Comparative Study of Random and Oriented Antibody Immobilization

Jul 13, 2010 - ACS eBooks; C&EN Global Enterprise ...... Nathan J. Alves , Matthew M. Champion , Jared F. Stefanick , Michael W. Handlogten , Demetri ...
0 downloads 0 Views 2MB Size
Download PDF
Anal. Chem. 2010, 82, 6401–6408

Comparative Study of Random and Oriented Antibody Immobilization Techniques on the Binding Capacity of Immunosensor A. Kausaite-Minkstimiene,† A. Ramanaviciene,†,‡ J. Kirlyte,† and A. Ramanavicius*,†,§ Nanotechnas - Centre of Nanotechnology and Material science, Vilnius University, Naugarduko 24, 03225 Vilnius, Lithuania, Department of Immunotechnology, State Research Institute “Centre of Innovative Medicine”, Zygimantu 9, 01102 Vilnius, Lithuania, and Institute of Chemistry, State Research Institute Centre for Physical and Technological Sciences, A. Gostauto g. 11, LT-01108 Vilnius, Lithuania A comparative study of four different antibody immobilization techniques that are suitable for modification of surface plasmon resonance (SPR) chip (SPR-chip) is reported. Antibodies against human growth hormone (anti-HGH) were used as the model system. The evaluated SPR-chip modification techniques were (i) random immobilization of intact anti-HGH (intact-anti-HGH) via selfassembled monolayer (SAM) based on 11-mercaptoundecanoic acid (MUA); (ii) random immobilization of intactanti-HGH within carboxymethyl dextran (CMD) hydrogel by direct covalent amine coupling technique; (iii) oriented coupling of intact-anti-HGH via Fc-fragment to protein-G layer assembled on SAM consisting of MUA (MUA/pG); (iv) oriented immobilization of fragmented anti-HGH antibodies (frag-anti-HGH) via their native thiol-groups directly coupled to the gold. To liberate these thiol groups, the intact-anti-HGH was chemically “divided” into two frag-anti-HGH fragments by chemical reduction with 2-mercaptoethylamine (2-MEA). Optimal concentration of 2-MEA for preparation of anti-HGH was 15 mM. The surface concentration of immobilized antibodies and the antigen binding capacity for all four differently modified SPR-chips was evaluated and compared. The maximum surface concentration of immobilized intact-anti-HGH was obtained by immobilizing the antibody within CMDhydrogel. The maximal antigen binding capacity was obtained by SPR-chip based on intact-anti-HGH immobilized via MUA/pG. The immobilization based on application of frag-anti-HGH was found to be the most suitable for design of SPR-immunosensor for HGH detection, due to its sufficient antigen binding capacity, simplicity, and low cost in respect to the currently evaluated techniques. Immunoassays and immunosensors based on antibodies have been used over thirty years and are still among the most important diagnostic tools, which are widely used in medical diagnostics, * Author to whom correspondence should be addressed. Phone: +370-52193115. Fax: +370-5-330987. E-mail: [email protected]. † Vilnius University. ‡ State Research Institute “Centre of Innovative Medicine”. § State Research Institute “Centre for Physical and Technological Sciences”. 10.1021/ac100468k  2010 American Chemical Society Published on Web 07/13/2010

environmental analysis, forensic medicine, and so on.1,2 Antibodies are used as probes in immunosensors.3,4 The major type of immunosensor relies on the ability of an immobilized antibody to recognize and to bind its associated target, which is known as an antigen. Recently, electrochemical impedance spectroscopy (EIS)5 and surface plasmon resonance (SPR)6 based immunosensors (SPR-immunosensors) have been used for the direct monitoring of antigen-antibody interactions. The SPR-immunosensors offer several significant advantages: there is no need for use of additional labels and for this reason the real-time detection is possible.7 Moreover curtailment of nonspecific binding is achievable, some immunosensors requires only a small volume of a sample and in some particular cases detection of very low concentrations of analytes is possible. Furthermore, by most successful immunosensors the analysis of antigen-antibody or other biomolecular interactions can be completed within a few minutes and very often it is possible to reuse the sensor chip for many times after proper regeneration of sensing surface.8 In principle, a surface plasmon oscillation is a localized wave that propagates along the interface between the gold film and the ambient medium and it is very sensitive to changes in the refractive index of media close to the gold surface when biomolecules binds to sensing surface.9 The efficient immobilization of an antibody on the solid surface of assay chips is an essential step in the preparation of these biosensors.10 A great variety of methods suitable for immobilization of antibodies has been described in the literature. Physical adsorption proves the easiest (1) Saerens, D.; Huang, L.; Bonroy, K.; Muyldermans, S. Sensors 2008, 8, 4669– 4686. (2) Suzuki, M.; Ozawa, F.; Sugimoto, W.; Aso, S. Anal. Sci. 2001, 17, i265i267. (3) Ramanavicius, A.; Kurilcik, N.; Jursenas, S.; Finkelsteinas, A.; Ramanaviciene, A. Biosens. Bioelectron. 2007, 23, 499–505. (4) Kurtinaitiene, B.; Ambrozaite, D.; Laurinavicius, V.; Ramanaviciene, A.; Ramanavicius, A. Biosens. Bioelectron. 2008, 23, 1547–1554. (5) Ramanavicius, A.; Finkelsteinas, A.; Cesiulis, H.; Ramanaviciene, A. Bioelectrochem. 2010, 79, 11–16. (6) Park, T. J.; Hyun, M. S.; Lee, H. J.; Lee, S. Y.; Ko, S. Talanta 2009, 79, 295–301. (7) Kausaite, A.; Van Dijk, M.; Castrop, J.; Ramanaviciene, A.; Baltrus, J. P.; Acaite, J.; Ramanavicius, A. Biochem. Mol. Biol. Educ. 2007, 35, 57–63. (8) Kausaite-Minkstimiene, A.; Ramanaviciene, A.; Ramanavicius, A. Analyst 2009, 134, 2051–2057. (9) Rusmini, F.; Zhong, Z.; Feijen, J. Biomacromolecules 2007, 8, 1775–1789. (10) Lee, J. M.; Park, H. K.; Jung, Y.; Kim, J. K.; Jung, S. O.; Chung, B. H. Anal. Chem. 2007, 79, 2680–2687.

Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

6401

immobilization method, but often suffers from random orientation and denaturation of immobilized antibodies, yielding poor reproducibility.11 More-stable immobilization of antibodies is obtained by covalent attachment. Furthermore, this immobilization method yields better reproducibility. Covalent immobilization of antibodies, however, causes disordered antibody orientation, which results in the loss of binding capability.12,13 The main reason for reduction of binding capability is based on random orientation of the antibody molecules and steric-hindrance induced by improper orientation of the antibody toward surface of solid substrate.14 Due to these reasons the selection/development of proper antibody immobilization method is strongly required for successful creation of immunosensors; to make a well oriented layer of antibodies and to minimize steric hindrances15 because the loss of binding capability reduces the sensitivity of SPR-immunosensor. Consequently the sensitivity of SPR-immunosensor can be increased by control of orientation and 2-dimensional (2D) configuration of antibodies immobilized on the SPR-chip surface.16 To develop an immunosensor with high performance the oriented immobilization of antibodies is most respectable.15 The advantages of oriented immobilization are 2-8 times higher antigen-binding capacity, which is resulting higher sensitivity of immunosensor and in many cases increased stability.17 Therefore the oriented antibody immobilization is one of the key issues in the development of the SPR-based and other biosensors. Several methods including self-assembled monolayer (SAM) technique have been proposed for the immobilization of oriented antibody molecules. This technique provides a powerful tool for immobilization of biological molecules on various solid substrates18 because SAM based immobilization technology provides stable covalent binding, it allows remarkable flexibility with respect to terminal functionality, oriented immobilization of ligands19 and it helps in minimization of nonspecific protein sorption on SPRchip.20 Furthermore, SAM-based technology provides (i) good stability under extreme pH and temperature; (ii) possible reusability and applicability in the flow-trough systems.21 In some cases covalent linking of antibodies to dextran-based hydrogels is also used due to the low nonspecific binding, an increase of antibody surface concentration if compared with SAM-based technology and versatility of derivation methods applied for dextran matri(11) Silvia, F.; Sally, P.; David, A. R.; Kim, E. S. Trends Anal. Chem. 2000, 19, 530–540. (12) Jyoung, J. Y.; Hong, S.; Lee, W.; Choi, J. W. Biosens. Bioelectron. 2006, 21, 2315–2319. (13) Nikin, P.; Martyn, C. D.; Mark, H.; Richard, J. H.; Clive, J. R.; Saul, J. B. T.; Philip, M. W. Langmuir 1997, 13, 6485–6490. (14) Tang, D. Q.; Zhang, D. J.; Tang, D. Y.; Ai, H. J. Immunol. Methods 2006, 316, 144–152. (15) Oh, B. K.; Kim, Y. K.; Lee, W.; Bae, Y. M.; Lee, W. H.; Choi, J. W. Biosens. Bioelectron. 2003, 18, 605–611. (16) Lee, S.; Sim, S. J.; Park, Ch.; Gu, M. B.; Hwang, U. Y.; Yi, J.; Oh, B. K.; Lee, J. J. Microbiol. Biotechnol. 2008, 18, 1695–1700. (17) Turkova´, J. J. Chromatogr. B 1999, 722, 11–31. (18) Lee, W.; Oh, B. K.; Bae, Y. M.; Paek, S. H.; Lee, W. H.; Choi, J. W. Biosens. Bioelectron. 2003, 19, 185–192. (19) Senaratne, W.; Andruzzi, L.; Ober, C. K. Biomacromolecules 2005, 6, 2427– 2448. (20) Uchida, K.; Otsuka, H.; Kaneko, M.; Kataoka, K.; Nagasaki, Y. Anal. Chem. 2005, 77, 1075–1080. (21) Kawaguchi, T.; Shankaran, D. R.; Kima, S. J.; Gobi, K. V.; Matsumoto, K.; Toko, K.; Miura, N. Talanta 2007, 72, 554–560.

6402

Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

ces.22 However these antibody molecules are randomly oriented and many of them are dysfunctional due to previously discussed steric factors. One of the best ways to achieve oriented immobilization of antibody molecules is their immobilization on a sublayer consisting of Fc-binding receptors, which specifically bind the Fc-part of antibody molecules.23,24 Protein-G and protein-A are cell wall proteins and they are found in most species of Streptococcus and Staphylococcus aureus, these proteins are most extensively studied as antibody-binding proteins, since they specifically interact with the Fc-part of antibodies that belongs to a class of immunoglobulins G (IgG).25-27 After this interaction the “paratope” of the antibody is located on the surface of the protein-G or protein-A. Therefore, it can be predicted that protein-G and protein-A based antibody immobilization leads to highly efficient immunosensors. However it is determined that the capacity of these proteins for binding of IgG varies with the species. In general, IgG’s have a higher affinity for protein-G than for protein-A, and protein-G can bind IgG’s from much wider variety of species. The affinity of various IgG subclasses, especially from mouse and human, for protein-A varies more than that for protein-G. In addition to these immobilization techniques the antibody fragments can be directly self-assembled onto hydrophilic gold via the terminal thiol functional groups. This immobilization is carried out using the native immunoglobulin thiol groups, which are liberated after the splitting of the intact antibody into two IgG fragments without the destruction of antigen-binding site of the antibody. The immobilization of IgG fragments on the gold surface is carried out by simple adsorption. This immobilization technique provides (i) oriented immobilization, (ii) proper orientation of IgG fragments, and (iii) high antigen binding capacity.28 Moreover, higher operational stability of IgG fragments modified gold surfaces is expected, which makes this immobilization technique very suitable for application in immunosensors. Human growth hormone (HGH) is a 22 kDa polypeptide secreted by the anterior pituitary gland. It is essential for normal growth and development. An excess of the growth hormone causes gigantism, hyperinsulinemia, impaired glucose tolerance, insulin resistance, and finally diabetes. A deficiency of HGH produces significantly different problems at various ages: in newborn infants it may cause hypoglycemia, growth failure in later infancy and childhood, and in adults lean body mass, poor bone density, and a number of physical and psychological symptoms, including poor memory, social withdrawal, and even depression.8 Consequently determination of HGH in blood serum is important for the diagnosis of disorders in HGH secretion. Detection of HGH is also used by sports authorities in doping detection because (22) Stigter, E. C. A.; de Jong, G. J.; van Bennekom, W. P. Biosens. Bioelectron. 2005, 21, 474–482. (23) Hoffman, W. L.; O’Shannessy, D. J. J. Immunol. Methods 1988, 112, 113– 120. (24) Oh, B. K.; Chun, B. S.; Park, K. W.; Lee, W.; Lee, W. H.; Choi, J. W. Mater. Sci. Eng., C 2004, 24, 65–69. (25) Boyle, M. D.; Reis, K. J. Biotechnol. 1987, 5, 697–703. (26) Bae, Y. M.; Oh, B. K.; Lee, W.; Lee, W. H.; Choi, J. W. Biosens. Bioelectron. 2005, 21, 103–110. (27) Di, G.; Nicole, M. B.; Jerome, S. S.; Yushan, Y.; Ashok, M.; Wilfred, C. J. Am. Chem. Soc. 2006, 128, 676–677. (28) Karyakin, A. A.; Presnova, G. V.; Rubtsova, M. Y.; Egorov, A. M. Anal. Chem. 2000, 72, 3805–3811.

HGH is one of the most habitual substances used to increase performance in some disciplines.29 To be able to determine amount of HGH suitable antibodies against human growth hormone (anti-HGH) immobilization protocol is required. There is variety of scientific papers and other informational issues related to immobilization of antibodies. Unfortunately, the most of this information is not showing clear comparisons between different immobilization techniques and are limited by presentation of purely empirical experiments. Consequently the objective of this study was to evaluate systematically the effect of few different immobilization techniques on the availability of binding sites of specific mouse anti-HGH antibodies. In the present paper the anti-HGH antibodies were immobilized by four different ways: (i) intact-anti-HGH via SAM based on MUA, (ii) intact-anti-HGH within CMD-hydrogel, (iii) intact-anti-HGH via protein-G coupled to MUA (MUA/pG), (iv) the anti-HGH antibodies were chemically divided into two parts of frag-anti-HGH and directly assembled on gold via terminal thiol-functional groups present in frag-anti-HGH. The surface concentration of immobilized intact-anti-HGH antibodies and frag-anti-HGH on differently modified surfaces was estimated by detection of antigen binding capacity. MATERIALS AND METHODS Materials. Purified human growth hormone (HGH) and monoclonal mouse antibodies against human growth hormone were purchased from AbD serotec (Oxford, United Kingdom). The protein-G and ethanolamine were purchased from Merck KGaA (Darmstadt, Germany). 11-mercaptoundecanoic acid (MUA), N-hydroxysuccinimide (NHS), N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC), sodium dodecylsulfate (SDS), and 2-mercaptoethylamine (2-MEA) were purchased from SIGMAAldrich (Steinheim, Germany). Unstained protein molecular weight marker was purchased from UAB FERMENTAS (Vilnius, Lithuania). Refractive index matching fluid (n ) 1.518) was obtained from Cargille Laboratories (Cedar Grove, NJ). All other chemicals were obtained from SIGMA-Aldrich (Steinheim, Germany) and were of analytical-reagent grade or better. All aqueous solutions were prepared in HPLC-grade water purified in a Purator-B Glas Keramic (Berlin, Germany). Preparation of Frag-Anti-HGH. The frag-anti-HGH were prepared by reducing the intact-anti-HGH antibodies with 15.0, 35.0, and 70.0 mM 2-MEA solution. To the Eppendorf tubes were added particular volumes of 10 mM Na-acetate buffer, pH 4.5, 3.7 mg/mL of intact-anti-HGH, and 1 mM of 2-MEA. The tubes were firmly closed and the reduction reaction was proceed for 1.5 h at 37 °C. Then the mixtures were dialyzed against Na-acetate for 4 h using tubes for dialysis with molecular weight cutoff membrane of 6-8 kDa, which were purchased from Novagen (Darmstadt, Germany). The frag-anti-HGH were used immediately after preparation. Electrophoresis. The equipment used for electrophoresis was purchased from UVItec (Cambridge, United Kingdom). The fraganti-HGH samples were analyzed using 9% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) by Laemmli.30 The current (29) Trevino, J.; Calle, A.; Rodriguez-Frade, J. M.; Mellado, M.; Lechuga, L. M. Talanta 2009, 78, 1011–1016. (30) Laemmli, U. K. Nature 1970, 227, 680–685.

strength of 10 mA was applied for 220 min to perform optimal protein separation. Surface Plasmon Resonance Measurements. A double channel SPR-analyzer “Autolab ESPRIT” obtained from ECO Chemie (Utrecht, Netherlands) was applied for SPR-measurements. One channel was used to perform assay. The second was used to run reference measurements. The geometrical setup of the instrument was in a Kretschmann configuration, and the detection was performed in the angular interrogation mode. A laser-diode was used as the light source to produce monochromatic light with a wavelength of 670 nm. The SPR-angle scan was performed around the manually fixed SPR-resonance position. The syringe and peristaltic pumps performed all liquid handling procedures. The syringe pump was used for sample mixing in the cuvette and for sample dispensing. The peristaltic pump was used for draining the cuvette, with the waste going into a waste flask. The SPR-angle was measured at a nonflow liquid condition, that is, with the circulating pump paused. All samples were monitored at a constant temperature of 20 °C. Cleaning of the SPR-Chips. Before the formation of selfassembled monolayer or immobilization of the frag-anti-HGH, bare sensor chip (SD AU, XanTec bioanalytics GmbH, Germany) was cleaned using piranha solution (1:3 30% H2O2 and concentrated H2SO4, respectively) at 60 °C for 5 min. Then, it was rinsed with pure ethanol solution and deionized water. Formation of MUA Self-Assembled Monolayer. To achieve this goal, the sensor chip was incubated in 1 mM MUA solution in methanol at 4 °C for 24 h. Afterward the MUA modified sensor disk was rinsed with deionized water and dried. Activation of MUA Self-Assembled Monolayer or CMDHydrogel. During this procedure functionally active NHS-esters were obtained due to the reaction of MUA or CMD-hydrogel (SD CMD20 M from XanTec bioanalytics GmbH, Germany) carboxyl groups with a mixture of 0.4 M EDC and 0.1 M NHS in water (Figure 1A-C, steps 1-2). The activation step was carried out for 5 min. Immobilization of the Intact-Anti-HGH by MUA Monolayer or CMD-Hydrogel. After activation of carboxyl functional groups immobilization of intact-anti-HGH was performed in both channels of the SPR-cell. The intact-anti-HGH was coupled covalently by uncharged primary amine functional groups (Figure 1A and B, step 3). Immobilization of 1.64 µM intact-anti-HGH antibodies dissolved in 10 mM Na-acetate buffer solution, pH 4.5 was carried out for 25 min. Coupling of the Protein-G. Coupling procedure was performed in both channels of the SPR-cell. The protein-G was coupled covalently via primary amine functional groups to activated carboxyl functional groups of the MUA self-assembled monolayer (Figure 1C, step 3). Coupling of 42.67 µM protein-G dissolved in 10 mM Na-acetate buffer solution, pH 4.5 was carried out for 25 min. After immobilization of the intact-anti-HGH or coupling of the protein-G it is essential to neutralize unreacted activated NHS-ester groups on the MUA self-assembled monolayer or CMD-hydrogel. Remaining reactive esters were deactivated with 1 M ethanolamine solution, pH 8.5 (Figure 1A-C, step 4). The deactivation step was carried out for 10 min. Immobilization of Intact-Anti-HGH by Protein-G. Immobilization was performed in both channels of the SPR-cell. The Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

6403

Figure 1. Principal immobilization schemes of (A-C) intact-anti-HGH and (B) frag-anti-HGH antibodies: (A) via self-assembled monolayer of MUA, (B) on carboxymethyldextran hydrogel, (C) on self-assembled MUA monolayer coupled protein-G, (D) directly on gold surface via native thiol functional groups present in frag-anti-HGH. Abbreviations R1 ) CH2CH3, R2 ) (CH2)3N(CH3)2, 2-MEA ) 2-mercaptoethylamine.

intact-anti-HGH was coupled covalently by uncharged primary amine functional groups to carboxy-terminal site of the protein-G (Figure 1C, step 5). Immobilization of 1.64 µM intact-anti-HGH antibodies dissolved in 10 mM PBS buffer, pH 6.0 was carried out for 25 min. Immobilization of Frag-Anti-HGH. The solution containing frag-anti-HGH was diluted with 10 mM Na-acetate buffer solution, pH 4.5, to a final concentration of 0.44 µM. Immobilization was performed in both channels of the SPR-cell. The frag-anti-HGH were coupled directly onto SPR-chip gold surface via terminal fraganti-HGH thiol functional groups (Figure 1D, step 2). Immobilization of frag-anti-HGH was carried out for 25 min. After the immobilization it is essential to block empty binding-able sites on the SPR-chip Au film. For this purpose modified sensor chip was incubated in PBS buffer, pH 7.4, containing 1 mg/mL BSA at room temperature for 8 h. Binding of HGH and Regeneration of SPR-Chip Surface. After the stable baseline was achieved the solution of HGH in 10 mM PBS buffer, pH 7.4, was injected. The interaction between immobilized intact-anti-HGH or frag-anti-HGH and HGH present in the solution was monitored for 15 min and followed by dissociation in PBS buffer solution, pH 7.4, for 3 min. Then sensor chip was regenerated with regeneration solution consisting of 50 mM of NaOH and 0.5% of SDS. The regeneration step was carried out for 5 min. RESULTS AND DISCUSSION Antibody immobilization is an essential process for the development of immunosensors because the choice of the im6404

Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

mobilization technique greatly affects antibody-antigen interactions. The surface concentration and orientation of antibodies on the sensor chip surface are important factors related to the sensitivity of SPR-immunosensor. To find optimal immobilization protocol in this work four antibody-immobilization techniques were compared. Three of them are typical and are based on the binding of intact-anti-HGH antibodies; the fourth technique is very innovative and based on application of frag-anti-HGH. These techniques were (i) random intact-anti-HGH immobilization over the MUA self-assembled monolayer (Figure 1A); (ii) random immobilization of intact-anti-HGH on CMD-hydrogel modified SPR-immunosensor chip by direct covalent amine coupling (Figure 1B); (iii) oriented intact-anti-HGH immobilization via MUA self-assembled monolayer coupled protein-G (MUA/pG) by indirect covalent coupling method due to affinity binding with an Fc-fragment of intact-antiHGH (Figure 1C); and (iv) oriented frag-anti-HGH immobilization on the gold by simple chemo-sorption of thiol-groups formed during splitting of anti-HGH (Figure 1D). Due to its advantages immobilization of biomolecules on SAMs is widely used in immunosensor design. One of the common strategies for immobilization of antibody on the electrode surface is covalent linking of amine groups of the antibody to the terminal carboxylic acid groups of a SAM formed on gold.31 Covalent linking of antibodies to dextran hydrogels is presently the most popular due to the low nonspecific binding and versatility of derivation methods suitable for modification of dextran matrices.22 (31) Fowler, J. M.; Stuart, M. C.; Wong, D. K. Y. Anal. Chem. 2007, 79, 350– 354.

Figure 2. SPR sensograms of the intact-anti-HGH immobilization via: (A) CMD-hydrogel; (B) self-assembled monolayer of MUA; and (C) the protein-G coupling to self-assembled monolayer of MUA. SPR chip modification/treatment steps represented in sensogram: 1, baseline in 10 mM Na-acetate buffer, pH 4.5 (from 1 until 120 s); 2, activation of the MUA or CMD-hydrogel with a mixture of EDC/NHS (from 120 until 420 s); 3, injection of Na-acetate buffer, pH 4.5 (from 420 until 450 s); 4, immobilization of intact-anti-HGH or coupling of protein-G (from 450 until 1960 s); 5, injection of Na-acetate buffer, pH 4.5 (from 1960 until 2070 s); 6, deactivation of remaining reactive NHS-esters with ethanolamine (from 2070 until 2670 s); 7, injection of Na-acetate buffer, pH 4.5 (from 2670 until 2700 s); 8, injection of regeneration solution consisting of 50 mM NaOH and 0.5% SDS (2700 until 2820 s); 9, injection of 10 mM PBS buffer, pH 7.4 (this step was continued until stabile baseline was obtained).

Furthermore, the hydrophilic dextran hydrogels offer a large surface area, which enables considerably higher biomolecule immobilization in comparison with SAMs. Figure 2 represents real SPR-data for intact-anti-HGH immobilization over the MUA monolayer (B) and CMD-hydrogel (A). The immobilization was supported by the electrostatic interaction between the negatively charged carboxyl functional groups present in the CMD-hydrogel matrix and MUA monolayer with the positively charged amino acid residues in the intact-anti-HGH. Hence, in this case the immobilization was achieved by the covalent coupling. Prior to immobilization, the modified SPR-chip surface was treated with coupling buffer (10 mM Na-acetate buffer, pH 4.5), until a stable resonance angle was maintained (Figure 2, step 1). Then terminal carboxyl groups were activated by incubation with the mixture of EDC and NHS in deionized distilled water to make an active intermediate of N-hydroxysuccinimide ester (Figure 1A and B, steps 1-2). The activation step was carried out for 5 min (Figure 2, line 2). This activation was essential for effective immobilization because the formation of a reactive group allowed the stable coupling of the intact-anti-HGH (Figure 1A and B, step 3). After the activation step the surface was washed with coupling buffer (Figure 2, step 3) and solution of 1.64 µM of intact-anti-HGH in coupling buffer was injected into the SPR-cell. At the same time registered sensorgrams showed a gradual increase in SPR-angle indicating binding of the intact-anti-HGH on the activated surface (Figure 2, step 4). The SPR-angle increased with time and reached steady-state condition within 25 min. Loosely bound intact-antiHGH was washed out from the surface by passing coupling buffer solution over the modified SPR-chip surface (Figure 2, step 5). However, after immobilization of intact-anti-HGH some activated N-hydroxysuccinimide ester groups left unreacted. To avoid the

binding of other proteins to these groups they were transformed into amides via reaction with ethanolamine.32 After the deactivation step (Figure 2, l step 6), extent of ethanolamine was washed out by passing 10 mM Na-acetate buffer solution over the intact-antiHGH modified surface (Figure 2, step 7). Along with that, the SPR-angle was registered during the flow of coupling buffer (Figure 2, step 7) ant it was slightly higher if compared with that registered after the immobilization step (Figure 2, step 5). The results illustrate that remaining unreacted N-hydroxysuccinimide ester groups are successfully blocked by ethanolamine (Figure 1A and B, step 4). After washing step the intact-anti-HGH modified SPR-chip with coupling buffer (Figure 2, step 7) the SPR-chip was subsequently treated by regeneration solution (Figure 2, step 8) and later by PBS buffer, pH 7.4 (Figure 2, step 9). Binding of the antibodies via protein-A or protein-G on the sensor chip surface is a promising option for oriented immobilization of antibodies. Both proteins specifically interacts with Fcdomain of antibodies and hence the paratope of IgG can face the opposite side of the immobilized proteins leading to a site-selective immobilization.26,33,34 However, the direct immobilization of these proteins on solid supports are less stable, for this reason they are generally immobilized through a spacer layer.35 Figure 2C shows real SPR-data for protein-G coupling. In our study the protein-G was coupled on the SPR-chip surface by chemical binding to a formed SAM consisting of MUA (Figure 1C, steps 1-4) by the same manner as described above. The shift of SPRangle obtained for the binding of protein-G to MUA was 418.43 ± 5 m°. From the shift of SPR-angle the surface concentration of the protein-G was calculated as 3.49 ± 0.04 ng/mm2. The first step of intact-anti-HGH immobilization via MUA/pG was based on treatment by 10 mM PBS buffer, pH 6.0, until a stable resonance angle was maintained (data not show). After injection into the SPR-cell of PBS buffer, pH 6.0, with 1.64 µM intact-anti-HGH a gradual increase of SPR-angle indicating binding of intact-anti-HGH on the MUA/pG modified sensor surface was detected in sensorgram (Figure 1C, step 4). The SPR-angle increased with time and reached steady-state conditions within 25 min. Any loosely bound intact-anti-HGH antibodies were removed from the surface by additional injection of PBS buffer, pH 4.5. The immobilization of biomolecules on gold supports is carried out using bifunctional reagents containing thiol groups, which bind strongly to the gold surface.36 On the other hand the biomolecule itself, can contain sulphide and/or disulfide groups that enables direct immobilization on gold. The anti-HGH antibodies belong to a special class of glycoproteins named imunoglobulin-G (IgG). The IgG molecule contains two couples of polypeptide chains with a molecular weight of ca. 150 kDa and has Y-shape. Each couple contains the heavy chain and the light chain. The two heavy chains are linked via disulfide bonds. The structure of IgG comprises two Fab-fragments used for binding specific antigens and an Fc(32) Subramanian, A.; Irudayaraj, J.; Ryan, T. Biosens. Bioelectron. 2006, 21, 998–1006. (33) Wua, B. Y.; Houb, S. H.; Huanga, L.; Yina, F.; Zhaoa, Z. X.; Anzaic, J. I.; Chen, Q. Mater. Sci. Eng., C 2008, 28, 1065–1069. (34) Jung, Y.; Jeong, J. Y.; Chung, B. H. Analyst 2008, 133, 697–701. (35) Shankaran, D. R.; Miura, N. J. Phys. D: Appl. Phys. 2007, 40, 7187–7200. (36) Lee, W.; Oh, B. K.; Lee, W. H.; Choi, J. W. Colloids Surf. B 2005, 40, 143– 148.

Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

6405

Figure 3. SDS-PAGE analysis of frag-anti-HGH fraction obtained by reduction of intact-anti-HGH antibodies with 15.0 (lane 2), 35.0 (lane 3), and 70.0 mM (lane 4) of 2-MEA. Molecular weight marker is on lane 1.

fragment.37 It is possible to split antibody by reducing agents and to obtain two fragments of antibody.38 It is expected that such procedure is not affecting the binding site of the antibody, which is formed by the coupling of heavy and light chains.28 To obtain frag-anti-HGH’s the intact-anti-HGH antibodies were chemically reduced with 2-MEA. Three different concentrations of 2-MEA (15.0, 35.0, and 70.0 mM) were used and prepared fraganti-HGH were analyzed using gel electrophoresis. Lanes 2, 3, and 4 of Figure 3 show three sharp bands with a molecular weight around of 25, 50, and 80 kDa. The band representing 80 kDa is equal to the half of molecular weight of intact-anti-HGH and can be assigned to the half-cleft anti-HGH (Figure 1D, step 1). 25 and 50 kDa bands indicate that during the reduction some disulfide bridges between the heavy and light chains of some anti-HGH molecules were dissociated and separated light and heavy chains at some extent were also present in the sample. Since the SDS-PAGE analysis did not show any significant differences in anti-HGH fragments obtained using various 2-MEA concentrations the surface concentration of immobilized frag-antiHGH and antigen binding capacity were investigated and compared. The immobilization of frag-anti-HGH was carried out using the thiol-groups of hinge part of anti-HGH (Figure 1D, step 2). Prior to immobilization, the bare SPR-chip surface was treated with 10 mM Na-acetate buffer, pH 4.5, until a stable resonance angle was maintained (data not show). Then Na-acetate buffer with frag-anti-HGH was injected into the SPR-cell. The SPR-angle indicating binding of the frag-anti-HGH on SPR-chip increased with time and reached steady-state conditions within 25 min. Loosely bound frag-anti-HGH were washed out from the surface by passing Na-acetate buffer solution over the modified SPR-chip surface. SPR-angle shifts for the binding of frag-anti-HGH obtained by reduction with 15.0, 35.0, and 70.0 mM of 2-MEA were 583.95 ± 9, 625.41 ± 10, and 636.08 ± 10 m° respectively. The surface concentration of the frag-anti-HGH calculated from these angle shifts was 4.87 ± 0.08, 5.21 ± 0.08, and 5.30 ± 0.08 ng/mm2. (37) Lu, B.; Smyth, M. R.; O’Kennedy, R. Analyst 1996, 121, 29R-32R. (38) Hermanson, G. T.; Mallia, A. K.; Smith, P. K. Immobilized Affinity Ligands Technique; Academic Press: San Diego, 1992.

6406

Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

Figure 4. Variation of SPR-angle shift with respect to concentration of 2-MEA (black bars) and the exposure of 1.59 µM of HGH (gray bars).

As shown in Figure 3, reduction of intact-anti-HGH with 2-MEA affected formation of frag-anti-HGH. In this study, both the cleavage of disulfide bridges between the two heavy chains of intact-anti-HGH and maintenance of the binding site of the antibody are important for the formation of biorecognition surface. The breakage of disulfide bridges of intact-anti-HGH increases probability for proper orientation of immobilized frag-anti-HGH and decreases loss of sensor selectivity, which is mainly caused by random orientation of biological recognition elements.39 Figure 4 shows the shift of SPR-angle caused by binding of frag-anti-HGH on the SPR-chip surface (black bar) and by formation of an immune-complex between immobilized frag-antiHGH and HGH present in the sample (gray bar) with respect to the concentration of 2-MEA. When frag-anti-HGH antibodies were produced by reduction in 15 mM of 2-MEA the lowest shift of SPR-angle (583.95 m°) was registered. As the concentration of 2-MEA was increased, the surface concentration of immobilized frag-anti-HGH increased and more significant shift of SPR-angle (625.41 m° and 636.08 m° for 35.0 mM and 70.0 mM of 2-MEA, respectively) was registered. But opposite effect was obtained for SPR-angle shifts, which were induced by HGH binding to immobilized frag-anti-HGH. When concentration of 2-MEA was increased, the SPR-angle shift induced by HGH binding was significantly decreased. Therefore, 15 mM of 2-MEA was found as the most optimal concentration for preparation of frag-anti-HGH. This concentration allowed us to reduce disulfide bridges between the two heavy chains of intact-anti-HGH and maintain high HGH binding capacity. Optimal condition for frag-anti-HGH immobilization was subsequently examined (data not shown) and 0.44 µM of frag-anti-HGH was considered to be the most optimal concentration for modification of SPR-chip surface and it was used throughout the study. Figure 5 shows the influence of different immobilization techniques used in this study on surface concentration of immobilized antibodies. The SPR-angle shifts for the binding of antibodies on SPR-chip surface were calculated as a difference between the SPR-signal prior to the binding and after treatment with coupling buffer. We found that absolute SPR-angle shifts caused by the immobilized molecules decreased by 807.53 ± 10 m° for the SPR-chip based on intact-anti-HGH immobilized within (39) Kortt, A. A.; Oddie, G. W.; Iliades, P.; Gruen, L. C.; Hudson, P. J. Anal. Biochem. 1997, 253, 103–111.

Figure 5. Dependence of SPR-angle shift on the technique of SPRchip modification: CMD, random immobilization of intact-anti-HGH within CMD-hydrogel; MUA/pG, oriented coupling of intact-anti-HGH via Fc-fragment to protein-G layer assembled on SAM consisting of MUA; Au, oriented immobilization of frag-anti-HGH via their native thiol-groups directly coupled to the gold; MUA, random immobilization of intact-anti-HGH via MUA based SAM.

CMD-hydrogel (CMD/intact-anti-HGH); 754.88 ± 9 m° for the SPR-chip based on intact-anti-HGH immobilized via MUA/pG (MUA/pG/intact-anti-HGH); 583.95 ± 6 m° for the SPR-chip based on frag-anti-HGH immobilized using native thiol groups of intactanti-HGH (Au/frag-anti-HGH); and 443.01 ± 5 m° for the SPRchip based on intact-anti-HGH immobilized via MUA monolayer (MUA/intact-anti-HGH). From the shift of SPR-angle, the surface concentration of the antibodies was calculated as 6.73 ± 0.08, 6.29 ± 0.08, 4.87 ± 0.05 and 3.69 ± 0.04 ng/mm2 for CMD, MUA/pG, Au and MUA modified SPR-chips respectively; whereas the shift of SPR-angle by 120 m° is equivalent to the change of surface concentration of protein by 1 ng/mm2.40 Coinciding with our results presented in Figure 5, the highest antibody immobilization level was found in the case of random immobilization of intact-anti-HGH within the CMD-hydrogel, whereas the least in the case of random immobilization of intactanti-HGH via MUA-based SAM. These results confirmed that dextran hydrogels offer a large surface area, which enables considerably higher surface density of immobilized antibodies in comparison with SAM-based technique.22 However despite the fact that the most of intact-anti-HGH antibodies were bound onto the CMD-hydrogel surface, the highest SPR-signal (184.75 m°) was observed by binding of HGH to the SPR-chip based on MUA/ pG/intact-anti-HGH (Figure 6). The SPR-signals induced by HGH binding to the SPR-chips based on CMD/intact-anti-HGH (20.73 m°) and MUA/intact-anti-HGH (16.51 m°) were, respectively, 8.9 and 11.2 times lower if compared with the signal obtained for MUA/pG/intact-anti-HGH. These results indicate that majority of antigen-binding sites on the randomly immobilized intact-anti-HGH were blocked due to incorrect antibody binding position, whereas in the oriented immobilization mode, the antigen-binding sites were better accessible from the solution front. It is expected that less than 10% of antibodies remain active when immobilized in a random orientation.41 Hence it can be concluded that CMDhydrogel is adsorbing relatively high number of antibodies, but (40) Karlsson, R.; Stahlberg, R. Anal. Biochem. 1995, 228, 274–280. (41) Brogan, K. L.; Wolfe, K. N.; Jones, P. A.; Schoenfisch, M. H. Anal. Chim. Acta 2003, 496, 73–80.

Figure 6. (A) The SPR-signals obtained by differently modified SPRchips. (B) Equilibrium angle dependence from antibody immobilization technique: a, MUA/intact-anti-HGH; b, CMD/intact-anti-HGH; c, MUA/ pG/intact-anti-HGH; d, Au/frag-anti-HGH. The HGH concentration in all tested samples was the same: 1.59 µM.

these antibodies might be (i) randomly oriented; many adsorbed antibodies are relatively far from sensing gold surface and some of them are completely out of range, which has influence on the SPR-signal; (ii) dextran especially if it is modified with adsorbed antibodies is forming relatively thick diffusion layer and it increases response time and at some extent it reduces sensitivity of SPR-sensor. Meanwhile the attached protein-G layer favored oriented immobilization of the intact-anti-HGH antibodies for sensitive detection of HGH. Since each protein-G molecule has more than one binding site for binding of antibody via Fc-domain, the binding of the intact-anti-HGH antibodies to protein-G independent of the orientation of the protein-G was possible. Because the binding of intact-anti-HGH antibodies occurred via Fc-domain, the antibody binding-site was not distorted. Thus, the immobilization of the intact-anti-HGH antibodies via MUA/pG results in uniform, stable and sterically accessible biological recognition layer. However, the limitation of intact-anti-HGH immobilization via MUA/pG is the necessity to immobilize the protein-G layer. Furthermore, we could not find such regeneration solution, which will just dissociate antigen-antibody complex and will not dissociate intact-anti-HGH complex with immobilized protein-G. The next most intensive SPR-signal (142.68 m°) was observed by application of SPR-chip based on Au/frag-anti-HGH (Figure 6d). It was only 1.3 times lower than the signal obtained for SPRchip based on MUA/pG/intact-anti-HGH (Figure 6c). This is in agreement with reports, which claimed that immunosensing surfaces based on half-cleft IgG result in enhanced antigen-binding activity when compared to surfaces based on intact IgG.42,43 The binding-activity of frag-anti-HGH compared with that of intact-antiHGH immobilized via MUA/pG may be influenced by partial inactivation of the antibody binding-site during the preparation of frag-anti-HGH and/or later by incubation of frag-anti-HGH modified SPR-chips in BSA solution with the aim to block empty nonspecific binding sites present on the SPR-chip. Nevertheless the immobilization technique based on interaction between the thiol-group of IgG fragment and gold surface is advantageous because the immobilized antibodies maintain both: (42) O’Brien, J. C.; Jones, V. W.; Porter, M. D.; Mosher, C. L.; Henderson, E. Anal. Chem. 2000, 72, 703–710. (43) Wang, H.; Wu, J.; Li, J.; Ding, Y.; Shen, G.; Yu, R. Biosens. Bioelectron. 2005, 20, 2210–2217.

Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

6407

high antigen binding constants and high stability. This immobilization technique provides proper orientation of IgG fragments; so that there is a similar distance between the binding site of the antibody and the surface of SPR-chip; it does not cause the distribution of the apparent affinity constants.28 Due to sufficient antigen binding capacity, simplicity of accomplishment and low cost in respect with others in this study evaluated techniques, the immobilization based on application of frag-antiHGH was found to be the most suitable for design of SPRimmunosensor devoted for HGH detection. Here evaluated results corresponds with results presented by other authors, that are stating that method based on immobilization of antibody fragments eliminates the random orientation of antibody on the surface, consequently increasing the antibodyantigen binding capacity, what increases the coupling of the antibody fragments, increases antigen binding capacity and decreases nonspecific binding.44,45 The comprehensive analysis of random immobilization of intact antibodies and oriented immobilization of fragmented antibodies via self-assembled monolayer were studied by surface plasmon resonance technique.46 For a particular antibody/antigen system, the optimized fragmentation protocol in combination with an oriented immobilization of Fab’ fragments on mixed SAMs leads to a >2-fold increase of the antigen binding signals when compared with signals registered with analytical systems based on randomly covalent immobilized intact antibodies.46 CONCLUSIONS The main objective of this study was to evaluate the effect of random and oriented antibody immobilization strategies on the (44) Vikholm-Lundin, I. Langmuir 2005, 21, 6473–6477. (45) Vikholm, I. Sens. Actuators B 2005, 106, 311–316. (46) Bonroy, K.; Frederix, F.; Reekmans, G.; Dewolf, E.; De Palma, R.; Borghs, G.; Declerck, P.; Goddeeris, B. J. Immunol. Methods 2006, 312, 167–181.

6408

Analytical Chemistry, Vol. 82, No. 15, August 1, 2010

availability of active antigen-binding sites. The surface concentration of immobilized intact-anti-HGH and frag-anti-HGH and their antigen binding capacity were tested by mean of surface plasmon resonance. Comparing random immobilization with oriented one, it was found that the random immobilization within CMD-hydrogel provides much higher surface concentration of immobilized intactanti-HGH antibodies, whereas the oriented immobilized antibodies provide considerably higher antigen binding capacity. Analytical signal registered by SPR-chip modified with intact-anti-HGH via MUA/pG was 8.9 and 11.2 times higher if compared with signal registered with SPR-chip modified with intact-anti-HGH immobilized within CMD hydrogel and via MUA, respectively. The analytical signal registered by SPR-chip modified with frag-antiHGH via native thiol groups was just 1.3 times lower than the signal registered by intact-anti-HGH immobilized via MUA/pG. These results indicate that most of antigen-binding sites on the randomly immobilized intact-anti-HGH were blocked due to random antigen-binding position, whereas in the oriented immobilization mode, the antigen-binding sites were more accessible from the solution front. Due to sufficient antigen binding capacity, simplicity of accomplishment and low cost in respect of other here evaluated techniques the immobilization based on application of frag-anti-HGH was found to be the most suitable for design of SPR-immunosensor devoted for HGH detection. ACKNOWLEDGMENT This work was supported by the Lithuanian Science Council.

Received for review February 21, 2010. Accepted June 17, 2010. AC100468K