Langmuir 2006, 22, 10889-10892
10889
Micropatterning of Cell-Repellent Polymer on a Glass Substrate for the Highly Resolved Virus Microarray Kyunga Na,† Jaeyeon Jung,† Byungcheol Shin,‡ and Jinho Hyun*,† Department of Biosystems and Biomaterials Science and Engineering, Seoul National UniVersity, Seoul 151-742, Korea, and AdVanced Materials DiVision, Korea Research Institute of Chemical Technology, Daejeon 305-600, Korea ReceiVed July 31, 2006. In Final Form: October 28, 2006 The development of a simple and easily accessible method to control cellular behavior under a spatially controlled surface is critical for fundamental studies in biotechnology. We fabricated a microarray of Spodoptera frugiperda 9 (Sf9) cells on a glass surface by microcontact printing cell-repellent polymeric molecules of poly(ethylene glycol)branched-poly(methyl methacrylate) as a template for cell micropatterning. The polymer micropatterns enabled the stable confinement of Sf9 cells on the surface, resulting in the formation of a cell microarray. Subsequently, the patterned Sf9 cells were infected with recombinant baculovirus modified with green fluorescent protein (GFP) to form a virus microarray, and GFP expression in the virus microarray was verified with confocal fluorescence microscopy.
Introduction Living-cell microarrays have been studied for applications in both cell-based sensors and drug discovery.1,2 A spatially welldesigned surface is required for the control of cell adhesion and cell spreading. Microfabrication techniques, combined with surface chemistry, have provided chemical structures on array surfaces capable of controlling the interaction between the cell and substrate.3,4 To generate patterned chemical structures on array surfaces, photolithographic, dip-pen lithographic, and soft lithographic techniques have been adopted. A soft lithographic technique, known as microcontact printing (µCP), is an especially attractive method for patterning cells and biomolecules on the substrate at micrometer-scale resolutions because of its simplicity, flexibility, and cost effectiveness.5 Because of these advantages, microstructures formed by µCP have been widely used to spatially control the adhesion and proliferation of biological species. Protein-resistant surfaces have been reported for applications involving blood-contacting devices,6 contact lenses,7 and a noninteractive background for biodiagnostic surfaces8 designed to elicit specific binding using several types of molecules, such as agarose,9 mannitol,10 albumin,11 and molecules including poly(ethylene glycol) (PEG).12 Among the aforementioned methods, PEG chemistry-based approaches have been the most thoroughly investigated because of their simplicity and effectiveness in * Corresponding author. E-mail:
[email protected]. † Seoul National University. ‡ Korea Research Institute of Chemical Technology. (1) Park, T. H.; Shuler, M. L. Biotechnol. Prog. 2003, 19, 243-253. (2) Itle, L. J.; Pishko, M. V. Anal. Chem. 2005, 77, 7887-7893. (3) Nagamine, K.; Onodera, S.; Torisawa, Y. S.; Yashkawa, T.; Shiku, H.; Matsue, T. Anal. Chem. 2005, 77, 4278-4281. (4) Hyun, J. H.; Ma, H. W.; Zhang, Z. P.; Beebe, T. P.; Chilkoti, A. AdV. Mater. 2003, 15, 576-579. (5) Koh, W. G.; Itle, L. J.; Pishko, M. V. Anal. Chem. 2003, 75, 5783-5789. (6) Tanaka, M.; Motomura, T.; Kawada, M.; Anzai, T.; Kasori, Y.; Shiroya, T.; Shimura, K.; Onishi, M.; Mochizuki, A. Biomaterials 2000, 21, 1471-1481. (7) Willis, S. L.; Court, J. L.; Redman, R. P.; Wang, J.-H.; Leppard, S. W.; O’Byrne, V. J.; Small, S. A.; Lewis, A. L.; Jones, S. A.; Stratford, P. W. Biomaterials 2001, 22, 3261-3272. (8) Dai, J.; Baker, G. L.; Bruening, M. L. Anal. Chem. 2006, 78, 135-140. (9) Nelson, C. M.; Chen, C. S. FEBS Lett. 2002, 514, 238-242. (10) Luk, Y.-Y.; Kato, M.; Mrksich, M. Langmuir 2000, 16, 9604-9608. (11) Ostuni, E.; Kane, R.; Chen, C. S.; Ingber, D. E.; Whitesides, G. M. Langmuir 2000, 16, 7811-7819. (12) Pasche, S.; De Paul, S. M.; Voros, J.; Spencer, N. D.; Textor, M. Langmuir 2003, 19, 9216-9225.
resisting the absorption of biomolecules through the hydrophilicity and characteristic movement of PEG chains.13,14 A variety of methods, including physisorption, chemisorption, covalent grafting of a polymer, and plasma glow discharge treatment, have been used to incorporate PEG at the surface by µCP in order to spatially localize biomolecules with well-controlled microstructures. Most of these methods, however, require multiple processing steps that must be optimized for the substrate of interest and hence do not offer a generic route for the surface micropatterning with PEG. PEG-b-PMMA used in the experiment is attractive because it can be bound to diverse substrates by a simple and generic one-step procedure to prevent the nonspecific adsorption of biomolecules providing a physicochemical barrier to cell attachment. In this letter, we describe a simple method of creating stable, highly resolved micropatterns of a PEGfunctionalized polymer on a glass surface, which has not been applied as a substrate for cell microarrays by direct µCP of amphiphilic polymers. The patterning of PEG microwells creates a highly ordered biointerface with modulating-cell or protein-repellent properties, and cells remain effectively isolated in the individual PEG microwell.15 The PEG length and graft density are important for the appropriate properties of nonfouling surfaces. The density of grafted chains must be sufficient to prevent proteins from diffusing through the grafted PEG layer and reaching the underlying substrate, onto which they can adsorb irreversibly. In addition, proteins must not adsorb on top of the PEG layer; such an absorption would, consequently, reduce the mobility of the hydrated grafted PEG chains.16 An antifouling effect can also be caused by osmotic and entropic repulsion as well as by interfacial charge screening.17 The PEG branched poly(methyl methacrylate) (PEG-b-PMMA) used in this study is an amphiphilic, comblike polymer consisting of a hydrophobic PMMA (13) Schlapak, R.; Pammer, P.; Armitage, D.; Zhu, R.; Hinterdorfer, P.; Vaupel, M.; Fruhwirth, T.; Howorka, S. Langmuir 2006, 22, 277-285. (14) Hong, B. J.; Oh, S. J.; Youn, T. O.; Kwon, S. H.; Park, J. W. Langmuir 2005, 21, 4257-4261. (15) Revzin, A.; Tompkins, R. G.; Toner, M. Langmuir 2003, 19, 9855-9862. (16) Pasche, S.; Textor, M.; Meagher, L.; Spencer, N. D.; Griesser, H. J. Langmuir 2005, 21, 6508-6520. (17) Schuler, M.; Owen, G. R.; Hamilton, D. W.; de Wild, M.; Textor, M.; Brunette, D. M.; Tosatti, S. G. P. Biomaterials 2006, 27, 4003-4015.
10.1021/la0622469 CCC: $33.50 © 2006 American Chemical Society Published on Web 11/15/2006
10890 Langmuir, Vol. 22, No. 26, 2006
Letters
Figure 1. Cell micropatterns on glass: (A) fabrication of the PDMS stamp, (B) the PDMS stamp, (C) PDMS with the comb polymer, (D) the micropattern of the comb polymer on the surface, and (E) the cell microarray.
backbone and hydrophilic PEG side chains that significantly reduce the nonspecific adsorption of biomolecules from the media.18,19 This letter demonstrates a simpler and more reliable PEG grafting process than has previously been reported in studies of other grafting methods. Sf9 cells were micropatterned on the surface and infected with recombinant baculovirus to demonstrate the utility of this grafting process for use in phenotype-based cell microarrays or in a viral plaque assay. This simple method can spatially control the adhesion of cells on a localized place and has the potential to be a useful biotechnical tool for the fabrication of cell-based sensing devices. Materials and Methods Synthesis of PEG-b-PMMA. PEG-b-PMMA was prepared from methyl methacrylate (MMA) (Aldrich), poly(ethylene glycol)methacrylate (POEM, Aldrich, Mn ≈ 526 g/mol, corresponding to n ≈ 10), and poly(ethylene glycol)methyl ether methacrylate(HPOEM, Aldrich, Mn ≈ 475 g/mol, corresponding to n ≈ 8.5) by free radical polymerization. The composition of the comb polymer was determined by 1H NMR in CDCl3: 4.14 ppm (-C(dO)-OCH2-), 3.6-3.8 ppm (-C(dO)-O-CH3 and -CH2-CH2-O-), 0.5-2 ppm(CH2-C-(CH3)-), and 3.39 ppm (-OCH3). The terpolymer was composed of 71 wt % MMA, 12 wt % HPOEM, and 17 wt % POEM. The molecular weight of the comb polymer was ∼30 000 Da with a polydispersity of ∼2.3, as measured by gel permeation chromatography. PDMS Mold Fabrication and Micropatterning. The master for the elastomeric mold was fabricated from polished Si wafers by spin coating the photoresist (AZ 1512, Clariant, Inc) to a thickness of ∼2 µm, followed by processing with contact photolithography. The elastomeric stamp was fabricated by casting polydimethyl siloxane (PDMS) against the photoresist on a silicon master with 40 µm features. The PDMS stamp was oxidized for 20 s using a commercially available hydrophilizer before µCP. The thin films of PEG-b-PMMA were spun cast from a 50/50 (v/v) mixture of H2O/ ethanol onto the oxidized PDMS stamp at 2000 rmp for 2 min. The stamp was brought into conformal contact with glass for 1 min, resulting in the transfer of PEG-b-PMMA to the glass surface that was in contact with the PDMS stamp. Atomic Force Microscope (AFM) Characterization of the Surface Pattern. AFM topographic images were collected in air using a MultiMode atomic force microscope (Digital Instruments, Santa Barbara, CA) in contact mode with V-shaped silicon nitride cantilevers (Nanoprobe, Veeco; spring constant 0.12 N/m; tip radius 20-60 nm). The typical observation of the rms noise of force was about 20 pN, which is in good agreement with the estimated thermal force fluctuations of 18 pN. GFP Vector Construction and Transfection. The enhanced green fluorescent protein (EGFP) gene was obtained by polymerase (18) Patel, S.; Thakar, R. G.; Wong, J.; McLeod, S. D.; Li, S. Biomaterials 2006, 27, 2890-2897. (19) Irvine, D. J.; Mayes, A. M. Biomarcromolecules 2001, 2, 85-94.
chain reaction (PCR) amplification of the pEGFP vector (Clontech.). The pBlue-BacHis2 baculovirus transfer vector (pB) (Invitrogen) was used as the cloning vector. pB-GFP was constructed by ligation of the EGFP gene into the pB baculovirus transfer vector. The EGFP gene sequence was confirmed by DAN sequence analysis. Expression of the construct was controlled by the Autographa California (Ac) MNPV polyhedron promoter. The recombinant baculovirus was obtained by co-transfecting pB-GFP with AcMNPV DNA. Cell Culture and Fabrication of Virus Microarray. Sf9 cells were maintained at 27 °C in TC-100, supplemented with 10% fetal bovine serum (Invitrogen Life Technologies), 100 U/mL penicillin, 100 µm/mL streptomycin, and 0.25 µm/mL amphotericin B. Cells were plated on the micropatterned surface at a density of 106 cells/ mL. The cells were incubated for 1 h and then gently rinsed with culture media. The cells seeded inside the micropatterns were infected with 0.5 mL of a virus solution (2 × 107 pfu/mL)) diluted to 10-7 pfu/mL for 3 h. The virus inoculums were removed from the cells, and 4 mL of 2% agarose with a tissue culture medium was overlaid on the surface. After hardening of the agarose overlay, it was incubated for 4 or 5 days at 27 °C. The cells were imaged under phase-contrast microscopy or confocal fluorescence microscopy.
Results and Discussion The controlled spatial distribution of cells is a prerequisite for the development of cell-based devices for applications such as high-throughput screening and cell-based biochips. µCP was routinely used for the spatial control of biomolecules because of its simplicity and accessibility. In this letter, µCP was adopted for the fabrication of a nonfouling polymeric template, and Sf9 cells were seeded in the micropatterns of the template. To form the cell microarray, PEG-b-PMMA was micropatterned on glass using a PDMS stamp. The micropatterning procedure is illustrated in Figure 1. First, to prepare the PDMS mold, a 10:1 mixture of silicon elastomer and the curing agent was poured onto the master template, which was fabricated from a polished Si wafer by spin coating, and cured at 70 °C for 1 h. The PDMS mold was inked with PEG-b-PMMA and dried for 30 s, followed by imprinting on glass. PEG-b-PMMA was easily imprinted on most surfaces, including, silicon, gold, and polymer, because of its amphiphilicity. Generally, a conventional method for cell micropatterning consists of a biofuctionalization and passivation step. For example, Oliva et al. patterned protein A through conventional µCP on substrates.20 The patterns were functionalized with a protein construct consisting of the extracellular domain of molecule L1 recombinantly linked to the Fc fragment of immunoglobulins (IgG) via the selective binding of protein A to the Fc fragment of IgG. Upon backfilling the surface (20) Oliva, A. A.; Kingman, C. D.; Craighead, H. G.; Banker, G. A. Neurochem. Res. 2003, 28, 1639-1648.
Letters
Langmuir, Vol. 22, No. 26, 2006 10891
Figure 2. PEG-b-PMMA micropatterned on the glass: (A) an AFM image of 1%, (B) the line profile of A, and (C) the effect of polymer concentration on the thickness of micropatterns. Figure 3. Optical image of the cell micropattern: (A) a micropattern with the comb polymer on the surface and (B) a cell microarray on the micropattern surface with the comb polymer.
with poly-L-lysine and plating neurons onto the surface, they observed selective axon growth over the patterns.21 In this letter, we used glass plates for cell adhesion that are similar to those used for the protein construct and created micropatterns of PEG-b-PMMA directly on the surface without additional surface modification, which made the process of cell micropatterning simple. The thickness of PEG-b-PMMA micropatterns was measured by AFM, as shown in Figure 2. The height at the edge of the micropatterned PEG-b-PMMA depends on the concentration of the solution. The increase in PEG-bPMMA concentration from 0.01 to 8% caused an increase in the micropattern height from 52 to 520 nm. In the cell seeding experiment, the PEG-b-PMMA micropatterns, formed at a concentration of less than 1%, did not provide a barrier to distribute the cells spatially, but concentrations above 1% showed distinctive chemophysical barriers during cell adhesion. The lack of spatial distribution in concentrations of less than 1% resulted because the Sf9 cells are not anchorage-based, and the physical barrier of the micropatterns should be high enough to confine the cells inside the patterned region. We postulate that physical barrier heights of ∼150 nm are critical in micropatterning Sf9 cells, and further investigation of this requirement is in progress. The concentration of the inking of the PEG-b-PMMA solution was subsequently fixed at 1% on the basis of our results. The optical microscopy image of PEG-b-PMMA micropatterning is shown in Figure 3A. The pattern size depends on the resolution of a silicon master for the preparation of PDMS molds, and we routinely created features from 0.5 to 200 µm by µCP. Several types of features, such as squares, circles, and stripes, were able to be created, but we chose the micropatterns of circle features that were 40 µm in diameter and 150 nm in height for the proper seeding of insect cells that were about 15 µm in diameter. The micropattern of PEG-b-PMMA created cell-adhesive and -repellent regions. For cell seeding, circle features 40 µm in diameter were micropatterned on the glass surface, which was cleaned with ethanol prior to micropatterning. The PEG-b-PMMA features were verified using phase-contrast optical microscopy before cell seeding. The PEG-b-PMMA pattern was stable in culture media for over 2 months because amphiphilic polymers are based on hydrophobic and hydrophilic structures, which are insoluble
in aqueous solutions, such as culture media, and because the methacrylate backbone of the comb polymer could be adsorbed onto the surface.4,19 PEG is biologically inert and has been approved by the FDA for use in containers for foods, cosmetics, and pharmaceuticals. PEG is a biocompatible hydrophilic polymer that has been widely used not only for modifying implant surfaces but also for other clinical applications such as advanced drug delivery and tissue engineering as a result of its ability to resist cell and protein adhesion. Cell adhesion studies on a range of PEG-modified polymer substrates revealed that even the addition of a very small amount of PEG to the modifying solution results in a dramatic reduction of cell attachment. Because of PEG side chains, PEG-b-PMMA resists the adhesion of the adhesive cells and suppresses nonspecific interactions between the surface and the protein-containing media. PEG side chains have a brushlike structure and are highly hydrated in aqueous solution, which decreases the attractive force between solid surfaces and proteins.22 The glass micropatterned with PEG-b-PMMA can be considered to be a plate with micrometer-scale wells that can be used to contain a specific number of Sf9 cells based on the size of a microwell (Figure 3A). We were able to routinely fabricate a microarray of insect Sf9 cells, a cell line derived from the lepidopteran insect Spodoptera frugiperda. These cells were selected because of their regular size and shape as well as their broad application in virologic and biomedical experiments. Moreover, unlike mammalian cells, Sf9 cells are relatively easy to maintain equally well in suspension and on surfaces as they grow; they do not require CO2 and can easily withstand temperature fluctuation. Sf9 (106 cells/mL) was incubated on the glass surface micropatterned with PEG-b-PMMA for 1 h. After the incubation, the glass surface was gently rinsed to remove loosely adherent cells and observed using phase-contrast optical microscopy. Figure 4B shows the well-localized Sf9 cell microarray created on the glass surface. The number of cells adhered inside a micropattern ranged from 4 to 8 cells with 40
(21) Tsourapas, G.; Rutten, F. J. M.; Briggs, D.; Davies, M. C.; Shakesheff, K. M. Appl. Surf. Sci. 2006, 15.
(22) Mu¨ller, M. T.; Yan, X.; Lee, S.; Perry, S. S.; Spencer, N. D. Macromolecules 2005, 38, 3861-3866.
10892 Langmuir, Vol. 22, No. 26, 2006
Letters
103 to 106 cells/mL with the same features on glass. Sf9 cells are commonly used as a host for the expression of baculovirus recombinant protein. For infection with baculovirus, culture media was removed, and then recombinant baculovirus, including pBGFP, was added to micropatterned Sf9 cells. After incubation for 3 h, the microarray was covered with soft agar to minimize the movement of cells and the diffusion of virus particles to the media. Green fluorescence in the microarray was observed by confocal fluorescence microscopy, confirming the expression of specific protein from the modified virus in the infected cells (Figure 4).
Conclusions This letter describes a simple method for the fabrication of an Sf9 cell microarray using soft lithographic stamping. The cells were efficiently seeded in the micropatterns because the micropatterns of cell-resistant PEG-b-PMMA provided chemophysical barriers for cell attachment. A GFP vector of modified baculovirus was successfully transfected into the cells micropatterned on the glass, and expression was confirmed by confocal fluorescence microscopy. This approach will be applicable for use in cell-based microarrays in high-throughput functional genomics as well as in the purification of a viral plaque from a mixture of different viruses.
Figure 4. Confocal fluorescence images of GFP expression in cells: (A) low magnification (×100) and (B) high magnification (×400).
µm features (Figure 3B). The cell number inside a feature also ranged from 1 to 5 by adjusting the cell seeding density from
Acknowledgment. This work was supported by grant no. R01-2006-000-10217-0 from the Basic Research Program of the Korea Science & Engineering Foundation. J.H. acknowledges that this work was partially supported by the Korea Ministry of Commerce, Industry and Energy. We thank Professor Y. Je for the valuable discussion of viral experiments. LA0622469
