Using Liquid Crystals to Amplify Protein−Receptor Interactions: Design

Langmuir 2003, 19, 1671-1680

1671

Using Liquid Crystals to Amplify Protein-Receptor Interactions: Design of Surfaces with Nanometer-Scale Topography that Present Histidine-Tagged Protein Receptors† Yan-Yeung Luk,‡ Matthew L. Tingey,‡ David J. Hall,§,| Barbara A. Israel,⊥ Christopher J. Murphy,# Paul J. Bertics,§ and Nicholas L. Abbott*,‡ Department of Chemical Engineering, Department of Surgical SciencesSchool of Veterinary Medicine, Department of Pathobiological SciencesSchool of Veterinary Medicine, and Department of Biomolecular Chemistry, University of WisconsinsMadison, Madison, Wisconsin 53706 Received June 28, 2002. In Final Form: December 24, 2002 This paper reports a strategy for the oriented immobilization of protein receptors on gold films possessing nanometer-scale topographies and the detection of protein binding events to these receptors by using liquid crystals. The approach revolves around the use of self-assembled monolayers (SAMs) formed from nitrilotriacetic acid (NTA)-terminated alkanethiols, 1, and tri(ethylene glycol)-terminated alkanethiols, 2. The SAMs are formed on ultrathin gold films that are deposited from a vapor onto silica substrates oriented at an oblique angle of incidence. Single-component SAMs formed from 2 on these gold films resist nonspecific protein adsorption (using cell lysates) and promote uniform planar anchoring of the nematic liquid crystal, 4-cyano-4′-pentylbiphenyl (5CB). Surprisingly, the azimuthal orientation of nematic 5CB is parallel to the direction of maximum roughness within the gold film when using SAMs formed from 2 but perpendicular to the direction of maximum roughness when tetra(ethylene glycol)-terminated SAMs are formed on the gold films. Mixed SAMs formed from 1 and 2 bind the hexahistidine-tagged protein MEK via specific complexation of the hexahistidine tags of MEK to the NiII-NTA complexes on the surface. When gold films are prepared by oblique deposition at an angle of 30° from the normal, we measure bound MEK to disrupt the uniform orientation of 5CB, thus leading to an easily visualized change in the optical appearance of the liquid crystal. However, by using gold films deposited at an angle of 40° from the normal, we report that bound MEK does not disrupt the alignment of the liquid crystal whereas anti-MEK IgG bound to the MEK does lead to a nonuniform alignment. These results, when combined with appropriate control experiments, suggest that nanostructured surfaces presenting NTA and ethylene glycol terminated SAMs form a useful interface for imaging proteins bound to histidine-tagged, surface-immobilized receptors.

Introduction The immobilization of oriented and functional proteins and the detection of specific binding events to immobilized proteins are two key challenges (among many) that must be addressed if surface-based proteomics tools are to be successfully developed.1 These tools possess the potential to substantially accelerate investigations aimed at understanding issues such as the roles of populations of regulatory proteins in cell signaling processes. Existing surface-based protein assays, such as enzymelink immunoassays (ELISA), typically exploit hydrophobic surfaces to promote the adsorption of desired proteins and use adsorbed bovine serum albumin (BSA) to prevent adsorption of nontargeted proteins.2 The surfaces used in these assays do not generally possess well-defined to* To whom correspondence should be addressed. Phone: (608) 265-5278. Fax: (608) 262-5434. E-mail: [email protected]. † Part of the Langmuir special issue entitled The Biomolecular Interface. ‡ Department of Chemical Engineering. # Department of Surgical Science-School of Veterinary Medicine. ⊥ Department of Pathobiological Science-School of Veterinary Medicine. § Department of Biomolecular Chemistry. | Current address: Department of Chemistry, Lawrence University, Appleton, WI 54911. (1) (a) Zhu, H.; Bilgin, M.; Bangham, R.; Hall, D.; Casamayor, A.; Bertone, P.; Lan, N.; Jansen, R.; Bidlingmaier, S.; Houfek, T.; Mitchell, T.; Miller, P.; Dean, R. A.; Gerstein, M.; Snyder, M. Science 2001, 293, 2101. (b) MacBeath, G.; Schreiber, S. L. Science 2000, 289, 1760.

pographies. Recently, however, several studies have reported on opportunities relevant to surface-based protein analysis tools that follow from the use of surfaces with well-defined chemistries and topographies.3-11 For example, a series of studies have demonstrated that selfassembled monolayers (SAMs) formed from ω-functionalized alkanethiols on gold can provide well-defined surfaces that resist nonspecific protein adsorption and cell attachment.3-5 We recently reported the fabrication of gold films with an anisotropic surface structure on the nanometer scale by physical vapor deposition of gold at an oblique angle of incidence and the formation of SAMs of ω-functionalized alkanethiols on these surfaces.6-11 These polycrystalline gold films possess both in-plane and out-of-plane crystal(2) Van Oss, C. J.; van Regenmortel, M. H. V. Immunochemistry; Marcel Dekker: New York, 1994. (3) (a) Gupta, V. K.; Abbott, N. L. Science 1997, 276, 1533. (b) Shah, R. R.; Abbott, N. L. J. Am. Chem. Soc. 1999, 121, 11300. (c) Shah, R. R.; Abbott, N. L. Science 2001, 293, 1296. (4) (a) Prime, K. L.; Whitesides, G. M. Science 1991, 252, 11641167. (b) Prime, K. L.; Whitesides, G. D. J. Am. Chem. Soc. 1993, 115, 10714-10721. (5) Luk, Y. Y.; Kato, M.; Mrksich, M. Langmuir 2000, 16, 9604. (6) Gupta, V. K.; Skaife, J. J.; Dubrovsky, T. B.; Abbott, N. L. Science 1998, 279, 2077. (7) Skaife, J. J.; Abbott, N. L. Langmuir 2000, 16, 3529. (8) Skaife, J. J.; Abbott, N. L. Langmuir 2001, 17, 5595. (9) Skaife, J. J.; Brake, J. M.; Abbott, N. L. Langmuir 2001, 17, 5448. (10) Gupta, V. K.; Abbott, N. L. Langmuir 1996, 12, 2587. (11) Everitt, D. L.; Miller, W. J.; Abbott, N. L.; Zhu, X. D. Phys. Rev. E 2000, 62, R4833.

10.1021/la026152k CCC: $25.00 © 2003 American Chemical Society Published on Web 02/06/2003

1672

Langmuir, Vol. 19, No. 5, 2003

Luk et al.

NTA-NiII complex can specifically bind proteins that present a sequence of six histidines (His-tag) at the amino end of the protein.12 His-tags are commonly incorporated into the amino end of the primary sequence of recombinant proteins to facilitate purification by metal-affinity chromatography.13 The tri(ethylene glycol)-terminated alkanethiol, 2, is used as a diluent of the NTA-terminated alkanethiol so as to minimize nonspecific protein adsorption.12 Figure 1. Schematic illustration of surface topography with and without protein bound to a SAM supported on a surface possessing nanometer-scale topography. The approximate wavelength and amplitude of the idealized corrugation of the gold film are indicated. The approximate topography of the surface supporting the bound protein is indicated by the dashed line.

lographic textures and an anisotropic topography that can be idealized as a corrugation with an amplitude of 1-2 nm and a wavelength of 10-40 nm.9,11 We have used these gold films to demonstrate that control of the nanometer-scale topography of surfaces can provide the basis of new principles for detection of protein-receptor interactions6 (and other chemical phenomena3). On such surfaces, nematic liquid crystals such as 4-cyano-4′pentylbiphenyl (5CB) and 4-methoxybenzylidene-4′butylaniline (MBBA) adopt a uniform planar orientation in an azimuthal direction that is either parallel or perpendicular to the direction of maximum roughness on these surfaces. The orientation depends on the specific functionality and orientation of the terminal groups of the SAMs.10 Thus, the orientation of the liquid crystal reports information about the chemical functionality of the SAMs and the nanometer-scale topography of the surface.3 Because the sizes of many proteins are comparable to the spatial scale of the topography of the surface, proteins bound to these SAMs can mask or erase the topography of the surface (Figure 1). The uniform alignment of liquid crystals, which is observed on SAMs not supporting bound protein, is disrupted by the presence of bound protein. In our past studies, we have also demonstrated that manipulation of the angle of deposition of the gold can be used to engineer the nanometer-scale topography of the gold films and thus tune the response of the liquid crystal to bound protein.9 In this paper, we report the results of an investigation that aimed to go beyond our past studies in several respects. First, we aimed to demonstrate a general scheme for the oriented immobilization of a range of protein receptors on surfaces that possess nanometer-scale topographies. Second, we aimed to explore strategies that would permit control of nonspecific adsorption of proteins on surfaces with nanometer-scale topography, while still permitting the immobilization of a particular protein receptor of interest. Third, we aimed to demonstrate that an experimental system that fulfilled the above criteria could be used in combination with liquid crystals to report protein binding events involving the immobilized receptor. Here we demonstrate this possibility by using a model system involving protein (MEK)-antibody (anti-MEK IgG) interactions. The approach reported in this paper is guided by past work of Whitesides and co-workers12 and uses mixed SAMs presenting nitrilotriacetic acid (NTA) (1) and tri(ethylene glycol) (2) (Figure 2). We used NTA because, when complexed with Ni(II) ions via a tetravalent chelate, the (12) Sigal, G. B.; Bamdad, C.; Barberis, A.; Strominger, J.; Whitesides, G. M. Anal. Chem. 1996, 68, 490-497.

Materials and Methods Materials. The glass microscope slides were Fisher’s Finest, premium grade obtained from Fisher Scientific (Pittsburgh, PA). N-benzyloxycarbonyl-L-lysine tert-butyl ester hydrochloride was purchased from Calbiochem-NovaBiochem Corp. (San Diego, CA). The nematic liquid crystal of 5CB, manufactured by BDH, was purchased from EM Industries (Hawthorne, NY). All other chemicals were purchased from Aldrich Chemicals (Milwaukee, WI) and used as received. His-tag MEK (55 kDa, the hexahistidine-tagged fusion protein containing the full length of MEK-1 of human) and anti-MEK IgG (∼150 kDa, polyclonal, developed in rabbit) were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); antistreptavidin IgG (polyclonal, developed in rabbit) was purchased from Sigma (St. Louis, MO) and used as received. All proteins used in this study were dissolved in phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4‚7H2O, 1.4 mM KH2PO4) solution at pH 7.2. All aqueous solutions were prepared with deionized water having a resistivity greater than 18.2 MΩ cm (Milli-Qplus, Millipore, Bedford, MA). 3T3-F442A Murine fibroblasts from American Type Tissue Culture Collection (Manassas, VA) were grown in Dulbecco’s modification of Eagle’s medium (DMEM) (Mediatech, VA) with 10% calf serum at 37 °C and 5% CO2. After washing the cells once with PBS, the cells were harvested by scraping and sonicated for 45 s in PBS containing 0.1% Triton X-100. The final concentration was 106 cells/mL. Synthesis. The NTA-terminated alkanethiol 1 was prepared by a convergent synthetic scheme comprised of routes A and B (Figure 3). In route A, the ammonium group of benzyl carbamate protected lysine is protected by the bromoacetic acid tert-butyl ester. A palladium-catalyzed hydrogenation is used to afford the primary amine, 4, for further coupling.14 In route B, n-undecylenic bromide was treated with neat tri(ethylene glycol) and sodium hydroxide to give monoalkylated tri(ethylene glycol), 5.15 The tri(ethylene glycol)-terminated alkene was treated with ethyl diazoacetate followed by a basic hydrolysis to give the desired acid, 7.16 The acid, 7, and amine, 4, were then coupled by a DCC/ NHS reaction to afford the desired alkene, 8.14 The terminal olefins of 8 were converted to the corresponding thioacetates, 9, by treatment with thiolacetic acid and AIBN under photolytic conditions, followed by an acidic hydrolysis under reflux to give the desired NTA-terminated alkanethiols 1. Nr,Nr-Bis[(tert-butyloxycarbonyl)methyl]-NE-benzyloxycarbonyl-L-lysine tert-Butyl Ester (3). A solution of N-benzyloxycarbonyl-L-lysine tert-butyl ester hydrochloride (813 mg, 2.181 mmol), triethylamine (2 mL) and bromoacetic acid tert-butyl ester (4.26 g, 21.81 mmol) was stirred in 45 mL of DMF under nitrogen for 3 days at 65 °C. The solvent and the excess of bromoacetic acid tert-butyl ester were removed in vacuo, and the remaining oil was extracted six times with hexane. The combined organic phases were then collected and concentrated in vacuo. Purification by flash chromatography (5% EtOAc/ hexane) gave 738 mg (1.31 mmol, 60%) of the desired dicarboxylate as a colorless oil: 1H NMR (250 MHz, CDCl3) δ1.42 (s, 18H), 1.45 (s, 9H), 1.53 (m, 4 H), 1.64 (m, 2H), 3.19 (q, 2 H), 3.31 (t, 1H), 3.46 (dd, 4H), 5.08 (s, 2H), 7.32 (m, 5H). (13) Hochuli, E.; Dobeli, H.; Schacher, A. J. J. Chromatogr. 1987, 411, 177-184. (14) Dorn, I. T.; Neumaier, K. R.; Tampe´, R. J. Am. Chem. Soc. 1998, 120, 2753. (15) Pale-Grosdemange, C.; Simon, E. S.; Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1991, 113, 12. (16) Houseman, B. T.; Mrksich, M. J. Org. Chem. 1998, 63, 7552.

Amplifying Protein-Receptor Interactions

Langmuir, Vol. 19, No. 5, 2003 1673

Figure 2. General scheme for immobilization of proteins on mixed SAMs formed from NTA-terminated alkanethiols, 1, and tri(ethylene glycol)-terminated alkanethiols, 2. Each nitrilotriacetic acid group on the SAM chelates a Ni(II) ion, which is subsequently recognized and bound by the histidine tags of the protein to be immobilized. The histidine-tagged protein functions as a receptor that binds a second protein. The tri(ethylene glycol)-terminated SAM resists nonspecific protein adsorption.

Figure 3. Scheme for synthesis of NTA-terminated alkanethiols 1: (a) i. triethylamine, ii. BrCH2CO2tBu; (b) H2, Pd/C; (c) i. NaOH, ii. triethylene glycol, 100 °C; (d) i. N2CH2CO2Et, ii. BF3Et2O, 0 °C; (e) NaOH/CH3OH; (f) i. DCC/NHS, ii. amine 4; (g) CH3COSH, AIBN, hν, THF; (h) H3O+/MeOH/H2O, reflux. Nr,Nr-Bis[(tert-butyloxycarbonyl)methyl]-L-lysine tertButyl Ester (4). The protected amine 3 (485 mg, 0.859 mmol) was dissolved in 50 mL of EtOH containing 15 mg of Pd/C (5%

Pd) as a suspension. The flask was sealed with a septum, and the gas in the flask was exchanged with hydrogen by applying a vacuum and a stream of H2 repeatedly three times. The reaction

1674

Langmuir, Vol. 19, No. 5, 2003

mixture was stirred under 1 atm of hydrogen overnight. The solid catalyst was filtered off, and the reaction mixture was extracted twice with 1 N NaOH. The combined organic phases were dried over anhydrous magnesium sulfate, and the organic solvent was removed in vacuo to give 332 mg (0.773 mmol, 90%) of amine 4: 1H NMR (250 MHz, CDCl3) δ1.43 (s, 18H), 1.44 (s, 9H), 1.46 (m, 2H), 1.63 (m, 2 H), 1.73 (m, 2H), 2.68 (t, 2H), 3.29 (t, 2H), 3.45 (dd, 4H). 11-Tri(ethylene glycol)-1-undecene (5) was synthesized as described in a previous work by Whitesides.15 Briefly, 11bromo-1-undecene was added to a solution containing excess tri(ethylene glycol) (neat, 20 equiv) and NaOH (50 wt %, 1 equiv) over 5 min. After stirring at 100 °C for 12 hours, the solution was mixed with hexane and water and neutralized with HCl (1 N). The aqueous phase was then extracted six times with hexane; the combined organic phases were dried with MgSO4, concentrated in vacuo, and purified by flash chromatography (ethyl acetate) to give olefin 5 as a clear oil: 1H NMR (250 MHz, CDCl3) δ1.26-1.29 (br s, 12H), 1.53-1.58 (qui, 2H), 2.00-2.04 (dd, 2H, J ) 7.02, 6.77 Hz), 3.43 (t, 2H, J ) 6.79 Hz), 3.57-3.76 (m, 12H), 4.88-4.96 (m, 2H), 5.74-5.84 (m, 1H). Tri(ethylene glycol) Ethyl Ester (6). A solution of 11-tri(ethylene glycol)-1-undecene 57 (1.02 g, 3.363 mmol) in dry dichloromethane (20 mL) was stirred at 0 °C for 10 min, followed by addition of ethyl diazoacetate (0.85 mL, 6.7246 mmol) and BF3‚Et2O (35 µL, 0.3363 mmol). After the mixture was stirred for 30 min at 0 °C, saturated aqueous ammonium chloride (10 mL) was added and the reaction mixture was placed in a separatory funnel. The organic phase was collected, and the aqueous phase was extracted with dichloromethane (5 × 30 mL). The combined organic phases were dried over magnesium sulfate and concentrated in vacuo to give a yellow oil. Purification by flash chromatography using gradient elution (1:1 ethyl acetate/ hexane f ethyl acetate) gave 783 mg (2.018 mmol, 60%) of ester 3 as a clear oil: 1H NMR (250 MHz, CDCl3) δ1.29-1.33 (m, 15H), 1.53-1.58 (m, 2H), 2.00-2.04 (dd, 2H, J ) 7.02, 6.77 Hz), 3.43 (t, 2H, J ) 6.79 Hz), 3.49-3.76 (m, 12H), 4.13 (s, 2H), 4.20 (q, 2H, J ) 7.14 Hz), 4.88-4.96 (m, 2H), 5.74-5.84 (m, 1H). Acid (7). A solution of ester 6 (304.5 mg, 0.785 mmol) in 30 mL of methanol/THF (50/50) and 2 mL of 1 N aqueous NaOH was stirred at room temperature for 4 h. The reaction mixture was then cooled to 0 °C and acidified by adding slowly over a period of 5 min ca. 2 mL of 1 N aqueous HCl. The solution was then extracted 4 times with ethyl acetate, rinsed with brine, dried over magnesium sulfate, and concentrated in vacuo to give 240 mg (0.667 mmol, 85%) of the desired acid 7 as a colorless oil: 1H NMR (250 MHz, CDCl ) δ1.29-1.33 (m, 12H), 1.53-1.58 (m, 3 2H), 2.00-2.04 (dd, 2H, J ) 7.02, 6.77 Hz), 3.43 (t, 2H, J ) 6.79 Hz), 3.49-3.76 (m, 12H), 4.13 (s, 2H), 4.88-4.96 (m, 2H), 5.745.84 (m, 1H). ESI/EMM calculated for [MNa]+, 383.2410; found, 383.2. Amide (8). A solution of acid 7 (437 mg, 1.21 mmol) and N-hydroxysuccinimide (NHS) (153 mg, 1.33 mmol) in 10 mL of DMF was mixed with dicyclohexylcarbodiimide (DCC) (275 mg, 1.33 mmol) in 15 mL of dry CH2Cl2. The mixture was stirred under nitrogen for ca. 1 h to give a white precipitate. The precipitate was filtered, and the filtrate was added directly to a solution of amine 4 (574 mg, 1.33 mmol) and triethylamine (0.19 mL, 1.33 mmol) in mL of DMF. The reaction mixture was stirred for 5 h and concentrated in vacuo. Purification by flash chromatography (5% ethyl acetate/hexane) gave 795 mg (1.03 mmol, 85%) of the desired amide 8 as a white solid: 1H NMR (250 MHz, CDCl3) δ1.29-1.33 (m, 14H), 1.4 (br s, 29) 1.53-1.58 (m, 2H), 1.72-1.75 (m, 2H), 2.00-2.04 (dd, 2H, J ) 7.02, 6.77 Hz), 3.23.22 (m, 6H) 3.43-3.45 (m, 3H), 3.49-3.76 (m, 12H), 4.13 (s, 2H), 4.88-4.96 (m, 2H), 5.74-5.84 (m, 1H). ESI/EMM calculated for [MNa]+, 795.5347; found, 795.5. Thioacetate (9). A solution of olefin 8 (410 mg, 0.79 mmol) in dry THF (20 mL) containing thiolacetic acid (0.13 mL, 1.96 mmol) and AIBN (13 mg, 0.077 mmol) was irradiated in a photochemical reactor (Rayonet reactor lamp, Southern New England Ultraviolet Co., model no. RPR-100) for 5 h under nitrogen (ca. 1 atm). Concentration of the reaction mixture in vacuo, followed by flash chromatography (30:1 CH2Cl2/MeOH), gave compound 4 as a clear oil (438 mg, 0.733 mmol, 93%): 1H NMR (250 MHz, CDCl3) δ1.29-1.33 (m, 14H), 1.4 (br s, 29) 1.53-

Luk et al. 1.58 (m, 2H), 1.72-1.75 (m, 2H), 2.00-2.04 (dd, 2H), 2.3 (s, 3H), 2.8 (t, 2H), 3.2-3.22 (m, 6H), 3.43-3.45 (m, 3H), 3.49-3.76 (m, 12H), 4.13 (s, 2H). ESI/EMM calculated for [MNa]+, 871.5330; found, 871.5. NTA-Terminated Alkanethiol (1). To a solution of 9 (87 mg, 0.146 mmol) in 20 mL of 1:1 methanol/H2O was added 5 drops of 12 N HCl. The solution was refluxed for 8 h and concentrated in vacuo. The residue was extracted with ether (6 × 10 mL), dried over sodium sulfate, and concentrated in vacuo. Purification by flash chromatography with CH2Cl2/MeOH/H2O (69:27:4) gave 49 mg (0.077 mmol, 53%) of the desired alkanethiol 1 as a colorless oil: 1H NMR (250 MHz, CD3OD) δ1.29-1.33 (m, 16H), 1.46 (m, 2H), 1.53-1.58 (m, 6H), 2.55 (t, 2H), 3.2-3.22 (m, 6H), 3.43-3.45 (m, 2H), 3.49-3.76 (m, 13H), 4.13 (s, 2H). ESI/ EMM calculated for [MNa]+, 661.3346; found, 661.3. Cleaning of Substrates. Microscope slides were cleaned sequentially in piranha (70% H2SO4, 30% H2O2) and base solutions (70% KOH, 30% H2O2) using nitrogen to provide agitation (1 h at ∼80 °C). Warning: Piranha solution should be handled with extreme caution; in some circumstances, most probably when it has been mixed with significant quantities of an oxidizable organic material, it has detonated unexpectedly. The slides were then rinsed thoroughly in deionized water (18.2 MΩ cm), ethanol, and methanol and dried under a stream of nitrogen. The clean slides were stored in a vacuum oven at 110 °C. All other glassware was cleaned in piranha solution prior to use. Uniform Deposition of Gold Films. Semitransparent films of gold with thicknesses of ∼100 Å were deposited onto glass slides mounted on rotating planetaries (no preferred direction or angle of incidence) by using an electron beam evaporator (VES3000-C manufactured by Tek-Vac Industries, Brentwood, NY). The rotation of the substrates on the planetaries ensured that the gold was deposited without a preferred direction of incidence. We refer to these gold films as “uniformly deposited gold films”. A layer of titanium (thickness of ∼50 Å) was used to promote adhesion between the glass microscope slide and the film of gold. The rates of deposition of gold and titanium were ∼0.2 Å/s. The pressure in the evaporator was less than 5 × 10-7 Torr before and during each deposition. The gold source was periodically cleaned by sequentially immersing it in aqua regia (70% HNO3, 30% HCl) and piranha solutions at 50 °C (30 min in each solution). The cycle was repeated three to four times, rinsing between cycles in deionized water. Oblique Deposition of Gold. Semitransparent films of gold with thicknesses of ∼130 Å were deposited onto glass microscope slides mounted on stationary holders by using the electron beam evaporator described above. The gold was deposited from a fixed direction of incidence and a fixed angle of incidence of either 30° or 40° (measured from the normal of the surface). We refer to these gold films as “obliquely deposited gold films”. A layer of titanium (thickness of ∼50 Å) was used to promote adhesion between the glass and the film of gold. Preparation of SAMs and Binding of Proteins. SAMs were formed on the surfaces of uniformly and obliquely deposited films of gold by immersing the films in ethanolic solutions containing 1 mM mixed alkanethiols 1 and 2 for 12 h. The percentages of 1 in ethanol were either 5% or 20%. After being rinsed in ethanol and dried under nitrogen, the SAMs were immersed in a 40 mM aqueous solution of NiSO4 (pH 7.2) for 1 h. The SAMs were then rinsed with a PBS solution (pH 8.2) for 10 s and dried with a stream of nitrogen. Protein binding was carried out by incubation of SAMs in PBS containing proteins (pH 7.4) for various lengths of time (indicated with each result). We used solutions containing 0.1 µM of each protein (His-tag MEK, anti-MEK IgG, and anti-streptavidin IgG). The substrates were rinsed with PBS (pH 8.2) and dried under a stream of nitrogen following removal from each solution of protein. X-ray Photoelectron Spectroscopy (XPS). X-ray photoelectron spectroscopy (Surface Science, Mountain View, CA) was used to determine the composition of SAMs formed by alkanethiols 1 chelated with Ni(II) ions.17 The chamber pressure during acquisition of all XPS spectra was ∼1.5 × 10-9 Torr, and the spot size was 250 µm × 1000 µm. The intensity of X-rays (17) Bain, C. D.; Whitesides, G. M. Langmuir 1989, 5, 1370.

Amplifying Protein-Receptor Interactions emitted from the Al KR anode varied by ∼2% over a 24 h period. Survey scans were acquired at each location on a sample. Acquisition was performed in Resolution 4 with 100 scans centered on the N 1s and Ni 2p3 peaks.18 Ellipsometry. Ellipsometric measurements were performed to determine the optical thicknesses of SAMs and films of proteins. The optical thickness reported for each sample is the average of 4-5 measurements performed at different locations on each substrate. The measurements were performed using a Rudolph Auto EL ellipsometer (Flanders, NJ) at a wavelength of 6320 Å and an angle of incidence of 70°. The gold substrates used for ellipsometric measurements were 1000 Å in thickness (Au) and were uniformly deposited. The ellipsometric thicknesses of SAMs and immobilized proteins were estimated by using a three-layer model and by assuming a refractive index of 1.45 for both the monolayer and protein. Fabrication of Liquid Crystal Cells. We fabricated optical cells from two films of gold, each of which supported the mixed SAMs. First, the two gold films were aligned (facing each other). For obliquely deposited films of gold, the direction of deposition of the gold in each film was parallel. Second, the gold films were clipped together (binder or bulldog clips) using a thin film of Mylar or Saran Wrap (nominal thickness of ∼12 µm) to keep the two surfaces apart. A drop of 5CB, heated into its isotropic phase (33 °C < T < 45 °C) was then drawn by capillarity into the cavity between the two surfaces of the optical cell. The cell was subsequently cooled to room temperature, and the optical texture was observed with an Olympus BX-60 polarizing light microscope (Tokyo, Japan) in transmission mode. Optical Characterization of Liquid Crystals. We measured the average and standard deviation of the intensity of light transmitted through liquid crystals supported on SAMs with and without bound protein. These indices of the appearance of the liquid crystal were extracted from brightness histograms of images obtained by using Adobe Photoshop (San Jose, CA).

Results and Discussion Synthesis of Compound 1. Compound 1 differs from the compound reported by Whitesides and co-workers12 by the presence of an additional methylene unit between the tri(ethylene glycol) linker and the NTA moiety (Figure 3). This methylene unit increased the ease of synthesis of 1 by keeping the polarity of each intermediate product low (the yield of each step was at least 50%). Each intermediate product was purified by column chromatography to give an analytically pure product (by NMR). We note that the presence of the amide moiety in compound 1, when compared to the carbamic acid ester bond reported in the past study,12 may provide additional stability under strongly acidic conditions. We also note that the susceptibility to hydrolysis of a moiety on a surface can be substantially different from that in bulk solution because the pKa of the moiety can change when immobilized on a surface.19 Characterization of SAMs Formed from 1 and 2. We first measured the ellipsometric thicknesses of SAMs formed from 1 or 2 (total concentration of 1 and 2 was 1 mM) and mixed SAMs formed from solutions containing different ratios of 1 and 2 (Table 1). Because the thickness measured by ellipsometry represents an average obtained over a macroscopic area (∼1 mm2), we expected the apparent thickness of the SAMs to increase with the percentage of 1 in the mixed alkanethiol solution. We measured the ellipsometric thickness of the tri(ethylene glycol)-terminated SAMs to be about 1.9 nm. With an increase in the percentage of 1 (18) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110, 6560. (19) (a) Wallwork, M. L.; Smith, D. A.; Zhang, J.; Kirkham, J.; Robinson, C. Langmuir 2001, 17, 1126-1131. (b) Zhang, H.; He, H. X.; Wang, J.; Mu, T.; Liu, Z. F. Appl. Phys. A 1998, 66, S269-S271, Suppl. S.

Langmuir, Vol. 19, No. 5, 2003 1675 Table 1. Ellipsometric Thicknesses of SAMs Formed from 1 or 2 SAMs 1

5% of 1a 20% of 1 20% of 1-NiII

thickness 33 ( 1 21 ( 0.8 24 ( 1.2 (Å)

24 ( 0.6

2 19 ( 0.4

a The mole percentage of 1 in ethanol from which the mixed SAMs were formed.

Figure 4. XPS spectrum of a SAM formed from the NTAterminated alkanethiols, 1, chelated with Ni2+. The various orbitals of atoms in the SAM and gold are labeled.

in a solution containing 1 and 2 (0%, 5%, 20%, and 100% of 1), we recorded a gradual increase in the ellipsometric thickness of the SAM (19, 21, 24, and 33 Å) with a standard deviation of about 1 Å. These values are consistent with those reported previously by Whitesides using the related compounds.12 Furthermore, immersion of the NTAterminated SAM into a solution containing Ni(II) ions did not change the ellipsometric thickness of the monolayer. This result is consistent with a specific chelation of Ni(II) by NTA ligands present on the surface and minimal deposition of excess salt, a conclusion that is confirmed by experiments reported below using XPS. We used XPS to characterize the chemical composition of the SAM presenting NTA-terminated alkanethiols (on uniformly deposited gold), including the extent of incorporation of the Ni(II).17 Figure 4 shows an XPS spectrum of a SAM formed from the NTA-terminated alkanethiol after treatment with Ni(II) ions. The regions corresponding to the Ni 2p3 and N 1s peaks are expanded for clarity. Integration of the Ni 2p3 and N 1s peaks, after correcting for the different cross sections of the orbitals, gives a stoichiometric ratio of 31/69. Because there are two nitrogen atoms in each molecule of 1, the closeness of this ratio to 1/2 suggests that the Ni(II) ions are chelated at the binding sites of each NTA ligand rather than being nonspecifically deposited. Alignment of Liquid Crystals on SAMs Formed from 2 on Obliquely Deposited Gold Films. Past studies have demonstrated that the orientations assumed by liquid crystals on SAMs supported on obliquely deposited gold depend on the orientation and functionality of the terminal group of the SAMs.3 Thus we first examined the orientations of liquid crystals on tri(ethylene glycol)terminated SAMs, 2. Because the conformations and dynamics of the ethylene glycol units on tri(ethylene glycol)-terminated SAMs likely depend on the choice of substrate and conditions used to form the SAM, and because the ethylene glycol groups are large compared to ω-functional groups explored in past studes,3 we first

1676

Langmuir, Vol. 19, No. 5, 2003

Figure 5. (A) Optical images (crossed polarizers) of nematic 5CB anchored on tri(ethylene glycol)-terminated SAMs formed on obliquely deposited gold films. The angle between the azimuthal direction of maximum roughness within the gold films and that of the polarizers is indicated below each image. (B) Optical images (with and without quarter wave plate) of the interference colors created by nematic 5CB in a wedged cell (see text for details). The orientation of the quarter wave plate and the wedged cell are shown to the left side of the optical images. The two gold films within the cell are aligned such that the direction of minimum roughness within each film is parallel to the other (A and B) but perpendicular to the long axis of the QWP (B).

determined if tri(ethylene glycol)-terminated SAMs would uniformly orient a liquid crystal. We examined the alignment of nematic 5CB on SAMs formed from 2 on gold films that were obliquely deposited at two different angles of deposition (30° and 40° measured from normal of the glass substrate). Figure 5A shows the optical appearance of the liquid crystal on SAMs formed from 2 on gold films deposited at 30°. The uniform optical texture and strong modulation of the intensity of light upon rotation of the sample between crossed polars indicates a uniform alignment of 5CB. On gold films prepared by oblique deposition at 40°, 5CB also assumed a uniform alignment on SAMs formed from 2. We next determined the azimuthal orientation of 5CB on the SAMs formed from 2. Figure 5B shows the change in interference colors created by the passage of white light through a wedge-shaped cell upon insertion of a quarter wave plate (QWP) into the optical path. The wedge-shaped cell was prepared by spacing two gold films, each of which supported SAMs formed from 2, with a spacer at one end but not the other (Figure 5B). When the wedge-shaped cell was oriented with the direction of minimum roughness in the gold film parallel to the long axis of the QWP, we observed a red-to-blue shift in the interference color to accompany insertion of the QWP. This shift in interference color indicates that the optical axis of the liquid crystal is perpendicular to the long axis of the QWP.20 We conclude, therefore, that the liquid crystal aligns in the azimuthal direction of maximum roughness on these surfaces (Figure 6A). We further confirmed this azimuthal orientation by examining the liquid crystal in optical cells comprised of surfaces supporting two different SAMs (hexadecanethiol and pentadecanethiol). Past studies have revealed that liquid crystals, when supported on obliquely deposited gold films, align in the azimuthal direction of minimum roughness on SAMs formed from pentadecanethiol but (20) Hartdshorne, N. H.; Stuart, A. Crystals and the Polarizing Microscope; Edward Arnold: New York, 1970.

Luk et al.

align in the azimuthal direction of maximum roughness on SAMs formed from hexadecanethiol. Therefore, we used an optical cell that supported tri(ethylene glycol)terminated SAMs on one surface and a SAM formed from either hexadecanethiol or pentadecanethiol on the other surface. While 5CB assumed a uniform alignment in cells presenting a tri(ethylene glycol)-terminated SAM on one surface and a SAM formed from hexadecanethiol on the other, a 90° twist distortion in the liquid crystal was observed when hexadecanethiol was replaced by pentadecanethiol.3 This result indicates that the azimuthal orientation of liquid crystals on a tri(ethylene glycol)terminated SAM is the same as that on SAMs formed from hexadecanethiol but orthogonal to that on SAMs formed from pentadecanethiol. It also confirms that 5CB aligns on tri(ethylene glycol)-terminated SAMs in the azimuthal direction of maximum roughness on the obliquely deposited gold film. The above-stated conclusion that nematic 5CB on tri(ethylene glycol)-terminated SAMs assumes an azimuthal orientation that is parallel to the direction of maximum roughness of the gold film is interesting for several reasons. First, the uniform alignment of liquid crystals on tri(ethylene glycol)-terminated SAMs supported on obliquely deposited gold films suggests that the bulky ethylene glycol units presented by these SAMs do not mask the underlying topography of the gold films. Second, bare gold films uniformly align the nematic phase of 5CB in the azimuthal direction of minimum roughness, presumably to minimize the distortion of the liquid crystal over the topography. Thus the azimuthal orientation observed on the SAMs formed from 2 suggests that molecular-level interactions between these SAMs and 5CB are important in determining the alignment of 5CB. Our past studies have demonstrated that the role of molecular-level interactions between SAMs and 5CB on the orientation of the liquid crystal can be tested by changing the structure of the SAMs.3 Thus, we next examined the orientation of 5CB on tetra(ethylene glycol)terminated SAMs. In contrast to SAMs presenting tri(ethylene glycol), 5CB adopts a uniform planar alignment that is parallel to the direction of minimum roughness in the obliquely deposited gold film (Figure 6B). This surprising result is further evidence that the orientation of the liquid crystal is highly sensitive to the molecularlevel details of the ethylene glycol groups presented by these SAMs. It also suggests that the SAMs are well organized when in contact with 5CB. Although we do not yet understand the origin of the molecular-level interaction between 5CB and the ethylene glycol-terminated SAMs that leads to the azimuthal orientations reported above, we note that Grunze and coworkers have reported tri(ethylene glycol)-terminated SAMs to assume a helical conformation when supported on gold.21 In addition, we note that hydrogen bonding between the nitrile group in 5CB and the terminal hydroxyl groups of these SAMs is possible and that this interaction may dominate the azimuthal orientation of the liquid crystal. Because this interaction would be influenced by the orientation of the terminal hydroxyl group, it would be expected to change with the number of ethylene glycol units in the terminal group of the alkanethiol. The main conclusions from the above observations are twofold. First, the topography of the obliquely deposited (21) (a) Feldman, K.; Hahner, G.; Spencer, N. D.; Harder, P.; Grunze, M. J. Am. Chem. Soc. 1999, 121, 10134-10141. (b) Pertsin, A. J.; Grunze, M. Langmuir 2000, 16, 8829-8841.

Amplifying Protein-Receptor Interactions

Langmuir, Vol. 19, No. 5, 2003 1677

Figure 6. Schematic illustration of azimuthal orientations of nematic phases of 5CB supported on ω-functionalized SAMs formed on obliquely deposited gold films. Nematic 5CB aligns in an azimuthal orientation that is perpendicular to the direction of maximum roughness on the gold film that supports (A) tri(ethylene glycol)-terminated SAMs but aligns parallel to the direction of maximum roughness when the gold film supports (B) SAMs formed from tetra(ethylene glycol)-terminated alkanethiols, mixed SAMs formed from 1 and 2 (20% of 1 in ethanolic solution from which mixed SAMs were formed), or mixed SAMs formed from 1 and 2 (20% of 1 in ethanolic solution from which mixed SAMs were formed) treated with Ni(II). The arrow above the schematic illustration of the topography of the gold indicates the azimuthal orientation of 5CB.

gold films is not masked by the ethylene glycol units of the oligo(ethylene glycol)-terminated SAMs, thus making this system potentially useful for detecting protein binding events using liquid crystals. Second, the azimuthal orientation assumed by the liquid crystal is sensitive to the structure of the oligo(ethylene glycol)-terminated SAMs. Alignment of Liquid Crystals on Mixed SAMs Formed from 1 and 2 on Obliquely Deposited Gold Films. We next examined the orientations of nematic 5CB on mixed SAMs formed from 1 and 2 on obliquely deposited gold films. We focused on SAMs possessing a low surface density of alkanethiol 1 (vs alkanethiol 2). Because the hexahistidine-tagged MEK protein is large compared to the NTA, only a low density of NTA is needed for immobilization of the MEK.22 Our principal results are threefold. First, we observed that mixed SAMs formed from 1 and 2 (20% of 1 and 80% of 2 in ethanol), when supported on obliquely deposited gold films, cause a uniform planar orientation of 5CB. Second, the liquid crystal assumes an azimuthal orientation that is parallel to the direction of minimum roughness on the surface (Figure 6B). This result indicates that while the NTA ligands are not large enough to mask the topography of the gold film and thus disrupt the uniform alignment of the liquid crystal, the presence of NTA is sufficient to disrupt the molecular-level interaction of 5CB and the tri(ethylene glycol) groups that leads to the alignment on 5CB on SAMs formed from 2 (see above). Third, by treating the mixed SAMs formed from 1 and 2 (20% of 1 and 80% of 2 in ethanol) with Ni(II) ions, we observed 5CB to assume an azimuthal orientation that was parallel to the direction (22) Houseman, B. T.; Mrksich, M. Angew. Chem., Int. Ed. 1999, 38, 782.

of the minimum roughness (i.e., the same as when Ni(II) ions were absent) (Figure 6B). We note that past studies have reported that some 5CB-metal ion interactions can influence the alignment of a liquid crystal on a surface.3c For example, surfaces presenting Cu(II) carboxylate groups cause homeotropic alignment of 5CB whereas surfaces presenting Ca(II) carboxylate do not. When using the NTA-terminated SAMs, we did not measure the incorporation of Ni(II) to cause a measurable change in alignment of 5CB. Nonspecific Adsorption of Proteins on SAMs Formed from 1 on Obliquely Deposited Gold Films. Past studies have reported oligo(ethylene glycol) and polyol-terminated SAMs to exhibit resistance toward nonspecific adsorption of proteins.4,5 The extent of resistance, however, appears to depend on details of the structure of the SAM that are not fully understood.23 Thus it was not obvious to us that SAMs formed on gold films that are prepared by oblique deposition would possess the same inertness to nonspecific adsorption as has been reported when using SAMs on gold films that do not possess an anisotropic topography. To address this issue, using obliquely deposited films of gold (deposited at 30° and 40° from normal of the glass substrate), we examined nonspecific adsorption of BSA and 3T3 fibroblast cell lysate onto SAMs formed from 2 and mixed SAMs formed from 1 and 2. Following incubation in cell lysate (106 fibroblasts per mL of PBS buffer) for 12 h, we rinsed SAMs formed from 2 with PBS and then contacted them with the liquid crystal (5CB). We measured the liquid crystals to assume an azimuthal orientation that was parallel to the direction (23) Zhu, H.; Snyder, M. Curr. Opin. Cell Biol. 2002, 14, 173.

1678

Langmuir, Vol. 19, No. 5, 2003

Luk et al.

Figure 7. Optical images (crossed polarizers) of nematic phases of 5CB supported on mixed SAMs formed from 1 and 2 (20% of 1 in ethanolic solution from which mixed SAMs were formed) that were prepared on obliquely deposited gold films. The mixed SAMs were incubated in either (A) deionized water or (B) 50 mM NiSO4 for 2 h prior to incubation in 0.5 µM His-tag MEK for 5 h. The mixed SAMs supported on the gold films were aligned with the direction of minimum roughness on each surface parallel to each other.

of maximum roughness (parallel to the direction of gold deposition). Furthermore, after incubation of mixed SAMs formed from alkanethiols 1 and 2 (prepared from 20% of 1 and 80% 2 in ethanol) in PBS containing BSA for 12 h, the alignment of the liquid crystal was the same as that observed prior to contact with BSA. These results indicate that nonspecific adsorption on both types of SAMs, when supported on obliquely deposited gold films, is sufficiently small so as not to disrupt the orientations of the liquid crystal on these surfaces. We concluded, therefore, that SAMs formed from 1 and 2 are likely to be useful in studies of specific protein binding on obliquely deposited gold films. Orientational Response of Liquid Crystals to Specific Immobilization of His-tag MEK. We next examined the immobilization of His-tag MEK on obliquely deposited gold films that presented mixed SAMs formed from 1 and 2 (following chelation of the Ni(II) with the surface). Because the chelation of Ni(II) ions by 1 within the mixed monolayer is a requirement for specific binding of the His-tag MEK, we also performed a control experiment in which we examined binding of His-tag MEK without chelated Ni(II) ions. Our choice of experimental conditions for these experiments was influenced by two observations. First, our past studies have demonstrated that the response of liquid crystals to bound protein can be enhanced by decreasing the angle of deposition of gold during preparation of the gold films.9 Hence, here we report the use of gold films prepared by oblique deposition at a low angles (30° from normal) to increase the sensitivity of the liquid crystal to bound protein. Second, we used mixed SAMs formed with low percentages of the NTA-terminated alkanethiols (5% 1 and 95% of 2 in ethanol) to ensure a low surface coverage of NTA. At low surface densities of NTA, we predicted the extent of nonspecific adsorption of protein on surfaces not treated with Ni(II) to be small compared to the amount of protein specifically immobilized via complexation with NiII-NTA on the surface. We immersed mixed SAMs of 1 and 2 (prepared from 5% of 1 and 95% of 2 in ethanol) in 50 mM NiSO4 for 2 h and then incubated the rinsed surfaces in PBS containing 0.5 µM His-tag MEK for 5 h. Upon removal of the samples from contact with the His-tag MEK, the surfaces were rinsed with PBS buffer (pH 8.2). The optical appearance of a nematic phase of 5CB contacted with a SAM pretreated

with the His-tag MEK is shown in Figure 7A. Inspection of Figure 7A reveals the optical appearance of the 5CB to be nonuniform. The extent of modulation of light transmitted through the sample during rotation of the sample between crossed polars was determined to be low, thus confirming the absence of a preferred azimuthal orientation of the liquid crystal within the sample. In contrast, when the SAM was pretreated with deionized water instead of 50 mM NiSO4 and then contacted with an aqueous solution of His-tag MEK, 5CB assumed a uniform alignment on the surface. A strong modulation of the intensity of transmitted light was measured as the sample was rotated between cross polarizers (Figure 7B). By comparing the optical appearance of the liquid crystal in parts A and B of Figure 7, we conclude that the His-tag MEK binds specifically to the NiII-NTA complex within the SAMs. This conclusion is supported by ellipsometric measurements (Table 2). We measured the ellipsometric thickness of the His-tag MEK on the SAMs treated with Ni(II) to be ∼1 nm, whereas there was no measurable His-tag MEK bound to the SAMs when Ni(II) ions were absent. Because the size of MEK is comparable to the scale of the topography of the obliquely deposited gold films, bound His-tag MEK masks the topography of the surface from the liquid crystal, thus resulting in nonuniform alignment of the liquid crystal. Orientational Response of Liquid Crystals to antiMEK IgG Bound to Hexahistidine-Tagged MEK. The protein MEK is representative of proteins involved in cell signaling pathways (such as the mitogen-activated protein kinase (MAPK) cascade).23,24 Detection and characterization of MEK is of substantial interest in cell biology. Here we report the use of anti-MEK immunoglobulin G (IgG) as a model binding partner of MEK to demonstrate the possibility of using liquid crystals to detect protein binding events that involve MEK. Our past studies have demonstrated that the alignment of liquid crystals on surfaces supporting bound proteins can be made uniform by increasing the angle of deposition of the gold films.9 Whereas a film of gold deposited at an angle of 30° possesses a topography that permits the (24) (a) Huang, C.-Y. F.; Ferrell, J. E. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 10078. (b) Asai, T.; Tena, G.; Plotnikova, J.; Willmann, M. R.; Chiu, W. L.; Gomez-Gomez, L.; Boller, T.; Ausubel, F. M.; Sheen, J. Nature 2002, 415, 977.

Amplifying Protein-Receptor Interactions

Langmuir, Vol. 19, No. 5, 2003 1679

Table 2. Ellipsometric Thickness and Indices of Optical Appearance of Liquid Crystals θa 1st protein mole % of 1c Ni(II) 2nd protein luminosityf (SD)g thicknessh (Å) ∆thickness (Å)

30° MEKb

His-tag 5 4 (4) 21 ( 0.8 -

30°

40°

40°

40°

His-tag MEK 5 + 78 (44) 29 ( 1.2 8 ( 1.2i

His-tag MEK 20 + 26 (36) 39 ( 0.5 15 ( 0.6i

His-tag MEK 20 + anti-MEK IgGd 133 (59) 72 ( 10 33 ( 10j

His-tag MEK 20 + anti-strep IgGe 21 (26) 39 ( 0.5 -

a The angle of deposition measured from the normal of the substrate. b His-tag MEK (MW of ∼55 kDa). c The mole percentage of 1 in ethanol from which the mixed SAMs were formed. d Anti-MEK IgG (MW of 150 kDa). e Anti-streptavidin IgG (MW of ∼150 kDa). f The average luminosity of the image of the liquid crystal. g The standard deviation of the luminosity of the image of the liquid crystal. h The measured ellipsometric thickness of the interfacial layer after binding of the protein. i The change in ellipsometric thickness upon immobilization of His-tag MEK. j The change in ellipsometric thickness upon binding of antibody to His-tag MEK.

Figure 8. Optical images (crossed polarizers) of nematic 5CB supported on mixed SAMs formed from 1 and 2 (5% of 1 in ethanolic solution from which mixed SAMs were formed) that were exposed to PBS solution containing the following proteins prior to contact with nematic 5CB: (A) His-tag MEK, (B) His-tag MEK and anti-MEK IgG, (C) His-tag MEK and anti-Streptavidin IgG. See text for details of concentrations and binding times.

reporting of bound MEK using the liquid crystal (see above), to obtain uniform alignment of liquid crystals on a surface presenting MEK, we prepared gold films that were deposited at an angle of 40° instead of 30°. On these gold films, we prepared mixed SAMs from a mixture of 20% of 1 and 80% of 2 (total concentration 1 mM in ethanol) to investigate the orientational response of liquid crystals to IgG bound to His-tag MEK immobilized on the surface. After incubating the SAMs in MEK (0.1 µM in PBS, pH 7.2) for 2 h, we used ellipsometry to determine that a layer of His-tag MEK with a thickness of ∼1.9 nm was bound to the SAMs. In contrast, a layer of MEK having a thickness of only ∼0.9 nm was bound to SAMs formed from 5% of 1 and 95% of 2. Our decision to base our experiments on the SAMs formed from 20% of 1 and 80% of 2 was a result of the higher level of binding of MEK to these SAMs as compared to the mixed SAMs from 5% of 1 and 95% of 2. We next incubated the substrates decorated with MEK in solutions containing either anti-MEK IgG (0.5 µM) or anti-streptavidin IgG (0.5 µM) for 5 h. Figure 8A shows the optical textures of nematic phases of 5CB in contact with the mixed SAM to which His-tag MEK was bound (as described above). A strong modulation in the intensity of light was observed as the sample was

rotated between crossed polarizers, indicating a largely uniform planar alignment of the liquid crystal. As mentioned above, the ellipsometric thicknesses of MEK bound to the mixed SAMs (0.5 µM His-tag MEK in PBS for 2 h) was 1.9 ( 0.5 nm. The result in Figure 8A indicates that this amount of immobilized MEK is not sufficient to mask the influence of the topography of the gold film on the orientation of the liquid crystal. This result contrasts to the result in Figure 7 in which MEK was immobilized on gold films deposited at 30° from the normal. Figure 8B shows the optical appearance of 5CB on surfaces presenting immobilized His-tag MEK that was treated with 0.1 µM anti-MEK IgG (∼150 kDa) in PBS for 5 h. Inspection of Figure 8B shows that the liquid crystal assumed a nonuniform orientation on these surfaces. In addition, we recorded little modulation in the intensity of light passing through the sample when the sample was rotated between the cross polarizers. Ellipsometric measurements on uniformly deposited gold films treated with MEK and anti-MEK IgG, as described above, revealed an increase in the ellipsometric thickness of ∼3.3 ( 1 nm upon binding of anti-MEK IgG. A control experiment in which anti-streptavidin IgG was used instead of anti-MEK IgG yielded a liquid crystal with a uniform alignment

1680

Langmuir, Vol. 19, No. 5, 2003

(Figure 8C). Ellipsometric measurements did not yield any detectable increase in the thickness of bound protein after treatment with anti-streptavidin IgG. These results are consistent with the specific binding between the immobilized MEK and the anti-MEK IgG in solution. When combined, the results above indicate that immobilized His-tag MEK participates in a specific binding event with anti-MEK IgG and that bound anti-MEK IgG is readily reported by the liquid crystal. Whereas the results in Figure 8 show that it is possible to determine the presence of anti-MEK IgG bound to the MEK by visual inspection of the liquid crystal, we briefly mention here that it is also possible to formulate numerical indices that can be used to characterize the appearance of the liquid crystal. We measured the average brightness and standard deviation in brightness (hereafter called the luminosity) of each image of the nematic film (with the optical axis of the liquid crystal aligned with the polarizer). The results are summarized in Table 2. These results indicate that the average luminosity and the standard deviation in luminosity are both useful indices of the appearance of the liquid crystal on surfaces. In general, both the average luminosity and the standard deviation in the luminosity increase with the amount of immobilized protein. Conclusions The main conclusions of this paper are threefold. First, we have reported an efficient synthesis of an NTAterminated alkanethiol in high yield and purity. This compound can be used to functionalize obliquely deposited gold films with NiII-NTA. Second, we formed mixed monolayers that present NTA and tri(ethylene glycol) on obliquely deposited gold films and demonstrated that

Luk et al.

liquid crystals can assume a uniform orientation on these surfaces. We measured the azimuthal orientation of the liquid crystal to be sensitive to the molecular details of the SAMs. Specifically, we report the surprising observation that nematic phases of 5CB assume orthogonal azimuthal orientations on surfaces that present SAMs formed from tri(ethylene glycol) versus tetra(ethylene glycol)-terminated alkanethiols. Third, we have demonstrated that mixed SAMs presenting NTA and tri(ethylene glycol) resist nonspecific protein adsorption and bind the histidine-MEK fusion protein (His-tag MEK) via the interaction of the His-tag and the NTA-NiII complex on the surface. By tuning the topography of the surface, we also demonstrate that it is possible to use liquid crystals to either (I) detect binding of the His-tag MEK to the NTANiII complex or (II) use the His-tag MEK as a receptor and detect binding of anti-MEK IgG to it. These results suggest that mixed SAMs presenting NTA and oligo(ethylene glycol) groups on obliquely deposited gold films, when combined with liquid crystals, may form the basis of a general and facile method for detecting binding events between proteins and histidine-tagged receptors on surfaces. Acknowledgment. This research was supported by funding from the Office of Naval Research (Presidential Early Career Award for Science and Engineering to N.L.A., N00014-99-1-0250), the Materials Research Science and Engineering Center on Nanostructured Materials and Interfaces (NSF-DMR-0079983) at the University of Wisconsin, the Biophotonics Partnership Initiative of NSF (ECS-0086902), and the University of Wisconsin Biotechnology Training Program (NIH5 T32GMO8349). LA026152K