456
Biomacromolecules 2008, 9, 456–462
Assembly of Multilayer Arrays of Viral Nanoparticles via Biospecific Recognition: A Quartz Crystal Microbalance with Dissipation Monitoring Study Nicole F. Steinmetz,*,†,4 Eva Bock,‡,§ Ralf P. Richter,‡,§,⊥ Joachim P. Spatz,‡,§ George P. Lomonossoff,† and David J. Evans† Department of Biological Chemistry, John Innes Centre, Norwich Research Park, Norwich, NR4 7UH, United Kingdom, Department of New Materials and Biosystems, Max-Planck Institute for Metals Research, Heisenbergstrasse 3, 70569 Stuttgart, Germany, and Department of Biophysical Chemistry, University of Heidelberg, Im Neuenheimer Feld 253, 69120 Heidelberg, Germany Received July 18, 2007; Revised Manuscript Received November 13, 2007
The development of multilayered thin film assemblies containing (bio)molecules is driven by the need to miniaturize sensors, reactors, and biochips. Viral nanoparticles (VNPs) have become popular nanobuilding blocks for material fabrication, and our research has focused on the well-characterized plant virus Cowpea mosaic virus (CPMV). In a previous study, we have reported the construction of multilayer VNP assemblies. Here we extend these studies by providing further details on the formation and properties of arrays that are made by the alternating deposition of biotinylated CPMV particles and streptavidin molecules. Array formation was followed in real time by a quartz crystal microbalance with dissipation monitoring. Our data provide indications that multiple interactions between biotin and streptavidin not only promote the assembly of a multilayered structure but also generate cross-links within each layer of CPMV particles. The degree of intralayer and interlayer cross-linking and hence the mechanical properties and order of the array can be modulated by the grafting density and spacer length of the biotin moieties on the CPMV particles.
Introduction The development of multilayered thin film assemblies containing (bio)molecules is of importance in the area of biomaterials science and technology. It is driven by the need to miniaturize sensors, reactors, and biochips. Multilayered arrays of functional (bio)molecules provide a higher load per unit surface area, thus providing higher specificity and/or activity on a smaller scale.1–3 During recent years, viral nanoparticles (VNPs) have become popular nanobuilding blocks for material fabrication. Straightforward crystallization procedures can lead to mesoscale selforganization, and 2-D and 3-D crystals can be readily obtained.4,5 Self-supporting crystalline thin films of rod-shaped VNPs can be fabricated by making use of the anisotropic nature of these particles.6,7 VNPs are unique in that they are made from regular biomolecular building units that self-assemble into a regular structure of well-defined size. The particle’s multivalency allows for the decoration of the particles with multiple functionalities. With such design tools at hand, functionalized VNPs offer new routes for the design and construction of novel biomaterials at the nano/microscale. Our research focuses on the utilization of capsids from the well-characterized plant virus Cowpea mosaic virus (CPMV) for the design and construction of new materials. CPMV has * Corresponding author. E-mail:
[email protected]. Telephone: +1 858 784 7124. Fax: +1 858 784 7979. † Department of Biological Chemistry, John Innes Centre. ‡ Department of New Materials and Biosystems, Max-Planck Institute for Metals Research. § Department of Biophysical Chemistry, University of Heidelberg. 4 Present address: Department of Cell Biology, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037. ⊥ Present address: Biosurfaces Unit, CIC BiomaGUNE, Paseo Miramon 182, 20009 San Sebastian, Spain.
appealing features that can be exploited for use in nanobiotechnology: the particles are ca. 30 nm in diameter, extremely robust, monodisperse, have a high degree of symmetry and polyvalency, and can be produced in gram quantities. The structure of CPMV is known to near-atomic resolution and the genetic properties are well characterized.8,9 The capsid surface is multidentate and provides binding sites for a wide range of chemical and biological moieties. We and others have demonstrated that CPMV particles can be decorated and functionalized with a broad range of biological and chemical functional groups including fluorescent dyes, metallic and semiconducting nanoparticles, redoxactive molecules, antibodies, and oligonucleotides.10–21 Besides chemical derivatization, the availability of cDNA clones allow sitedirected and insertional mutagenesis.22–24 Although plant viral capsids have been extensively studied and utilized in the area of nanobiosciences/technology, most studies focus on their chemical or biological modification in solution. To date only a few articles have been published concerning the immobilization of VNPs on solid supports. While CPMV monolayers have been fabricated by immobilization of the particles by chemisorption, covalently, and by biospecific recognition,25–28 only one report describes the construction of arrays consisting of CPMV multilayers. We have recently demonstrated the construction of CPMV bi- and trilayer assemblies consisting of virions that were labeled with two different ligands: fluorescent dyes and biotin molecules.16 The biotin functionalization allowed for the bottom-up assembly of arrays via a layer-by-layer approach, exploiting the molecular recognition between streptavidin (SAv) and biotinylated virions (CPMV-biotin). Because of the extraordinarily high affinity constant between SAv and biotin (Ka > 1014 M-1),29 chemical stability of the system is expected. The fluorescent labels enabled detection and characterization of the assemblies by fluorescence
10.1021/bm700797b CCC: $40.75 2008 American Chemical Society Published on Web 01/16/2008
VNP Multilayer Arrays
Figure 1. Schematic of Cowpea mosaic virus (CPMV) particles decorated with biotin. (a) CPMV labeled with biotin via an LC linker; (b) CPMV labeled with biotin via an LCLC linker.
microscopy. This study demonstrated that multilayer arrays of CPMV can be constructed. Data about the assembly mechanism and the properties of the resulting assembly were not available. Here, we extend these studies by providing further and novel details on the kinetics of multilayer assembly and on the mechanical properties of the resulting arrays. Array formation via alternating deposition of SAv and CPMV-biotin on a solidsupported biotinylated lipid bilayer was followed by quartz crystal microbalance with dissipation monitoring (QCM-D). QCM-D is an analytical surface-sensitive technique that is based on an acoustomechanic transducer principle. It is a unique method for observing events on surface-confined films in situ and in real time. By measuring the shifts in the resonance frequency (∆f) and energy dissipation (∆D) of an oscillating sensor crystal, information about adsorbed masses and the mechanical properties, dynamics and hydration of resulting films can be obtained. To a first approximation, the frequency shift is related to the mass of the deposited film, including coupled water, while the dissipation is sensitive to viscoelastic properties of the adlayer.30 The QCM-D technology has been applied for a wide range of studies, among them the adsorption of peptides and proteins on surfaces and the characterization of interactions between biological binding partners including ligand–receptor, antibody–antigen, oligonucleotide, and carbohydrate interactions. QCM-D has also been used as a tool to develop detection assays for viral and microbial infections.31 To the best of the authors’ knowledge, the QCM-D technique has so far not been applied to characterize the properties of multilayer arrays of viral nanoparticles or nanoparticles in general. In this study, we have analyzed the buildup of multilayers of CPMV particles displaying various densities of biotin molecules that were attached to the CPMV particles via shorter and longer linkers. CPMV particles were covalently decorated at solventexposed exterior lysine residues with either biotin-LC-Nhydroxysuccinimide (NHS) or biotin-LCLC-NHS (Figure 1). The theoretical maximum length of the spacer, LC or LCLC, was around 22.4 and 30.5 Å, respectively.
Materials and Methods Virus Growth and Purification. The propagation and purification of CPMV wild type virions were performed by standard procedures.32 Purified virions were stored at 4 °C in 10 mM sodium phosphate buffer pH 7.0. The concentration of purified virions was determined by Bradford assay or photometrically; the molar extinction coefficient of CPMV at a wavelength of λ ) 260 nm is ) 8.1 mL mg-1 cm-1.33 UV-visible spectra were recorded using a Perkin-Elmer Lambda 25 UV-visible spectrometer and UVWINLab software. Chemical Modification of CPMV with Biotin. VNPs were decorated with the NHS-activated biotin molecules using standard
Biomacromolecules, Vol. 9, No. 2, 2008 457 protocols for the modification of primary amines with NHS-activated compounds. In brief, particles were mixed with an excess of NHSactivated biotin (biotin-LC-NHS and biotin-LCLC-NHS; Pierce) of 3000 biotin conjugates per 1 CPMV particle and incubated overnight at 4 °C in order to allow for full decoration of the particles (up to 240 out of a total of 300 solvent-exposed lysines are biotinylated). To achieve partial labeling, the particles were incubated with a molar excess of 600 moles biotin per 1 mol CPMV. The reaction time was limited to two hours at room temperature, resulting in labeling with 30–40 biotin molecules. The density of conjugated biotins could not be determined by a direct method and was therefore estimated based on labeling studies with NHS-activated fluorescent dyes using the same reaction condition and also by double labeling reactions, as previously described.16 In the following text, particles labeled partially (P) and fully (F) with biotin-LC are designated CPMV-(LC-BIO)P and CPMV-(LCBIO)F, respectively. Particles decorated with biotin-LCLC are described as CPMV-(LCLC-BIO)P and CPMV-(LCLC-BIO)F, respectively. Transmission Electron Microscopy (TEM). TEM studies were performed using a JEOL JEM-1200 EX EM (Philips). For visualization the particles were negatively stained with 2% (w/v) uranyl acetate. Samples were prepared on carbon-coated copper grids (400 mesh; Agar Scientific). Native Gel Electrophoresis. CPMV (10 µg) particles in 10 mM sodium phosphate buffer pH 7.0 were analyzed (added loading dye; MBI Fermentas) on 1.2% (w/v) agarose gels in an electric field strength of 1–5 V/cm. Ethidium bromide (0.1 µg/mL) in 1× TBE buffer (10× TBE-Buffer: 900 mM Tris-base, 900 mM boric acid, 25 mM EDTA, filled up with MilliQ water to a final amount of 1000 mL) was added to the gel. After completion of the electrophoretic separation, the viral particles were visualized on a UV transilluminator at a wavelength of 302 nm. Documentation of the bands was carried out by a video documentation system from Gene Genius Bio Imaging Systems using the software Gene Snap (Syngene). Dot Blot Studies. To test for successful biotinylation, 10 µg of modified and unmodified CPMV particles were spotted on nitrocellulose membranes (Amersham) and air-dried for 30–60 min.. The membranes were blocked for two hours at room temperature with PBS (phosphate buffered saline) + 0.05% (v/v) Tween20 (Sigma) + 5% (w/v) dry milk (Marvel). The blots were then probed with streptavidin-labeled horseradish peroxidase (SAv-HRP) (Zymed) diluted 1:1000 in PBS + 0.05% (v/v) Tween20 + 5% (w/v) dry milk. After washing with 3× PBS + 0.05% (v/v) Tween20 and 1× PBS for 10 min at room temperature, signals were visualized by electrochemiluminescence (ECL) using a Curix 60 film processor (Agfa Gevaert). QCM-D Measurements. The experiments were performed using a Q-Sense E4 system (Q-Sense, Gothenburg, Sweden). Briefly, upon interaction of (soft) matter with the surface of a sensor crystal, changes in its resonance frequency (∆f) related to attached mass (including coupled water), and in its dissipation (∆D) related to frictional (viscous) losses in the adlayer, were measured with a time resolution of better than 1 s. The system was operated in flow mode, with a flow rate ranging from 5 to 100 µL/min. Sample solution was continuously delivered to the measurement chamber by the aid of a peristaltic pump (ISM597D; Ismatec, Zürich, Switzerland). To switch between sample liquids, the flow was interrupted for a few seconds without disturbing the QCM-D signal. With this setup, adsorption and interfacial processes can be followed in situ while successively exposing different solutions to the surface. Resonance frequency and dissipation were measured at six harmonics (15, 25, . . . 65 MHz) simultaneously. For simplicity, only changes in dissipation and normalized frequency (∆f ) ∆fn/n, with n being the overtone number) of the fifth overtone (n ) 5, i.e., 25 MHz) are presented. The working temperature was 24 °C. Adsorbed (wet) masses, m, were calculated according to Sauerbrey’s equation,34 ∆m ) -C × ∆f, with C ) 17.7 ng cm-2 Hz-1. For thicker nonrigid films, the Sauerbrey relation may underestimate the adsorbed mass.30 We estimated the error to be below 25% for the investigated
458 Biomacromolecules, Vol. 9, No. 2, 2008
Steinmetz et al.
Figure 3. Presentation of results from dot blot studies of native and biotinylated Cowpea mosaic virus (CPMV) particles. CPMV was spotted on nitrocellulose and detected via streptavidin-labeled horseradish peroxidase. A dark signal shows successful biotinylation. 1 ) PBS buffer, 2 ) CPMV, 3 ) CPMV-(LC-BIO)P, LC ) short linker; P ) partial decoration, i.e., 30–40 biotin molecules per CPMV particle, 4 ) CPMV-(LCLC-BIO)P, LCLC ) long linker, 5 ) CPMV-(LC-BIO)F, F ) full decoration with up to 240 biotin labels per CPMV, and 6 ) CPMV-(LCLC-BIO)F.
Figure 2. Transmission electron micrographs of biotinylated Cowpea mosaic virus (CPMV) particles. The scale bar represents 100 nm. (a) CPMV-(LCLC-BIO)F; (b) CPMV-(LC-BIO)F; (c) CPMV-(LCLCBIO)P; (d) CPMV-(LC-BIO)P. LC ) short linker, LCLC ) long linker, F ) full decoration with up to 240 biotin labels per CPMV, P ) partial decoration, i.e., 30–40 biotin molecules per CPMV particle.
films, as tested by comparison with the viscoelastic model35,36 implemented in the software QTools 2 (Q-Sense, Gothenburg, Sweden). Silica-coated QCM-D sensors (Q-Sense, Gothenburg, Sweden) were cleaned by immersion in a 2% SDS solution for 30 min, rinsing with MilliQ water, blow-drying with a stream of nitrogen, and exposure to oxygen plasma (0.4 mbar, 150 W) (100-E Plasma System; TePla, Feldkirchen, Germany) for 30 min. Cleaned substrates were stored in air and again exposed to oxygen plasma (5 min) prior to use. The QCM-D sensors were functionalized by formation of a biotindoped supported lipid bilayer (SLB) via the method of vesicle spreading. To this end, small unilamellar vesicles containing fractions of 10 mol % dioleoylphospatidylethanolamine-CAP-biotin (DOPE-CAP-Biotin) and 90 mol % dioleoylphosphatidylcholine (DOPC) were exposed to the sensor surface, resulting in the formation of SLBs that provide functional sites for the immobilization of SAv.37 SLB-formation, as well as binding of SAv and biotinylated CPMV samples, was directly monitored and controlled by QCM-D. Vesicles were added at a concentration of approximately 50 µg/mL in HEPES buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 2 mM EDTA, and 3 mM NaN3). Stock solutions of SAv and biotinylated CPMV particles in PBS were diluted with HEPES buffer and added at a concentration of approximately 10 µg/mL.
Results and Discussion Modification of CPMV Particles. The integrity of CPMV particles after chemical modification with biotin was verified by TEM (Figure 2). All electron micrographs show intact particles with a uniform size of around 30 nm, as expected for CPMV particles. The assembly of VNPs into clusters is likely due to capillary effects during deposition and preparation of the samples on the EM grids. Integrity was further confirmed by native gel electrophoresis (Figure 4) showing only single bands. Successful attachment of biotin was confirmed by dot blot tests on nitrocellulose probed with SAv-HRP followed by ECL detection (Figure 3) and native gel electrophoresis (Figure 4). The latter also indicated that particles reacted under forcing conditions were labeled with more biotin molecules when
Figure 4. Cowpea mosaic virus (CPMV) particles after separation in an electric field on a 1.2% agarose gel with ethidium bromide staining visualized under UV light. Lane 1 ) CPMV, 2 ) CPMV-(LC-BIO)P, LC ) short linker, P ) partial decoration, i.e., 30–40 biotin molecules per CPMV particle, 3 ) CPMV-(LCLC-BIO)P, LCLC ) long linker, 4 ) CPMV-(LC-BIO)F, F ) full decoration with up to 240 biotin labels per CPMV), and 5 ) CPMV-(LCLC-BIO)F.
compared to particles incubated with a lower excess of chemicals (600 versus 3000 biotin moieties per CPMV) and a shorter reaction time (two hours versus overnight). The biotinylated particles showed a higher mobility toward the anode in native gels when compared to unmodified native particles. This higher mobility can be explained by a charge effect. Under physiological conditions, the lysines were protonated and thus contribute to the overall charge of the VNPs. Biotinylation of solvent-exposed lysines leads to the deletion of these positive charges, which renders the particles more negatively charged and thereby increases their mobility. Further, the more biotins attached, the more lysines are substituted (equates to the deletion of positive charge), and thus the higher mobility of particles fully decorated with biotin in comparison to those that were partially labeled with biotin. Construction and Characterization of Multilayer Arrays. Alternating layers of SAv and biotinylated CPMV particles were constructed on silica surfaces that were functionalized with a biotin-doped SLB. All deposition steps were monitored by QCM-D. We first focus on the results obtained for CPMV particles that were fully decorated with biotin using the long spacers, i.e., CPMV-(LCLC-BIO)F (Figure 5a). The QCM-D response upon exposure of biotin-doped small unilamellar vesicles (9-t32 min.) to silica-coated sensors showed a characteristic two-phase behavior, reflecting the initial adsorption of intact vesicles, which is followed by the formation of a SLB.38,39 The shifts in frequency and dissipation at saturation of -25 Hz and
