Biosensors and Bioelectronics 65 (2015) 71–77
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
One-step conjugation of aminoferrocene to phosphate groups as electroactive probes for electrochemical detection of sequence-specific DNA Qiong Hu a, Xianbao Deng a, Xuehua Yu a, Jinming Kong a,n, Xueji Zhang a,b,nn a b
School of Environmental and Biological Engineering, Nanjing University of Science & Technology, Nanjing 210094, PR China Chemistry Department, College of Arts and Sciences, University of South Florida, East Fowler Ave, Tampa, FL 33620-4202, United States
art ic l e i nf o
a b s t r a c t
Article history: Received 28 July 2014 Received in revised form 7 October 2014 Accepted 7 October 2014
A straightforward electrochemical DNA biosensing approach based on exploiting organometallic compound, aminoferrocene (AFC), as electroactive probes was firstly demonstrated, where the probes could be directly labeled to the free phosphate groups of the hybridized PNA/DNA heteroduplexes merely through one-step conjugation in the presence of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and imidazole. Briefly, mercapto-terminated peptide nucleic acid (PNA) was firstly immobilized onto gold electrode and used as the capture probes for the specific recognition of target single-stranded DNA (ssDNA). After hybridization, AFC probes were directly labeled to the free 5′-terminal phosphate groups, which were activated by EDC and imidazole, of the hybridized PNA/DNA heteroduplexes, and then they were exploited as the electroactive probes to monitor the hybridization. As the captured ssDNA was labeled with AFC in the stoichiometric ratio of 1:1, thus the electrochemical analysis of the proportionally labeled AFC based on differential pulse voltammetry (DPV) enabled a quantitative determination of sequence-specific DNA. Under optimal conditions, the approach presented a good linear relationship between the current intensities and logarithm of ssDNA concentrations in the range from 0.1 nM to 100 nM with a detection limit of 93 pM, and it rendered satisfactory analytical performance in serum samples. Furthermore, it exhibited excellent specificity toward single-nucleotide polymorphism (SNP) and precluded complicated protocols. More importantly, the simplicity of this approach together with its compatibility with standard micro-fabrication techniques makes it great potential in practical applications, especially in microarray areas where simple procedures are preferred. & Elsevier B.V. All rights reserved.
Keywords: Electrochemical DNA Organometallic compounds Aminoferrocene Single-nucleotide polymorphism Electrochemical impedance spectroscopy
1. Introduction Organometallic compounds are usually defined as a specific class of metal complexes containing at least one metal–carbon bond (Patra and Gilles, 2012). They are renowned for their remarkable applications in the fields including chemical biology, catalysis, and therapy because of their outstanding physicochemical properties such as structural diversity, chemical stability and unique photo- and electrochemical properties (Heinze and Lang, 2013; Noffke et al., 2012; Patra and Gilles, 2012). Presently, they have been preliminarily investigated as enzyme inhibitors and luminescent agents in chemical biology, while they have also been extensively exploited as stereoselective and asymmetric catalysts n
Corresponding author. Corresponding author at: School of Environmental and Biological Engineering, Nanjing University of Science & Technology, Nanjing 210094, PR China. E-mail addresses: [email protected] (J. Kong), [email protected] (X. Zhang). n
http://dx.doi.org/10.1016/j.bios.2014.10.015 0956-5663/& Elsevier B.V. All rights reserved.
in organic synthesis and as therapeutic agents in biomedicine (Che and Sun, 2011; Debreczeni et al., 2006; Feng et al., 2011; Heinze and Lang, 2013; Noffke et al., 2012; Patra and Gilles, 2012; Ringenberg and Ward, 2011). More interestingly, some findings have nicely revealed that a variety of organometallic compounds can effectively accumulate in specific compartments of cellular organelles due to their selective reactions with certain substances in such compartments or a consequence of hydrophobic/hydrophilic/electrostatic interactions (Barnard et al., 2006; Li et al., 2011; Murphy et al., 2010; Yu et al., 2008). For example, an Ir(III) complex with selective turn-on phosphorescence for nuclei in living cells has been reported, it tends to specifically stain nuclei through luminescence enhancement by interaction with nucleic acids via its rapid reaction with histidine/histidine-containing proteins (Li et al., 2011). In addition, the conjugation of metal complexes, especially organometallic compounds, to biomolecules such as carbohydrates, polypeptides, proteins, and DNA have also been intensively investigated over the past years (Kong et al., 2007,
72
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
2013). Thus, apart from the well-established applications as efficient luminescent probes for bioimaging (Clède et al., 2012; Meister et al., 2010; Patra and Gilles, 2012), they also show great potential as labeling probes in analytical chemistry (Fang et al., 2011; Liu et al., 2014). With the increasing improvement of living standards and the further deepening of globalization, urgent demands relevant to the early diagnosis of hereditary diseases in clinical medicines and the fast screening of pathogenic microorganisms in customs quarantine and food quality supervision have prompted the fast development of highly sensitive, highly selective, and even real-time biosensing approaches with superior specificity in differentiating sequence-specific DNA (Grabowska et al., 2014; Hu et al., 2015; Liu et al., 2013a, 2013b; Qian et al., 2015). Compared with conventional optical approaches, electrochemical approaches have drawn considerable attentions by virtue of their high sensitivity, uncomplicated instrumentation, low cost, good portability, and excellent compatibility with micro-manufacturing technology (Hansen et al., 2006; Hu et al., 2015; Liu et al., 2008; Qian et al., 2015). Generally, the quantitative analysis of DNA can be conveniently achieved via monitoring the proportionally labeled electroactive probes after hybridization. For example, owing to their favorable redox properties, ferrocene-based derivatives, a versatile family of stable organometallic compounds, have emerged as high-profile electroactive probes in the field of electrochemical DNA biosensing (Fang et al., 2011; Seiwert and Karst, 2008). To exploit ferrocene-based derivatives as electroactive probes, several available strategies have been reported. Examples include attempts to modify them to the terminus of molecular beacons (MBs) (Miao et al., 2014; Qian et al., 2014); labeling target ssDNA with them before/after hybridization through versatile coupling linkers (Fang et al., 2011); hybridization with the auxiliary signal probes that have been labeled with them (Wang et al., 2003; Wei et al., 2014), and so on. These strategies are efficient and have been extensively investigated. Nevertheless, complicated protocols for the preparation, separation and purification of the labeled ssDNA are tedious and time-consuming, which prevent its versatility in practical applications (Xu et al., 2001). To overcome the enumerated drawbacks, carbodiimides, especially 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC), can be exploited as the crosslinking mediators to directly label aminecontaining ferrocene to the captured target ssDNA among the electrochemical detection of sequence-specific DNA. Carbodiimides are capable of activating the carboxylate groups and phosphate groups efficiently to form highly reactive intermediates, while the latter can be immediately substituted by amine-containing nucleophiles (Hermanson, 2008). More importantly, the crosslinking between carboxylates/phosphates and amine-containing nucleophiles is nearly impossible without carbodiimides as the activation mediators due to their poor reactivity (Chu et al., 1983; Ghosh et al., 1990; Hermanson, 2008). Among all carbodiimides, EDC is the frequently used crosslinking mediator in bioconjugation, and this benefits most from its favorable watersolubility. Consequently, EDC shows great potential as the crosslinking mediator in mediating the direct conjugation of target ssDNA and amine-containing electroactive probes (Chu et al., 1983; Ghosh et al., 1990; Hermanson, 2008; Ralph et al., 1962). To the best of our knowledge, there has been no report about the electrochemical detection of sequence-specific DNA based on organometallic compounds merely through one-step conjugation. Herein, we demonstrated a straightforward electrochemical DNA biosensing approach, which was applicable to detect sequencespecific DNA, by exploiting organometallic compound, aminoferrocene (AFC), as electroactive probes for the first time. In this work, mercapto-terminated peptide nucleic acid (PNA) probes
were firstly immobilized onto the clean surface of the pretreated gold electrode via the formation of self-assembled monolayer (SAM). Then, they were utilized as the capture probes for the specific recognition of target ssDNA to be determined in the following step. After hybridization, AFC probes were directly labeled to the free 5′-terminal phosphate groups of the hybridized PNA/ DNA heteroduplexes merely through one-step conjugation in the presence of EDC and imidazole, subsequently, they were exploited as the electroactive probes to monitor the hybridization via differential pulse voltammetry (DPV). As the hybridized ssDNA was labeled with AFC in the stoichiometric ratio of 1:1, thus the electrochemical analysis of the proportionally labeled AFC enabled a reliable quantitative determination of sequence-specific DNA. Under optimal conditions, the approach presented a good linear relationship between the current intensities and logarithm of ssDNA concentrations in the range from 0.1 nM to 100 nM, and it rendered satisfactory analytical performance in serum samples. As the hybridized ssDNA could be effectively labeled with AFC merely through one-step conjugation, thus the approach demonstrated here was straightforward and labour-saving, which could effectively preclude complicated protocols accordingly. Results also revealed that it exhibited excellent specificity toward single-nucleotide polymorphism (SNP). More importantly, the simplicity of this approach together with its compatibility with standard microfabrication techniques make it great potential in practical applications, especially in microarray areas where simple procedures are preferred.
2. Material and methods 2.1. Materials and reagents Mercapto-terminated peptide nucleic acid with sequence 5′-HS-(CH2)11-AAC CAT ACA ACC TAC TAC CTC A-3′ (PNA) was custom-made by Panagene Inc. (South Korea). Target complementary ssDNA 5′-TGA GGT AGT AGG TTG TAT GGT T-3′ (cDNA), single-base mismatched ssDNA (mismatch underlined) 5′-TGA GGT AGT AGG TTG TGT GGT T-3′ (SBM), three bases mismatched ssDNA (mismatches underlined) 5′-TGA GGT ATT AGA TTG TGT GGT T-3′ (TBM), and completely non-complementary ssDNA 5′ACT TAC CTT TGC TCA TTG ACG A-3′ (Control) were all purchased from Invitrogen Biotechnology Co., Ltd. (Shanghai, China). 1-Ethyl3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC), and imidazole were purchased from J&K Scientific Ltd. (Shanghai, China). Aminoferrocene (AFC) was purchased from TCI (Shanghai, China). 6-Mercapto-1-hexanol (MCH) was purchased from SigmaAldrich (St. Louis, MO). Fetal bovine serum (FBS, Defined) was purchased from Shanghai YiJi Industrial Co., Ltd. (Shanghai, China). All other chemical reagents were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China) and used without further purification. Ultrapure water obtained from a Millipore Milli-Q water purification system (Z18.25 MΩ) was used in all experiments. Tris–EDTA buffer (TE, 10 mM Tris–HCl, 1 mM EDTA, pH ¼ 8.0) was prepared and utilized as the stocking solution and washing buffer for ssDNA, and various concentrations of ssDNA samples were prepared by serial dilution. 0.1 M imidazole solution was prepared by dissolving it in 80% ethanol solution with its pH adjusted by HCl solution and used as the buffer for the labeling of ssDNA. 2.2. Apparatus All electrochemical experiments were performed with a conventional three-electrode system consisted of modified gold
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
electrode (Φ ¼2 mm) as working electrode, saturated calomel electrode (SCE) as reference electrode and platinum wire as counter electrode. A CHI 760D electrochemical workstation (Shanghai, China) was used for differential pulse voltammetry (DPV) and cyclic voltammetry (CV) measurements. Electrochemical impedance spectroscopy (EIS) measurements were performed on an Autolab/PGSTAT30 (Eco Chemie, Netherlands). And all the electrochemical experiments were carried out at room temperature.
73
2.5. Electrochemical measurement To carry out the electrochemical measurement, the gold electrode was finally immersed in 10 mL of 1.0 M KNO3 solution (nitrogen-saturated), and then DPV was performed under the potential from 0 V to 0.5 V to record the oxidation current of the covalently labeled AFC (Fang et al., 2011). The scheme for electrochemical detection of sequence-specific DNA by exploiting AFC as electroactive probes through one-step conjugation is illustrated in Scheme 1.
2.3. Pretreatment of the gold electrode and immobilization of capture probes 3. Results and discussion Prior to modification, the gold electrode was polished to a mirror-like surface with 0.3 and 0.05 μm alumina slurry, and then successively ultrasonically cleaned with absolute ethanol and ultrapure water (Hu et al., 2015; Liu et al., 2013a, 2013b). Then, it was chemically cleaned in freshly prepared piranha solution (a mixture of 98% H2SO4 and 30% H2O2, 3:1 v/v, CAUTION: piranha solution is strongly corrosive and must be handled with extreme care) under ultrasonication (Hu et al., 2015; Liu et al., 2013a, 2013b). Afterwards, it was electrochemically cleaned in a fresh 0.5 M H2SO4 solution by cycling the electrode potential between 0.2 V and 1.5 V with a scan rate of 100 mV s 1 until a reproducible cyclic voltammogram was achieved to remove any remaining impurities (Hu et al., 2015; Liu et al., 2013a, 2013b). Finally, it was rinsed thoroughly with absolute ethanol and ultrapure water and dried with nitrogen prior to modification. Immobilization of capture probes was accomplished by incubating the clean electrode in 1 μM PNA aqueous solution to allow the formation of well-ordered SAM (Hu et al., 2015). After washing with ultrapure water, it was immersed in 2 mM MCH solution (dissolved in 70% ethanol solution) to passivate the possible nonspecific binding sites, and then successively rinsed with 70% ethanol solution and ultrapure water to remove excessive MCH (Hu et al., 2015; Liu et al., 2013a, 2013b). 2.4. DNA hybridization and labeling of electroactive probes Hybridization was performed in 10 μL of TE buffer that contained certain concentration of ssDNA at 37 °C for 1.5 h, followed by washing with TE buffer to remove the unhybridized ssDNA (Hu et al., 2015). After that, 20 μL of freshly prepared mixed solution of AFC and EDC with the concentration of EDC kept at 10 mg mL 1 (freshly dissolved in 0.1 M imidazole solution) was added on the electrode and then incubated at 25 °C for the labeling of electroactive probes. (The schematic diagram of covalent labeling of ssDNA with AFC in the presence of EDC and imidazole was shown in Fig. 1.) Finally, the resulting electrode was moderately washed with 80% ethanol solution and ultrapure water to remove the remaining reactants.
3.1. Electrochemical characterization of the covalently labeled AFC In the absence of imidazole, the water-soluble carbodiimide EDC alone tends to rapidly activate the free phosphate group at the 5′-terminus of ssDNA to form an active phosphomonoester intermediate, which is highly reactive to couple with amine-containing nucleophilic compounds (Ghosh et al., 1990; Ralph et al., 1962; Shabarova, 1988). However, EDC is labile in water and the resulting intermediate shows a short half-life in aqueous solution due to its hydrolysis (Hermanson, 2008). In addition, the generated phosphorimidazolide intermediate in the presence of imidazole provides better reactivity toward amine-containing nucleophiles than the corresponding EDC phosphomonoester intermediate (Hermanson, 2008). As a result, the supplement of imidazole to the reaction medium could significantly improve the derivatization yield of ssDNA labeled with AFC probes over the carbodiimide-only reactions (Hermanson, 2008). In this work, EDC and imidazole were simultaneously adopted as the synergistic crosslinking mediators to directly label AFC probes to the hybridized PNA/DNA heteroduplexes via the EDC-mediated one-step conjugation. To evaluate the feasibility and effectivity of the proposed strategy for labeling of target ssDNA with AFC probes merely through one-step conjugation in the presence of EDC and imidazole, DPV was used to characterize the presence of AFC probes on the electrode. As shown in Fig. 2A, an apparent oxidation peak would appear at the potential around 0.25 V provided that AFC probes had been covalently labeled to the hybridized heteroduplexes (Fig. 2A,a), the resulting oxidation potential was in agreement with the reported value (Miao et al., 2014; Shiddiky et al., 2010), and the oxidation current could be interpreted as the electrochemical oxidation of ferrocene into ferrocenium (Brisset et al., 2006; Heinze and Lang, 2013). On the contrary, there merely feeble background current due to the charging process would be observed without the labeling of AFC probes (Fig. 2A,b) or before the hybridization had been performed (Fig. 2A,c). The obtained
Fig. 1. Covalent labeling of ssDNA with AFC in the presence of EDC and imidazole.
74
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
Scheme 1. Schematic illustration of electrochemical detection of sequence-specific DNA by exploiting AFC as electroactive probes through one-step conjugation.
Fig. 2. Differential pulse voltammograms of the electrode after (a) and before (b) labeling of target ssDNA with AFC as well as before the hybridization (c) (A); the cyclic voltammograms of the labeled AFC probes on the electrode at scan rates from 10 to 200 mV s 1 (B), (Inset) The linear relationship been anodic (black line) and cathodic (red line) peak currents and scan rates toward 5 nM cDNA in 1.0 M KNO3; the Nyquist plots of EIS of the bare gold electrode (Au) (a), Au–PNA (b), Au–PNA–MCH (c), Au–PNA– MCH–DNA (d), and Au–PNA–MCH9DNA–AFC (e) (C).
results indicated that AFC probes had been successfully labeled to the free phosphate groups at the 5′-terminus of the hybridized ssDNA via the EDC-mediated one-step conjugation. More importantly, the oxidation current appeared only when the hybridization had been accomplished and AFC probes had been successfully conjugated to the hybridized PNA/DNA heteroduplexes, and it in turn strongly guaranteed the reliability of the response signal. Thus, the proposed strategy was quite feasible and effective. In addition, the covalently labeled AFC probes were further characterized by cyclic voltammetry (CV). As shown in Fig. 2B, cyclic voltammograms obtained for the labeled AFC probes at different scan rates from 10 to 200 mV s 1 exhibited that the ferrocene/ferrocenium redox process was quasi reversible at low scan rates, and then it became more irreversible parallel with the further increasing of scan rate (Grabowska et al., 2014). In addition, the linear relationship between the anodic and cathodic peak currents and scan rates further indicated that the redox process was not diffusion dependent, and thus confirmed that the AFC probes were confined to the electrode surface with a slow electron transfer (Brisset et al., 2006; Grabowska et al., 2014; Fan et al., 2003). Electrochemical impedance spectroscopy (EIS) is an effective, informative, and amenable technique to examine the microscopic interfacial changes (Shervedani and Pourbeyram, 2009). Here, it was performed to provide further evidence for the preparation process of the demonstrated approach by measuring the impedance change of the electrode surface during the modification procedure. The impedance was carried out within the frequency range of 0.01 Hz to 100 kHz in a solution of 0.1 M KNO3 containing 5 mM [Fe(CN)6]3 /4 as the redox probe at a potential of 0.18 V (vs. SCE), with a voltage amplitude of 5 mV. Fig. 2C shows the Nyquist plots of impedance spectra at different stages of the preparation process. The semicircle portion observed at high
frequencies corresponded to the electron transfer limited process and its diameter was equal to the electron transfer resistance, which could reflect the electron transfer kinetics of the redox probe at the electrode surface (Alfonta et al., 2001). The bare gold electrode exhibited the smallest semicircle at high frequency region due to the fast electron-transfer process (Fig. 2C,a) (Wang et al., 2011). After the immobilization of mercapto-terminated PNA, the resistance increased accordingly (Fig. 2C,b). This could be attributed to the steric effect of the well-ordered self-assembled monolayer (Liu et al., 2010). After MCH was employed to passivate the nonspecific binding sites, the resistance further increased (Fig. 2C,c). This was mainly because that the more densely packed monolayer rendered a relatively slow electron-transfer. Since the negatively charged phosphate backbone of ssDNA electrostatically repelled [Fe(CN)6]3 /4 from reaching the electrode surface (Liu et al., 2005; Noorbakhsh and Salimi, 2011), remarkable increase in the electron transfer resistance could be observed when the hybridization had been accomplished (Fig. 2C,d). Finally, the steric effect of the covalently labeled AFC probes resulted in a further enhancement in the electron transfer resistance (Fig. 2C,e). 3.2. Optimization of detection conditions of the electrochemical DNA biosensing approach To achieve the best analytical performance of the demonstrated electrochemical DNA biosensing approach, several experimental parameters associated with the current intensity, which was in turn dominated by the amount of AFC probes that had been labeled to the hybridized PNA/DNA heteroduplexes, were carefully investigated. Apart from the pH value, the labeling time and the concentration of AFC were also inspected. The activation of phosphate groups with EDC in aqueous solution occurs most effectively at acidic reaction medium, while it is also favorable for the formation of phosphoramidate linkages
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
75
Fig. 3. Effects of pH (A), labeling time (B), and AFC concentration (C) on the analytical performance of the electrochemical DNA biosensing approach toward 10 nM cDNA in 1.0 M KNO3.
(Nakajima and Ikada, 1995). However, EDC is susceptible to be violently hydrolyzed under the conditions (Nakajima and Ikada, 1995). Thus, the covalent labeling of ssDNA with AFC probes is indeed dominated by the competitive equilibrium process of these mechanisms mentioned above. Therefore, the effect of pH value on current intensity was firstly investigated. As shown in Fig. 3A, the current intensity increased drastically with the pH value varying from 4.8 to 6.0, and then it encountered a remarkable decrease with the further increasing of pH value. The obtained results indicated that the EDC-mediated formation of phosphoramidate linkage in the presence of imidazole benefited its maximal yield at pH 6.0. As a result, the 0.1 M imidazole solution at pH 6.0 was adopted for the labeling of ssDNA with AFC probes in the subsequent experiments. The analytical performance of the electrochemical DNA biosensing approach was largely depended upon the amount of AFC probes that had been covalently conjugated to the hybridized PNA/ DNA heteroduplexes, which was partially controlled by the labeling time. Admittedly, more labeling time resulted in more conjugated AFC probes, however, the biosensing time would also be extended accordingly. To optimize the analytical performance, especially the sensitivity, the effect of labeling time on the current intensity was further investigated. As shown in Fig. 3B, the current intensity increased significantly with the increasing of labeling time until 120 min, afterwards, no significant increase in current intensity could be observed with the further increasing of labeling time. Thus, it could be concluded that the optimal labeling time was 120 min, and was adopted in the following experiments.
Obviously, the concentration of AFC was another important parameter that would significantly affect the amount of the labeled electroactive probes, namely, the analytical performance could be further optimized by choosing an optimal AFC concentration. Meanwhile, the labeling time would be shortened accordingly. As shown in Fig. 3C, the current intensity increased rapidly with the increasing of AFC concentration until 1.4 mg mL 1 and then it trended to reach a plateau until 1.6 mg mL 1. From the result mentioned above, it could be concluded that the optimal concentration of AFC was 1.6 mg mL 1. In addition, as the concentration of AFC at 1.6 mg mL 1 is around 7.96 mM, which is much higher than the concentration of ssDNA to be determined, thus it can be accepted as the saturated concentration in the following experiments. 3.3. Analytical performance of the electrochemical DNA biosensing approach Under optimal conditions, the analytical performance of the electrochemical DNA biosensing approach was examined with varying cDNA concentrations. As shown in Fig. 4, the current intensities increased linearly along with the increasing concentration of cDNA. The calibration plot showed a good linear relationship between the current intensities and the logarithm of cDNA concentrations in the range from 0.1 nM to 100 nM. The linear regression equation was I (μA)¼ 0.284þ0.170 lg [CDNA/nM] with a correlation coefficient of 0.9984. The limit of detection at a signalto-noise ratio of 3 was calculated to be 93 pM, which was much lower than those previously reported approaches (Miao et al.,
Fig. 4. Differential pulse voltammograms (A), and calibration plot (B) of the electrochemical DNA biosensing approach toward varying concentrations of cDNA in the range from 0.1 nM to 100 nM.
76
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
was 3.8% for inter-assay. The results indicated that the demonstrated approach had acceptable repeatability; and it might be attributed to the relatively simple experimental procedures, which could effectively eliminate the potential interferences originated from the complicated protocols. In addition, the stability of the electrochemical DNA biosensing approach was further investigated by long-term storage assay. The modified electrode could be stored in a moisture-saturated environment at 4 °C, and over 93.4% of the initial response retained after a storage period of three weeks. The loss of response signal might be a consequence of the irreversible oxidation of the conjugated ferrocene. The results demonstrated that the approach possessed satisfactory stability. 3.5. Application in analysis of serum samples
Fig. 5. DPV responses of the electrochemical DNA biosensing approach toward four types of 10 nM ssDNA.
2014; Xu et al., 2001; Watanabe et al., 2011; Wang et al., 2012). The improved sensitivity might be attributed to the excellent discrimination ability of DPV in distinguishing faradaic current from background current when the target analytes to be determined existed at ultralow level (Habibi et al., 2011). In addition, compared with other electrochemical DNA biosensing approaches, the labeling of electroactive probes in the demonstrated approach excluded complicated protocols and could be conveniently achieved merely through one-step conjugation. Therefore, the analytical performance of the demonstrated electrochemical DNA biosensing approach was acceptable. 3.4. Selectivity, repeatability, and stability of the electrochemical DNA biosensing approach The selectivity, namely specificity, of the electrochemical DNA biosensing approach was investigated by using a single-base mismatched ssDNA (SBM), three bases mismatched ssDNA (TBM), and completely non-complementary ssDNA (Control). As shown in Fig. 5, the current intensities from SBM, TBM, and Control were approximately 83.0%, 62.3%, and 7.9% of the value obtained from cDNA, respectively. The difference could be attributed to the hybridization efficiency of the four types of ssDNA with the immobilized capture probes and the prominent selectivity of PNA, as the cDNA could specifically and robustly bind to the complementary capture probes while the others couldn't (Hu et al., 2015). More importantly, the introduction of AFC probes to the system happened only when ssDNA had been successfully captured by the immobilized capture probes, because there were no additional possible reactive sites available for the binding of AFC probes. The obtained results highlighted that the novel approach demonstrated here could effectively differentiate complementary and mismatched oligonucleotide fragments, indicating that it was highly specific and even showed great potential for the genotyping of SNP. The excellent selectivity could be attributed to the utilization of highly specific PNA as the immobilized capture probes (Hu et al., 2015). The repeatability of the electrochemical DNA biosensing approach was evaluated by the intra-assay and inter-assay coefficients of variation. The intra-assay and inter-assay were independently conducted by measuring two identical samples both contained 5 nM cDNA under the same conditions. Each assay was repeatedly conducted 10 times using 10 parallelly prepared electrodes. The coefficient of variation of intra-assay was 4.9%, while it
To evaluate the analytical reliability and application potential of the electrochemical DNA biosensing approach, the interference effect of complicated serum samples on analytical performance was investigated. The current intensities originated from 5 nM cDNA in serum samples were compared with that obtained from 5 nM cDNA in TE buffer. The assay results from 10 repeated experiments are displayed in Fig. 6. The current intensities from 1% (B) and 5% (C) serum samples were approximately 94.4% and 88.4% of that from TE buffer (A), respectively. All these obtained results highlighted that the demonstrated approach showed acceptable analytical reliability and application potential for clinical applications, allowing for its good detectable capability in complicated systems such as samples in serum.
4. Conclusion In summary, a straightforward electrochemical DNA biosensing approach based on exploiting organometallic compound, aminoferrocene (AFC), as electroactive probes was firstly demonstrated, where the probes could be directly labeled to the free phosphate groups of the hybridized PNA/DNA heteroduplexes merely through one-step conjugation in the presence of EDC and imidazole. Using the mercapto-terminated PNA as the immobilized capture probes for the specific recognition of target ssDNA significantly improved the selectivity, leading to the great potential in the genotyping of SNP. The utilization of differential pulse voltammetric technique improved the sensitivity, enabled a detection limit of 93 pM
Fig. 6. DPV responses of the electrochemical DNA biosensing approach toward 10 nM cDNA in TE buffer (A), 1% serum sample (B), and 5% serum sample (C).
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
achieved under optimal conditions. In addition, it excluded complicated protocols for the monitoring of hybridization and could be successfully applied for the quantification of sequence-specific DNA in serum samples. The outstanding analytical performance including labour- and time-saving, low cost, satisfactory repeatability, acceptable stability and reliability, implicated that it showed great potential in clinical applications. More importantly, the demonstrated approach based on exploiting one-step conjugation of organometallic compounds to target analytes as electroactive probes in the presence of EDC and imidazole has presented a novel strategy for the electrochemical determination of target analytes that contained free carboxylate groups, phosphate groups, and even primary amines. Furthermore, it could be integrated with standard micro-fabrication techniques to develop microfluidic chips and realized multiplexed detection capabilities at the same time.
Acknowledgments We are grateful to Nanjing University of Science and Technology for its start-up funding and also National Natural Science Foundation of China (No. 21345002) for funding this project.
References Alfonta, L., Willner, I., Throckmorton, D.J., Singh, A.K., 2001. Anal. Chem. 73, 5287–5295. Brisset, H., Navarro, A.E., Spinelli, N., Chaix, C., Mandrand, B., 2006. Biotechnol. J. 1, 95–98. Barnard, P.J., Wedlock, L.E., Baker, M.V., Berners-Price, S.J., Joyce, D.A., Skeleton, B. W., Steer, J.H., 2006. Angew. Chem. Int. Ed. 45, 5966–5970. Chu, B.C.F., Wahl, G.M., Orgel, L.E., 1983. Nucleic Acids Res. 11, 6515–6529. Che, C.M., Sun, R.W.Y., 2011. Chem. Commun. 47, 9554–9560. Clède, S., Lambert, F., Sandt, C., Gueroui, Z., Réfrégiers, M., Plamont, M.A., Dumas, P., Vessières, A., Policar, C., 2012. Chem. Commun. 48, 7729–7731. Debreczeni, J.E., Bullock, A.N., Atilla, G.E., Williams, D.S., Bregman, H., Knapp, S., Meggers, E., 2006. Angew. Chem. Int. Ed. 45, 1580–1585. Fan, C., W. Plaxco, K., J. Heeger, A., 2003. Proc. Natl. Acad. Sci. USA 100, 9134–9137. Feng, L., Geisselbrecht, Y., Blanck, S., Wilbuer, A., Atilla-Gokcumen, G.E., Filippakopoulos, P., Krling, K., Celik, M.A., Harms, K., Maksimoska, J., Mar-morstein, R., Frenking, G., Knapp, S., Essen, L.-O., Meggers, E., 2011. J. Am. Chem. Soc. 133, 5976–5986. Fang, C., Ji, H., Karen, W.Y., Rafei, S.R.M., 2011. Biosens. Bioelectron. 26, 2670–2674. Ghosh, S.S., Kao, P.M., McCue, A.W., Chappelle, H.L., 1990. Bioconjug. Chem. 1, 71–76.
77
Grabowska, I., Singleton, D.G., Stachyra, A., Góra-Sochacka, A., Sirko, A., ZagórskiOstoja, W., Radecka, H., Stulz, E., Radecki, J., 2014. Chem. Commun. 50, 4196–4199. Hansen, J.A., Mukhopadhyay, R., Hansen, J.Ø, Gothelf, K.V., 2006. J. Am. Chem. Soc. 128, 3860–3861. Hermanson, G.T., 2008. Bioconjugate Techniques, 2nd ed. Academic Press, Amsterdam. Habibi, B., Jahanbakhshi, M., Pournaghi-Azar, M.H., 2011. Anal. Biochem. 411, 167–175. Heinze, K., Lang, H., 2013. Organometallics 32, 5623–5625. Hu, Q., Hu, W., Kong, J., Zhang, X., 2015. Biosens. Bioelectron. 63, 269–275. Kong, K.V., Chew, W., Lim, L.H.K., Fan, W.Y., Leong, W.K., 2007. Bioconjug. Chem 18, 1370–1374. Kong, K.V., Lam, Z., Lau, W.K.O., Leong, W.K., Olivo, M., 2013. J. Am. Chem. Soc. 135, 18028–18031. Liu, J., Tian, S., Nielsen, P.E., Knoll, W., 2005. Chem. Commun., 2969–2971. Liu, G., Wan, Y., Gau, V., Zhang, J., Wang, L., Song, S., Fan, C., 2008. J. Am. Chem. Soc. 130, 6820–6825. Liu, X., Qu, X., Fan, H., Ai, S., Han, R., 2010. Electrochim. Acta. 55, 6491–6495. Liu, Y.H., Li, H.N., Chen, W., Liu, A.L., Lin, X.H., Chen, Y.Z., 2013a. Anal. Chem. 85, 273–277. Liu, S., Wang, C., Zhang, C., Wang, Y., Tang, B., 2013b. Anal. Chem. 85, 2282–2288. Liu, S., Lin, Y., Wang, L., Liu, T., Cheng, C., Wei, W., Tang, B., 2014. Anal. Chem. 86, 4008–4015. Li, C., Yu, M., Sun, Y., Wu, Y., Huang, C., Li, F., 2011. J. Am. Chem. Soc. 133, 11231–11239. Murphy, L., Congreve, A., Palsson, L.-O., Williams, J.A.G., 2010. Chem. Commun. 46, 8743–8745. Meister, K., Niesel, J., Schatzschneider, U., Metzler-Nolte, N., Schmidt, D.A., Havenith, M., 2010. Angew. Chem. Int. Ed. 49, 3310–3312. Miao, X., Guo, X., Xiao, Z., Ling, L., 2014. Biosens. Bioelectron. 59, 54–57. Nakajima, N., Ikada, Y., 1995. Bioconjugate. Chem. 6, 123–130. Noorbakhsh, A., Salimi, A., 2011. Biosens. Bioelectron. 30, 188–196. Noffke, A.L., Habtemariam, A., M. Pizarro, A., Sadler, P., J., 2012. Chem. Commun. 48, 5219–5246. Patra, M., Gilles, G., 2012. ChemBioChem 13, 1232–1252. Qian, Y., Wang, C., Gao, F., 2014. Talanta 130, 33–38. Qian, Y., Wang, C., Gao, F., 2015. Biosens. Bioelectron. 63, 425–431. Ralph, R.K., Young, R.J., Khorana, H.G., 1962. J. Am. Chem. Soc. 84, 1490–1491. Ringenberg, M.R., Ward, T.R., 2011. Chem. Commun. 47, 8470–8476. Shabarova, Z.A., 1988. Biochimie 70, 1323–1334. Seiwert, B., Karst, U., 2008. Anal. Bioanal. Chem. 390, 181–200. Shervedani, R.K., Pourbeyram, S., 2009. Biosens. Bioelectron. 24, 2199–2204. Shiddiky, M.J., Torriero, A.A., Zeng, Z., Spiccia, L., Bond, A.M., 2010. J. Am. Chem. Soc. 132, 10053–10063. Wang, J., Li, J., Baca, A.J., Hu, J., Zhou, F., Yan, W., Pang, D.-W., 2003. Anal. Chem. 75, 3941–3945. Wang, K., Sun, Z., Feng, M., Liu, A., Yang, S., Chen, Y., Lin, X., 2011. Biosens. Bioelectron. 26, 2870–2876. Wang, Q.X., Gao, F., Gao, F., Li, S.X., Weng, W., Liu, F.Q., Jiao, K., 2012. Biosens. Bioelectron. 32, 50–55. Watanabe, S., Sato, S., Ohtsuka, K., Takenaka, S., 2011. Anal. Chem. 83, 7290–7296. Wei, W., Ni, Q., Pu, Y., Yin, L., Liu, S., 2014. J. Electroanal. Chem. 714, 25–29. Xu, C., Cai, H., He, P., Fang, Y., 2001. Analyst 126, 62–65. Yu, M., Zhao, Q., Shi, L., Li, F., Zhou, Z., Yang, H., Yi, T., Huang, C., 2008. Chem. Commun., 2115–2117.
Contents lists available at ScienceDirect
Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios
One-step conjugation of aminoferrocene to phosphate groups as electroactive probes for electrochemical detection of sequence-specific DNA Qiong Hu a, Xianbao Deng a, Xuehua Yu a, Jinming Kong a,n, Xueji Zhang a,b,nn a b
School of Environmental and Biological Engineering, Nanjing University of Science & Technology, Nanjing 210094, PR China Chemistry Department, College of Arts and Sciences, University of South Florida, East Fowler Ave, Tampa, FL 33620-4202, United States
art ic l e i nf o
a b s t r a c t
Article history: Received 28 July 2014 Received in revised form 7 October 2014 Accepted 7 October 2014
A straightforward electrochemical DNA biosensing approach based on exploiting organometallic compound, aminoferrocene (AFC), as electroactive probes was firstly demonstrated, where the probes could be directly labeled to the free phosphate groups of the hybridized PNA/DNA heteroduplexes merely through one-step conjugation in the presence of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and imidazole. Briefly, mercapto-terminated peptide nucleic acid (PNA) was firstly immobilized onto gold electrode and used as the capture probes for the specific recognition of target single-stranded DNA (ssDNA). After hybridization, AFC probes were directly labeled to the free 5′-terminal phosphate groups, which were activated by EDC and imidazole, of the hybridized PNA/DNA heteroduplexes, and then they were exploited as the electroactive probes to monitor the hybridization. As the captured ssDNA was labeled with AFC in the stoichiometric ratio of 1:1, thus the electrochemical analysis of the proportionally labeled AFC based on differential pulse voltammetry (DPV) enabled a quantitative determination of sequence-specific DNA. Under optimal conditions, the approach presented a good linear relationship between the current intensities and logarithm of ssDNA concentrations in the range from 0.1 nM to 100 nM with a detection limit of 93 pM, and it rendered satisfactory analytical performance in serum samples. Furthermore, it exhibited excellent specificity toward single-nucleotide polymorphism (SNP) and precluded complicated protocols. More importantly, the simplicity of this approach together with its compatibility with standard micro-fabrication techniques makes it great potential in practical applications, especially in microarray areas where simple procedures are preferred. & Elsevier B.V. All rights reserved.
Keywords: Electrochemical DNA Organometallic compounds Aminoferrocene Single-nucleotide polymorphism Electrochemical impedance spectroscopy
1. Introduction Organometallic compounds are usually defined as a specific class of metal complexes containing at least one metal–carbon bond (Patra and Gilles, 2012). They are renowned for their remarkable applications in the fields including chemical biology, catalysis, and therapy because of their outstanding physicochemical properties such as structural diversity, chemical stability and unique photo- and electrochemical properties (Heinze and Lang, 2013; Noffke et al., 2012; Patra and Gilles, 2012). Presently, they have been preliminarily investigated as enzyme inhibitors and luminescent agents in chemical biology, while they have also been extensively exploited as stereoselective and asymmetric catalysts n
Corresponding author. Corresponding author at: School of Environmental and Biological Engineering, Nanjing University of Science & Technology, Nanjing 210094, PR China. E-mail addresses: [email protected] (J. Kong), [email protected] (X. Zhang). n
http://dx.doi.org/10.1016/j.bios.2014.10.015 0956-5663/& Elsevier B.V. All rights reserved.
in organic synthesis and as therapeutic agents in biomedicine (Che and Sun, 2011; Debreczeni et al., 2006; Feng et al., 2011; Heinze and Lang, 2013; Noffke et al., 2012; Patra and Gilles, 2012; Ringenberg and Ward, 2011). More interestingly, some findings have nicely revealed that a variety of organometallic compounds can effectively accumulate in specific compartments of cellular organelles due to their selective reactions with certain substances in such compartments or a consequence of hydrophobic/hydrophilic/electrostatic interactions (Barnard et al., 2006; Li et al., 2011; Murphy et al., 2010; Yu et al., 2008). For example, an Ir(III) complex with selective turn-on phosphorescence for nuclei in living cells has been reported, it tends to specifically stain nuclei through luminescence enhancement by interaction with nucleic acids via its rapid reaction with histidine/histidine-containing proteins (Li et al., 2011). In addition, the conjugation of metal complexes, especially organometallic compounds, to biomolecules such as carbohydrates, polypeptides, proteins, and DNA have also been intensively investigated over the past years (Kong et al., 2007,
72
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
2013). Thus, apart from the well-established applications as efficient luminescent probes for bioimaging (Clède et al., 2012; Meister et al., 2010; Patra and Gilles, 2012), they also show great potential as labeling probes in analytical chemistry (Fang et al., 2011; Liu et al., 2014). With the increasing improvement of living standards and the further deepening of globalization, urgent demands relevant to the early diagnosis of hereditary diseases in clinical medicines and the fast screening of pathogenic microorganisms in customs quarantine and food quality supervision have prompted the fast development of highly sensitive, highly selective, and even real-time biosensing approaches with superior specificity in differentiating sequence-specific DNA (Grabowska et al., 2014; Hu et al., 2015; Liu et al., 2013a, 2013b; Qian et al., 2015). Compared with conventional optical approaches, electrochemical approaches have drawn considerable attentions by virtue of their high sensitivity, uncomplicated instrumentation, low cost, good portability, and excellent compatibility with micro-manufacturing technology (Hansen et al., 2006; Hu et al., 2015; Liu et al., 2008; Qian et al., 2015). Generally, the quantitative analysis of DNA can be conveniently achieved via monitoring the proportionally labeled electroactive probes after hybridization. For example, owing to their favorable redox properties, ferrocene-based derivatives, a versatile family of stable organometallic compounds, have emerged as high-profile electroactive probes in the field of electrochemical DNA biosensing (Fang et al., 2011; Seiwert and Karst, 2008). To exploit ferrocene-based derivatives as electroactive probes, several available strategies have been reported. Examples include attempts to modify them to the terminus of molecular beacons (MBs) (Miao et al., 2014; Qian et al., 2014); labeling target ssDNA with them before/after hybridization through versatile coupling linkers (Fang et al., 2011); hybridization with the auxiliary signal probes that have been labeled with them (Wang et al., 2003; Wei et al., 2014), and so on. These strategies are efficient and have been extensively investigated. Nevertheless, complicated protocols for the preparation, separation and purification of the labeled ssDNA are tedious and time-consuming, which prevent its versatility in practical applications (Xu et al., 2001). To overcome the enumerated drawbacks, carbodiimides, especially 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC), can be exploited as the crosslinking mediators to directly label aminecontaining ferrocene to the captured target ssDNA among the electrochemical detection of sequence-specific DNA. Carbodiimides are capable of activating the carboxylate groups and phosphate groups efficiently to form highly reactive intermediates, while the latter can be immediately substituted by amine-containing nucleophiles (Hermanson, 2008). More importantly, the crosslinking between carboxylates/phosphates and amine-containing nucleophiles is nearly impossible without carbodiimides as the activation mediators due to their poor reactivity (Chu et al., 1983; Ghosh et al., 1990; Hermanson, 2008). Among all carbodiimides, EDC is the frequently used crosslinking mediator in bioconjugation, and this benefits most from its favorable watersolubility. Consequently, EDC shows great potential as the crosslinking mediator in mediating the direct conjugation of target ssDNA and amine-containing electroactive probes (Chu et al., 1983; Ghosh et al., 1990; Hermanson, 2008; Ralph et al., 1962). To the best of our knowledge, there has been no report about the electrochemical detection of sequence-specific DNA based on organometallic compounds merely through one-step conjugation. Herein, we demonstrated a straightforward electrochemical DNA biosensing approach, which was applicable to detect sequencespecific DNA, by exploiting organometallic compound, aminoferrocene (AFC), as electroactive probes for the first time. In this work, mercapto-terminated peptide nucleic acid (PNA) probes
were firstly immobilized onto the clean surface of the pretreated gold electrode via the formation of self-assembled monolayer (SAM). Then, they were utilized as the capture probes for the specific recognition of target ssDNA to be determined in the following step. After hybridization, AFC probes were directly labeled to the free 5′-terminal phosphate groups of the hybridized PNA/ DNA heteroduplexes merely through one-step conjugation in the presence of EDC and imidazole, subsequently, they were exploited as the electroactive probes to monitor the hybridization via differential pulse voltammetry (DPV). As the hybridized ssDNA was labeled with AFC in the stoichiometric ratio of 1:1, thus the electrochemical analysis of the proportionally labeled AFC enabled a reliable quantitative determination of sequence-specific DNA. Under optimal conditions, the approach presented a good linear relationship between the current intensities and logarithm of ssDNA concentrations in the range from 0.1 nM to 100 nM, and it rendered satisfactory analytical performance in serum samples. As the hybridized ssDNA could be effectively labeled with AFC merely through one-step conjugation, thus the approach demonstrated here was straightforward and labour-saving, which could effectively preclude complicated protocols accordingly. Results also revealed that it exhibited excellent specificity toward single-nucleotide polymorphism (SNP). More importantly, the simplicity of this approach together with its compatibility with standard microfabrication techniques make it great potential in practical applications, especially in microarray areas where simple procedures are preferred.
2. Material and methods 2.1. Materials and reagents Mercapto-terminated peptide nucleic acid with sequence 5′-HS-(CH2)11-AAC CAT ACA ACC TAC TAC CTC A-3′ (PNA) was custom-made by Panagene Inc. (South Korea). Target complementary ssDNA 5′-TGA GGT AGT AGG TTG TAT GGT T-3′ (cDNA), single-base mismatched ssDNA (mismatch underlined) 5′-TGA GGT AGT AGG TTG TGT GGT T-3′ (SBM), three bases mismatched ssDNA (mismatches underlined) 5′-TGA GGT ATT AGA TTG TGT GGT T-3′ (TBM), and completely non-complementary ssDNA 5′ACT TAC CTT TGC TCA TTG ACG A-3′ (Control) were all purchased from Invitrogen Biotechnology Co., Ltd. (Shanghai, China). 1-Ethyl3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC), and imidazole were purchased from J&K Scientific Ltd. (Shanghai, China). Aminoferrocene (AFC) was purchased from TCI (Shanghai, China). 6-Mercapto-1-hexanol (MCH) was purchased from SigmaAldrich (St. Louis, MO). Fetal bovine serum (FBS, Defined) was purchased from Shanghai YiJi Industrial Co., Ltd. (Shanghai, China). All other chemical reagents were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China) and used without further purification. Ultrapure water obtained from a Millipore Milli-Q water purification system (Z18.25 MΩ) was used in all experiments. Tris–EDTA buffer (TE, 10 mM Tris–HCl, 1 mM EDTA, pH ¼ 8.0) was prepared and utilized as the stocking solution and washing buffer for ssDNA, and various concentrations of ssDNA samples were prepared by serial dilution. 0.1 M imidazole solution was prepared by dissolving it in 80% ethanol solution with its pH adjusted by HCl solution and used as the buffer for the labeling of ssDNA. 2.2. Apparatus All electrochemical experiments were performed with a conventional three-electrode system consisted of modified gold
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
electrode (Φ ¼2 mm) as working electrode, saturated calomel electrode (SCE) as reference electrode and platinum wire as counter electrode. A CHI 760D electrochemical workstation (Shanghai, China) was used for differential pulse voltammetry (DPV) and cyclic voltammetry (CV) measurements. Electrochemical impedance spectroscopy (EIS) measurements were performed on an Autolab/PGSTAT30 (Eco Chemie, Netherlands). And all the electrochemical experiments were carried out at room temperature.
73
2.5. Electrochemical measurement To carry out the electrochemical measurement, the gold electrode was finally immersed in 10 mL of 1.0 M KNO3 solution (nitrogen-saturated), and then DPV was performed under the potential from 0 V to 0.5 V to record the oxidation current of the covalently labeled AFC (Fang et al., 2011). The scheme for electrochemical detection of sequence-specific DNA by exploiting AFC as electroactive probes through one-step conjugation is illustrated in Scheme 1.
2.3. Pretreatment of the gold electrode and immobilization of capture probes 3. Results and discussion Prior to modification, the gold electrode was polished to a mirror-like surface with 0.3 and 0.05 μm alumina slurry, and then successively ultrasonically cleaned with absolute ethanol and ultrapure water (Hu et al., 2015; Liu et al., 2013a, 2013b). Then, it was chemically cleaned in freshly prepared piranha solution (a mixture of 98% H2SO4 and 30% H2O2, 3:1 v/v, CAUTION: piranha solution is strongly corrosive and must be handled with extreme care) under ultrasonication (Hu et al., 2015; Liu et al., 2013a, 2013b). Afterwards, it was electrochemically cleaned in a fresh 0.5 M H2SO4 solution by cycling the electrode potential between 0.2 V and 1.5 V with a scan rate of 100 mV s 1 until a reproducible cyclic voltammogram was achieved to remove any remaining impurities (Hu et al., 2015; Liu et al., 2013a, 2013b). Finally, it was rinsed thoroughly with absolute ethanol and ultrapure water and dried with nitrogen prior to modification. Immobilization of capture probes was accomplished by incubating the clean electrode in 1 μM PNA aqueous solution to allow the formation of well-ordered SAM (Hu et al., 2015). After washing with ultrapure water, it was immersed in 2 mM MCH solution (dissolved in 70% ethanol solution) to passivate the possible nonspecific binding sites, and then successively rinsed with 70% ethanol solution and ultrapure water to remove excessive MCH (Hu et al., 2015; Liu et al., 2013a, 2013b). 2.4. DNA hybridization and labeling of electroactive probes Hybridization was performed in 10 μL of TE buffer that contained certain concentration of ssDNA at 37 °C for 1.5 h, followed by washing with TE buffer to remove the unhybridized ssDNA (Hu et al., 2015). After that, 20 μL of freshly prepared mixed solution of AFC and EDC with the concentration of EDC kept at 10 mg mL 1 (freshly dissolved in 0.1 M imidazole solution) was added on the electrode and then incubated at 25 °C for the labeling of electroactive probes. (The schematic diagram of covalent labeling of ssDNA with AFC in the presence of EDC and imidazole was shown in Fig. 1.) Finally, the resulting electrode was moderately washed with 80% ethanol solution and ultrapure water to remove the remaining reactants.
3.1. Electrochemical characterization of the covalently labeled AFC In the absence of imidazole, the water-soluble carbodiimide EDC alone tends to rapidly activate the free phosphate group at the 5′-terminus of ssDNA to form an active phosphomonoester intermediate, which is highly reactive to couple with amine-containing nucleophilic compounds (Ghosh et al., 1990; Ralph et al., 1962; Shabarova, 1988). However, EDC is labile in water and the resulting intermediate shows a short half-life in aqueous solution due to its hydrolysis (Hermanson, 2008). In addition, the generated phosphorimidazolide intermediate in the presence of imidazole provides better reactivity toward amine-containing nucleophiles than the corresponding EDC phosphomonoester intermediate (Hermanson, 2008). As a result, the supplement of imidazole to the reaction medium could significantly improve the derivatization yield of ssDNA labeled with AFC probes over the carbodiimide-only reactions (Hermanson, 2008). In this work, EDC and imidazole were simultaneously adopted as the synergistic crosslinking mediators to directly label AFC probes to the hybridized PNA/DNA heteroduplexes via the EDC-mediated one-step conjugation. To evaluate the feasibility and effectivity of the proposed strategy for labeling of target ssDNA with AFC probes merely through one-step conjugation in the presence of EDC and imidazole, DPV was used to characterize the presence of AFC probes on the electrode. As shown in Fig. 2A, an apparent oxidation peak would appear at the potential around 0.25 V provided that AFC probes had been covalently labeled to the hybridized heteroduplexes (Fig. 2A,a), the resulting oxidation potential was in agreement with the reported value (Miao et al., 2014; Shiddiky et al., 2010), and the oxidation current could be interpreted as the electrochemical oxidation of ferrocene into ferrocenium (Brisset et al., 2006; Heinze and Lang, 2013). On the contrary, there merely feeble background current due to the charging process would be observed without the labeling of AFC probes (Fig. 2A,b) or before the hybridization had been performed (Fig. 2A,c). The obtained
Fig. 1. Covalent labeling of ssDNA with AFC in the presence of EDC and imidazole.
74
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
Scheme 1. Schematic illustration of electrochemical detection of sequence-specific DNA by exploiting AFC as electroactive probes through one-step conjugation.
Fig. 2. Differential pulse voltammograms of the electrode after (a) and before (b) labeling of target ssDNA with AFC as well as before the hybridization (c) (A); the cyclic voltammograms of the labeled AFC probes on the electrode at scan rates from 10 to 200 mV s 1 (B), (Inset) The linear relationship been anodic (black line) and cathodic (red line) peak currents and scan rates toward 5 nM cDNA in 1.0 M KNO3; the Nyquist plots of EIS of the bare gold electrode (Au) (a), Au–PNA (b), Au–PNA–MCH (c), Au–PNA– MCH–DNA (d), and Au–PNA–MCH9DNA–AFC (e) (C).
results indicated that AFC probes had been successfully labeled to the free phosphate groups at the 5′-terminus of the hybridized ssDNA via the EDC-mediated one-step conjugation. More importantly, the oxidation current appeared only when the hybridization had been accomplished and AFC probes had been successfully conjugated to the hybridized PNA/DNA heteroduplexes, and it in turn strongly guaranteed the reliability of the response signal. Thus, the proposed strategy was quite feasible and effective. In addition, the covalently labeled AFC probes were further characterized by cyclic voltammetry (CV). As shown in Fig. 2B, cyclic voltammograms obtained for the labeled AFC probes at different scan rates from 10 to 200 mV s 1 exhibited that the ferrocene/ferrocenium redox process was quasi reversible at low scan rates, and then it became more irreversible parallel with the further increasing of scan rate (Grabowska et al., 2014). In addition, the linear relationship between the anodic and cathodic peak currents and scan rates further indicated that the redox process was not diffusion dependent, and thus confirmed that the AFC probes were confined to the electrode surface with a slow electron transfer (Brisset et al., 2006; Grabowska et al., 2014; Fan et al., 2003). Electrochemical impedance spectroscopy (EIS) is an effective, informative, and amenable technique to examine the microscopic interfacial changes (Shervedani and Pourbeyram, 2009). Here, it was performed to provide further evidence for the preparation process of the demonstrated approach by measuring the impedance change of the electrode surface during the modification procedure. The impedance was carried out within the frequency range of 0.01 Hz to 100 kHz in a solution of 0.1 M KNO3 containing 5 mM [Fe(CN)6]3 /4 as the redox probe at a potential of 0.18 V (vs. SCE), with a voltage amplitude of 5 mV. Fig. 2C shows the Nyquist plots of impedance spectra at different stages of the preparation process. The semicircle portion observed at high
frequencies corresponded to the electron transfer limited process and its diameter was equal to the electron transfer resistance, which could reflect the electron transfer kinetics of the redox probe at the electrode surface (Alfonta et al., 2001). The bare gold electrode exhibited the smallest semicircle at high frequency region due to the fast electron-transfer process (Fig. 2C,a) (Wang et al., 2011). After the immobilization of mercapto-terminated PNA, the resistance increased accordingly (Fig. 2C,b). This could be attributed to the steric effect of the well-ordered self-assembled monolayer (Liu et al., 2010). After MCH was employed to passivate the nonspecific binding sites, the resistance further increased (Fig. 2C,c). This was mainly because that the more densely packed monolayer rendered a relatively slow electron-transfer. Since the negatively charged phosphate backbone of ssDNA electrostatically repelled [Fe(CN)6]3 /4 from reaching the electrode surface (Liu et al., 2005; Noorbakhsh and Salimi, 2011), remarkable increase in the electron transfer resistance could be observed when the hybridization had been accomplished (Fig. 2C,d). Finally, the steric effect of the covalently labeled AFC probes resulted in a further enhancement in the electron transfer resistance (Fig. 2C,e). 3.2. Optimization of detection conditions of the electrochemical DNA biosensing approach To achieve the best analytical performance of the demonstrated electrochemical DNA biosensing approach, several experimental parameters associated with the current intensity, which was in turn dominated by the amount of AFC probes that had been labeled to the hybridized PNA/DNA heteroduplexes, were carefully investigated. Apart from the pH value, the labeling time and the concentration of AFC were also inspected. The activation of phosphate groups with EDC in aqueous solution occurs most effectively at acidic reaction medium, while it is also favorable for the formation of phosphoramidate linkages
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
75
Fig. 3. Effects of pH (A), labeling time (B), and AFC concentration (C) on the analytical performance of the electrochemical DNA biosensing approach toward 10 nM cDNA in 1.0 M KNO3.
(Nakajima and Ikada, 1995). However, EDC is susceptible to be violently hydrolyzed under the conditions (Nakajima and Ikada, 1995). Thus, the covalent labeling of ssDNA with AFC probes is indeed dominated by the competitive equilibrium process of these mechanisms mentioned above. Therefore, the effect of pH value on current intensity was firstly investigated. As shown in Fig. 3A, the current intensity increased drastically with the pH value varying from 4.8 to 6.0, and then it encountered a remarkable decrease with the further increasing of pH value. The obtained results indicated that the EDC-mediated formation of phosphoramidate linkage in the presence of imidazole benefited its maximal yield at pH 6.0. As a result, the 0.1 M imidazole solution at pH 6.0 was adopted for the labeling of ssDNA with AFC probes in the subsequent experiments. The analytical performance of the electrochemical DNA biosensing approach was largely depended upon the amount of AFC probes that had been covalently conjugated to the hybridized PNA/ DNA heteroduplexes, which was partially controlled by the labeling time. Admittedly, more labeling time resulted in more conjugated AFC probes, however, the biosensing time would also be extended accordingly. To optimize the analytical performance, especially the sensitivity, the effect of labeling time on the current intensity was further investigated. As shown in Fig. 3B, the current intensity increased significantly with the increasing of labeling time until 120 min, afterwards, no significant increase in current intensity could be observed with the further increasing of labeling time. Thus, it could be concluded that the optimal labeling time was 120 min, and was adopted in the following experiments.
Obviously, the concentration of AFC was another important parameter that would significantly affect the amount of the labeled electroactive probes, namely, the analytical performance could be further optimized by choosing an optimal AFC concentration. Meanwhile, the labeling time would be shortened accordingly. As shown in Fig. 3C, the current intensity increased rapidly with the increasing of AFC concentration until 1.4 mg mL 1 and then it trended to reach a plateau until 1.6 mg mL 1. From the result mentioned above, it could be concluded that the optimal concentration of AFC was 1.6 mg mL 1. In addition, as the concentration of AFC at 1.6 mg mL 1 is around 7.96 mM, which is much higher than the concentration of ssDNA to be determined, thus it can be accepted as the saturated concentration in the following experiments. 3.3. Analytical performance of the electrochemical DNA biosensing approach Under optimal conditions, the analytical performance of the electrochemical DNA biosensing approach was examined with varying cDNA concentrations. As shown in Fig. 4, the current intensities increased linearly along with the increasing concentration of cDNA. The calibration plot showed a good linear relationship between the current intensities and the logarithm of cDNA concentrations in the range from 0.1 nM to 100 nM. The linear regression equation was I (μA)¼ 0.284þ0.170 lg [CDNA/nM] with a correlation coefficient of 0.9984. The limit of detection at a signalto-noise ratio of 3 was calculated to be 93 pM, which was much lower than those previously reported approaches (Miao et al.,
Fig. 4. Differential pulse voltammograms (A), and calibration plot (B) of the electrochemical DNA biosensing approach toward varying concentrations of cDNA in the range from 0.1 nM to 100 nM.
76
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
was 3.8% for inter-assay. The results indicated that the demonstrated approach had acceptable repeatability; and it might be attributed to the relatively simple experimental procedures, which could effectively eliminate the potential interferences originated from the complicated protocols. In addition, the stability of the electrochemical DNA biosensing approach was further investigated by long-term storage assay. The modified electrode could be stored in a moisture-saturated environment at 4 °C, and over 93.4% of the initial response retained after a storage period of three weeks. The loss of response signal might be a consequence of the irreversible oxidation of the conjugated ferrocene. The results demonstrated that the approach possessed satisfactory stability. 3.5. Application in analysis of serum samples
Fig. 5. DPV responses of the electrochemical DNA biosensing approach toward four types of 10 nM ssDNA.
2014; Xu et al., 2001; Watanabe et al., 2011; Wang et al., 2012). The improved sensitivity might be attributed to the excellent discrimination ability of DPV in distinguishing faradaic current from background current when the target analytes to be determined existed at ultralow level (Habibi et al., 2011). In addition, compared with other electrochemical DNA biosensing approaches, the labeling of electroactive probes in the demonstrated approach excluded complicated protocols and could be conveniently achieved merely through one-step conjugation. Therefore, the analytical performance of the demonstrated electrochemical DNA biosensing approach was acceptable. 3.4. Selectivity, repeatability, and stability of the electrochemical DNA biosensing approach The selectivity, namely specificity, of the electrochemical DNA biosensing approach was investigated by using a single-base mismatched ssDNA (SBM), three bases mismatched ssDNA (TBM), and completely non-complementary ssDNA (Control). As shown in Fig. 5, the current intensities from SBM, TBM, and Control were approximately 83.0%, 62.3%, and 7.9% of the value obtained from cDNA, respectively. The difference could be attributed to the hybridization efficiency of the four types of ssDNA with the immobilized capture probes and the prominent selectivity of PNA, as the cDNA could specifically and robustly bind to the complementary capture probes while the others couldn't (Hu et al., 2015). More importantly, the introduction of AFC probes to the system happened only when ssDNA had been successfully captured by the immobilized capture probes, because there were no additional possible reactive sites available for the binding of AFC probes. The obtained results highlighted that the novel approach demonstrated here could effectively differentiate complementary and mismatched oligonucleotide fragments, indicating that it was highly specific and even showed great potential for the genotyping of SNP. The excellent selectivity could be attributed to the utilization of highly specific PNA as the immobilized capture probes (Hu et al., 2015). The repeatability of the electrochemical DNA biosensing approach was evaluated by the intra-assay and inter-assay coefficients of variation. The intra-assay and inter-assay were independently conducted by measuring two identical samples both contained 5 nM cDNA under the same conditions. Each assay was repeatedly conducted 10 times using 10 parallelly prepared electrodes. The coefficient of variation of intra-assay was 4.9%, while it
To evaluate the analytical reliability and application potential of the electrochemical DNA biosensing approach, the interference effect of complicated serum samples on analytical performance was investigated. The current intensities originated from 5 nM cDNA in serum samples were compared with that obtained from 5 nM cDNA in TE buffer. The assay results from 10 repeated experiments are displayed in Fig. 6. The current intensities from 1% (B) and 5% (C) serum samples were approximately 94.4% and 88.4% of that from TE buffer (A), respectively. All these obtained results highlighted that the demonstrated approach showed acceptable analytical reliability and application potential for clinical applications, allowing for its good detectable capability in complicated systems such as samples in serum.
4. Conclusion In summary, a straightforward electrochemical DNA biosensing approach based on exploiting organometallic compound, aminoferrocene (AFC), as electroactive probes was firstly demonstrated, where the probes could be directly labeled to the free phosphate groups of the hybridized PNA/DNA heteroduplexes merely through one-step conjugation in the presence of EDC and imidazole. Using the mercapto-terminated PNA as the immobilized capture probes for the specific recognition of target ssDNA significantly improved the selectivity, leading to the great potential in the genotyping of SNP. The utilization of differential pulse voltammetric technique improved the sensitivity, enabled a detection limit of 93 pM
Fig. 6. DPV responses of the electrochemical DNA biosensing approach toward 10 nM cDNA in TE buffer (A), 1% serum sample (B), and 5% serum sample (C).
Q. Hu et al. / Biosensors and Bioelectronics 65 (2015) 71–77
achieved under optimal conditions. In addition, it excluded complicated protocols for the monitoring of hybridization and could be successfully applied for the quantification of sequence-specific DNA in serum samples. The outstanding analytical performance including labour- and time-saving, low cost, satisfactory repeatability, acceptable stability and reliability, implicated that it showed great potential in clinical applications. More importantly, the demonstrated approach based on exploiting one-step conjugation of organometallic compounds to target analytes as electroactive probes in the presence of EDC and imidazole has presented a novel strategy for the electrochemical determination of target analytes that contained free carboxylate groups, phosphate groups, and even primary amines. Furthermore, it could be integrated with standard micro-fabrication techniques to develop microfluidic chips and realized multiplexed detection capabilities at the same time.
Acknowledgments We are grateful to Nanjing University of Science and Technology for its start-up funding and also National Natural Science Foundation of China (No. 21345002) for funding this project.
References Alfonta, L., Willner, I., Throckmorton, D.J., Singh, A.K., 2001. Anal. Chem. 73, 5287–5295. Brisset, H., Navarro, A.E., Spinelli, N., Chaix, C., Mandrand, B., 2006. Biotechnol. J. 1, 95–98. Barnard, P.J., Wedlock, L.E., Baker, M.V., Berners-Price, S.J., Joyce, D.A., Skeleton, B. W., Steer, J.H., 2006. Angew. Chem. Int. Ed. 45, 5966–5970. Chu, B.C.F., Wahl, G.M., Orgel, L.E., 1983. Nucleic Acids Res. 11, 6515–6529. Che, C.M., Sun, R.W.Y., 2011. Chem. Commun. 47, 9554–9560. Clède, S., Lambert, F., Sandt, C., Gueroui, Z., Réfrégiers, M., Plamont, M.A., Dumas, P., Vessières, A., Policar, C., 2012. Chem. Commun. 48, 7729–7731. Debreczeni, J.E., Bullock, A.N., Atilla, G.E., Williams, D.S., Bregman, H., Knapp, S., Meggers, E., 2006. Angew. Chem. Int. Ed. 45, 1580–1585. Fan, C., W. Plaxco, K., J. Heeger, A., 2003. Proc. Natl. Acad. Sci. USA 100, 9134–9137. Feng, L., Geisselbrecht, Y., Blanck, S., Wilbuer, A., Atilla-Gokcumen, G.E., Filippakopoulos, P., Krling, K., Celik, M.A., Harms, K., Maksimoska, J., Mar-morstein, R., Frenking, G., Knapp, S., Essen, L.-O., Meggers, E., 2011. J. Am. Chem. Soc. 133, 5976–5986. Fang, C., Ji, H., Karen, W.Y., Rafei, S.R.M., 2011. Biosens. Bioelectron. 26, 2670–2674. Ghosh, S.S., Kao, P.M., McCue, A.W., Chappelle, H.L., 1990. Bioconjug. Chem. 1, 71–76.
77
Grabowska, I., Singleton, D.G., Stachyra, A., Góra-Sochacka, A., Sirko, A., ZagórskiOstoja, W., Radecka, H., Stulz, E., Radecki, J., 2014. Chem. Commun. 50, 4196–4199. Hansen, J.A., Mukhopadhyay, R., Hansen, J.Ø, Gothelf, K.V., 2006. J. Am. Chem. Soc. 128, 3860–3861. Hermanson, G.T., 2008. Bioconjugate Techniques, 2nd ed. Academic Press, Amsterdam. Habibi, B., Jahanbakhshi, M., Pournaghi-Azar, M.H., 2011. Anal. Biochem. 411, 167–175. Heinze, K., Lang, H., 2013. Organometallics 32, 5623–5625. Hu, Q., Hu, W., Kong, J., Zhang, X., 2015. Biosens. Bioelectron. 63, 269–275. Kong, K.V., Chew, W., Lim, L.H.K., Fan, W.Y., Leong, W.K., 2007. Bioconjug. Chem 18, 1370–1374. Kong, K.V., Lam, Z., Lau, W.K.O., Leong, W.K., Olivo, M., 2013. J. Am. Chem. Soc. 135, 18028–18031. Liu, J., Tian, S., Nielsen, P.E., Knoll, W., 2005. Chem. Commun., 2969–2971. Liu, G., Wan, Y., Gau, V., Zhang, J., Wang, L., Song, S., Fan, C., 2008. J. Am. Chem. Soc. 130, 6820–6825. Liu, X., Qu, X., Fan, H., Ai, S., Han, R., 2010. Electrochim. Acta. 55, 6491–6495. Liu, Y.H., Li, H.N., Chen, W., Liu, A.L., Lin, X.H., Chen, Y.Z., 2013a. Anal. Chem. 85, 273–277. Liu, S., Wang, C., Zhang, C., Wang, Y., Tang, B., 2013b. Anal. Chem. 85, 2282–2288. Liu, S., Lin, Y., Wang, L., Liu, T., Cheng, C., Wei, W., Tang, B., 2014. Anal. Chem. 86, 4008–4015. Li, C., Yu, M., Sun, Y., Wu, Y., Huang, C., Li, F., 2011. J. Am. Chem. Soc. 133, 11231–11239. Murphy, L., Congreve, A., Palsson, L.-O., Williams, J.A.G., 2010. Chem. Commun. 46, 8743–8745. Meister, K., Niesel, J., Schatzschneider, U., Metzler-Nolte, N., Schmidt, D.A., Havenith, M., 2010. Angew. Chem. Int. Ed. 49, 3310–3312. Miao, X., Guo, X., Xiao, Z., Ling, L., 2014. Biosens. Bioelectron. 59, 54–57. Nakajima, N., Ikada, Y., 1995. Bioconjugate. Chem. 6, 123–130. Noorbakhsh, A., Salimi, A., 2011. Biosens. Bioelectron. 30, 188–196. Noffke, A.L., Habtemariam, A., M. Pizarro, A., Sadler, P., J., 2012. Chem. Commun. 48, 5219–5246. Patra, M., Gilles, G., 2012. ChemBioChem 13, 1232–1252. Qian, Y., Wang, C., Gao, F., 2014. Talanta 130, 33–38. Qian, Y., Wang, C., Gao, F., 2015. Biosens. Bioelectron. 63, 425–431. Ralph, R.K., Young, R.J., Khorana, H.G., 1962. J. Am. Chem. Soc. 84, 1490–1491. Ringenberg, M.R., Ward, T.R., 2011. Chem. Commun. 47, 8470–8476. Shabarova, Z.A., 1988. Biochimie 70, 1323–1334. Seiwert, B., Karst, U., 2008. Anal. Bioanal. Chem. 390, 181–200. Shervedani, R.K., Pourbeyram, S., 2009. Biosens. Bioelectron. 24, 2199–2204. Shiddiky, M.J., Torriero, A.A., Zeng, Z., Spiccia, L., Bond, A.M., 2010. J. Am. Chem. Soc. 132, 10053–10063. Wang, J., Li, J., Baca, A.J., Hu, J., Zhou, F., Yan, W., Pang, D.-W., 2003. Anal. Chem. 75, 3941–3945. Wang, K., Sun, Z., Feng, M., Liu, A., Yang, S., Chen, Y., Lin, X., 2011. Biosens. Bioelectron. 26, 2870–2876. Wang, Q.X., Gao, F., Gao, F., Li, S.X., Weng, W., Liu, F.Q., Jiao, K., 2012. Biosens. Bioelectron. 32, 50–55. Watanabe, S., Sato, S., Ohtsuka, K., Takenaka, S., 2011. Anal. Chem. 83, 7290–7296. Wei, W., Ni, Q., Pu, Y., Yin, L., Liu, S., 2014. J. Electroanal. Chem. 714, 25–29. Xu, C., Cai, H., He, P., Fang, Y., 2001. Analyst 126, 62–65. Yu, M., Zhao, Q., Shi, L., Li, F., Zhou, Z., Yang, H., Yi, T., Huang, C., 2008. Chem. Commun., 2115–2117.
