58
Bioconjugate Chem. 2011, 22, 58–66
Chemoselective Modification of Turnip Yellow Mosaic Virus by Cu(I) Catalyzed Azide-Alkyne 1,3-Dipolar Cycloaddition Reaction and Its Application in Cell Binding Qingbing Zeng,*,†,‡ Sharmistha Saha,‡ L. Andrew Lee,‡ Hannah Barnhill,‡ Jerry Oxsher,‡ Theo Dreher,§ and Qian Wang*,‡ Biomaterial Research Center, School of Pharmaceutical Sciences, Southern Medical University, 1023 Southern Shatai Road, Guangdong, GD, 510515, China, Department of Chemistry and Biochemistry and Nanocenter, University of South Carolina, 631 Sumter Street, Columbia, South Carolina 29208, United States, and Department of Microbiology, Oregon State University, Corvallis, Oregon 97331, United States. Received August 1, 2010; Revised Manuscript Received November 15, 2010
Turnip yellow mosaic virus (TYMV) is an icosahedral plant virus with a diameter of 28-30 nm that can be isolated in gram quantities from turnip or Chinese cabbage inexpensively. In this study, TYMV combined with spatially addressable surface chemistries was selected as a prototype bionanoparticle for modulating patterns of cell adhesion, morphology, and proliferation. We exploited the chemical reactivity of TYMV using the mild conditions of Cu(I) catalyzed azide-alkyne cycloaddition (CuAAC) reaction, the best example of “click” chemistry. Oligo-ethylene glycol (OEG) short chain, coumarintriazole, and RGD-containing peptide were grafted on the surface of TYMV via carbodiimide activation and CuAAC reaction. The bioconjugation to intact viral particles was confirmed by MS, TEM, FPLC, and SDS-PAGE with fluorescence visualization analysis. Therefore, this method is a generally useful means of incorporating various types of functionalities onto the TYMV surface. Further studies were done to learn the behavior of NIH-3T3 fibroblast cells on the modified or unmodified TYMV surfaces. OEG-modified TYMV surfaces retarded cell attachment and growth, while cell adhesion, spreading, and proliferation were dramatically enhanced on RGD-modified TYMV surfaces. Compared with RGD immobilized 3-aminopropyltriethoxysilane-coated glass surface, the cells are more ready to spread fully and proliferate on TYMV-RGD coated surface, which thus provides a more cell-friendly environment with nanometer-scale surface features. This illustrates the potential application of plant virus based materials in tissue engineering, drug delivery, and biosensing.
INTRODUCTION Recently, there has been increasing focus in biomaterials science on producing nanometer-sized particles for a variety of biomedical applications like drug delivery, tissue engineering, and biosensing (1-4). A particular desire of modern biomedical engineering is to integrate biomaterials design with new insights emerging from studies of cell-matrix interactions and cellular signaling processes (5). Microfabrication technology, a new and powerful tool in the manufacturing of various types of biomedical microdevices such as those developed for neural stimulation, implantable encapsulation, sensing, and drug/gene delivery, is making a considerable impact in recent biotechnological research. In designing these microdevices, the control of initial cell attachment to the substrate via integrin binding and subsequent cell proliferation is a critical parameter (6, 7), which can be modulated by surface immobilization of poly(ethylene glycol) (PEG), a water-soluble polymer that inhibits cell binding (6), or oligopeptide such as RGD, IKVAV, YIGSR, and RNIAEIIKDI to promote cell binding (8). Several techniques, including physical adsorption, graft polymerization, and chemical coupling have been applied to modify substrate surfaces. However, most of the substrates used to develop such microde* Corresponding author. Qingbing Zeng: Tel: +086-020-62789462, Fax: +086-020-62789462, E-mail:
[email protected]. Qian Wang: Tel: ++001-803-777-2680, Fax: + +001-803-777-9521, E-mail:
[email protected]. † Southern Medical University. ‡ University of South Carolina. § Oregon State University.
vices are polymer- or silicon-based, and the rigidity and the thickness of the substrate limit their viability in applications where nanometer-scale dimensional requirements must be enforced, such as in the field of biomedical microdevices (6-8). Virus particles, as self-assembled and uniformly ordered nanostructures with well-defined geometries, can be ideal polyvalent scaffolds for ligand display. With the goal of introducing new functions to these structures, numerous studies have been performed to tailor the interior or exterior surfaces of viruses with sensing units (9-15), drug (16), polymers (17-20), carbohydrates (21, 22), oligonucleotides (23, 24), peptides, and proteins (14, 18, 24). Viral nanoparticles or virus-like particles, including cowpea mosaic virus (CPMV) (9-11, 24-27), cowpea chlorotic mottle virus (28), tobacco mosaic virus (TMV) (18), bacteriophage MS2 (14, 29-31), and polyoma virus (32), have also been employed for surface functionalization at the atomic level by both chemical and genomic manipulation. Such strategies have also been extended to a broad array of various viruses including bacteriophage P22, Sulfolobus islandicus rodshaped virus 2, and Hibiscus chlorotic ringspot virus (16, 33, 34). Recently, it was shown in the Wang lab that plant viruses can offer a special nanoplatform, which can regulate the differentiation of bone marrow stromal cells (BMSCs) into osteoblasts. These data indicated that the gene expression, cell behavior, and levels of osteo-differentiation were significantly affected by the presence of turnip yellow mosaic virus (TYMV) nanoparticles coated on 2D substrate as compared to the cells grown under standard conditions (35). Although research is ongoing to establish how the nanotopography of the ordered viral capsid enhances cellular differentiation, we are also
10.1021/bc100351n 2011 American Chemical Society Published on Web 12/20/2010
Controlled Cell Binding via Modification of TYMV
Bioconjugate Chem., Vol. 22, No. 1, 2011 59
Figure 1. (Left) Structure of TYMV capsid with an asymmetric unit cut out. Protein subunits are colored as blue (chain A), red (chain B), and green (chain C) (38). (Right) Enlarged ribbon diagram of TYMV asymmetric unit, which includes three subunits. The five and sixfold axes are labeled where they form on the intact capsid (37).
interested in finding out whether surface modification of TYMV can affect the cellular response by altering the initial adhesion of the cells to the substrate. Thus, the bioconjugation of the TYMV viral capsid with various cell recognition motifs would give us an edge not only to modulate but also control various cellular behaviors. TYMV is a nonenveloped plant virus with a diameter of 28-30 nm comprising a single-stranded RNA of 1.9 × 106 Da and 180 chemically identical protein subunits of 20 KDa. The subunits arrange into 60 trimeric asymmetric units assembled in T ) 3 icosahedral symmetry (Figure 1) (36-39). The structure of TYMV has been extensively studied by electron microscopy (40) and X-ray crystallographic analysis (41-43). It can be isolated from either turnip or Chinese cabbage in gram quantities. Previous study in the Wang lab has demonstrated that the surface amino groups and carboxyl groups of TYMV can be selectively and orthogonally modified via amidation reaction, and TYMV has some advantageous properties compared with other plant viruses (12, 13). In this paper, Cu(I)catalyzed Huisgen azide-alkyne 1,3-dipolar cycloaddition (CuAAC) reaction, a type of “click” chemistry (44), that has been developed and utilized broadly in recent years (24, 25, 45-50), has been employed to modify TYMV. The CuAAC reaction is a versatile and modular process and provides a method for coupling a wide range of molecules in a regiospecific fashion under relatively mild reaction conditions with little byproduct. Finn and co-workers have reported that CPMV can be efficiently conjugated with both small molecules and macromolecules via CuAAC reaction (22, 24, 25, 45). The Wang group has recently developed a tandem CuAAC reaction to address tyrosine residues of TMV chemoselectively (51). In this paper, we report our effort to extend the CuAAC reaction to tailor TYMV and the corresponding cell binding studies using TYMV decorated with relevant motifs.
EXPERIMENTAL PROCEDURES General. Most bioreagents were purchased from Bio-Rad or Fisher and used without further purification. 3-Aminopropyltriethoxysilane-coated glass slides (APTES slides) were from Labscientific. Azide-terminated GRGDS oligopeptide 3 was prepared by standard solid-phase peptide synthesis (49). General desalting and removal of other small molecules from protein samples were achieved using Bio-Spin disposable chromatog-
raphy centrifuge columns (Bio-Rad). Ultracentrifugation was performed at the indicated rpm values using a Beckman Optima L-90K ultracentrifuge equipped with either SW41 or 50.2 Ti rotors. TEM analyses were carried out by depositing 20 µL aliquots of each sample at a concentration of 0.1-0.3 mg/mL onto 100 mesh carbon-coated copper grids for 2 min followed by rinsing with ddH2O. The grids were then stained with 20 µL of 2% uranyl acetate and viewed with a Hitachi H-8000 TEM electron microscope at 200 kV accelerating voltage. Fluorescent emission spectra were recorded on a FP-6200 spectrofluorometer. FPLC analyses were performed on an AKTA Explorer (GE Biotech) using a Superose-6 size exclusion column. Potassium phosphate buffer (0.05 M, pH 7.0) with 0.15 M NaCl was used as eluent, and the intact virions show retention volume of approximately 9.5 mL at an elution rate of 0.5 mL/ min. SDS-PAGE analysis was carried out in a Bio-Rad MiniPROTEAN 3 gel electrophoresis cell. The proteins were resolved on a 15% polyacrylamide gel at 120 V for 2.5 h. Fluorescently labeled TYMV was visualized with UVP Epi Chemi II imager before Coomassie blue staining. Matrix assisted laser desorption-ionization time of flight (MALDI-TOF) mass spectra were obtained on a Bruker Ultraflex II mass spectrometer (Berman Germany) with MS-grade sinapinic acid in 70% acetonitrile and 0.1% TFA as the matrix. X-ray photoelectron spectroscopic (XPS) analysis was performed on Thermo ESCA LAB 250 XPS instrument. Three spots per sample were analyzed with XPS; the average of the values was used for reasoning. An X-ray beam with an 800 µm diameter was used in the analysis. The zeta potential of proteins was measured by Zetasizer Nano ZS (Malvern Instrument) using buffer adjusted to ionic strength of 40 mM. A multimode AFM Nanoscope IIIa (Digital Instruments, Santa Barbara, CA) was used to observe the nanoparticles in situ in tapping mode using OTESPA silicon cantilevers (Veeco Instruments, Dourdan, France). Water contact angles for the modified coating were recorded on OCA 40 Microsystem, Dataphysics. Isolation of TYMV. Chinese cabbage was grown for 3 weeks after which it was inoculated with TYMV. After 1 week, leaves started showing mosaic symptoms. After another 2 weeks, the leaves were picked and purified for virus. Infected leaves were blended with 3× volume of 10 mM potassium phosphate buffer (pH 7.0) and 0.1% β-mercaptoethanol. The mixture was filtered, and the filtrate was subjected to centrifugation to remove bulk
60 Bioconjugate Chem., Vol. 22, No. 1, 2011
Zeng et al.
Scheme 1. Bioconjugation on TYMV Using Cu(I) Catalyzed Alkyne-Azide Cycloaddition Reactiona
a Conditions: (i) EDC, Sulfo-NHS, K-phos buffer pH 7.8, 4 °C, 24 h; (ii) CuSO4, TCEP, BCDS, 10 mM Tris Buffer pH 7.8/DMSO ) 4/1, 4 °C, 24 h.
plant material. The supernatant was collected and clarified by adding an equal volume of CHCl3/1-butanol (v/v 1:1). The aqueous layer was collected, and the virus was precipitated with 8% PEG 8K and 0.2 M NaCl. The pellet was centrifuged and resuspended in buffer. In general, the virus was stored in 10 mM potassium phosphate buffer (pH 7.0) at a concentration of 10 mg/mL at 4 °C and was stable for months. TYMV Bioconjugation Protocol. TYMV modification with amine was carried out by incubating TYMV at 1 mg/mL with 5 mM propargyl-amine or 3-{2-[2-(2-methoxyethoxy)ethoxy]ethoxy}propylamine aided with 1-ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride (EDC) and N-hydroxysulfosuccinimide sodium salt (sulfo-NHS) in potassium phosphate buffer (10 mM, pH 7.8). The product was isolated by sucrose gradient sedimentation, ultracentrifugation pelleting, and resuspension in 0.1 M Tris buffer (pH 7.8). The CuAAC conjugation after propargyl amine modification was carried out by incubating TYMV-alkyne at 1 mg/ mL with 3-azido-7-hydroxycoumarin 2, azide-terminated RGD containing oligopeptides 3, in the presence of copper sulfate (1 mM), tris(2-carboxyethyl)phosphine hydrochloride (TCEP, 2 mM), and ligand bathocuproindisulfonic acid disodium salt (BCDS, 3 mM) in 20% DMSO and 80% buffer for 24 h at 4 °C as shown in Scheme 1. The products were purified by two successive series of sucrose gradient sedimentation, ultracentrifugation pelleting, and resuspension in 10 mM potassium phosphate buffer (pH 7.0). Reactions were analyzed by MALDI-TOF MS, TEM, FPLC, SDSPAGE, and fluorescence spectroscopy. Immobilization of TYMV, TYMV-OEG, and TYMVRGD on APTES Slides. Cage-shaped proteins were adsorbed onto APTES slides using solutions of TYMV, TYMV-OEG, or TYMV-RGD in pure water with a protein concentration of 0.5 mg/mL. 500 µL virus solutions were dropped onto the substrates (10 × 10 mm2) so that they fully covered the surface, and the samples were set at room temperature in a biosafe hood for 10 min. The samples were washed in pure water and dried at room temperature. The resultant systems were denoted as APTES-TYMV, APTES-TYMV-OEG, and APTES-TYMVRGD, respectively. The adsorption behavior of virus particles onto the APTES slides were characterized by atomic force microscope (AFM) and water contact angles. Immobilization of RGD Peptide on APTES Slides. The RGD peptide immobilization was achieved on APTES slides via CuAAC reaction by using a propiolic acid linker group as shown in Scheme 2 through the following two steps: (1) APTES
slides were immersed in 10 mM potassium phosphate buffer (pH 7) with 0.3% (w/v) propiolic acid 5, EDC, and sulfo-NHS at room temperature for 12 h, washed with the buffer three times after the reaction. (2) The slides treated with 5 treated were put into 3 containing (150 µg/mL) solution (20% DMSO and 80% 10 mM, pH 7.8 Tris-buffer), then CuSO4 (1 mM) and NaAsc (2 mM) were added and the mixture was incubated at room temperature for 12 h, after that the slides were rinsed with 10 mM potassium phosphate buffer (pH 7.0) three times. The resultant system was denoted as APTES-RGD. Reaction were monitored by XPS analysis and water contact angles. Cell Culture. NIH 3T3 fibroblasts were cultured in complete Dulbecco’s Modified Eagles Medium (DMEM) supplied with 10% fetal calf serum, 1 mM sodium pyruvate, 1× antibiotics (penicillin-streptomycin 100× stock from Hyclone), 4 mM L-glutamine, at 37 °C in a humidified 5% CO2 incubator. Monolayers of fibroblasts in their growth phase (ca. 90% confluence) were peeled from dishes with 0.05% trypsin-EDTA in HBSS (Fisher) for 2 min, and resuspended in the complete medium for subsequent cell adhesion studies. APTES-TYMV, APTES-TYMV-OEG, and APTES-TYMVRGD slides (10 × 10 mm2) were put into 12 well plates; 2 mL of NIH3T3 cell suspension in complete DMEM at a density of 2 × 104 cells/mL was added to each well of the 12-well plate, and incubated at 37 °C in a humidified 5% CO2 incubator. APTES slides and APTES-RGD slides were used as control groups. Cell adhesion, spreading, and proliferation on precoated substrates was examined at 5 and 24 h postseeding. Nonadherent cells on slides were removed by washing with PBS, and adherent cells were fixed using formaldehyde and stained with Giemsa stain. Photomicrographs of the cytoplasmicstained cells were obtained (100× total magnification using an Olympus CKX41 culture microscope with a CCD camera). The INFINITY analysis software (Lumenera corporation, Ottawa, Canada) was used to classify the cell morphology into three categories: round, partially spread, or fully spread. Cells with surface area of 200 µm2 were classified as fully spread and had multiple cytoplasmic extensions in different directions.
Controlled Cell Binding via Modification of TYMV
Bioconjugate Chem., Vol. 22, No. 1, 2011 61
Scheme 2. Immobilization of RGD Peptide to Silane Surface via CuAAC Reaction
As previously described by Zhang and Davis et al (6, 7), the change in the number of attached cells on different substrates was also monitored. After designated incubation time, the sample substrates were transferred to empty plate wells and rinsed twice with 2 mL PBS to remove unattached cells. Substrates were then incubated with 0.2 mL of 0.05% trypsin-EDTA to remove attached cells. Next, trypsin was neutralized with 0.3 mL DMEM+10% FBS. Cells detached from the substrate were counted with a hemacytometer.
Statistical analysis was performed using a two-tailed unpaired Student’s t test. Values of p < 0.05 were considered to be significant.
RESULTS AND DISCUSSION Introduction of Alkyne Group and Oligo-Ethylene Glycol Chain on TYMV. From structural analysis of TYMV, we found that the reactive lysine is highly buried inside (12, 13). Therefore, the more exposed carboxyl groups on the TYMV
Figure 2. (a) MALDI-TOF MS of whole subunit of wt-TYMV (black line, wt-TYMV peak ) 20 176 m/z, peak at 20 390 m/z is a matrix adduct to wt-TYMV), TYMV-alkyne (blue line, the matrix adduct peak was merged with acetylene adduction peaks, the difference in mass was consistent to the attachment of acetylene, i.e., 37 Da), TYMV-OEG (red line, modified peak ) 20 382 m/z; the peak at 20 563 m/z can be either a matrix adduct or the dual modification of 4, which has a mass of 213 Da), TYMV-coumarin (green line, modified peak ) 20412 m/z, addition of 2, i.e., 205 Da; the peak at 20 209 m/z belongs to TYMV-alkyne), TYMV-RGD (purple line, modified peak ) 20867 m/z, addition of 3, i.e., 667 Da; the peak at 21 520 m/z indicates the dual modification). (b,c) TEM image of TYMV-OEG and TYMV-RGD. (d) Size exclusion FPLC analysis: intact wt-TYMV, TYMV-OEG, and TYMV-RGD particles elute at 9.5 mL.
62 Bioconjugate Chem., Vol. 22, No. 1, 2011
Zeng et al.
Figure 3. (a) Size-exclusion chromatogram of TYMV-coumarin showed a single intact peak at 9.5 mL with strong absorption at 340 nm, indicating the integrity of the particles and the coumarin dye attachment. (b) Fluorescent emission spectra (λexc ) 340 nm) of TYMV-coumarin (black line) and the control mixture of TYMV-alkyne and coumarin without reaction (red line). (c) Fluorescent quantification analysis of covalently attached coumarin to TYMV as a function of increasing concentration of reagent 3-azido-7-hydroxycoumarin. (d) SDS-PAGE of coumarin-modified TYMV visualized under UV irradiation (right) or upon Coomassie blue staining (left).
exterior surface were chosen for further derivatization in this study. Many studies have shown that the carboxyl residues of various viral capsids are amenable to reaction with primary amines under activation (15, 18, 52). The reactivity of TYMV carboxyl groups was tested with fluorescein amines and biotin amine (13) under activation with EDC and sulfo-NHS, which revealed that approximately 90 to 120 carboxylic groups could be modified in high yields. To explore the potential of introducing multifunctional group on the surface of TYMV, the viral capsids were labeled with acetylene or oligo-ethylene glycol chain at surface-exposed carboxylic groups (Scheme 1). In all cases, the reaction yield (the percent of virus recovered after purification of protein away from small molecules) and purity (intact virus particle vs disassembled viral protein) was high. More than 90% of the protein was recovered in each case, and the size-exclusion FPLC and TEM analysis indicated that more than 95% of the virions were intact particles (Figure 2). A distinct modification signal in the MALDI-TOF MS spectra confirms the attachment of alkyne on TYMV-alkyne (Figure 2a, blue line). Encouraged by this result, the CuAAC reaction between TYMV-alkyne and azides was explored; data supporting the CuAAC reaction on TYMV capsid also proves the success of alkyne riveting on TYMV-alkyne (see discussion below). The attachment of oligo-ethylene glycol chain to the viral particle is also monitored by MALDI-TOF MS, as can be seen in Figure 2a, red line. The peak at 20 382 m/z corresponds to OEG riveting protein subunits. The peak at 20 563 m/z can be either a matrix adduct or the dual modification of OEG. CuAAC Reaction on TYMV-Alkyne Capsid. The acetylene functional group on the TYMV-alkyne surface was further modified by the CuAAC reaction to attach fluorogenic or peptide motif via triazole connection (Scheme 1). Initial attempts using CuSO4-ascorbic acid as catalytic system were unsuccessful; TYMV either aggregated or denatured in the presence of these reagents, which may be attributed to the initial oxidation product of ascorbate, dehydroascorbate, acting as a potent electrophile.
Additionally, ascorbate can also be hydrolyzed to form reactive aldehydes such as 2,3-diketogulonate and presumably glyoxal. These species can make connection with arginine, N-terminal cysteine, and lysine side-chains and lead to protein adduct formation, cross-linking, and precipitation (45). Thus, an alternative protocol developed by Finn and co-workers (24, 25) was adopted, in which CuSO4, TCEP, and BCDS was employed as Cu(I) source. 3-Azido-7-hydroxycoumarin 2 was first employed as the azido counterpart in the CuAAC reaction, which could be easily monitored by the increase of fluorescence at 475 nm upon formation of triazole ring. A solution of TYMV-alkyne and 2 was incubated for 24 h in the presence of TCEP, BCDS, and a catalytic amount of CuSO4. Unbound 2 showed no fluorescence, whereas the purified and diluted cycloaddition product TYMVcoumarin produced a strong detectable fluorescent signal at 475 nm upon excitation at 340 nm (Figure 3b, black line) (53, 54). A control sample of a mixture of TYMV-alkyne and 2 without Cu(I) as a catalyst showed little fluorescence (Figure 3b, red line). Quantification of the fluorescent signal arising from the newly formed cycloaddition product revealed that the number of coumarintriazoles attached per TYMV particle was a function of increasing concentration of 2 (Figure 3c), the curve reaching a plateau at approximately 120 dye molecules per particle. The TYMV particles were stable to the click reaction conditions and remained in the assembled state as shown by FPLC; the coumarin absorbance peak at 340 nm coeluted precisely with the virus particle (Figure 3a). TEM also confirmed the integrity of the particle after two-step modification (data not shown). The TYMV-coumarin sample was also analyzed using SDS-PAGE with fluorescence visualization of the attached coumarin, followed by Coomassie staining; the result revealed one dyelabeled band corresponding to the subunit of TYMV, indicating that the subunits of the virus were covalently attached to 2 (Figure 3d). The coupling of 2 onto the TYMV-alkyne surface
Controlled Cell Binding via Modification of TYMV
Figure 4. Topographical AFM micrographs of TYMV-RGD coated APTES substrate surface 3 × 3 µm2.
with high reaction efficiency was also confirmed by MALDITOF MS (Figure 2a, green line). Taking advantage of the high coupling efficiency of the click reaction, azide-terminated GRGDS oligopeptide 3 was grafted onto acetylene-functionalized TYMV with the same procedure as described for 2. The success of this reaction was confirmed by MALDI-TOF MS analysis of viral capsid subunit showing an increase in the molecular weight that corresponds to the attachment of 3 (Figure 2a, purple line). The presence of the peak at m/z ) 20 867 corresponds to the addition of single 3 on the TYMV capsid subunit. There is also some sign of dual decoration with the oligopeptide moiety on TYMV capsid subunit as can be seen from the m/z ) 21 520 small peak in the MALDI-TOF MS spectra. The integrity of TYMV-RGD capsids was confirmed by FPLC and TEM analysis (Figure 2c,d. Immobilization of TYMV, TYMV-OEG, TYMV-RGD, and RDG on APTES Slides. Virus particles need to be immobilized onto 2D surfaces prior to cell binding studies. The isoelectric point of TYMV is about 3.7. At around neutral pH, the zeta potential of TYMV and TYMV-OEG was approximately -18 mV, while the zeta potential of TYMV-RGD was -20 mV. The slight shift in the negative direction can be attributed to the negative charges on aspartic acid residues in the oligopeptide moiety. This indicates that TYMV, TYMVOEG, and TYMV-RGD display negative surface charges in neutral solutions. Therefore, APTES slides were used as the underlying substrate (55). APTES is the positively charged molecule which holds negatively charged virus particles on the slides via electrostatic interactions at around neutral pH. After drop coating and setting according to the conditions described in experimental procedures, TYMV, TYMV-OEG, and TYMVRGD particles anchored strongly on the APTES substrate due to electrostatic interactions. APTES surfaces decorated with TYMV, TYMV-OEG, and TYMV-RGD were imaged by tapping-mode AFM. The AFM micrographs of the TYMV-RGD adsorption on APTES substrate surface is shown in Figure 4. APTES-TYMV and APTES-TYMV-OEG surfaces yielded similar results. The virus coating height was observed to be around 30 nm, and the adsorption densities were around 1.2 × 1011 cm-2, about 85% of the highest theoretical density, which indicate the 10 min drop-coating method resulted in the formation of a monolayer of TYMV on the APTES surface. Tracing of the surface as well as the topographical view indicates that the majority of the virus particles were present as isolated units with very little aggregation. As a control, the GRGDS peptide was also immobilized on a silane surface using a propiolic acid linker via CuAAC reaction
Bioconjugate Chem., Vol. 22, No. 1, 2011 63
as shown in Scheme 2. Theoretically, if all APTES molecules were covalently bound to the RGD peptide there would be six nitrogen atoms for every silicon atom on the APTES slide surface. Following the reaction conditions described in the Experimental Procedures, we found that the N/Si ratio is 4.9 according to XPS analysis result, which is about 80% of the theoretical value. With higher amounts of 5 and 3 in the reaction, the XPS data indicate no increase in the N/Si ratio. After the CuAAC reaction, the oligopetide moiety anchor yield increases from 26% as reported by Davis et al. (7) using maleimide attachment to 80% as described above in our study. “Click chemistry” strategy yielded a promising result and was consistent with what we expected. Siloxane in APTES has low surface energy and shows hydrophobicity, the water contact angle of APTES slides being 76.2°. After RGD modification, the water contact angle of RGD immobilized APTES slides decreased to 49.1° due to hydrophilic amino and carboxyl groups present in RGD. After TYMV, TYMV-OEG, or TYMV-RGD coating, the water contact angle of APTES slides decreased to
