Proximity Ligation Assay with Three-Way Junction-Induced Rolling

Jul 1, 2014 - Circle Amplification for Ultrasensitive Electronic Monitoring of. Concanavalin A. Bingqian Liu, Bing Zhang, Guonan Chen, Huanghao Yang,*...
1 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/ac

Proximity Ligation Assay with Three-Way Junction-Induced Rolling Circle Amplification for Ultrasensitive Electronic Monitoring of Concanavalin A Bingqian Liu, Bing Zhang, Guonan Chen, Huanghao Yang,* and Dianping Tang* Key Laboratory of Analysis and Detection for Food Safety (Fujian Province and Ministry of Education of China), Research Institute of Nanomedicine and Nanobiosensing, Department of Chemistry, Fuzhou University, Fuzhou, Fujian 350108, P.R. China ABSTRACT: A new signal amplification strategy based on target-induced proximity ligation assay accompanying threeway junction-based rolling chain amplification was designed for ultrasensitive detection of concanavalin A (Con A) by coupling with a sequential injection mode. To construct such a proximity ligation assay system, two types of magnetic sensing probes including glucosamine/DNA1-conjugated magnetic bead (GA-MB-DNA1) and glucosamine/DNA2-labeled magnetic bead (GA-MB-DNA2) were first synthesized and prepared through a typical carbodiimide coupling. In the presence of target Con A, GA-MB-DNA1 and GA-MB-DNA2 were ligated together based on the interaction between Con A and the conjugated glucosamine on the MB, thereby resulting in the formation of a three-way DNA junction because of partial base pairing on the DNA1/DNA2. With the aid of ligase and polymerase, the formed three-way DNA junction could be used as the primer to produce numerous repeated oligonucleotide sequences through rolling circle amplification (RCA) reaction. The formed long oligonucleotide strand could cause the intercalation of numerous positively charged methylene blue molecules with a negatively charged DNA backbone. During the electrochemical measurement, each of the intercalated indicators could produce an electrochemical signal within the applied potentials, resulting in the amplification of detectable electronic signal. By monitoring the change in the signal, we could indirectly determine the concentration of target Con A in the sample. Under the optimal conditions, the developed sensing platform exhibited high sensitivity for detection of Con A with a wide dynamic range of 1.96 pM to 98 nM and a low limit of detection (LOD) of 1.5 pM at the 3sB level. Intra-assay and interassay coefficients of variation were less than 8.9% and 9.7%, respectively. In addition, the methodology was validated by assaying Con A spiked samples including newborn cattle serum and peanuts, and the recovery in all cases was 88.8−134.7%.

A

foundation of pathogen detection and prevention of bacterial infection.12 Measurement of the binding affinity of glycoconjugates toward lectins in solution is routinely realized, e.g., using agglutination inhibition assays,13 enzyme-linked lectin assays,14 enzyme-linked immunosorbent assay,15 isothermal titration calorimetry,16 colorimetry,17 scattered light,18 surface plasmon resonance,19,20 and fluorescence spectra.12 Despite the high sensitivity of these methods, they have some limitations, such as expensive instrumentation and complex sample pretreatment, being time-consuming, and requiring skilled personnel. In contrast, the electrochemical method with simple instrumentation and easy signal quantification has become the predominant analytical technique for highly sensitive detection of biomolecules.21 To successfully develop an electrochemistry-based sensing platform with high sensitivity, the signal-produced

s the carbohydrate recognition domain, lectins are carbohydrate-binding proteins or glycoproteins of nonimmune origin that bind to mono/oligosaccharides reversibly with a high-degree stereo specificity.1,2 Lectin−carbohydrate interactions play the key roles in a variety of important biological processes, e.g., cell-surface recognition, cell−cell communications, cancer, and host−pathogen infection.3−6 The detection of the carbohydrates and lectins in disease states and related lectin−carbohydrate interactions are thus of importance in the areas of glycomics and early diagnostics of cancer and other diseases based on the determination of glycobiomarkers.7 Concanavalin A (Con A), extracted from jack bean seeds, is one of the most commonly studied lectins that exists as a dimmer below pH 5.5 and a tetramer between pH 5.8 and pH 7.0 (molecular mass, 104 000 for tetramer).8 Under neutral conditions, each subunit of Con A contains four binding sites.9 Con A binds exclusively to α-glucose and shows no significant affinity toward other carbohydrates (galactose and lactose).2,10,11 Hence, understanding and mimicing carbohydrate and bacterial lectin interactions are of great interest as the © 2014 American Chemical Society

Received: May 6, 2014 Accepted: July 1, 2014 Published: July 1, 2014 7773

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

Scheme 1. Schematic Illustration of the Proximity Ligation Assay with Three-Way Junction-Induced Rolling Circle Amplification for Electrochemical Detection of Concanavalin A

isothermal conditions by a special DNA polymerase with strong strand-displacement ability.28,29 Compared with other isothermal amplification methods, RCA technology with an intrinsically wide dynamic range is a robust and simple procedure that can provide a universal platform for the localization of a wide variety of molecules as a function of either proteins or nucleic acid sequences. In this process, the design of DNA primer for the RCA is very important.30 However, most works reported previously were based on an entire DNA primer sequence for RCA development. Such a protocol was difficult to realize in the same-nanoparticle-based assay. Significantly, proximity ligation assay (PLA) is a powerful method capable of detecting proteins, protein−protein interactions, and post-translational modifications.31,32 A protein-recognition event can be converted into detectable DNA molecules. Binding of two or more conjugates to the target results in the assembly of an assay-specific DNA sequence with unique DNA configuration. As expected, the three-way DNA junction is an important building block to construct DNA architecture and dynamic assembly.33−35 In this regards, our starting point is to exploit target-induced formation of a three-way DNA junction with the RCA primer by the proximity ligation assay. Herein, we design a novel proximity ligation assay protocol accompanying three-way DNA junction-induced rolling circle amplification for ultrasensitive monitoring of concanavalin A on protein/half-primer-functionalized magnetic beads (Scheme 1). The protein assay system consists of two magnetic sensing probes, i.e., glucosamine and single-strand DNA1-conjugated magnetic bead (GA-MB-DNA1) and glucosamine/DNA2labeled magnetic bead (GA-MB-DNA 2 ). The assay is implemented as follows: (i) target-induced aggregation of GA-MB-DNA1 and GA-MB-DNA2 accompanying the formation of a three-way junction due to partial base pairing of DNA1/DNA2, (ii) RCA reaction triggered by the primer on the three-way junction, (iii) the indicator intercalation of electro-

protocol and signal-transduction tag are two basic concerns of interest. The emerging research field of nanoparticle-based assay protocols provides an excitingly new possibility for advanced development of newly signal-produced protocols.22,23 Magnetic beads (MB) are attractive because they have good biocompatibility and can be separated readily from a complex system by applying a local magnetic field gradient.24,25 In this case, the protocol can pull biomolecules bound to magnetic beads from one laminar flow path to another and selectively remove them from flowing biological fluids without any washing steps. To build such a magnetic sorting protein assay system, detectable signal amplification and noise reduction are very crucial. The nanoparticle-based assay allows for detection of low-concentration levels with signal amplification and reduces the sample pretreatment requirement because of the presence of magnetic particles.26 Lee et al. developed a DNA-modified gold nanoparticle-based assay system for the detection of microRNA at aM level without enzymatic amplification.27 Typically, the nanoparticle-based assay consists of two sets of particles (magnetic bead and gold nanoparticle). The signal was derived from gold nanoparticles decorated with proteins and a unique DNA sequence. However, the immobilized proteins on the nanogold particles by the noncovalent bonds (e.g., physical adsorption) were not stable and were easily affected by the external conditions (e.g., pH and temperature). To tackle this issue, our motivation is to covalently conjugate protein and oligonucleotide sequence on the magnetic bead for the development of a nanoparticle-based detection method without the participation of gold nanoparticles. Another important point for the nanoparticle-based assay is to achieve a high sensitivity with a low background signal. Rolling circle amplification (RCA), as a unique enzymatic process that can be used to generate extremely long singlestranded DNA with repeating sequences, is carried out with a short DNA primer and a circular DNA template under 7774

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

BaFe12O19 magnet with a pot shape (10 mm in diameter and 5 mm in depth, 410−430 mT). Finally, the poly(methacrylic acid)-coated magnetic beads (designated as PMA-MB) were dispersed into 10 mL of PBS (pH 6.0) for further use (C[MB] ≈ 25 mg mL−1). Preparation of Magnetic Sensing Probes (GA-MBDNA1,2). Magnetic sensing probes were prepared through a typical carbodiimide coupling. Initially, the poly(methacrylic acid)-coated magnetic beads prepared above (1.0 mL, C[MB] ≈ 25 mg mL−1) were diluted with 1.0 mL of Tris-HCl buffer solution (pH 7.4, 50 mM). Following that, 40 mM N-(3(dimethylamino)propyl)-N′-ethyl-carbodiimide hydrochloride and 10 mM N-hydroxysuccinimide were injected into the suspension. The resulting mixture was left to gently shake for 1 h at RT to activate the carboxyl group. Afterward, glucosamine aqueous solution (5.0 μL, 30 μM) and DNA1 (50 μL, 5 μM) were thrown into the mixture and incubated for 12 h at 4 °C on an end-over-end shaker (MS, IKA GmbH, Staufen, Germany). After completion of the reaction, the conjugates (designated as GA-MB-DNA1) were collected by using an external magnet. Finally, the obtained GA-MB-DNA1 was redispersed into 1 mL of Tris-HCl buffer solution (pH 7.4) for a subsequent experiment (C[MB] ≈ 25 mg mL−1). Conversely, glucosamine and DNA2-conjugated MB (designated as GA-MB-DNA2, C[MB] ≈ 25 mg mL−1) were synthesized and prepared by using the same conjugation protocol. Flow-Through Electrochemical Measurement. In this work, target Con A was assayed in a homemade magnetic detection cell with a sequential injection mode. (The schematic illustration is shown in our previous report.37) All electrochemical measurements were performed on a μAutoLab AUTIII.FRA2.v electrochemical workstation (Eco Chemie, The Netherlands) consisting of an O-ring gold-disk working work (2 mm in diameter), a platinum wire auxiliary electrode, and an Ag/AgCl reference electrode. The flow-through detection device was composed of a six-way connection valve equipped with a 1 mL syringe pump and connected through a Teflon tubing into the flow cell. The analytical flow stream entered from the other side into the center of the flow cell. The gold working electrode was installed at the bottom of the detection cell attached with a removable external magnet. All carrier buffers and samples were introduced in the detection cell at 0.5 mL min−1. Scheme 1 represents the electrochemical measurement protocol toward target Con A. All reactions, incubations, and measurements were carried out in the detection cell. The assay could be simply summarized as follows: (i) 100 μL of magnetic sensing probes including GA-MB-DNA1 and GA-MB-DNA2 (C[MB] ≈ 25 mg mL−1, 1:1) was initially flowed into the detection cell, and then, 100 μL of target Con A standards or samples with different concentrations in the presence of Mn2+ and Ca2+ ions was injected into the detection cell using a pipet and incubated for 30 min at 37 °C without the magnet. (Note: During this process, target-induced aggregation of GA-MBDNA1,2 was executed through the interaction between Con A and glucosamine. Meanwhile, the three-way DNA junction between DNA1 and DNA2 was fulfilled owing to partial base pairing.) (ii) 100 μL of RCA reaction buffer including 1 nM padlock DNA, 2.0 unit mL−1 phi29 DNA polymerase, and 500 mM dNTP [40 mM, pH 7.4 Tris-HCl buffer, 50 mM KCl, 10 Mm MgCl2, and 5 mM (NH4)2SO4)] was introduced into the detection cell and incubated for 60 min at 37 °C for the progression of RCA reaction without the magnet. (iii) 100 μL

active methylene blue with DNA backbone, and (iv) electrochemical measurement into a homemade magnetic detection cell with a sequential injection mode. The signal is amplified by the intercalated indicator within the RCA product. With the increasing target concentration, the formed primers increase, thereby resulting in the amplification of electrochemical signal. By monitoring the change in the signal, we might quantitatively determine the concentration of target analyte in the sample.

■

EXPERIMENTAL SECTION Materials and Reagents. All oligonucleotides used in this work were synthesized by Beijing Dingguo Biotechnol. Co., Ltd. (Beijing, China), which were purified by HPLC and confirmed by mass spectrometry. DNA stock solution was obtained by dissolving oligonucleotides in Tris-HCl buffer solution (pH 7.4). Each oligonucleotide was heated to 90 °C for 5 min and slowly cooled down to room temperature (RT) before use. The sequences of oligonucleotides are listed as follows: DNA 1 : 5′-NH 2 -TTTTTTTTTTTTTTTTGTGAGGGAACGGTCCTTG-3′ DNA 2 : 3′-NH 2 -TTTTTTTTTTTTTTTTCACTCCCTGCCAGATTAG-5′ Padlock DNA: 5′-p-GACGGTCTAATCCCATGACTTTGGGTAGGGCGGGTTGGGCTTTCTTTCCAAGGACCGTTC-3′ The 5′ end of padlock DNA was modified with the phosphate group. The underlined sequence of DNA1 was complementary to the underlined sequence of DNA2. The primer for RCA development consisted of the bold letters of DNA1 and DNA2, which was partially complementary to the bold letters of padlock DNA, respectively. Concanavalin A (Con A) from Canavalia ensiformis (Jack bean) (Type VI, lyophilized power, MW 26.5 kDa) was obtained from SigmaAldrich (Shanghai, China). D-(+)-Glucosamine hydrochloride (GA) was received from Sinopharm Chem. Re. Co., Ltd. (Beijing, China). Monomer methacrylic acid (99 wt %) was purchased from Aladdin (Shanghai, China). Magnetic Fe3O4 nanoparticles (particle size: 100 nm) in an aqueous suspension with a concentration of 25 mg mL−1 were obtained from Chemical GmbH (Berlin, Germany). All other reagents were of analytical grade and used as received without further purification. Ultrapure water obtained from a Millipore water purification system (≥18 MΩ, Milli-Q, Millipore) was used in all runs. Phosphate-buffered saline (PBS, 0.1 M) solution with various pH values was prepared by mixing the stock solutions of 0.1 M NaH2PO4 and 0.1 M Na2HPO4, and 0.1 M KCl was added as the supporting electrolyte. Synthesis of Poly(methacrylic acid)-Coated Magnetic Beads. Before functionalization with methacrylic acid, magnetic beads were separated by using an external magnet and dried in the vacuum at 80 °C for 1 h. Methacrylic acidmodified magnetic beads were prepared according to the literature with a little modification.36 Initially, 250 mg of magnetic beads and 575 mg of sodium dodecyl sulfate were added into 100 mL of distilled water in a 200 mL flask under mechanical stirring. Following that, the mixture was heated to 70 °C, and 0.48 mL of methacrylic acid (99 wt %) was introduced into the flask. After reaction for 45 min, 0.99 g of K2S2O8 was injected into the resulting mixture under the protection of nitrogen and polymerized for 2 h at 70 °C. Afterward, the resulting suspension was cooled to RT, washed with distilled water, and purified by using a permanent 7775

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

Figure 1. (A) UV−vis adsorption spectra of (a) magnetic beads, (b) PMA-MB, and (c) GA-MB-DNA1,2 (Insets: Zeta potential profiles of MB and PMA-MB and the corresponding photographs after storing for different times at RT). (B) DLS data of GA-MB-DNA1,2 before (top) and after (bottom) reaction with excess Con A. (C−E) Typical TEM images of (C) MB, (D) GA-MB-DNA1,2, and (E) GA-MB-DNA1,2 after reaction with 1.96 pM Con A (Insets: the corresponding magnified images).

DNA2. In the presence of target Con A, GA-MB-DNA1 and GA-MB-DNA2 are ligated together through the interaction between the labeled glucosamine on the MB and target Con A. In this case, DNA1 and DNA2 on the magnetic beads are brought into close proximity to form a three-way DNA junction owing to partial base pairing. The noncomplementary bases away from the magnetic beads can form an entire primer. With the help of T4 ligase and phi29 polymerase, the formed primer can trigger the RCA reaction by the template padlock DNA, thereby resulting in the formation of a long single-strand DNA on the magnetic bead. Upon addition of electroactive methylene blue, large numbers of methylene blue molecules are intercalated into DNA backbone. During the electrochemical measurement, each of the methylene blue molecules produces an electrochemical signal (current) within the applied potentials. The signal increases with the increasing target concentration in the sample. By monitoring the change in the electrochemical signal, we can indirectly determine the concentration of target analyte. Importantly, the conjugated DNA1 or DNA2 on the magnetic bead can not trigger the RCA reaction in the absence of target Con A because of the vacancy of the primer. To realize our design, one precondition for the successful development of RCA-based PLA was whether oligonucleotide and glucosamine were conjugated onto the magnetic bead. To demonstrate this issue, we used UV−vis absorption spectroscopy (UV 1102, Techcomp, China) to investigate the magnetic beads before and after modification with oligonucleotide and glucosamine (Figure 1A). As seen from curve “a”, the absorption spectra of the as-prepared magnetic beads were complicated and increased with the decrease of wavelength in the range of 400−200 nm. Moreover, the spectra were slightly changed when methacrylic acid was polymerized onto the

of methylene blue aqueous solution (0.5 mM) was thrown into the detection cell and incubated for 30 min at RT. (Note: During this process, the positively charged methylene blue was intercalated with negatively charged DNA backbone.) (iv) The collected magnetic sensing probes on the gold working electrode with an external magnet were monitored in pH 7.0 PBS by using square wave voltammetry (SWV) from −400 to −0.1 mV (vs. Ag/AgCl) (Amplitude: 25 mV; Frequency: 15 Hz; Increase E: 4 mV). After each step, the magnetic sensing probes were washed with pH 7.4 PBS in the presence of the magnet. All incubations and reactions were carried out at the stopped-flow status without the magnet. Electrochemical measurements were performed with the aid of the external magnet. Analyses were always made in triplicate.

■

RESULTS AND DISCUSSION Design and Characteristics of PLA with Three-Way Junction-Induced RCA. To achieve a high-performance electrochemical sensing platform, design and preparation of the sensing probes are very important. In this work, the magnetic beads not only act as the substrate for the conjugation of oligonucleotides and glucosamine but also enable the rapid separation and purification of bionanocomposites after synthesis. Oligonucleotides and glucosamine are covalently immobilized onto the poly(methacrylic acid)-coated magnetic beads by using a typical carbodiimide coupling. The conjugated glucosamine is utilized for the detection of target Con A, while the oligonucleotides are employed for RCA development. To fulfill target-induced PLA with the formation of a three-way DNA junction for RCA progression, oligonucleotide sequences of DNA1 and DNA2 are partially complementary each other (6 nt). Meanwhile, the primer for RCA development can consist of 12 bases at the 3′ end of DNA1 and 12 bases at the 5′ end of 7776

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

Figure 2. (A) Nyquist diagrams for (a) bare gold working electrode, (b) GA-MB-DNA1,2-modified electrode, (c) probe “b” after reaction with 19.6 pM Con A, (d) probe “c” after the RCA reaction in 5 mM Fe(CN)64−/3− + 0.1 M KCl with the range from 10−2 Hz to 106 Hz at an alternate voltage of 5 mV. (B) Cyclic voltammograms of GA-MB-DNA1,2-modified electrode after reaction with the following components: (a) GA-MB-DNA1,2 + 19.6 pM Con A + RCA + methylene blue, (b) GA-MB-DNA1,2 + 0 pM Con A + RCA + methylene blue, (c) GA-MB-DNA1,2 + methylene blue, and (d) GA-MB-DNA1,2 + 19.6 pM Con A + methylene blue, in pH 7.0 PBS at 50 mV s−1. (C) SWV curves of the GA-MB-DNA1.2-modified electrode after incubation with (a) 19.6 pM Con A + Fc-DNA and (b) 0 pM Con A + Fc-DNA in pH 7.0 PBS, respectively. (D) SWV curves of (a) GA-MBDNA1.2 + 19.6 pM Con A + RCA + Fc-DNA and (b) GA-MB-DNA1.2 + 0 pM Con A + RCA + Fc-DNA in pH 7.0 PBS, respectively.

development of the flow-through nanoparticle-based sensing platform. Another important precondition was whether target analyte could induce the proximity ligation of GA-MB-DNA1 and GAMB-DNA2. To verify this point, we used dynamic light scattering (DLS, Malvern Zetasizer 3000 HS, UK) and transmission electron microscopy (TEM, H-7650, Hitachi Instrument, Japan) to investigate the as-prepared GA-MBDNA1,2 in the absence and presence of Con A. As seen from the DLS analysis in Figure 1B, the mean size (≈378.7 nm) of the as-prepared GA-MB-DNA1,2 in the presence of target Con A was obviously larger than that in the absence of target Con A (≈100.3 nm), indicating that target Con A could make the GAMB-DNA1,2 aggregate together. Meanwhile, we also noticed that magnetic beads before (Figure 1C) and after (Figure 1D) modification with glucosamine and oligonucleotides could homogeneously disperse in the distilled water. However, the conjugated glucosamine and oligonucleotides were difficult to observe on the surface of magnetic beads owing to the small biomolecules. Significantly, when 1.96 pM Con A was added into the incubation solution including GA-MB-DNA1,2, partial GA-MB-DNA1,2 conjugates were aggregated together (Figure 1E). The results preliminarily revealed that the target analyte could be brought into close proximity to the functionalized magnetic beads. Elucidation of Target-Induced Proximity Ligation Assay Protocol. In this work, the assay mainly consists of three protocols: target-induced PLA, PLA-assisted formation of

magnetic beads, which was in agreement with that of pure methacrylic acid (curve “b”).38 To further monitor the functionalized magnetic beads, the as-prepared PMA-MB before and after modification with methacrylic acid was investigated by microelectrophoresis (Microelectrophoresis Apparatus Mk II, Rank Brothers Ltd., England). As shown from the inset of Figure 1A, the zeta potentials were −9.3 and −14.7 mV for the MB and PMA-MB, respectively. The more negative potential might be ascribed to the immobilized methacrylic acid on the particle surface. Significantly, one obvious characteristic peak at 260 nm (characteristic peak of oligonucleotides) was achieved when oligonucleotide and glucosamine were covalently conjugated onto the magnetic beads (curve “c”). However, we did not observe the absorption peak of glucosamine (i.e., 254 nm). The reason might be most likely as a consequence of the fact that the amount of the conjugated glucosamine on the magnetic beads was relatively less than that of oligonucleotides. More importantly, it can also be seen that unmodified magnetic beads were more easily aggregated (left photograph in Figure 1A, storing for 30 min at RT for the newly prepared magnetic beads) than that of oligonucleotide and glucosamine-modified magnetic beads (right photograph in Figure 1A, storing for 48 h at RT for the newly prepared GA-MB-DNA1,2), indicating that the highly negatively charged oligonucleotides could help stabilize the magnetic beads in the end. The long-time and homogeneous dispersion of the GA-MB-DNA1,2 was conducive for the 7777

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

Figure 3. Effects of (A) the molar ratio of glucosamine and oligonucleotide, (B) incubation time for glucosamine-Con A reaction, and (C) RCA reaction time on the developed sensing platform.

of the GA-MB-DNA1,2 after incubation with 19.6 pM Con A and methylene blue without RCA reaction. As shown from curve “d”, the peak currents were very close to those of curve “b” and curve “c”. More importantly, the peak currents (curve “d”) were largely lower than that of curve “a”. The results suggested that (i) target Con A could not adsorb the methylene blue in solution and (ii) the RCA could cause the amplification of detectable signal. Thus, our designed proximity ligation assay could be triggered by target analyte. Elucidation of PLA-Induced Formation of Three-Way DNA Junction. To further demonstrate that the signal was derived from the three-way junction-induced RCA reaction, the successful formation of the three-way DNA junction was completely necessary. To verify this point, we designed a referenced ferrocene (Fc)-labeled DNA strand (Fc-DNA: 5′Fc-TTTCCGTTCGACGGT-3′), which was used as the trace tag for three-way DNA junction. Theoretically, in the presence of target analyte, the formed primer by the target-induced three-way DNA junction paired with the Fc-DNA. The carried ferrocene molecule could produce the corresponding electrochemical signal. Initially, the prepared GA-MB-DNA1,2 was incubated with 1.96 pM Con A, and then, the resulting magnetic sensing probes were reincubated with the Fc-DNA. As indicated from curve “a” in Figure 2C, a strong electrochemical signal could be acquired at +320 mV in pH 7.0 PBS, which mainly originated from the labeled ferrocene on the hairpin probe. For comparison, magnetic sensing probes were also utilized for the detection of zero analyte. As seen from curve “b” in Figure 2C, no peak current was achieved in pH 7.0 PBS. The results also indicated that the Fc-DNA could be conjugated onto the magnetic bead in the presence of target analyte. A mostly possible explanation for this issue was that the developed PLA could induce the formation of the three-way DNA junction, and the formed junction could be used as the template for the hybridization of Fc-DNA. Validation of Three-Way Junction-Induced Rolling Circle Amplification. As described above, the electrochemical signal was amplified through the intercalated methylene blue in the RCA product. In this case, two different sensing protocols were investigated with and without the RCA by using the above-designed Fc-DNA as the trace tag (1.96 pM Con A used as an example). As seen from curve “a” in Figure 2D, magnetic sensing probes without the RCA exhibited only a weak current response (86 nA). In contrast, a high peak current was achieved by using the RCA as the signal amplification strategy (221 nA, curve “b”). Relative to curve “a”, the use of RCA could cause a 256.9% signal increase of the sensing platform. This is most

a three-way DNA junction, and a junction-triggered RCA reaction. The electrochemical signal was amplified through the intercalated methylene blue in the RCA product. To smoothly implement the assay procedure, target-induced close proximity of GA-MB-DNA1 and GA-MB-DNA2 should be the most key step. Prior to validation, we first investigated the whole process of magnetic sensing probes for the detection of 19.6 pM Con A after each step by using electrochemical impedance spectroscopy (EIS) (Figure 2A). Curve “a” shows the EIS of the cleaned gold working electrode. It can be seen that gold electrode exhibited an almost straight line that was characteristic of a diffusional limiting step of the electrochemical process. When magnetic sensing probes were simultaneously immobilized on the gold substrate, the resistance largely increased (720 Ω, curve “b”), implying that the as-prepared GA-MB-DNA1,2 was not conducive for the transfer of the negatively charged [Fe(CN)6]3−/4− from the solution to the base electrode. Moreover, the resistance further increased when magnetic sensing probes reacted with 19.6 pM Con A (1180 Ω, curve “c”). As expected, we also observed that the resistance successfully increased after the RCA reaction (2300 Ω, curve “d”). Such an increase in the resistance was basically attributed to the electrostatic repulsion between [Fe(CN)6]3−/4− redox couple and the negatively charged DNA backbone. Logically, one puzzling question arises as to whether the added target analyte could induce the formation of a three-way DNA junction for the development of RCA reaction. To demonstrate this concern, the as-prepared magnetic sensing probes were also employed for the detection of 19.6 pM Con A and zero analyte with cyclic voltammetry, respectively, by coupling with the three-way junction-induced RCA strategy. As seen from curve “a” in Figure 2B, a couple of strong redox peaks at −228 and −276 mV was observed in pH 7.0 PBS in the presence of 19.6 pM Con A. The redox peaks were mainly derived from the intercalated methylene blue in the DNA molecules. In the absence of target Con A, favorably, the peak currents (curve “b”) were almost the same as those of magnetic sensing probes (curve “c”), indicating that the two partially complementary oligonucleotides could not form a duplex upon mixing in solution to induce the RCA reaction. The reason might be the facts that (i) the conjugated oligonucleotides on the MB had no ability to have close proximity in the absence of target analyte and form the three-way DNA junction for RCA development and (ii) the slight change in the peak current might be ascribed to the conjugated oligonucleotides on the MB toward the absorption of a few methylene blue molecules. For comparison, we also investigated the cyclic voltammogram 7778

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

Figure 4. (A) SWV curves of the developed sensing platform toward Con A with various concentrations in pH 7.0 PBS and (B) the corresponding calibration curve.

standard deviation for 11 determinations of blank solution and K is the slope of the calibration plot). The precision of the electrochemical sensor was assessed by estimating the variation coefficients (CVs) of intra- and interassays. The intra-assay precision of the analytical method was evaluated by analyzing 4 concentrations, 5 times per run. The CVs were 5.6%, 7.4%, and 8.9% at 49, 1.96, and 0.0196 nM Con A, respectively. Similarly, the interassay CVs using different GA-MB-DNA with various batches were 9.7, 8.4%, and 9.5% with the above-mentioned analyte. The specificity of the newly developed method was evaluated by challenging it against other biomolecules or proteins. The results are shown in Figure 5. With an amount of nontarget

likely a consequence of the fact that the hybridized amount of hairpin DNA on one RCA product was much more than that of one three-way DNA junction. Importantly, every primer formed on the MB could trigger the RCA reaction, thereby resulting in the intercalation of numerous indicators and enhancing the electrochemical signal. Optimization of Experimental Conditions. To achieve an optimal analytical performance, some key parameters including the conjugation ratio between oligonucleotide and glucosamine, the incubation time for glucosamine-Con A, and RCA reaction time should be investigated in detail. Owing to the coconjugation of glucosamine and oligonucleotide on the MB, the conjugated ratio should be one of the most important factors influencing the sensitivity of the magnetic sensing platform. Usually, a highly carried amount of glucosamine on the MB could increase the capture ability of target analyte, but it was not conducive to RCA development. As seen from Figure 3A, the maximum electrochemical signal could be obtained at the molar ratio of 3:5. Thus, 5.0 μL of glucosamine aqueous solution (30 μM) and 50 μL of oligonucleotide (5 μM) were used for the preparation of GA-MB-DNA (1.0 mL, 25 mg mL−1). Figure 3B shows the effect of different incubation times for the glucosamine-Con A reaction on the electrochemical signal of magnetic sensing probes. The electrochemical signal increased with the increasing incubation time from 20 to 60 min and then tended to level off. To save assay time, 60 min was chosen for the reaction between glucosamine and target Con A. Conversely, we also investigated the effect of RCA time on the electrochemical signal of the developed method. As indicated in Figure 3C, the optimal RCA time was 60 min. Longer RCA time did not result in the significant increase of the electrochemical signal. Thus, 60 min was selected for RCA development in this study. Analytical Performance. Under the optimal conditions, the analytical properties of the developed electrochemical sensing method were investigated and evaluated by assaying routine Con A samples with various concentrations based on the designed assay protocol. As seen from Figure 4A, the electrochemical signals increased with the increasing Con A concentration in the sample. A linear dependence between the peak current and the logarithm of Con A concentration was obtained in the range from 1.96 pM to 98 nM (Figure 4B). The linear regression equation was i = 0.1952 × C[Con A] + 8.4578 (nM, R2 = 0.9966, n = 33) with a limit of detection (LOD) of 1.5 pM (estimated from the expression of 3S/K, where S is the

Figure 5. Specificity of the developed electrochemical sensing platform against bovine serum albumin (BSA), horseradish peroxidase (HRP), glucose oxidase (GOx), and target Con A.

molecules, such as BSA, HRP, and GOx, no apparent change in the current was observed in comparison with that of the blank test. However, the presence of target analyte resulted in the dramatic increase in the electrochemical signal, indicating the specificity of magnetic sensing probes. Analysis of Real Samples. To test the applicable possibility of the designed magnetic sensing probes for real samples, two types of sample matrices including newborn cattle serum and peanuts were used for spiking Con A standards with different concentrations. The obtained samples were measured by using the developed strategy. The results are shown in Table 1. As seen from Table 1, the recovery in all the cases was lower than 135%. The results suggested that the developed method could be considered as an optional scheme for the detection of target Con A in real samples by coupling with the target7779

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

Notes

Table 1. Recovery Evaluation for Con A Spiked Samples in Newborn Cattle Serum and Peanuts by Using the PLA-Based Electrochemical Sensing Platform substrate newborn cattle serum

blank peanut sampleb

sample no.

spiked (nM)

found (nM)a

recovery (%)

1 2 3 4 5 6 7 8 9 10 11 12

0.0098 0.098 0.98 9.8 49 73.5 0.0098 0.098 0.98 9.8 49 73.5

0.0123 0.087 1.03 11.4 46.5 82.3 0.0119 0.132 0.873 12.4 53.8 70.2

125.5 88.8 105.1 116.3 94.9 111.9 121.4 134.7 89.1 126.5 109.8 95.5

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by the National “973” Basic Research Program of China (2010CB732403), the National Natural Science Foundation of China (41176079, 21125524), the National Science Foundation of Fujian Province (2011J06003), and the Program for Changjiang Scholars and Innovative Research Team in University (IRT1116).

■

a The average value for three measurements. bThe blank peanut samples were made available as aqueous slurries, and the spiking process was described in detail in our previous reports.39,40

induced proximity ligation assay accompanying three-way DNA junction-assisted rolling circle amplification.

■

CONCLUSIONS In summary, we for the first time demonstrate the ability of magnetic bead-based sensing probes for highly sensitive electrochemical detection of Con A by coupling with a sequential injection mode. The signal was amplified through target-induced proximity of magnetic sensing probes accompanying formation of a three-way DNA junction for RCA reaction. Compared with traditional amplification strategies, the method is sensitive and feasible without the need of bioactive enzymes, thus representing an excellent isothermal cycling signal-amplification protocol. In contrast, thermal cycling-based amplification methods, e.g., polymerase chain reaction (PCR), usually need repeat thermal cycling in an exponential way and thereby are time-consuming and limited to a thermostable enzyme and a laboratory setting. Meanwhile, the nanoparticlebased sensing probes are easily controlled by using an external magnetic switch. Attraction of the functionalized magnetic beads to the electrode surface with the aid of an external magnet activates the electrical contact between the immobilized biomolecules and the base electrode, and the sensor’s circuit is switched on. Positioning the magnet away from the detection cell retracts the magnetic sensing probes from the substrate surface, and the circuit is switched off. Importantly, targetinduced formation of the detectable signal tag through the proximity ligation assay can decrease the background signal to some extent and achieve a high signal-to-noise ratio for enhancing the assay sensitivity. By controlling and tuning the conjugated protein to immobilize other biomolecules (e.g., streptavidin, antigen, or antibody and specific oligonucleotide), the assay can be easily extended for use with other target analytes, thus representing a versatile detection scheme.

■

REFERENCES

(1) Lis, H.; Sharon, N. Chem. Rev. 1998, 98, 637−674. (2) He, X.; Wang, X.; Jin, X.; Zhou, H.; Shi, X.; Chen, G.; Long, Y. J. Am. Chem. Soc. 2011, 133, 3649−3657. (3) Dwek, R. Chem. Rev. 1996, 96, 683−720. (4) Ritchie, G.; Moffatt, B.; Sim, R.; Morgan, B.; Dwek, R.; Rudd, P. Chem. Rev. 2002, 102, 305−319. (5) Roth, J. Chem. Rev. 2002, 102, 285−303. (6) Zachara, N.; Hart, G. Chem. Rev. 2002, 102, 431−438. (7) Jelinek, R.; Kolusheva, S. Chem. Rev. 2004, 104, 5987−6015. (8) Cella, L.; Chen, W.; Myung, N.; Mulchandani, A. J. Am. Chem. Soc. 2010, 132, 5024−5026. (9) Yonzon, C.; Jeoung, E.; Zou, S.; Schatz, G.; Mrksich, M.; Duyne, R. J. Am. Chem. Soc. 2004, 126, 12669−12676. (10) Samanta, A.; Stuart, M.; Ravoo, B. J. Am. Chem. Soc. 2012, 134, 19909−19914. (11) Gou, Y.; Slavin, S.; Geng, J.; Voorhaar, L.; Haddleton, D.; Becer, C. ACS Macro Lett. 2012, 1, 180−183. (12) Zhang, H.; Zhang, L.; Liang, R.; Huang, J.; Qiu, J. Anal. Chem. 2013, 85, 10969−10976. (13) Carrotta, R.; Vilasi, S.; Librizzi, F.; Martorana, V.; Bulone, D.; San Biagio, P. J. Phys. Chem. B 2012, 116, 14700−12707. (14) Hussein, F.; Daniels, R. Biomed. Anal. 2007, 43, 758−762. (15) Kamada, Y.; Kinoshita, N.; Tsuchiya, Y.; Kobayashi, K.; Fujii, H.; Terao, N.; Kamihagi, K.; Koyama, N.; Yamada, S.; Daigo, Y.; Nakamura, Y.; Taniguchi, N.; Miyoshi, E. Clin. Chim. Acta 2013, 417, 48−53. (16) Wang, X.; Matei, E.; Gronenborn, A.; Ramström, O.; Yan, M. Anal. Chem. 2012, 84, 4248−4252. (17) Tsai, C.; Yu, T.; Chen, C. Chem. Commun. 2005, 41, 4273− 4275. (18) Babu, P.; Sinha, S.; Surolia, A. Bioconjugate Chem. 2007, 18, 146−151. (19) Jiménez-Castells, C.; Defaus, S.; Moise, A.; Przbylski, M.; Andreu, D.; Gutiérrez-Gallego, R. Anal. Chem. 2012, 84, 6515−6520. (20) Bellapadrona, G.; Tesler, A.; Grünstein, D.; Hossain, L.; Kikkeri, R.; Seeberger, P.; Vaskevich, A.; Rubinstein, I. Anal. Chem. 2012, 84, 232−240. (21) Zhang, B.; Liu, B.; Liao, J.; Chen, G.; Tang, D. Anal. Chem. 2013, 85, 9245−9252. (22) Gunnarsson, A.; Sjovall, P.; Hook, F. Nano Lett. 2010, 10, 732− 737. (23) Thaxton, C.; Elghanian, R.; Thomas, A.; Stoeva, S.; Lee, J.; Smith, N.; Schaeffer, A.; Klocker, H.; Horninger, W.; Bartsch, G.; Mirkin, C. Proc. Natl. Acad. Sci. U.S.A. 2009, 3, 18437−18442. (24) Lai, W.; Tang, D.; Zhuang, J.; Chen, G.; Yang, H. Anal. Chem. 2014, 86, 5061−5068. (25) Gao, Z.; Xu, M.; Hou, L.; Chen, G.; Tang, D. Anal. Chem. 2013, 85, 6945−6952. (26) Nam, J.; Wise, A.; Groves, J. Anal. Chem. 2005, 77, 6985−6988. (27) Lee, H.; Park, J.; Nam, J. Nat. Commun. 2014, 5, 3367. (28) Hamblin, G.; Carneiro, K.; Fakhoury, J.; Bujold, K.; Sleiman, H. J. Am. Chem. Soc. 2012, 134, 2888−2891. (29) Liu, H.; Li, L.; Duan, L.; Wang, X.; Xie, Y.; Tong, L.; Wang, Q.; Tang, B. Anal. Chem. 2013, 85, 7941−7949.

AUTHOR INFORMATION

Corresponding Authors

*Phone: +86-591-2286 6125. Fax: +86-591-2286 6135. E-mail: [email protected]. *E-mail: [email protected]. 7780

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781

Analytical Chemistry

Article

(30) Russell, C.; Welch, K.; Jarvius, J.; Cai, Y.; Brucas, R.; Nikolajeff, F.; Svedlindh, P.; Nilsson, M. ACS Nano 2014, 8, 1147−1153. (31) Gao, X.; Hannoush, R. J. Am. Chem. Soc. 2014, 136, 4544−4550. (32) Hu, J.; Wang, T.; Kim, J.; Shannon, C.; Easley, C. J. Am. Chem. Soc. 2012, 134, 7006−7072. (33) Goodman, R.; Schaap, I.; Tardin, C.; Erben, C.; Berry, R.; Schmidt, C.; Turberfield, A. Science 2005, 310, 1661−1665. (34) Boer, D.; Kerckhoffs, J.; Parajo, Y.; Pascu, M.; Uson, I.; Lincoln, P.; Hannon, M.; Coll, M. Angew. Chem., Int. Ed. 2010, 49, 2336−2339. (35) Shu, D.; Shu, Y.; Haque, F.; Abdelmawla, S.; Guo, P. Nat. Nanotechnol. 2011, 6, 658−667. (36) Shi, Y.; Gan, M. J. Mater. Chem. 2004, 14, 2781−2786. (37) Li, Q.; Tang, D.; Tang, J.; Su, B.; Chen, G.; Wei, M. Biosens. Bioelectron. 2011, 27, 153−159. (38) Azhgozhinova, G.; Guven, O.; Pekel, N.; Dubolazov, A.; Mun, G.; Nurkeeva, Z. J. Colloid Interface Sci. 2004, 278, 155−159. (39) Tang, D.; Yu, Y.; Niessner, R.; Miro, M.; Knopp, D. Analyst 2010, 135, 2661−2667. (40) Tang, D.; Liu, B.; Niessner, R.; Li, P.; Knopp, D. Anal. Chem. 2013, 85, 10589−10596.

7781

dx.doi.org/10.1021/ac501690v | Anal. Chem. 2014, 86, 7773−7781