Stainless steel surface functionalization for immobilization of antibody fragments for cardiovascular applications  ska,3 A. Foerster,1 I. Hołowacz,1 G. B. Sunil Kumar,2 S. Anandakumar,2 J. G. Wall,2 M. Wawrzyn 4 4 5 1 6 7  ska-Janus, S. J. Hinder, D. Bialy, M. Paprocka, A. Kantor, H. Kraskiewicz, S. Olsztyn  ska1 H. Podbielska,1 M. Kopaczyn 1

Departament of Biomedical Engineering, Faculty of Fundamental Problems of Technology, Wroclaw University of Technology, Poland 2  Microbiology and Centre for Research in Medical Devices (CURAM), NUI Galway, Galway, Ireland 3 Department of Medical Emergency, Wroclaw Medical University, Wrocław, Poland 4 Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, Wrocław, Poland 5 Balton Ltd, Warsaw, Poland 6 Department of Mechanical Engineering Sciences, University of Surrey, England 7 Clinic of Cardiology, Wroclaw Medical University, Wrocław, Poland Received 19 January 2015; revised 23 September 2015; accepted 11 November 2015 Published online 15 December 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.35616 Abstract: Stainless steel 316 L material is commonly used for the production of coronary and peripheral vessel stents. Effective biofunctionalization is a key to improving the performance and safety of the stents after implantation. This paper reports the method for the immobilization of recombinant antibody fragments (scFv) on stainless steel 316 L to facilitate human endothelial progenitor cell (EPC) growth and thus improve cell viability of the implanted stents for cardiovascular applications. The modification of stent surface was conducted in three steps. First the stent surface was coated with titania based coating to increase the density of hydroxyl groups for successful silanization. Then silanization with 3 aminopropyltriethoxysilane (APTS) was performed to provide the surface with amine groups which presence was verified using FTIR, XPS, and fluorescence microscopy. The maximum density of amine groups (4.8*1025 mol/cm2) on the surface was reached after reaction taking place in ethanol for

1 h at 608 C and 0.04M APTS. On such prepared surface the glycosylated scFv were subsequently successfully immobilized. The influence of oxidation of scFv glycan moieties and the temperature on scFv coating were investigated. The fluorescence and confocal microscopy study indicated that the densest and most uniformly coated surface with scFv was obtained at 378C after oxidation of glycan chain. The results demonstrate that the scFv cannot be efficiently immobilized without prior aminosilanization of the surface. The effect of the chemical modification on the cell viability of EPC line 55.1 (HucPEC-55.1) was performed indicating that the modifications to the 316 L stainless steel are non-toxic to C 2015 Wiley Periodicals, Inc. J Biomed Mater Res Part A: EPCs. V 104A: 821–832, 2016.

Key Words: aminosilanization, surface modification, antibody fragments, immobilization, progenitor cells

 ska M, Paprocka M, How to cite this article: Foerster A, Hołowacz I, Sunil Kumar GB, Anandakumar S, Wall JG, Wawrzyn  ska-Janus S, Hinder SJ, Bialy D, Podbielska H, Kopaczyn  ska M. 2016. Stainless steel surface Kantor A, Kraskiewicz H, Olsztyn functionalization for immobilization of antibody fragments for cardiovascular applications. J Biomed Mater Res Part A 2016:104A:821–832.

INTRODUCTION

Coronary artery disease is the most common and severe type of cardiovascular disease worldwide.1 Stent implantation is an efficient and safe intravascular intervention used to treat coronary artery disease.2–4 Although the use of coronary stents minimizes the risk of complications during medical procedures,4 restenosis frequently occurs in patients in vivo after stent implantation. This results in a need for repeated procedures with associated reduced quality of life for patients and high healthcare costs.5–7 Several potential solutions have been reported, including

pharmacological treatments and the application of additional devices, but none has successfully eliminated the risk of post-stenting restenosis.8 One approach to solve this problem, which we describe in this work, is to coat endothelial precursor cell-binding antibody fragments on the stent surface in order to facilitate EPC growth and thus improve cell viability of the implanted stents. Our work presents a method for binding antibody fragments on amine functionalized stent surface. First, the amine groups are attached to the titania-coated solid surface followed by immobilization of glycosylated antibody

 ska; e-mail: [email protected] Correspondence to: M. Kopaczyn Contract grant sponsor: European Union (FP7) “EPICstent’’ Project; contract grant number: FP7-PEOPLE-2012-IAPP-324514

C 2015 WILEY PERIODICALS, INC. V

821

fragments specific toward vascular endothelial growth factor. Vascular endothelial growth factor (VEGF) is a family of endothelial cell-associated glycoproteins9,10 which are key factors in blood vessel formation and development of the vascular and lymphatic systems.11–13 Five different growth factors have been identified which bind to three different receptor tyrosine kinases called VEGFR-1, VEGFR-2 and VEGFR-3.10,14 Of these, VEGFR-2 is critical for the development of vascular endothelial cells.15–18 Single chain antibody fragments (scFv) are small (5 3 4 3 5 nm3), 25–30 kDa monovalent recombinant antibody fragments that are formed by connecting antibody heavy and light chain variable domains using a short polypeptide linker. The immobilization of scFvs on solid supports is an important step in diverse applications including biosensors, affinity chromatography, and immunoassays19–22 in which the use of scFvs rather than whole immunoglobulin molecules can greatly improve detection sensitivities because of their small size and increased coating densities.23,24 Common methods for attachment of antibodies to solid surfaces include physical adsorption, chemical cross linking, covalent bonding, or entrapment in a gel network.25,26 Numerous studies of the use of chemical methods for sitespecific immobilization of whole antibody molecules onto inorganic substrates have been reported. One of the most common methods employs the prior immobilization of antibody-binding protein A or G,27 or the use of polymers with carefully designed functional groups to control antibody orientation.28 Tan et al. described the preparation of a novel polyhedral oligomeric silsesquioxane poly(carbonateurea) urethane (POSS-PCU) nanocomposite polymer with covalently attached anti-CD34 antibodies and demonstrated their use for capture of circulating endothelial progenitor cells (EPC).29 In the case of recombinant antibody fragments, the absence of useable moieties from whole antibodies such as naturally occurring glycan chains or protease cleavage sites can be overcome by the ability to design and engineer specific functional groups for covalent, oriented immobilization.30 Surface-bonded biomolecules can also be obtained by treating the substrate with silane molecules. One of the best studied silane molecules for protein attachment is 3aminopropyltriethoysilane (APTES).31–33 APTES may form polysiloxanes in contact with water as a result of its quick hydrolysis and subsequent condensation of silane triols. Contact with the substrate surface occurs initially via hydrogen bonding, and later when the material dries, bonds between the metal surface, oxygen, and silicon are generated.34 Another approach reported in the literature35 involves immobilization of murin monoclonal anti-human CD34 antibodies covalently bonded with polysaccharide to attract endothelial progenitor cells circulating in the blood stream. As prepared surface of stainless steel stent has been showed to be safe and feasible. In our work, in contrast to their work, we have used antibody fragments (scFv), rather than whole immunoglobulin molecules. This approach is

822

FOERSTE ET AL.

aiming to greatly improve detection sensitivities due to the small size and increased coating densities of scFv. In this study, a 316 L stainless steel metal surface was biofunctionalized with VEGFR2-binding antibody fragments. First, the surface was coated with titania-based coating to increase the density of hydroxyls groups. The surface with these groups was then silinized with 3-(aminopropyl) triethoxysilane (APTES) to provide amine groups for covalent bond formation with the aldehydes groups introduced into the antibody fragment glycan chains upon oxidation. Oxidation of the glycan chain attached to the scFv generates aldehydes from the branched glucose monomer, providing reactive potential with the surface amines for covalent attachment.30 There are three main goals toward which this work was directed. The first one was to establish the optimal conditions for covering of 316 L stainless steel surfaces with amino groups necessary for covalent attachment of scFvs. Second one, to investigate the influence of oxidation of scFv glycan moieties and the immobilization temperature on antibody fragment uniformity and density coverage on the surface. The third one was to investigate the effect of each stage of chemical modification on cell viability. MATERIALS AND METHODS

Aminosilanization of substrates Stainless steel disks (316 L) (Balton Sp. Z o. o.), 10 mm in diameter and 0.5 mm in thickness, were coated with a titania-based material prior to use. The titania-based coating was prepared using a sol-gel method36 to introduce OH groups on the disk surface for amine group attachment. For sols preparation tetraethyl orthotitanate (TEOT) (Sigma Aldrich) and ethyl alcohol (EtOH) (POCH) were used at a molar ratio of 1:44 for TEOT:EtOH. Hydrochloric acid (36%) was used for acid hydrolysis (pH 2). Prepared substrates were immersed in (3-aminpropyl)triethoxysilane (APTS) (Sigma Aldrich) water and ethanol solutions with varying concentrations (0.001–0.04M), reaction times (15 min, 1 h), and temperatures (25, 60, 1208C) to introduce amine groups on their surfaces. Substrates thus prepared were rinsed with ethanol and deionized water and dried under air. FTIR-ATR spectroscopy To identify functional groups on the surface of stent material infrared spectroscopy has been used. Modified surfaces of 316 L discs were evaluated by FTIR-ATR spectroscopy. ATR spectra were taken on a Nicolet 6700 FTIR spectrometer (Thermo Scientific, USA) based on the attenuated total reflection method.37 A sample was placed directly on the Diamond Top-Plate of the Golden Gate MK II ATR Accessory (Pike, USA). Spectra were recorded at a constant temperature of (21 6 1) 8C. The number of scans was 32 and the resolution was 4 cm21. The FTIR-ATR spectra were recorded using OMNIC software. Standard spectral analysis in the range 4000–500 cm21 was performed by the Origin program, as previously described.38,39 Second-order derivatives were calculated to obtain the precise positions of the vibrations bands.

IMMOBILIZATION OF RECOMBINANT ANTIBODY FRAGMENTS ON STAINLESS STEEL 316 L

ORIGINAL ARTICLE

X-ray photoelectron spectroscopy XPS analyses were performed on a ThermoFisher Scientific Instruments (East Grinstead, UK) Theta Probe spectrometer. XPS spectra were acquired using a monochromated Al Ka Xray source (hm 5 1486.6 eV). An X-ray spot of 400 lm radius was employed. Survey spectra were acquired using a Pass Energy of 300 eV. C1s, O1s and Ti2p high resolution, core level spectra were acquired using a Pass Energy of 50 eV. All other spectra were acquired employing a Pass Energy of 100 eV. Charge compensation was achieved using a low energy electron flood gun. All spectra were charge referenced against the C1s peak at 285 eV to correct for charging effects during acquisition. Quantitative surface chemical analyses were calculated from the high resolution, core level spectra following the removal of a non-linear (Shirley) background. The manufacturers Avantage (v4.88) software was used which incorporates the appropriate sensitivity factors and corrects for the electron energy analyser transmission function. Fluorescence microscopy Fluorescence microscopy was used to determine the density of amine groups on coated titania-based disks. Functional amine groups on disk surfaces were detected by reaction with 1 mM fluorescein isothiocyanate (FITC) (Sigma Aldrich) at 48C overnight after which samples were washed with Phosphate Buffered Saline (PBS) (Sigma Aldrich) and viewed under a fluorescence microscope (Leica FL-800 microscope). The intensity of fluorescence released from the surface was directly proportional to the number of amino groups that could react with FITC. Fluorescence measurements were performed with Image J. scFv expression and purification The VEGFR2-binding scFv #32 was isolated from a human scFv phage-antibody library40,41 and expressed in Escherichia coli CLM37 cells in the format ompA leader peptidescFv-6His-glycosylation tag where the bacterial leader peptide directs the scFv to the oxidizing periplasmic space of the bacterial cell for stable disulphide bond formation, the hexahistidine tag allows immunodetection and purification of the translated protein, and the C-terminal glycosylation recognition site enables glycosylation of the translated scFv in E. coli cells containing the pgl protein glycosylation machinery from Campylobacter jejuni.42,43 For expression, a single, freshly-transformed E. coli colony containing the scFv expression vector and the pACYC-pgl protein glycosylation vector42,43 was used to inoculate 400 mL of ZYP-5052 autoinducing medium44 containing 34 lg/mL chloramphenicol and 100 lg/mL ampicillin, followed by incubation with shaking at 378C until an OD600 of 0.9 was reached. Induction was then allowed to proceed at 258C for 24 h, followed by harvesting of cells by centrifugation and fractionation as described elsewhere45 to isolate periplasmic proteins. Samples were passed through a 0.45 lm filter to remove particulate material and the scFv was purified via its hexahistidine tag by immobilized metal affinity chromatography (IMAC) on a nickel affinity column.46 Eluted fractions were

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | APR 2016 VOL 104A, ISSUE 4

dialyzed against phosphate-buffered saline and analyzed by SDS–PAGE and immunodetection using an anti-polyhistidine horse radish peroxidase (HRP)-conjugated antibody before immobilization. scFv immobilization The VEGFR2-binding scFv #32 antibody fragments were selected because of their selective binding toward VEGFR2 on the surface of endothelial cells. The use of scFv fragments rather than whole immunoglobulin molecules removes multiple non-antigen-binding domains that could mediate non-specific protein adsorption in vivo. Therefore protein adsorption is expected to be significantly reduced. Immobilization of oxidized and nonoxidized anti-VEGFR2 scFv fragment #32 was carried out by depositing 20 lL of 60 lg/mL scFv in 5 mM phosphate buffer solution on nonaminated T15 and aminated (T15-NH2) surfaces for comparison. Antibody fragments were allowed to react with functionalized surfaces for 1 h at 258 or 378 C. Oxidation of the glycosylated scFv (Gly-scFv) was carried out as previously reported.30 Briefly, 10 mM meta-sodium periodate was dissolved in sodium acetate buffer (0.1 M) and GlyscFvs were added and incubated for 30 min at room temperature. After dialysis against 100 mM NaCl, scFvs were used for immobilization. Immobilized scFvs were detected using an anti-6X His reporter antibody labeled with DyLight 488 at 258C over night, followed by extensive rinsing with PBS and drying prior to microscopy. Confocal scanning laser microscopy Fluorescence detection and localization of antibody fragments were carried out using a Leica TCS SPE confocal scanning laser microscope with LAS AF software was used. For excitation the laser line at 488 nm was operated at 25% of maximum power and an optimal gain setting of 800 V was used. The emission signal was detected at 520 nm. Using the software the intensity of fluorescence was determined along lines drawn on the image. Atomic force microscopy Surface characterization was done by using AFM technique. Antibody fragments were dissolved in phosphate buffered saline (PBS), then 150 mM magnesium chloride, 150 mM potassium chloride and 20 mM HEPES were added. The outcome protein concentration was 100 mM. The immobilization of antibody fragments was obtained on two types of surface: on aminated and non-aminated surface (T15). AFM images were acquired in the tapping mode using a Nanoscope IIId scanning probe microscope with Extender Module (Brucker) in the dynamic modus. Olympus etched silicon cantilevers were used with a typical resonance frequency in the range of 200–400 kHz and a spring constant of 42 N/m. All samples were measured at room temperature in air. The sample was first adjusted with an optical light microscope (Nanoscope, Optical Viewing System). Structural analysis and 3D projections of acquired images were obtained by using Nanoscope v.6.13 software.

823

FIGURE 1. Schematic of step by step procedure for 316 L stainless steel chemical surface modification with 3- aminopropyltriethoxysilane (APTS) including characterization methods taken at each stage.

Cell viability study The cell viability study of the modified 316 L materials was assessed using human endothelial progenitor cell line 55.1 (HucPEC-55.1). Metal disks with and without surface modifications were sterilized by 15 min incubation in 70% ethanol followed by washing once in dH2O and twice in PBS. There are various methods that could be used to sterilize the stents surface such as using heat, chemicals, irradiation or high pressure. Further studies will be performed to find the best method for our stents to avoid any damage to scFv fragments. Each disk was placed in a 24-well plate (n 5 2 per group) and cells were seeded at 187,000 cells per well in 0.5 mL of medium. Following 24 h incubation at 378C, 5%

CO2 disks were transferred into new wells and cells were trypsinized for 5 min at room temperature. To deactivate trypsin 400 lL of Optimum medium supplemented with 3% fetal calf serum (FCS) and 1% Pen/Strep was added to each well and cells were counted using a culture microscope (Olympus, CK40-SL, Japan).

RESULTS

A silane agent terminated with amine a group was used to enable antibody fragments to be bond to stainless steel surface. The OH groups necessary for successful aminosilanization were provided to the substrate by coating a surface with titania based coating prior to silanization. Figure 1

FIGURE 2. Normalized FTIR spectrum (ATR mode) of T15 surface (a) before and (b) after aminosilanization for 1 h at 608C in 0.04M APTS in ethanol. The absorbance band in spectrum (a) at 3390 cm21 indicates the presence of OH groups in the titania-based coating that are involved in aminosilanization process. The absorbance bands observed at 3400 and 3303 cm21 and 1605 cm21 in spectrum (b) confirm the presence of amine groups on the surface of T15 coating.

824

FOERSTE ET AL.

IMMOBILIZATION OF RECOMBINANT ANTIBODY FRAGMENTS ON STAINLESS STEEL 316 L

ORIGINAL ARTICLE

FIGURE 3. X-ray photoelectron spectroscopy survey spectra of T15 and T15-NH2 surfaces. Survey spectra identified elements C (51.76%), O (37.08), and Ti (10%) in T15, and C (46.1%), O (28.97%), Si (14.69%), and N (10.25%) in T15-NH2.

shows schematic process of stent surface functionalization with amine groups. Figure 2 presents an infrared spectrum obtained from the titanium-coated stainless steel surface before (T15) and after (T15-NH2) aminosilanization. The broad absorption band centered at 3220 cm21 in spectrum (a) corresponding to the surface prior to aminosilanization is assigned to stretching vibration of molecular water and OH groups.47 Additional peaks observed at 2921 cm21 and 2847 cm21 are attributed to the asymmetric and symmetric stretching modes of CH2 groups.48,49 The absorption band noted at 1646 cm21 is assigned to asymmetric COO stretching vibrations. Carbonyl groups are the products of photocatalytic oxidation of ethanol residues in the presence of TiO2. TiO2 particles are well known from their catalytic activity upon light illumination.50,51 The band at 1090 cm21 is because of the presence of Ti-O-C bridging vibrations.52,53 Vibrations of Ti-O and Ti-O-Ti are represented in the spectrum by the peaks observed at 817 and 660 cm21, respectively.54–56 In spectrum (b) in Figure 2, representing the same surface after aminosilanization (T15-NH2), peaks at 3355 and 3294 cm21 are assigned to asymmetric and symmetrically N-H stretching modes, respectively. The additional absorbance band observed at 1570 cm21 is assigned to amine groups while the peak at 1098 cm21 is because of Si-O-Si vibrations57,58 and is attributed to siloxane units formed during the silanization process. Based on these results it is apparent that the titanium-coated 316 L stainless steel discs were successfully aminosilanized.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | APR 2016 VOL 104A, ISSUE 4

XPS analysis was performed to investigate the surface chemistry of the titanium-coated stainless steel surface before (T15) and after aminosilanization (T15-NH2). Figure 3 shows the survey spectra of the samples. The atomic percentages of the elements are calculated from these spectra and are also presented in the figure. It can be observed in Figure 3(a) that the T15 surface is mainly composed of C (51.76%), O (37.08), and Ti (10%). There have been also detected small residues (1.165) of Cl which can originate from hydrochloric acid that was used for the synthesis. Figure 3(b) shows the survey spectra of T15-NH2 surface where the presence of C, O, Si and N elements can be seen in atomic percentage of 46.1, 28.97, 14.69, and 10.25 respectively. High resolution spectra of detected C 1s, O 1s, N 1s and Si 2p on the surface of T15-NH2 is summarized in Figure 4. Only one pick was observed for O 1s at 532.4 eV related to O-Si bond (1). In XPS spectra of C 1s two pics were observed, at 285 and 286 eV for C-C and C-N bonds respectively. In the XPS spectra of N 1s the pick at 399.3 is observed that is assigned to amine groups. Additionally the presence of Si 2p peak at 102.6 indicates silane grafting. To determine the optimal conditions for the densest functional groups coverage of the stent surface various parameters were investigated. The influence of reaction time (15 and 60 min), temperature (25, 60, and 1208C), and concentration (0.001, 0.004, 0.04 mol/dm3) and type of solvent on the concentration of amine groups introduced was evaluated by fluorescence microscopy using FITC to quantify theNH2 groups on disc surfaces. As can be seen in Figure 5,

825

FIGURE 4. X-ray photoelectron spectroscopy analysis of the T15-NH2 surfaces. High resolution spectra of Si 1s, N 1s, C 1s and O 1s. The observed pick in spectra of N at 399.3 eV assigned to amine groups together with the presence of Si 2p peak at 102.6 eV indicates silane grafting and succesful aminosilanization.

the density of amine groups increases with increasing temperature from 25 to 608C, and further to 1208C, in particular for samples prepared at low APTS concentration (0.001M). The loss of amino groups could be because of the presence of non-covalent silane bounds and their evaporation, or catalysis of amine oxidation resulting in the loss of nitrogencontaining molecules.45,46 As can be seen in Figure 5(a) lower surface density of amines resulted from reactions carried out in water compared with ethanol; this occurred after 15 and 60 min incubations and at both temperatures for all three APTS concentrations and may be explained by the higher solubility of APTS in ethanol than in water. Meanwhile, the density of amine groups increased with aminosilanization time for both solvents up to maximal values at 1 h of 3*10–5, 3.6*10–5, and 4.8*10–5 mol/cm2 for 0.001, 0.004, and 0.04M solutions, respectively, in ethanol and 1.27*10–5, 1.3*10–5, and 2.4*10–5 mol/cm2 for the corresponding reactions in water. As can be seen in Figure 5, the density of amino groups on the titania-coated substrates increased with increasing APTS concentrations at each temperature and all incubation times in both ethanol and water, with a maximum NH2 density (4.8*10–5 mol/cm2) reached after 1 h at 608C and with an APTS concentration of 0.04M in ethanol.

826

FOERSTE ET AL.

Immobilization of antibody fragments on aminated surface was performed. By oxidizing glycan chain incorporated to antibody fragment, it was possible to covalently attach the antibodies fragments to the surface. For comparison immobilization of non oxidized antibody was performed as well. The temperature effect on immobilization efficiency, antibody attachment was also investigated. Antibody fragment immobilization was characterized by fluorescence microscopy using an anti-6X His tag reporter antibody labeled with DyLight 488 for detection. Glycosylated scFvs with or without oxidation of the linked glycan moiety were immobilized on the aminated surfaces at 258C and 378C. As can be seen in Figure 6, the fluorescence intensity increases with increasing temperature for both oxidized (Gly-Oxy) and non-oxidized (Gly) antibody fragments, indicating higher numbers of immobilized scFv molecules. Higher (1.2 times) fluorescence intensity was observed for Gly-Oxy scFvs at 378C than at 258C, while fluorescence was 1.7 times higher in the case of the nonoxidized antibody. Meanwhile immobilization of antibody fragments with an oxidized glycan moiety yielded 1.8 and 1.2 fold higher fluorescence intensities than non-oxidized scFvs at 258C and 378C, respectively. The proposed

IMMOBILIZATION OF RECOMBINANT ANTIBODY FRAGMENTS ON STAINLESS STEEL 316 L

ORIGINAL ARTICLE

FIGURE 5. Influence of reaction temperature on the density of amine groups introduced using three APTS concentration5s in various solvents. Reactions were carried out in ethanol for (a) 15 min and (b) 60 min and in water for (c) 15 min and (d) 60 min. Symbols S1, S2, and S3 represent APTS concentrations of 0.001, 0.004 and 0.04M, respectively.

mechanism of immobilization of the oxidized scFvs on the modified surface, mediated by a covalent interaction between an oxidized glucose monomer in the scFv glycan chain and amine groups introduced into the surface upon silanization with APTS, is presented in Figure 7. To highlight the role of amine groups in antibody fragments bonding to the surface, the results of scFv immobilization on T15 surface were investigated. The attachment of the Gly-Oxy and Gly scFvs to the T15 surface before and after aminosilanization at 378C was visualized by confocal laser scanning microscopy. T15 surfaces before and after aminosilanization were incubated with antibody fragments, which were then detected by the anti-6x His tag DyLight 488-labeled antibody and observed using a confocal laser scanning microscope. As can be seen in Figure 8, adsorption of antibody fragments was greater on surfaces containing introduced amine groups [Fig. 8(b,d)] than on non-aminated T15 surfaces [Fig. 8(a,c)], while adsorption of oxidized antibody fragments on the aminated surface [Fig. 8(d)] was more efficient than nonoxidized fragments [Fig. 8(b)]. Figure 9(a,b) presents images of 2D topography of scFv proteins immobilized on nonaminated [Fig. 9(a)] and aminated [Fig. 9(b)] surface. The antibodies immobilized on nonaminated surface tend to aggregate into higher organized structures. Proteins are uniformly scattered all over the mica surface. The proteins immobilized on aminated surface are homogeneously distributed on the surface. The antibody fragments form mostly monomeric structures and the aggregates are not observed. Materials for in vivo use must not be toxic upon contact with the human body. Therefore, the cell viability on the

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | APR 2016 VOL 104A, ISSUE 4

aminated and nonaminated surfaces was evaluated in vitro using human endothelial progenitor cell line 55.1 (HucPEC55.1) cultured in direct contact with the 316 L, T15, and T15-NH2 materials. As can be seen in Figure 10 the cellular viabilities of the cell lines slightly differ on three tested surfaces. The results indicate that the cells are more vulnerable to cell death when exposed to either 316 L or 316 LNH2, whereas with the 316-T15 the viability remained rather high. The different behaviour of the cells observed on these three surfaces could be explained by the different surface charges of the surfaces. Slightly positive charge of the surface coming from amine groups might influence the cell behaviour. It has been reported that increased toxicity in cell culture can be caused by strong cell interactions.1,2 This

FIGURE 6. Effects of temperature on immobilization of glycosylated scFvs before (“Gly”) and after (“Gly-Oxy”) oxidation of the scFvlinked glycan.

827

FIGURE 7. Schematic of (a) oxidation of glucose monomer in glycan chain linked to the scFv antibody fragment and (b) immobilization of the oxidized scFv on the aminofunctionalized surface via the oxidized glycan.

could indicate that the positive T15-NH2 surface charge tends to attract the cells more and leading to strong interactions with the cells, thus reducing the cell viability when compared to 316 L and T15 surfaces. Further studies of cell viability at various times are required to verify this hypothesis. The obtained results indicate that the viability of EPC cells on the aminated surface increased about 20% in comparison to bare 316 L stainless steel. Such a small difference in change in EPC cells viability can be explained by natural

low toxicity of 316 L stainless steel which is a common material to be used for medical application. DISCUSSION

The overall aim of this study was to develop the method for the immobilization of scFv on stainless steel 316 L material to improve cell viability of the implanted stents. The presented method for surface stent modification was created to enable a strong binding of antibody fragments to the stent

FIGURE 8. Confocal images of scFv immobilized on (a) nonaminated (T15) and (b) aminated surfaces, and oxidized scFv immobilized on (c) nonaminated and (d) aminated surfaces. Immobilization was carried out at 378C for 1 h in 5 mM phosphate buffer containing 100 mM NaCl, pH 7.0.

828

FOERSTE ET AL.

IMMOBILIZATION OF RECOMBINANT ANTIBODY FRAGMENTS ON STAINLESS STEEL 316 L

ORIGINAL ARTICLE

FIGURE 9. AFM images of scFv immobilized on (a) nonaminated (T15) and (b) aminated surfaces. Immobilization was carried out at 378C for 1 h in 5 mM phosphate buffer containing 100 mM NaCl, pH 7.0 on mica surface.

surface. The nucleophilic addition reaction which amine groups attached to the stents surface undergo with the oxidized glycan chain of antibody fragments provides a strong covalent bond between the molecules. This bond is strong enough to withstand both, the force generated during the balloon deployment and the pressure of blood flowing through the artery. To achieve the uniform biocompatible coating of scFv on the stent material surface three tasks were performed. First the optimal conditions for covering of 316 L stainless steel surfaces with amino groups were established. The OH groups necessary for successful aminosilanization were provided to the substrate by coating a surface with titania based coating prior to silanization. The presence of the amino groups on the surface has been proved to exist (Figs. 2–4). It has been showed that using 0.04M solution of APTS in ethanol for 1 h at 608C (Fig. 5) provide the 316 L surface with the densest functional groups coverage. It has been discussed in literature59–61 that APTS molecule can attach to the hydroxyl groups on the titania based surface in two ways, either with Si end or with the amine groups. It has been concluded that if the successful surface silanization (Si end attached to the hydroxyl groups) takes place, the free amine groups should be observed around 399.4 eV. In our XPS spectra of N 1s the pick at 399.2 eV is observed that is assigned to amine groups. Additionally the presence of Si 2p peak at 102.6 eV indicates silane grafting. Based on the obtained results from XPS analysis it can be said that successful silanization with the amine groups on the top of the T15 surface was achieved. The concern of hydrolytic stability of silane layer was limited in this study. Using a heat treatment in the silanization process helps to remove water from the surface and enables the silane molecules to inter-crosslink. Consequently a relatively robust silane layer is created. Although the

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | APR 2016 VOL 104A, ISSUE 4

presence of amine groups can catalyze reaction of degradation of the covalently attached to the substrate aminosilane layers in aqueous media at 408C62 blocking the amine groups by their reaction with antibody fragments provides enough resistance to hydrolysis. The lifespan of stents with modified surface will be further investigated in our future work. The mechanical test we have done (not published yet) demonstrated no changes in the physical properties of the coating during stent expansion. The presence of Triton X helped to reduce brittleness of the coating. The alternative to TiO2 would be plasma treatment of the bare metal. However it is more expensive method and requires special equipment. In the study the influence of oxidation of scFv glycan moieties and the immobilization temperature on antibody

FIGURE 10. Viable counts of human endothelial progenitor cell line 55.1 (HucPEC-55.1) incubated for 24 h on unmodified stainless steel (“316 L”), titania-coated 316 L (“T15”) and aminofunctionalized T15 (“T15-NH2”). Bars represent the mean of four measurements.

829

fragment uniformity and density coverage on the aminated surface were determined next. The results (Fig. 6) suggest that immobilization of the oxidized antibody fragments occurred via temperature-dependent covalent bond formation between free amines on the functionalized surface and aldehyde groups introduced into the antibody fragment glycan chains upon oxidation. Immobilization of non-oxidized antibody fragments most likely occurred via ionic adsorption with the weakly attached antibody fragments easily removed by the washing step, resulting in fewer antibody fragments remaining attached to the surface after immobilization studies. There was a uniform distribution of both antibody fragments on the aminated surface whereas weak, non-uniform fluorescence was observed on the non-modified T15 surface [Fig. 8(a,c)] incubated with the same antibody fragments, demonstrating that the antibody fragments cannot be efficiently immobilized without prior aminosilanization of the T15 surface. Moreover AFM characterization [Fig. 9(a,b)] indicates that the antibody fragments are homogeneously distributed and aggregates are not observed on aminated surface. Homogeneity of scFv antibodies distribution may have significant impact on cell binding and adhesion. The uniform biofunctionalization of the surface may favor its biocompatibility increase because protein self-aggregation tendency is minimized and the antibody fragments may interact with circulating EPCs through exposed receptor sites. Finally, the effect of each stage of chemical modification on cell viability was investigated (Fig. 10). The results proved that neither surface modification of the 316 L substrate resulted in a detectable cytotoxic effect, indicating a satisfactory cell viability of both modified surfaces for in vivo use. Further studies will be performed to prove the specific binding ability of scFv. CONCLUSION

A surface modification method was successfully developed to functionalize the 316 L stainless steel surface for enhanced immobilization of VEGFR2-binding recombinant antibody fragments and potential application in cardiovascular settings. Aminosilanization of discs of titania-coated 316 L stainless steel typical of coronary and peripheral vessel stents was influenced by temperature, time, silane concentration and solvent. Fluorometric evaluation of the density of introduced amine groups by FTIR-ATR spectroscopy allowed optimization of the modified surfaces while fluorescence and confocal microscopy of glycosylated antibody fragments immobilized on the functionalized surfaces revealed four times more non-oxidized and 17 times more oxidized antibody fragments bound on the aminofunctionalized compared to unmodified surface. AFM study indicated homogeneously and aggregates free distribution of scFv on aminated surface. Furthermore, cell viability test indicated the aminofunctionalized and antibody-coated surfaces were not toxic to endothelial progenitor cells, suggesting their potential suitability for use in cardiovascular stents or other biomedical implants.

830

FOERSTE ET AL.

ACKNOWLEDGMENT

The authors thank Balton Z.o.o. for providing 316 L stainless steel discs. REFERENCES 1. Fox KA, Carruthers KF, Dunbar DR, Graham C, Manning JR, Raedt De, Buysschaert H, Lambrechts I, Van de Werf DF. Underestimated and under-recognized: The late consequences of acute coronary syndrome (GRACE UK–Belgian Study). Eur Heart J 2010; 31:2755–2764. 2. Carrozza JP Jr, Kuntz RE, Levine MJ, Pomerantz RM, Fishman RF, Mansour M, Gibson CM, Senerchia CC, Diver DJ, Safian RD. Angiographic clinical outcome of intracoronary stenting: Immediate and long-term results from a large single center experience. J Am Coll Cardiol 1992; 20:328–337. 3. Savage MP, Fischman DL, Schatz RA, Teirstein PS, Leon MB, Baim D, Ellis SG, Topol EJ, Hirshfeld JW, Cleman MW, et al. Long-term angiographic and clinical outcome after implantation of a balloon-expandable stent in the native coronary circulation. J Am Coll Cardiol 1994; 24:1207–1212. 4. Fischman DL, Leon MB, Baim DS, Schatz RA, Savage MP, Penn I, Detre K, Veltri L, Ricci D, Nobuyoshi M, et al. A randomized comparison of coronary-stent placement and balloon angioplasty in the treatment of coronary artery disease. N Engl J Med 1994; 331: 496–501. 5. Hoffmann R, Mintz GS, Dussaillant GR, Popma JJ, Pichard AD, Satler LF, Kent KM, Griffin J, Leon MB. Patterns and mechanisms of in-stent restenosis. A serial intravascular ultrasound study. Circulation 1996; 94:1247–1254. 6. Schwartz RS, Huber KC, Murphy JG, Edwards WD, Camrud AR, Vlietstra RE, Holmes DR. Restenosis and the proportional neointimal response to coronary artery injury: Results i na porcine model. J Am Coll Cardiol 1992; 19:267–274. 7. Kleinedler J, Foley JD, Orchard EA, Dugas TR. Novel nanocomposite stent coatingreleasing resveratrol and quercetin reduces neointimal hyperplasia and promotes reendothelialization. J Control Release 2012; 125:27–33. 8. Hoffmann R, Mintz GS. Coronary in-stent restenosis—Predictors, treatment and prevention. Eur Heart J 2000; 21:1739–1749. 9. Ferrara N, Carvermoore K, Chen H. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature (London) 1996; 380:439–442. € ldicke T, Tesar M, Griesel C, Rohde M, Gro € ne HJ, Waltenberger 10. Bo J, Kollet O, Lapidot T, Yayon A, Weich H. Anti-VEGFR-2 scFvs for cell isolation. Single-chain antibodies recognizing the human vascular endothelial growth factor receptor-2 (VEGFR-2/flk-1) on the surface of primary endothelial cells and preselected CD341 cells from cord blood. Stem Cell 2011; 19:24–36. 11. Leung DW, Cachianes G, Kuang WJ. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989; 246:1306– 1309. 12. Wilting J, Christ B, Weich AW. The effects of growth factors on the day 13 chorioallantoic membrane (CAM): A study of VEGF165 and PDGF-BB. Anat Embryol 1992; 186:251–257. 13. Holmes K, Roberts OL, Thomas AM, Cross MJ. Vascular endothelial growth factor receptor-2: Structure, function, intracellular signaling and therapeutic inhibition. Cell Signal 2007; 19:2003–2012. 14. Veikkola T, Alitalo K. VEGFs, receptors and angiogenesis. Cancer Biol (Berlin) 1999; 9:211–220. 15. Millauer B, Wizigmann-Voss S, Schnurch H, et al. High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 1993; 72:835–846. 16. Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science 1997; 275:964– 967. 17. Peichev M, Naiyer AJ, Pereira D, et al. Expression of VEGFR-2 and AC133 by circulating human CD341 cells identifies a population of functional endothelial precursors. Blood 2000; 95:952– 958. 18. Ohara R, Knappik A, Shimada K, Frisch C, Ylera F, Koga H. Antibodies for proteomic research: Comparison of traditional

IMMOBILIZATION OF RECOMBINANT ANTIBODY FRAGMENTS ON STAINLESS STEEL 316 L

ORIGINAL ARTICLE

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

immunization with recombinant antibody technology. Proteomics 2006; 6:2638–2646. Torrance L, Ziegler A, Pittman H, Paterson M, Toth R, Eggleston I. Oriented immobilization of engineered single-chain antibodies to develop biosensors for virus detection. J Virol Methods 2006; 134:164–170. Markoglou N, Wainer IW. Synthesis and characterization of an immobilized phenylethanolamine N-methyltransferase liquid chromatographic stationary phase. Anal Biochem 2000; 288:83– 88. Shen Z, Stryker GA, Mernaugh RL, Yu L, Yan H, Zeng X. Singlechain fragment variable antibody piezoimmunosensors. Anal Chem 2005; 136:77797277805. Hu X, Spada S, White S, Hudson S, Magner E, Wall JG. Adsorption and activity of a domoic acid binding antibody fragment on mesoporous silicates. J Phys Chem B 2006; 110:18703–18709. € lfl S. Preparation of DNA and proAfanassiev V, Hanemann V, Wo tein micro arrays on glass slides coated with an agarose film. Nucleic Acids Res 2000; 28:66. Charles PT, Goldman ER, Rangasammy JG, Schauer CL, Chen MSTaitt CR. Fabrication characterization of 3D hydrogel microarrays to measure antigenicity and antibody functionality for biosensor application. Biosens Bioelectron 2004; 20:753–764. € kler J, Sobek J, Lehrach H, Cahill DJ. Next genAngenendt P, Glo eration of protein microarray support materials: Evaluation for protein and antibody microarray applications. J Chromatog A 2003; 1009:97–104.  pez-Gallego F, Betancor L, Mateo C, Hidalgo A, Alonso-Morales Lo n JM, Ferna ndez-Lafuente R. Enzyme N, Dellamora-Ortiz G, Guisa stabilization by glutaraldehyde crosslinking of adsorbed proteins on animated supports. J Biotechnol 2005; 119:70–75. Ha TH, Jung SO, Lee JM, Lee KY, Lee Y, Park JS, Chung BH. Oriented immobilization of antibodies with GST-fused multiple Fcspecific B-domains on a gold surface. Anal Chem 2007; 79:546– 556. Mouri M, Ikawa T, Narita M, Hoshino F, Watanabe O. Orientation control of photoimmobilised antibodies on the surface of azobnzene-containing polymers by the introduction of functional groups. Micromol Biosci 2010; 10:612–620. Tan A, Farhatnia Y, Goh D, Achala de Mel NG, Lim J, Teoh S-H, Malkovskiy AV, Chawla R, Rajadas J, Cousins BG, Hamblin MR, Alavijeh MS, Seifalian AM. Surface modification of a polyhedral oligomeric silsesquioxane poly(carbonate-urea) urethane (POSSPCU) nanocomposite polymer as a stent coating for enhanced capture of endothelial progenitor cells. Biointerphases 2013; 8:23 Hu X, Hortig€ uela MJ, Robin S, Lin H, Li Y, Moran AP, Wang W, Wall JG. Covalent and oriented immobilization of scFv antibody fragments via an engineered glycan moiety. Biomacromolecules 2013; 14:153–159. Martin HJ, Schulz KH, Bumgardner JD, walters KB. XPS study on the use of 3- aminopropyltriethoxysilane to bond chitosan to a titanium surface. Langmuir 2007; 23:6645–6651. 2007. Weetall HH. preparation of immobilized proteins covalently coupled through silane coupling agents to inorganic supports. Appl Biochem Biotechnol 1993; 41:157–188. Nanci A, Wuest JD, Peru L, Brunet P, Sharma V, Zalzal S, McKee MD. Chemical modification of titanium surfaces for covalent attachment of biological molecules. J Biomed Mater Res 1998; 40: 324–335. thivier C, Zanna S, Minier M, Salmain M, Yacoubi N, Barbes L, Me Pradier CM. Covalent immobilization of lysozyme on stainless steel. interface spectroscopic characterization and measurement of enzymatic activity. Langmuir 2005; 21:5957–5965. Aoki J, Serruys PW, Beusekom H, Ong ATL, McFadden EP, Sianos G, Giessen WJ, Regar E, Feyter PJ, Davis HR, Rowland S, Kutryk MJB. Endothelial progenitor cell capture by stents coated with antibody against CD34. J Am Coll Cardiol 2005;45: 1574–1579.  ska M, Sobieszczan  ska B, Ulatowska-Jar_za A, Hołowacz Kopaczyn  ska I, Buzalewicz I, Wasyluk Ł, Tofail SAM, Biały D, Wawrzyn MPodbielska H. Photoactivated titaniabased nanomaterials for potential application as cardiovascular stent coatings. Biocybernetics Biomed Eng 2014; 34:189–197.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | APR 2016 VOL 104A, ISSUE 4

37. Fahrenfort J. Attenuated total reflection: A new principle for the production of useful infra-red reflection spectra of organic compounds. Spectrochim Acta 1961; 17:698–709.  ska-Janus S, Wilk I. Studies of the kinetics 38. Paszkowski M, Olsztyn of lithium grease microstructure regeneration by means of dynamic oscillatory rheological tests and FTIR-ATR spectroscopy. Tribol Lett 2014; 56:107–117.  ska-Janus S, Komorowska M. Conformational changes of 39. Olsztyn L-phenylalanine induced by Near Infrared radiation. ATR-FTIR studies. Struct Chem 2012; 23:1399–1407. 40. Pansri P, Jaruseranee N, Rangnoi K, Kristensen P, Yamabhai M. A compact phage display human scFv library for selection of antibodies to a wide variety of antigens. BMC Biotechnol 2009; 9:6 41. Cunningham C, Srivastava A, Collin E, Grad S, Alini M, Pandit A, Wall JG. Isolation and characterization of a recombinant antibody fragment that binds NCAM1-expressing intervertebral disc cells. Plos One 2013; 8:e83678 42. Wacker M, Linton D, Hitchen PG, Nita-Lazar M, Haslam SM, North SJ, Panico M, Morris HR, Dell A, Wren BW, Aebi M. Nlinked glycosylation in Campylobacter jejuni and its functional transfer into E. coli. Science 2002; 98:1790–1793. 43. Hu X, O’Hara L, White S, Magner E, Kane M, Wall JG. Optimization of production of a domoic acid-binding scFv antibody fragment in E. coli using molecular chaperones and functional immobilization on a mesoporous silicate support. Protein Expr Purif 2007; 52:194–201. 44. Studier FW. Protein production by auto-induction in high density shaking cultures. Protein Expr Purif 2005; 41:207–234. € ela MJ, Wall JG. A novel antibody engineering approach 45. Hortigu for improved detection of marine neurotoxins. Marine Drugs 2013; 11:881–895. 46. Lang X, Ji H, Chen C, Ma W, Zhao J. Selective formation of imines by aerobic photocatalytic oxidation of amines on TiO2. Angew Chem Int Ed 2011; 50:3934–3937. 47. Klitgaard SK, Egeblad K, Mentzel UV, Popov AG, Jensen T, Taarning E, Nielsen IS, Christensen CH. Oxidations of amines with molecular oxygen using bifunctional gold–titania catalysts. Green Chem 2008; 10:419–423. 48. Wood DL, Rabinovich EM, Johnson DW, Mac-Chesney JB, Voge M. Preparation of high-silica glasses from colloidal gels. Infrared spectrophotometric studies. J Am Ceram Soc 1983; 66:693– 699. 49. Urlaub R, Posset U, Thull R. FT-IR spectroscopic investigations on sol-gel-derived coatings from acid-modified titanium alkoxides. J Non-Cryst Solids 2000; 265:276–284. 50. Vasconcelos DCL, Costa VC, Nunes EHM, Sabioni ACS, Gasparon M, Vasconcelos WL. Infrared spectroscopy of titania sol-gel coatings on 316L stainless steel. Mater Sci Appl 2011; 2:1375–1382. 51. Pilkenton S, Hwang SJ, Raftery D. Ethanol photocatalysis on TiO2-coated optical microfiber, supported monolayer, and powdered catalysts: An in situ NMR Study. J Phys Chem B 1999; 103: 11152–11160. 52. Iseda K. Oxygen effect on photocatalytic reaction of ethanol over some titanium dioxide photocatalysts. Bull Chem Soc Jpn 1991; 64:1160–1166. 53. Burgos M, Langlet M. The sol-gel transformation of TIPT coatings: A FTIR study. Thin Solid Films 1999; 349:19–23. 54. Castellano FN, Stipkala JM, Friedman LA, Meyer GL. Spectroscopic and excited-state properties of titanium-dioxide gels. Chem Mater 1994; 6:2123–2129. 55. Kumar DA, Shyla JM, Xavier FP. Synthesis and characterization of TiO2/SiO2 nano composites for solar cell applications. Appl Nanosci 2012; 2:429–436. 56. Izutsu H, Nair PK, Maeda K, Kiyozumi Y, Mizukami F. “Structure and properties of TiO2-SiO2 prepared by sol-gel method in the presence of tartaric acid. Mater Res Bull 1997; 32:1303–1311. 57. Phillippi CM, Lyon SR. Longitudinal-optical phonons in TiO2 (Rutile) thin-film spectra. Phys Rev B 1971; 3:2086–2087. 58. Tan G, Zhang L, Ning C, Liu X, Liao j. Preparation and characterization of APTES films on modification titanium by SAMs. Thin Solid Films 2011; 519:4997–5001.  ndez-Caldero  nb MC, Angel 59. Rodrıguez-Canoa A, Cintasa P, Ferna lezPacha-Olivenz M, Crespoe L, Salda L, Vilaboa N, Gonza

831

Martınd ML, Babiano R. Controlled silanization– amination reactions on the Ti6Al4V surface for biomedical applications. Colloids Surf B Biointerfaces 2013; 106:248–257. 60. Song Y-Y, Hildebrand H, Schmuki P. Optimized monolayer grafting of 3-aminopropyltriethoxysilane onto amorphous, anatase and rutile TiO2. Surface Sci 2010; 604:346–353. 61. Kallury KMR, Macdonald PM, Thompson M. Effect of surface water and base catalysis on the silanization of silica by (aminopropyl)alkoxysilanes studied by X-ray photoelectron

832

FOERSTE ET AL.

spectroscopy and13C cross-polarization/magic angle spinning nuclear magnetic resonance. Langmuir 1994; 10:492 62. Arranz A, Palacio C, Garcia-Fresnadillo D, Orellana G, Navarrom A, Munoz E. Influence of surface hydroxylation on 3-aminopropyltriethoxysilane growth mode during chemical functionalization of gan surfaces: an angle-resolved x-ray photoelectron spectroscopy study. Langmuir 2008; 24:8667 63. Smith EA, Chen W. How to prevent the loss of surface functionality derived from aminosilanes. Langmuir 2008; 24:12405–12409.

IMMOBILIZATION OF RECOMBINANT ANTIBODY FRAGMENTS ON STAINLESS STEEL 316 L