Photochemical Micropatterning of Carbohydrates on a Surface

Gregory T. Carroll , Gábor London , Tatiana Fernández Landaluce , Petra Rudolf , and ..... Henning S. G. Beckmann , Andrea Niederwieser , Manfred Wi...
0 downloads 0 Views 341KB Size
Download PDF
Langmuir 2006, 22, 2899-2905

2899

Photochemical Micropatterning of Carbohydrates on a Surface Gregory T. Carroll,† Denong Wang,‡ Nicholas J. Turro,†,§ and Jeffrey T. Koberstein*,§ Department of Chemistry, Columbia UniVersity, 3000 Broadway, MC 3157, 10027, New York, New York, Carbohydrate Microarray Laboratory, Departments of Genetics, Neurology, and Neurological Sciences, Stanford UniVersity School of Medicine, Beckman Center B007, 94305, Stanford, California, and Department of Chemical Engineering, Columbia UniVersity, 500 West 120th Street, 10027, New York, New York 10027 ReceiVed NoVember 16, 2005. In Final Form: January 9, 2006 In this report, we demonstrate a versatile method for the immobilization and patterning of unmodified carbohydrates onto glass substrates. The method employs a novel self-assembled monolayer to present photoactive phthalimide chromophores at the air-monolayer interface. Upon exposure to UV radiation, the phthalimide end-groups graft to surface-adsorbed carbohydrates, presumably by a hydrogen abstraction mechanism followed by radical recombination to form a covalent bond. Immobilized carbohydrate thin films are evidenced by fluorescence, ellipsometry and contactangle measurements. Surface micropatterns of mono-, oligo-, and polysaccharides are generated by exposure through a contact photomask and are visualized by condensing water onto the surface. The efficiency of covalent coupling is dependent on the thermodynamic state of the surface. The amount of surface-grafted carbohydrate is enhanced when carbohydrate surface interactions are increased by the incorporation of amine-terminated molecules into the monolayer. Glass substrates modified with mixed monolayers of this nature are used to construct carbohydrate microarrays by spotting the carbohydrates with a robot and subsequently illuminating them with UV light to covalently link the carbohydrates. Surface-immobilized polysaccharides display well-defined antigenic determinants for antibody recognition. We demonstrate, therefore, that this novel technology combines the ability to create carbohydrate microarrays using the current state-of-the-art technology of robotic microspotting and the ability to control the shape of immobilized carbohydrate patterns with a spatial resolution defined by the UV wavelength and a shape defined by a photomask.

Introduction Carbohydrates, like nucleic acids and proteins, carry important biological information. The development of high-throughput technologies for generating DNA and protein microarrays has been vigorously explored and has contributed greatly to the fields of genomics and proteomics. A newer field that explores the information content of carbohydrates, called glycomics, has recently emerged and has been facilitated by the relatively recent development of carbohydrate microarrays.1-10 Already, carbohydrate microarrays have been used to investigate the SARS6 and HIV7 viruses. In addition, enzyme activity,3,11 glycome sequencing,2 and carbohydrate interactions with cells,12,13 * Corresponding author. † Department of Chemistry, Columbia University. ‡ Stanford University School of Medicine. § Department of Chemical Engineering, Columbia University. (1) Wang, D.; Liu, S.; Trummer, B. J.; Deng, C.; Wang, A. Nat. Biotechnol. 2002, 20 (3), 275-281. (2) Fukui, S.; Feizi, T.; Galustian, C.; Lawson Alexander, M.; Chai, W. Nat. Biotechnol. 2002, 20 (10), 1011-1017. (3) Fazio, F.; Bryan, M. C.; Blixt, O.; Paulson, J. C.; Wong, C.-H. J. Am. Chem. Soc. 2002, 124 (48), 14397-14402. (4) Park, S.; Shin, I. Angew. Chem., Int. Ed. 2002, 41 (17), 3180-3182. (5) Willats, W. G. T.; Rasmussen, S. E.; Kristensen, T.; Mikkelsen, J. D.; Knox, J. P. Proteomics 2002, 2 (12), 1666-1671. (6) Wang, D.; Lu, J. Physiol. Genomics 2004, 18 (2), 245-248. (7) Adams, E. W.; Ratner, D. M.; Bokesch, H. R.; McMahon, J. B.; O’Keefe, B. R.; Seeberger, P. H. Chem. Biol. 2004, 11 (6), 875-881. (8) Blixt, O.; Head, S.; Mondala, T.; Scanlan, C.; Huflejt, M. E.; Alvarez, R.; Bryan, M. C.; Fazio, F.; Calarese, D.; Stevens, J.; Razi, N.; Stevens, D. J.; Skehel, J. J.; van Die, I.; Burton, D. R.; Wilson, I. A.; Cummings, R.; Bovin, N.; Wong, C.-H.; Paulson, J. C. Proc. Natl. Acad. Sci. U.S.A. 2004, 101 (49), 17033-17038. (9) Houseman, B. T.; Mrksich, M. Chem. Biol. 2002, 9 (4), 443-454. (10) Ko, K.-S.; Jaipuri, F. A.; Pohl, N. L. J. Am. Chem. Soc. 2005, 127 (38), 13162-13163. (11) Bryan, M. C.; Lee, L. V.; Wong, C.-H. Bioorg. Med. Chem. Lett. 2004, 14 (12), 3185-3188. (12) Nimrichter, L.; Gargir, A.; Gortler, M.; Altstock, R. T.; Shtevi, A.; Weisshaus, O.; Fire, E.; Dotan, N.; Schnaar, R. L. Glycobiology 2004, 14 (2), 197-203.

antibodies,1 and proteins4,14 have been studied with carbohydrate microarrays. Immobilizing carbohydrates on surfaces has become a major preliminary challenge in the area. Most current methods involve either a noncovalent immobilization that becomes less stable as the molecular weight (MW) decreases, or synthetic methods in which each carbohydrate to be spotted must first be chemically modified. To develop a simple and universal approach to carbohydrate microarray fabrication, it is very important to devise methods that allow for covalent immobilization of carbohydrates on a surface without prior chemical derivatization. Only a few methods have been reported that demonstrate this goal. Underivatized carbohydrates have been covalently attached to monolayers bearing phenylboronic acid groups,15 polysaccharide films bearing diazirine groups,16 and hydrazide-coated glass slides.17 Only the latter two methods were used to construct microarrays. Carbohydrates also contain important physical and chemical properties that may find utility in biotechnology and novel devices. Surface-immobilized carbohydrates are potential components in biological sensors,18 scaffolds for tissue engineering,19 templates for studying cell behavior in a confined space,20 suprabiomolecular structures on surfaces,21 host-guest complexes on a surface, (13) Disney, M. D.; Seeberger, P. H. Chem. Biol. 2004, 11 (12), 1701-1707. (14) Park, S.; Lee, M.-R.; Pyo, S.-J.; Shin, I. J. Am. Chem. Soc. 2004, 126 (15), 4812-4819. (15) Takahashi, S.; Anzai, J. Langmuir 2005, 21 (11), 5102-5107. (16) Angeloni, S.; Ridet, J. L.; Kusy, N.; Gao, H.; Crevoisier, F.; Guinchard, S.; Kochhar, S.; Sigrist, H.; Sprenger, N. Glycobiology 2005, 15 (1), 31-41. (17) Lee, M.-R.; Shin, I. Org. Lett. 2005, 7 (19), 4269-4272. (18) Jelinek, R.; Kolusheva, S. Chem. ReV. 2004, 104 (12), 5987-6015. (19) Yeong, W.-Y.; Chua, C.-K.; Leong, K.-F.; Chandrasekaran, M. Trends Biotechnol. 2004, 22 (12), 643-652. (20) Chen, C. S.; Mrksich, M.; Huang, S.; Whitesides, G. M.; Ingber, D. E. Science 1997, 276 (5317), 1425-1428. (21) Liu, G.-Y.; Amro, N. A. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (8), 5165-5170.

10.1021/la0531042 CCC: $33.50 © 2006 American Chemical Society Published on Web 02/15/2006

2900 Langmuir, Vol. 22, No. 6, 2006

Carroll et al.

Figure 1. Schematic of a phthalimide derivative undergoing a photochemical hydrogen abstraction reaction followed by recombination to form a covalent bond.

and three-dimensional biochips.22 It is desirable to have spatial control over surface immobilization when incorporating carbohydrates into many of these devices. One widely accessible approach involves pattern transfer via a photomask or stamp. Although patterning biological materials on surfaces using these versatile approaches has been demonstrated,23,24 there are relatively few reports focusing on carbohydrates.25 Irradiating photoactive surfaces in the presence of a photomask26 is a well-known technology that employs photons as traceless reagents for pattern formation without the use of a spotter, microstamps, or an atomic force microscope and has been used to control the spatial deposition of various materials including proteins,27 DNA,28 cells,29 and colloidal and nanoparticles.30 The resolution of such patterns is controlled by the size of the illumination pattern and is not dependent on the size of the drop placed on the surface by a spotter. Such features allow greater control over the size and shape of a micron-sized architecture. As the size of such patterns continues to decrease, scaffolds of individual macromolecules could be produced, allowing for interactions at the single-molecule or few-molecules level to be interrogated. In addition, reducing the size of the pattern reduces the amount of material consumed. In this report, we demonstrate a versatile method for covalent immobilization and patterning of unmodified mono-, oligo-, and polysaccharides onto glass substrates. This technology involves self-assembly of a new class of photoactive monolayers onto glass substrates. The monolayers present phthalimide chromophores31 at the surface that, upon exposure to light, graft surface adsorbed carbohydrates by hydrogen abstraction followed by radical recombination. Using a robotic spotter, we are able to generate a microarray of carbohydrates and demonstrate highthroughput characterization of antigen-antibody interactions. The surface-immobilized carbohydrates retain their immunological properties. In comparison with nitrocellulose-coated substrates, an established technology for carbohydrate microarrays,1,2,6,32 this novel approach is much less dependent on the MW of the spotted carbohydrates and shows a higher grafting efficiency for lower MWs. The photochemical patterning method described herein requires no chemical modification of the sugars prior to deposition, is applicable for carbohydrates of different MWs, requires no chemical reagents for covalent coupling of carbohydrates on the surfaces, and uses existing microspotting devices (22) Blawas, A. S.; Reichert, W. M. Biomaterials 1998, 19 (7-9), 595-609. (23) Whitesides, G. M.; Ostuni, E.; Takayama, S.; Jiang, X.; Ingber, D. E. Annu. ReV. Biomed. Eng. 2001, 3, 335-373. (24) Pirrung, M. C. Angew. Chem., Int. Ed. 2002, 41 (8), 1276-1289. (25) Chevolot, Y.; Bucher, O.; Leonard, D.; Mathieu, H. J.; Sigrist, H. Bioconjugate Chem. 1999, 10 (2), 169-175. (26) Fodor, S. P. A.; Read, J. L.; Pirrung, M. C.; Stryer, L.; Lu, A. T.; Solas, D. Science 1991, 251 (4995), 767-773. (27) Rozsnyai, L. F.; Fodor, S. P. A.; Schultz, P. G.; Benson, D. R. Angew. Chem. 1992, 104 (6), 801-802. (28) Pease, A. C.; Solas, D.; Sullivan, E. J.; Cronin, M. T.; Holmes, C. P.; Fodor, S. P. A. Proc. Natl. Acad. Sci. U.S.A. 1994, 91 (11), 5022-5026. (29) Dillmore, W. S.; Yousaf, M. N.; Mrksich, M. Langmuir 2004, 20 (17), 7223-7231. (30) Lee, K.; Pan, F.; Carroll, G. T.; Turro, N. J.; Koberstein, J. T. Langmuir 2004, 20 (5), 1812-1818. (31) Kanaoka, Y. Acc. Chem. Res. 1978, 11 (11), 407-413. (32) Wang, D. Proteomics 2003, 3 (11), 2167-2175.

for high-throughput microarray construction. The methodology we present has potential applications in materials, biological, and medical research.

Results and Discussion To covalently link carbohydrates to a surface without prior derivatization, we created self-assembled monolayers (SAMs) containing aromatic carbonyls that can react with C-H groups upon absorption of a photon to form a covalent bond.33 Phthalimide derivatives can undergo all the major photochemical reactions of aromatic carbonyls.31 Exposure to UV light produces an excited n-π* state that can abstract a hydrogen atom from a nearby molecule. The resulting radicals can then recombine, forming a covalent bond as shown in Figure 1. Other secondary processes are also possible, including disproportionation and back-transfer. Carbohydrate substrates can undergo pH-dependent and independent rearrangements, depending on the structure of the carbohydrate.34 Facile incorporation of potassium phthalimide into bromine-terminated silanes allows for self-assembly on silicon, glass, or quartz substrates. Terminal groups other than silanes could readily be employed in the synthesis to create phthalimides for self-assembly onto other substrates. To create a novel surface suitable for immobilizing carbohydrates, a phthalimide-derivatized silane was synthesized in one step by reacting 11-bromoundecanetrimethoxy silane with potassium phthalimide in dimethylformamide (DMF) to produce 11-phthalimidoundecanetrimethoxy silane (compound 1). Compound 1 was self-assembled on silicon, glass, and quartz in anhydrous toluene to produce SAM 1 as shown in Figure 2. The self-assembly of compound 1 on the surface was verified by UV/Visible (UV-vis) spectroscopy as shown in Figure 3. Under the rough assumption that the extinction coefficient of the chromophore on the surface is the same as that in solution, the approximate surface coverage was calculated to be 5.5 molecules/ nm2.35 A rough calculation using Chem.3D suggests that about 4.9 aliphatic phthalimides can fit in a space of 1 nm2, a value that is the same order of magnitude as the experimental value, suggesting that SAM 1 is densely packed. In addition, an H2O contact angle of 65 (1° and an ellipsometric thickness of 1.4 ( 0.1 nm indicated the self-assembly of compound 1 on silicon. To test the ability of surface-bound phthalimides to photochemically immobilize sugars, 2000 kDa fluorescein isothiocyanate (FITC)-conjugated R(1,6)dextran polysaccharide films were spin-coated onto SAM 1 from an aqueous solution and irradiated for approximately 1 h with a 300 nm rayonet bulb in an inert environment. Two controls were also prepared. In the first, polysaccharides were spin-coated onto SAM 1 and left in the dark. In the second, polysaccharides were spin-coated onto an underivatized silicon wafer. All three samples were placed in water-filled vials for 12 h. After removing the samples and (33) Turro, N. J. Modern Molecular Photochemistry; University Science Books: Sausalito, CA, 1991. (34) Gilbert, B. C.; King, D. M.; Thomas, C. B. Carbohydr. Res. 1984, 125 (2), 217-235. (35) Moon, J. H.; Shin, J. W.; Kim, S. Y.; Park, J. W. Langmuir 1996, 12 (20), 4621-4624.

Photochemical Micropatterning of Carbohydrates

Langmuir, Vol. 22, No. 6, 2006 2901

Figure 2. Synthesis of compound 1 and SAM 1.

Figure 3. UV/vis spectra of compound 1 in ethanol (dashed line) and SAM 1 (solid line).

Figure 4. Fluorescence spectra of 2000 kDa FITC-conjugated R(1,6)dextran films under three conditions: irradiated SAM 1 (dashed line), dark SAM 1 (dotted line), and underivatized silicon (solid line). Each spectrum was obtained after washing the substrates for 12 h in H2O.

rinsing with water and methanol followed by blow-drying with argon, the fluorescence spectrum of each sample was obtained, as shown in Figure 4. Preferential retention of polysaccharides on the irradiated sample relative to the two controls indicates the photochemical immobilization of the polysaccharides on SAM 1. The film thicknesses of the three samples were measured using a Beaglehole ellipsometer in variable angle mode. A refractive index value of 1.5 was used for the organic layer. The irradiated sample retained 7.1 ( 0.3 nm of material after the rinse. The thickness of the material on SAM 1 unexposed to light was 0.7 ( 0.3 nm, and the thickness on the underivatized silicon wafer was 0.4 ( 0.3 nm. The reported thicknesses do not include the thickness of SAM 1. The surfaces were further investigated with water contact-angle measurements. The hydrophilic nature of the sugars reduced the water contact angle from 65 ( 1 to 28

( 1° on the irradiated SAM. Inefficient immobilization on the dark control is evident from a post-rinse contact angle of 62 ( 1°. The higher retention of material on the irradiated SAM demonstrates that self-assembled phthalimide monolayers are capable of photochemically bonding to an overlayer sugar film, despite any spatial restrictions on the chromophore as a result of placement in a constrained environment. We speculate that the nature of the bonding is covalent and results from radicalradical recombination following hydrogen abstraction. The above experiments were also performed on SAMs comprised of benzophenone chromophores, another class of aromatic carbonyls that can photochemically abstract hydrogen from C-H groups and have been shown to graft polymers to surfaces.36 Although the benzophenone monolayers were able to graft the sugars, the resulting sugar film thickness and fluorescence intensity were lower, and the contact angle was higher than that of the films on SAM 1. The lower performance may be due to the radical center in the benzophenone SAM residing further from the surface than that in the phthalimides, self-quenching of the excited state, or a higher interfacial tension between the more hydrophobic benzophenone monolayer and the sugar film compared to the phthalimide-sugar interaction. Benzophenone SAMs have more hydrophobic character than phthalimide SAMs, as evidenced by a higher water contact angle of about 85°. Preliminary experiments with a microarray spotter have shown that hydrophilic surfaces are more easily spotted than hydrophobic substrates. We found that more material physisorbed onto SAM 1 in comparison to a benzophenone-terminated SAM. In any case, other photoactive carbonyl groups capable of abstracting hydrogen atoms can be substituted and may enhance or retard the reaction because of the efficiency of self-assembly, steric, and thermodynamic constraints. In addition to covalently attaching underivatized sugars to a substrate, we are also able to generate patterns of grafted sugars. Our strategy for immobilizing carbohydrates on SAM 1 in a spatially controlled fashion is presented in Figure 5. Spin-coated polysaccharide films were covered with a photomask consisting of a copper grid with spacings of 280 µm and irradiated for 2 h as described above. The photoreaction is restricted to the opaque regions of the mask, leaving the pattern of the mask written to the surface via attached carbohydrates. We removed ungrafted sugars by sonicating films in water for 15 min, changing the water and vial every 5 min. (36) Prucker, O.; Naumann, C. A.; Ruehe, J.; Knoll, W.; Frank, C. W. J. Am. Chem. Soc. 1999, 121 (38), 8766-8770.

2902 Langmuir, Vol. 22, No. 6, 2006

Carroll et al.

Figure 5. The strategy for direct chemical patterning of a surface with carbohydrates involves a photolithographic technique. Carbohydrates are spin-coated onto SAM 1 and covered with a mask. Irradiation induces a photochemical reaction that covalently links the carbohydrates to SAM 1.

Figure 6. Optical microscope images of water condensation on photochemically generated patterns of polysaccharides on SAM 1.

Condensing water onto the substrate provided a quick way to visualize the hydrophilic patterns using optical microscopy.37 Figure 6 presents an optical microscope image of water condensation onto patterned R(1,6)dextran polysaccharides with a MW of 2000 kDa. The hydrophilic attraction between water and the polysaccharides relative to the unmodified masked regions of the monolayer causes water to preferentially reside on the areas of the substrate containing polysaccharide. The results were similar when 20 kDa R(1,6)dextrans were patterned. Other strategies involving immobilization without prior derivatization have been successful with carbohydrate-containing macromolecules noncovalently adsorbed on nitrocellulose1 or oxidized polystyrene (PS);5 however, the nitrocellulose study showed that the immobilization efficiency decreases with the MW of the polysaccharides.1 To show the versatility of our method, we tested glucose and sucrosestwo simple sugars at the low extreme of MW, containing both six- and five-membered sugar moieties. The resulting water condensation images are presented in Figure 7. The visible patterns clearly show that our method extends to sugars of the lowest MWs. (37) Lopez, G. P.; Biebuyck, H. A.; Frisbie, C. D.; Whitesides, G. M. Science 1993, 260 (5108), 647-649.

Figure 7. Optical microscope images of breath condensation on photochemically generated patterns of (a) sucrose and (b) glucose.

To show that our method is applicable for the high-throughput production of carbohydrate microarrays, we investigated whether carbohydrates could be microspotted and subsequently photoimmobilized using a robotic spotter. We applied the FITCconjugated polysaccharides as probes to monitor the spotting

Photochemical Micropatterning of Carbohydrates

Langmuir, Vol. 22, No. 6, 2006 2903

Figure 8. Immobilization of mono-, oligo-, and polysaccharides on PAM. A FAST slide is included for comparison. (A) Fluorescence images and intensity values of the spotted polysaccharides, the FITC-conjugated R(1,6)dextrans of 20-, 70-, and 2000 kD, before treatment with light. The three-dimensional FAST slide adsorbs more material than does the two-dimensional PAM. (B) Fluorescence images and intensity values after treatment with light, rinsing, and staining with a biotinylated anti-dextran antibody (16.4.12E), followed by staining with a Streptavidin-Cy3 conjugate. Immobilization on PAM is not dependent on MW, and a greater amount of 20 kD polysaccharides are retained even though much less material could initially be spotted. (C) Fluorescence intensity values of mono- and oligosaccharide arrays after treatment with light, rinsing, and staining with a biotinylated lectin, Con A, followed by staining with a Streptavidin-Cy5 conjugate. IM3, IM5, and IM7 refer to isomaltotriose, isomaltopentose, and isomaltoheptaose, respectively.

process. After irradiation for 1 h, extensive washing, and “blocking” with bovine serum albumin (BSA), we introduced specific antibodies and lectins to detect immobilized carbohydrates. Bound antibodies were revealed with a streptavidin-Cy3 conjugate. We found that the thermodynamic parameters of the surface needed to be adjusted to transfer a detectable amount of carbohydrates from the pin of the spotter to SAM 1. To make the surface more attractive to carbohydrates, we made mixed phthalimide-amine monolayers (PAM) from a solution containing a 5:1 ratio of aminopropyltrimethoxy silane to compound 1. Presumably, the hydrophilic amine group interacts more favorably with the carbohydrates compared to the more hydrophobic phenyl ring of compound 1, decreasing the interfacial tension between the carbohydrate and the substrate, which allows a sufficient amount of carbohydrates to be adsorbed to the surface for subsequent photoimmobilization. Figure 8a,b presents the results after spotting FITC-conjugated R(1,6)dextrans with MWs of 20, 70, and 2000 kDa on PAM and nitrocellulose-coated FAST slides. The FAST slide was treated to provide a comparison of our new method with an established platform. By examining the fluorescent signals of the spotted slides before irradiation and washing (Figure 8a), we found that the amounts of carbohydrates adsorbed onto PAM are significantly less than those spotted on the FAST slide. This may be attributed to the two-dimensional nature of PAM, which allows less polysaccharides to be delivered and adsorbed in comparison to nitrocellulose surfaces with thicker three-dimensional coatings. However, staining the slides with an anti-R(1,6)dextran antibody (16.4.12E), which is specific for the terminal nonreducing end epitopes displayed by all three dextran conjugates1 revealed that

the PAM surface retains a similar amount of polysaccharides regardless of the MW of the polysaccharides spotted (Figure 8b). Neither an underivatized glass substrate nor PAM without UV irradiation showed a detectable signal with anti-R(1,6)dextran antibodies under the same experimental conditions. These results were reproduced in multiple microarray assays (data not shown). Thus, not only is PAM suitable for use in the high-throughput construction of polysaccharide microarrays, but the photoimmobilized carbohydrates also retain their immunological properties, as defined by a specific antibody, after immobilization. We further examined a panel of mono- and oligosaccharide arrays on PAM and FAST slides. The spotted arrays were probed with a biotinylated lectin, Concanavalin A (Con A; Figure 8c), which is Man- and/or Glc-specific and requires the C-3, C-4, and C-5 hydroxyl groups of the Man or Glc ring for binding. We found that oligosaccharides with three (IM3), five (IM5), and seven glucoses (IM7) are reactive to Con A on the PAM slide but not on the FAST slide. However, none of the spotted monosaccharides were reactive to the lectin on these surfaces. The method of photocoupling, which can target any CH- group on the sugar rings with varying specificity depending on the structure of the ring34,38,39 (Figure 5), may interfere significantly with the lectin binding of monosaccharides, Man, or Glc. The limited specificity of the reaction and the lesser amount of saccharide epitopes present for smaller carbohydrates reduces (38) Madden, K. P.; Fessenden, R. W. J. Am. Chem. Soc. 1982, 104 (9), 25782581. (39) Shkrob, I. A.; Depew, M. C.; Wan, J. K. S. Chem. Phys. Lett. 1993, 202 (1-2), 133-140.

2904 Langmuir, Vol. 22, No. 6, 2006

the probability that a biologically active epitope presents itself at the air-monolayer interface. These results show that PAM offers a plausible alternative to nitrocellulose for displaying polysaccharides and oligosaccharides on glass chips. Larger panels of carbohydrates with structural and immunological diversities must be introduced to further validate and explore the potential of this novel chip substrate for microarray technologies. Although this report focuses on the application of immobilizing and patterning carbohydrates, phthalimide-containing monolayers are suitable for patterning virtually any material containing C-H groups. In addition to various carbohydrates, we are also able to pattern a variety of polymers. PS, poly(methyl methacrylate) (PMMA), and poly(vinyl alcohol) (PVA) were all immobilized and patterned on SAM 1 (see Supporting Information for images). The versatility of our method allows materials with varying surface tensions and chemical functionalities to be immobilized and patterned on a surface, allowing for a fast and simple approach to design organic scaffolds for novel materials and devices.

Summary We have shown that a new class of SAMs containing phthalimide chromophores is capable of photochemically immobilizing carbohydrates on a flat substrate. The method requires no chemical modification of the carbohydrates prior to deposition and is not limited to carbohydrates of high MWs. Further, immobilized carbohydrate antigens are shown to retain their ability to interact with the corresponding antibody or lectin. The photochemical nature of the technique allows patterns to be created and makes the method adaptable to the full potential of photolithography, which is currently used in industry for the high-throughput fabrication of computer chips and nanoscale patterning. Multiple carbohydrate patterns can be immobilized by repeating the reaction with a different carbohydrate in a previously masked region. In conjunction with a microarray spotter, large libraries of carbohydrates can be immobilized on our surface without previous derivatization. The versatility and ease of the method provides a platform for biologists, chemists, and engineers to investigate and create new biological materials as well as characterize carbohydrate interactions in a rapid manner. Methods Synthesis of Compound 1. A 3.3 mmol portion of 11bromoundecanetrimethoxysilane (Gelest) was added to a solution of an equimolar amount of potassium phthalimide (Aldrich) in 60 mL of anhydrous DMF (Aldrich). The solution was stirred overnight at room temperature (RT) under argon. Chloroform (50 mL) was added. The solution was transferred to a separatory flask containing 50 mL of H2O. The aqueous layer was separated and then extracted with two 20 mL portions of chloroform. The combined chloroform extract was washed with several 20 mL portions of H2O. The chloroform was removed by rotoevaporation, and residual DMF was removed on a high vacuum line to give a pale yellow liquid (0.99 g, 72% yield). The compound was used without further purification. Note that, for self-assembly experiments, residual DMF was not removed. 1H NMR: (CDCl3) δ 7.82 (m, 2H), 7.69 (m, 2H), 3.66 (t, J ) 7 Hz, 2H), 3.55 (s, 9H), 1.44-1.15 (m, 18H), 0.71-0.51 (m, 2H). LRMS-FAB+ (m/z): (M - H) 420.2 (experimental), 420.2 (calculated); (M - OCH3) 390.1 (experimental), 390.2 (calculated). Self-Assembly of SAM 1. The substrates consisted of glass (ArrayIt), quartz (SPI), or silicon (wafer world). The substrates and glassware were cleaned by being boiled in a “piranha” solution (7:3 sulfuric acid/H2O2) for 1 h followed by an extensive rinse with water and methanol. Substrates were dried with a stream of argon and immersed in a 1 mmol solution of compound 1 in anhydrous toluene (Aldrich). The solution was kept under argon and left undisturbed

Carroll et al. for 12 h. The resulting SAMs were rinsed with toluene and sonicated three times for 2 min each in toluene, toluene/methanol (1:1), and methanol. Substrates were kept in argon-purged vials until further use. Preparation of PAM. PAM was made in the same manner as SAM 1, except that a 5× molar amount of aminopropyltrimethoxy silane (Gelest) was simultaneously added with compound 1. The contact angle of the resulting surface was 72 ( 1°. Photochemical Grafting of Polysaccharide Films. FITCconjugated R(1,6)dextrans weighing 20 or 2000 kD (Dextran-FITC) (Sigma) were spin-coated from a 10 mg/mL aqueous solution at 3000 rpm for 90 s and placed in argon-purged quartz tubes. Irradiation was carried out for 70 min with a Rayonet photochemical reactor equipped with lamps that emit at 300 nm. For ellipsometry and fluorescence experiments, the surface was rinsed by placing in H2O for 12 h followed by rinsing with methanol. Substrates were blown dry with argon. Instrumental Measurements. UV-vis spectra were obtained using a Shimadzu (UV-2401PC) UV-vis recording spectrophotometer. Contact-angle measurements were performed with a RameHart 100-00 contact-angle goniometer using Millipore Milli-Q water. At least three droplets were measured on each sample and averaged. Thicknesses were measured with a Beaglehole ellipsometer in variable angle mode. A refractive index of 1.5 was used for all samples. Measurements were performed three times in different locations on the surface and averaged. Fluorescence spectra were obtained using a Jobin Yvon Fluorolog 3 spectrofluorometer in front face mode. The surface was placed at an angle of 20° to a line parallel to the plane of the detector. Photochemical Patterning of Carbohydrates. A 75-mesh transmission electron microscopy (TEM) grid (Electron Microscopy) was used as a photomask for all patterning experiments. DextranFITC (2000 kD) and 20 kD polysaccharide films were prepared as described above. Glucose (Aldrich) was spin-coated from a solution of 26 mg in 1 mL of acetonitrile at 3000 rpm for 90 s. One drop of a sucrose (Aldrich) solution containing 1.5 g in 1 mL H2O was placed on the substrate using a pipet. Approximately three-fourths of the drop was removed with a pipet. In all cases, the photomask was placed on top of the sugar film or droplet and pressed down with a quartz plate. Irradiation was carried out in an argon-filled glovebag with a desktop lamp containing a 300 nm Rayonet bulb for approximately 2 h. Samples were rinsed by sonication in H2O for 15 min, with the water and vial being changed every 5 min. Sonication was accompanied by extensive rinsing with water and methanol. Samples were blown dry with argon. Visualization of the Chemically Patterned Surface. Patterns were visualized by condensing water onto the pattern and imaging with a Nikon Eclipse optical microscope equipped with an INSIGHT digital camera. Two methods were used to condense water onto the surface. In the first, the surface was exposed to an extended breath. In the second, the substrate was held over boiling water for approximately 10 s. Microarray Construction. Antigen preparations were dissolved in saline (0.9% NaCl) at a given concentration and were spotted as triplet replicate spots in parallel. The initial amount of antigen spotted was approximately 0.35 ng/spot and was diluted by serial dilutions of 1:5 thereafter (see also the microarray images inserted in Figure 8). A high-precision robot designed to produce cDNA microarrays (PIXSYS 5500C, Cartesian Technologies, Irvine, CA) was utilized to spot carbohydrate antigens onto chemically modified glass slides as described.1,6 Both FAST slides (Schleicher & Schuell, Keene, NH) and PAM slides were spotted. The printed FAST slides were air-dried and stored at RT. The printed PAM slides were subjected to UV irradiation to activate the photocoupling of carbohydrates to the surface. Photocoupling of Carbohydrates on the Chips. After microarray spotting, the PAM slides were air-dried and placed in a quartz tube. The sealed tube was subsequently purged with argon or nitrogen before irradiation. UV irradiation was conducted by placing the quartz tube under a desktop lamp containing a 300 nm Rayonet bulb

Photochemical Micropatterning of Carbohydrates for 1 h. Precaution was made to avoid skin and eye contact with the radiation during the irradiation process. Microarray Staining, Scanning, and Data-Processing. Immediately before use, the printed microarrays were rinsed and washed with phosphate-buffered saline (PBS) (pH 7.4) and with 0.05% Tween 20 five times, with 5 min of incubation in each washing step. They were then “blocked” by incubating the slides in 1% BSA in PBS containing 0.05% NaN3 at RT for 30 min. Antibody staining was conducted at RT for 1 h at given dilutions in 1% BSA/PBS containing 0.05% NaN3 and 0.05% Tween 20. Since a biotinylated anti-dextran antibody (mAb 16. 4.12E, adapted from the late Professor Elvin A. Kabat at Columbia University) and lectin Con A (EY Laboratories, San Mateo, CA) were applied in this study, streptavidin-Cy3 conjugate or streptavidin-Cy5 conjugate (Amersham Pharmasia) were applied to reveal the bound anti-dextran antibodies or lectin Con A, respectively. The stained slides were rinsed five times with PBS and with 0.05% Tween 20 after each staining step. A ScanArray 5000A standard biochip scanning system (Perkin Elmer, Torrance, CA), equipped with multiple lasers, emission filters, and ScanArray acquisition software, was used to scan the microarray. Fluorescence intensity values for each array spot and its background were calculated using ScanArray Express (Perkin Elmer, Torrance, CA).

Langmuir, Vol. 22, No. 6, 2006 2905

Acknowledgment. This material is based upon work supported by, or in part by, the U.S. Army Research Laboratory and the U.S. Army Research Office under Contract/Grant No. DA W911NF-04-1-0282, the National Science Foundation under Grant Nos. DMR-02-14263, IGERT-02-21589, and CHE-0415516 to N.J.T. and J.T.K. at Columbia University, and the Phil N. Allen Trust and the Herzenberg Trust to D.W. at Stanford University. This work used the shared experimental facilities that are supported primarily by the MRSEC Program of the National Science Foundation under Award No. DMR-0213574 and the New York State Office of Science, Technology and Academic Research (NYSTAR). G.T.C. acknowledges an IGERT fellowship. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. Supporting Information Available: Optical microscope images of patterns of PVA, poly(tert-butyl acrylate) (PTBA), and PS. This material is available free of charge via the Internet at http://pubs.acs.org. LA0531042