Research Article www.acsami.org
Control of Protein Affinity of Bioactive Nanocellulose and Passivation Using Engineered Block and Random Copolymers Maija Vuoriluoto,† Hannes Orelma,*,†,§ Baolei Zhu,‡ Leena-Sisko Johansson,† and Orlando J. Rojas† †
Biobased Colloids and Materials (BiCMat), Department of Forest Products Technology, School of Chemical Technology, Aalto University, FI-00076, Espoo, Finland ‡ DWI−Leibniz-Institute for Interactive Materials Research, Forckenbeckstr. 50, D-52056 Aachen, Germany S Supporting Information *
ABSTRACT: We passivated TEMPO-oxidized cellulose nanofibrils (TOCNF) toward human immunoglobulin G (hIgG) by modification with block and random copolymers of poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA) and poly(oligo(ethylene glycol) methyl ether methacrylate) (POEGMA). The block copolymers reversibly adsorbed on TOCNF and were highly effective in preventing nonspecific interactions with hIgG, especially if short PDMAEMA blocks were used. In such cases, total protein rejection was achieved. This is in contrast to typical blocking agents, which performed poorly. When an anti-human IgG biointerface was installed onto the passivated TOCNF, remarkably high affinity antibody−antigen interactions were observed (0.90 ± 0.09 mg/m2). This is in contrast to the nonpassivated biointerface, which resulted in a significant false response. In addition, regeneration of the biointerface was possible by low pH aqueous wash. Protein A from Staphylococcus aureus was also utilized to successfully increase the sensitivity for human IgG recognition (1.28 ± 0.11 mg/m2). Overall, the developed system based on TOCNF modified with multifunctional polymers can be easily deployed as bioactive material with minimum fouling and excellent selectivity. KEYWORDS: biosurfaces, human IgG, cellulose nanofibrils, antifouling, TEMPO-oxidation, PDMAEMA, POEGMA, nonspecific adsorption
■
poly(vinyl alcohol),12 and polyacrylamide.13 Blocking agents widely utilized against nonspecific protein and bacteria adsorption also include macromolecules based on polyethylene glycol (PEG).9,14−19 A promising method to control specific and nonspecific adsorption involves the installation of multifunctional polymers incorporating surface-binding and surface-passivating domains.9,20−22 Triblock polymers with PEG blocks have been shown to passivate hydrophobic substrates,23 whereas cationic poly(L-lysine) grafted with multiple PEG side chains has yielded a similar effect for negatively charged metal oxide surfaces.9,15,16 It is established, however, that PEG has low affinity for cellulose24,25 but various indirect routes mediate their interactions.18,25−27 We have previously reported on the adsorption of block copolymers of cationic28 poly(2-(dimethylamino)ethyl methacrylate) (PDMAEMA) and highly hydrophilic poly(oligo(ethylene glycol) methyl ether methacrylate) (POEGMA) on low-, medium-, and highly charged cellulose (regenerated cellulose, CNF and on highly charged TEMPO-oxidized
INTRODUCTION Recent efforts have been aimed at the development of biobased, biocompatible materials from renewable resources. In parallel, there are increasing demands in biomedical and biomolecular engineering to design biosensors, diagnostic assays, and implants displaying high performance at low cost. Recently, cellulose and cellulosic materials, such as paper, nanopaper and films based on cellulose nanofibrils (CNF) have been shown to be suitable substrates for diagnostics.1−7 These materials compete with nitrocellulose, a flammable and brittle cellulose derivative with high affinity toward proteins, which has been utilized for decades as a supporting material in bioassays.8 Moreover, wood-derived CNF offers a sustainable platform for bioassays due to its many inherent qualities, including availability, biocompatibility, porosity, hydrophilicity, and low toxicity.1 Biomedical applications often require controlled immobilization of recognition molecules on the substrate and antifouling properties, mainly to lower nonspecific biomolecule adsorption and minimize interferences.9 Nonspecific adsorption is problematic because it may block otherwise active binding sites or result in a false positive bioassay.10 Prominent methods to minimize nonspecific protein adsorption include proteinresistant coatings based on polymers such as dextran,11 © 2016 American Chemical Society
Received: December 3, 2015 Accepted: February 4, 2016 Published: February 4, 2016 5668
DOI: 10.1021/acsami.5b11737 ACS Appl. Mater. Interfaces 2016, 8, 5668−5678
Research Article
ACS Applied Materials & Interfaces CNF).29 The adsorption mechanism of PDMAEMA-POEGMA block copolymers was attributed to electrostatic interaction between the cationic segment and the anionic cellulose surfaces as the adsorption increased linearly with surface charge of the substrate. In this study, we propose similar block copolymers to control protein affinity on highly anionic TEMPO-oxidized CNF (TOCNF). This is from the anticipation that cationic PDMAEMA blocks act as anchors for hydrophilic POEGMA segments, thereby passivating the cellulosic support. Here, we report on the reversibility of PDMAEMA-bPOEGMA copolymer adsorption on highly charged TOCNF as studied by in situ multiparametric surface plasmon resonance (MP-SPR) measurements. Block copolymers were found to adsorb partially reversibly on the highly charged TOCNF. Compared to common commercial blocking agents, copolymers with short PDMAEMA segments and large POEGMA blocks were found to be very effective in preventing nonspecific human immunoglobulin G (hIgG) binding (Scheme 1). The
RAFT polymerization and random copolymers by ATRP with size exclusion chromatography is described in detail elsewhere.29 The block and random copolymers were quaternized with methyl iodide. The average number of ethylene glycol units per oligo(ethylene glycol) methyl ether methacrylate monomer was 8−9. All together, six PDMAEMA-b-POEGMA copolymers with varying amounts of DMAEMA (33, 58, or 74) and OEGMA monomers (10−137) were utilized in this work. The utilized two random copolymers had welldefined molar percentage ratios of DMAEMA and OEGMA (10−90% and 70−30%). The block and random copolymers are herein referred to as Dn-b-EGMAm or Dn-rnd-EGMAm, where n and m represent the number of DMAEMA and OEGMA units in the copolymer, respectively. A detailed analysis of the utilized copolymers can be found in the Supporting Information, sections S1 and S2. Preparation of Cellulose Nanofibril Thin Films. The method of Ahola et al.30 was followed to prepare the CNF thin film substrates that were later used for TEMPO-oxidation. Briefly, CNF-gel (0.190 wt % in water) was ultrasonicated (25% amplitude, 400W tip sonicator, Brandon 450 Digital Sonifier, Branson Ultrasonics, Danbury, CT) for 10 min and sequentially centrifuged (10 400 rpm, 45 min, 25 °C). From the clear supernatant individualized nanofibrils were collected. Gold SPR crystals (Oy BioNavis Ltd., Ylöjärvi, Finland) were UV/ ozonized and preadsorbed with a thin anchoring layer of PEI and consecutively spin-coated with the nanofibril suspension at 3000 rpm and 90 s of spinning time. The spin-coated CNF films were annealed at 80 °C for 10 min and stored at ambient conditions. Prior to measurements in MP-SPR, the crystals were stabilized in water overnight. Multiparametric Surface Plasmon Resonance (MP-SPR). A multiparametric Surface Plasmon Resonance instrument (MP-SPR Model Navi 200, Oy BioNavis Ltd., Ylöjärvi, Finland) was utilized in monitoring real-time block copolymer adsorption with regeneration and protein adsorption on TOCNF. The light beam in MP-SPR is directed at the crystal producing a sharp attenuation of reflectivity when energy is lost as part of the incident energy of certain wavelengths and angles of the incident light is coupled with the surface plasmon wave traveling along the interface between the analyzed solution and gold layer.31 The angles and wavelengths at which the surface plasmon resonance effect takes place correlate very closely with the refractive index of media in contact with the SRP crystal surface.32−34 The thickness of the adsorbed layer is calculated with eq 1:33
Scheme 1. Schematic Illustration of (1) PDMAEMA-bPOEGMA Adsorption on Anionic TEMPO-oxidized CNF (TOCNF) and (2−3) Passivation of Non-specific Human IgG Adsorption
topographical changes on TOCNF surfaces were studied by atomic force microscopy (AFM). Anti-human IgG biointerface preparation for hIgG detection was also investigated by MPSPR and compositional changes followed at different stages of biointerface preparation by X-ray photoelectron spectroscopy (XPS).
■
d=
ld Δangle 2 m(na − n0)
(1)
where Δangle is the change in the MP-SPR angle, ld is a characteristic evanescent electromagnetic field decay length, estimated as 0.37 of the light wavelength (240 nm), m is a sensitivity factor for the sensor obtained after calibration of the MP-SPR (109.94°/RIU), na is the refractive index of the adsorbed substance), and n0 is the refractive index of the bulk solution (1.334 RIU). The refractive indices, 1.57 for proteins33 and 1.46 for the block and random copolymers (based on POEGMA35), were estimated based on literature. Eq 234 was utilized to estimate the adsorbed amount per unit area:
EXPERIMENTAL SECTION
Materials. Cellulose nanofibrils were obtained by processing bleached sulphite birch pulp (Consistency 1.5%) through a Masuko grinder three times followed by further fluidization (at least 12 passes) in a M110P microfluidizer (Microfluidics Corp., Newton, MA) equipped with a chamber pair (200 and 100 μm chambers) operated at 2000 bar pressure. Human IgG (#I4506), anti-human IgG (#I9764), protein A (#P6031), bovine serum albumin (BSA #A2934), methoxypolyethylene glycol amine (PEG-amine #06679), N-hydroxysuccinimide, (NHS #130672) 1-ethyl-3-[3-(dimethylamino)propyl]cardodiimide hydrochloride (EDC #03450), (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO #426369) and ethanolamine (#398136) were all purchased from Sigma-Aldrich (Helsinki, Finland). SuperBlock (#37545), a commercial protein solution, was obtained from Pierce (Rockford, IL, USA). All used laboratory chemicals were of analytical grade. Deionized water purified with a Millipore Synergy UV unit (Milli-Q) was used in all experiments. Preparation of PDMAEMA-POEGMA Block and Random Copolymers. The preparation and characterization of the poly(2(dimethylamino)ethyl methacrylate) (PDMAEMA) and poly(oligo(ethylene glycol) methyl ether methacrylate) block copolymers by
Δm = dρ
(2)
where d is the thickness of the adsorbed layer and ρ is the packing density of the adsorbed species. The estimated packing densities based on literature were 1.3 g/cm3 for proteins34 and 1.15 g/cm3 for the block and random copolymers (based on POEGMA36). It is important to note that eq 2 assumes that the adsorbed layer density is constant with the layer thickness, and therefore the calculation carries some uncertainty. For instance, the de Feijter equation typically used in ellipsometry or optical reflectometry could be used to account for the fact that the refractive index used in the analysis to represent the adsorbed layer is that of the whole layer (with the contributions from both protein and water, or polymer and water) and is not the refractive index of a protein molecule (or a block copolymer molecule). Furthermore, PDMAEMA and POEGMA and their random 5669
DOI: 10.1021/acsami.5b11737 ACS Appl. Mater. Interfaces 2016, 8, 5668−5678
Research Article
ACS Applied Materials & Interfaces
Preparation of Oriented Anti-human IgG Biointerface on TOCNF Utilizing Protein A from Staphylococcus aureus. An oriented anti-hIgG biointerface was prepared following a similar to that presented before for nonoriented systems. Protein A from S. aureus was conjugated on TOCNF via EDC/NHS chemistry and monitored with MP-SPR. EDC/NHS activated TOCNF surface was conjugated with protein A (0.1 mg/mL) from 10 mM acetate buffer solution (pH 5). After ethanolamine treatment (0.1 M, pH 8.5), antihuman IgG (0.1 mg/mL) in 10 mM phosphate buffer solution (pH 7.4) was adsorbed on the surface. Nonspecific hIgG adsorption was prevented with selected block copolymer D33-b-EGMA137 (0.5 g/L) in PB (10 mM, pH 7.4). Finally, hIgG (0.1 mg/mL) was adsorbed on the prepared biointerface. AFM and XPS Analyses. Substrates for AFM and XPS experiments were prepared following a procedure similar to that discussed previously but on silicon wafers, which are convenient substitute for MP-SPR gold surfaces. Nanoscope IIIa multimode scanning probe microscope (Digital Instruments, Santa Barbara, CA) was utilized to analyze topographical changes on dried TOCNF films without and with adsorbed block copolymers and human IgG. Images (1 × 1 μm) were scanned in air with silicon cantilevers operating tapping mode. Each sample was imaged from three separate areas and flattening was the only image processing step applied. The RMS roughness values were calculated from the 1 × 1 μm images. X-ray photoelectron spectroscopy was carried out with an AXIS Ultra spectrometer by Kratos Analytical with an X-ray source of monochromatic Al Kα at 100 W. The surface chemical composition was determined for dried samples at different stages of anti-human IgG biointerface preparation. The analysis depth and area were
