Article pubs.acs.org/Biomac
Lectin-Functionalized Poly(glycidyl methacrylate)-blockpoly(vinyldimethyl azlactone) Surface Scaffolds for High Avidity Microbial Capture Ryan R. Hansen,† Juan Pablo Hinestrosa,† Katherine R. Shubert,‡ Jennifer L. Morrell-Falvey,‡ Dale A. Pelletier,‡ Jamie M. Messman,† S. Michael Kilbey, II,†,§ Bradley S. Lokitz,† and Scott T. Retterer*,†,‡ †
Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States Biosciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States § Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996, United States ‡
S Supporting Information *
ABSTRACT: Microbial exopolysaccharides (EPS) play a critical and dynamic role in shaping the interactions between microbial community members and their local environment. The capture of targeted microbes using surface immobilized lectins that recognize specific extracellular oligosaccharide moieties offers a nondestructive method for functional characterization of EPS content. In this report, we evaluate the use of the block copolymer, poly(glycidyl methacrylate)-block-4,4-dimethyl-2vinylazlactone (PGMA-b-PVDMA), as a surface scaffold for lectin-specific microbial capture. Three-dimensional polymer films were patterned on silicon substrates to provide discrete, covalent coupling sites for Triticum vulgare and Lens culinaris lectins. This material increased the number of Pseudomonas f luorescens microbes captured by up to 43% compared to control scaffolds that did not contain the copolymer. These results demonstrate that PGMA-b-PVDMA scaffolds provide a platform for improved microbe capture and screening of EPS content by combining high avidity lectin surfaces with three-dimensional surface topography.
1. INTRODUCTION Cell surface carbohydrates play an important role in a variety of processes related to cell growth and proliferation. In microbial systems, extracellular carbohydrates are critical in biofouling, host−microbe interactions, cell motility, and immune recognition processes.1 Due to the highly dynamic nature of extracellular glycan expression, an understanding of the role of exopolysaccharides (EPS) in many microbial systems is lacking.2,3 Lectins are a class of proteins that have aided in the study of cell-surface glycosylation patterns and their effects on cell growth and proliferation, as they recognize oligosaccharides with high specificity.4 In recent years, lectin-functionalized surfaces have also been implemented to selectively isolate cell subpopulations in both microarray formats and in flow-based assays.2,5−10 Prominent examples can be seen in mammalian systems aimed at isolating circulating tumor cells and in microbial systems aimed at isolating food-borne pathogens.6−8,11 This approach is attractive because the nondestructive and reversible nature of capture allows for isolation of functional target cells. Using lectins as surface-bound capture proteins requires a surface immobilization strategy that offers a high protein density while also preserving the protein conformation and activity. Lectin-coated coverslips that rely on surface © 2013 American Chemical Society
physisorption for immobilization have been used to characterize extracellular microbial carbohydrates.12,13 Covalent immobilization has also been implemented in efforts to reduce the loss in activity associated with random orientation, heterogeneous surface coverage, and denaturation inherent in nonspecific protein immobilization approaches.14 This has been demonstrated with the development of lectin microarrays fabricated on reactive hydrogel-coated surfaces.2 Functionalizing lectins in three-dimensional microstructures may improve cell capture and proliferation by providing a larger active contact area. Such an approach has been taken using lectin-functionalized PDMS posts in microfluidic systems.7 However, a low percentage of captured target cells has been reported, likely due to inherently weak carbohydrate−lectin association (KD ∼ 10−100 μM).15,16 An attractive approach to developing surfaces that provide high-avidity, lectin-based cellular capture is to use multivalent polymer films. In particular, azlactone-functionalized polymer films have gained increased use as a reactive platform that can modulate surfaces with a secondary functionality through postpolymerization, ring-opening addition reactions with Received: July 31, 2013 Revised: August 30, 2013 Published: September 4, 2013 3742
dx.doi.org/10.1021/bm4011358 | Biomacromolecules 2013, 14, 3742−3748
Biomacromolecules
Article
Scheme 1. General Approach for Using PGMA-b-PVDMA Scaffolds for Microbial Capture
nucleophiles.17−19 The advantages of azlactone-based polymers for surface modifications have recently been reviewed.20 With respect to generating protein surfaces, azlactones react efficiently and selectively with external lysine residues under mild reaction conditions while remaining highly inert to competing hydrolysis reactions, as opposed to other commonly used functional groups such as N-hydroxy-succinimide. Cullen et al. have reported modification of 4,4-dimethyl-2-vinylazlactone (PVDMA) brushes with RNase A, glucose oxidase, DNase I, glucoamylase, and trypsin. These immobilized proteins showed activity similar to or higher than values reported using other surface immobilization strategies.21 Buck et al. have also used azlactone-based polymer films to generate patterns of immobilized proteins or monosaccharides that guide microbial and mammalian cell adhesion at solid interfaces.22,23 Our group recently developed a dually reactive block copolymer, poly(glycidyl methacrylate)-block-poly(vinyldimethyl azlactone) (PGMA-b-PVDMA), using RAFT polymerization that allows biological surface functionality to be tailored based on polymer block length, surface density, and film thickness.24 Motivated by the demonstrated utility of azlactone-based polymer films for facilitating biological surface modifications, we utilize the PGMA-b-PVDMA copolymer as a surface support for the lectin-based capture of selected Pseudomonas gammaproteobacteria. To accomplish this, the PGMA-bPVDMA copolymer is lithographically patterned onto silicon substrates forming three-dimensional, circular polymer films. Upon annealing, these copolymer films undergo microphase segregation, forming a GMA-rich layer at the substrate interface for covalent surface attachment, and a VDMA-rich layer at the polymer−air interface allowing for further modification.24 We herein refer to these patterned films as “PGMA-b-PVDMA scaffolds”. The pendant VDMA groups within PGMA-bPVDMA scaffolds are used to couple high concentrations of lectins which then promote the selective adhesion and aggregation of microbes based on EPS content. The overall approach is summarized in Scheme 1. By integrating high local lectin concentrations with three-dimensional surface structure, the PGMA-b-PVDMA scaffolds increase the number of
captured microbes relative to nonpolymeric scaffolds, suggesting that this approach can be used to address the previous limitations associated with inefficient, lectin-based cellular capture.
2. EXPERIMENTAL SECTION 2.1. Materials. Triticum vulgare lectin (Wheat germ agglutinin) conjugated with Alexa Fluor 488 (WGA-A488; Invitrogen) and Lens culinaris lectin conjugated with texas red isothiocyanate (LcH-TRITC; EY laboratories) were diluted to 1 mg/mL concentrations in 1× PBS and stored at −20 °C. Bovine serum albumin (BSA) and all other chemicals were purchased from Sigma-Aldrich and used as received. Some 4-in. silicon (Si) wafers were purchased from Silicon Quest. A poly(glycidyl methacrylate)-block-poly(vinyldimethyl azlactone) copolymer (PGMA-b-PVDMA) having block lengths (number of monomer units in each block) of 56 and 175, respectively, was synthesized as described previously.24 Pseudomonas f luorescens GM30, a gram negative plant growth promoting rhizobacteria (PGPR) that plays a role in plant pathogen protection, was isolated from the Populus deltoides rhizosphere and used in this study as a model microbe.25,26 This microbe was stored in glycerol stocks at −80 °C until use. 2.2. Lithographic Patterning of PGMA-b-PVDMA and Si Control Scaffolds. Si wafers were patterned to contain arrays of circles 10 or 50 μm in diameter with a pitch (edge-to-edge distance) of 20 or 100 μm, respectively. Si wafers were first spin coated with ShinEtsu MicroSi MicroPrime P20 at 3000 rpm for 45 s, followed by S1818 (Microchem Corp.) positive resist at 3000 rpm for 45 s and baked on a hot plate at 115 °C for 60 s. Wafers were then exposed for 6 s with a contact mask aligner and developed in CD-26 for 2 min, rinsed with DI H2O, and dried with N2. Wafers were then treated with O2 plasma using a Tepla Ion Wave 10 plasma processing system to remove residual resist and diced into 10 × 10 mm substrates. Substrates were again treated with 3 min of O2 plasma to provide surface hydroxyl groups for reaction with epoxy groups present on the PGMA block. A 100 μL aliquot of a 1 wt % solution of PGMA56-b-PVDMA175 in anhydrous CHCl3 was then spin-coated over the substrate (Laurell WS-400B-6NPP/LITE) at 1500 rpm for 15 s and annealed at 110 °C under vacuum for 18 h to allow for microphase segregation and surface attachment.24 The polymer-coated substrates were sonicated in acetone for 5 min to remove the photoresist, rinsed in isopropanol, and dried with N2. Patterned substrates were stored under vacuum in a desiccator until use. 3743
dx.doi.org/10.1021/bm4011358 | Biomacromolecules 2013, 14, 3742−3748
Biomacromolecules
Article
Figure 1. The 10 μm diameter PGMA-b-PVDMA surface scaffolds. (A) SEM images of scaffold arrays and scaffold edges (inset), (B) AFM contact mode image, and (C) cross-sectional height profile across the center of a scaffold corresponding to the blue line in B. For the fabrication of Si control scaffolds without PGMA-bPVDMA, identical pattern arrays were used. Si wafers were spin-coated with Shin-Etsu MicroSi MicroPrime P20 at 3000 rpm for 45 s followed by JSR NFR-016D2 55 cP negative resist at 3000 rpm for 45 s, baked on a hot plate at 90 °C for 90 s. Wafers were then exposed for 3 s using a contact mask aligner, baked on a hot plate at 115 °C for 90 s, and developed in CD-26 developer for 1 min. Wafers were etched as previously described using an Oxford Plasmalab 100 reactive ion etching system to provide posts 1−2 μm in height.27 After etching, substrates were diced into 10 × 10 mm substrates and cleaned with Piranha solution (3:1 v/v H2SO4/30% H2O2 at 25 °C for 10 min) to remove residual resist material (Caution! strongly corrosive). No polymer coating was applied on the Si control scaffolds. 2.3. Protein Functionalization of Patterned Substrates. Substrates containing PGMA-b-PVDMA scaffolds or Si control scaffolds were contacted with 100 μL of 1 mg/mL solutions of WGA-A488, LcH-TRITC, or BSA in 1× PBS for varied incubation times in a humidified environment. After protein functionalization, substrates were washed for 5 min with a 0.05% solution of Tween 20 in 1× PBS and then stored in 1× PBS until further use. In control experiments where it was desired to block protein coupling to the PGMA-b-PVDMA scaffolds, substrates were submerged in a solution of 50 mM ethanolamine in 50 mM borate buffer, pH 8.5 for 18 h prior to protein incubation. These conditions should allow for the aminolysis of accessible azlactone groups.24 2.4. Bacterial Culture and Adhesion Conditions. Pseudomonas f luorescens GM30 was used in this study. Microbes were cultured on tryptone yeast (TY) agar plates (10 g tryptone, 5 g yeast extract, 15 g agar per liter) for 24−48 h at 28 °C and then maintained at ambient conditions for up to one week. For growth in liquid, a single colony was used to inoculate liquid TY media in sterile 20 mL glass tubes. The microbes were grown to logarithmic phase in a shaking incubator set at 28 °C and 200 rpm, harvested by centrifugation, and resuspended in 1× PBS at an O.D600 of 0.1. A 100 μL aliquot of the microbe solution was incubated onto protein functionalized substrates in a humidified environment for 1 h under gentle rocking. To remove unattached microbes, substrates were washed for 5 min with a 0.05% solution of Tween 20 in 1× PBS. The remaining adherent microbes were then chemically fixed to the surface by incubating the substrate in a 2.5% solution of glutaraldehyde in H2O for 2 min and then rinsed in H2O. Finally, the substrates were dried by gently aspirating water off the surface. 2.5. Solution-Phase Lectin Binding Assay. To investigate the solution-phase binding of lectins, 1 mL of the washed microbe solution from Section 2.4 (OD600 = 0.1) was incubated with 10 μg/mL quantities of WGA-A488 and LcH-TRITC for 1 h at 25 °C, washed by centrifugation to remove unattached lectins, and then resuspended back to an OD600 = 0.1. A 10 μL aliquot of this solution was then placed between a 75 × 25 mm glass slide and a 20 × 20 mm coverslip and imaged. 2.6. Instrumentation. Brightfield and Fluorescence Microscopy. All images of substrates containing arrays of PGMA-b-PVDMA or Si control scaffolds were taken in brightfield or fluorescence (20×, NA
0.40/50×, NA 0.50/100×, NA 0.95) with an upright microscope (BX51, Olympus). For characterizations of microbial adhesion levels on the substrates, 20−30 representative images of each substrate were taken near the center of the 10 × 10 mm substrates. Atomic Force Microscopy (AFM). A Park Systems atomic force microscope was used to image the PGMA-b-PVDMA scaffolds. NanoWorld PNP-TR B cantilevers with a resonance frequency of 17 kHz and a force constant of 0.08 N m−1 were used in contact mode. Scan areas were varied between 4 and 1600 μm2. Images were acquired at a scan rate of 0.2−0.6 Hz. Scaffold heights and surface roughness were analyzed using the Park Systems built-in analysis software. Scanning Electron Microscopy (SEM). All SEM images of PGMAb-PVDMA scaffolds were taken using a Carl Zeiss Merlin SEM instrument operating at 1.7 kV. Samples were not pretreated with a conducting layer, and charge compensation was used during sample imaging. Fourier Transform Infrared Spectroscopy (FT-IR). Unpatterned polymer films on Si substrates were characterized with ATR-FTIR using a Bruker Optics Vertex 70 spectrometer with a Harrick Scientific VariGATR accessory and a narrow-band MCT detector. Spectra were analyzed using OPUS software. A background spectra of 128 scans of the clean germanium crystal was first collected. Spectra of the polymer films were acquired using 512 scans. The crystal was cleaned with methanol prior to measurement of each sample. All spectra were background-subtracted and baseline-corrected. 2.7. Image Analysis. All images were analyzed using ImageJ software. Fluorescent signal intensities for WGA-A488 functionalized PGMA-b-PVDMA scaffolds were averaged across 100 replicate scaffolds. Microbial adhesion levels on PGMA-b-PVDMA or Si control scaffolds were analyzed using image thresholding with the triangle method. To quantify the microbial growth from the scaffold exterior, the increase in scaffold area with microbial deposition was specifically measured (see Supporting Information, Figure S1). A total of 200 representative surface scaffolds were measured under each condition investigated. Microbes present in unpatterned regions that were not attached to the scaffolds were not included in the measurement. 2.8. Statistical Analysis of Data. All data are reported as the mean ± standard deviation from replicate measurements. The Statistical Analysis Toolbox in MATLAB was used for all statistical tests. Differences between microbial surface coverage on different substrates was identified using the Student’s t-test (p < 0.01).
3. RESULTS AND DISCUSSION 3.1. Formation and Characterization of PGMA-bPVDMA Scaffolds. The lithographic patterning of PGMA-bPVDMA onto oxidized Si surfaces results in well-defined, threedimensional copolymer films of 10 μm diameter (referred to as “PGMA-b-PVDMA scaffolds”). All PGMA-b-PVDMA scaffolds remained attached to the substrate through each processing step, including sonication in organic solvents and washing in aqueous solvents, providing a robust platform for surface 3744
dx.doi.org/10.1021/bm4011358 | Biomacromolecules 2013, 14, 3742−3748
Biomacromolecules
Article
modifications. Figure 1 shows SEM and AFM images of the PGMA-b-PVDMA scaffolds. As evident in Figure 1A, the interior of the scaffold appears smooth, while the outer edges are of a higher surface roughness. AFM images and data in Figure 1B and C show varied internal film thickness with respect to radial direction resulting from patterning using the liftoff process. At the scaffold center, the average film thickness is 115 ± 32 nm which approaches the expected thickness for a single layer of a PGMA-b-PVDMA film.24 Moving from the center toward the exterior of the scaffold, the film thickness increases nonlinearly and is greatest at the edges where the maximum polymer thickness ranged between 500 nm to 2 μm. The formation of stable polymer films at higher thicknesses (greater than 90 nm) is influenced by the height of the templating photoresist and is likely stabilized by the cross-linking of GMA blocks away from the substrate interface, which is favored using an annealing temperature of 110 °C. Annealing at temperatures
