 0.5 μM S2, lane 4: 0.5 μM S0 þ0.5 μM S2, lane 5: 0.5 μM S0 þ0.5 μM S1, lane 6: 0.5 μM S0 þ 0.5 μM S1 þ 0.5 μM S2, lane 7: DNA marker). (B) UV–vis absorption spectra of (a) gold colloids, (b) DNA, (c) Ab2 antibody and (d) Ab2-AuNP-S0.
3.2. Impedimetric characterization Fig. 2A shows the Nyquist diagrams of variously modified electrode after each step in pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 and 0.1 M KCl. As shown from the inset, the impedimetric results could be fitted to a Randles equivalent circuit
including four parameters (Rs: the resistance of the electrolyte solution; Cdl: the double-layer capacitance; Ret: the electron transfer resistance; Zw: Warburg impedance) (Katz and Willner, 2003). The Nyquist diagrams consisted of a semicircle at the highfrequency side corresponding to the electron transfer limiting process and a straight line with a phase angle of 45° at the
Fig. 2. (A) Nyquist diagrams for (a) Ab1-CD-GCE, (b) electrode ‘a’ after incubation with 5 ng mL 1 CEA, (c) electrode ‘b’ after incubation with excess Ab2-AuNP-S0, (d) electrode ‘c’ after hybridization with S1 and S2, (e) electrode ‘d’ after hybridization with hemin, and (f) electrode ‘e’ after incubation with 4-CN and H2O2 in pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 þ0.1 M KCl with the range from 10 2 Hz to 106 Hz at an alternate voltage of 5 mV (Inset: equivalent circuit); (B) Nyquist diagrams for (a) Ab1-CD-GCE, and (b,c) electrode ‘a’ after incubation with (b) zero analyte and (c) 0.05 ng mL 1 CEA, respectively, using Ab2-AuNP-S0 as the signal tag; (C) resistance response of the Ab2-AuNP-S0/CEA/Ab1-CD-GCE after alternate hybridization with S1 and S2 with various cycles in pH 7.4 PBS containing 1.0 mM 4-CN and 0.15 mM H2O2; and (D) the relationship between the resistance and hybridization cycle in figure C. Error bar represents the standard deviation (n¼ 3).
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
low-frequency side resulting from the diffusion limiting step of the electrochemical process. The straight line with a 45° phase angle, known as Warburg impedance, is closely associated with the diffusion of the redox probes in solution (Gao et al., 2013). The semicircle diameter in the impedance spectrum is equal to the electron transfer resistance, Ret, which reflects the electron transfer kinetics of the redox-probe at electrode interface (Wang et al., 2013). As seen from curve’‘a’, a relatively large resistance was obtained at the Ab1-CD-GCE (Ret E778 Ω). To realize our design, the as-prepared Ab1-CD-GCE was utilized for the detection of 5 ng mL 1 CEA (as an example) by using the Ab2-AuNP-S0 as the signal-transduction tag. As indicated from curve ‘b’ in Fig. 2A, the resistance of the Ab1-CD-GCE was increased (Ret E1248 Ω) after incubating with 5 ng mL 1 CEA. Moreover, the subsequent resistances were constantly increased when the CEA/Ab1-CD-GCE reacted with Ab2-AuNP-S0 (Ret E1754 Ω, curve ‘c’), S1 þS2 (Ret E1866 Ω, curve ‘d’’), and hemin (Ret E 2150 Ω, curve ‘e’). The reason might be most likely a consequence of the fact that the negatively charged DNA backbones, and hindered the electrode transfer (Yang et al., 2012). When the resulting immunosensor reacted with 4-CN and H2O2, significantly, the resistance was heavily increased (Ret E5659 Ω, curve ‘f’). The increase in the resistance mainly derived from DNAzyme toward the 4-CN catalytic oxidation, and produced an insoluble benzo-4chlorohexadienone precipitation on the electrode. The absence of Warburg impedance and the presence of extremely large Ret of the electrode implied that a significantly large amount of charge transfer impeding material was brought to the electrode (Gao et al., 2013). These results revealed that our design should be
491
feasible for the detection of target CEA. Another question arises as to whether the as-prepared Ab2-AuNP-S0 could be non-specifically adsorbed to the Ab1-CD-GCE. To demonstrate this issue, the Ab1-CD-GCE was used for the detection of 0 and 0.05 ng mL 1 CEA by using the abovementioned protocol, respectively (Fig. 2B). As shown from Fig. 2B, the resistance in the absence of target CEA (curve ‘b’) was almost the same as that of Ab1-CD-GCE (curve ‘a’). In contrast with curve ‘b’, the resistance largely increased in the presence of 0.05 ng mL 1 CEA (curve ‘c’). The results indicated that the impedimetric signal really originated from target-induced DNAzyme for biocatalytic precipitation of 4-CN. To further demonstrate that the hybridization chain reaction could be carried out on the electrode by the initiator strand, the formed Ab2-AuNP-S0/CEA/Ab1-CD-GCE (5 ng mL 1 CEA used in this case) was alternately incubated with S1 and S2 alone. Each round hybridization with S1 and S2 was designated as one cycle (n). After each cycle, the resistance was monitored. As shown in Fig. 2C, the resistance increased with the increasing cycle ‘n’. Fig. 4D shows the linear relationship between the resistance and the number of hybridization cycle (n): Ret (Ω) ¼242.93nþ 1647 (n ¼2, 4, 6, 8, 10, 12), indicated that the initiator strand could induce the progression of hybridization chain reaction. 3.3. Optimization of experimental conditions Fig. 3A shows the effect of the incubation time for the antigen– antibody reaction on the resistance of the immunosensor from 5 min to 35 min (5 ng mL 1 CEA used in this case). The resistances
Fig. 3. The effect of (A) incubation time for the antigen–antibody reaction, (B) hybridization reaction time between Ab2-AuNP-S0/CEA/Ab1-CD-GCE and S1 þS2, (C) binding time between G-quadruplex and hemin, and (D) deposition time of 4-CN for the developed immunosensor by using 5 ng mL 1 CEA as an example. Error bar represents the standard deviation (n¼ 3).
492
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
increased with the increasing incubation time, and tended to level off after 20 min. Further, we also investigated dependence of the impedimetric signal on the hybridization time between Ab2-AuNP-S0 and S1 þS2. As seen from Fig. 3B, the resistance increased with the reaction time and reached the maximum value after 60 min. Therefore, 20 and 60 min were used as the incubation time for the formation of immunocomplex and DNA concatamer, respectively. Next, we also monitored the effect of binding time between hemin and the aptamer on the signal of the impedimetric immunosensor. As shown in Fig. 3C, the resistance increased with the increasing incubation time, and reached a plateau at 40 min. Take consideration of the reaction efficiency, 40 min were used for the formation of DNAzyme on the AuNP. To leverage the cumulative nature of 4-CN deposition for sensitivity improvement, the deposition time of 4-CN was studied. As shown in Fig. 3D, an optimal impedimetric signal could be obtained after 20 min. A long incubation time did not cause the large change in the resistance. So, 20 min was selected for 4-CN deposition. 3.4. Analytical performance Under optimal condition, we investigated the analytical performance of the developed immunosensing system toward CEA standards with different concentration on Ab1-CD-GCE by using Ab2-AuNP-S0 as the signal-transduction tag, coupling HCR-stimulated formation of DNAzyme concatamer on the gold nanoparticle with enzymatic biocatalytic precipitation. Fig. 4A shows the linear
relationship between the ΔRet values and logarithm of CEA level in the range from 1.0 pg mL 1 to 20 ng mL 1. The linear regression equation was ΔRet (kΩ)¼1.3272 lg (C[CEA]/ng mL 1)þ3.8501 (R2 ¼0.9813, n ¼30) (Note. Analyses were made in triplicate, and the data were obtained as mean values of three assays). The detection limit (LOD) was 0.42 pg mL 1 estimated at the 3sblank criterion (where sblank is the standard deviation of a blank sample, n¼ 11) (Note. The corresponding Nyquist plots are shown in Fig. 4B). The LOD was obviously lower than those of gold nanoparticle-functionalized carbon nitride hybrid nanosheets-based electrochemiluminescence (ECL) immunoassay (LOD: 6.8 pg mL 1) (Chen et al., 2014), dendritic bipodal scaffold-based electrochemical immunoassay (LOD: 0.2 ng mL 1) (Laboria et al., 2010), carbon nanoparticle/poly(ethylene imine)-based electrochemical immunoassay (LOD: 32 pg mL 1) (Ho et al., 2009), and inductively coupled plasma mass spectrometric method. (LOD: 0.03 ng mL 1) (Liu et al., 2011). The sensitivity was far lower than the cutoff value of CEA in normal human serum (3.0 ng mL 1). Nevertheless, only one disadvantage of the developed strategy needed a longer incubation time ( 140 min) to implement all steps for one sample in comparison with the above-mentioned methods, but which was almost the same as that of the commercialized ELISA kit ( 180 min). Thus, future work should focus on the improvement of the detection time. The reproducibility of the impedimetric immunoassay was evaluated toward two CEA levels including 0.05 and 5 ng mL 1 CEA. Experimental results revealed that the relative standard deviations (RSD) by using the same-batch Ab1-CD-GCE and
Fig. 4. (A) Calibration plots (log C vs. ΔRet) and (B) Nyquist diagrams of the impedimetric immunosensor for target CEA from 1.0 pg mL 1 to 20 ng mL 1 by using Ab2-AuNP-S0 in pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 þ0.1 M KCl; (C) the specificity of the impedimetric immunosensors; and (D) comparison of assayed results using the impedimetric immunosensor and ECL referenced method. Error bar represents the standard deviation (n¼ 3).
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
Table 1 Comparison of the assay results for real samples using the impedimetric immunosensor and the referenced electrochemiluminescent (ECL) method. Sample Method;a Concentration [mean 7 SD (RSD), n¼ 3, ng mL 1
1 2 3 4 5 6
Found by impedimetric immunosensor
Found by ECL immunoassay
2.83 7 0.32 (11.3%) 4.23 7 0.27 (6.4%) 3.62 7 0.43 (11.9%) 1.23 7 0.13 (10.6%) 1.127 0.086 (7.7%) 2.23 7 0.091 (4.1%)
2.92 70.21 (7.2%) 4.18 70.14 (3.3%) 3.3570.27 (8.1%) 1.09 70.11 (10.1%) 1.02 70.095 (9.3%) 2.38 70.065 (2.7%)
texp
0.41 0.28 0.92 1.42 1.35 2.32
a
The regression equation (linear) for these data is as follows: y¼ 0.9939xþ 0.0686 (R2 ¼0.9851, n¼ 18) (x-axis: by impedimetric immunosensor; y-axis: by the ECL).
Ab2-AuNP-S0 were 9.6% and 8.9% for 0.05 and 5 ng mL 1 CEA, respectively, while those of using different-batch Ab1-CD-GCE and Ab2-AuNP-S0 were 9.5%, and 10.4% toward the mentioned-above concentrations, respectively. When Ab1-CD-GCE and Ab2-AuNP-S0 were stored at 4 °C for 15 days, no significant change in the resistance was observed. Therefore, the impedimetric immunoassay had acceptable reproducibility and stability. Further, we also investigated the specificity of the impedimetric immunoassay toward other possible biomarkers, e.g., alpha-fetoprotein (AFP), prostate-specific antigen (PSA) and prolactin (PRL). As seen from Fig. 4C, a relatively high signal could be observed toward target CEA than other biomarkers. Compared with background signal, the presence of interfering materials in the complex matrixes did not cause the significant change in the resistance. So, the specificity of the impedimetric immunosensor was relatively satisfactory. 3.5. Monitoring of real samples To evaluate the method accuracy with standard method, e.g., electrochemiluminescent immunoassay, the as-prepared immunosensor was used for 6 clinical serum specimens (from the local Fujian Provincial Hospital). As seen from Table 1, the texp values in all samples were less than tcrit (tcrit[4, 0.05] ¼2.77). Moreover, the intercept and slope of the regression equation between two methods were close to the ideal situation of ‘0’ and ‘1’ (Fig. 4D). The results indicated a highly matched correction between two methods.
4. Conclusion In the present work, we report on the proof-of-concept of a new and highly sensitive impedimetric immunoassay for the determination of low-abundance protein. The assay was amplified by using HCR-triggered formation of DNAzyme concatamers on single gold nanoparticle accompanying enzymatic biocatalytic precipitation strategy. Experimental results indicated that HCR-stimulated impedimetric immunosensor was capable of continuously carrying out all steps in less than 2.5 h for one sample, including incubation, washing, enzymatic reaction, and measurement. The RSD values for intra- and inter-assay were less than 10.5%. No significant different at the 0.05 significance level was encountered in the analysis of 6 clinical serum specimens between the developed impedimetric immunoassay and the referenced ECL immunoassay for detecting target CEA. Compared with conventional enzyme labels and nanolabels, single gold nanoparticle toward the small-sized DNAzyme could conjugate more than that of
493
macromolecular enzyme. Meanwhile, the DNAzyme could be formed on the gold nanoparticle in the three-dimensional structures. Highlight of this work is to adequately utilize the nano-label, enzyme label and the molecular biological amplification technology for immunoassay development. Future work should be focused on the determination of other biomarkers by tuning the corresponding capture antibody and detection antibody, thereby representing a versatile immunosensing scheme.
Acknowledgments Support by the National Natural Science Foundation of China (Grant nos. 41176079, 21475025 and 21305029), the National Science Foundation of Fujian Province (Grant no. 2014J07001), and the Program for Changjiang Scholars and Innovative Research Team in University (Grant no. IRT1116) is gratefully acknowledged.
References Alves, R., Pimentel, F., Nouws, H., Marques, R., González-García, M., Oliveira, M., Delerue-Matos, C., 2015. Biosens. Bioelectron. 64, 19–24. Catherine, A., Shishido, S., Robbins-Welty, G., Diegelman-Parente, A., 2014. FEBS Open Bio 4, 788–795. Chen, L., Zeng, X., Si, P., Chen, Y., Chi, Y., Kim, D., Chen, G., 2014. Anal. Chem. 86, 4188–4195. Fang, Y., Huang, X., Wang, L., Wang, J., 2015. Biosens. Bioelectron. 64, 324–332. Filippov, N., Lomzov, A., Pyshni, D., 2009. Biophysics 54, 280. Gao, Z., Deng, H., Shen, W., Ren, Y., 2013. Anal. Chem. 85, 1624–1630. Ge, C., Luo, Q., Wang, D., Zhao, S., Liang, X., Yu, L., Xing, X., Zeng, L., 2014. Anal. Chem. 86, 6387–6392. Hou, L., Gao, Z., Xu, M., Cao, X., Wu, X., Chen, G., Tang, D., 2014. Biosens. Bioelectron. 54, 365–371. Ho, J., Lin, Y., Wang, L., Hwang, K., Chou, P., 2009. Anal. Chem. 81, 1340–1346. Huang, P., Vazin, M., Liu, J., 2014. Anal. Chem. 86, 9993–9999. Huang, Y., Ma, Y., Chen, Y., Wu, X., Fang, L., Zhu, Z., Yang, C., 2014. Anal. Chem. 86, 11434–11439. Jia, D., Dai, J., Yuan, H., Lei, L., Xiao, D., 2011. Talanta 85, 2344–2351. Katz, E., Willner, I., 2003. Electroanalysis 15, 913–947. Kavosi, B., Hallaj, R., Teymourian, H., Salimi, A., 2014. Biosens. Bioelectron. 59, 389–396. Laboria, N., Fragoso, A., Kemmner, W., Latta, D., Nilsson, O., Botero, M., Drese, K., O’Sullivan, C., 2010. Anal. Chem. 82, 1712–1719. Li, Y., Ma, M., Zhu, J., 2012. Anal. Chem. 84, 10492–10499. Liang, G., Liu, S., Zou, G., Zhang, X., 2012. Anal. Chem. 84, 10645–10649. Liu, R., Liu, X., Tang, Y., Wu, L., Hou, X., Lv, Y., 2011. Anal. Chem. 83, 2330–2336. Marisonneuve, M., Valsecchi, C., Wang, C., Brolo, A., Meunier, M., 2015. Biosens. Bioelectron. 63, 80–85. Martic, S., Gabriel, M., Turowec, J., Lichfield, D., Kraatz, H., 2012. Anal. Chem. Soc. 134, 17036–17045. Sheng, Q., Zheng, J., 2009. Biosens. Bioelectron. 24, 1621–1628. Smith, S., 1994. FASEB J. 8, 1249–1259. Song, C., Chen, J., Zhao, Y., Wang, L., 2014. J. Mater. Chem. B 2, 7488–7494. Tang, J., Hou, L., Tang, D., Zhang, B., Zhou, J., Chen, G., 2012a. Chem. Commun. 48, 8180–8182. Tang, J., Tang, D., Zhou, J., Yang, H., Chen, G., 2012b. Chem. Commun. 48, 2627–2629. Wang, G., Gang, X., Zhou, X., Zhang, G., Huang, H., Zhang, X., 2013. Talanta 103, 75–80. Won, M., Rahman, M., Kwon, N., Shankaran, D., Shim, Y., 2005. Electroanalysis 17, 2231–2238. Yang, L., Liu, C., Ren, W., Li, Z., 2012. Appl. Mater. Interfaces 4, 6450–6453. Yang, N., Cao, Y., Han, P., Zhu, X., Sun, L., Li, G., 2012. Anal. Chem. 84, 2492–2497. Yuan, L., Xu, L., Liu, S., 2012. Anal. Chem. 84, 10737–10744. Yuan, Y., Yuan, R., Chai, Y., Zhuo, Y., Ye, X., Gan, X., Bai, L., 2012. Chem. Commun. 48, 4621–4623. Zangheri, M., Cevenini, L., Anfossi, L., Baggiani, C., Simoni, P., Di Nardo, F., Roda, A., 2015. Biosens. Bioelectron. 64, 63–68. Zhang, B., Liu, B., Tang, D., Niessner, R., Chen, G., Knopp, D., 2012. Anal. Chem. 84, 5392–5399. Zhao, W., Ma, Z., Yu, P., Dong, X., Xu, J., Chen, H., 2012. Anal. Chem. 84, 917–923. Zhu, J., Yang, X., Zhang, L., Zhang, L., Lou, B., Dong, S., Wang, E., 2013. Chem. Commun. 49, 5459–5461. Zong, C., Wu, J., Liu, M., Yang, L., Yan, F., Ju, H., 2014. Anal. Chem. 86, 9939–9944.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
HCR-stimulated formation of DNAzyme concatamers on gold nanoparticle for ultrasensitive impedimetric immunoassay Li Hou a, Xiaoping Wu a,n, Guonan Chen a, Huanghao Yang a, Minghua Lu b,nn, Dianping Tang a,n a Institute of Nanomedicine and Nanobiosensing, Key Laboratory of Analysis and Detection for Food Safety (MOE & Fujian Province), Department of Chemistry, Fuzhou University, Fuzhou 350116, PR China b Institute of Environmental and Analytical Science, School of Chemistry and Chemical Engineering, Henan University, Kaifeng 475004, Henan, PR China
art ic l e i nf o
a b s t r a c t
Article history: Received 29 November 2014 Received in revised form 13 January 2015 Accepted 19 January 2015 Available online 20 January 2015
A novel signal-amplified impedimetric immunosensing strategy was successfully developed for ultrasensitive detection of low-abundance proteins (carcinoembryonic antigen, CEA, used as a model) based on hybridization chain reaction (HCR)-stimulated formation of DNAzyme concatamers on nanogold particle accompanying enzyme-triggered biocatalytic precipitation. The assay was carried out on capture antibody-modified electrode by using gold nanoparticle heavily functionalized with initiator strand and detection antibody as the signal-transduction tag with a sandwich-type immunosensing format. In the presence of target CEA, the formed immunocomplex underwent an unbiased strand-displacement reaction by the initiator strand on the gold nanoparticle between two auxiliary single-stranded DNA with the hemin aptamer. Upon hemin introduction, numerous DNAzyme molecules were formed on the concatamers and nanogold particle, which could catalyze 4-chloro-1-naphthol to produce an insoluble precipitation on the electrode, thereby resulting in the amplification of impedimetric signal. Under the optimal conditions, the immuno-HCR assay exhibited good impedimetric responses for the detection of target CEA in the working range from 1.0 pg mL 1 to 20 ng mL 1 with a detection limit of 0.42 pg mL 1. In addition, the immuno-HCR assay was validated by measuring six human serum specimens for target CEA, receiving a highly matched correction between the obtained results by the immuno-HCR assay and the commercialized ELC-based immunoassay method. & 2015 Elsevier B.V. All rights reserved.
Keywords: Impedimetric immunosensor Biocatalytic precipitation Immuno-HCR assay DNAzyme concatamer Carcinoembryonic antigen
1. Introduction The detection of trace protein biomarkers poses a formidable challenge in clinical diagnosis, biomedical research, food quality control, and environmental analysis (Zangheri et al., 2015; Marisonneuve et al., 2015). Immunoassay based on the specific antigen–antibody recognition event is a valuable tool for this purpose (Zhang et al., 2012; Song et al., 2014). Recent research has shown that electrochemical immunoassays attract an increasing attention due to their high sensitivity and specificity, rapid detection, low cost and manpower requirement, and simple instrumentation (Kavosi et al., 2014). Routine approaches on the signal amplification during the measurement mainly consist of nano-signal amplification and enzyme-assisted signal amplification strategies (Liang et al., 2012; Yuan et al., 2012; Li et al., 2012; Martic et al., n
Corresponding authors. Fax: þ 86 591 2286 6135. Corresponding author. Fax: þ 86 378 3881 599. E-mail addresses: [email protected] (X. Wu), [email protected] (M. Lu), [email protected] (D. Tang). nn
http://dx.doi.org/10.1016/j.bios.2015.01.043 0956-5663/& 2015 Elsevier B.V. All rights reserved.
2012). Owing to the limitation of available sites on the antibody for the conjugation of bioactive enzyme, however, nanolabels and enzyme labels are co-employed usually for the construction of electrochemical immunosensing schemes to achieve a high sensitivity and a low detection limit (Alves et al., 2015; Fang et al., 2015). Theoretically, the covalent conjugation of the antibody with the enzyme decreases their intrinsic bioactivity to some extent (Yang et al., 2012). In contrast, the emergence of nanobiotechnology opens a new horizon for the development of enzyme-labeling strategies with the functionalized nanostructures. Typically, bioactive enzymes [e.g., horseradish peroxidase (HRP) and alkaline phosphatase (ALP)] are usually applied for the development of enzyme immunoassays (Zhu et al., 2013). In general, enzymes are globular proteins and range from just 62 amino acid residues in size to over 2500 residues (Smith, 1994). For such a large-sized enzyme protein, hence, the immobilized amount on one nanoparticle is relatively limited. Moreover, most enzymes are much larger than the substrates they act on, and only small portion of the enzyme (around 2–4 amino acids) is directly involved in catalysis. Aptamers are nucleic acid sequences that have been
488
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
selected for their ability to bind specific molecular targets, which can range from small organic molecules to large proteins (Catherine et al., 2014). Inspiringly, some G-quadruplex DNA aptamers have been found to strongly bind the hemin to form the DNAzymes with peroxidase-like activity toward H2O2-mediated oxidation (Yuan et al., 2012). Meanwhile, DNAzymes exhibit many excellent advantages, e.g., simple synthesis, good stability, easy labeling and design flexibility. Ge et al. designed a colorimetric biosensor for rapid detection of Cu2 þ by using hemin-G-quadruplex peroxidase-mimicking DNAzyme (Ge et al., 2014). Yang's group prepared a Pb2 þ -responsive hydrogel by using a DNAzyme and its substrate as cross-linker for sensitive colorimetric detection of Pb2 þ (Huang et al., 2014). Huang and co-workers developed biosensors in vitro selection of RNA-cleaving DNAzyme for ratiometric sensing lanthanides (Huang et al., 2014). Ju's group devised an array-based chemiluminescence imaging method for detection of protein targets by using proximity-dependent DNAzyme formation (Zong et al., 2014). Recently, we also found that the hemin-based DNAzyme could catalyze 4-chloro-1-naphthol (4-CN) to produce an insoluble product on the electrode (Hou et al., 2014). Nevertheless, the immobilized amount of DNAzyme on the gold–palladium hybrid nanostructure was limited. To improve this concern, our motivation in this work is to construct massive DNAzyme concatamers on single nanoparticle to enhance the catalytic efficiency toward H2O2-mediated oxidation. DNA concatamer, one of linear polymeric structures that arise by self-association of short DNA fragments through specific interaction, can be utilized for the formation of DNAzyme concatamer (Filippov et al., 2009). To the best of our knowledge, there is no report focusing on DNAzyme concatamers for the development of impedimetric immunoassay until now. Herein we combine the nanolabel (gold nanoparticle label used in this case due to its easy preparation and good biocompatibility with proteins) with DNAzyme concatamer to construct a novel impedimetric immunosensor for ultrasensitive determination of low-abundant proteins (carcinoembryonic antigen, CEA, as a model) in biological fluids by coupling with enzymatic biocatalytic precipitation (BCP)
technique (Scheme 1). BCP, as a versatile and effective method to form an insoluble product on the support through DNAzyme, can be used to enhance the interfacial electron-transfer feature (Zhao et al., 2012; Sheng et al., 2009; Tang et al., 2012a). Initially, DNA concatamers are triggered on the initiator strand-modified nanogold particle between two auxiliary single-stranded DNA after the formation of the sandwiched immunocomplex. Then DNAzyme can be generated on the DNA concatamer in the presence of hemin, thereby resulting in the formation of DNAzyme concatamer. Gold nanoparticle heavily functionalized with initiator strand and detection antibody is expected to enhance the amount of DNAzyme. Two auxiliary DNA strands (S1 and S2) are used to participate in the concatamer reaction. Upon addition of 4-chloro1-naphthol and H2O2, the concatenated DNAzyme catalyzes 4-chloro-1-naphthol to produce an insoluble benzo-4-chlorohexadienone product and precipitates on the electrode, thus hindering heavily the electron transfer of redox probe in the solution, which can be quantitatively monitored using electrochemical impedance spectroscopy. The impedimetric signal indirectly depends on the concentration of target CEA in the sample.
2. Experimental 2.1. Chemicals Monoclonal rabbit anti-human CEA antibody (designated as Ab1) and CEA standards were obtained from Biocell Biotechnol. Co., Ltd. (Zhengzhou, China). Polyclonal rabbit anti-human CEA antibody (designated as Ab2) was purchased from Bioss Biosynth. Biotechnol. Co., Ltd. (Beijing, China). Hemin was from Tokyo Chem. Inc. (Japan). β-cyclodextrin (CD), bovine serum albumin (BSA), HAuCl4 4H2O and 4-chloro-1-naphthol (4-CN) were achieved from Sinopharm Chem. Re. Co. (Shanghai, China). Oligonucleotides designed in this study were synthesized by Sangon Biotech. Co., Ltd. (Shanghai, China), and the sequences were as follows: Initiator strand (S0): 5′-SH-GTACTACAGCAGCTG-3′
Scheme 1. Schematic illustration of HCR-stimulated formation of DNAzyme concatamers on gold nanoparticle for ultrasensitive impedimetric immunoassay.
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
Strand I (S1): 5′-GGGTAGGGCGGGTTGGGTATCTCCTAATAGCAGCAGCTGCTGTAGTAC-3′ Strand II (S2): 5′-CTGCTATTAGGAGATGTACTACAGCAGCTG-3′ In the strand I, the hemin-binding aptamer is underlined. DNA stock solution was obtained by dissolving the corresponding oligonucleotide in 0.01 M phosphate-buffered saline (PBS, pH 7.4) solution. Each oligonucleotide was heated to 90 °C for 5 min, and slowly cooled down to room temperature (RT) before use. All other chemicals were of analytical grade. All solutions were prepared with deionized water obtained from a Milli-Q water purifying system (18.2 MΩ cm 1, Millipore). 2.2. Preparation of gold nanoparticle heavily functionalized with S0 and Ab2 Gold nanoparticle heavily functionalized with the initiator strand and detection antibody (designated as Ab2-AuNP-S0) was synthesized similar to our previous report (Zhang et al., 2012). Briefly, detection antibody (Ab2, 200 μL, 100 μg mL 1) was initially added to colloidal gold nanoparticles (AuNP with 16 nm in diameter, 1.0 mL, C[Au] ¼24 μM, pH 9.0–9.5) and incubated for 20 min at RT. Afterward, the initiator strand (S0, 1.0 OD) was injected into the resulting mixture and incubated overnight at 4 °C with gentle shaking on an end-over-end shaker (MS, IKA GmbH, Staufen, Germany) to make S0 and Ab2 adsorbed on the nanogold particle through the Au–S or Au–NH2 bond. On the following day, the resultant suspension was centrifuged (12,000g) for 10 min at 4 °C to remove the excess S0 and Ab2. Finally, the as-prepared Ab2-AuNP-S0 was dispersed into 1-mL PBS (0.01 M, pH 7.4) containing 1.0 wt % BSA for use. 2.3. Fabrication of impedimetric immunosensor The impedimetric immunosensor was prepared according to your previous report (Tang et al., 2012b). Briefly, a cleaned glassy carbon electrode (GCE, 3 mm in diameter) was scanned for five cycles in 0.1 M H2SO4 within the potential range from 0 to 1.5 V to make the –OH or –COOH formed on the electrode (Won et al., 2005). After being washed with distilled water, β-cyclodextrin aqueous solution (5 μL, 10 mg mL 1) was dropped onto the pretreated GCE, and dried for 2 h at RT to form a CD-modified surface (Jia et al., 2011). Following that, 10 μL of Ab1 in pH 7.4 PBS (100 μ g mL 1) was thrown on the upstanding electrode and reacted for 6 h at 4 °C to capture the antibody on the β-cyclodextrin based on the host–guest chemistry. Finally, the as-prepared immunosensor (i.e., Ab1-CD-GCE) was utilized for the detection of target CEA. 2.4. Electrochemical measurement In this work, all impedimetric measurements were carried out by an AutoLab electrochemical workstation (μAUTIII.FRA2.v, Eco Chemie, The Netherlands) with a three-electrode system containing a modified glassy carbon working electrode, platinum wire-counter electrode, and a saturated calomel electrode (SCE) reference electrode. Before measurement, a biocatalytic precipitation solution was prepared in 0.01 M pH 7.4 PBS including 1.0 mM 4-CN, 0.15 mM H2O2 and 2% (v/v) ethanol. The assay was implemented as follows: (i) dropping an incubation solution including CEA standards/samples (3 μL) and Ab2-AuNP-S0 (7 μL) onto a upstanding Ab1-CD-GCE and incubating for 20 min at RT to form a sandwiched immune complex due to the specific antigen– antibody reaction; (ii) immersing into the hybridization solution containing 0.5 μM S1 and 0.5 μM S2, and incubating for 60 min at RT to form DNA concatamer on the AuNP because of the stranddisplacement reaction; (iii) suspending the resultant electrode into 0.2 mM hemin solution and reacting for 40 min at RT to form
489
the DNAzyme concatamer; (iv) dipping the electrode into the asprepared precipitation solution and incubating for 20 min at RT (Note. The resulting electrode was washed with pH 7.4 PBS after each step). Finally, the impedimetric measurement was carried out in 0.01 M pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 and 0.1 M KCl (Frequency: 10 2–106 Hz; Potential: 220 mV; Alternating voltage: 10 mV). A Nyquist plot (Zre vs. Zim) was drown to analyze the impedimetric results. The shift in the resistance (ΔRet) was calculated relative to zero analyte. Analyses were always made in triplicate.
3. Results and discussion 3.1. Characterization of DNAzyme concatamer and Ab2-AuNP-S0 Scheme 1 gives the fabrication process of the impedimetric immunosensor. During the measurement, the successful construction of DNAzyme concatamers on the gold nanoparticle should be very crucial. Single-stranded S1 and S2 are initially used for the progression of DNA concatamers, and then introduction of hemin results in formation of DNAzyme concatamers. The initial 15 bases at the 3′ end of S1 are complementary to the initiator strand and the 3′ end of S2, while the 15 bases in the middle of S1 are complementary to the 5′ end of S2. The initial 18 bases at the 5′ end of S1 are the hemin-binding aptamer. In the presence of target CEA, the formed sandwich-type immunocomplex can trigger the strand-displacement reaction by the labeled initiator strand on the gold nanoparticle between two auxiliary DNA strands. Upon hemin introduction, numerous DNAzyme molecules were formed on the concatamers. In this way, each initiator strand on single gold nanoparticle propagates a HCR event between S1 and S2 to form a long DNAzyme concatamer. Upon addition of 4-chloro-1naphthol and H2O2, the concatenated DNAzyme can catalyze 4-chloro-1-naphthol to produce an insoluble product and precipitates on the electrode, thus resulting in the amplification of impedimetric signal. To investigate whether the designed initiator strand could trigger the progression of hybridization chain reaction between S1 and S2, the process was characterized by using gel electrophoresis (Fig. 1A). Lanes 1–3 represent the electrophoresis images of S0, S1 and S2 alone, respectively. The corresponding base number was almost the same as our design. Significantly, two similar spots with S0 and S2 could be appeared at the mixture containing 0.5 μM S0 and 0.5 μM S2 (lane 4), suggesting that strands S0 and S2 could not self-hybridize with each other. When 0.5 μM S0 was incubated with 0.5 μM S1, one spot at 60 nt was observed (lane 5), indicating that strand S0 could hybridize with strand S1. When 0.5 μM S0 was reacted with 0.5 μM S1 and 0.5 μM S2 in sequence, however, one strong spot was achieved (lane 6). Obviously, the base number almost summed to S0, S1 and S2. The strong spot mainly derived from the formed long-nicked DNA concatamer due to the hybridization chain reaction. Thus, DNA concatamer could be triggered through the guidance of S0 primer between S1 and S2. To further demonstrate the formation of Ab2-AuNP-S0, the UV– vis absorption spectroscopic characteristics of gold colloids before and after modification were investigated (Fig. 1B). Curve ‘a’ shows UV–vis absorption spectroscope of the newly prepare gold colloids, and a characteristic peak at 518 nm was achieved. After formation of Ab2-AuNP-S0, two absorption peaks at 262 and 521 nm were observed (curve ‘d’). Compared with Ab2 (260 nm, curve ‘b’), S0 (278 nm, curve ‘c’) and gold colloids (518 nm, curve ‘a’) alone, the shift in the absorption peak was mainly attributed to the strong interaction between S0/Ab2 and AuNP.
490
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
Fig. 1. (A) Gel electrophoresis (lane 1: 0.5 μM S0, lane 2: 0.5 μM S1, lane 3: 0.5 μM S2, lane 4: 0.5 μM S0 þ0.5 μM S2, lane 5: 0.5 μM S0 þ0.5 μM S1, lane 6: 0.5 μM S0 þ 0.5 μM S1 þ 0.5 μM S2, lane 7: DNA marker). (B) UV–vis absorption spectra of (a) gold colloids, (b) DNA, (c) Ab2 antibody and (d) Ab2-AuNP-S0.
3.2. Impedimetric characterization Fig. 2A shows the Nyquist diagrams of variously modified electrode after each step in pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 and 0.1 M KCl. As shown from the inset, the impedimetric results could be fitted to a Randles equivalent circuit
including four parameters (Rs: the resistance of the electrolyte solution; Cdl: the double-layer capacitance; Ret: the electron transfer resistance; Zw: Warburg impedance) (Katz and Willner, 2003). The Nyquist diagrams consisted of a semicircle at the highfrequency side corresponding to the electron transfer limiting process and a straight line with a phase angle of 45° at the
Fig. 2. (A) Nyquist diagrams for (a) Ab1-CD-GCE, (b) electrode ‘a’ after incubation with 5 ng mL 1 CEA, (c) electrode ‘b’ after incubation with excess Ab2-AuNP-S0, (d) electrode ‘c’ after hybridization with S1 and S2, (e) electrode ‘d’ after hybridization with hemin, and (f) electrode ‘e’ after incubation with 4-CN and H2O2 in pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 þ0.1 M KCl with the range from 10 2 Hz to 106 Hz at an alternate voltage of 5 mV (Inset: equivalent circuit); (B) Nyquist diagrams for (a) Ab1-CD-GCE, and (b,c) electrode ‘a’ after incubation with (b) zero analyte and (c) 0.05 ng mL 1 CEA, respectively, using Ab2-AuNP-S0 as the signal tag; (C) resistance response of the Ab2-AuNP-S0/CEA/Ab1-CD-GCE after alternate hybridization with S1 and S2 with various cycles in pH 7.4 PBS containing 1.0 mM 4-CN and 0.15 mM H2O2; and (D) the relationship between the resistance and hybridization cycle in figure C. Error bar represents the standard deviation (n¼ 3).
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
low-frequency side resulting from the diffusion limiting step of the electrochemical process. The straight line with a 45° phase angle, known as Warburg impedance, is closely associated with the diffusion of the redox probes in solution (Gao et al., 2013). The semicircle diameter in the impedance spectrum is equal to the electron transfer resistance, Ret, which reflects the electron transfer kinetics of the redox-probe at electrode interface (Wang et al., 2013). As seen from curve’‘a’, a relatively large resistance was obtained at the Ab1-CD-GCE (Ret E778 Ω). To realize our design, the as-prepared Ab1-CD-GCE was utilized for the detection of 5 ng mL 1 CEA (as an example) by using the Ab2-AuNP-S0 as the signal-transduction tag. As indicated from curve ‘b’ in Fig. 2A, the resistance of the Ab1-CD-GCE was increased (Ret E1248 Ω) after incubating with 5 ng mL 1 CEA. Moreover, the subsequent resistances were constantly increased when the CEA/Ab1-CD-GCE reacted with Ab2-AuNP-S0 (Ret E1754 Ω, curve ‘c’), S1 þS2 (Ret E1866 Ω, curve ‘d’’), and hemin (Ret E 2150 Ω, curve ‘e’). The reason might be most likely a consequence of the fact that the negatively charged DNA backbones, and hindered the electrode transfer (Yang et al., 2012). When the resulting immunosensor reacted with 4-CN and H2O2, significantly, the resistance was heavily increased (Ret E5659 Ω, curve ‘f’). The increase in the resistance mainly derived from DNAzyme toward the 4-CN catalytic oxidation, and produced an insoluble benzo-4chlorohexadienone precipitation on the electrode. The absence of Warburg impedance and the presence of extremely large Ret of the electrode implied that a significantly large amount of charge transfer impeding material was brought to the electrode (Gao et al., 2013). These results revealed that our design should be
491
feasible for the detection of target CEA. Another question arises as to whether the as-prepared Ab2-AuNP-S0 could be non-specifically adsorbed to the Ab1-CD-GCE. To demonstrate this issue, the Ab1-CD-GCE was used for the detection of 0 and 0.05 ng mL 1 CEA by using the abovementioned protocol, respectively (Fig. 2B). As shown from Fig. 2B, the resistance in the absence of target CEA (curve ‘b’) was almost the same as that of Ab1-CD-GCE (curve ‘a’). In contrast with curve ‘b’, the resistance largely increased in the presence of 0.05 ng mL 1 CEA (curve ‘c’). The results indicated that the impedimetric signal really originated from target-induced DNAzyme for biocatalytic precipitation of 4-CN. To further demonstrate that the hybridization chain reaction could be carried out on the electrode by the initiator strand, the formed Ab2-AuNP-S0/CEA/Ab1-CD-GCE (5 ng mL 1 CEA used in this case) was alternately incubated with S1 and S2 alone. Each round hybridization with S1 and S2 was designated as one cycle (n). After each cycle, the resistance was monitored. As shown in Fig. 2C, the resistance increased with the increasing cycle ‘n’. Fig. 4D shows the linear relationship between the resistance and the number of hybridization cycle (n): Ret (Ω) ¼242.93nþ 1647 (n ¼2, 4, 6, 8, 10, 12), indicated that the initiator strand could induce the progression of hybridization chain reaction. 3.3. Optimization of experimental conditions Fig. 3A shows the effect of the incubation time for the antigen– antibody reaction on the resistance of the immunosensor from 5 min to 35 min (5 ng mL 1 CEA used in this case). The resistances
Fig. 3. The effect of (A) incubation time for the antigen–antibody reaction, (B) hybridization reaction time between Ab2-AuNP-S0/CEA/Ab1-CD-GCE and S1 þS2, (C) binding time between G-quadruplex and hemin, and (D) deposition time of 4-CN for the developed immunosensor by using 5 ng mL 1 CEA as an example. Error bar represents the standard deviation (n¼ 3).
492
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
increased with the increasing incubation time, and tended to level off after 20 min. Further, we also investigated dependence of the impedimetric signal on the hybridization time between Ab2-AuNP-S0 and S1 þS2. As seen from Fig. 3B, the resistance increased with the reaction time and reached the maximum value after 60 min. Therefore, 20 and 60 min were used as the incubation time for the formation of immunocomplex and DNA concatamer, respectively. Next, we also monitored the effect of binding time between hemin and the aptamer on the signal of the impedimetric immunosensor. As shown in Fig. 3C, the resistance increased with the increasing incubation time, and reached a plateau at 40 min. Take consideration of the reaction efficiency, 40 min were used for the formation of DNAzyme on the AuNP. To leverage the cumulative nature of 4-CN deposition for sensitivity improvement, the deposition time of 4-CN was studied. As shown in Fig. 3D, an optimal impedimetric signal could be obtained after 20 min. A long incubation time did not cause the large change in the resistance. So, 20 min was selected for 4-CN deposition. 3.4. Analytical performance Under optimal condition, we investigated the analytical performance of the developed immunosensing system toward CEA standards with different concentration on Ab1-CD-GCE by using Ab2-AuNP-S0 as the signal-transduction tag, coupling HCR-stimulated formation of DNAzyme concatamer on the gold nanoparticle with enzymatic biocatalytic precipitation. Fig. 4A shows the linear
relationship between the ΔRet values and logarithm of CEA level in the range from 1.0 pg mL 1 to 20 ng mL 1. The linear regression equation was ΔRet (kΩ)¼1.3272 lg (C[CEA]/ng mL 1)þ3.8501 (R2 ¼0.9813, n ¼30) (Note. Analyses were made in triplicate, and the data were obtained as mean values of three assays). The detection limit (LOD) was 0.42 pg mL 1 estimated at the 3sblank criterion (where sblank is the standard deviation of a blank sample, n¼ 11) (Note. The corresponding Nyquist plots are shown in Fig. 4B). The LOD was obviously lower than those of gold nanoparticle-functionalized carbon nitride hybrid nanosheets-based electrochemiluminescence (ECL) immunoassay (LOD: 6.8 pg mL 1) (Chen et al., 2014), dendritic bipodal scaffold-based electrochemical immunoassay (LOD: 0.2 ng mL 1) (Laboria et al., 2010), carbon nanoparticle/poly(ethylene imine)-based electrochemical immunoassay (LOD: 32 pg mL 1) (Ho et al., 2009), and inductively coupled plasma mass spectrometric method. (LOD: 0.03 ng mL 1) (Liu et al., 2011). The sensitivity was far lower than the cutoff value of CEA in normal human serum (3.0 ng mL 1). Nevertheless, only one disadvantage of the developed strategy needed a longer incubation time ( 140 min) to implement all steps for one sample in comparison with the above-mentioned methods, but which was almost the same as that of the commercialized ELISA kit ( 180 min). Thus, future work should focus on the improvement of the detection time. The reproducibility of the impedimetric immunoassay was evaluated toward two CEA levels including 0.05 and 5 ng mL 1 CEA. Experimental results revealed that the relative standard deviations (RSD) by using the same-batch Ab1-CD-GCE and
Fig. 4. (A) Calibration plots (log C vs. ΔRet) and (B) Nyquist diagrams of the impedimetric immunosensor for target CEA from 1.0 pg mL 1 to 20 ng mL 1 by using Ab2-AuNP-S0 in pH 7.4 PBS containing 5.0 mM Fe(CN)64 /3 þ0.1 M KCl; (C) the specificity of the impedimetric immunosensors; and (D) comparison of assayed results using the impedimetric immunosensor and ECL referenced method. Error bar represents the standard deviation (n¼ 3).
L. Hou et al. / Biosensors and Bioelectronics 68 (2015) 487–493
Table 1 Comparison of the assay results for real samples using the impedimetric immunosensor and the referenced electrochemiluminescent (ECL) method. Sample Method;a Concentration [mean 7 SD (RSD), n¼ 3, ng mL 1
1 2 3 4 5 6
Found by impedimetric immunosensor
Found by ECL immunoassay
2.83 7 0.32 (11.3%) 4.23 7 0.27 (6.4%) 3.62 7 0.43 (11.9%) 1.23 7 0.13 (10.6%) 1.127 0.086 (7.7%) 2.23 7 0.091 (4.1%)
2.92 70.21 (7.2%) 4.18 70.14 (3.3%) 3.3570.27 (8.1%) 1.09 70.11 (10.1%) 1.02 70.095 (9.3%) 2.38 70.065 (2.7%)
texp
0.41 0.28 0.92 1.42 1.35 2.32
a
The regression equation (linear) for these data is as follows: y¼ 0.9939xþ 0.0686 (R2 ¼0.9851, n¼ 18) (x-axis: by impedimetric immunosensor; y-axis: by the ECL).
Ab2-AuNP-S0 were 9.6% and 8.9% for 0.05 and 5 ng mL 1 CEA, respectively, while those of using different-batch Ab1-CD-GCE and Ab2-AuNP-S0 were 9.5%, and 10.4% toward the mentioned-above concentrations, respectively. When Ab1-CD-GCE and Ab2-AuNP-S0 were stored at 4 °C for 15 days, no significant change in the resistance was observed. Therefore, the impedimetric immunoassay had acceptable reproducibility and stability. Further, we also investigated the specificity of the impedimetric immunoassay toward other possible biomarkers, e.g., alpha-fetoprotein (AFP), prostate-specific antigen (PSA) and prolactin (PRL). As seen from Fig. 4C, a relatively high signal could be observed toward target CEA than other biomarkers. Compared with background signal, the presence of interfering materials in the complex matrixes did not cause the significant change in the resistance. So, the specificity of the impedimetric immunosensor was relatively satisfactory. 3.5. Monitoring of real samples To evaluate the method accuracy with standard method, e.g., electrochemiluminescent immunoassay, the as-prepared immunosensor was used for 6 clinical serum specimens (from the local Fujian Provincial Hospital). As seen from Table 1, the texp values in all samples were less than tcrit (tcrit[4, 0.05] ¼2.77). Moreover, the intercept and slope of the regression equation between two methods were close to the ideal situation of ‘0’ and ‘1’ (Fig. 4D). The results indicated a highly matched correction between two methods.
4. Conclusion In the present work, we report on the proof-of-concept of a new and highly sensitive impedimetric immunoassay for the determination of low-abundance protein. The assay was amplified by using HCR-triggered formation of DNAzyme concatamers on single gold nanoparticle accompanying enzymatic biocatalytic precipitation strategy. Experimental results indicated that HCR-stimulated impedimetric immunosensor was capable of continuously carrying out all steps in less than 2.5 h for one sample, including incubation, washing, enzymatic reaction, and measurement. The RSD values for intra- and inter-assay were less than 10.5%. No significant different at the 0.05 significance level was encountered in the analysis of 6 clinical serum specimens between the developed impedimetric immunoassay and the referenced ECL immunoassay for detecting target CEA. Compared with conventional enzyme labels and nanolabels, single gold nanoparticle toward the small-sized DNAzyme could conjugate more than that of
493
macromolecular enzyme. Meanwhile, the DNAzyme could be formed on the gold nanoparticle in the three-dimensional structures. Highlight of this work is to adequately utilize the nano-label, enzyme label and the molecular biological amplification technology for immunoassay development. Future work should be focused on the determination of other biomarkers by tuning the corresponding capture antibody and detection antibody, thereby representing a versatile immunosensing scheme.
Acknowledgments Support by the National Natural Science Foundation of China (Grant nos. 41176079, 21475025 and 21305029), the National Science Foundation of Fujian Province (Grant no. 2014J07001), and the Program for Changjiang Scholars and Innovative Research Team in University (Grant no. IRT1116) is gratefully acknowledged.
References Alves, R., Pimentel, F., Nouws, H., Marques, R., González-García, M., Oliveira, M., Delerue-Matos, C., 2015. Biosens. Bioelectron. 64, 19–24. Catherine, A., Shishido, S., Robbins-Welty, G., Diegelman-Parente, A., 2014. FEBS Open Bio 4, 788–795. Chen, L., Zeng, X., Si, P., Chen, Y., Chi, Y., Kim, D., Chen, G., 2014. Anal. Chem. 86, 4188–4195. Fang, Y., Huang, X., Wang, L., Wang, J., 2015. Biosens. Bioelectron. 64, 324–332. Filippov, N., Lomzov, A., Pyshni, D., 2009. Biophysics 54, 280. Gao, Z., Deng, H., Shen, W., Ren, Y., 2013. Anal. Chem. 85, 1624–1630. Ge, C., Luo, Q., Wang, D., Zhao, S., Liang, X., Yu, L., Xing, X., Zeng, L., 2014. Anal. Chem. 86, 6387–6392. Hou, L., Gao, Z., Xu, M., Cao, X., Wu, X., Chen, G., Tang, D., 2014. Biosens. Bioelectron. 54, 365–371. Ho, J., Lin, Y., Wang, L., Hwang, K., Chou, P., 2009. Anal. Chem. 81, 1340–1346. Huang, P., Vazin, M., Liu, J., 2014. Anal. Chem. 86, 9993–9999. Huang, Y., Ma, Y., Chen, Y., Wu, X., Fang, L., Zhu, Z., Yang, C., 2014. Anal. Chem. 86, 11434–11439. Jia, D., Dai, J., Yuan, H., Lei, L., Xiao, D., 2011. Talanta 85, 2344–2351. Katz, E., Willner, I., 2003. Electroanalysis 15, 913–947. Kavosi, B., Hallaj, R., Teymourian, H., Salimi, A., 2014. Biosens. Bioelectron. 59, 389–396. Laboria, N., Fragoso, A., Kemmner, W., Latta, D., Nilsson, O., Botero, M., Drese, K., O’Sullivan, C., 2010. Anal. Chem. 82, 1712–1719. Li, Y., Ma, M., Zhu, J., 2012. Anal. Chem. 84, 10492–10499. Liang, G., Liu, S., Zou, G., Zhang, X., 2012. Anal. Chem. 84, 10645–10649. Liu, R., Liu, X., Tang, Y., Wu, L., Hou, X., Lv, Y., 2011. Anal. Chem. 83, 2330–2336. Marisonneuve, M., Valsecchi, C., Wang, C., Brolo, A., Meunier, M., 2015. Biosens. Bioelectron. 63, 80–85. Martic, S., Gabriel, M., Turowec, J., Lichfield, D., Kraatz, H., 2012. Anal. Chem. Soc. 134, 17036–17045. Sheng, Q., Zheng, J., 2009. Biosens. Bioelectron. 24, 1621–1628. Smith, S., 1994. FASEB J. 8, 1249–1259. Song, C., Chen, J., Zhao, Y., Wang, L., 2014. J. Mater. Chem. B 2, 7488–7494. Tang, J., Hou, L., Tang, D., Zhang, B., Zhou, J., Chen, G., 2012a. Chem. Commun. 48, 8180–8182. Tang, J., Tang, D., Zhou, J., Yang, H., Chen, G., 2012b. Chem. Commun. 48, 2627–2629. Wang, G., Gang, X., Zhou, X., Zhang, G., Huang, H., Zhang, X., 2013. Talanta 103, 75–80. Won, M., Rahman, M., Kwon, N., Shankaran, D., Shim, Y., 2005. Electroanalysis 17, 2231–2238. Yang, L., Liu, C., Ren, W., Li, Z., 2012. Appl. Mater. Interfaces 4, 6450–6453. Yang, N., Cao, Y., Han, P., Zhu, X., Sun, L., Li, G., 2012. Anal. Chem. 84, 2492–2497. Yuan, L., Xu, L., Liu, S., 2012. Anal. Chem. 84, 10737–10744. Yuan, Y., Yuan, R., Chai, Y., Zhuo, Y., Ye, X., Gan, X., Bai, L., 2012. Chem. Commun. 48, 4621–4623. Zangheri, M., Cevenini, L., Anfossi, L., Baggiani, C., Simoni, P., Di Nardo, F., Roda, A., 2015. Biosens. Bioelectron. 64, 63–68. Zhang, B., Liu, B., Tang, D., Niessner, R., Chen, G., Knopp, D., 2012. Anal. Chem. 84, 5392–5399. Zhao, W., Ma, Z., Yu, P., Dong, X., Xu, J., Chen, H., 2012. Anal. Chem. 84, 917–923. Zhu, J., Yang, X., Zhang, L., Zhang, L., Lou, B., Dong, S., Wang, E., 2013. Chem. Commun. 49, 5459–5461. Zong, C., Wu, J., Liu, M., Yang, L., Yan, F., Ju, H., 2014. Anal. Chem. 86, 9939–9944.
