Talanta 138 (2015) 273–278
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Bi-functionalized aptasensor for ultrasensitive detection of thrombin Liping Lu n, Jiao Li, Tianfang Kang, Shuiyuan Cheng Key Laboratory of Beijing on Regional Air Pollution Control, Beijing University of Technology, Beijing 100124, China
art ic l e i nf o
a b s t r a c t
Article history: Received 27 November 2014 Received in revised form 7 March 2015 Accepted 10 March 2015 Available online 18 March 2015
A novel bi-functionalized aptasensor for thrombin detection was fabricated by using electrochemiluminescence (ECL) and electrochemical impedance spectroscopy (EIS) techniques. A thiol-terminated aptamer with 15 oligonucleotides was hybridized with its complementary oligonucleotides to form a double-stranded DNA (dsDNA). Then, the thiol-labeled dsDNA was assembled on a Au electrode via Au–S bond; the other distal of the dsDNA labeled with biotin was bound to QDs through a biotin–avidin conjugation. When thrombin is present in the detection solution, the aptamer can combine with thrombin, resulting in loss of CdSe QDs from the electrode surface and thereby decreasing the ECL intensity and increasing the impedance. The decreased ECL and increased EIS signals are logarithmically linear with respect to the concentration of thrombin. The linear range was 10 10–10 3 μg mL 1 (R¼ 0.9924) for the ECL signal and 10 10–10 1 μg mL 1 (R¼0.9875) for the EIS method with a common detection limit of 10 10 μg mL 1 (2.7 aM). In addition, the bi-functionalized aptasensor exhibited excellent selectivity, super sensitivity, a low detection limit and a wide linear range. & 2015 Elsevier B.V. All rights reserved.
Keywords: Aptasensor Thrombin ECL DNA Quantum dots
1. Introduction In the early 1990s, three research groups (Szostak [1], Tuerk [2] and Joyce [3]) in the US independently reported that there were oligonucleotide sequences with high affinities for proteins, nucleic acids or other small molecules that could be isolated via in vitro selection and amplification. The selected oligonucleotide sequences were named aptamers. Due to the high specificity between aptamers and target molecules, aptamers have received wide attention and have been used for broad application in various scientific fields, such as sensor preparations, drug screenings, and disease diagnoses and treatments. Aptasensors, which are based on aptamers as molecular recognition substances, have since been widely applied to detect proteins, genes and drugs. Thrombin, a very important serine protease in blood coagulation, can directly transform soluble fibrinogen into insoluble fibrin, accelerate blood coagulation and facilitate blood hemostasis [4]. If the level of thrombin in the body is not maintained within a certain range, thromboembolic diseases or even death would result. The concentration and activity of thrombin can also be used as an important indicator of the clotting mechanism, which is also of great importance to the early diagnosis, efficacious examination and prognosis of some types of diseases. Under normal circumstances, the thrombin concentration is nmol/L. The establishment of a rapid and accurate thrombin detection method with high sensitivity is of n
Corresponding author. Tel.: þ 86 10 67391659; fax: þ86 10 67391983. E-mail address: [email protected] (L. Lu).
http://dx.doi.org/10.1016/j.talanta.2015.03.016 0039-9140/& 2015 Elsevier B.V. All rights reserved.
vital significance in medicine. Currently, there are multiple ways to detect thrombin, such as high performance liquid chromatography (HPLC) [5], chromatography [6], fluorescence-based methods [7], surface plasmon resonance [8], electrochemical methods [9], and the electrochemiluminescence (ECL) method [10]. As we know, electrochemical and ECL methods have widely used in various scientific fields for its simple device requirements, reduced sample and reagent consumption and use easily. In this study, we designed a bi-functionalized aptasensor based on the aptamer-specific binding of thrombin to detect thrombin with high sensitivity and high specificity. The selected thrombin aptamer-15 (Apt15) has a high affinity to fibrinogen binding sites of thrombin. The ECL intensity and EIS were proportional to the concentration of thrombin; therefore, we used both ECL and EIS methods to detect thrombin. This bi-functionalized method is a more reliable detection system in various scientific and medical fields. These advantages show the great potentiality of our sensor for application in anticoagulation drugs screening and cardiovascular and thromboembolic disease diagnosis in clinical field.
2. Experimental section 2.1. Chemicals and materials CdCl2 2.5H2O and K2S2O8 were purchased from Beijing Chemical Reagent Corporation. Selenium was purchased from Shanghai Meixing Chemical Corporation. NaBH4, Na2HPO4 12H2O,
274
L. Lu et al. / Talanta 138 (2015) 273–278
NaH2PO4 2H2O and KCl were purchased from Tianjin Fuchen Chemical Corporation. MgCl2 was purchased from Shanghai Sangon Biotechnology Corporation. Mercaptopropionic acid (MPA), 6-mercapto-1-hexanol (MCH), avidin (from egg white), adenosine 5′-triphosphate (ATP), L-histidine, lysozyme, bovine serum albumin (BSA), mouse IgG, casein and thrombin (from human plasma) were purchased from Sigma-Aldrich (USA). All reagents were of analytical grade. Labeled DNA oligonucleotides were synthesized by Beijing SBS Gene Tech. Co. The sequences of the two oligonucleotides used are provided below: ssDNA-1: 5′-biotin-CCA ACC ACA CCA ACC-3′ ssDNA-2: 5′-SH-GGT TGG TGT GGT TGG-3′
2.2. Apparatus Cyclic voltammetry (CV) and EIS were performed with a CHI 660a Electrochemical Analyzer (Shanghai Chenhua Instrument Corporation). ECL emission measurements were conducted on an MPI-A ECL analyzer (Xi'an Remex Analysis Instrument Corporation). All experiments were carried out with a three-electrode system: an Au electrode as the working electrode, an Ag–AgCl electrode as the reference electrode, and a platinum wire as the counter electrode. Field emission electron microscopy was performed with a JEOL model 2100F instrument. UV–vis absorption spectra were obtained on a UV-2450 spectrophotometer (Shimadzu). Fluorescence emission spectra were obtained on a F-4600 fluorescence spectrophotometer (Hitachi). 2.3. Synthesis of CdSe/MPA quantum dots CdSe/MPA QDs were synthesized using a slightly modified procedure previously reported in [11]. Briefly, 0.0094 g CdCl2 2.5H2O was dissolved in 60 mL ultrapure water, followed by the addition of 9 μL MPA. The pH of the solution was adjusted to 11.0 by the dropwise addition of 1.0 mol/L NaOH solution. Afterward, the mixture was transferred to a three-necked flask and bubbled with pure N2 for 30 min. Then, 500 μL of freshly prepared NaHSe solution was slowly injected, and the solution was refluxed for 2 h at 100 °C to obtain a clear, light yellow solution of CdSe/MPA QDs. After cooling, the CdSe/MPA QDs solution was centrifuged at 6000 rpm for 5 min. The final CdSe/MPA QDs solution was stored in a refrigerator at 4 °C.
2.4. Pretreatment of the gold electrode Prior to use, the Au electrode was polished with 0.02–0.05 μm Al2O3 power on a polishing pad. Then, the electrode was cleaned with freshly made piranha solution (98% H2SO4:30% H2O2 ¼7:3) for 10 min, followed by an adequate ultrasonic cleaning in absolute ethanol and ultrapure water. Subsequently, the electrode was immersed in 0.5 mol/L H2SO4 for cyclic voltammetry scan with a potential range of 0.3 to 1.5 V. Then, it was washed with ultrapure water and dried with N2. 2.5. Preparation of bi-functionalized aptasensor The schematic illustration of the bi-functionalized aptasensor for the detection of thrombin is shown in Scheme 1. First, 25 μL of 50 μmol/L dsDNA was dropped on a pre-cleaned Au electrode surface and incubated for 16–24 h in a humidified environment. Thus, thiol-terminated dsDNA was assembled on the Au electrode via a Au–S bond; this electrode was denoted as the DNA/Au electrode. The dsDNA was obtained by mixing ssDNA-1 and ssDNA-2 with the same concentration and volume, and then heated at 90 °C and slowly cooled to room temperature. After a thorough rinse with 0.1 mol/L PBS (0.1 mol/L HPO42 /H2PO4 , 0.1 mol/L KCl, pH 7.4) buffer solution, the DNA/Au electrode was backfilled with 0.1 mol/L MCH for 1 h to block the nonspecifically bound oligonucleotides. Then, the DNA/ Au electrode was adequately rinsed with 0.1 mol/L PBS buffer solution and immersed in a QDs–avidin solution under shocking conditions for 1 h. Thus, the other end of the dsDNA was firmly bound with CdSe QDs through a biotin–avidin conjugation. The electrode was denoted as the CdSe QDs/DNA/Au electrode. In the presence of thrombin in the solution, the aptamer would combine with thrombin, and CdSe QDs with complementary oligonucleotides would leave the surface of electrode. The CdSe QDs–avidin solution was obtained by mixing CdSe QDs and avidin in a certain proportion [12] and centrifuged at 13,000 rpm for 80 min. The supernatant was discarded and the precipitate was CdSe QDs–avidin which was dissolved in ultrapure water. 2.6. ECL and EIS methods detection of thrombin For the detection of thrombin, bi-functionalized aptasensors were incubated in a thrombin solution for 40 min at 37 °C, and then washed with a 0.1 mol/L PBS buffer solution.
Scheme 1. The preparation of bi-functionalized aptasensor and the detection of thrombin.
L. Lu et al. / Talanta 138 (2015) 273–278
b
1.0
2250
614nm
1500
506nm a
0.5
750
0.0
0 480
560 640 Wavelength(nm)
PL intensity
1.5
Absorbance
The measurement of thrombin was performed with the ECL method at a potential ranging from 1.5 to 0.0 V at a scan rate of 100 mV/s with a three electrode system. Prior to and after the incubation in the thrombin solution, the ECL intensity was recorded in a 0.1 mol/L PBS (pH 7.4) solution containing 0.1 mol/L K2S2O8 as a coreactant. The EIS measurements were carried out on a CHI 660a Electrochemical Analyzer. Prior to and after the incubation in the thrombin solution, the impedance spectra at a frequency range from 1 to 105 Hz and an amplitude of 5 mV was recorded in a 0.1 mol/L KCl solution containing 5.0 mmol/L K3[Fe(CN)6]/K4[Fe(CN)6].
275
720
Fig. 2. The absorbance spectra (a) and fluorescence spectra (b) of CdSe QDs.
3. Results and discussion
2.1
-6
The morphologies of CdSe QDs were characterized by high resolution transmission electron microscopy (HRTEM) imaging (Fig. 1). As seen from the image, the average size of the prepared CdSe QDs was approximately 2–3 nm with a spherical shape and good dispersion. Fig. 2(a) and (b) shows the UV–vis absorbance spectra and fluorescence spectra of CdSe QDs, respectively. From the UV–vis absorbance spectra, we can see that the absorption peak occurred at 506 nm, suggesting the size of CdSe QDs to be 2.4 nm [13], which corresponded with the diameter of CdSe QDs observed from HRTEM images. From the fluorescence spectra, we can see that the photoluminescence peak occurred at 614 nm, indicating the consequence of quantum confinement.
Charge× 10 /C
3.1. Characterization of the prepared CdSe QDs
a
1.4
0.7
b
0.0 0.00
0.15
0.30 1/2
0.45
1/2
Time /s Fig. 3. Chronocoulometric response RuHex þ5 mM PBS and (b) 5 mM PBS.
curves
of
dsDNA/Au:
(a)
100 μM
3.2. Coverage of dsDNA on Au electrode surface
450 ECL intensity
To confirm the density of dsDNA on the electrode surface, Tarlov [14] first calculated the redox charge on the surface of the Ru(NH3)63 þ (RuHex) molecule by chronocoulometry. They proposed a mechanism wherein the RuHex molecule is electrostatically absorbed to DNA phosphate backbone, which carries a negative charge. Because the density of RuHex fixed on the surface of the electrode is directly proportional to the density of DNA, we can calculate the surface density of the dsDNA. When the electrode
300
150
0 0
10
20
30
t/s
ECL intensity
450
300
150
0
5
10 -1
Cthrombin / (ug.mL )
Fig. 1. High resolution transmission electron microscopy (HRTEM) images of CdSe QDs.
Fig. 4. (A) The ECL intensity of bi-functionalized aptasensor with different concentrations of thrombin (from top to bottom: 10 10, 10 9, 10 8, 10 7, 10 6, 10 5, 10 4, and 10 3 μg mL 1). (B) Relationship between ECL intensity and thrombin concentrations. Inset: logarithmic calibration curve for thrombin.
L. Lu et al. / Talanta 138 (2015) 273–278
ΓDNA =
Q z NA , nFA m
where Γ is the surface coverage of DNA, n is the number of electrons per redox event, F is the Faraday constant, A is the area of the working electrode, m is the number of bases in the duplex DNA, z is the charge number, and Q is the capacitive charge. To quantify the surface density of dsDNA immobilized on Au electrode, we used chronocoulometry to calculate the redox capacitive charge in 5 mmol/L PBS buffer solution containing 100 μmol/L RuHex or without RuHex. Chronocoulometric response curves of dsDNA/Au are shown in Fig. 3. The coverage of DNA on the Au electrode was approximately 12.40 pmol/cm2, which corresponded with the theoretical estimation of 8.64 pmol/cm2 [15]. 3.3. Bi-functionalized aptasensor for high sensitivity detection of thrombin 3.3.1. ECL method detection of thrombin Electrochemically reduced or oxidized CdSe QDs react with coreactants to generate an ECL signal. In this experiment, the CdSe QDs and coreactant S2O82 were reduced to QD and SO4 , respectively upon a negative potential scan. When QD reacted with SO4 , an excited state QD* was generated and the ECL signal was emitted as the excited state QD* went back to the ground state QD.
(q) (p) (o) (n) (m) (l) (k) (j) (i) (h) (g) (f) (e) (d) (c) (b) (a)
-Z''/ohm
15000
10000
5000
0 0
9000
18000 Z''/ohm
480
320
160
0 0
100
200
300
t/s Fig. 6. ECL intensity–time curve of thrombin/QDs/DNA/Au electrode under continuous cyclic voltammetry scanning for 10 cycles with thrombin concentrations of 10 10 μg mL 1.
300
150
0
27000
Fig. 7. Selectivity of the ECL to thrombin by comparing the interfering species at 10-fold concentration: ATP, BSA, L-histidine, lysozyme, IgG, casein. The error bars show the standard deviation of three replicate determination. (ΔECL intensity is the difference value of ECL signal before and after the sensor incubated in samples.)
24000 diameter
Fig. 4(A) illustrates the ECL intensity of this aptasensor for the detection of different concentrations of thrombin. The ECL signal was dependent on the amount of CdSe QDs on the electrode surface. When a thrombin molecule affinity bound with an aptamer, one single-stranded DNA modified CdSe QDs would leave the electrode surface resulting in an ECL intensity decrease. Fig. 4(B) shows the relationship between ECL intensity and thrombin concentrations. As seen from the curve, the ECL intensity gradually reduced with the addition of thrombin. The ECL intensity was proportional to the logarithm of the thrombin concentration within the range of 10 10–10 3 μg mL 1 (R¼ 0.9924) with a detection limit of 10 10 μg mL 1 (2.7 amol/L).
ECL intensity
modified with DNA was soaked in an electrolyte of low ionic strength containing multiple redox cations, the cations electrostatically stabilized on the interface and induced a positive potential. At steady state, the potential would be fairly constant because the electroactivity would be reduced. The DNA density equation is as follows:
Δ ECL intensity
276
21000
21000
16000
14000
14000 7000
8000
-10
-5
0
5
10
-1 Cthrombin / (ug.mL ) Fig. 5. (A) The EIS image of bi-functionalized aptasensor with different concentrations of thrombin. (a) Au bar electrode, (b) DNA/Au electrode. (c–q) used the CdSe QDs/DNA/Au electrodes for various concentrations of thrombin in solution: (c) 0, (d) 10 10, (e) 10 9, (f) 10 8, (g) 10 7, (h) 10 6, (i) 10 5, (j) 10 4, (k) 10 3, (l) 10 2, (m) 10 1, (n) 1, (o) 2, (p) 5, and (q) 10 μg/mL thrombin. (B) Relationship between EIS and thrombin concentrations. Inset: logarithmic calibration curve for thrombin.
Δ
7000
0
0
Fig. 8. Selectivity of the EIS to thrombin by comparing to the interfering species at 10-fold concentration: ATP, BSA, L-histidine, lysozyme, IgG, casein. The error bars show the standard deviation of three replicate determination. (ΔRet value is the difference of Ret signal before and after the sensor incubated in samples.)
L. Lu et al. / Talanta 138 (2015) 273–278
277
Table 1 Comparison of various assay methods for the detection of thrombin. Methods
Sensing probe
Detection limit
Linear range
Reproducibility (RSD)
Ref.
Electrochemistry ECL Fluorescent EIS Surface plasmon resonance Chronocoulometry Colorimetry ECL and EIS
Methylene blue CdS:Eu Green fluorescence protein [Fe(CN)6]3 /4
0.143 pM 50 aM 3.0 10 4 U mL 1 0.06 nM 0.1 nM 30 pM 3.2 fM 2.7 aM
10 5–10 12 M 50 aM–1 pM 3.0 10 4–5.0 10 2 U mL 1 0.12–30 nM 0.1–75 nM 0.1–18.5 nM 0.01–0.5 pM 2.7 aM–2.7 nM
Not mentioned ‘Stable enough’ 1.4% 4.5% 13% 5.8% 9.43% 5.6%
[16] [17] [18] [19] [20] [21] [22] This work
[Ru(NH3)6]3 þ HAuCl4–NH2OH CdSe QDs
3.3.2. EIS method detection of thrombin EIS can provide information about surface properties of different modified electrodes. Fig. 5(A) illustrates the EIS detection method of the modified Au electrode for different stages. Fig. 5(B) displays the relationship between the electrode resistance and thrombin concentrations. As seen, the resistance of the DNA/Au electrode (b) is greater than that of the Au bar electrode (a). This difference could be attributed to the electrostatic interaction between DNA phosphate backbones, which carries negative charges, and electroactive [Fe(CN)6]3 /4 molecules. When CdSe QDs were immobilized on the DNA/Au electrode through biotin–avidin conjugation, the resistance was further increased. This is mainly because the CdSe QD molecule is a semi-conductor. In Fig. 5(A), curves d–q represent the impedance measurements for different concentrations of thrombin; the resistance gradually increases with increasing concentrations of thrombin. When a thrombin molecule combines with an aptamer, the resulting complex would remain on the electrode surface. Because the complex cannot transfer electrons, the conductivity of the overall electrode decreases. Therefore, the resistance increases with the thrombin concentration. Additionally, impedance is proportional to the concentration of thrombin. Fig. 5(B) shows the linear relationship between the thrombin concentration and impedance. Impedance was proportional to the logarithm of the thrombin concentration within the range of 10 10–10 1 μg mL 1 (R¼ 0.9875) with a detection limit of 10 10 μg mL 1 (2.7 aM). As we know, the thrombin concentration varies from nmol/L to μmol/L in blood. So, this assay is a promising candidate for simple, sensitive and costsaving thrombin detection in clinic field. 3.4. Stability of bi-functionalized aptasensor Fig. 6 shows the ECL image of bi-functionalized aptasensors under continuous cyclic voltammetry scanning for 10 cycles with a thrombin concentration of 10 10 μg mL 1. As is clearly shown in the figure, the ECL intensity of aptasensors had no apparent change during continuous cyclic voltammetry scanning, which demonstrated its good stability. Additionally, when the thrombin/ QDs/DNA/Au electrode was stored at 4 °C for 28 days, the ECL intensity of the modified electrode decreased approximately 8–10%, which further illustrated the good stability of the bi-functionalized aptasensor. The fabrication reproducibility of five electrodes, made independently, showed an acceptable reproducibility with average relative standard deviation(RSD) of 5.6% for detection thrombin at 1 ug/mL. 3.5. Selectivity of bi-functionalized aptasensor As signaling is based on a specific binding between thrombin and aptamer, the sensor should be insensitive to nonspecific binding. In this work, we tested six potential interfering species including ATP, BSA, L-histidine, lysozyme, mouse IgG, and casein at a concentration 10-fold higher than thrombin, using the same experimental procedures as those for thrombin. The results showed that these species
did not induce any great changes of signal. As shown in Fig.7, the change of ECL intensities (ΔECL), which was ECL difference between before and after the aptasensor incubation in species, were much smaller than the thrombin. Moreover, a mixed sample (thrombin coexisted with 10-fold concentration ATP, BSA, L-histidine, lysozyme, mouse IgG, and casein ) did not display large change compared with that of thrombin alone. Similarly, the changes of EIS signal (ΔEIS) were much smaller than thrombin (Fig. 8). The above results fully demonstrated that the bi-functionalized aptasensor was good selectivity to thrombin. We compared our results with various thrombin detection methods published in recent years (Table 1). Our bi-functionalized aptasensor, which employed ECL and EIS methods, performed optimally displaying a lower detection limit, higher sensitivity, strong specificity, and good selectivity.
4. Conclusions This article described a novel bi-functionalized aptasensor for the detection of thrombin. ECL emitter QDs were assembled on the electrode surface through biotin–avidin conjugation to immobilized dsDNA. In the presence of a thrombin solution, thrombin molecules quickly associated with aptamers, which decreased the ECL intensity and increased the EIS signal. The ECL intensity and impedance were proportional to the thrombin concentration; therefore, we utilized both ECL and EIS measurement methods for the detection of thrombin. Additionally, the bi-functionalized aptasensor displays potential scientific and medical applications due to its good specificity and sensitivity for thrombin.
Acknowledgments We thank the Foundation of National Natural Science Foundation of China (Nos. 21005005, 21375005, and 21475006), Beijing Nova Program (No. 2010B009) and the Program for New Century Excellent Talents in University (NCET-12-0603).
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
A.E. Ellington, J.W. Szostak, Nature 346 (1990) 818–822. C. Tuerk, L. Gold, Science 249 (1990) 505–510. D.L. Robertson, G.F. Joyce, Nature 344 (1990) 467–468. K.H. Hsieh, Thromb. Res. 86 (1997) 301–316. M. Michaud, E. Jourdan, A. Villet, A. Ravel, C. Grosset, E. Peyrin, J. Am. Chem. Soc. 125 (2003) 8672–8679. Q. Zhao, X. Li, Y. Shao, Anal. Chem. 80 (2008) 7586–7593. V. Pavlov, B. Shlyahovsky, J. Am. Chem. Soc. 127 (2005) 6522–6523. Q. Tang, X. Su, K.P. Loh, J. Colloid Interface Sci. 315 (2007) 99–106. H. Yi, W. Xu, Y. Yuan, Y. Wu, Y. Chai, R. Yuan, Biosens. Bioelectron. 47 (2013) 368–372. X. Wang, A. Gao, C. Lu, X. He, X. Yin, Biosens. Bioelectron. 48 (2013) 120–125. Y. Yu, Y. Liang, Y. Lai, Y. Cao, J. Anal. Sci. 25 (2009) 59–62.
278
[12] [13] [14] [15] [16] [17] [18]
L. Lu et al. / Talanta 138 (2015) 273–278
M. Stanisavljevic, L. Janu, K. Smerkova, Chromatographia 76 (2013) 335–343. W.W. Yu, L. Qu, W. Guo, X. Peng, Chem. Mater. 15 (2003) 2854–2860. A.B. Steel, T.M. Herne, M.J. Tarlov, Anal. Chem. 70 (1998) 4670–4677. H. Yu, C. Luo, C.G. Sankar, Anal. Chem. 75 (2003) 3902–3907. H. Bai, F.J. Campo, Y.C. Tsai, Biosens. Bioelectron. 42 (2013) 17–22. L. Deng, Y. Du, J.J. Xu, H.Y. Chen, Biosens. Bioelectron. 59 (2014) 58–63. Y. Zhang, B. Li, Y. Jin, Analyst 136 (2011) 3268–3273.
[19] [20] [21] [22]
X. Li, L. Shen, D. Zhang, Biosens. Bioelectron. 23 (2008) 1624–1630. Y. Bai, F. Feng, L. Zhao, Biosens. Bioelectron. 43 (2013) 265–270. X. Jiao, J. Chen, X. Zhang, H. Luo, N. Li, Anal. Biochem. 441 (2013) 95–100. Y. Niu, P. Wang, Y. Zhao, A. Fan, Analyst 138 (2003) 1475–1482.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Bi-functionalized aptasensor for ultrasensitive detection of thrombin Liping Lu n, Jiao Li, Tianfang Kang, Shuiyuan Cheng Key Laboratory of Beijing on Regional Air Pollution Control, Beijing University of Technology, Beijing 100124, China
art ic l e i nf o
a b s t r a c t
Article history: Received 27 November 2014 Received in revised form 7 March 2015 Accepted 10 March 2015 Available online 18 March 2015
A novel bi-functionalized aptasensor for thrombin detection was fabricated by using electrochemiluminescence (ECL) and electrochemical impedance spectroscopy (EIS) techniques. A thiol-terminated aptamer with 15 oligonucleotides was hybridized with its complementary oligonucleotides to form a double-stranded DNA (dsDNA). Then, the thiol-labeled dsDNA was assembled on a Au electrode via Au–S bond; the other distal of the dsDNA labeled with biotin was bound to QDs through a biotin–avidin conjugation. When thrombin is present in the detection solution, the aptamer can combine with thrombin, resulting in loss of CdSe QDs from the electrode surface and thereby decreasing the ECL intensity and increasing the impedance. The decreased ECL and increased EIS signals are logarithmically linear with respect to the concentration of thrombin. The linear range was 10 10–10 3 μg mL 1 (R¼ 0.9924) for the ECL signal and 10 10–10 1 μg mL 1 (R¼0.9875) for the EIS method with a common detection limit of 10 10 μg mL 1 (2.7 aM). In addition, the bi-functionalized aptasensor exhibited excellent selectivity, super sensitivity, a low detection limit and a wide linear range. & 2015 Elsevier B.V. All rights reserved.
Keywords: Aptasensor Thrombin ECL DNA Quantum dots
1. Introduction In the early 1990s, three research groups (Szostak [1], Tuerk [2] and Joyce [3]) in the US independently reported that there were oligonucleotide sequences with high affinities for proteins, nucleic acids or other small molecules that could be isolated via in vitro selection and amplification. The selected oligonucleotide sequences were named aptamers. Due to the high specificity between aptamers and target molecules, aptamers have received wide attention and have been used for broad application in various scientific fields, such as sensor preparations, drug screenings, and disease diagnoses and treatments. Aptasensors, which are based on aptamers as molecular recognition substances, have since been widely applied to detect proteins, genes and drugs. Thrombin, a very important serine protease in blood coagulation, can directly transform soluble fibrinogen into insoluble fibrin, accelerate blood coagulation and facilitate blood hemostasis [4]. If the level of thrombin in the body is not maintained within a certain range, thromboembolic diseases or even death would result. The concentration and activity of thrombin can also be used as an important indicator of the clotting mechanism, which is also of great importance to the early diagnosis, efficacious examination and prognosis of some types of diseases. Under normal circumstances, the thrombin concentration is nmol/L. The establishment of a rapid and accurate thrombin detection method with high sensitivity is of n
Corresponding author. Tel.: þ 86 10 67391659; fax: þ86 10 67391983. E-mail address: [email protected] (L. Lu).
http://dx.doi.org/10.1016/j.talanta.2015.03.016 0039-9140/& 2015 Elsevier B.V. All rights reserved.
vital significance in medicine. Currently, there are multiple ways to detect thrombin, such as high performance liquid chromatography (HPLC) [5], chromatography [6], fluorescence-based methods [7], surface plasmon resonance [8], electrochemical methods [9], and the electrochemiluminescence (ECL) method [10]. As we know, electrochemical and ECL methods have widely used in various scientific fields for its simple device requirements, reduced sample and reagent consumption and use easily. In this study, we designed a bi-functionalized aptasensor based on the aptamer-specific binding of thrombin to detect thrombin with high sensitivity and high specificity. The selected thrombin aptamer-15 (Apt15) has a high affinity to fibrinogen binding sites of thrombin. The ECL intensity and EIS were proportional to the concentration of thrombin; therefore, we used both ECL and EIS methods to detect thrombin. This bi-functionalized method is a more reliable detection system in various scientific and medical fields. These advantages show the great potentiality of our sensor for application in anticoagulation drugs screening and cardiovascular and thromboembolic disease diagnosis in clinical field.
2. Experimental section 2.1. Chemicals and materials CdCl2 2.5H2O and K2S2O8 were purchased from Beijing Chemical Reagent Corporation. Selenium was purchased from Shanghai Meixing Chemical Corporation. NaBH4, Na2HPO4 12H2O,
274
L. Lu et al. / Talanta 138 (2015) 273–278
NaH2PO4 2H2O and KCl were purchased from Tianjin Fuchen Chemical Corporation. MgCl2 was purchased from Shanghai Sangon Biotechnology Corporation. Mercaptopropionic acid (MPA), 6-mercapto-1-hexanol (MCH), avidin (from egg white), adenosine 5′-triphosphate (ATP), L-histidine, lysozyme, bovine serum albumin (BSA), mouse IgG, casein and thrombin (from human plasma) were purchased from Sigma-Aldrich (USA). All reagents were of analytical grade. Labeled DNA oligonucleotides were synthesized by Beijing SBS Gene Tech. Co. The sequences of the two oligonucleotides used are provided below: ssDNA-1: 5′-biotin-CCA ACC ACA CCA ACC-3′ ssDNA-2: 5′-SH-GGT TGG TGT GGT TGG-3′
2.2. Apparatus Cyclic voltammetry (CV) and EIS were performed with a CHI 660a Electrochemical Analyzer (Shanghai Chenhua Instrument Corporation). ECL emission measurements were conducted on an MPI-A ECL analyzer (Xi'an Remex Analysis Instrument Corporation). All experiments were carried out with a three-electrode system: an Au electrode as the working electrode, an Ag–AgCl electrode as the reference electrode, and a platinum wire as the counter electrode. Field emission electron microscopy was performed with a JEOL model 2100F instrument. UV–vis absorption spectra were obtained on a UV-2450 spectrophotometer (Shimadzu). Fluorescence emission spectra were obtained on a F-4600 fluorescence spectrophotometer (Hitachi). 2.3. Synthesis of CdSe/MPA quantum dots CdSe/MPA QDs were synthesized using a slightly modified procedure previously reported in [11]. Briefly, 0.0094 g CdCl2 2.5H2O was dissolved in 60 mL ultrapure water, followed by the addition of 9 μL MPA. The pH of the solution was adjusted to 11.0 by the dropwise addition of 1.0 mol/L NaOH solution. Afterward, the mixture was transferred to a three-necked flask and bubbled with pure N2 for 30 min. Then, 500 μL of freshly prepared NaHSe solution was slowly injected, and the solution was refluxed for 2 h at 100 °C to obtain a clear, light yellow solution of CdSe/MPA QDs. After cooling, the CdSe/MPA QDs solution was centrifuged at 6000 rpm for 5 min. The final CdSe/MPA QDs solution was stored in a refrigerator at 4 °C.
2.4. Pretreatment of the gold electrode Prior to use, the Au electrode was polished with 0.02–0.05 μm Al2O3 power on a polishing pad. Then, the electrode was cleaned with freshly made piranha solution (98% H2SO4:30% H2O2 ¼7:3) for 10 min, followed by an adequate ultrasonic cleaning in absolute ethanol and ultrapure water. Subsequently, the electrode was immersed in 0.5 mol/L H2SO4 for cyclic voltammetry scan with a potential range of 0.3 to 1.5 V. Then, it was washed with ultrapure water and dried with N2. 2.5. Preparation of bi-functionalized aptasensor The schematic illustration of the bi-functionalized aptasensor for the detection of thrombin is shown in Scheme 1. First, 25 μL of 50 μmol/L dsDNA was dropped on a pre-cleaned Au electrode surface and incubated for 16–24 h in a humidified environment. Thus, thiol-terminated dsDNA was assembled on the Au electrode via a Au–S bond; this electrode was denoted as the DNA/Au electrode. The dsDNA was obtained by mixing ssDNA-1 and ssDNA-2 with the same concentration and volume, and then heated at 90 °C and slowly cooled to room temperature. After a thorough rinse with 0.1 mol/L PBS (0.1 mol/L HPO42 /H2PO4 , 0.1 mol/L KCl, pH 7.4) buffer solution, the DNA/Au electrode was backfilled with 0.1 mol/L MCH for 1 h to block the nonspecifically bound oligonucleotides. Then, the DNA/ Au electrode was adequately rinsed with 0.1 mol/L PBS buffer solution and immersed in a QDs–avidin solution under shocking conditions for 1 h. Thus, the other end of the dsDNA was firmly bound with CdSe QDs through a biotin–avidin conjugation. The electrode was denoted as the CdSe QDs/DNA/Au electrode. In the presence of thrombin in the solution, the aptamer would combine with thrombin, and CdSe QDs with complementary oligonucleotides would leave the surface of electrode. The CdSe QDs–avidin solution was obtained by mixing CdSe QDs and avidin in a certain proportion [12] and centrifuged at 13,000 rpm for 80 min. The supernatant was discarded and the precipitate was CdSe QDs–avidin which was dissolved in ultrapure water. 2.6. ECL and EIS methods detection of thrombin For the detection of thrombin, bi-functionalized aptasensors were incubated in a thrombin solution for 40 min at 37 °C, and then washed with a 0.1 mol/L PBS buffer solution.
Scheme 1. The preparation of bi-functionalized aptasensor and the detection of thrombin.
L. Lu et al. / Talanta 138 (2015) 273–278
b
1.0
2250
614nm
1500
506nm a
0.5
750
0.0
0 480
560 640 Wavelength(nm)
PL intensity
1.5
Absorbance
The measurement of thrombin was performed with the ECL method at a potential ranging from 1.5 to 0.0 V at a scan rate of 100 mV/s with a three electrode system. Prior to and after the incubation in the thrombin solution, the ECL intensity was recorded in a 0.1 mol/L PBS (pH 7.4) solution containing 0.1 mol/L K2S2O8 as a coreactant. The EIS measurements were carried out on a CHI 660a Electrochemical Analyzer. Prior to and after the incubation in the thrombin solution, the impedance spectra at a frequency range from 1 to 105 Hz and an amplitude of 5 mV was recorded in a 0.1 mol/L KCl solution containing 5.0 mmol/L K3[Fe(CN)6]/K4[Fe(CN)6].
275
720
Fig. 2. The absorbance spectra (a) and fluorescence spectra (b) of CdSe QDs.
3. Results and discussion
2.1
-6
The morphologies of CdSe QDs were characterized by high resolution transmission electron microscopy (HRTEM) imaging (Fig. 1). As seen from the image, the average size of the prepared CdSe QDs was approximately 2–3 nm with a spherical shape and good dispersion. Fig. 2(a) and (b) shows the UV–vis absorbance spectra and fluorescence spectra of CdSe QDs, respectively. From the UV–vis absorbance spectra, we can see that the absorption peak occurred at 506 nm, suggesting the size of CdSe QDs to be 2.4 nm [13], which corresponded with the diameter of CdSe QDs observed from HRTEM images. From the fluorescence spectra, we can see that the photoluminescence peak occurred at 614 nm, indicating the consequence of quantum confinement.
Charge× 10 /C
3.1. Characterization of the prepared CdSe QDs
a
1.4
0.7
b
0.0 0.00
0.15
0.30 1/2
0.45
1/2
Time /s Fig. 3. Chronocoulometric response RuHex þ5 mM PBS and (b) 5 mM PBS.
curves
of
dsDNA/Au:
(a)
100 μM
3.2. Coverage of dsDNA on Au electrode surface
450 ECL intensity
To confirm the density of dsDNA on the electrode surface, Tarlov [14] first calculated the redox charge on the surface of the Ru(NH3)63 þ (RuHex) molecule by chronocoulometry. They proposed a mechanism wherein the RuHex molecule is electrostatically absorbed to DNA phosphate backbone, which carries a negative charge. Because the density of RuHex fixed on the surface of the electrode is directly proportional to the density of DNA, we can calculate the surface density of the dsDNA. When the electrode
300
150
0 0
10
20
30
t/s
ECL intensity
450
300
150
0
5
10 -1
Cthrombin / (ug.mL )
Fig. 1. High resolution transmission electron microscopy (HRTEM) images of CdSe QDs.
Fig. 4. (A) The ECL intensity of bi-functionalized aptasensor with different concentrations of thrombin (from top to bottom: 10 10, 10 9, 10 8, 10 7, 10 6, 10 5, 10 4, and 10 3 μg mL 1). (B) Relationship between ECL intensity and thrombin concentrations. Inset: logarithmic calibration curve for thrombin.
L. Lu et al. / Talanta 138 (2015) 273–278
ΓDNA =
Q z NA , nFA m
where Γ is the surface coverage of DNA, n is the number of electrons per redox event, F is the Faraday constant, A is the area of the working electrode, m is the number of bases in the duplex DNA, z is the charge number, and Q is the capacitive charge. To quantify the surface density of dsDNA immobilized on Au electrode, we used chronocoulometry to calculate the redox capacitive charge in 5 mmol/L PBS buffer solution containing 100 μmol/L RuHex or without RuHex. Chronocoulometric response curves of dsDNA/Au are shown in Fig. 3. The coverage of DNA on the Au electrode was approximately 12.40 pmol/cm2, which corresponded with the theoretical estimation of 8.64 pmol/cm2 [15]. 3.3. Bi-functionalized aptasensor for high sensitivity detection of thrombin 3.3.1. ECL method detection of thrombin Electrochemically reduced or oxidized CdSe QDs react with coreactants to generate an ECL signal. In this experiment, the CdSe QDs and coreactant S2O82 were reduced to QD and SO4 , respectively upon a negative potential scan. When QD reacted with SO4 , an excited state QD* was generated and the ECL signal was emitted as the excited state QD* went back to the ground state QD.
(q) (p) (o) (n) (m) (l) (k) (j) (i) (h) (g) (f) (e) (d) (c) (b) (a)
-Z''/ohm
15000
10000
5000
0 0
9000
18000 Z''/ohm
480
320
160
0 0
100
200
300
t/s Fig. 6. ECL intensity–time curve of thrombin/QDs/DNA/Au electrode under continuous cyclic voltammetry scanning for 10 cycles with thrombin concentrations of 10 10 μg mL 1.
300
150
0
27000
Fig. 7. Selectivity of the ECL to thrombin by comparing the interfering species at 10-fold concentration: ATP, BSA, L-histidine, lysozyme, IgG, casein. The error bars show the standard deviation of three replicate determination. (ΔECL intensity is the difference value of ECL signal before and after the sensor incubated in samples.)
24000 diameter
Fig. 4(A) illustrates the ECL intensity of this aptasensor for the detection of different concentrations of thrombin. The ECL signal was dependent on the amount of CdSe QDs on the electrode surface. When a thrombin molecule affinity bound with an aptamer, one single-stranded DNA modified CdSe QDs would leave the electrode surface resulting in an ECL intensity decrease. Fig. 4(B) shows the relationship between ECL intensity and thrombin concentrations. As seen from the curve, the ECL intensity gradually reduced with the addition of thrombin. The ECL intensity was proportional to the logarithm of the thrombin concentration within the range of 10 10–10 3 μg mL 1 (R¼ 0.9924) with a detection limit of 10 10 μg mL 1 (2.7 amol/L).
ECL intensity
modified with DNA was soaked in an electrolyte of low ionic strength containing multiple redox cations, the cations electrostatically stabilized on the interface and induced a positive potential. At steady state, the potential would be fairly constant because the electroactivity would be reduced. The DNA density equation is as follows:
Δ ECL intensity
276
21000
21000
16000
14000
14000 7000
8000
-10
-5
0
5
10
-1 Cthrombin / (ug.mL ) Fig. 5. (A) The EIS image of bi-functionalized aptasensor with different concentrations of thrombin. (a) Au bar electrode, (b) DNA/Au electrode. (c–q) used the CdSe QDs/DNA/Au electrodes for various concentrations of thrombin in solution: (c) 0, (d) 10 10, (e) 10 9, (f) 10 8, (g) 10 7, (h) 10 6, (i) 10 5, (j) 10 4, (k) 10 3, (l) 10 2, (m) 10 1, (n) 1, (o) 2, (p) 5, and (q) 10 μg/mL thrombin. (B) Relationship between EIS and thrombin concentrations. Inset: logarithmic calibration curve for thrombin.
Δ
7000
0
0
Fig. 8. Selectivity of the EIS to thrombin by comparing to the interfering species at 10-fold concentration: ATP, BSA, L-histidine, lysozyme, IgG, casein. The error bars show the standard deviation of three replicate determination. (ΔRet value is the difference of Ret signal before and after the sensor incubated in samples.)
L. Lu et al. / Talanta 138 (2015) 273–278
277
Table 1 Comparison of various assay methods for the detection of thrombin. Methods
Sensing probe
Detection limit
Linear range
Reproducibility (RSD)
Ref.
Electrochemistry ECL Fluorescent EIS Surface plasmon resonance Chronocoulometry Colorimetry ECL and EIS
Methylene blue CdS:Eu Green fluorescence protein [Fe(CN)6]3 /4
0.143 pM 50 aM 3.0 10 4 U mL 1 0.06 nM 0.1 nM 30 pM 3.2 fM 2.7 aM
10 5–10 12 M 50 aM–1 pM 3.0 10 4–5.0 10 2 U mL 1 0.12–30 nM 0.1–75 nM 0.1–18.5 nM 0.01–0.5 pM 2.7 aM–2.7 nM
Not mentioned ‘Stable enough’ 1.4% 4.5% 13% 5.8% 9.43% 5.6%
[16] [17] [18] [19] [20] [21] [22] This work
[Ru(NH3)6]3 þ HAuCl4–NH2OH CdSe QDs
3.3.2. EIS method detection of thrombin EIS can provide information about surface properties of different modified electrodes. Fig. 5(A) illustrates the EIS detection method of the modified Au electrode for different stages. Fig. 5(B) displays the relationship between the electrode resistance and thrombin concentrations. As seen, the resistance of the DNA/Au electrode (b) is greater than that of the Au bar electrode (a). This difference could be attributed to the electrostatic interaction between DNA phosphate backbones, which carries negative charges, and electroactive [Fe(CN)6]3 /4 molecules. When CdSe QDs were immobilized on the DNA/Au electrode through biotin–avidin conjugation, the resistance was further increased. This is mainly because the CdSe QD molecule is a semi-conductor. In Fig. 5(A), curves d–q represent the impedance measurements for different concentrations of thrombin; the resistance gradually increases with increasing concentrations of thrombin. When a thrombin molecule combines with an aptamer, the resulting complex would remain on the electrode surface. Because the complex cannot transfer electrons, the conductivity of the overall electrode decreases. Therefore, the resistance increases with the thrombin concentration. Additionally, impedance is proportional to the concentration of thrombin. Fig. 5(B) shows the linear relationship between the thrombin concentration and impedance. Impedance was proportional to the logarithm of the thrombin concentration within the range of 10 10–10 1 μg mL 1 (R¼ 0.9875) with a detection limit of 10 10 μg mL 1 (2.7 aM). As we know, the thrombin concentration varies from nmol/L to μmol/L in blood. So, this assay is a promising candidate for simple, sensitive and costsaving thrombin detection in clinic field. 3.4. Stability of bi-functionalized aptasensor Fig. 6 shows the ECL image of bi-functionalized aptasensors under continuous cyclic voltammetry scanning for 10 cycles with a thrombin concentration of 10 10 μg mL 1. As is clearly shown in the figure, the ECL intensity of aptasensors had no apparent change during continuous cyclic voltammetry scanning, which demonstrated its good stability. Additionally, when the thrombin/ QDs/DNA/Au electrode was stored at 4 °C for 28 days, the ECL intensity of the modified electrode decreased approximately 8–10%, which further illustrated the good stability of the bi-functionalized aptasensor. The fabrication reproducibility of five electrodes, made independently, showed an acceptable reproducibility with average relative standard deviation(RSD) of 5.6% for detection thrombin at 1 ug/mL. 3.5. Selectivity of bi-functionalized aptasensor As signaling is based on a specific binding between thrombin and aptamer, the sensor should be insensitive to nonspecific binding. In this work, we tested six potential interfering species including ATP, BSA, L-histidine, lysozyme, mouse IgG, and casein at a concentration 10-fold higher than thrombin, using the same experimental procedures as those for thrombin. The results showed that these species
did not induce any great changes of signal. As shown in Fig.7, the change of ECL intensities (ΔECL), which was ECL difference between before and after the aptasensor incubation in species, were much smaller than the thrombin. Moreover, a mixed sample (thrombin coexisted with 10-fold concentration ATP, BSA, L-histidine, lysozyme, mouse IgG, and casein ) did not display large change compared with that of thrombin alone. Similarly, the changes of EIS signal (ΔEIS) were much smaller than thrombin (Fig. 8). The above results fully demonstrated that the bi-functionalized aptasensor was good selectivity to thrombin. We compared our results with various thrombin detection methods published in recent years (Table 1). Our bi-functionalized aptasensor, which employed ECL and EIS methods, performed optimally displaying a lower detection limit, higher sensitivity, strong specificity, and good selectivity.
4. Conclusions This article described a novel bi-functionalized aptasensor for the detection of thrombin. ECL emitter QDs were assembled on the electrode surface through biotin–avidin conjugation to immobilized dsDNA. In the presence of a thrombin solution, thrombin molecules quickly associated with aptamers, which decreased the ECL intensity and increased the EIS signal. The ECL intensity and impedance were proportional to the thrombin concentration; therefore, we utilized both ECL and EIS measurement methods for the detection of thrombin. Additionally, the bi-functionalized aptasensor displays potential scientific and medical applications due to its good specificity and sensitivity for thrombin.
Acknowledgments We thank the Foundation of National Natural Science Foundation of China (Nos. 21005005, 21375005, and 21475006), Beijing Nova Program (No. 2010B009) and the Program for New Century Excellent Talents in University (NCET-12-0603).
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
A.E. Ellington, J.W. Szostak, Nature 346 (1990) 818–822. C. Tuerk, L. Gold, Science 249 (1990) 505–510. D.L. Robertson, G.F. Joyce, Nature 344 (1990) 467–468. K.H. Hsieh, Thromb. Res. 86 (1997) 301–316. M. Michaud, E. Jourdan, A. Villet, A. Ravel, C. Grosset, E. Peyrin, J. Am. Chem. Soc. 125 (2003) 8672–8679. Q. Zhao, X. Li, Y. Shao, Anal. Chem. 80 (2008) 7586–7593. V. Pavlov, B. Shlyahovsky, J. Am. Chem. Soc. 127 (2005) 6522–6523. Q. Tang, X. Su, K.P. Loh, J. Colloid Interface Sci. 315 (2007) 99–106. H. Yi, W. Xu, Y. Yuan, Y. Wu, Y. Chai, R. Yuan, Biosens. Bioelectron. 47 (2013) 368–372. X. Wang, A. Gao, C. Lu, X. He, X. Yin, Biosens. Bioelectron. 48 (2013) 120–125. Y. Yu, Y. Liang, Y. Lai, Y. Cao, J. Anal. Sci. 25 (2009) 59–62.
278
[12] [13] [14] [15] [16] [17] [18]
L. Lu et al. / Talanta 138 (2015) 273–278
M. Stanisavljevic, L. Janu, K. Smerkova, Chromatographia 76 (2013) 335–343. W.W. Yu, L. Qu, W. Guo, X. Peng, Chem. Mater. 15 (2003) 2854–2860. A.B. Steel, T.M. Herne, M.J. Tarlov, Anal. Chem. 70 (1998) 4670–4677. H. Yu, C. Luo, C.G. Sankar, Anal. Chem. 75 (2003) 3902–3907. H. Bai, F.J. Campo, Y.C. Tsai, Biosens. Bioelectron. 42 (2013) 17–22. L. Deng, Y. Du, J.J. Xu, H.Y. Chen, Biosens. Bioelectron. 59 (2014) 58–63. Y. Zhang, B. Li, Y. Jin, Analyst 136 (2011) 3268–3273.
[19] [20] [21] [22]
X. Li, L. Shen, D. Zhang, Biosens. Bioelectron. 23 (2008) 1624–1630. Y. Bai, F. Feng, L. Zhao, Biosens. Bioelectron. 43 (2013) 265–270. X. Jiao, J. Chen, X. Zhang, H. Luo, N. Li, Anal. Biochem. 441 (2013) 95–100. Y. Niu, P. Wang, Y. Zhao, A. Fan, Analyst 138 (2003) 1475–1482.
