Talanta 131 (2015) 706–711
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Electrochemical biotin detection based on magnetic beads and a new magnetic flow cell for screen printed electrode Julien Biscay, María Begoña González García, Agustín Costa García n Department of Physical and Analytical Chemistry, Faculty of Chemistry, University of Oviedo, 33006 Oviedo, Spain
art ic l e i nf o
a b s t r a c t
Article history: Received 25 April 2014 Received in revised form 3 August 2014 Accepted 5 August 2014 Available online 27 August 2014
The use of the first flow-cell for magnetic assays with an integrated magnet is reported here. The flow injection analysis system (FIA) is used for biotin determination. The reaction scheme is based on a one step competitive assay between free biotin and biotin labeled with horseradish peroxidase (B-HRP). The mixture of magnetic beads modified with streptavidin (Strep-MB), biotin and B-HRP is left 15 min under stirring and then a washing step is performed. After that, 100 μL of the mixture is injected and after 30 s 100 μL of 3,30 ,5,50 -Tetramethylbenzidine (TMB) is injected and the FIAgram is recorded applying a potential of 0.2 V. The linear range obtained is from 0.01 to 1 nM of biotin and the sensitivity is 758 nA/nM. The modification and cleaning of the electrode are performed in an easy way due to the internal magnet of the flow cell. & 2014 Elsevier B.V. All rights reserved.
Keywords: Biotin detection Flow injection analysis Flow cell with an integrated magnet Screen printed electrode Magnetic beads
1. Introduction Flow injection analysis (FIA) techniques have been employed to automate a wide variety of chemical and biochemical analyses since their invention in the middle of the seventies [1,2]. FIA techniques are classified into the basic mode (also named normal flow injection analysis) and different special modes as stopped flow [3,4] or reverse flow [5,6]. FIA is one of the most popular continuous flow techniques and its simplicity, and flexibility allow its application in various chemistries. Moreover the flexibility in the different created formats and designs helps its introduction in laboratories using low cost instrumentation but each one has to comply with the three cornerstones of the FIA: (i) injection of discrete and well defined volume of sample solution into a flowing carrier stream; (ii) reproducible and precise timing of the manipulation that the injected sample zone is subjected to in the system, from the point of the injection to the point of detection; and (iii) the creation of a concentration gradient of the injected sample, providing a transient, but reproducible readout of the recorded signal. The second generation of FIA including stopped flow injection [7], bead injection [8–10], sequential injection with lab on valve [11] has opened new perspectives with many analysis systems. The majority of these systems have in common the detection of the product before the equilibrium state is reached, which considerably reduces the analysis time. To overcome
n
Corresponding author. E-mail address: [email protected] (A. Costa García).
http://dx.doi.org/10.1016/j.talanta.2014.08.013 0039-9140/& 2014 Elsevier B.V. All rights reserved.
shortcomings such as contamination of the surface of the detector or the lack of sensitivity, an appealing tool is used called the beadinjection analysis [12,13]. Magnetic beads (MBs) are generally used for it high surface area per volume. The ability to accommodate higher numbers of immobilized molecules helps to improve sensitivity and consequently detection limit of the assay. Another benefit is their easy manipulation with external magnets which makes it possible to perform biological reaction events away from the electrode surface [14] and reduce the complexity for sensing application. Moreover their uses minimize the matrix effect thanks to the washing procedures and faster assay kinetics that are achieved since the beads are in suspension [15]. Many articles have been published using both the advantages of magnetic beads and screen printed electrodes [16,17]; articles about FIA and its application in pharmaceutical [18,19] and biomedical analysis can be found [20] but articles on any assay using a magnetic flow cell with an integrated magnet for screen printed electrodes have not been reported. In this paper an electrochemical biotin assay using a flow injection system which includes for the first time a flow cell for magnetic assay with an integrated magnet is described. The present strategy is based on a competition scheme where biotin and the biotin-HRP compete for the binding sites of streptavidin, which are immobilized on the magnetic beads surface. After molecular recognition, the beads are injected into the FIA system and immobilized on the electrode surface with the help of the integrated magnet of the flow cell which is exactly situated underneath the working electrode of the screen printed electrode. After the injection of the substrate, enzymatic product reduc-
J. Biscay et al. / Talanta 131 (2015) 706–711
tion current is obtained, which is inversely related to biotin concentration. The electrochemical response is measured using the amperometric technique applying a constant potential. Moreover the packing of MBs on the SPCE surface also implied that the enzyme reaction product was generated very close to the electrode surface, thus allowing the steady-state to be reached rapidly (which implies faster measurements), and minimizing diffusion limitations of the electroactive species.
2. Experimental 2.1. Chemicals Sodium hydroxide (1.064.1000) was delivered by MERCK (Spain), magnetic beads of 1 μm diameter modified with streptavidin (Dynabeads s MyOne™ Streptavidin C1) (Strep-MB) (ref. 650.01) were purchased from Invitrogen, Biotinylated Horseradish Peroxidase (HRP-B) (ref. 29139) was supplied by Thermo Scientific and 3,30 ,5,50 tetramethylbenzidine (TMB) (ref. T0440), and Biotin (ref. B4501) were purchased from SIGMA. All chemicals were of analytical reagent grade, and the Milli-Q water used was obtained from a Millipore Direct-Q™ 5 purification system. Stock solution of 5 10 6 M of B-HRP, 5 10 3 M of Biotin and 7 108 magnetic beads per mL was daily prepared in 0.1 M phosphate buffer (PB) pH 7.2 and stored at 4 1C in a refrigerator. 2.2. Apparatus and electrodes Chronoamperometric measurements were performed using an ECO Chemie μAutolab type II potentiostat interfaced with a Pentium 166 computer system and controlled by the Autolab GPES software version 4.8 for Windows 98. All measurements were carried out at room temperature. Screen-Printed Carbon Electrodes (ref DRP-110), an edge connector (ref. DRP-CAC), were purchased from DropSens, S.L (Oviedo, Spain). The screen-printed electrodes consist of a Carbon working (4 mm diameter), carbon auxiliary and silver pseudo reference electrodes printed on an alumina substrate. An insulating layer serves to delimit the electrochemical cell and electric contacts. The magnetic flow cell (ref. CFLWCL-MAGN) was designed and purchased from DropSens, S.L (Oviedo, Spain). It is a methacrylate wall-jet Flow-Cell for FIA, with an unscrewed open–close system which allows an easy electrode replacement. This cell is designed to obtain an inlet flow perpendicular to the electrode0 s surface, and an outlet flow forming a 451 angle and an O-ring which limits the volume of the electrochemical cell. The integrated magnet is a cylindrical magnet which in the upper position is situated exactly above the working electrode and permits the modified magnetic beads to fix on the surface of the electrode. When the magnet is in the lower position, the magnetic beads are eliminated from the surface of the electrode. This new magnetic flow cell is represented in Fig. 1A. The cell is one of the parts of the flow injection analysis system which is schematically represented in Fig. 1B The FIA system used for the detection of biotin is a 12 cylinder Perimax Spetec peristaltic pump (Spetec GmbH, Germany) which allows the 0.1 M PB pH 7.2 stream to flow through the system. Desired solutions are injected by means of a six port rotary valve, Model 1106 (Omnifit Ltd., UK) equipped with a 100 μL loop. 2.3. Assay procedure The reaction scheme is based on a one step competitive assay between free biotin and biotin labeled with horseradish peroxidase (B-HRP). The mixture of magnetic beads modified with
707
streptavidin (Strep-MB), biotin and B-HRP is left for 15 min under stirring and then a washing step is performed. It consists of a resuspension of the beads in the 0.1 M PB solution at pH 7.2 and then a separation with the magnet to remove the supernatant. This operation is repeated 3 times. The scheme of the assay is reported in Fig. 2. 2.4. Electrochemical measurement Amperometric detection was performed applying a constant potential to the modified electrode. The carrier solution (phosphate buffer, pH 7.2) was pumped at a constant flow rate (1 mL/ min) until a stable baseline was recorded. Then, the flow rate was changed to 0.5 mL/min and the magnet was moved to the upper position. Finally, 30 s after the injection of 100 μL of the sample, the flow rate was changed to 1 mL/min and then 100 μL of TMB diluted 5 times (this dilution was necessary to avoid the oxidation peak due to the TMB itself) were injected into the system and the reduction current was observed and recorded 10 s after the injection of the substrate. Finally, the magnet is moved to the lower position and for 3 min at a flow rate of 3 mL/min the carrier is pumped in order to wash the electrode surface. After the washing step another analytical run can be performed. 2.5. Real sample measurement The methodology was tested in two pharmaceuticals with a known concentration of biotin. The first one is MEDEBIOTIN tablets (pharmaceutical 1) and second is MEDEBIOTIN vials (pharmaceutical 2). Each tablet (0.072 g) contains 5 mg of biotin, and each vial of 1 mL contains 4.6 mg of biotin. So 0.01 g of the pharmaceutical 1 was dissolved in 100 mL of the buffer. Then, an aliquot of this solution was diluted 1:50,000 times. Concerning the pharmaceutical 2, the vial was diluted in 1 L of the buffer and from this solution, a second dilution was performed (1:50,000). Finally mixtures of Strep-MB/B-HRP/Pharmaceutical were prepared and left under stirring for 15 min. Then a washing step was performed and finally the amperometric measurement was performed as explained in Section 2.4. Both samples were measured in triplicate.
3. Result and discussion 3.1. Optimization of the parameters of the flow injection analysis system 3.1.1. Flow rate 3.1.1.1. Injection of the mixture. The flow rate obtained when the mixture of reaction is injected in the FIA system is very important because there must be equilibrium between the pulling power of the flow carrier and the power of magnet in the flow cell that retains the magnetic beads. To carry out this study, a mixture of Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M), prepared as explained in Section 2.3 was injected at different flow rates of the carrier (0.5, 1, 1.5, 2 mL/min). Then TMB was injected and the reduction current was recorded applying a potential of 0.20 V and using a flow rate of 1 mL/min. The results obtained are summarized in Table 1. The analytical signal decreases when the flow rate increases because the number of Strep-MB fixed on the electrode surface is lower. At higher flow rates the pulling power of the flow carrier is higher than the capacity of the magnet to retain all magnetic beads on the electrode surface. In other words, the effect of cleaning is higher than the retention of the Strep-MB on the electrode surface and consequently, lower analytical signals are obtained. Thus, for further studies, a flow
708
J. Biscay et al. / Talanta 131 (2015) 706–711
Injection valve
Carrier
Sample Pump
Detector with integrated magnet Waste
Potentiostat
xxxxxx
Fig. 1. (A) Images of the magnetic flow cell, with the magnet up and down. (B) Scheme of the FIA system.
Fig. 2. Scheme of the competitive Biotin assay.
Table 1 Influence of the flow rate, during the injection of the mixture Strep-MBs/HRP-B, on the analytical response; Eapplied ¼ 0.2 V; TMB diluted 5 times; Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M). Each signal is the mean of three measurements. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2. Flow rate (mL/min) 0.5
1
1.5
2
( 55737 64)nA
( 5100 7100)nA
( 2850 7 132)nA
(-13757 106)nA
rate of 0.5 mL/min, for the injection of the mixtures, where higher analytical signals were obtained, has been chosen.
3.1.1.2. Cleaning of the electrode. The influence of the flow rate on the efficiency of the washing step also has been studied. The washing step consists of placing the magnet in the lower position, so that the carrier washes the surface of the electrode sending the magnetic beads to the waste. Without magnetic field, the StrepMBs are not retained on the electrode and the carrier can clean it.
A mixture of Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M) was prepared as explained in Section 2.3, and was injected with a flow rate of 0.5 mL/min. After the injection of the TMB and recording the reduction current at 0.20 V, a washing step is performed pumping a flow of 1, 2 and 3 mL/min (Fig. 3A, B and C, respectively) for 3 min in each case. To check the efficiency of the washing step, TMB was injected again. In the case of 1 mL/min, the washing step was totally inefficient because a similar reduction peak was observed after the second injection of TMB, whereas in the case of 2 mL/min a decrease in the intensity of the reduction peak was noted. Finally if the washing step was performed with a flow rate of 3 mL/min, the cleaning was more efficient because after the second injection of TMB, no reduction peak can be observed. All these results are summarized in Fig. 3. To complete this study, the washing step was performed, applying a flow rate of 3 mL/min, for 1, and 2 and 3 min. The reduction peak after the second injection of TMB totally disappeared with a washing step of 3 min (data not shown). It means that the surface of the electrode is enzyme free. So for further experiments the washing step will be performed for 3 min with a flow rate of 3 mL/min. 3.2. Optimization of the parameters that affect the analytical response 3.2.1. Optimization of the applied potential To determine the best detection potential, 40 μL of TMB was added on a SPCE and a cyclic voltammetry was recorded between 0.40 and þ0.80 V at scan rate of 50 mV/s. In Fig. 4A both
J. Biscay et al. / Talanta 131 (2015) 706–711
709
Intensity (nA)
0
-1500
-3000
-4500
0
0
-1500
-1500
Intensity (nA)
Intensity (nA)
-6000
-3000
-4500
Second injection ofTMB
-3000
-4500
-6000
-6000
Fig. 3. Optimization of the washing flow rate; Flow rate 1 ml/min (A), 2 ml/min (B) and 3 ml/min (C). Flow carrier 0.1 M phosphate buffer pH 7.2; TMB diluted 5 times; Eapplied ¼ 0.2 V; Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M).
not occur. Previously, a mixture of Strep-MB (7 107 per mL)/ B-HRP (5 10 7 M) was prepared as explained in Section 2.3 and injected into the FIA system. The results are summarized in Fig. 4B. The highest analytical response was obtained for 0.20 V. This potential was chosen for further studies.
-0.4
-0.2
0
0.2
0.4
0.6
0.8
Potential (V vs. Ag pseudoreference)
Potential (V vs Ag pseudoreference)
Intensity (nA)
0
0
-0.1
-0.2
-0.3
-2000
IðnAÞ ¼ 505:96C BHRP ðnMÞ 30:8
-4000 -4940 -6000
3.2.2. Concentration of HRP-B To determine the best concentration of B-HRP for the competitive assay, different mixtures of Strep-MB and HRP-B were prepared. The number of Strep-MB per mL was fixed to 7 107 and the concentration of B-HRP was between 5 10 12 and 5 10 8 M. The mixture was prepared as previously explained and the FIAgram was recorded applying -0.20 V as explained in Section 2.4. A linear relationship between current and B-HRP concentration in the range of 0.05 and 5 nM was obtained with a correlation coefficient of 0.9997 (n ¼7) according to the following equation:
-5025 -5575
-5585
Fig. 4. (A) Cyclic voltammogram of TMB in 0.1 M phosphate buffer solution pH 7.2. ν ¼ 50 mV/s. (B) Influence of the applied potential on the analytical response; StrepMB (7 107 per mL)/ B-HRP (5 10 7 M); TMB diluted 5 times. Each signal is the mean of three measurements. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2.
oxidation and reduction peaks of TMB are obtained between þ0.00 and þ0.20 V. So, a hydrodynamic curve was registered whenTMB is injected into the FIA system and potentials between 0.00 and 0.30 V are applied, where the oxidation of TMB does
This result indicates that the saturation of the binding sites of Strep-MB with the enzymatic conjugate B-HRP is reached for a concentration of 5 nM. For further experiments, 5 nM has been chosen as the concentration for B-HRP. By selecting this concentration, it is insured that the maximum slope is achieved. 3.2.3. Optimization of the number of magnetic beads per mL For the purpose to determine the best number of magnetic beads per mL,a mixture of Strep-MB/B-HRP with different number of magnetic beads per mL and a constant concentration of B-HRP were prepared. So in this study a number of magnetic beads comprised between 7 105 and 7 108 beads per mL were used and the concentration of B-HRP was 5 nM. The different mixtures were left under stirring for 15 min, and after washing, the chronoamperogram was recorded. The results are presented in
710
J. Biscay et al. / Talanta 131 (2015) 706–711
C Biotina (nM)
Table 2 Influence of the concentration of MBs on the analytical response; Eapplied ¼ -0.2 V; TMB diluted 5 times; B-HRP (5 10 9 M). Each signal is the mean of three measurements. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2.
0
7 108
7 107
7 106
7 105
( 35507 50) nA
( 1336 715) nA
( 546 715) nA
( 1337 15) nA
Intensity (nA)
Magnetic beads (per mL)
0
Intensity (nA)
-200 -400 -600
0
0.2
0.4
0.6
0.8
1
-400
-800
-1200 Fig. 6. Calibration curve of biotin under the optimized conditions assay. Strep-MB (7 107 per mL)/B-HRP (5 10 9 M); TMB diluted 5 times. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2. Each point is the mean of three measurements.
-800 -1000 -1200 -1400
Fig. 5. Amperometric detection of TMB in the flow injection analysis system with the magnetic flow-cell; TMB diluted 5 times; Eapplied ¼ 0.2 V; Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2 Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 9 M).
Table 2 and represent the mean of 3 signals obtained with the same SPCE. As it can be seen in Table 2 that the analytical response increases with the number of magnetic beads. Morever, for lower number of magnetic beads than 7 107, an important decrease in the signal can be noticed. Nevertheless, in order to be economically viable, 7 107 magnetic beads per mL for the following experiments was selected. 3.2.4. Reproducibility To check the reproducibility of the assay, a mixture of 7 107 magnetic beads per mL and 5 nM of B-HRP was prepared. After 15 min under stirring and a washing step, the mixture was injected into the FIA system. Then TMB diluted 5 times was injected and the peak current was recorded applying 0.20 V. After a washing step, the injection of the mixture and the TMB were repeated 5 times. The reduction peaks obtained are presented in Fig. 5. An analytical signal of 1320 719 nA (RSD: 1.5%) was obtained. These results demonstrate the robustness of the electrode and the efficiency of the washing step. 3.3. Biotin assay 3.3.1. Calibration curve The biotin assay was carried as explained in Section 2.3. Different mixtures of magnetic beads, B-HRP, and biotin were prepared. In each solution, the number of magnetic beads and the concentration of B-HRP were 7 107 per mL and 5 nM respectively, and the concentration of biotin was comprised between 5 10 11 and 10 7 M. The calibration plot is shown in Fig. 6. A linear relationship between peak current and biotin concentration in the range between 0.01 and 1 nM was obtained with a correlation coefficient of 0.998 according to the following equation: IðnAÞ ¼ 758:8C biotin ðnMÞ–939:6 The detection limit of this method was 7.5 10 3 nM of biotin and the reproducibility obtained was 2% (n ¼3). In comparison with other methods for biotin determination using magnetic
Table 3 Measurement of biotin in real samples. Data are given as average7SD (n¼ 3). Type
Reference value
Value obtained
MEDEBIOTIN (tablet) MEDEBIOTIN (vial)
5 mg 4.6 mg/ mL
4.8 7 0.3 mg 4.5 7 0.1 mg/mL
beads, it is probably one of the fastest methods because only 4 min are necessary to obtain one analytical response (13 analytical signals can be obtained per hour). This short analysis time is due to the extremely short enzymatic reaction time. Once the TMB is injected into the FIA system, the enzymatic reaction occurs in a very short time (few seconds) instead of minutes. The present biotin assay presents better limit of detection in comparison with other electrochemical methods for biotin determination [21–23]. Last but not least, the SPCE does not need tedious pretreatment or previous modification and the washing step only performed with the buffer does not damage the electrode surface. Moreover , the first assay using a magnetic flow cell with an integrated magnet is reported here. In the literature few articles are reported on coupling magnetic beads and flow injection analysis system. For example Krejcova et al. [24] have coupled paramagnetic particles with a flow injection analysis, but the magnetic beads are used for the isolation of the complex H5N1–CdS and are not deposited on the detector. On the other hand Pan and Yang [25] used functionalized magnetic beads for the carcinoembryonic antigen. But they used external magnet and glassy carbon electrode. 3.3.2. Application to real sample Two pharmaceuticals with a known value of biotin have been used to validate our methods. These samples have been prepared as explained in Section 2.5. These two pharmaceuticals present a very high concentration of biotin, because their use is to compensate a lack of biotin in the organism; hence important dilutions have to be performed. Taking in account all the dilutions, the accuracy of the result obtained for both samples the accuracy obtained is satisfactory and the standard deviation obtained is also very good. The results are tabulated in Table 3.
4. Conclusion Here, a new biotin assay using magnetic beads and a flow analysis system with the first flow cell with an integrated magnet is reported. The biotin assay described here is easy to use, costeffective, time saving, and robust. It is obtained by a one step competitive assay and after the biological reaction a first analytical
J. Biscay et al. / Talanta 131 (2015) 706–711
signal can be obtained after 1 min thanks to the flow injection analysis system. This technique coupled to a new cell with an integrated magnet allows the modification and the cleaning of the electrode surface in a very easy way and much analysis can be obtained in a very short time. Moreover, it has been reported here that in order to demonstrate the viability of the method, and to validate it, a biotin assay is employed to determine biotin in two pharmaceuticals available in the market. In the future a lot of assays based on the magnetic beads and flow injection analysis can be obtained.
[18]
References
[19]
[10] [11] [12] [13] [14] [15]
[16] [17]
[20] [1] J. Ruzicka, H. Hansen, Anal. Chim. Acta 78 (1975) 145–157. [2] K.K. Stewart, G.R. Beecher, P.E. Harc, Anal Biochem. 70 (1976) 167–173. [3] P. Dugoa, N. Fawzya, F. Cichelloa, F. Cacciolac, P. Donatob, Luigi Mondello, J. Chromatogr. A 1278 (2013) 46–53. [4] M. Maciejewskaa, A. Szczureka, Ł. Ochromowicz, Sens. Actuat. B 141 (2009) 417–423. [5] F.R. Mansour, N.D. Danielson, Trends Anal. Chem. 40 (2012) 1–14. [6] S. Feng, M. Zhang, Y. Huang, D. Yuann, Y. Zhu, Talanta 117 (2013) 456–462. [7] R.O. Hassan, A.T. Faizullah, Arab. J. Chem. 6 (2013) 393–400. [8] E.C. Vidotti, V.C. Almeida, C.C. Oliveira, Talanta 64 (2004) 993–999. [9] M.J. Ruedas Rama, A. Ruiz Medina, A. Molina Díaz, J. Pharm. Biomed. Anal. 35 (2004) 1027–1034.
[21]
[22] [23] [24] [25]
711
Y. Yu, Y. Jiang, R. He, Talanta 88 (2012) 352–357. J. Wang, E. Hansena, M. Miró, Anal. Chim. Acta 499 (2003) 139–147. S.S.M.P. Vidigal, I.V. Tóth, A.O.S.S. Rangel, Talanta 84 (2011) 1298–1303. K. Jitmanee, S. Kradtap Hartwell, J. Jakmunee, S. Jayasvasti, J. Ruzicka, K. Grudpan, Talanta 57 (2002) 187–192. E. Palecek, M. Bartosik, Chem. Rev. 112 (6) (2012) 3428–3461. F. Bettazzi, E. Hamid-Asl, C.L. Esposito, C. Quintavalle, N. Formisano, S. Laschi, S. Catuogno, M. Iaboni, G. Marrazza, M. Mascini, L. Cerchia, V. De Franciscis, G. Condorelli, I. Palchetti, Anal. Bioanal. Chem. 405 (2–3) (2013) 1025–1034. A. Erdem, G. Congur, Talanta 118 (2014) 7–13. R.K. Mendes, S. Laschi, D.R. Stach-Machado, L.T. Kubota, G. Marrazza, Sens. Actut. B 166–167 (2012) 135–140. K. Mervartova, Miroslav Polasek, J.M. Calatayud, J. Pharm. Biomed. Anal. 45 (2007) 367–381. E. Khaled, I.A. El-Sabbagh, N.G. El-Kholy, E.Y.A. Ghahni, Talanta 87 (2011) 40–45. M.F. Bergamini, A.L. Santos, N.R. Stradiotto, M.V.B. Zanoni, J. Pharma. Biomed. Anal. 43 (2007) 315–319. R. Kizek, M. Masarik, K.J. Kramer, D. Potesil, M. Bailey, J.A. Howard, B. Klejdus, R. Mikelova, V. Adam, L. Trnkova, F. Jelen, Anal. Bioanal. Chem. 381 (2005) 1167–1178. S. Tanaka, F. Yamamoto, K. Sugarwara, H. Nakumura, Talanta 44 (1997) 357–363. J. Ho, W. Hsu, W. Liao, J. Chiu, M. Chen, H. Chang, C. Li, Bioelectron 26 (2010) 1021–1027. L. Krejcova1, D. Dospivova1, M. Ryvolova, P. Kopel, D. Hynek, S. Krizkova, J. Hubalek, V. Adam, R. Kizek, Electrophoresis 33 (2012) 3195–3204. J. Pan, Q. Yang, Anal. Bioanal. Chem. 388 (2007) 279.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Electrochemical biotin detection based on magnetic beads and a new magnetic flow cell for screen printed electrode Julien Biscay, María Begoña González García, Agustín Costa García n Department of Physical and Analytical Chemistry, Faculty of Chemistry, University of Oviedo, 33006 Oviedo, Spain
art ic l e i nf o
a b s t r a c t
Article history: Received 25 April 2014 Received in revised form 3 August 2014 Accepted 5 August 2014 Available online 27 August 2014
The use of the first flow-cell for magnetic assays with an integrated magnet is reported here. The flow injection analysis system (FIA) is used for biotin determination. The reaction scheme is based on a one step competitive assay between free biotin and biotin labeled with horseradish peroxidase (B-HRP). The mixture of magnetic beads modified with streptavidin (Strep-MB), biotin and B-HRP is left 15 min under stirring and then a washing step is performed. After that, 100 μL of the mixture is injected and after 30 s 100 μL of 3,30 ,5,50 -Tetramethylbenzidine (TMB) is injected and the FIAgram is recorded applying a potential of 0.2 V. The linear range obtained is from 0.01 to 1 nM of biotin and the sensitivity is 758 nA/nM. The modification and cleaning of the electrode are performed in an easy way due to the internal magnet of the flow cell. & 2014 Elsevier B.V. All rights reserved.
Keywords: Biotin detection Flow injection analysis Flow cell with an integrated magnet Screen printed electrode Magnetic beads
1. Introduction Flow injection analysis (FIA) techniques have been employed to automate a wide variety of chemical and biochemical analyses since their invention in the middle of the seventies [1,2]. FIA techniques are classified into the basic mode (also named normal flow injection analysis) and different special modes as stopped flow [3,4] or reverse flow [5,6]. FIA is one of the most popular continuous flow techniques and its simplicity, and flexibility allow its application in various chemistries. Moreover the flexibility in the different created formats and designs helps its introduction in laboratories using low cost instrumentation but each one has to comply with the three cornerstones of the FIA: (i) injection of discrete and well defined volume of sample solution into a flowing carrier stream; (ii) reproducible and precise timing of the manipulation that the injected sample zone is subjected to in the system, from the point of the injection to the point of detection; and (iii) the creation of a concentration gradient of the injected sample, providing a transient, but reproducible readout of the recorded signal. The second generation of FIA including stopped flow injection [7], bead injection [8–10], sequential injection with lab on valve [11] has opened new perspectives with many analysis systems. The majority of these systems have in common the detection of the product before the equilibrium state is reached, which considerably reduces the analysis time. To overcome
n
Corresponding author. E-mail address: [email protected] (A. Costa García).
http://dx.doi.org/10.1016/j.talanta.2014.08.013 0039-9140/& 2014 Elsevier B.V. All rights reserved.
shortcomings such as contamination of the surface of the detector or the lack of sensitivity, an appealing tool is used called the beadinjection analysis [12,13]. Magnetic beads (MBs) are generally used for it high surface area per volume. The ability to accommodate higher numbers of immobilized molecules helps to improve sensitivity and consequently detection limit of the assay. Another benefit is their easy manipulation with external magnets which makes it possible to perform biological reaction events away from the electrode surface [14] and reduce the complexity for sensing application. Moreover their uses minimize the matrix effect thanks to the washing procedures and faster assay kinetics that are achieved since the beads are in suspension [15]. Many articles have been published using both the advantages of magnetic beads and screen printed electrodes [16,17]; articles about FIA and its application in pharmaceutical [18,19] and biomedical analysis can be found [20] but articles on any assay using a magnetic flow cell with an integrated magnet for screen printed electrodes have not been reported. In this paper an electrochemical biotin assay using a flow injection system which includes for the first time a flow cell for magnetic assay with an integrated magnet is described. The present strategy is based on a competition scheme where biotin and the biotin-HRP compete for the binding sites of streptavidin, which are immobilized on the magnetic beads surface. After molecular recognition, the beads are injected into the FIA system and immobilized on the electrode surface with the help of the integrated magnet of the flow cell which is exactly situated underneath the working electrode of the screen printed electrode. After the injection of the substrate, enzymatic product reduc-
J. Biscay et al. / Talanta 131 (2015) 706–711
tion current is obtained, which is inversely related to biotin concentration. The electrochemical response is measured using the amperometric technique applying a constant potential. Moreover the packing of MBs on the SPCE surface also implied that the enzyme reaction product was generated very close to the electrode surface, thus allowing the steady-state to be reached rapidly (which implies faster measurements), and minimizing diffusion limitations of the electroactive species.
2. Experimental 2.1. Chemicals Sodium hydroxide (1.064.1000) was delivered by MERCK (Spain), magnetic beads of 1 μm diameter modified with streptavidin (Dynabeads s MyOne™ Streptavidin C1) (Strep-MB) (ref. 650.01) were purchased from Invitrogen, Biotinylated Horseradish Peroxidase (HRP-B) (ref. 29139) was supplied by Thermo Scientific and 3,30 ,5,50 tetramethylbenzidine (TMB) (ref. T0440), and Biotin (ref. B4501) were purchased from SIGMA. All chemicals were of analytical reagent grade, and the Milli-Q water used was obtained from a Millipore Direct-Q™ 5 purification system. Stock solution of 5 10 6 M of B-HRP, 5 10 3 M of Biotin and 7 108 magnetic beads per mL was daily prepared in 0.1 M phosphate buffer (PB) pH 7.2 and stored at 4 1C in a refrigerator. 2.2. Apparatus and electrodes Chronoamperometric measurements were performed using an ECO Chemie μAutolab type II potentiostat interfaced with a Pentium 166 computer system and controlled by the Autolab GPES software version 4.8 for Windows 98. All measurements were carried out at room temperature. Screen-Printed Carbon Electrodes (ref DRP-110), an edge connector (ref. DRP-CAC), were purchased from DropSens, S.L (Oviedo, Spain). The screen-printed electrodes consist of a Carbon working (4 mm diameter), carbon auxiliary and silver pseudo reference electrodes printed on an alumina substrate. An insulating layer serves to delimit the electrochemical cell and electric contacts. The magnetic flow cell (ref. CFLWCL-MAGN) was designed and purchased from DropSens, S.L (Oviedo, Spain). It is a methacrylate wall-jet Flow-Cell for FIA, with an unscrewed open–close system which allows an easy electrode replacement. This cell is designed to obtain an inlet flow perpendicular to the electrode0 s surface, and an outlet flow forming a 451 angle and an O-ring which limits the volume of the electrochemical cell. The integrated magnet is a cylindrical magnet which in the upper position is situated exactly above the working electrode and permits the modified magnetic beads to fix on the surface of the electrode. When the magnet is in the lower position, the magnetic beads are eliminated from the surface of the electrode. This new magnetic flow cell is represented in Fig. 1A. The cell is one of the parts of the flow injection analysis system which is schematically represented in Fig. 1B The FIA system used for the detection of biotin is a 12 cylinder Perimax Spetec peristaltic pump (Spetec GmbH, Germany) which allows the 0.1 M PB pH 7.2 stream to flow through the system. Desired solutions are injected by means of a six port rotary valve, Model 1106 (Omnifit Ltd., UK) equipped with a 100 μL loop. 2.3. Assay procedure The reaction scheme is based on a one step competitive assay between free biotin and biotin labeled with horseradish peroxidase (B-HRP). The mixture of magnetic beads modified with
707
streptavidin (Strep-MB), biotin and B-HRP is left for 15 min under stirring and then a washing step is performed. It consists of a resuspension of the beads in the 0.1 M PB solution at pH 7.2 and then a separation with the magnet to remove the supernatant. This operation is repeated 3 times. The scheme of the assay is reported in Fig. 2. 2.4. Electrochemical measurement Amperometric detection was performed applying a constant potential to the modified electrode. The carrier solution (phosphate buffer, pH 7.2) was pumped at a constant flow rate (1 mL/ min) until a stable baseline was recorded. Then, the flow rate was changed to 0.5 mL/min and the magnet was moved to the upper position. Finally, 30 s after the injection of 100 μL of the sample, the flow rate was changed to 1 mL/min and then 100 μL of TMB diluted 5 times (this dilution was necessary to avoid the oxidation peak due to the TMB itself) were injected into the system and the reduction current was observed and recorded 10 s after the injection of the substrate. Finally, the magnet is moved to the lower position and for 3 min at a flow rate of 3 mL/min the carrier is pumped in order to wash the electrode surface. After the washing step another analytical run can be performed. 2.5. Real sample measurement The methodology was tested in two pharmaceuticals with a known concentration of biotin. The first one is MEDEBIOTIN tablets (pharmaceutical 1) and second is MEDEBIOTIN vials (pharmaceutical 2). Each tablet (0.072 g) contains 5 mg of biotin, and each vial of 1 mL contains 4.6 mg of biotin. So 0.01 g of the pharmaceutical 1 was dissolved in 100 mL of the buffer. Then, an aliquot of this solution was diluted 1:50,000 times. Concerning the pharmaceutical 2, the vial was diluted in 1 L of the buffer and from this solution, a second dilution was performed (1:50,000). Finally mixtures of Strep-MB/B-HRP/Pharmaceutical were prepared and left under stirring for 15 min. Then a washing step was performed and finally the amperometric measurement was performed as explained in Section 2.4. Both samples were measured in triplicate.
3. Result and discussion 3.1. Optimization of the parameters of the flow injection analysis system 3.1.1. Flow rate 3.1.1.1. Injection of the mixture. The flow rate obtained when the mixture of reaction is injected in the FIA system is very important because there must be equilibrium between the pulling power of the flow carrier and the power of magnet in the flow cell that retains the magnetic beads. To carry out this study, a mixture of Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M), prepared as explained in Section 2.3 was injected at different flow rates of the carrier (0.5, 1, 1.5, 2 mL/min). Then TMB was injected and the reduction current was recorded applying a potential of 0.20 V and using a flow rate of 1 mL/min. The results obtained are summarized in Table 1. The analytical signal decreases when the flow rate increases because the number of Strep-MB fixed on the electrode surface is lower. At higher flow rates the pulling power of the flow carrier is higher than the capacity of the magnet to retain all magnetic beads on the electrode surface. In other words, the effect of cleaning is higher than the retention of the Strep-MB on the electrode surface and consequently, lower analytical signals are obtained. Thus, for further studies, a flow
708
J. Biscay et al. / Talanta 131 (2015) 706–711
Injection valve
Carrier
Sample Pump
Detector with integrated magnet Waste
Potentiostat
xxxxxx
Fig. 1. (A) Images of the magnetic flow cell, with the magnet up and down. (B) Scheme of the FIA system.
Fig. 2. Scheme of the competitive Biotin assay.
Table 1 Influence of the flow rate, during the injection of the mixture Strep-MBs/HRP-B, on the analytical response; Eapplied ¼ 0.2 V; TMB diluted 5 times; Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M). Each signal is the mean of three measurements. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2. Flow rate (mL/min) 0.5
1
1.5
2
( 55737 64)nA
( 5100 7100)nA
( 2850 7 132)nA
(-13757 106)nA
rate of 0.5 mL/min, for the injection of the mixtures, where higher analytical signals were obtained, has been chosen.
3.1.1.2. Cleaning of the electrode. The influence of the flow rate on the efficiency of the washing step also has been studied. The washing step consists of placing the magnet in the lower position, so that the carrier washes the surface of the electrode sending the magnetic beads to the waste. Without magnetic field, the StrepMBs are not retained on the electrode and the carrier can clean it.
A mixture of Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M) was prepared as explained in Section 2.3, and was injected with a flow rate of 0.5 mL/min. After the injection of the TMB and recording the reduction current at 0.20 V, a washing step is performed pumping a flow of 1, 2 and 3 mL/min (Fig. 3A, B and C, respectively) for 3 min in each case. To check the efficiency of the washing step, TMB was injected again. In the case of 1 mL/min, the washing step was totally inefficient because a similar reduction peak was observed after the second injection of TMB, whereas in the case of 2 mL/min a decrease in the intensity of the reduction peak was noted. Finally if the washing step was performed with a flow rate of 3 mL/min, the cleaning was more efficient because after the second injection of TMB, no reduction peak can be observed. All these results are summarized in Fig. 3. To complete this study, the washing step was performed, applying a flow rate of 3 mL/min, for 1, and 2 and 3 min. The reduction peak after the second injection of TMB totally disappeared with a washing step of 3 min (data not shown). It means that the surface of the electrode is enzyme free. So for further experiments the washing step will be performed for 3 min with a flow rate of 3 mL/min. 3.2. Optimization of the parameters that affect the analytical response 3.2.1. Optimization of the applied potential To determine the best detection potential, 40 μL of TMB was added on a SPCE and a cyclic voltammetry was recorded between 0.40 and þ0.80 V at scan rate of 50 mV/s. In Fig. 4A both
J. Biscay et al. / Talanta 131 (2015) 706–711
709
Intensity (nA)
0
-1500
-3000
-4500
0
0
-1500
-1500
Intensity (nA)
Intensity (nA)
-6000
-3000
-4500
Second injection ofTMB
-3000
-4500
-6000
-6000
Fig. 3. Optimization of the washing flow rate; Flow rate 1 ml/min (A), 2 ml/min (B) and 3 ml/min (C). Flow carrier 0.1 M phosphate buffer pH 7.2; TMB diluted 5 times; Eapplied ¼ 0.2 V; Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 7 M).
not occur. Previously, a mixture of Strep-MB (7 107 per mL)/ B-HRP (5 10 7 M) was prepared as explained in Section 2.3 and injected into the FIA system. The results are summarized in Fig. 4B. The highest analytical response was obtained for 0.20 V. This potential was chosen for further studies.
-0.4
-0.2
0
0.2
0.4
0.6
0.8
Potential (V vs. Ag pseudoreference)
Potential (V vs Ag pseudoreference)
Intensity (nA)
0
0
-0.1
-0.2
-0.3
-2000
IðnAÞ ¼ 505:96C BHRP ðnMÞ 30:8
-4000 -4940 -6000
3.2.2. Concentration of HRP-B To determine the best concentration of B-HRP for the competitive assay, different mixtures of Strep-MB and HRP-B were prepared. The number of Strep-MB per mL was fixed to 7 107 and the concentration of B-HRP was between 5 10 12 and 5 10 8 M. The mixture was prepared as previously explained and the FIAgram was recorded applying -0.20 V as explained in Section 2.4. A linear relationship between current and B-HRP concentration in the range of 0.05 and 5 nM was obtained with a correlation coefficient of 0.9997 (n ¼7) according to the following equation:
-5025 -5575
-5585
Fig. 4. (A) Cyclic voltammogram of TMB in 0.1 M phosphate buffer solution pH 7.2. ν ¼ 50 mV/s. (B) Influence of the applied potential on the analytical response; StrepMB (7 107 per mL)/ B-HRP (5 10 7 M); TMB diluted 5 times. Each signal is the mean of three measurements. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2.
oxidation and reduction peaks of TMB are obtained between þ0.00 and þ0.20 V. So, a hydrodynamic curve was registered whenTMB is injected into the FIA system and potentials between 0.00 and 0.30 V are applied, where the oxidation of TMB does
This result indicates that the saturation of the binding sites of Strep-MB with the enzymatic conjugate B-HRP is reached for a concentration of 5 nM. For further experiments, 5 nM has been chosen as the concentration for B-HRP. By selecting this concentration, it is insured that the maximum slope is achieved. 3.2.3. Optimization of the number of magnetic beads per mL For the purpose to determine the best number of magnetic beads per mL,a mixture of Strep-MB/B-HRP with different number of magnetic beads per mL and a constant concentration of B-HRP were prepared. So in this study a number of magnetic beads comprised between 7 105 and 7 108 beads per mL were used and the concentration of B-HRP was 5 nM. The different mixtures were left under stirring for 15 min, and after washing, the chronoamperogram was recorded. The results are presented in
710
J. Biscay et al. / Talanta 131 (2015) 706–711
C Biotina (nM)
Table 2 Influence of the concentration of MBs on the analytical response; Eapplied ¼ -0.2 V; TMB diluted 5 times; B-HRP (5 10 9 M). Each signal is the mean of three measurements. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2.
0
7 108
7 107
7 106
7 105
( 35507 50) nA
( 1336 715) nA
( 546 715) nA
( 1337 15) nA
Intensity (nA)
Magnetic beads (per mL)
0
Intensity (nA)
-200 -400 -600
0
0.2
0.4
0.6
0.8
1
-400
-800
-1200 Fig. 6. Calibration curve of biotin under the optimized conditions assay. Strep-MB (7 107 per mL)/B-HRP (5 10 9 M); TMB diluted 5 times. Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2. Each point is the mean of three measurements.
-800 -1000 -1200 -1400
Fig. 5. Amperometric detection of TMB in the flow injection analysis system with the magnetic flow-cell; TMB diluted 5 times; Eapplied ¼ 0.2 V; Flow rate 1 ml/min; Flow carrier 0.1 M phosphate buffer pH 7.2 Strep-MB (7 107 magnetic beads per mL)/B-HRP (5 10 9 M).
Table 2 and represent the mean of 3 signals obtained with the same SPCE. As it can be seen in Table 2 that the analytical response increases with the number of magnetic beads. Morever, for lower number of magnetic beads than 7 107, an important decrease in the signal can be noticed. Nevertheless, in order to be economically viable, 7 107 magnetic beads per mL for the following experiments was selected. 3.2.4. Reproducibility To check the reproducibility of the assay, a mixture of 7 107 magnetic beads per mL and 5 nM of B-HRP was prepared. After 15 min under stirring and a washing step, the mixture was injected into the FIA system. Then TMB diluted 5 times was injected and the peak current was recorded applying 0.20 V. After a washing step, the injection of the mixture and the TMB were repeated 5 times. The reduction peaks obtained are presented in Fig. 5. An analytical signal of 1320 719 nA (RSD: 1.5%) was obtained. These results demonstrate the robustness of the electrode and the efficiency of the washing step. 3.3. Biotin assay 3.3.1. Calibration curve The biotin assay was carried as explained in Section 2.3. Different mixtures of magnetic beads, B-HRP, and biotin were prepared. In each solution, the number of magnetic beads and the concentration of B-HRP were 7 107 per mL and 5 nM respectively, and the concentration of biotin was comprised between 5 10 11 and 10 7 M. The calibration plot is shown in Fig. 6. A linear relationship between peak current and biotin concentration in the range between 0.01 and 1 nM was obtained with a correlation coefficient of 0.998 according to the following equation: IðnAÞ ¼ 758:8C biotin ðnMÞ–939:6 The detection limit of this method was 7.5 10 3 nM of biotin and the reproducibility obtained was 2% (n ¼3). In comparison with other methods for biotin determination using magnetic
Table 3 Measurement of biotin in real samples. Data are given as average7SD (n¼ 3). Type
Reference value
Value obtained
MEDEBIOTIN (tablet) MEDEBIOTIN (vial)
5 mg 4.6 mg/ mL
4.8 7 0.3 mg 4.5 7 0.1 mg/mL
beads, it is probably one of the fastest methods because only 4 min are necessary to obtain one analytical response (13 analytical signals can be obtained per hour). This short analysis time is due to the extremely short enzymatic reaction time. Once the TMB is injected into the FIA system, the enzymatic reaction occurs in a very short time (few seconds) instead of minutes. The present biotin assay presents better limit of detection in comparison with other electrochemical methods for biotin determination [21–23]. Last but not least, the SPCE does not need tedious pretreatment or previous modification and the washing step only performed with the buffer does not damage the electrode surface. Moreover , the first assay using a magnetic flow cell with an integrated magnet is reported here. In the literature few articles are reported on coupling magnetic beads and flow injection analysis system. For example Krejcova et al. [24] have coupled paramagnetic particles with a flow injection analysis, but the magnetic beads are used for the isolation of the complex H5N1–CdS and are not deposited on the detector. On the other hand Pan and Yang [25] used functionalized magnetic beads for the carcinoembryonic antigen. But they used external magnet and glassy carbon electrode. 3.3.2. Application to real sample Two pharmaceuticals with a known value of biotin have been used to validate our methods. These samples have been prepared as explained in Section 2.5. These two pharmaceuticals present a very high concentration of biotin, because their use is to compensate a lack of biotin in the organism; hence important dilutions have to be performed. Taking in account all the dilutions, the accuracy of the result obtained for both samples the accuracy obtained is satisfactory and the standard deviation obtained is also very good. The results are tabulated in Table 3.
4. Conclusion Here, a new biotin assay using magnetic beads and a flow analysis system with the first flow cell with an integrated magnet is reported. The biotin assay described here is easy to use, costeffective, time saving, and robust. It is obtained by a one step competitive assay and after the biological reaction a first analytical
J. Biscay et al. / Talanta 131 (2015) 706–711
signal can be obtained after 1 min thanks to the flow injection analysis system. This technique coupled to a new cell with an integrated magnet allows the modification and the cleaning of the electrode surface in a very easy way and much analysis can be obtained in a very short time. Moreover, it has been reported here that in order to demonstrate the viability of the method, and to validate it, a biotin assay is employed to determine biotin in two pharmaceuticals available in the market. In the future a lot of assays based on the magnetic beads and flow injection analysis can be obtained.
[18]
References
[19]
[10] [11] [12] [13] [14] [15]
[16] [17]
[20] [1] J. Ruzicka, H. Hansen, Anal. Chim. Acta 78 (1975) 145–157. [2] K.K. Stewart, G.R. Beecher, P.E. Harc, Anal Biochem. 70 (1976) 167–173. [3] P. Dugoa, N. Fawzya, F. Cichelloa, F. Cacciolac, P. Donatob, Luigi Mondello, J. Chromatogr. A 1278 (2013) 46–53. [4] M. Maciejewskaa, A. Szczureka, Ł. Ochromowicz, Sens. Actuat. B 141 (2009) 417–423. [5] F.R. Mansour, N.D. Danielson, Trends Anal. Chem. 40 (2012) 1–14. [6] S. Feng, M. Zhang, Y. Huang, D. Yuann, Y. Zhu, Talanta 117 (2013) 456–462. [7] R.O. Hassan, A.T. Faizullah, Arab. J. Chem. 6 (2013) 393–400. [8] E.C. Vidotti, V.C. Almeida, C.C. Oliveira, Talanta 64 (2004) 993–999. [9] M.J. Ruedas Rama, A. Ruiz Medina, A. Molina Díaz, J. Pharm. Biomed. Anal. 35 (2004) 1027–1034.
[21]
[22] [23] [24] [25]
711
Y. Yu, Y. Jiang, R. He, Talanta 88 (2012) 352–357. J. Wang, E. Hansena, M. Miró, Anal. Chim. Acta 499 (2003) 139–147. S.S.M.P. Vidigal, I.V. Tóth, A.O.S.S. Rangel, Talanta 84 (2011) 1298–1303. K. Jitmanee, S. Kradtap Hartwell, J. Jakmunee, S. Jayasvasti, J. Ruzicka, K. Grudpan, Talanta 57 (2002) 187–192. E. Palecek, M. Bartosik, Chem. Rev. 112 (6) (2012) 3428–3461. F. Bettazzi, E. Hamid-Asl, C.L. Esposito, C. Quintavalle, N. Formisano, S. Laschi, S. Catuogno, M. Iaboni, G. Marrazza, M. Mascini, L. Cerchia, V. De Franciscis, G. Condorelli, I. Palchetti, Anal. Bioanal. Chem. 405 (2–3) (2013) 1025–1034. A. Erdem, G. Congur, Talanta 118 (2014) 7–13. R.K. Mendes, S. Laschi, D.R. Stach-Machado, L.T. Kubota, G. Marrazza, Sens. Actut. B 166–167 (2012) 135–140. K. Mervartova, Miroslav Polasek, J.M. Calatayud, J. Pharm. Biomed. Anal. 45 (2007) 367–381. E. Khaled, I.A. El-Sabbagh, N.G. El-Kholy, E.Y.A. Ghahni, Talanta 87 (2011) 40–45. M.F. Bergamini, A.L. Santos, N.R. Stradiotto, M.V.B. Zanoni, J. Pharma. Biomed. Anal. 43 (2007) 315–319. R. Kizek, M. Masarik, K.J. Kramer, D. Potesil, M. Bailey, J.A. Howard, B. Klejdus, R. Mikelova, V. Adam, L. Trnkova, F. Jelen, Anal. Bioanal. Chem. 381 (2005) 1167–1178. S. Tanaka, F. Yamamoto, K. Sugarwara, H. Nakumura, Talanta 44 (1997) 357–363. J. Ho, W. Hsu, W. Liao, J. Chiu, M. Chen, H. Chang, C. Li, Bioelectron 26 (2010) 1021–1027. L. Krejcova1, D. Dospivova1, M. Ryvolova, P. Kopel, D. Hynek, S. Krizkova, J. Hubalek, V. Adam, R. Kizek, Electrophoresis 33 (2012) 3195–3204. J. Pan, Q. Yang, Anal. Bioanal. Chem. 388 (2007) 279.
