Subscriber access provided by Fudan University
Article
Self-Assembly of Recombinant Silk as a Strategy for ChemicalFree Formation of Bioactive Coatings – a Real-Time Study Linnea Nilebäck, Jesper Niels Hedin, Mona Widhe, Lotta S Floderus, Annika Krona, Helena Bysell, and My Hedhammar Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.6b01721 • Publication Date (Web): 13 Feb 2017 Downloaded from http://pubs.acs.org on February 14, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biomacromolecules
Self-Assembly of Recombinant Silk as a Strategy for Chemical-Free Formation of Bioactive Coatings – a Real-Time Study Linnea Nilebäck†, Jesper Hedin‡, Mona Widhe†, Lotta S. Floderus†, Annika Krona§, Helena Bysell‡, My Hedhammar*,† †
KTH Royal Institute of Technology, School of Biotechnology, AlbaNova University Center,
SE-106 91 Stockholm, Sweden ‡
SP Technical Research Institute of Technology, SP Chemistry, Materials and Surfaces,
Drottning Kristinas väg 45, SE-114 86 Stockholm, Sweden §
SP Technical Research Institute of Technology, SP Food and Bioscience, Soft Materials
Science, Box 5401, SE-402 29 Gothenburg, Sweden
ACS Paragon Plus Environment
1
Biomacromolecules
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 38
ABSTRACT
Functionalization of biomaterials with biologically active peptides can improve their performance after implantation. By genetic fusion to self-assembling proteins, the functional peptides can easily be presented on different physical formats. Herein, a chemical-free coating method based on self-assembly of the recombinant spider silk protein 4RepCT is described and used to prepare functional coatings on various biomaterial surfaces. The silk assembly was studied in real-time, revealing occurrence of continuous assembly of silk proteins onto surfaces and formation of nanofibrillar structures. The adsorbed amounts and viscoelastic properties were evaluated, and the coatings were shown to be stable against wash with hydrogen chloride, sodium hydroxide, and ethanol. Titanium, stainless steel, and hydroxyapatite were coated with silk fused to an antimicrobial peptide or a motif from fibronectin. Human primary cells cultured on the functional silk coatings show good cell viability and proliferation, implying potential to improve implant performance and acceptance by the body.
KEYWORDS coatings, recombinant spider silk, self-assembly, functionalization, implants
INTRODUCTION Hard implants, such as dental and orthopaedic prostheses, are associated with a risk of implant failure. This can be due to infections or aseptic loosening, a result of e.g. poor osseointegration.1–
ACS Paragon Plus Environment
2
Page 3 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biomacromolecules
3
Improving cell adhesion to implants and providing antimicrobial agents are two important
approaches to address these issues. Strategies to improve osseointegration include sandblasting, acid etching, anodization, and chemical oxidation of metal implants to tailor the nanotopography and physicochemical properties of the surfaces.4–6 Additionally, different immobilization strategies to attach cell-binding motifs on implant surfaces have been reported (e.g. through spin coating with elastin-like protein containing RGD-motifs, followed by stabilization through UV crosslinking).7 Vidal et al. presented a strategy in which both RGD-peptides and titanium binding peptides were chemically coupled to silkworm fibroin so that the silk proteins could adsorb specifically onto titanium disks.8 Anti-infective strategies include physical release of antibiotics.9,10 As an alternative, in order to avoid further resistance development toward conventional antibiotics, antimicrobial peptides are gaining increased interest. Several successful cases of tethering such peptides to surfaces through covalent attachment via various linker molecules have been reported.11,12 Recombinant silk is considered as a promising biomaterial for several applications due to its strength, elasticity, cell compatibility, and low immunogenicity.13,14 The partial spider silk protein 4RepCT derived from the major ampullate spidroin 1 of Euprosthenops australis can be recombinantly produced together with biologically active peptides such as cell binding motifs.15,16 These recombinant silk proteins can be processed into different formats such as fibers, meshes, and foams,17 and have shown to be suitable as scaffolds for cell culturing (e.g. human primary cells and pancreatic islets).16,18,19 Recently, it was shown that silk proteins recombinantly produced together with larger functional domains such as affinity domains and enzymes also retain the ability to assemble into macroscopic silk structures, as well as exhibiting activity from the functional domains.20–22 In order to address in vivo complications associated
ACS Paragon Plus Environment
3
Biomacromolecules
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 38
with surface properties of hard implants, a thin silk coating could constitute a viable option to modify the implant surface properties. The functionality of such silk coatings could easily be altered by using silk proteins produced recombinantly in fusion with functional peptide motifs or domains. Moreover, silk proteins with different motifs can be mixed to allow formation of multifunctional silk. In this work, we investigated the possibility to utilize the self-assembly propensity of the recombinant silk protein 4RepCT to form stable silk coatings on surfaces without covalent attachment, so that no additional chemicals are needed during the coating process. First, we characterized the surface adsorption and assembly behavior of non-functionalized silk proteins (wild-type 4RepCT, hereafter denoted WT-silk). The assembly of the proteins on surfaces was studied in real-time by several techniques. Quartz Crystal Microbalance with Dissipation (QCMD) can be used to monitor molecular interactions on a sensor surface in real-time with high sensitivity (ng/cm2).23 A decrease in the resonance frequencies of the sensor corresponds to an increase in the adsorbed mass, and the dissipation of oscillation gives information about the viscoelastic properties of the coating. Surface Plasmon Resonance (SPR) and ellipsometry are optical techniques that detect changes in reflection angles and light polarization changes due to interactions of proteins close to the surface, respectively, and can be compared to QCM-D data to reveal the influence of entrapped water, contributing to the response of the latter technique only. To study the nanostructure of the silk coatings, Atomic Force Microscopy (AFM), which allows for high-resolution topographic imaging at the nanoscale, was used. This was performed in aqueous solution, to maintain the native structure during the characterization. Approaching the goal of introducing biological activity to implant surfaces, silk proteins recombinantly fused to a peptide motif from fibronectin16 (FN-silk), and the antimicrobial
ACS Paragon Plus Environment
4
Page 5 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biomacromolecules
peptide Magainin I24 (Mag-silk), respectively, were used for coating formation and their functions were evaluated. Both silk fusion proteins were shown to possess the ability to assemble on titanium, hydroxyapatite, stainless steel, and polystyrene, materials that are commonly used in orthopaedic and dental implants, coronary stents, and in vitro cell cultures, respectively.
EXPERIMENTAL SECTION Materials. Recombinantly produced WT-silk (wild-type 4RepCT), Mag-silk (antimicrobial motif with the sequence GIGKFLHSAGKFGKAFVGEIMKS fused to 4RepCT), FN-silk (cell binding motif with the sequence CTGRGDSPAC fused to 4RepCT) and Protease 3C were kindly provided by Spiber Technologies AB. Major secondary structure motifs were compared based on an available structure for Protease 3C (PDB-ID: 3ZYD) and a previous investigation verifying a β-sheet rich secondary structure of 4RepCT in silk format15. Proteins were used in 20 mM Tris(hydroxymethyl)aminomethane (Tris) buffer (pH 8.0) and kept cool on ice during adsorption measurements. Alkylthiol solutions of 2 mM 1-undecanethiol were prepared in 99.5% ethanol. 2% polystyrene solution was prepared by dissolving petri dish pieces in 99.5% toluene. Quartz Crystal Microbalance with Dissipation Monitoring, QCM-D, takes advantage of the piezoelectric properties of AT-cut quartz crystals to monitor changes in its oscillation frequency and dissipation of oscillation through application of a pulsating voltage. Upon adsorption of mass onto the crystal sensor, the frequency decreases and the dissipation increases. By comparing dissipation changes to frequency changes, viscoelastic properties of the adsorbed layer can be assessed. QCM-D sensors coated with titanium, stainless steel of type SS2343, and hydroxyapatite (Biolin Scientific) were cleaned according to the manufacturer recommendations and used
ACS Paragon Plus Environment
5
Biomacromolecules
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 38
without further modification. Gold coated QCM-D sensors were cleaned by a three-step procedure starting with immersion in 98% formic acid in a fume hood for 10 minutes followed by extensive rinsing in Milli-Q water, 5 minutes of plasma treatment at maximum power in a Harrick Plasma PDC-3XG plasma cleaner, and subsequent incubation in a 6:1:1 mixture of Milli-Q water, 32% ammonia, and 30% hydrogen peroxide in a fume hood for 8 minutes at 80 °C. After extensive rinsing in Milli-Q water, the surfaces were dried in nitrogen gas and incubated in alkylthiol solution overnight. Excessive alkylthiols were removed by 5 minutes of ultra-sonication in 99.5% ethanol, repeated 5 times. The presence of an alkylthiol layer on the gold was verified by measuring the contact angle of Milli-Q water drops (2 µL) on each sensor on a DataPhysics OCA40 instrument, verifying hydrophobic surfaces with a water contact angle of 94°-104°. Polystyrene sensors were prepared by cleaning gold sensors as described above, followed by spin coating with 2% polystyrene in toluene (10 µl) in a Headway Research Inc. PWM32 spin coater at 1600 rpm, 60 seconds, two times per sensor. Spin coated sensors were dried in 150 °C for 15 minutes. QCM-D measurements were conducted in an E4 instrument (Q-Sense AB). The flow was set to 20 µL/min and the temperature was fixed at 20.0 °C. The third frequency overtone was chosen for representation in the results section. The system was equilibrated with 20 mM Tris buffer until the baseline was stabilized. Protein solution was flowed over the sensors for 120 minutes, followed by a flow of 20 mM Tris buffer to confirm that proteins had adsorbed well to the sensors and was not washed away. Repeated experiments (N) were performed with parallel replicates (n), with N=3, n=2 for WT-silk and N=4, n=1 for Protease 3C on alkylthiol surfaces, n=1 for WT-silk on polystyrene and hydroxyapatite, and n=1 for FN-silk and Mag-silk on polystyrene, hydroxyapatite, stainless steel and titanium, respectively. Concentration dependence
ACS Paragon Plus Environment
6
Page 7 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biomacromolecules
was evaluated by subsequent flow of increasing concentrations of WT-silk over alkylthiolmodified gold sensors with 0.05, 0.1, 0.3, and 0.5 g/L protein solutions, respectively, with a Tris buffer flow in between each protein solution (N=1, n=2). The chemical wash resilience of four protein coatings was investigated in the same settings. After 120 minutes of protein adsorption onto alkylthiol-modified surfaces and 60 minutes of rinsing with 20 mM Tris buffer, the sensors were rinsed with Phosphate Buffered Saline (PBS, 7.8 mM sodium hydrogen phosphate, 1.5 mM potassium dihydrogen phosphate, 2.7 mM potassium chloride, 137 mM sodium chloride, pH 7.4), as well as sodium hydroxide (NaOH, 0.1 and 0.5 M, n=1), hydrochloric acid (HCl, 0.1 and 0.5 M, n=1), or ethanol (20% and 70%, n=2). Each solution was flowed over the sensors for 30 minutes, directly followed by 30 minutes of 20 mM Tris buffer to regain the Tris baseline for evaluation of net frequency changes corresponding to changes in protein amounts on the surfaces. Ellipsometry. Ellipsometry monitors changes in polarization of light that is reflected at the surface. As proteins adsorb onto the surface, the light polarization is changed. By monitoring this, changes in refractive index and adsorbed mass can be obtained. Alkylthiol-modified gold sensors were mounted in a Q-Sense Ellipsometry module to allow simultaneous data collection using an E1 instrument (Q-sense AB) for QCM-D monitoring and a Multiscope ellipsometer (Optrel GBR), Null PCSA operating mode, to record changes in dielectric properties of the protein coatings during adsorption. For the ellipsometry, a 532 nm NdYAG laser was used at 65° angle of incidence. The same settings were used as for measurements in the E4 instrument, except for the flow rate, which was set to 25 µL/min. The QTools software (Biolin Scientific) was used to calculate the coating thicknesses and the mass of adsorbed proteins from the frequency and dissipation shifts based on
ACS Paragon Plus Environment
7
Biomacromolecules
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 38
a one layer model and a Voigt representation. In QCM-D, water that is incorporated into the coatings contributes to the frequency and dissipation shifts so that the mass obtained with these calculations is the wet mass. With ellipsometry, the refractive index (n) and layer thickness (d) can be obtained from the changes in light polarization during protein adsorption onto the surface. The absorbed amount (Γ) was calculated according to Equation (1).25
where n0 is the refractive index of the bulk solution and dn/dc is the refractive index increment (0.154 cm3/g). Since the incorporated water molecules do not influence the change in refractive index of the protein layer relative to the bulk solution, the adsorbed mass obtained by this method can be considered as dry mass. Thus, the water content of the coatings can be calculated as the percentage of wet mass from the total mass through the difference between the total mass, as extracted from QCM-D data, and the dry mass, as extracted from ellipsometry data, using Equation (2).26
Calculations were done on data from triplicate measurements (N=3, n=1) for WT-silk and Protease 3C, respectively. Surface Plasmon Resonance, SPR, is a technique that registers the angle of incidence upon which plasmon resonance occurs in a thin gold layer on a crystal chip. As molecules interact with the chip surface, the incident angle changes, which correlates to changes in refractive index. It thereby indirectly correlates to mass changes close to the chip surface. ProteOn™ GLM sensor chips were demounted from their holders, cleaned and alkylthiolmodified as described above for QCM-D gold sensors. The ProteOn™ XPR36 Protein
ACS Paragon Plus Environment
8
Page 9 of 38
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Biomacromolecules
Interaction Array System (Bio-Rad) was used with the temperature set to 25.0 °C for both the chip and the rack, and the flow rate set to 25 µL/min. The maximum injection volume was used, leading to 960 s of protein injection at the given flow rate. Three protein injections were performed. After each injection, 20 mM Tris buffer was flowed over the chip during the automated syringe refill periods. For WT-silk, N=3, n≥2, whereas for Protease 3C, N=1, n=5. Atomic Force Microscopy. After QCM-D analyses, sensors with adsorbed proteins were used for AFM imaging, in which the surface topography is mapped by measuring the deflection changes of a tip that scans selected areas on the sample in close proximity to the surface. Protein coatings were imaged in droplets of 20 mM Tris buffer using PeakForce Tapping mode in a Bruker Dimension FastScan instrument and ScanAsyst Fluid+ tips. A scan rate of 1.00 Hz and peak force frequency of 1 kHz was used for all images. Imaging in liquid ensures that the peptides retain their native conformation, and improves the resolution of the image by minimizing tip-surface adhesion. Silk Fibers and Coatings for Cell Cultures and Bacterial Cultures. Silk fibers were made in accordance to previous methods27 from protein solutions of 4RepCT (0.05, 0.1, 0.3, and 0.5 g/L, respectively), FN-silk (0.1 g/L), and Mag-silk (0.1 g/L), allowed to assemble overnight. Fibers were photographed in an inverted fluorescence microscope (Nikon Ti-S). Coatings for cell cultures were prepared by incubation of sterile filtered FN-silk, Mag-silk or WT-silk proteins (0.3 g/L) in 96 well non-treated plates (Sarstedt) for 0.5 hours. The liquid was removed and the wells were washed twice with 20 mM Tris buffer before set to dry overnight under sterile conditions. Polystyrene disks (0.8 cm in diameter) for bacterial studies were immersed in Fairy surfactant solution (
