Protein Micropatterns Using a pH-Responsive Polymer and Light

Langmuir 2005, 21, 8389-8393

8389

Protein Micropatterns Using a pH-Responsive Polymer and Light Karen L. Christman and Heather D. Maynard* Department of Chemistry and Biochemistry and California Nanosystems Institute, 607 E. Charles E. Young Drive, Los Angeles, California 90095-1569 Received March 9, 2005. In Final Form: May 26, 2005 Protein and peptide microarrays are popular candidates for medical diagnostics because of the possibility for high sensitivity and simultaneous marker screening. To realize the potential of these arrays, new strategies for ligand patterning are needed. We report a method for patterning proteins that utilizes a pH-responsive polymer, deep ultraviolet (DUV) light, and a photoacid generator (PAG). Poly(3,3′diethoxypropyl methacrylate) (PDEPMA) contains reactive acetal side chains which are converted to aldehydes following treatment with acid. PDEPMA was spin-coated onto Si-SiO2 substrates and was either chemically deprotected with 1 M HCl or photochemically deprotected by exposure to DUV in the presence of triphenylsulfonium triflate. Conversion to aldehyde groups was confirmed with Purpald and by reaction with a green fluorescent hydroxylamine. Protein microarrays were demonstrated by incubating photochemically patterned surfaces with an aldehyde-reactive biotin followed by red fluorescent streptavidin. This methodology provides a new substrate for the precise patterning of both peptides and proteins for various biological applications including medical sensors.

Introduction It has been established that mRNA expression levels do not always correlate with protein expression or disease diagnosis. In contrast, levels of various proteins have been directly correlated with disease detection, progression, and prognosis.1 Thus, there is significant interest in protein detection. Many protein biomarkers are present at low concentrations; however, current diagnostic techniques that measure protein levels suffer from poor sensitivity. The emerging technology of protein micro- and nanoarrays, which require less ligand material and less analyte, offers the possibility of increased sensitivity.2,3 Improvements in assay sensitivity could lead to earlier disease detection and discovery of currently undetectable diagnostic biomarkers. Additionally, miniaturization provides the ability to analyze hundreds or thousands of markers simultaneously, producing a unique diagnostic pattern or biosignature for a patient or disease.3,4 Protein arrays have been typically developed using robotic printing techniques, which require physical spotting onto substrates to produce array features, thus limiting the resolution.5 Ultraviolet (UV) or visible irradiation provides an alternative way to selectively pattern films with well-defined features at both the micron and nanometer scale.6 For example, UV radiation has been employed to remove photolabile amine-protecting onitrobenzyl groups.7-9 The amine patterns produced could * To whom correspondence should be addressed. Tel.: (310) 267 5162; fax: (310) 825 7385; e-mail: [email protected]. (1) Rocken, C.; Ebert, M. P. A.; Roessner, A. Pathol., Res. Pract. 2004, 200, 69-82. (2) Silzel, J. W.; Cercek, B.; Dodson, C.; Tsay, T.; Obremski, R. J. Clin. Chem. 1998, 44, 2036-2043. (3) Wilson, D. S.; Nock, S. Angew. Chem., Int. Ed. 2003, 42, 494500. (4) Walter, G.; Bussow, K.; Lueking, A.; Glokler, J. Trends Mol. Med. 2002, 8, 250-253. (5) Choudhuri, S. J. Biochem. Mol. Toxicol. 2004, 18, 171-179. (6) Mendes, P. M.; Preece, J. A. Curr. Opin. Colloid. Interface Sci. 2004, 9, 236-248. (7) Nakagawa, M.; Ichimura, K. Colloids Surf., A 2002, 204, 1-7. (8) Vossmeyer, T.; Jia, S.; Delonno, E.; Diehl, M. R.; Kim, S. H.; Peng, X.; Alivisatos, A. P.; Heath, J. R. J. Appl. Phys. 1998, 84, 3664-3670.

be modified with biotin and subsequently used to pattern an antibiotin antibody.9 Formation of carboxylic acid derivatives has been achieved through exposure of alkysilane self-assembled monolayers (SAMs) to vacuum UV light,10,11 as well as through exposure of a SAM of a tertbutyl ester azobenzene alkanethiol derivative and a photoacid generator (PAG) to UV light.12 Polystyrene colloidal particles and nanoparticles were then patterned on the hydrophilic regions on the SAM.12 Hydrolysis of hyperbranched poly(tert-butyl acrylate)-coated surfaces using UV light and a PAG has also been reported.13 The resultant poly(acrylic acid) regions were subsequently covalently grafted with amine-functionalized dyes.13 Oxidation of thiol-terminated monolayers by exposure to UV light in the presence of oxygen has also been used for the patterning of proteins and gold nanoparticles.14-16 While these studies nicely illustrate methods to pattern surfaces, to our knowledge, patterning of proteins using lithography techniques to produce aldehyde groups has not been described. Formation of aldehydes as a result of phototransformation of chloromethylphenyl end groups on SAMs has been reported;17,18 however, attachment of proteins was not demonstrated. Aldehydes are a versatile group to attach biomolecules to surfaces because they react with many proteins and (9) Ryan, D.; Parviz, B. A.; Linder, V.; Semetey, V.; Sia, S. K.; Su, J.; Mrksich, M.; Whitesides, G. M. Langmuir 2004, 20, 9080-9088. (10) Hong, L.; Hayashi, K.; Sugimura, H.; Takai, O.; Nakagiri, N.; Okada, M. Surf. Coat. Technol. 2003, 169, 211-214. (11) Hong, L.; Sugimura, H.; Furukawa, T.; Takai, O. Langmuir 2003, 19, 1966-1969. (12) Lee, K.; Pan, F.; Carroll, G. T.; Turro, N. J.; Koberstein, J. T. Langmuir 2004, 20, 1812-1818. (13) Aoki, A.; Ghosh, P.; Crooks, R. M. Langmuir 1999, 15, 74187421. (14) Liu, J.; Hlady, V. Colloids Surf., B 1996, 8, 25-37. (15) Bhatia, S. K.; Hickman, J. J.; Ligler, F. S. J. Am. Chem. Soc. 1992, 114, 4432-4433. (16) Ichinose, N.; Sugimura, H.; Uchida, T.; Shimo, N.; Masuhara, H. Chem. Lett. 1993, 1961-1964. (17) Brandow, S. L.; Chen, M. S.; Fertig, S. J.; Chrisey, L. A.; Dulcey, C. S.; Dressick, W. J. Chem. Eur. J. 2001, 7, 4495-4499. (18) Brandow, S. L.; Chen, M. S.; Aggarwal, R.; Dulcey, C. S.; Calvert, J. M.; Dressick, W. J. Langmuir 1999, 15, 5429-5432.

10.1021/la050646a CCC: $30.25 © 2005 American Chemical Society Published on Web 07/27/2005

8390

Langmuir, Vol. 21, No. 18, 2005

peptides under mild conditions in aqueous solutions.19 This reactivity has been exploited on surfaces to attach amines via reductive amination to form stable bonds and selectively with aminooxy-modified compounds to form oxime-linked conjugates, hydrazides to form hydrazones, and β-amino thiols to form thiazolidines. For example, aminooxy and 1,2-amino-thiol terminated peptides or biotin have been conjugated to aldehyde-functionalized glass slides by spotting techniques.20 Alkylamines have been immobilized onto aldehyde-terminated SAMs via formation of the Schiff bases.21 4-H-Benzo[d][1,3]dioxinol terminated SAMs, which reveal aldehyde functionalities following application of an oxidative potential, have been utilized to coat streptavidin after reaction of the biotin hydrazide.22 Additionally, DNA arrays have been fabricated using aldehyde monolayers on both gold23 and glass24 by conjugating amine-terminated oligonucleotidies via reductive amination. We describe patterning of surfaces using chemical transformation lithography on polymer films to form aldehyde groups and subsequent attachment of biomolecules to form protein microarrays; this procedure should allow for higher resolutions than spotting techniques. Recently, we prepared poly(3,3′-diethoxypropyl methacrylate) (PDEPMA), a pH-responsive polymer.25 PDEPMA contains acetal groups, which upon exposure to acid in solution, hydrolyze to form aldehyde groups.26 This led us to hypothesize that the same transformation could be effected on the polymer films using a PAG and light. Furthermore, selective transformation to aldehydes through a mask should provide a reactive substrate for protein array formation. In this paper, we introduce this method and demonstrate site-specific micropatterning of the protein streptavidin onto a film of PDEPMA using deep UV (DUV) light and a PAG. Experimental Section Preparation of pH-Responsive Polymer. Monomer 3,3′diethoxypropyl methacrylate was synthesized as described in the literature.27 Using conventional free-radical polymerization with 2,2′-azobisisobutyronitrile (AIBN) as the initiator, the monomer was then polymerized with an initial ratio of monomer to AIBN of 25:1. The reaction was allowed to proceed under inert atmosphere for 12 h in toluene at 65 °C. The solvent was subsequently evaporated, yielding the product. The polymer identity was confirmed by 1H NMR using a Bruker ARX500 spectrometer. 1H NMR (toluene-d8, 500 MHz, all peaks were broad multiplets): δ 4.69 (-COOCH2CH2CH(OCH2CH3)2, 1H), 4.27 (-COOCH2CH2CH(OCH2CH3)2, 2H), 3.60, 3.49 (COOCH2CH2CH(OCH2CH3)2, 4H), 2.20-1.99 (CH2C(CH3)(COOCH2CH2CH(OCH2CH3)2), 4H), 1.40-1.18 (CH2C(CH3)(COOCH2CH2CH(OCH2CH3)2), 9H). The polymer was further characterized by gel permeation chromatography (GPC) conducted on a Shimadzu HPLC system equipped with a refractive index detector RID10A and two Polymer Laboratories PLgel 5-µm mixed D columns (with guard column). LiBr (0.1 M) in DMF at 40 °C was used as a solvent (flow rate: 0.6 mL/min). Near-monodisperse poly(methyl methacrylate) standards (Polymer Laboratories) were employed (19) Lemieux, G. A.; Bertozzi, C. R. Trends. Biotechnol. 1998, 16, 506-513. (20) Falsey, J. R.; Renil, M.; Park, S.; Li, S.; Lam, K. S. Bioconjugate Chem. 2001, 12, 346-353. (21) Horton, R. C.; Herne, T. M.; Myles, D. C. J. Am. Chem. Soc. 1997, 119, 12980-12981. (22) Yeo, W. S.; Mrksich, M. Adv. Mater. 2004, 16, 1352-1356. (23) Peelen, D.; Smith, L. M. Langmuir 2005, 21, 266-271. (24) Zammatteo, N.; Jeanmart, L.; Hamels, S.; Courtois, S.; Louette, P.; Hevesi, L.; Remacle, J. Anal. Biochem. 2000, 280, 143-150. (25) Hwang, J.; Maynard, H. D. Polym. Prepr. 2004, 45, 1083-1084. (26) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic Synthesis; Wiley-Interscience: New York, 1999. (27) Zabransky, J.; Houska, M.; Plichta, Z.; Kalal, J. Makromol. Chem. 1985, 186, 231-245.

Christman and Maynard for calibration. Chromatograms were processed with the EZStart 7.2 chromatography software resulting in a number-average molecular weight (Mn) for the polymer of 34 900. Chemical and Photochemical Patterning. For those experiments involving chemical deprotection, a 2% w/w solution of PDEPMA in chloroform was spin-coated onto Si-SiO2 (∼1000 Å silicon dioxide on a silicon wafer) substrates at 3000 rpm for 30 s with a ramp time of 0.5 s. A 2% w/w solution of PDEPMA in chloroform containing triphenylsulfonium triflate (Alrdich; 5 wt % PAG/polymer) was spin-coated onto Si-SiO2 substrates for use in photochemical patterning. PDEPMA coated substrates were either completely or partially immersed in 1 M HCl for 15 min to chemically deprotect the acetal groups. Samples were then rinsed with Millipore Milli-Q H2O. For photochemical deprotection, substrates coated with PDEPMA and PAG were either half covered with a steel shield, covered with a 1000 mesh nickel TEM grid (Structure Probe, Inc.), or completely uncovered and exposed to UV radiation for 2 min from a Deep UV Light Exposure Tool (248 nm; 4 mW/cm2; A B Manufacturing) at the University of California, Santa Barbara Nanofabrication Facility, which is part of the NSF-funded National Nanofabrication Infrastructure Network (NNIN). Film Characterization. PDEPMA film thicknesses were measured with a Filmetrics F20 thin film analyzer. Contact angle measurements were performed on protected and deprotected aldehyde surfaces following exposure to acid or UV light with a contact angle goniometer (First Ten Angstroms, Inc.) after placing a drop of 1 µL Milli-Q H2O on each sample. Conversion to aldehyde functionality was confirmed using Purpald (4-amino3-hydrazino-5-mercapto-1,2,4-triazole, 4-amino-5-hydrazino-4H1,2,4-triazole-3-thiol, Aldrich). Purpald solution was prepared by dissolving 10 mg in 1 mL of 1 N NaOH. Sixty microliters of the solution was then placed on samples coated with PDEPMA, PDEPMA plus PAG, PDEPMA exposed to acid, or PDEPMA plus PAG exposed to UV light. After 20 min, the solution was removed and absorbance was determined using a BioMate 5 (Thermospectronic) spectrophotometer. Unreacted Purpald solution was subtracted as the background. Visualization of Patterned Films. All reagents were diluted in Milli-Q H2O. To visualize the aldehyde surfaces, samples were first rinsed with Milli-Q H2O (3 × 5 min) and then were incubated with Alexa Fluor 488 hydroxylamine (C5-aminooxyacetamide, bis(triethylammonium) salt; 100 µg/mL; Molecular Probes), which is known to react with aldehydes, for 1 h at 37 °C. For protein patterning, deprotected films were incubated with a biotinylated, aminooxy containing aldehyde reactive probe (ARP, N-(aminooxyacetyl)-N′-(D-biotinoyl) hydrazine, trifluoroacetic acid salt; 2 mg/mL; Molecular Probes) for 1 h at 37 °C. Samples were again rinsed with Milli-Q H2O (3 × 5 min) and then were incubated with red fluorescent Alexa Fluor 568 conjugated streptavidin (5 µg/mL; Molecular Probes) for 1 h at 37 °C. Control samples were stained with either Alexa Fluor 568 conjugated streptavidin (1 h, 37 °C), biotin (no aminooxy; 0.5 mg/mL) for 1 h at 37 °C followed by Alexa Fluor 568 streptavidin (1 h, 37 °C), or ARP followed by Alexa Fluor 568 streptavidin that had been presaturated in solution with 0.5 mg/mL biotin (1 h, 37 °C). As a further control, samples which were exposed to UV light, but did not contain a PAG, were also stained with ARP and Alexa Fluor 568 streptavidin. To confirm that the patterned streptavidin retained bioactivity, samples were first stained with ARP and streptavidin and were finally incubated with fluorescein biotin (5 µg/mL; 1 h, 37 °C; Molecular Probes). Following final rinsing with Milli-Q H2O (3 × 5 min), fluorescence in all samples was visualized with a Nikon E400 fluorescent microscope equipped with a Spot Camera (7.2 Color Mosaic). Pictures were acquired and processed using Spot 3.5.1 software (Diagnostic Instruments). Pictures were normalized using unstained areas in each picture. The optical image of the TEM grid was obtained on the Nikon E400 microscope. Determining Protein Surface Density. Solutions of Alexa Fluor 568 streptavidin with the following dilutions were prepared: 8, 6, 4, and 2 µg/mL. A 5-µL drop of each dilution was placed on PDEPMA plus PAG samples, which had been floodexposed with DUV light. Three samples were prepared for each dilution. The samples were placed in an incubator at 37 °C and the water was allowed to evaporate, leaving the streptavidin

Protein Micropatterns Using a pH-Responsive Polymer

Langmuir, Vol. 21, No. 18, 2005 8391

Figure 1. Acetal groups of PDEPMA are converted to aldehydes through chemical deprotection via incubation with aqueous HCl or photochemical deprotection with a PAG and light. adsorbed on the surface. The above dilutions corresponded to a streptavidin surface density of 560, 420, 280, and 150 ng/cm2. Three pictures were taken with the 40× objective for each sample and included both a stained and unstained area. The fluorescence intensity was measured using NIH Image, and samples were normalized using the intensity of the unstained area in each picture. A standard curve was then generated using the average intensity of each dilution compared to the density of streptavidin. To determine the streptavidin surface density of the 18 × 18 µm arrays, NIH image was used to measure the fluorescence intensity of the array features, normalized to an unstained area. Fluorescence intensity was translated to surface density using the standard curve described above.

Results and Discussion Hydrolysis of acetal side chains in the pH-responsive polymer film to aldehydes (Figure 1) was explored. Chemical deprotection was employed to confirm that conversion to aldehydes occurs on the PDEPMA films, while photochemical deprotection was conducted to allow for site-specific conversion to aldehyde groups and subsequent attachment of proteins. The polymer films were immersed in acid to cause chemical deprotection. Alternatively, photochemical deprotection was induced by exposure of the film to DUV radiation in the presence of the PAG, triphenylsulfonium triflate. This PAG forms triflic acid upon exposure to DUV radiation.28 PDEPMA alone or PDEPMA plus PAG were spin-coated onto Si-SiO2 substrates, producing films with thicknesses of 190 ( 10 nm. The water contact angle for both films was 88 ( 1°. Samples consisting of solely PDEPMA were partially immersed in 1 M HCl to chemically deprotect one-half of the film. The contact angle for the portion of the film exposed to acid was 82 ( 2°. Those films containing PDEPMA plus PAG were exposed to DUV light to convert the acetal groups to aldehydes. Samples were exposed for 2 min to achieve maximum conversion to aldehydes, without resulting in significant damage to the polymer.29 Following deprotection, the water contact angle was 78 ( 3°. The contact angle of films exposed to HCl or DUV both exhibited a small decline indicative of the expected slight increase in hydrophilicity. To confirm the presence of the expected aldehyde functionality, samples were incubated with Purpald, which forms a purple product with a maximum absorbance between 520 and 560 nm.30 Purpald reaction and release of product is specific for aldehydes and does not produce a purple product with any other functional group, including ketones. Prior to exposure to acid or DUV light, samples coated with PDEPMA or PDEPMA plus PAG remained visibly unchanged. Furthermore, the removed Purpald solution exhibited no absorbance in the range of 450-600 (28) Dektar, J. L.; Hacker, N. P. J. Am. Chem. Soc. 1990, 112, 60046015. (29) Moore, J. A.; Choi, J. O. ACS Symp. Ser. 1991, 475, 156-192. (30) Hopps, H. B. Aldrichimica Acta 2000, 33, 28-30.

Figure 2. Visualization of (a) aqueous HCl and (b) DUV radiation deprotected PDEPMA films. Samples were stained with green fluorescent Alexa Fluor 488 hydroxylamine. Green fluorescence was visualized only at locations of acid or DUV exposure. Scale bar ) 25 µm.

nm. In contrast, the solutions on both chemically and photochemically deprotected samples were visibly purple after 20 min. Basic conditions of the Purpald assay resulted in ester cleavage, thus releasing the purple product into solution. The removed solution exhibited the expected absorbance at a λmax of 524 nm. The formation of a purple product provides evidence that conversion of the acetal groups to aldehydes occurs following both chemical and photochemical deprotection. Aldehydes react with aminooxy-modified compounds selectively to form oxime-linked conjugates in aqueous solution.31 Compared to reaction with amines in aqueous solutions which necessitates reduction to form a stable linkage,19 the reaction requires no other reagents and the (31) Jencks, W. P. J. Am. Chem. Soc. 1959, 81, 475-481.

8392

Langmuir, Vol. 21, No. 18, 2005

Christman and Maynard

Figure 3. Protein patterning on PDEPMA film. Following exposure to DUV light through a mask (1), acetal groups were converted to aldehydes (2). A biotinylated hydroxylamine (ARP) was reacted with converted aldehydes (3). Films were then incubated with red fluorescent streptavidin (SA) which bound specifically to the attached biotin (4).

linkage is stable at physiological pH.32 Therefore, to visualize the deprotected films, samples were stained with a green fluorescent hydroxylamine. For those samples that were chemically deprotected, the half of the film exposed to 1 M HCl exhibited strong green fluorescence, while the remaining half showed no fluorescence (Figure 2a). Films which were photochemically deprotected likewise displayed strong fluorescence (Figure 2b). This data provides further evidence that aldehydes are formed. Also, these results illustrate that aminooxy-containing compounds will selectively attach to exposed regions of the film, indicating that the films could be used to produce protein microarrays. Micropatterning of proteins on the polymer surface was demonstrated by exposing PDEPMA plus PAG films to DUV light through a 1000-mesh Ni TEM grid. Following exposure, the entire film was incubated with a biotin containing aminooxy functionality (ARP) followed by red fluorescent streptavidin (Figure 3). ARP reacts with aldehydes to form oxime linkages and thus should attach only to regions exposed to DUV light. Streptavidin binds strongly to biotin, and as expected, strong red fluorescence was observed only in areas exposed to DUV light through openings in the TEM grid (Figure 4c). The streptavidin was arrayed into 18 × 18 µm patterns with close fidelity to the original TEM grid mask (Figure 4a). The protein surface density in the spots was approximately 200 ng/cm2, which is close to the surface density of a twodimensional crystal monolayer of streptavidin as measured by electron crystallography.33 This data demonstrates that streptavidin is successfully patterned onto the surface of PDEPMA films by first attaching the ligand for the protein. A high surface density of streptavidin at the regions exposed to DUV results, indicating that the reaction of the aminooxy-biotin to the aldehyde-functionalized surface is efficient. (32) Rose, K. J. Am. Chem. Soc. 1994, 116, 30-33.

As controls, films were stained with biotin (no aminooxy) followed by Alexa Fluor 568 streptavidin or just Alexa Fluor 568 streptavidin (Figure 4b). Neither film exhibited strong fluorescence. Samples stained first with ARP and then with streptavidin preincubated with biotin displayed even less fluorescence. In addition, samples which were exposed to DUV light but did not contain the PAG also displayed minimal fluorescence following incubation with ARP and Alexa Fluor 568 streptavidin. Taken together, these results suggest that PAG-induced hydrolysis to aldehydes must first take place and that biotin then conjugates to the surface specifically via aminooxy functionality, likely through an oxime linkage. Nonspecific adsorption of streptavidin on the surface is minimal and the background fluorescence can be removed with image processing for easier visualization (Figure 4d). Streptavidin has four binding sites for biotin,34 and thus the array produced on the PDEPMA surface may be utilized to bind other biotinylated proteins or antibodies. However, for this to be viable, the surface-bound streptavidin must maintain bioactivity; namely, the protein arrayed on the surface should still bind biotin. To test this, the streptavidin array was incubated with fluoresceinmodified biotin to confirm bioactivity of the patterned protein. As anticipated, the green fluorescent biotin bound specifically to locations of arrayed streptavidin (Figure 4e), thus illustrating that this methodology for site-specific attachment does not suppress protein bioactivity. These results indicate that the streptavidin arrays may be used to bind biotinylated ligands for biosensors and other similar applications. We report direct evidence that proteins can be patterned into microarrays on PDEPMA films. We demonstrated (33) Darst, S. A.; Ahlers, M.; Meller, P. H.; Kubalek, E. W.; Blankenburg, R.; Ribi, H. O.; Ringsdorf, H.; Kornberg, R. D. Biophys. J. 1991, 59, 387-396. (34) Weber, P. C.; Ohlendorf, D. H.; Wendoloski, J. J.; Salemme, F. R. Science 1989, 243, 85-88.

Protein Micropatterns Using a pH-Responsive Polymer

Langmuir, Vol. 21, No. 18, 2005 8393

that exposure of the polymer film to DUV radiation in the presence of a PAG results in conversion to aldehyde groups. Aldehyde-specific probes including biotinylated ARP and Alexa Fluor 488 hydroxylamine conjugated to exposed areas without the need for any other reagent, likely by forming the oxime bonds. The conversion to aldehydes can be made spatially specific by use of a mask, serving as an excellent template to produce microarrays of proteins. In this way, streptavidin, which binds strongly to biotin, was micropatterned at high density on the polymer surface while retaining its bioactivity. The spot size for the array produced in this study was 18 × 18 µm; however, the minimal resolution for the methodology presented here is in the submicrometer range corresponding to the low absorbance wavelength of triphenylsulfonium triflate. Therefore, smaller feature sizes may also be possible. This methodology should be readily extendable to patterning other proteins. For example, biotinylated proteins could be attached to the arrays produced in this study because of the multiple binding sites for biotin available on the streptavidin. Other aminooxy-modified peptides and proteins could also be arrayed by direct reaction with the aldehydes. Preferential reaction with aminooxy compounds occurs in aqueous solutions in the presence of amines.19,32 Thus, it may be possible to use the aldehyde-functionalized surfaces to conjugate aminooxymodified peptides and proteins in a specific orientation. Nonetheless, the unmodified ligands could also be reacted with aldehyde-patterned polymer films through amine functionality in the presence of a reducing agent such as sodium cyanoborohydride to form the stable bonds. Summary and Conclusions We have demonstrated a novel method for protein patterning that utilizes a pH-responsive polymer film. The polymer PDEPMA contains reactive acetal groups which hydrolyze to aldehyde groups following incubation with acid or exposure to DUV in the presence of a PAG. The use of a PAG and DUV light allows for the spatially resolved conversion to aldehyde groups and patterning of streptavidin specifically at locations of light exposure. Because aldehydes are formed, this methodology should be easily extended to other proteins and peptides. Also, different patterns and feature sizes should be possible, leading to many applications in biotechnology. Figure 4. Site-specific protein patterning. Films consisting of PDEPMA plus PAG were exposed to UV light through a 1000mesh nickel TEM grid (a), which was used as a mask. Nonspecific binding of red streptavidin to exposed films was minimal (b). Red fluorescent streptavidin was specifically patterned to locations of UV exposure after first incubating with a biotinylated aldehyde reactive probe (c). The background can easily be removed with image processing for easier visualization (d). Fluorescein biotin bound specifically to locations of patterned streptavidin, indicating the protein is bioactive (e). Scale bar ) 25 µm.

Acknowledgment. This work was supported by the NSF (DMI-0327077) and a CNSI/Hewlett-Packard Postdoctoral Scholarship. We thank Professor Yong Chen for kindly providing the Si-SiO2 substrates and for the use of his Nikon microscope. We would also like to thank Vimary Va´zquez-Dorbatt for helping with the polymerization. LA050646A