Controllable Construction of Carbohydrate Microarrays by Site

Feb 8, 2010 - (1, 2) They serve as sites for docking other cells, biomacromolecules, and pathogens in a .... the surface morphology of films after bei...
0 downloads 0 Views 4MB Size
Download PDF
pubs.acs.org/Langmuir © 2010 American Chemical Society

Controllable Construction of Carbohydrate Microarrays by Site-Directed Grafting on Self-Organized Porous Films Bei-Bei Ke, Ling-Shu Wan,* and Zhi-Kang Xu Key Laboratory of Macromolecular Synthesis and Functionalization (Ministry of Education), Department of Polymer Science and Engineering, State Key Laboratory of Chemical Engineering, Zhejiang University, Hangzhou 310027, China Received December 16, 2009. Revised Manuscript Received January 13, 2010 Carbohydrate-protein interactions are critical in many biological processes. However, the interactions between individual carbohydrates and proteins are often of low affinity and difficult to study. Recent development of carbohydrate microarrays provides an effective tool to explore the interaction. In this work, carbohydrate microarrays were controllably constructed by grafting of a carbohydrate-containing monomer on self-organized honeycombpatterned films. The films were prepared from an amphiphilic block copolymer, poly(styrene-block-(2-hydroxyethyl methacrylate)), by a breath figure method. Three-dimensional fluorescence results demonstrate that the hydroxyl groups aggregate mainly inside the pores, which afford a chance of site-directed surface modification. 2-(2,3,4,6-TetraO-acetyl-β-D-glucosyloxy)ethyl methacrylate was selectively grafted in the pores by a surface-initiated atom transfer radical polymerization. Characterization by attenuated total reflectance Fourier transform infrared spectroscopy, X-ray photoelectron spectroscopy, scanning electron microscopy, atomic force microscopy, and contact angle measurements confirms the site-directed growth of the glycopolymer chains. Further specific recognition of the carbohydrate microarrays to lectin (concanavalin A) leads to an organized microarray of protein, and hence this approach also opens a new route to fabricating other functional microarrays such as protein-patterned surfaces.

Introduction Carbohydrates are found on the surface of nearly every cell in the forms of polysaccharides and glycoconjugates (e.g., glycoproteins, glycopeptides, and glycolipids).1,2 They serve as sites for docking other cells, biomacromolecules, and pathogens in a specific recognition process, which is eventually triggered by carbohydrate-protein interaction. The interaction has been widely recognized to be the key in a variety of biological processes and the first step in numerous phenomena based on cell-cell interactions, such as blood coagulation, immune response, viral infection, inflammation, embryogenesis, and cellular signal transfer. As we know, monosaccharide ligands can only weakly bind to protein receptors,3 whereas multivalent interaction can greatly enhance the carbohydrate-protein binding strength, which is normally accepted as the “cluster glycoside effect”.4 Therefore, carbohydrate microarrays that consist of immobilized saccharides with proper density and orientation are very useful for elucidating the recognition events between carbohydrates and proteins at a molecular level.5 Furthermore, carbohydrate microarrays can interact with many kinds of proteins at once, offering a highthroughput approach for the study of carbohydrate-protein interaction.6 Accordingly, the preparation of carbohydrate microarrays has received considerable interest with the development of glycomics. The most common method to fabricate carbohydrate *Corresponding author: e-mail [email protected]; Tel þ86-571-87952605.

(1) Gruner, S. A. W.; Locardi, E.; Lohof, E.; Kessler, H. Chem. Rev. 2002, 102, 491–514. (2) Wang, Q.; Dordick, J. S.; Linhardt, R. J. Chem. Mater. 2002, 14, 3232–3244. (3) Nagahori, N.; Nishimura, S. I. Biomacromolecules 2001, 2, 22–24. (4) Lundquist, J. J.; Toone, E. J. Chem. Rev. 2002, 102, 555–578. (5) Park, S. J.; Shin, I. J. Angew. Chem., Int. Ed. 2002, 41, 3180–3182. (6) Wang, D. N.; Liu, S. Y.; Trummer, B. J.; Deng, C.; Wang, A. L. Nat. Biotechnol. 2002, 20, 275–281. (7) Miura, Y.; Sato, H.; Ikeda, T.; Sugimura, H.; Takai, O.; Kobayashi, K. Biomacromolecules 2004, 5, 1708–1713.

8946 DOI: 10.1021/la904729b

microarrays is based on lithography techniques including photolithography,7 inkjet printing,5 and microcontact printing.8 Recently, it has been reported that honeycomb-patterned films can be prepared by a breath figure method, which is based on evaporative cooling and subsequent water-droplet templating to form an ordered array of breath figures.9,10 Some block copolymers and homopolymers have been used to prepare exquisite honeycomb-patterned films with controlled sizes and separated distances.11-16 This water-templating technique has many advantages over the lithography methods: it is simple and cost-saving, and no extra steps are needed to remove the water templates. Therefore, the breath figure method as an alternative approach is much promising for the preparation of microlenses,17 superhydrophobic materials,18 and tissue engineering scaffolds.19,20 Most recently, honeycomb-patterned films were utilized to fabricate protein microarrays by attaching proteins into the pores with high functionality.21,22 In contrast to protein (8) Michel, O.; Ravoo, B. J. Langmuir 2008, 24, 12116–12118. (9) Bunz, U. H. F. Adv. Mater. 2006, 18, 973–989. (10) Stenzel, M. H.; Barner-Kowollik, C.; Davis, T. P. J. Polym. Sci., Part A: Polym. Chem. 2006, 44, 2363–2375. (11) Liu, C. H.; Gao, C.; Yan, D. Y. Angew. Chem., Int. Ed. 2007, 46, 4128–4131. (12) Sun, W.; Ji, J.; Shen, J. C. Langmuir 2008, 24, 11338–11341. (13) Peng, J.; Han, Y. C.; Yang, Y. M.; Li, B. Y. Polymer 2004, 45, 447–452. (14) Wong, K. H.; Davis, T. P.; Bamer-Kowollik, C.; Stenzel, M. H. Polymer 2007, 48, 4950–4965. (15) Jenekhe, S. A.; Chen, X. L. Science 1999, 283, 372–375. (16) Yu, C. L.; Zhai, J.; Li, Z.; Wan, M. X.; Gao, M. Y.; Jiang, L. Thin Solid Films 2008, 516, 5107–5110. (17) Yabu, H.; Shimomura, M. Langmuir 2005, 21, 1709–1711. (18) Yabu, H.; Takebayashi, M.; Tanaka, M.; Shimomura, M. Langmuir 2005, 21, 3235–3237. (19) Sunami, H.; Ito, E.; Tanaka, M.; Yamamoto, S.; Shimomura, M. Colloids Surf., A 2006, 284, 548–551. (20) Nishikawa, T.; Nishida, J.; Ookura, R.; Nishimura, S. I.; Wada, S.; Karino, T.; Shimomura, M. Mater. Sci. Eng., C 1999, 10, 141–146. (21) Min, E.; Wong, K. H.; Stenzel, M. H. Adv. Mater. 2008, 20, 3550–3556. (22) Zhang, Y.; Wang, C. Adv. Mater. 2007, 19, 913–916.

Published on Web 02/08/2010

Langmuir 2010, 26(11), 8946–8952

Ke et al.

Article

microarrays, carbohydrate microarrays are unique since the density of saccharide residues will remarkably affect their functions.23,24 Generally the saccharide residues should reach a high density that ensures the “cluster glycoside effect”. Therefore, it is a challenge to fabricate carbohydrate microarrays with adjustable density of saccharide residues. Previously, honeycomb-patterned films were directly fabricated by casting copolymers containing carbohydrate moieties in humid environment.25-27 However, the protein recognition results showed that the saccharide residues were randomly distributed in the films, leading to a carbohydrate network but not microarrays.27 In this work, we propose a controllable approach to carbohydrate microarrays with the combination of the breath figure method and surface-initiated atom transfer radical polymerization (ATRP). Amphiphilic block copolymers were adopted to generate breath figures with hydrophilic functional groups aggregated mainly inside the pores, which affords a chance of site-directed surface grafting. Then, through surface-initiated ATRP glycopolymers were selectively grafted inside the pores but not on the top surface of the films. This strategy can be used not only for the preparation of carbohydrate microarrays but also for a range of other functional microarrays.

Experimental Section Materials. Poly(styrene-block-(2-hydroxyethyl methacrylate)), PS187-b-PHEMA15 (Mn =20 500 g/mol, MWD=1.10), was synthesized by ATRP using a reported procedure in chlorobenzene solution.28 Polystyrene (PS, Mn =235 000 g/mol, MWD=2.89) was provided by Zhenjiang Chiemei Chemicals. The carbohydrate-containing monomer, 2-(2,3,4,6-tetra-O-acetyl-β-D-glucosyloxy)ethyl methacrylate (AcGEMA), was synthesized using a procedure described previously.29 5-Aminofluorescein (5-AF), N,N,N0 ,N00 ,N00 -pentamethyldiethylenetriamine (PMDETA), and 2-bromoisobutyryl bromide were used as received from Aldrich. 2-[4-(2-Hydroxyethyl)-1-piperazinyl]ethanesulfonic acid (HEPES) was a commercial product of Nanjing Robiot (China) and was used as received. CuBr was purified by subsequently washing with acetic acid, methanol, and drying under reduced pressure. CuBr2 (99.9%) was a commercial product and was used without further purification. Triethylamine (TEA) was purified by distillation. Poly(ethylene terephthalate) (PET) film was kindly provided by Hangzhou Tape Factory and cleaned with acetone for 2 h before use. Fluorescein-labeled concanavalin A (FL-Con A) and peanut agglutinin (FL-PNA) (Vector) were used as received. Water used in all experiments was deionized and ultrafiltrated to 18 MΩ with an ELGA LabWater system. All other reagents were analytical grade and used without further purification. Preparation of Honeycomb-Patterned Films. The films were cast by the breath figure method. In a typical experiment, PS-b-PHEMA was dissolved in carbon disulfide and PS was dissolved in chloroform with a concentration of 5 mg/mL. An aliquot of 100 μL for each polymer solution was drop-cast onto a PET substrate placed under a 2 L/min humid airflow. The (23) Kiessling, L. L.; Pohl, N. L. Chem. Biol. 1996, 3, 71–77. (24) Seeberger, P. H.; Werz, D. B. Nature 2007, 446, 1046–1051. (25) Stenzel, M. H.; Davis, T. P.; Fane, A. G. J. Mater. Chem. 2003, 13, 2090– 2097. (26) Nishikawa, T.; Nishida, J.; Ookura, R.; Nishimura, S. I.; Scheumann, V.; Zizlsperger, M.; Lawall, R.; Knoll, W.; Shimomura, M. Langmuir 2000, 16, 1337– 1342. (27) Nishida, J.; Nishikawa, K. A.; Nishimura, S.; Wada, S.; Karino, T.; Nishikawa, T.; Ijiro, K.; Shimomura, M. Polym. J. 2002, 34, 166–174. (28) Wang, T. L.; Liu, Y. Z.; Jeng, B. C.; Cai, Y. C. J. Polym. Res. 2005, 12, 67– 75. (29) Fleming, C.; Maldjian, A.; Da Costa, D.; Rullay, A. K.; Haddleton, D. M.; John, J. S.; Penny, P.; Noble, R. C.; Cameron, N. R.; Davis, B. G. Nat. Chem. Biol. 2005, 1, 270–274.

Langmuir 2010, 26(11), 8946–8952

Figure 1. FESEM images of honeycomb-patterned films prepared from 5 mg/mL PS-b-PHEMA solutions in CS2 at 25 °C and RH 80% with a 2 L/min airflow: (a) before and (b) after the removal of the top layer of the film using adhesive tape. humidity of the airflow was maintained to be above 60%. After solidification, the film was dried at room temperature.

Suborder Arrangement of the Amphiphilic Copolymer. Rearrangement of PHEMA blocks was evaluated by confocal laser scanning microscopy (CLSM). The honeycomb films were treated with 0.05 g of p-benzoquinone dissolved in 5 mL of 20% ethanol aqueous for 2 h at room temperature. The activated films were successively washed with 20% ethanol and water, incubated in a 10 μg/mL 5-AF solution (20% ethanol as solvent) for 3 h, washed with water, and then observed using CLSM. CLSM was performed on a Leica TCS SP5 confocal setup mounted on a Leica DMI 6000 CS inverted microscope (Leica Microsystems, Wetzlar, Germany) and was operated under the Leica Application Suite Advanced Fluorescence (LAS AF) program.

Surface-Initiated ATRP on the Honeycomb-Patterned Films. Immobilization of ATRP initiators on PS-b-PHEMA film surface was achieved by the reaction between 2-bromoisobutyryl bromide and the hydroxyl groups. PS-b-PHEMA films were immersed in 10 mL of n-heptane at 0 °C, and 0.12 mL of dry TEA (0.86 mmol) was added. Then 0.1 mL (0.81 mmol) of 2-bromoisobutyryl bromide dissolved in 10 mL of n-heptane was added dropwise under a nitrogen atmosphere. The mixture was stirred for another 3 h at 0 °C, followed by stirring at room temperature for 12 h. After the reaction, PS-b-PHEMA films were thoroughly washed with methanol-water-methanol in sequence and dried under reduced pressure at 40 °C. In a typical surfaceinitiated ATRP experiment, a piece of Br-immobilized film and 9 mg of CuBr (0.063 mmol) were added to a 50 mL Schlenk flask. The flask was evacuated and backfilled with N2 three times. AcGEMA (0.46 g, 1 mmol), PMDETA (14.6 μL, 0.070 mmol), and CuBr2 (1.4 mg, 0.006 mmol) were dissolved in methanol (5 mL) and purged with nitrogen for 20 min. Afterward, the monomer solution was transferred into the flask, and the reaction mixture was shaken at 25 °C for a predetermined time. The surface-initiated ATRP was terminated by exposure to air, and then the honeycomb-patterned film with glycopolymer (PS-bPHEMA-AG) was taken out from the reaction mixture and thoroughly washed with methanol and water. The PS-b-PHEMA-AG film was then dipped into 10 mL of dry methanol containing 0.05 g of sodium methoxide and vibrated for 90 min at 25 °C to deprotect the acetyl groups of glucose pentaacetate (PS-b-PHEMA-G). Thereafter, the film was washed with water and dried under reduced pressure at 60 °C. Characterization. Attenuated total reflectance Fourier transform infrared spectroscopy (FTIR/ATR) measurements were carried out on a Nicolet 6700 FTIR spectrometer equipped with an ATR cell (ZnSe, 45°). Thirty-two scans were taken for each spectrum at a nominal resolution of 4 cm-1. X-ray photoelectron spectroscopy (XPS) measurements were recorded on a PHI5000C ESCA system (Perkin-Elmer) with Al KR excitation radiation (1486.6 eV). The pressure in the analysis chamber was maintained at about 10-6 Pa during measurement. All survey and core-level spectra were referenced to the C1s hydrocarbon peak at 284.7 eV to compensate for the surface charging effect. Field emission scanning electron microscope (FESEM, Sirion-100, DOI: 10.1021/la904729b

8947

Article

Ke et al.

Figure 2. Optical images of the films prepared from PS-b-PHE-

MA solutions in CS2 with different concentration at 25 °C and RH 80% with a 2 L/min airflow: (a) 2, (b) 5, (c) 10, and (d) 20 mg/mL.

Figure 3. Optical images of the films prepared from 5 mg/mL PSb-PHEMA solutions in CS2 at 25 °C under different RH with a 2 L/ min airflow: (a) 90%, (b) 80%, (c) 70%, and (d) 60%.

Figure 4. Optical images of the films prepared from 5 mg/mL PSb-PHEMA solutions in CS2 at 25 °C and RH 80% with different airflow speed: (a) 1, (b) 2, (c) 3, and (d) 4 L/min. 8948 DOI: 10.1021/la904729b

Figure 5. Confocal laser scanning microscopy micrographs of PS (a, b) and PS-b-PHEMA (c, d, e) honeycomb-patterned films after staining with 5-AF: (a, c) optical images, (b, d) fluorescence images, (e) 3D confocal image reconstruction. The scale bars are 10 μm. FEI) was used to observe the surface morphology of films after being sputtered with gold using ion sputter JFC-1100. Height images were recorded by atomic force microscopy (AFM, SEIKO SPI3800N) under tapping conditions. The water contact angle of porous films was analyzed by a DropMeter A-200 contact angle system (MAIST Vision Inspection & Measurement Ltd. Co., China) at room temperature. Static contact angles were measured by sessile drop method. First, a 2 μL drop of water was set onto the dry film with a microsyringe. Digital images for the droplet were then recorded. Contact angles were calculated from these images with software. Each reported value was an average of at least eight independent measurements. Recognition of Lectins. After being fully wetted in HEPES buffer solution (pH 7.5, containing 10 mM HEPES, 0.15 M NaCl, 0.1 mM Ca2þ, 0.01 mM Mn2þ (not for PNA), 0.08% sodium azide), a piece of glycosylated film was dipped into 200 mL of FLConA or FL-PNA HEPES buffer solution (0.1 mg/mL) and incubated at 25 °C for 2 h. The film was then washed with HEPES buffer solution six times. After being dried under reduced pressure at room temperature, high-resolution images of the lectin-adsorbed film were recorded by CLSM. Langmuir 2010, 26(11), 8946–8952

Ke et al.

Article

Figure 6. Schematic illustration of the construction of carbohydrate microarrays and lectin patterns from the self-organized honeycombpatterned films.

Preparation of Honeycomb-Patterned Films. Honeycombpatterned films have been prepared from a variety of polymers including polystyrene star or comb polymers,25,30 rod-coil block copolymers,15 amphiphilic copolymers,14 and homopolymers.13 Earlier work demonstrates that a random copolymer composed from styrene and 2-hydroxyethyl methacrylate (HEMA) can be cast into a regular porous array.31 Nevertheless, it seems impossible to effectively rearrange random copolymers around the water droplets. In contrast, amphiphilic block copolymers yield pores enriched with hydrophilic functionality whereas the film surface remains mostly hydrophobic.21 Therefore, an amphiphilic block copolymer poly(styrene-block-(2-hydroxyethyl methacrylate)), PS187-b-PHEMA15 (Mn = 20 500 g/mol, MWD = 1.10), was synthesized by ATRP for the film preparation. PS-b-PHEMA with higher HEMA contents was also synthesized for comparison. However, it could not be dissolved in suitable casting solvent. When PS187-b-PHEMA15 was cast from carbon disulfide (CS2) in a humid atmosphere, films with highly ordered pores can be formed as verified by FESEM images (Figure 1a). The patterned pores have an average diameter of about 2.3 μm. The

top layer of the film can be easily removed with adhesive tape to expose the honeycomb structure underneath (Figure 1b). It is obvious that the pores beneath the top layer also organize into a hexagonal array with a bigger diameter of about 3.0 μm, and these pores are isolated from each other by the continuous wall with a thickness of about 0.1 μm. The top layer of the ordered pores is supported by this “polymer ring” structure, indicating that the pores are disconnected. The thickness of the honeycomb-structured film is about 3.0 μm according to the cross-section observation. The films were obtained over a range of polymer concentration (2-20 mg/mL) (Figure 2). The lower the polymer concentration is, the larger the pore size will be. When the concentration was too low (2 mg/mL), irregular patterns were formed (Figure 2a). Casting conditions including relative humidity and airflow speed also influence the pore size (Figures 3 and 4). Generally, increased humidity and reduced airflow speed lead to large pore size. The effect of solvent on the morphology of the resulted films was also investigated (see the Supporting Information). The films using CH2Cl2 or CHCl3 as the casting solvent also show similar honeycomb morphology, but with less regularity. Suborder Arrangement of PS-b-PHEMA. The rearrangement of the amphiphilic blocks around the water droplets provides potential reaction sites for the generation of carbohydrate microarrays. Up to now, this suborder arrangement can be characterized using XPS,32 contact angle measurements,33 and bacterial culture34 as previously described. In the present work, 5-aminofluorescein (5-AF) staining was used to locate the hydroxyl groups (see the Supporting Information). The hydroxyl groups were activated using p-benzoquinone35 and then reacted with 5-AF. Fluorescence patterns are observed inside the pores, but not on the top surface of the film, indicating that the top surface mainly consists of hydrophobic polystyrene blocks whereas the pore wall contains hydrophilic blocks (Figure 5). The three-dimensional presentation of the fluorescence distribution further demonstrates that the hydroxyl groups locate only in the pores. A control sample prepared from pure PS does not show obvious fluorescence (Figure 5b). It is to be noted that in the experiments an ethanol/water mixture was used as the solvent to ensure the penetration of the solution into the pores while maintaining the structure of the honeycomb-patterned films. Generally, the top surface is easier to be wetted by the 5-AF

(30) Srinivasarao, M.; Collings, D.; Philips, A.; Patel, S. Science 2001, 292, 79–83. (31) Hernandez-Guerrero, M.; Davis, T. P.; Barner-Kowollik, C.; Stenzel, M. H. Eur. Polym. J. 2005, 41, 2264–2277.

(32) Stenzel, M. H.; Davis, T. P. Aust. J. Chem. 2003, 56, 1035–1038. (33) Nygard, A.; Davis, T. P.; Barner-Kowollik, C.; Stenzel, M. H. Aust. J. Chem. 2005, 58, 595–599. (34) Dalton, H. M.; Stein, J.; March, P. E. Biofouling 2000, 15, 83–94. (35) Huang, X. J.; Yu, A. G.; Xu, Z. K. Bioresour. Technol. 2008, 99, 5459–5465.

Figure 7. FTIR/ATR spectra of (a) PS-b-PHEMA, (b) Br-immobilized, and (c) glycosylated films.

Results and Discussion

Langmuir 2010, 26(11), 8946–8952

DOI: 10.1021/la904729b

8949

Article

Ke et al. Table 1. Elemental Composition of the Honeycomb-Patterned Films As Obtained by XPS Analysis before the removal of the surface

after the removal of the surface

sample

C (%)

O (%)

Br (%)

C (%)

O (%)

Br (%)

PS-b-PHEMA film Br-immobilized film Glycosylated film

95.79 93.90 86.67

4.21 6.00 13.22

0.10 0.11

92.74 90.73 79.86

7.26 9.00 20.04

0.27 0.10

Figure 8. High-resolution FESEM images of (a) PS-b-PHEMA, (b) Br-immobilized, and (c) glycosylated films.

solution than the pore surface and then easier to be stained. Therefore, the results suggest the aggregation of the PHEMA block in the pores. Surface-Initiated ATRP on the Honeycomb-Patterned Films. The self-organized hydroxyl groups in the pores can be applied to further glycosylation, as schematically shown in Figure 6. ATRP is especially attractive and garners much more attention because of its living characteristic, tolerance for impurities, compatibility with a wide range of monomers, and relatively mild polymerization conditions. In this work, site-directed glycosylation was performed on the honeycomb-patterned films by surface-initiated ATRP of a carbohydrate-containing monomer. To introduce ATRP initiators, the hydroxyl groups were reacted with 2-bromoisobutyryl bromide. The reaction was carried out for 12 h at room temperature. For an ATRP process, the deactivating species are crucial to the equilibrium between the dormant species and the active sites. However, differing from bulk ATRP, surfaceinitiated ATRP has relatively low surface initiator concentration, which prevents the formation of a sufficient amount of the deactivator Cu(II) complex and results in decreased control of the polymerization process.36,37 To achieve good control over surface-initiated ATRP and to modulate the speed of polymerization, 10 mol % of CuBr2 (relative to CuBr) was added to the reaction mixture. FTIR/ATR spectra for the PS-b-PHEMA, the Br-immobilized, and the glycosylated films are shown in Figure 7. The PS-bPHEMA film shows a characteristic peak at 1716 cm-1 due to the stretching vibration of carbonyl groups. After bromination, the intensity of this peak strengthens because of the introduction of more carbonyl groups. The peak also shows a blue shift from 1716 to 1735 cm-1.38 It can be seen from Figure 7c that the carbonyl peak increases and shifts back to 1716 cm-1, ascribing to the ester bond connecting the saccharide residues to the grafted chains. An increase in the broad absorption band at about 3100-3600 cm-1 is attributed to the presence of a large amount of hydroxyl groups of the carbohydrates. Therefore, the FTIR results confirm that the glycopolymer has been successfully grafted on the film by surface-initiated ATRP. X-ray photoelectron spectroscopy (XPS) analysis provides further insight into the reactions on the surface of the film. For PS-b-PHEMA film, the calculated C and O contents are 95.79% (36) Yang, Q.; Tian, J.; Hu, M. X.; Xu, Z. K. Langmuir 2007, 23, 6684–6690. (37) Jeyaprakash, J. D.; Samuel, S.; Dhamodharan, R.; Ruhe, J. Macromol. Rapid Commun. 2002, 23, 277–281. (38) Wan, L. S.; Lei, H.; Ding, Y.; Fu, L.; Li, I.; Xu, Z. K. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 92–102.

8950 DOI: 10.1021/la904729b

and 4.21%, respectively (Table 1). As reported previously,21,33 the XPS cannot capture the full picture of the element distribution when using porous materials. The removal of the top surface layer with adhesive tape helps to explore the structure underneath. The PS-b-PHEMA film after the removal possesses higher oxygen content, indicating an increased concentration of HEMA units around the pores. After reacting with 2-bromoisobutyryl bromide, bromine is introduced onto the pore wall. The bromine contents are 0.10% and 0.27% before and after the removal of the top surface layer, respectively. Because of the introduction of more carbonyl groups, the oxygen content of the Br-immobilized films increases. It further increases to 13.22% when the grafting takes place. Moreover, the glycosylated film has much higher oxygen content (20.04%) after the removal of top surface layer, indicating that the glycosylation mainly occurs inside the pores. The surface morphologies of the original, the Br-immobilized, and the glycosylated films were investigated by FESEM and tapping mode atomic force microscopy (AFM). As seen from Figure 8, no significant change in the topography was observed before and after the immobilization of the ATRP initiator. In contrast, the glycosylated film shows a quite different topography (Figure 8c). As expected, the growth of polymer is mainly confined in the pores of the film. The roughness of the top surface also slightly increases. This can be suppressed by shortening the polymerization time or increasing the concentration of CuBr2 (see the Supporting Information). Tapping mode AFM was employed to obtain more detailed information about the surface topography (Figure 9). The AFM image of the original film is almost the same with that of the Br-immobilized one. The section analysis indicates that the pore diameter is about 2.3 μm and the depth is about 500 nm. After polymerization of AcGEMA, the pore depth of the glycosylated film decreases to about 300 nm and the pore diameter decreases slightly, confirming the presence of grafted glycopolymer. Contact angle measurements provide additional evidence for the site-directed grafting. Depending on the air entrapped in the pores, the contact angle (θM) can be theoretically calculated using Cassie and Baxter’s law:39 cos θM ¼ ð1 -fpore Þ cos θpolymer þ fpore cos 180° where fpore is the area fraction of pores on the surface and θpolymer is the water contact angle of a polymer in the form of a thin and smooth film. According to Cassie and Baxter’s law, the contact (39) Cassie, A. B. D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 0546–0550.

Langmuir 2010, 26(11), 8946–8952

Ke et al.

Article

Figure 9. Top views and section analysis of AFM images of (a) PS-b-PHEMA, (b) Br-immobilized, and (c) glycosylated films.

angle is only dependent on the area fraction of pores and the polymer on the top layer. As estimated from the FESEM images, the fpore is about 0.51. Considering that θPS is 89° and θPS-b-PHEMA is 78°, the calculated θM is 120° and 114°, respectively. Besides, because the glycosylated surface is highly hydrophilic (the water contact angle is about 30°),40,41 the θM of a film that has a glycosylated top layer will be about 95°, which is a quite low value. Experimentally, the contact angles of the PS-b-PHEMA, the Brimmobilized, and the glycosylated films are 118°, 121°, and 117°, respectively. That is to say, the three films have no significant difference in water contact angle values. Therefore, it can be concluded that the top layer of the films is mainly composed of PS whereas the pores consist of glycopolymer; i.e., carbohydrate (40) Wulff, G.; Schmidt, H.; Zhu, L. M. Macromol. Chem. Phys. 1999, 200, 774– 782. (41) Hu, M. X.; Wan, L. S.; Fu, Z. S.; Fan, Z. Q.; Xu, Z. K. Macromol. Rapid Commun. 2007, 28, 2325–2331.

Langmuir 2010, 26(11), 8946–8952

microarrays have been prepared using this site-directed surfaceinitiated ATRP on the honeycomb-patterned films. Recognition of Lectins. Carbohydrate microarrays are useful for studying the recognition events between carbohydrates and proteins. The use of a porous film rather than a solid substrate can enhance the effective probe density and hence the sensitivity of the microarrays. The specific recognition to lectin was performed on these carbohydrate microarrays. Concanavalin A (Con A), a plant lectin, is commonly used to probe R-D-mannose, R-Dglucose, and β-D-glucose. Another lectin, peanut agglutinin (PNA), binds preferentially to β-D-galactose or galactosyl-(β1,3)-N-acetylgalactosamine. The carbohydrate microarrays immobilized with β-D-glucose residues are expected to specifically recognize Con A instead of PNA.42 The immersion of a control (42) Hu, M. X.; Wan, L. S.; Liu, Z. M.; Dai, Z. W.; Xu, Z. K. J. Mater. Chem. 2008, 18, 4663–4669.

DOI: 10.1021/la904729b

8951

Article

Ke et al.

Figure 10. Fluorescence images of honeycomb films after adsorption of fluorescent-labeled lectins: (a) adsorption of FL-Con A on PS-b-

PHEMA film, (b) adsorption of FL-Con A on glycosylated film, and (c) adsorption of FL-PNA on glycosylated film. The scale bars are 5 μm.

film without saccharide residues into a fluorescent-labeled Con A (FL-Con A) solution followed by a standard washing step does not result in obvious adsorption of Con A (Figure 10a). In contrast, β-D-glucose microarray yields significantly different results, as shown in Figure 10b. Fluorescence patterns can be clearly observed, and the spots are solid or nearly solid circles. It is to be noted that ring-shaped fluorescence circles appearing on the protein patterned film is reasonable because of the limitation of focus. Moreover, the glycosylated film interacted with the fluorescent-labeled PNA (FL-PNA) solution only shows very weak fluorescence, indicating that the glucose-based microarrays have good protein selectivity. It is well-known that the specifically adsorbed Con A can be easily desorbed from the film using glucose solution.42 Therefore, protein microarrays could be reversibly constructed by adsorbing and desorbing the lectins on the established carbohydrate microarrays.

Conclusions In summary, we present a facile approach to carbohydrate microarrays by surface-initiated ATRP on the self-organized breath figure films. The arrangement of amphiphilic block around the breath figures results in honeycomb-patterned films with hydrophilic and functional pores, as confirmed by the threedimensional fluorescence micrographs. The enrichment of the

8952 DOI: 10.1021/la904729b

hydrophilic groups makes it feasible for site-directed grafting. Surface-initiated ATRP of a carbohydrate-containing monomer, as a nonlithographic method, generates well-controlled carbohydrate microarrays. Further specific recognition of the carbohydrate microarrays to lectin (Con A) can lead to an organized microarray of protein. Therefore, this convenient and cost-saving approach is not only versatile to the preparation of carbohydrate microarrays with other kinds of sugar residues but also opens a new route to fabricating other functional microarrays such as protein-patterned surfaces. Acknowledgment. Financial support from the National Natural Science Foundation of China (Grant No. 50803053), the National Natural Science Foundation of China for Distinguished Young Scholars (Grant No. 50625309), and the National Basic Research Program of China (2009CB623401) is gratefully acknowledged. Dr. Ling-Shu Wan also thanks the financial support from China Postdoctoral Science Foundation (Grant No. 20081466). Supporting Information Available: GPC and NMR results of the copolymer PS-b-PHEMA; images of a series of films prepared with different solvents. This material is available free of charge via the Internet at http://pubs.acs.org.

Langmuir 2010, 26(11), 8946–8952