- Email: [email protected]
4-Aminophenyl boronic acid modified gold platforms for influenza diagnosis
Materials Science and Engineering C 33 (2013) 824–830
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
4-Aminophenyl boronic acid modified gold platforms for influenza diagnosis Sibel Emir Diltemiz ⁎, Arzu Ersöz, Deniz Hür, Rüstem Keçili, Ridvan Say a
Department of Chemistry, Anadolu University, Eskişehir, Turkey
a r t i c l e
i n f o
Article history: Received 17 April 2012 Received in revised form 31 July 2012 Accepted 5 November 2012 Available online 13 November 2012 Keywords: Influenza Quartz crystal microbalance (QCM) Surface plasmon resonance (SPR) 4-Aminophenyl boronic acid (4-APBA)
a b s t r a c t As a potential pandemic threat to human health, there has been an urgent need for rapid, sensitive, simpler and less expensive detection method for the highly pathogenic influenza A virus. For this purpose, Quartz Crystal Microbalance (QCM) and Surface Plasmon Resonance (SPR) sensors have been developed for the recognition of hemagglutinin (HA) which is a major protein of influenza A virus. 4-Aminophenyl boronic acid (4-APBA) has been synthesized and used as a new ligand for binding of sialic acid (SA) via boronic acid–sugar interaction. SA has an important role in binding of HA. QCM and SPR sensor surfaces have been modified with thiol groups and then 4-APBA and SA have been immobilized on sensor surfaces, respectively. Sensor surfaces have been screened with AFM and used for the determination of HA from aqueous solution. The selective recognition of the QCM and SPR sensors toward Concanavalin A has been reported in this work. Also, the binding capacity and detection limits of QCM and SPR sensors have been calculated and detection limits were found to be 4.7 × 10 −2 μM, (0.26 μg ml −1) and 1.28 × 10 −1 μM, (0.72 μg ml − 1) in the 95% confidence interval, respectively. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Influenza, commonly called “the flu,” is a highly contagious viral infection of the respiratory tract. Compared with most other viral respiratory infections, such as the common cold, influenza infection often causes a more severe illness. Influenza virus uses a receptor that binds to human erythrocytes, causing hemagglutination. This is the first step in infection. This attachment step is mediated by the interaction of the viral envelope glycoprotein hemagglutinin (HA) with cell-surface molecules containing sialic acid (SA) [1]. Influenza virus particles are spherical with a diameter of about 100 nm and present two components [2]. HA, which has the function to bind to SA and agglutinate the erythrocytes, and neuraminidase (NA), which catalyzes removal of terminal SA linked to glycoproteins and glycolipids and allows newly created viruses to leave the cell [3]. Sialic acids that are present on cellular surface structures (glycoproteins and glycolipids) represent the targets for binding by HA. HA is the major component of the virus that has a SA binding site located in the distal top of the molecule and it is defined as a pocket of aminoacids that are highly conserved among influenza virus strains [4,5]. According to literature, it has been considered that the carboxylate anion in SA is indispensable to bind with HA on the influenza virus (Fig. 1) [6]. However, it does not bind to neuraminidase.
⁎ Corresponding author at: Anadolu Üniversitesi, Fen Fakültesi, Kimya Bölümü, Yunus Emre Kampüsü, 26470 Eskişehir, Turkey. Tel.: +90 222 3350580/4789; fax: +90 222 3204910. E-mail address: [email protected] (S.E. Diltemiz). 0928-4931/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2012.11.007
Quartz crystal microbalance (QCM) immunosensors have approved widespread applications in the analysis of clinical targets [7–9], the monitoring of environmental contaminants, such as pathogen and bacteria [10–12] and the detection of biomolecular interaction [13–15]. Surface plasmon resonance (SPR) biosensors have made great strides both in terms of technology and applications [16] such as characterizing and quantifying biomolecular interactions [17–19], determination of affinity and binding constants [20–22], monitoring [23,24], diagnosis [25–28] and DNA sensing [29,30]. There is much interest in developing new influenza sensors for quick and reliable testing for influenza virus. One of the strategies is to develop single step direct sensing methods that eliminate separation, incubation or use of any signal-reporting agents. In recent years, non-labeling techniques such as SPR and QCM have attracted a great deal of attention in detection of viral samples [31]. These methods have the advantage of simplifying the analytical method by excluding the labeling procedures [32–35]. Sato et al. used QCM to study the binding of influenza A virus to monosialoganglioside in membranes and explore the influence of membrane composition on receptor functions of gangliosides GM3 reconstituted in sphingomyelin (SM) and glucosylceramide (GlcCer) monolayers were used as the viral receptor [36]. On the other hand, SPR has been used for the detection of influenza virus and the study of interactions that involve viral proteins and receptors. The first use of SPR in influenza virus detection was reported by Schofield and Dimmock [33]. This study was followed by Critchley and Dimmock who studied the binding of influenza A virus to a neomembrane composed of bovine brain lipids that contains sialoglycolipids [37].
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
825
Fig. 1. Illustration of influenza A binding to sialic acids on cell surface.
In this study, novel detection methods based on boronic acid– sugar interaction to use in the medical field for diagnosis influenza virus have improved. For this purpose, surfaces of quartz crystals and SPR chips have modified with thiol groups for providing 4-APBA–SA interaction and SA has immobilized. Then, HA detection was achieved by using prepared QCM and SPR sensors. 2. Materials and methods 2.1. Materials Boric acid, 11-mercaptoundecanoic acid, N-(3-dimethylaminopropyl)N′-ethylcarbodiimide, N-hydroxysuccinimide, thionyl chloride, butyllithium, isopropyl alcohol, sodium carbonate and hemagglutinin were supplied by Aldrich (Milwaukee, WI, USA). All glassware was extensively washed with dilute HNO3 before use. All other chemicals were of analytical grade purity and purchased from Merck AG (Darmstadt, Germany). All water used in the experiments was purified using a Barnstead (Dubuque, IA) ROpure LP® reverse osmosis unit with a high flow cellulose acetate membrane (Barnstead D2731) followed by a Barnstead D3804 NANO pure® organic/colloid removal and ion exchange packed-bed system. 2.2. Synthesis of 4-aminophenyl boronic acid (4-APBA) Firstly, tri-isopropyl borate was synthesized. For this purpose, boric acid was used as a starting compound. As seen in Fig. 2a, thionyl chloride (SOCI2), (3), (105.29 g, 885 mmol) was slowly added to suspension of boric acid, 1, (9.12 g, 147 mmol) in excess of isopropyl alcohol, (2), under nitrogen atmosphere. After the all SOCI2 was added, the reaction mixture was refluxed. Then, distillation apparatus built under nitrogen atmosphere and excess of isopropyl alcohol was distilled at 80–82 °C and distillation carried out at 110–120 °C as well. Finally, tri-isopropyl borate, (4), was obtained (colorless liquid product, 8.20 g, 93% yield), as a product. Then, in order to synthesize
N,N-dibenzyl-4-bromo aniline, (7), sodium carbonate (4.62 g, 43.59 mmol) was added to 4-bromo aniline, (5), (5 g, 29.06 mmol) and benzyl bromide, (6), (10.31 g, 61.03 mmol) solutions in 150 ml of DMF under nitrogen atmosphere. Obtained suspension was stirred for 10 h at 100–110 °C. After the reaction completed, pieces of ice were added to suspension and stirred. The precipitate of N, N-dibenzyl-4-bromo aniline compound, (7), was filtered, washed with water and dried under vacuum (10 g product, 98% yield) (Fig. 2b). For the synthesis of trimeric-4-(N,N-dibenzylamino)phenylboronic acid, (8), 2.5 M of butyllithium (BuLi) (13.27 ml, 33.19 mmol) in hexane was added dropwise to N,N-dibenzyl4-bromo aniline, (7), (7.70 g, 22.12 mmol) in distilled THF (100 ml) at − 78 °C under nitrogen atmosphere. Then, tri-isopropyl borate, (4), (8.30 g, 44.24 mmol) was added dropwise into reaction mixture. The reaction mixture was stirred for 1 h at − 78 °C and 30 min at room temperature. The solvent was evaporated and obtained solid was dissolved in ethyl acetate (150 ml) and extracted by water. Organic phase was dried by MgSO4 and filtered. Obtained white solid was suspended in hexane and filtered again. The precipitate was dried by vacuum and the product is trimeric-4-(N,N-dibenzylamino)phenylboronic acid, (8), (10 g, 76% yield) (Fig. 2c). Catalytic amount of Pd/C (10%) and concentrated HCl were added to trimeric-4-(N, N-dibenzylamino)-phenylboronic acid (5 g, 5.76 mmol), (8), in dry methanol at the last step. Reaction mixture was stirred for 3 h under 14 bar H2(g) atmosphere. At the end of the reaction, the catalyst was filtered, the solvent was evaporated and light brown 4-aminophenyl boronic acid (4-APBA) was obtained (2.27 g, 96% yield) (Fig. 2d). In this synthesis, benzyl groups which are protective for amine group have turned into toluene, trimeric structure has broken up and 4-APBA, (9), has synthesized.
2.3. Characterizations Characterization of synthesized compounds was carried using H-NMR (500 MHz), 13C-NMR (125 MHz) and 11B-NMR (166 MHz).
1
826
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
(a) HO
OH B + OH
+
HO
1
O
SOCl2
B O
O
3
2
4
(b) NH2
N Br
+
+
Na2CO3
DMF 100-110 oC
Br
Br 7
6
5
(c) N
N
-78 oC
O
O B
THF, Nitrogen atmosphere 2h
O B
+ BuLi +
O
Br
B O
O B
N
7
N
4 8
(d) N NH2 O B N
B O
Pd/C (%10) O B
H2(g), 14 bar, MeOH, HCl N
8
HO
B
OH
9
Fig. 2. (a) Tri-isopropyl borate synthesis. (b) N,N-Dibenzyl-4-bromo aniline synthesis. (c) Trimeric-4-(N,N-dibenzylamino)–phenylboronic acid synthesis. (d) 4-Aminophenyl boronic acid (4-APBA) synthesis.
1 H-NMR and 13C-NMR analysis were carried out according to TMS standard in borosilicate NMR tube. 11B-NMR analysis was carried out according to BF3.OEt2 in quartz NMR tube.
3. Experimental 3.1. Quartz crystal microbalance (QCM) Binding events were followed using a Research Quartz Crystal Microbalance (RQCM) with phaselock oscillator, Kynar crystal holder, 100-μl cell volume flow cell, and 1-in., Ti/Au, AT-cut, 5-MHz quartz crystals (all purchased from Maxtek, Inc.). The holder was mounted with crystal face positioned 90° to ground to minimize gravity precipitation onto the surface. A variable-flow minipump (Minipuls 3) peristaltic pump with 0.51 mm PVC tubing (Tygon R 3603) was used with the flow cell at rates in the range of 0.1–0.5 ml min −1. Fresh tubing was cut before each run in order to keep contamination at a minimum level and to limit flow rate deviations. The RQCM phase-lock oscillator provided loading resistance measurements and allowed for the
examination of crystal damping resistance during frequency measurements. All measurements were recorded at room temperature. Sensitivity is known to be 56.6 Hz cm 2 μg −1 for a 5-MHz crystal. 3.2. Surface plasmon resonance (SPR) Surface plasmon resonance curves were obtained by a SPR device provided by SPRiLab (GenOptics, Orsay, France). Gold-coated (thickness 50 nm) SPR chips (25 mm × 12.5 mm) were also supplied from GenOptics. SPRview software was used to obtain plasmon curves. These values were investigated by SPR1001 software and surface plasmon curves were plotted as the angle of incident light versus percent diffraction amount. SPRview software was used as kinetic monitoring program for kinetic analysis studies. 3.3. Surface modification of the QCM electrodes and SPR chips Surfaces of the QCM electrode and SPR chip were modified with 11-mercaptoundecanoic acid. Before modification, electrodes and
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
827
chips were immersed in 20 ml of alkaline piranha solution (1:1:5 deionized water:H2O2:NH3, v/v) for 30 s. Then, they were washed with pure ethyl alcohol and dried in vacuum oven (200 mm Hg, 40 °C) for 3 h. In order to introduce thiol groups onto the QCM electrodes and SPR chips, the electrodes and chips were immersed in a 10 mM 11-mercaptoundecanoic acid solution in ethanol for 24 h. Then, they were thoroughly rinsed three times with ethanol for the removal of unbounded thiol groups onto the sensor surfaces and dried with nitrogen gas under vacuum (200 mm Hg, 40 °C). 3.4. Immobilization of 4-aminophenyl boronic acid and sialic acid onto the surfaces of QCM electrodes and SPR chips After carboxyl fictionalization by using 11-mercaptoundecanoic acid (MUA) solution, QCM electrodes and SPR chips were modified with imide groups via using N-(3-dimethylaminopropyl)-N′ethylcarbodiimide and N-hydroxysuccinimide for immobilization of 4-APBA and SA. The applied experimental procedure can be briefly summarized as follow: QCM electrodes and SPR chips were washed with equilibration buffer (pH: 7.4 HEPES buffer (10 mM HEPES, 150 mM NaCl, 3.4 mM EDTA), 30 ml, 1.5 ml min −1 flow-rate) and then, interacted with 0.4 M N-(3-dimethylaminopropyl)-N′ethylcarbodiimide and 0.1 M N-hydroxysuccinimide (1:1, v/v; 1 ml, 1.5 ml min −1 flow-rate). For the homogenous immobilization of 4-APBA onto the surface of QCM electrodes and SPR chips, sensor surfaces were interacted with 16 mM 4-APBA and 16 mM SA solutions, respectively, The binding efficiencies were determined by monitoring frequency shift and reflection angle for QCM and SPR. Fig. 3a and b show QCM and SPR sensors responses after binding of 4-APBA and SA, respectively. As schematically seen in Fig. 3c, SA was bound to 4-APBA through hydroxyl groups. 3.5. Monitoring of QCM electrode and SPR chip responses for detection of hemagglutinin 4-APBA and SA immobilized QCM electrode and SPR chips were used for real time detection of HA from aqueous solution. The applied experimental procedure can be briefly summarized as follow: the QCM electrodes and SPR chips were washed with equilibration buffer (pH: 7.4 HEPES buffer, 30 ml, 1.5 ml min −1 flow-rate) and then, the sensor responses were monitored until they reached stable baseline. After that, aqueous HA solutions in different concentrations in the range of 0.01 mM–0.16 mM were applied to the QCM electrodes and SPR chips and frequency shifts and reflectivity were monitored, evaluations were repeated as triplicate. Then, desorption was done by applying pH 2.0 glycine/HCl buffer (0.5 ml min −1, 60 min). After desorption step, the QCM electrodes and SPR chips were washed with deionized water and equilibration buffer. For each HA sample application, adsorption–desorption–cleaning step was repeated. The real time Concanavalin A detection studies were also carried out to show specificity of 4-APBA and SA immobilized QCM and SPR nanosensors as given above. Concanavalin A was used because it is lectin and lectins are proteins that react with specific terminal sugar residues. Also, some viruses use lectins to attach themselves to the cells of the host organism during infection. Lectins are commonly known hemagglutinins.
Fig. 3. (a) Frequency change of QCM electrodes for 4-APBA and SA interaction. (b) Reflectivity response of SPR chips for 4-APBA and SA interaction. (c) Binding of SA onto the 4-APBA immobilized sensor surfaces.
trimeric structure. On the other hand, wide peak (− 2 H) at 10.5 ppm shows that there are boronic acid− OH groups in the structure. Peaks at between 7.2 and 7.6 ppm show p-substituted benzene ring. In addition, the signal at 19.68 ppm in 11B NMR spectrum of product shows the existence of boron in the structure. Therefore, it can be said that the product is 4-APBA as expected.
3.6. Characterization of 4-aminophenyl boronic acid (4-APBA)
3.7. Evaluation of 4-aminophenyl boronic acid–sialic acid immobilized sensors response for recognizing of hemagglutinin
Characterization of synthesized compounds was carried by using H-NMR (500 MHz), 13C-NMR (125 MHz) and 11B-NMR (166 MHz). 1 H-NMR spectra of 9 has showed no signals corresponding to these assigned to the benzyl–CH2 group of trimeric compound 8 at 4.7 ppm and 13C-NMR spectra of 9 also no longer showed benzyl − CH2 signal of 8 at 53.86 ppm (spectrum was given in Supplementary data). This is the indication of the cleavage of benzyl groups in the
3.7.1. QCM evaluations The 4-APBA–SA immobilized QCM electrodes were mounted in the holder/flow cell, rinsed with pH 7.4 HEPES buffer (30 ml, 1.5 ml min − 1 flow-rate), and brought to stable resonant frequency. HA was dissolved in HEPES buffer (pH 7.4) to have a concentration from 0.01 to 0.16 mM and pumped through the flow cell at 0.5 ml min − 1 flow rate. The frequency of the sensor was monitored
1
828
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
until it became stable. The frequency shift for each concentration of HA was determined and the evaluation was performed in triplicate. After each assay, HA was removed from the coating by washing with pH 2.0 glycine/HCl buffer (0.5 ml min − 1, 60 min) and then three times with phosphate buffer. The frequency of the sensor approximately recovered to the value of beginning resonant frequency. Fig. 4a shows real-time change in HA detection response of QCM electrode with respect to concentration. AFM image of HA immobilized QCM sensor surface is shown in Fig. 4b. As seen from the figure, it can be said that HA molecule was homogeneously bound to 4-APBA–SA on the sensor surface. Experiments were repeated three times and standard statistical methods were performed to calculate mean values and standard deviation for each data serial. Detection limit of graph was calculated as 4.7 × 10 − 2 μM in the 95% confidence interval.
approximately. After that, desorption was done by applying 0.1 M glycine–HCl solution. After desorption, SPR chip was washed with deionized water and isotonic solution (50 ml and 1.0 ml min − 1 flow-rate). For each HA concentration, adsorption–desorption– cleaning steps were repeated. In order to analyze the kinetic data obtained, SPR1001 software was used. Fig. 5a, shows the reflectivity (%) sensogram with increasing HA concentrations when SA immobilized SPR sensor was used. Binding interactions between SA immobilized sensor surface and HA were determined by using Langmuir adsorption isotherm (Fig. 5b). Eq. (1) was used for this purpose. 1 Requi
¼
1 R max K A
1
. C
þ
1 : R max
ð1Þ
In this equation; 3.7.2. SPR evaluations Sialic acid immobilized SPR chip was used for kinetic analysis. The experimental procedure can be briefly summarized as follow: SPR chip was washed with pH 7.4 phosphate buffer (50 ml, 1 ml min − 1 flow-rate) and then, with deionized water (50 ml, 1 ml min − 1 flow-rate). During the water circulation, surface plasmon curves were taken and the resonance angle was determined. The mirror system was arranged to the resonance angle and kinetic studies were applied at this angle value. After taking plasmon curves and arranging the mirror at the resonance angle, water circulation through the SPR system was preceded for 5 more min. Then, various HA concentrations in the range of 0.01–0.16 mM were applied to SPR system (25 ml and 1.0 ml min − 1 flow rate). Reflectivity (%) changes were monitored instantly and reached to plateau value in 30 min,
a
Requi C Rmax KA
SPR signal after HA injection. HA concentration. Equilibrium signal for infinite HA concentration. Affinity constant.
By using Langmuir graph given in Fig. 4b, affinity constant and the detection limit of SA immobilized sensor for HA binding was found 1.79 × 10 6 M −1 and 1.28 × 10 −1 μM within the 95% confidence interval, respectively. Fig. 5c shows AFM image of SPR sensor surface after immobilization of 0.16 mM HA. As seen from the figure, it can be said HA molecules were homogeneously bound to SA molecules on the sensor surface.
Concentration of Hemagglutinin (mM) 0
Frequency Shift (ΔHz)
-100
0
0,05
0,1
0,15
0,2
-200 -300 -400 -500 -600 -700
y = -4000,5x - 10,299 R2 = 0,9959
b
Fig. 4. (a) Concentration–frequency relationship of 4-APBA–SA immobilized QCM electrode. (b) Morphology of QCM sensor surface after adsorption of 0.16 mM HA.
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
829
Fig. 6. (a) Comparison of selectivity of QCM nanosensors toward to Concanavalin A. (b) Comparison of selectivity of SPR nanosensors toward to Concanavalin A.
Fig. 5. (a) Change of concentration between reflectivity in SA immobilized SPR sensor. (b) Langmuir graph of 4-APBA–SA immobilized SPR sensor. (c) Surface morphology of SPR sensor after immobilization of 0.16 mM HA.
3.8. Selectivity and specificity of 4-APBA and SA immobilized QCM and SPR nanosensors In order to indicate the selectivity and specificity of 4-APBA and SA immobilized QCM and SPR nanosensors, experiments were performed with Concanavalin A. Thus, Concanavalin A solutions (0.01 mM, pH: 7.4 phosphate buffer) were applied to QCM and SPR nanosensors. As seen from Fig. 6a and b the QCM and SPR nanosensors did not give any responses to Concanavalin A solution (Δf = 3.331 and ΔR = 0.261) respectively. Also, the specific responses of the nanosensors to HA molecules, Δf = 39.951 and ΔR = 2.027 are excessively higher than to Concanavalin A. The selectivity ratio was calculated by dividing the QCM and SPR responses data of the Concanavalin A molecules and found 11.99 and 7.77 respectively. 4. Conclusions In the present study, affinity based-novel QCM and SPR sensors which recognize HA were improved by using 4-APBA–SA molecule
that has strong interactions with HA. For this purpose, 4-APBA was prepared. The preparation of 4-APBA is unique; by means of starting from cheap and easy to handle boronic acid and this is the first preparation and application of 4-APBA in literature. QCM electrodes and SPR chips were modified with thiol groups for the immobilization of 4-APBA. Then, HA detection was achieved by using prepared QCM and SPR sensors. Detection limits of QCM and SPR sensors for HA were found to be 4.7 × 10 −2 μM, (0.26 μg ml −1) and 1.28 × 10 −1 μM, (0.72 μg ml −1) in the 95% confidence interval, respectively. Also, the selectivity ratios were calculated for QCM and SPR nanosensors and found as 11.99 and 7.77 respectively. In literature, SPR and QCM based sensors frequently have been reported to provide information on molecular interactions among HA, natural receptors and influenza virus [31]. For example, Sato et al. used QCM to study the binding of influenza A virus to monosialoganglioside in membranes [36], Schofield and Dimmock used monoclonal antibody HC10 immobilized carboxylated dextran polymer matrix coated SPR chip for capture of influenza virus [33]. They reported the dissociation rate constant and association rate constant were found similar to those obtained with an affinity ELISA and Takemoto et al. developed a binding assay to study the interaction between influenza HA and its cell surface receptor SA [38]. But any study has not been reported about detection limit of HA via interaction with SA and HA molecules. We can say that, SA and 4-APBA based synthetic receptor method have compatible detection limit
830
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
than the other HA detection methods which have been mentioned in literature. Also, most of the featured studies which have been used for detection of influenza have shortcoming of short shelf life, denaturizing problems, specific equipment's and expensive antibody–antigen interactions. Therefore, researching of new methods is strongly needed to overcome these disadvantages. In this study, novel biosensor systems which have long shelf life, easy storage conditions, simple and low cost properties to detect influenza were obtained. This new system can also be a very good alternative to biologic receptors. Our results not only demonstrate that HA detection is achieved using QCM and SPR nanosensors, but they can also be used potentially to detection of influenza with special conditions (training personel and cabinet). Acknowledgments Authors thank to Turkey National Boron Research Institute (BOREN) (project number: 2008.Ç0155) for supporting this study. Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.msec.2012.11.007. References [1] W. Weiss, J.H. Brown, S. Cusack, J.C. Paulson, J.J. Skehel, D.C. Wiley, Nature 333 (1988) 426–431. [2] J.J. Skehel, D.C. Wiley, Annu. Rev. Biochem. 69 (2000) 531–569. [3] R.G. Webster, W.J. Bean, O.T. Gorman, T.M. Chambers, Y. Kawaoka, Microbiol. Rev. 56 (1992) 152–179. [4] B. Sukla, T. Milan, W.C. Richard, Virus Res. 2 (1985) 61–68. [5] C. Rohm, N. Zhou, J. Suss, J. Mackenzie, R.G. Webster, Virology 217 (1996) 508–516. [6] Y. Suzuki. 33 (1994) 429–457. [7] L.L. Pang, J.S. Li, J.H. Jiang, Y. Le, G.L. Shen, R.Q. Yu, Sensors Actuators B 127 (2007) 311–316. [8] M. Plomer, G.G. Guilbault, B. Hock Enzyme, Microb. Technol. 14 (1992) 230–235. [9] E.P. Sochazewski, J.H.T. Luong, G.G. Guilbault, Enzyme Microb. Technol. 12 (1990) 173–177. [10] M. Lazerges, H. Perrot, N. Zeghib, E. Antoine, C. Compere, Sensors Actuators B 120 (2006) 329–337. [11] J.L.N. Harteveld, M.S. Nieuwenhuizen, E.R.J. Wils, Biosens. Bioelectron. 12 (1997) 661–667. [12] Y.S. Fung, Y.Y. Wong, Anal. Chem. 73 (2001) 5302–5309. [13] R. Say, A. Gultekin, A.A. Ozcan, Anal. Chim. Acta 640 (2009) 82–86. [14] A. Janshoff, H.J. Galla, C. Steinem, Angew. Chem. Int. Ed. Engl. 39 (2000) 4004–4032. [15] A. Dolatshahi-Pirouz, N. Kolman, A. Arpanaei, T. Jensen, M. Foss, J. Chevallier, P. Kingshott, J. Baas, K. Søballe, F. Besenbacher, Mater. Sci. Eng. C 31 (2011) 514–522.
[16] J. Homola, Chem. Rev. 108 (2008) 462–493. [17] U. Aldinger, P. Pfeifer, G. Schwotzer, P. Steinrücke, Sensors Actuators B 51 (1998) 298–304. [18] A. Tlili, M. Ali Jarboui, A. Abdelghani, D.M. Fathallah, M.A. Maaref, Mater. Sci. Eng. C 25 (2005) 490–495. [19] S.E. Diltemiz, R. Say, S. Büyüktiryaki, D. Hür, A. Denizli, A. Ersöz, Talanta 75 (2008) 890–896. [20] E.A. Turkoglu, K. Caktu, L. Uzun, H. Yavuz, Curr. Opin. Biotechnol. 22 (2011) 63–64. [21] L. Zhang, J. Liu, E. Wang, Biochem. Biophys. Res. Commun. 373 (2008) 202–205. [22] L. Quan, D. Wei, X. Jiang, Y. Liu, Z. Li, N. Li, K. Li, F. Liu, L. Lai, Anal. Biochem. 378 (2008) 144–150. [23] C.F. Mandenius, R. Wang, A. Alden, G. Bergstrom, S. Thebault, C. Lutsch, S. Ohlson, Anal. Chim. Acta 623 (2008) 66–75. [24] M. Del Toro, R. Gargallo, R. Eritja, J. Jaumot, Anal. Biochem. 379 (2008) 8–15. [25] L. Uzun, R. Say, S. Unal, A. Denizli, Biosens. Bioelectron. 24 (2009) 2878–2884. [26] T. Nagel, N. Gajovic-Eichelmann, S. Tobisch, U. Schulte-Spechtel, F.F. Bier, Clin. Chim. Acta 394 (2008) 110–113. [27] K.Y. Park, Y.S. Sohn, C.K. Kim, H.S. Kim, Y.S. Bae, S.Y. Choi, Biosens. Bioelectron. 23 (2008) 1904–1907. [28] S.E. Diltemiz, A. Denizli, A. Ersoz, R. Say, Sensors Actuators B 133 (2008) 484–488. [29] S. Peeters, T. Stakenborg, G. Reekmans, W. Laureyn, L. Lagae, A. Van Aerschot, M. Van Ranst, Biosens. Bioelectron. 24 (2008) 72–77. [30] A. Ersoz, S.E. Diltemiz, A.A. Ozcan, A. Denizli, R. Say, Biosens. Bioelectron. 24 (2008) 742–747. [31] Y. Amano, Q. Cheng, Anal. Bioanal. Chem. 381 (2005) 156–164. [32] X. Su, J. Zhang, Sensors Actuators B 100 (2004) 309–314. [33] D.J. Schofield, N.J. Dimmock, J. Virol. Methods 62 (1996) 33–42. [34] A.A. Kortt, E. Nice, Anal. Biochem. 273 (1999) 133–141. [35] E. Nice, T.L. McInerney, D.C. Jackson, Mol. Immunol. 33 (1996) 659–670. [36] T. Sato, M. Ishii, F. Ohtake, K. Nagata, T. Terabayashi, Y. Kawanishi, Y. Okahata, Glycoconj. J. 16 (1999) 223–227. [37] P. Critchley, N.J. Dimmock, Bioorg. Med. Chem. Lett. 12 (2004) 2773–2780. [38] D.K. Takemoto, J.J. Skehel, D.C. Wiley, Virology 217 (1996) 452–458. Sibel Emir Diltemiz was born in Turkey in 1977. She received the BS degree and PhD from the Department of Chemistry at Anadolu University, Eskisehir, Turkey, in 2002 and 2006, respectively. She is currently an assistant professor in Anadolu University. Her main research area is focused on SPR and QCM sensors and their applications. Arzu Ersöz was born in Turkey in 1961. She is the professor of the Department of Chemistry, Anadolu University, Turkey. She received her PhD from Hacettepe University in 1998. Her current research area is based on molecular imprinted polymers, synthetic receptors and quartz crystal microbalance. Deniz Hür was born in Turkey in 1975. He is an associate professor of the Department of Chemistry, Anadolu University, Turkey. He received his PhD from Anadolu University in 2004. His current researches are based on organic heterocyclic synthesis, microwave assisted organic synthesis and boron heterocyclics. Rüstem Keçili was born in Turkey in 1981. He is the PhD student of the Department of Chemistry, Anadolu University, Turkey. His current research includes the study of imprinted polymers and synthetic receptors. Ridvan Say was born in Turkey in 1967. He is the professor of the Department of Chemistry, Anadolu University, Turkey. He received his PhD from Hacettepe University in 1998. His current research includes the study of removal of heavy metal ions, polymeric sorbents, synthetic receptors, molecular imprinted polymers and biosensors.
Contents lists available at SciVerse ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
4-Aminophenyl boronic acid modified gold platforms for influenza diagnosis Sibel Emir Diltemiz ⁎, Arzu Ersöz, Deniz Hür, Rüstem Keçili, Ridvan Say a
Department of Chemistry, Anadolu University, Eskişehir, Turkey
a r t i c l e
i n f o
Article history: Received 17 April 2012 Received in revised form 31 July 2012 Accepted 5 November 2012 Available online 13 November 2012 Keywords: Influenza Quartz crystal microbalance (QCM) Surface plasmon resonance (SPR) 4-Aminophenyl boronic acid (4-APBA)
a b s t r a c t As a potential pandemic threat to human health, there has been an urgent need for rapid, sensitive, simpler and less expensive detection method for the highly pathogenic influenza A virus. For this purpose, Quartz Crystal Microbalance (QCM) and Surface Plasmon Resonance (SPR) sensors have been developed for the recognition of hemagglutinin (HA) which is a major protein of influenza A virus. 4-Aminophenyl boronic acid (4-APBA) has been synthesized and used as a new ligand for binding of sialic acid (SA) via boronic acid–sugar interaction. SA has an important role in binding of HA. QCM and SPR sensor surfaces have been modified with thiol groups and then 4-APBA and SA have been immobilized on sensor surfaces, respectively. Sensor surfaces have been screened with AFM and used for the determination of HA from aqueous solution. The selective recognition of the QCM and SPR sensors toward Concanavalin A has been reported in this work. Also, the binding capacity and detection limits of QCM and SPR sensors have been calculated and detection limits were found to be 4.7 × 10 −2 μM, (0.26 μg ml −1) and 1.28 × 10 −1 μM, (0.72 μg ml − 1) in the 95% confidence interval, respectively. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Influenza, commonly called “the flu,” is a highly contagious viral infection of the respiratory tract. Compared with most other viral respiratory infections, such as the common cold, influenza infection often causes a more severe illness. Influenza virus uses a receptor that binds to human erythrocytes, causing hemagglutination. This is the first step in infection. This attachment step is mediated by the interaction of the viral envelope glycoprotein hemagglutinin (HA) with cell-surface molecules containing sialic acid (SA) [1]. Influenza virus particles are spherical with a diameter of about 100 nm and present two components [2]. HA, which has the function to bind to SA and agglutinate the erythrocytes, and neuraminidase (NA), which catalyzes removal of terminal SA linked to glycoproteins and glycolipids and allows newly created viruses to leave the cell [3]. Sialic acids that are present on cellular surface structures (glycoproteins and glycolipids) represent the targets for binding by HA. HA is the major component of the virus that has a SA binding site located in the distal top of the molecule and it is defined as a pocket of aminoacids that are highly conserved among influenza virus strains [4,5]. According to literature, it has been considered that the carboxylate anion in SA is indispensable to bind with HA on the influenza virus (Fig. 1) [6]. However, it does not bind to neuraminidase.
⁎ Corresponding author at: Anadolu Üniversitesi, Fen Fakültesi, Kimya Bölümü, Yunus Emre Kampüsü, 26470 Eskişehir, Turkey. Tel.: +90 222 3350580/4789; fax: +90 222 3204910. E-mail address: [email protected] (S.E. Diltemiz). 0928-4931/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2012.11.007
Quartz crystal microbalance (QCM) immunosensors have approved widespread applications in the analysis of clinical targets [7–9], the monitoring of environmental contaminants, such as pathogen and bacteria [10–12] and the detection of biomolecular interaction [13–15]. Surface plasmon resonance (SPR) biosensors have made great strides both in terms of technology and applications [16] such as characterizing and quantifying biomolecular interactions [17–19], determination of affinity and binding constants [20–22], monitoring [23,24], diagnosis [25–28] and DNA sensing [29,30]. There is much interest in developing new influenza sensors for quick and reliable testing for influenza virus. One of the strategies is to develop single step direct sensing methods that eliminate separation, incubation or use of any signal-reporting agents. In recent years, non-labeling techniques such as SPR and QCM have attracted a great deal of attention in detection of viral samples [31]. These methods have the advantage of simplifying the analytical method by excluding the labeling procedures [32–35]. Sato et al. used QCM to study the binding of influenza A virus to monosialoganglioside in membranes and explore the influence of membrane composition on receptor functions of gangliosides GM3 reconstituted in sphingomyelin (SM) and glucosylceramide (GlcCer) monolayers were used as the viral receptor [36]. On the other hand, SPR has been used for the detection of influenza virus and the study of interactions that involve viral proteins and receptors. The first use of SPR in influenza virus detection was reported by Schofield and Dimmock [33]. This study was followed by Critchley and Dimmock who studied the binding of influenza A virus to a neomembrane composed of bovine brain lipids that contains sialoglycolipids [37].
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
825
Fig. 1. Illustration of influenza A binding to sialic acids on cell surface.
In this study, novel detection methods based on boronic acid– sugar interaction to use in the medical field for diagnosis influenza virus have improved. For this purpose, surfaces of quartz crystals and SPR chips have modified with thiol groups for providing 4-APBA–SA interaction and SA has immobilized. Then, HA detection was achieved by using prepared QCM and SPR sensors. 2. Materials and methods 2.1. Materials Boric acid, 11-mercaptoundecanoic acid, N-(3-dimethylaminopropyl)N′-ethylcarbodiimide, N-hydroxysuccinimide, thionyl chloride, butyllithium, isopropyl alcohol, sodium carbonate and hemagglutinin were supplied by Aldrich (Milwaukee, WI, USA). All glassware was extensively washed with dilute HNO3 before use. All other chemicals were of analytical grade purity and purchased from Merck AG (Darmstadt, Germany). All water used in the experiments was purified using a Barnstead (Dubuque, IA) ROpure LP® reverse osmosis unit with a high flow cellulose acetate membrane (Barnstead D2731) followed by a Barnstead D3804 NANO pure® organic/colloid removal and ion exchange packed-bed system. 2.2. Synthesis of 4-aminophenyl boronic acid (4-APBA) Firstly, tri-isopropyl borate was synthesized. For this purpose, boric acid was used as a starting compound. As seen in Fig. 2a, thionyl chloride (SOCI2), (3), (105.29 g, 885 mmol) was slowly added to suspension of boric acid, 1, (9.12 g, 147 mmol) in excess of isopropyl alcohol, (2), under nitrogen atmosphere. After the all SOCI2 was added, the reaction mixture was refluxed. Then, distillation apparatus built under nitrogen atmosphere and excess of isopropyl alcohol was distilled at 80–82 °C and distillation carried out at 110–120 °C as well. Finally, tri-isopropyl borate, (4), was obtained (colorless liquid product, 8.20 g, 93% yield), as a product. Then, in order to synthesize
N,N-dibenzyl-4-bromo aniline, (7), sodium carbonate (4.62 g, 43.59 mmol) was added to 4-bromo aniline, (5), (5 g, 29.06 mmol) and benzyl bromide, (6), (10.31 g, 61.03 mmol) solutions in 150 ml of DMF under nitrogen atmosphere. Obtained suspension was stirred for 10 h at 100–110 °C. After the reaction completed, pieces of ice were added to suspension and stirred. The precipitate of N, N-dibenzyl-4-bromo aniline compound, (7), was filtered, washed with water and dried under vacuum (10 g product, 98% yield) (Fig. 2b). For the synthesis of trimeric-4-(N,N-dibenzylamino)phenylboronic acid, (8), 2.5 M of butyllithium (BuLi) (13.27 ml, 33.19 mmol) in hexane was added dropwise to N,N-dibenzyl4-bromo aniline, (7), (7.70 g, 22.12 mmol) in distilled THF (100 ml) at − 78 °C under nitrogen atmosphere. Then, tri-isopropyl borate, (4), (8.30 g, 44.24 mmol) was added dropwise into reaction mixture. The reaction mixture was stirred for 1 h at − 78 °C and 30 min at room temperature. The solvent was evaporated and obtained solid was dissolved in ethyl acetate (150 ml) and extracted by water. Organic phase was dried by MgSO4 and filtered. Obtained white solid was suspended in hexane and filtered again. The precipitate was dried by vacuum and the product is trimeric-4-(N,N-dibenzylamino)phenylboronic acid, (8), (10 g, 76% yield) (Fig. 2c). Catalytic amount of Pd/C (10%) and concentrated HCl were added to trimeric-4-(N, N-dibenzylamino)-phenylboronic acid (5 g, 5.76 mmol), (8), in dry methanol at the last step. Reaction mixture was stirred for 3 h under 14 bar H2(g) atmosphere. At the end of the reaction, the catalyst was filtered, the solvent was evaporated and light brown 4-aminophenyl boronic acid (4-APBA) was obtained (2.27 g, 96% yield) (Fig. 2d). In this synthesis, benzyl groups which are protective for amine group have turned into toluene, trimeric structure has broken up and 4-APBA, (9), has synthesized.
2.3. Characterizations Characterization of synthesized compounds was carried using H-NMR (500 MHz), 13C-NMR (125 MHz) and 11B-NMR (166 MHz).
1
826
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
(a) HO
OH B + OH
+
HO
1
O
SOCl2
B O
O
3
2
4
(b) NH2
N Br
+
+
Na2CO3
DMF 100-110 oC
Br
Br 7
6
5
(c) N
N
-78 oC
O
O B
THF, Nitrogen atmosphere 2h
O B
+ BuLi +
O
Br
B O
O B
N
7
N
4 8
(d) N NH2 O B N
B O
Pd/C (%10) O B
H2(g), 14 bar, MeOH, HCl N
8
HO
B
OH
9
Fig. 2. (a) Tri-isopropyl borate synthesis. (b) N,N-Dibenzyl-4-bromo aniline synthesis. (c) Trimeric-4-(N,N-dibenzylamino)–phenylboronic acid synthesis. (d) 4-Aminophenyl boronic acid (4-APBA) synthesis.
1 H-NMR and 13C-NMR analysis were carried out according to TMS standard in borosilicate NMR tube. 11B-NMR analysis was carried out according to BF3.OEt2 in quartz NMR tube.
3. Experimental 3.1. Quartz crystal microbalance (QCM) Binding events were followed using a Research Quartz Crystal Microbalance (RQCM) with phaselock oscillator, Kynar crystal holder, 100-μl cell volume flow cell, and 1-in., Ti/Au, AT-cut, 5-MHz quartz crystals (all purchased from Maxtek, Inc.). The holder was mounted with crystal face positioned 90° to ground to minimize gravity precipitation onto the surface. A variable-flow minipump (Minipuls 3) peristaltic pump with 0.51 mm PVC tubing (Tygon R 3603) was used with the flow cell at rates in the range of 0.1–0.5 ml min −1. Fresh tubing was cut before each run in order to keep contamination at a minimum level and to limit flow rate deviations. The RQCM phase-lock oscillator provided loading resistance measurements and allowed for the
examination of crystal damping resistance during frequency measurements. All measurements were recorded at room temperature. Sensitivity is known to be 56.6 Hz cm 2 μg −1 for a 5-MHz crystal. 3.2. Surface plasmon resonance (SPR) Surface plasmon resonance curves were obtained by a SPR device provided by SPRiLab (GenOptics, Orsay, France). Gold-coated (thickness 50 nm) SPR chips (25 mm × 12.5 mm) were also supplied from GenOptics. SPRview software was used to obtain plasmon curves. These values were investigated by SPR1001 software and surface plasmon curves were plotted as the angle of incident light versus percent diffraction amount. SPRview software was used as kinetic monitoring program for kinetic analysis studies. 3.3. Surface modification of the QCM electrodes and SPR chips Surfaces of the QCM electrode and SPR chip were modified with 11-mercaptoundecanoic acid. Before modification, electrodes and
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
827
chips were immersed in 20 ml of alkaline piranha solution (1:1:5 deionized water:H2O2:NH3, v/v) for 30 s. Then, they were washed with pure ethyl alcohol and dried in vacuum oven (200 mm Hg, 40 °C) for 3 h. In order to introduce thiol groups onto the QCM electrodes and SPR chips, the electrodes and chips were immersed in a 10 mM 11-mercaptoundecanoic acid solution in ethanol for 24 h. Then, they were thoroughly rinsed three times with ethanol for the removal of unbounded thiol groups onto the sensor surfaces and dried with nitrogen gas under vacuum (200 mm Hg, 40 °C). 3.4. Immobilization of 4-aminophenyl boronic acid and sialic acid onto the surfaces of QCM electrodes and SPR chips After carboxyl fictionalization by using 11-mercaptoundecanoic acid (MUA) solution, QCM electrodes and SPR chips were modified with imide groups via using N-(3-dimethylaminopropyl)-N′ethylcarbodiimide and N-hydroxysuccinimide for immobilization of 4-APBA and SA. The applied experimental procedure can be briefly summarized as follow: QCM electrodes and SPR chips were washed with equilibration buffer (pH: 7.4 HEPES buffer (10 mM HEPES, 150 mM NaCl, 3.4 mM EDTA), 30 ml, 1.5 ml min −1 flow-rate) and then, interacted with 0.4 M N-(3-dimethylaminopropyl)-N′ethylcarbodiimide and 0.1 M N-hydroxysuccinimide (1:1, v/v; 1 ml, 1.5 ml min −1 flow-rate). For the homogenous immobilization of 4-APBA onto the surface of QCM electrodes and SPR chips, sensor surfaces were interacted with 16 mM 4-APBA and 16 mM SA solutions, respectively, The binding efficiencies were determined by monitoring frequency shift and reflection angle for QCM and SPR. Fig. 3a and b show QCM and SPR sensors responses after binding of 4-APBA and SA, respectively. As schematically seen in Fig. 3c, SA was bound to 4-APBA through hydroxyl groups. 3.5. Monitoring of QCM electrode and SPR chip responses for detection of hemagglutinin 4-APBA and SA immobilized QCM electrode and SPR chips were used for real time detection of HA from aqueous solution. The applied experimental procedure can be briefly summarized as follow: the QCM electrodes and SPR chips were washed with equilibration buffer (pH: 7.4 HEPES buffer, 30 ml, 1.5 ml min −1 flow-rate) and then, the sensor responses were monitored until they reached stable baseline. After that, aqueous HA solutions in different concentrations in the range of 0.01 mM–0.16 mM were applied to the QCM electrodes and SPR chips and frequency shifts and reflectivity were monitored, evaluations were repeated as triplicate. Then, desorption was done by applying pH 2.0 glycine/HCl buffer (0.5 ml min −1, 60 min). After desorption step, the QCM electrodes and SPR chips were washed with deionized water and equilibration buffer. For each HA sample application, adsorption–desorption–cleaning step was repeated. The real time Concanavalin A detection studies were also carried out to show specificity of 4-APBA and SA immobilized QCM and SPR nanosensors as given above. Concanavalin A was used because it is lectin and lectins are proteins that react with specific terminal sugar residues. Also, some viruses use lectins to attach themselves to the cells of the host organism during infection. Lectins are commonly known hemagglutinins.
Fig. 3. (a) Frequency change of QCM electrodes for 4-APBA and SA interaction. (b) Reflectivity response of SPR chips for 4-APBA and SA interaction. (c) Binding of SA onto the 4-APBA immobilized sensor surfaces.
trimeric structure. On the other hand, wide peak (− 2 H) at 10.5 ppm shows that there are boronic acid− OH groups in the structure. Peaks at between 7.2 and 7.6 ppm show p-substituted benzene ring. In addition, the signal at 19.68 ppm in 11B NMR spectrum of product shows the existence of boron in the structure. Therefore, it can be said that the product is 4-APBA as expected.
3.6. Characterization of 4-aminophenyl boronic acid (4-APBA)
3.7. Evaluation of 4-aminophenyl boronic acid–sialic acid immobilized sensors response for recognizing of hemagglutinin
Characterization of synthesized compounds was carried by using H-NMR (500 MHz), 13C-NMR (125 MHz) and 11B-NMR (166 MHz). 1 H-NMR spectra of 9 has showed no signals corresponding to these assigned to the benzyl–CH2 group of trimeric compound 8 at 4.7 ppm and 13C-NMR spectra of 9 also no longer showed benzyl − CH2 signal of 8 at 53.86 ppm (spectrum was given in Supplementary data). This is the indication of the cleavage of benzyl groups in the
3.7.1. QCM evaluations The 4-APBA–SA immobilized QCM electrodes were mounted in the holder/flow cell, rinsed with pH 7.4 HEPES buffer (30 ml, 1.5 ml min − 1 flow-rate), and brought to stable resonant frequency. HA was dissolved in HEPES buffer (pH 7.4) to have a concentration from 0.01 to 0.16 mM and pumped through the flow cell at 0.5 ml min − 1 flow rate. The frequency of the sensor was monitored
1
828
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
until it became stable. The frequency shift for each concentration of HA was determined and the evaluation was performed in triplicate. After each assay, HA was removed from the coating by washing with pH 2.0 glycine/HCl buffer (0.5 ml min − 1, 60 min) and then three times with phosphate buffer. The frequency of the sensor approximately recovered to the value of beginning resonant frequency. Fig. 4a shows real-time change in HA detection response of QCM electrode with respect to concentration. AFM image of HA immobilized QCM sensor surface is shown in Fig. 4b. As seen from the figure, it can be said that HA molecule was homogeneously bound to 4-APBA–SA on the sensor surface. Experiments were repeated three times and standard statistical methods were performed to calculate mean values and standard deviation for each data serial. Detection limit of graph was calculated as 4.7 × 10 − 2 μM in the 95% confidence interval.
approximately. After that, desorption was done by applying 0.1 M glycine–HCl solution. After desorption, SPR chip was washed with deionized water and isotonic solution (50 ml and 1.0 ml min − 1 flow-rate). For each HA concentration, adsorption–desorption– cleaning steps were repeated. In order to analyze the kinetic data obtained, SPR1001 software was used. Fig. 5a, shows the reflectivity (%) sensogram with increasing HA concentrations when SA immobilized SPR sensor was used. Binding interactions between SA immobilized sensor surface and HA were determined by using Langmuir adsorption isotherm (Fig. 5b). Eq. (1) was used for this purpose. 1 Requi
¼
1 R max K A
1
. C
þ
1 : R max
ð1Þ
In this equation; 3.7.2. SPR evaluations Sialic acid immobilized SPR chip was used for kinetic analysis. The experimental procedure can be briefly summarized as follow: SPR chip was washed with pH 7.4 phosphate buffer (50 ml, 1 ml min − 1 flow-rate) and then, with deionized water (50 ml, 1 ml min − 1 flow-rate). During the water circulation, surface plasmon curves were taken and the resonance angle was determined. The mirror system was arranged to the resonance angle and kinetic studies were applied at this angle value. After taking plasmon curves and arranging the mirror at the resonance angle, water circulation through the SPR system was preceded for 5 more min. Then, various HA concentrations in the range of 0.01–0.16 mM were applied to SPR system (25 ml and 1.0 ml min − 1 flow rate). Reflectivity (%) changes were monitored instantly and reached to plateau value in 30 min,
a
Requi C Rmax KA
SPR signal after HA injection. HA concentration. Equilibrium signal for infinite HA concentration. Affinity constant.
By using Langmuir graph given in Fig. 4b, affinity constant and the detection limit of SA immobilized sensor for HA binding was found 1.79 × 10 6 M −1 and 1.28 × 10 −1 μM within the 95% confidence interval, respectively. Fig. 5c shows AFM image of SPR sensor surface after immobilization of 0.16 mM HA. As seen from the figure, it can be said HA molecules were homogeneously bound to SA molecules on the sensor surface.
Concentration of Hemagglutinin (mM) 0
Frequency Shift (ΔHz)
-100
0
0,05
0,1
0,15
0,2
-200 -300 -400 -500 -600 -700
y = -4000,5x - 10,299 R2 = 0,9959
b
Fig. 4. (a) Concentration–frequency relationship of 4-APBA–SA immobilized QCM electrode. (b) Morphology of QCM sensor surface after adsorption of 0.16 mM HA.
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
829
Fig. 6. (a) Comparison of selectivity of QCM nanosensors toward to Concanavalin A. (b) Comparison of selectivity of SPR nanosensors toward to Concanavalin A.
Fig. 5. (a) Change of concentration between reflectivity in SA immobilized SPR sensor. (b) Langmuir graph of 4-APBA–SA immobilized SPR sensor. (c) Surface morphology of SPR sensor after immobilization of 0.16 mM HA.
3.8. Selectivity and specificity of 4-APBA and SA immobilized QCM and SPR nanosensors In order to indicate the selectivity and specificity of 4-APBA and SA immobilized QCM and SPR nanosensors, experiments were performed with Concanavalin A. Thus, Concanavalin A solutions (0.01 mM, pH: 7.4 phosphate buffer) were applied to QCM and SPR nanosensors. As seen from Fig. 6a and b the QCM and SPR nanosensors did not give any responses to Concanavalin A solution (Δf = 3.331 and ΔR = 0.261) respectively. Also, the specific responses of the nanosensors to HA molecules, Δf = 39.951 and ΔR = 2.027 are excessively higher than to Concanavalin A. The selectivity ratio was calculated by dividing the QCM and SPR responses data of the Concanavalin A molecules and found 11.99 and 7.77 respectively. 4. Conclusions In the present study, affinity based-novel QCM and SPR sensors which recognize HA were improved by using 4-APBA–SA molecule
that has strong interactions with HA. For this purpose, 4-APBA was prepared. The preparation of 4-APBA is unique; by means of starting from cheap and easy to handle boronic acid and this is the first preparation and application of 4-APBA in literature. QCM electrodes and SPR chips were modified with thiol groups for the immobilization of 4-APBA. Then, HA detection was achieved by using prepared QCM and SPR sensors. Detection limits of QCM and SPR sensors for HA were found to be 4.7 × 10 −2 μM, (0.26 μg ml −1) and 1.28 × 10 −1 μM, (0.72 μg ml −1) in the 95% confidence interval, respectively. Also, the selectivity ratios were calculated for QCM and SPR nanosensors and found as 11.99 and 7.77 respectively. In literature, SPR and QCM based sensors frequently have been reported to provide information on molecular interactions among HA, natural receptors and influenza virus [31]. For example, Sato et al. used QCM to study the binding of influenza A virus to monosialoganglioside in membranes [36], Schofield and Dimmock used monoclonal antibody HC10 immobilized carboxylated dextran polymer matrix coated SPR chip for capture of influenza virus [33]. They reported the dissociation rate constant and association rate constant were found similar to those obtained with an affinity ELISA and Takemoto et al. developed a binding assay to study the interaction between influenza HA and its cell surface receptor SA [38]. But any study has not been reported about detection limit of HA via interaction with SA and HA molecules. We can say that, SA and 4-APBA based synthetic receptor method have compatible detection limit
830
S.E. Diltemiz et al. / Materials Science and Engineering C 33 (2013) 824–830
than the other HA detection methods which have been mentioned in literature. Also, most of the featured studies which have been used for detection of influenza have shortcoming of short shelf life, denaturizing problems, specific equipment's and expensive antibody–antigen interactions. Therefore, researching of new methods is strongly needed to overcome these disadvantages. In this study, novel biosensor systems which have long shelf life, easy storage conditions, simple and low cost properties to detect influenza were obtained. This new system can also be a very good alternative to biologic receptors. Our results not only demonstrate that HA detection is achieved using QCM and SPR nanosensors, but they can also be used potentially to detection of influenza with special conditions (training personel and cabinet). Acknowledgments Authors thank to Turkey National Boron Research Institute (BOREN) (project number: 2008.Ç0155) for supporting this study. Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.msec.2012.11.007. References [1] W. Weiss, J.H. Brown, S. Cusack, J.C. Paulson, J.J. Skehel, D.C. Wiley, Nature 333 (1988) 426–431. [2] J.J. Skehel, D.C. Wiley, Annu. Rev. Biochem. 69 (2000) 531–569. [3] R.G. Webster, W.J. Bean, O.T. Gorman, T.M. Chambers, Y. Kawaoka, Microbiol. Rev. 56 (1992) 152–179. [4] B. Sukla, T. Milan, W.C. Richard, Virus Res. 2 (1985) 61–68. [5] C. Rohm, N. Zhou, J. Suss, J. Mackenzie, R.G. Webster, Virology 217 (1996) 508–516. [6] Y. Suzuki. 33 (1994) 429–457. [7] L.L. Pang, J.S. Li, J.H. Jiang, Y. Le, G.L. Shen, R.Q. Yu, Sensors Actuators B 127 (2007) 311–316. [8] M. Plomer, G.G. Guilbault, B. Hock Enzyme, Microb. Technol. 14 (1992) 230–235. [9] E.P. Sochazewski, J.H.T. Luong, G.G. Guilbault, Enzyme Microb. Technol. 12 (1990) 173–177. [10] M. Lazerges, H. Perrot, N. Zeghib, E. Antoine, C. Compere, Sensors Actuators B 120 (2006) 329–337. [11] J.L.N. Harteveld, M.S. Nieuwenhuizen, E.R.J. Wils, Biosens. Bioelectron. 12 (1997) 661–667. [12] Y.S. Fung, Y.Y. Wong, Anal. Chem. 73 (2001) 5302–5309. [13] R. Say, A. Gultekin, A.A. Ozcan, Anal. Chim. Acta 640 (2009) 82–86. [14] A. Janshoff, H.J. Galla, C. Steinem, Angew. Chem. Int. Ed. Engl. 39 (2000) 4004–4032. [15] A. Dolatshahi-Pirouz, N. Kolman, A. Arpanaei, T. Jensen, M. Foss, J. Chevallier, P. Kingshott, J. Baas, K. Søballe, F. Besenbacher, Mater. Sci. Eng. C 31 (2011) 514–522.
[16] J. Homola, Chem. Rev. 108 (2008) 462–493. [17] U. Aldinger, P. Pfeifer, G. Schwotzer, P. Steinrücke, Sensors Actuators B 51 (1998) 298–304. [18] A. Tlili, M. Ali Jarboui, A. Abdelghani, D.M. Fathallah, M.A. Maaref, Mater. Sci. Eng. C 25 (2005) 490–495. [19] S.E. Diltemiz, R. Say, S. Büyüktiryaki, D. Hür, A. Denizli, A. Ersöz, Talanta 75 (2008) 890–896. [20] E.A. Turkoglu, K. Caktu, L. Uzun, H. Yavuz, Curr. Opin. Biotechnol. 22 (2011) 63–64. [21] L. Zhang, J. Liu, E. Wang, Biochem. Biophys. Res. Commun. 373 (2008) 202–205. [22] L. Quan, D. Wei, X. Jiang, Y. Liu, Z. Li, N. Li, K. Li, F. Liu, L. Lai, Anal. Biochem. 378 (2008) 144–150. [23] C.F. Mandenius, R. Wang, A. Alden, G. Bergstrom, S. Thebault, C. Lutsch, S. Ohlson, Anal. Chim. Acta 623 (2008) 66–75. [24] M. Del Toro, R. Gargallo, R. Eritja, J. Jaumot, Anal. Biochem. 379 (2008) 8–15. [25] L. Uzun, R. Say, S. Unal, A. Denizli, Biosens. Bioelectron. 24 (2009) 2878–2884. [26] T. Nagel, N. Gajovic-Eichelmann, S. Tobisch, U. Schulte-Spechtel, F.F. Bier, Clin. Chim. Acta 394 (2008) 110–113. [27] K.Y. Park, Y.S. Sohn, C.K. Kim, H.S. Kim, Y.S. Bae, S.Y. Choi, Biosens. Bioelectron. 23 (2008) 1904–1907. [28] S.E. Diltemiz, A. Denizli, A. Ersoz, R. Say, Sensors Actuators B 133 (2008) 484–488. [29] S. Peeters, T. Stakenborg, G. Reekmans, W. Laureyn, L. Lagae, A. Van Aerschot, M. Van Ranst, Biosens. Bioelectron. 24 (2008) 72–77. [30] A. Ersoz, S.E. Diltemiz, A.A. Ozcan, A. Denizli, R. Say, Biosens. Bioelectron. 24 (2008) 742–747. [31] Y. Amano, Q. Cheng, Anal. Bioanal. Chem. 381 (2005) 156–164. [32] X. Su, J. Zhang, Sensors Actuators B 100 (2004) 309–314. [33] D.J. Schofield, N.J. Dimmock, J. Virol. Methods 62 (1996) 33–42. [34] A.A. Kortt, E. Nice, Anal. Biochem. 273 (1999) 133–141. [35] E. Nice, T.L. McInerney, D.C. Jackson, Mol. Immunol. 33 (1996) 659–670. [36] T. Sato, M. Ishii, F. Ohtake, K. Nagata, T. Terabayashi, Y. Kawanishi, Y. Okahata, Glycoconj. J. 16 (1999) 223–227. [37] P. Critchley, N.J. Dimmock, Bioorg. Med. Chem. Lett. 12 (2004) 2773–2780. [38] D.K. Takemoto, J.J. Skehel, D.C. Wiley, Virology 217 (1996) 452–458. Sibel Emir Diltemiz was born in Turkey in 1977. She received the BS degree and PhD from the Department of Chemistry at Anadolu University, Eskisehir, Turkey, in 2002 and 2006, respectively. She is currently an assistant professor in Anadolu University. Her main research area is focused on SPR and QCM sensors and their applications. Arzu Ersöz was born in Turkey in 1961. She is the professor of the Department of Chemistry, Anadolu University, Turkey. She received her PhD from Hacettepe University in 1998. Her current research area is based on molecular imprinted polymers, synthetic receptors and quartz crystal microbalance. Deniz Hür was born in Turkey in 1975. He is an associate professor of the Department of Chemistry, Anadolu University, Turkey. He received his PhD from Anadolu University in 2004. His current researches are based on organic heterocyclic synthesis, microwave assisted organic synthesis and boron heterocyclics. Rüstem Keçili was born in Turkey in 1981. He is the PhD student of the Department of Chemistry, Anadolu University, Turkey. His current research includes the study of imprinted polymers and synthetic receptors. Ridvan Say was born in Turkey in 1967. He is the professor of the Department of Chemistry, Anadolu University, Turkey. He received his PhD from Hacettepe University in 1998. His current research includes the study of removal of heavy metal ions, polymeric sorbents, synthetic receptors, molecular imprinted polymers and biosensors.