Facile Construction of Sulfanyl-Terminated Poly(ethylene glycol

Jul 29, 2008 - Graduate School of Pure and Applied Science, University of Tsukuba, 1-1-1 ... Master's School of Medical Science, Graduate School of ...
0 downloads 0 Views 743KB Size
Download PDF
Langmuir 2008, 24, 9623-9629

9623

Facile Construction of Sulfanyl-Terminated Poly(ethylene glycol)-Brushed Layer on a Gold Surface for Protein Immobilization by the Combined Use of Sulfanyl-Ended Telechelic and Semitelechelic Poly(ethylene glycol)s Keitaro Yoshimoto,†,‡,§ Takumi Hirase,† Seiko Nemoto,| Tamao Hatta,| and Yukio Nagasaki*,†,‡,§,⊥,# Graduate School of Pure and Applied Science, UniVersity of Tsukuba, 1-1-1 Ten-noudai, Tsukuba, Ibaraki 305-8573, Japan, Center for Tsukuba AdVanced Research Alliance (TARA), UniVersity of Tsukuba, 1-1-1 Ten-noudai, Tsukuba 305-8577, Japan, Tsukuba Research Center for Interdisciplinary Materials Science (TIMS), UniVersity of Tsukuba, 1-1-1 Ten-noudai, Tsukuba, Ibaraki, 305-8573, Japan, International Research Center for Agricultural Sciences (JIRCAS), 1-1 Ohwashi, Tsukuba, Ibaraki 305-8686, Japan, Master’s School of Medical Science, Graduate School of ComprehensiVe Human Sciences, UniVersity of Tsukuba, 1-1-1 Ten-noudai, Tsukuba, Ibaraki 305-8573, Japan, and Satellite Laboratory, International Center for Materials Nanoarchilectonics (MANA), National Institute of Materials Science (NIMS), 1-1-1 Ten-noudai, Tsukuba, Ibaraki 305-8573, Japan ReceiVed April 23, 2008. ReVised Manuscript ReceiVed June 15, 2008 A sulfanyl-terminated poly(ethylene glycol) (PEG)-brushed layer was constructed on a gold sensor platform by consecutive treatment with a sulfanyl-ended semitelechelic PEG (2 kDa, hereafter “MeO-PEG-SH (2k)”) and a sulfanyl-ended telechelic PEG (5 kDa, hereafter “SH-PEG-SH (5k)”). Our strategy of constructing the sulfanyl-terminated PEG-brushed gold surface is based on mixed-PEG-brush formation from the longer SH-PEG-SH (5k) and the shorter MeO-PEG-SH (2k), where the preimmobilized shorter MeO-PEG-SH (2k) prevents loop formation in the longer SH-PEG-SH (5k) on the surface and the free sulfanyl group at one end of the longer SH-PEG-SH is exposed to the mixed-PEG tethered-chain surface. From the experimental results obtained from surface plasmon resonance analysis, it became apparent that the immobilization density and the orientation of the longer SH-PEG-SH (5k) on the gold surface could be controlled by the amount of preimmobilized shorter MeO-PEG-SH (2k). Under the optimized conditions of MeO-PEG-SH (2k) premodification, the constructed MeOPEG-SH (2k)/SH-PEG-SH (5k) mixed layer conjugated efficiently with the maleimide-installed proteins and the antibody Fab′ fragments, accompanied by an appreciable nonfouling characteristic against bovine serum albumin as strong as that of the MeO-PEG-SH (5k)/MeO-PEG-SH (2k) mixed surface, which was reported in our previous work; it also showed a superior nonfouling characteristic compared to the commercially available carboxymethylated dextran surface (Uchida, K.; et al. Biointerphase 2007, 2 (4), 126-130). Furthermore, from the experimental results of the X-ray photoelectron spectrometry analysis, the presence of both Au-bound and Au-unbound sulfur species was confirmed on the SH-PEG-SH (5k)/MeO-PEG-SH (2k)-modified gold surface. These results clearly indicate that the preimmobilized shorter MeO-PEGSH (2k) not only increased the nonfouling characteristic of the PEG tethered-chain surface but also prevented loop formation in the longer SH-PEG-SH (5k) on the gold surface. Since the protein-installed SH-PEG-SH (5k)/MeO-PEG-SH (2k)modified surface showed a strongly nonfouling characteristic and recognized the target molecules selectively, this new mixed-brush-formation technique using longer sulfanyl-ended telechelic PEGs and shorter semitelechelic PEGs is a simple yet effective method of constructing a strongly nonfouling terminal-functionalized gold surface for protein immobilization.

Introduction A metal surface is one of the most valuable platforms for molecular immobilization, because sulfanyl group1 containing species can be immobilized firmly and easily on a metal surface via covalent bonds due to the strong interaction between the metal surface and the sulfanyl group.2 In particular, the gold surface has been widely used in bioanalytic devices due to its * To whom correspondence should be addressed. Graduate School of Pure and Applied Science, University of Tsukuba, 1-1-1 Ten-noudai, Tsukuba, Ibaraki 305-8573, Japan. Phone: +81-29-853-5749. Fax: +81-29-853-5749. E-mail:[email protected]. † Graduate School of Pure and Applied Science, University of Tsukuba. ‡ Center for Tsukuba Advanced Research Alliance (TARA), University of Tsukuba. § Tsukuba Research Center for Interdisciplinary Materials Science (TIMS), University of Tsukuba. | Japan International Research Center for Agricultural Sciences (JIRCAS). ⊥ Master’s School of Medical Science, Graduate School of Comprehensive Human Sciences, University of Tsukuba. # Satellite Laboratory, International Center for Materials Nanoarchilectonics (MANA), National Institute of Materials Science (NIMS).

conductivity, resistance to oxidation, and simplicity with which thin films can be applied onto inorganic substrates. Additionally, the surface plasmon resonance (SPR) angle shift on the gold substrate, which is a refractive index that changes in the very close vicinity of a thin metal film surface, has been found to be a useful analytical signal for monitoring the molecular binding events on the sensor surface in existing spectroscopies3 and nanoparticle-based assays.4 Under the proper conditions, the (1) It was well-known that “mercapto-” and “-thiol” have been used in the scientific literature as prefix and suffix, respectively. However, according to A Guide to IUPAC Nomenclature of Organic Compounds: Recommendations 1993 (Blackwell Scientific Publications: Boca Raton, FL, 1993), the prefix “sulfanyl-” is preferred to “mercapto-” that was used in previous editions of the IUPAC Nomenclature of Organic Chemistry. Therefore, we used “sulfanyl” as prefix in our papers instead of “mercapto-” in this paper. You can check this recommendation in the web site http://www.acdlabs.com/iupac/nomenclature/93/r93_296.htm and http://www.acdlabs.com/iupac/nomenclature/93/r93_302.htm. (2) (a) Prime, K. L.; Whitesides, G. M. Science 1991, 252, 1164–1167. (b) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. ReV. 2005, 105, 1103–1170. (3) Homola, J. Chem. ReV. 2008, 108, 462–493. (4) Eugenii, K.; Itamar, W. Angew. Chem., Int. Ed. 2004, 43, 6042–6180.

10.1021/la8012798 CCC: $40.75  2008 American Chemical Society Published on Web 07/29/2008

9624 Langmuir, Vol. 24, No. 17, 2008

reflectivity of a thin metal film is extremely sensitive to optical variations in the medium on one side of it. This property makes the gold surface particularly attractive; however, since a bare gold surface does not have specific sites for interaction with biomolecules, all proteins and most other biocomponents adhered to it spontaneously and nonselectively. Therefore, terminalfunctionalization techniques for gold surfaces have played a critical role in a wide variety of analytical applications, including electrochemical analysis,5 SPR sensors,3,6 quartz crystal microbalance measurement,7 and so on. Self-assembled monolayers (SAMs) of alkanethiols have been used for the modification of gold surfaces, and functional groups such as carboxyl (-COOH),8 amino (-NH2),9 and sulfanyl (-SH)10 have been bound to gold surfaces via terminalfunctionalized alkanethiols. Although the surface modification technique using these alkanethiol is one of the most robust methods of constructing terminal-functionalized gold surfaces, in order to develop high-performance analytic devices for biomolecular applications, it is very important to achieve a good balance between the terminal functionalization and the preparation of a nonfouling surface on the gold sensor platform. Regarding the issue of constructing a nonfouling surface with biological media, poly(ethylene glycol) (PEG) has been used in numerous biomedically relevant systems to control protein adsorption on surfaces.11 Surface modification with PEG tethered chains dramatically decreases the nonspecific adsorption of highmolecular-weight biomolecules on the surface,11f and sulfanylended semitelechelic PEGs (PEG-SH) have been utilized to construct a PEG tethered-chain layer on a gold surface.11g It is believed that the appreciable flexibility, high hydrophilicity, and nonionic character of the PEG chain are important for keeping proteins away from the surface by steric exclusion.11h In our previous work, the mixed-PEG-brushed layer composed of the shorter MeO-PEG-SH (2k) and the longer MeO-PEG-SH (5k) on a gold surface was found to have a superior nonfouling character.12 Interestingly, as compared to commercial carboxymethylated dextran and conventional MeO-PEG-SH (5k) tetheredchain-modified gold surfaces, this mixed-PEG-modified gold surface showed an almost complete inhibition of the nonspecific adsorption not only of high-molecular-weight but also of lowmolecular-weight peptides.12b Moreover, we have succeeded in constructing a terminal-functionalized PEG-modified surface using heterotelechelic PEGs possessing an acetal group at one end13,14 which has made it possible to construct reactive PEG(5) Zhang, J.; Chi, Q.; Albrecht, T.; Kuznetsov, A. M.; Grubb, M.; Hansen, A. G.; Wackerbarth, H.; Welinder, A. C.; Ulstrup, J. Electrochim. Acta 2005, 50, 3143–3159. (6) Homola, J.; Yee, S. S.; Gauglitz, G. Sens. Actuators, B 1999, 54, 3–15. (7) Cooper, M. A.; Singleton, V. T. J. Mol. Recognit. 2007, 20, 154–184. (8) Li, J.; Liang, K. S.; Scoles, G.; Ulman, A. Langmuir 1995, 11, 4418–4427. (9) Delamarche, E.; Sundarababu, G.; Biebuyck, H.; Michel, B.; Gerber, Ch.; Sigrist, H.; Wolf, H.; Ringsdorf, H.; Xanthopoulos, N.; Mathieu, H. J. Langmuir 1996, 12, 1997–2006. (10) Brust, M.; Blass, P. M.; Bard, A. J. Langmuir 1997, 13, 5602–5607. (11) (a) Feller, L. M.; Cerritelli, S.; Textor, M.; Hubbell, J. A.; Tosatti, S. G. P. Macromolecules 2005, 38, 10503–10510. (b) Kitano, H.; Miyamoto, T.; Kawasaki, H. J. Colloid Interface Sci. 2004, 279, 425–432. (c) Bearinger, J. P.; Terrettaz, S.; Michel, R.; Tirelli, N.; Vogel, H.; Textor, M.; Hubbell, J. A. Nat. Mater. 2003, 2, 259–264. (d) Vermette, P.; Meagher, L. Colloids Surf., B 2003, 28, 153–198. (e) Seto, F.; Tahara, K.; Kishida, A.; Muraoka, Y.; Akashi, M. J. Appl. Polym. Sci. 1999, 74, 1516–1523. (f) Poly(ethylene glycol) Chemistry: Biotechnical and Biomedical Application; Harris, J., Ed.; Plenum Press: New York, 1992. (g) Glass, J. E. Ed. Hydrophilic Polymers: Poly(ethylene oxide) and Protein Resistance;American Chemical Society: Washington, DC, 1996. (h) Lee, J. H.; Lee, H. B.; Andrade, J. D. Prog. Polym. Sci. 1995, 20, 1043–1079. (12) (a) Uchida, K.; Otsuka, H.; Kaneko, M.; Kataoka, K.; Nagasaki, Y. Anal. Chem. 2005, 77, 1075–1080. (b) Uchida, K.; Hoshino, Y.; Tamura, A.; Yoshimoto, K.; Kojima, S.; Yamashita, K.; Yamanaka, I.; Otsuka, H.; Kataoka, K.; Nagasaki, Y. Biointerphases 2007, 2, 126–130. (13) Nagasaki, Y.; Ogawa, R.; Yamamoto, S.; Kato, M.; Kataoka, K. Macromolecules 1997, 30, 6489–6493.

Yoshimoto et al.

brushed layers on gold surfaces12a and nanoparticles.15,16 Although the acetal group on the PEGylated surface thus prepared can react with biocomponents,12a,16,17 much needs to be improved in their reactivity on the PEGylated surface.12a For example, the conjugation efficiencies of the acetal group terminated PEGpoly(D,L-lactide) micelles to amino group containing molecules have been reported to be between 20% and 50%.18 This low conversion efficiency is due to the reaction mechanism between the aldehyde group and the amino group, where the generation of the Schiff base (-CHdN) group is an equilibrium reaction, and it is difficult to accomplish the quantitative generation of the -CH2-NH- bond by reduction using NaBH3CN. Furthermore, the conjugation of target molecules with the reactive terminus group is difficult to accomplish on a PEG-modified surface, because the PEG tethered-chains strongly prevent their access to the surface. To overcome these problems, the use of another functional group is an effective way to construct a highly reactive surface. The reaction of the sulfanyl group is particularly useful, because the sulfanyl group reacts rapidly and very efficiently with many functional groups such as iodoacetamide, maleimide, sulfanyl, and benzylic halide at or below room temperature under physiological conditions. To our knowledge, while several types of sulfanyl-terminated surfaces have been proposed,19 a sulfanylterminated PEGylated gold surface has never been reported. We present here a new strategy for constructing a sulfanyl-terminated mixed-PEG-brushed layer on a gold surface using sulfanyl-ended telechelic PEG (SH-PEG-SH) (5k) and semitelechelic PEG (MeOPEG-SH) (2k). Our strategy of constructing sulfanyl-terminated PEGylated gold surfaces is based on the SH-PEG-SH (5k) brush formation on the gold surface; however, it is very difficult because the formation of loop in R,ω-disulfanyl compounds on a gold surface has been pointed out and discussed in previous reports.20 For example, in the case of a sulfanyl-terminated gold surface fabricated using R,ω-disulfanyl compounds such as R,ωalkanedithiol and R,ω-bis(sulfanylmethylthienyl)alkanes, some experimental results have indicated that a lie-down or a looped structure was induced in the R,ω-disulfanyl compounds on the gold surface by the simultaneous chemisorption of both of the sulfanyl termini on the surface, depending on the extent of the adsorbate-substrate interaction.20b In this study, in order to accomplish the facile construction of a sulfanyl-terminated PEGmodified gold surface based on SH-PEG-SH (5k) brush formation, preimmobilization of the shorter MeO-PEG-SH (2k) with a controlled chain density on the gold surface was performed. To evaluate the performance of the constructed MeO-PEG-SH (2k)/ SH-PEG-SH (5k) mixed-PEG tethered-chain surface as a new platform for protein immobilization, the assessment of the nonfouling character, the conjugation reactivity with specific proteins, and the thiolate species on the MeO-PEG-SH (2k)/ SH-PEG-SH (5k) mixed-PEG tethered-chain surface was carried (14) Akiyama, Y.; Otsuka, H.; Nagasaki, Y.; Kato, M.; Kataoka, K. Bioconjugate Chem. 2000, 11, 947–950. (15) Otsuka, H.; Akiyama, Y.; Nagasaki, Y.; Kataoka, K. J. Am. Chem. Soc. 2001, 47, 8226–8230. (16) Ishii, T.; Otsuka, H.; Kataoka, K.; Nagasaki, Y. Langmuir 2004, 20, 561–564. (17) Otsuka, H.; Nagasaki, Y.; Kataoka, K. Langmuir 2004, 20, 11285–11287. (18) Yamamoto, Y.; Nagasaki, Y.; Kato, M.; Kataoka, K. Colloids Surf., B 1999, 16, 135–146. (19) (a) He, H. X.; Zhang, H.; Li, Q. G.; Zhu, T.; Li, S. F. Y.; Liu, Z. F. Langmuir 2000, 16, 3846–3851. (b) Brust, M.; Blass, P. M.; Bard, A. J. Langmuir 1997, 13, 5602–5607. (c) Cheng, F.; Gamble, L. J.; Castner, D. G. Anal. Chem. 2008, 80, 2564–2573. (d) Castner, D. G.; Hinds, K.; Grainger, D. W. Langmuir 1996, 12, 5083–5086. (20) (a) Murty, K. V. G. K.; Venkataramanan, M.; Pradeep, T. Langmuir 1998, 14, 5446–5456. (b) Yang, J.-S.; Lee, C.-C.; Yau, S.-L.; Chang, C.-C.; Lee, C.-C.; Leu, J.-M. J. Org. Chem. 2000, 65, 871–877.

Sulfanyl-Terminated PEG-Brushed Layer on a Gold Surface

Langmuir, Vol. 24, No. 17, 2008 9625

Materials. Commercial tetrahydrofuran (THF), triethylamine, N,N′-dimethylformamide (DMF), sodium hydride (NaH), and methanesulfonyl chloride were purchased from Kanto Chemical Co., Inc. (Tokyo, Japan), and purified by conventional methods21 before use. Hydroxyl-ended telechelic poly(ethylene glycol) (OH-PEGOH (MW 4600); Sigma-Aldrich, Inc., St. Louis, MO, USA), sulfanylended semitelechelic poly(ethylene glycol) (MeO-PEG-SH (MW 2000); NOF Corp., Tokyo, Japan), potassium O-ethyldithiocarbonate (DTC), n-proplyamine and dithiothreitol (DTT; Wako Pure Chemical Industries, Ltd., Osaka, Japan), sulfosuccinimidyl 4-(N-maleimidomethyl)-cyclohexane-1-carboxylate (Sulfo-SMCC; Pierce Biotechnology, Inc., IL, USA), bovine serum albumin (BSA), fluorescein isothiocyanate labelled BSA (FITC-BSA; Sigma-Aldrich), antifluorescein isothiocyanate and goat-poly (anti-FITC; Bethyl Laboratories, Inc., Montgomery, TX, USA), and anti-FITC antibody F(ab′)2 (Rockland Immunochemicals, Inc., PA, USA) were used without any purification. The water used in this study was purified using a Milli-Q system (Nihon Millipore Co., Tokyo, Japan). Synthesis of r,ω-Disulfanyl-PEG (SH-PEG-SH). An outline of the synthetic root of SH-PEG-SH is described in Scheme S1. To a THF solution (20 mL) of 1 mmol of PEG with hydroxyl groups at both ends (MW 4600) in a 100 mL round-bottom flask equipped with a three-way stopcock, NaH (20 mmol), triethylamine (9 mmol), and methanesulfonyl chloride (7 mmol) were added under a nitrogen atmosphere. After the mixture was allowed to react for several hours with magnetic stirring, chloroform was added to the solution, which was then washed with a saturated NaCl aqueous solution several times to eliminate the impurities. The organic phase was concentrated by evaporation after drying with sodium sulfate, followed by pouring into an excess amount of diethyl ether. The precipitate, dimethansulfonyl-PEG (MS-PEG-MS), was dried in vacuo and then analyzed by size exclusion chromatography (SEC). The number-average molecular weight (Mn) and the molecular weight distribution (MWD) of MS-PEG-MS were determined to be 4600 and 1.02, respectively. An end functionality of approximately 98% was confirmed by 1H NMR spectroscopy (400 MHz, DMSO-d6). The SEC chart and NMR spectra of Ms-PEG-Ms (5k) are shown in Figures S1 and S2 in the Supporting Information, respectively. To convert from the methanesulfonyl precursor end group to the O-ethyldithiocarbonate group, the MS-PEG-MS (0.44 mmol) was reacted with potassium O-ethyldithiocarbonate (0.88 mmol) in a mixture of dry THF and DMF (14:1, vol:vol) cosolvent (50 mL). After the mixture was allowed to react for several hours with magnetic stirring, chloroform was added to the mixture, and then it was washed with a saturated NaCl aqueous solution several times to eliminate the impurities. The organic phase was concentrated by evaporation after drying with sodium sulfate followed by pouring into an excess amount of diethyl ether. The precipitate, dithiocarbonate-PEG (DTCPEG-DTC), was dried in vacuo and then analyzed by SEC and 1H NMR spectroscopy (400 MHz, DMSO-d6). An end functionality of approximately 97% was confirmed. The SEC chart and NMR spectra of DTC-PEG-DTC (5k) are shown in Figures S3 and S4 in the Supporting Information, respectively. The deprotection of the dithiocarbonate group of DTC-PEGDTC was carried out using n-propylamine. DTC-PEG-DTC (0.02 mmol) and n-propylamine (14 mmol) were dissolved in 10 mL of dry THF. After the mixture was allowed to react for several hours with magnetic stirring, chloroform was added to the mixture, and then it was washed with a saturated NaCl aqueous solution several times to eliminate the impurities. The organic phase was concentrated by evaporation after drying with sodium sulfate, followed by pouring

into an excess amount of diethyl ether. Finally, the obtained polymer was freeze-dried with benzene (98% yield). The precipitate, SHPEG-SH, was analyzed by 1H NMR spectroscopy (400 MHz, DMSOd6), and an end-functionality of approximately 95% was confirmed. The SEC chart and NMR spectra of SH-PEG-SH (5k) are shown in Figures S5 and S6 in the Supporting Information, respectively. Preparation of a MeO-PEG-SH (2k) Tethered-Chain Surface Having Different Chain Densities on a Gold Sensor Chip. The MeO-PEG-SH (2k) modification of the gold sensor surface was done on a Biacore 3000 SPR instrument (Biacore AB, Uppsala, Sweden). Bare gold sensor chips (SIA Kit Au) for SPR measurement were purchased from Biacore AB (Uppsala, Sweden). The bare gold sensor chips were cleaned with piranha solution (3:1 volume of concentrated H2SO4 and hydrogen peroxide (30% w/v)) at room temperature and then rinsed with a copious amount of water prior to use. After a bare gold sensor chip was placed in the SPR instrument, MeO-PEG-SH (2k) solution containing 0.05 M phosphate buffer solution (pH 7.4, containing 1 M NaCl) was injected at a constant flow rate of 5 µL/min at 37 °C. To construct a MeO-PEG-SH (2k) tethered-chain surface with low chain density, we monitored the changes in the SPR sensorgram caused by the MeO-PEG-SH (2k) adsorption and stopped the sample injection manually when the value of the SPR angle shift was between 0.025 and 0.2°. In this experiment, MeO-PEG-SH (2k) concentrations between 0.001 and 1 mg/mL were used, and all the injection times for the MeO-PEGSH (2k) solution were less than 10 min. Optimization of MeO-PEG-SH (2k) Brush Density on the Gold Surface for the Construction of a Highly Reactive MeO-PEGSH (2k)/SH-PEG-SH (5k) Mixed Tethered-Chain Surface. As shown in Figure S1, since the SEC chart of SH-PEG-SH (5k) indicated the formation of SH-PEG-SH oxidized multimers, column purification of SH-PEG-SH (5k) was carried out as follows: a mixture of n-propylamine (0.14 mmol) and DTC-PEG-DTC (0.2 µmol) in 500 µL aqueous solution stirred after 30 min at 37 °C was purified on a Sephadex G-25 column (GE Healthcare, UK, Ltd., Buckinghamshire, England) using 0.05 M phosphate buffer solution (pH 7.4, containing 1 M NaCl). To construct the MeO-PEG-SH (2k)/SHPEG-SH (5k) mixed tethered-chain surface, a purified SH-PEG-SH solution containing 0.05 M phosphate buffer (pH 7.4, containing 1 M NaCl) was flowed onto MeO-PEG-SH (2k) preimmobilized sensor surfaces having different chain densities at a constant flow rate of 5 µL/min for 30 min at 37 °C. Additionally, the shorter MeOPEG-SH (2k) was also layered as the filler on the constructed surfaces thus prepared by repetitive injection. To construct a highly reactive MeO-PEG-SH (2k)/SH-PRGSH(5k) mixed tethered-chain surface, the amount of MeO-PEG-SH for preimmobilization was optimized in an experiment of BSAmaleimide conjugation on the surface. The BSA-maleimide solution was prepared as follows: After 54.4 µL of 1 mM sulfo-SMCC solution was added to 500 µL of 10 µM BSA solution, the mixture was stirred for 30 min at 37 °C and the obtained BSA-maleimide was purified on a Sephadex G-25 column using 0.05 M phosphate buffer (pH 7.4, containing 0.15 M NaCl). The conjugation of BSAmaleimide with the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface was analyzed by SPR analysis. The SPR angle shift of the conjugated protein on the PEG-graft surface was converted using the following relationship: 0.01° ) 8.8 ng/cm;2 this relationship been determined through the radiolabeling assay of streptavidine conjugation with a PEG-modified gold surface.12b Since the conjugation between the maleimide group and the sulfanyl group requires a long reaction time,22 BSA-maleimide was reacted with the MeO-PEG-SH (2k)/SH-PEG-SH (5k) surface for 3 h. Thus, a solution of 5 µM BSA-maleimide and BSA in 0.05 M phosphate buffer (pH 7.4, containing 0.15 M NaCl) was flowed onto MeOPEG-SH (2k)/SH-PRG-SH (5k) surfaces having various SH-PEGSH (5k) densities for 3 h at a flow rate of 1 µL/min at 25 °C and for 30 min at a flow rate of 5 µL/min at 25 °C. The most reactive MeO-PEG-SH (2k)/SH-PRG-SH (5k) surface was obtained when

(21) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals; Pergamon: New York, 1980.

(22) Houseman, B. T.; Gawalt, E. S.; Mrksich, M. Langmuir 2003, 19, 1522– 1531.

out by SPR analysis and X-ray photospectroscopy. Additionally, the performance of the protein-installed MeO-PEG-SH (2k)/ SH-PEG-SH (5k) surface as a protein sensor platform is also discussed in the Results and Discussion section.

Experimental Section

9626 Langmuir, Vol. 24, No. 17, 2008

Yoshimoto et al.

Figure 1. Schematic illustration of the preparation of a sulfanyl-terminated mixed-PEG tethered-chain surface on the gold surface by consecutive treatment with MeO-PEG-SH (2k) and SH-PEG-SH (5k). In this strategy, the density of the free sulfanyl groups at one end of SH-PEG-SH (5k) on the surface can be controlled by the chain density of the preimmobilized MeO-PEG-SH (2k) on the surface.

the SPR angle shift for the immobilization of MeO-PEG-SH (2k) was about 0.11°. Evaluation of the Protein Conjugation with the Most Reactive MeO-PEG-SH (2k)/SH-PEG-SH (5k) Mixed Tethered-Chain Surface. The selective conjugation of BSA-maleimide and the antiFITC antibody Fab′ fragments with the most reactive MeO-PEGSH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface was carried out in an SPR instrument. As a control, the same experiments were performed using a SH-PEG-SH (5k)-modified surface and a conventional MeO-PEG-SH (2k)/MeO-PEG-SH (5k) mixed tetheredchain surface. Four kinds of protein solutions, 10 µM BSA-maleimide, 10 µM BSA, 1 µM FITC antibody Fab′ fragments, and 1 µM antibody in 0.05 M phosphate buffer (pH 7.4, containing 0.15 M NaCl), were prepared. The anti-FITC antibody Fab′ fragments were obtained from a 2 h reaction of 12 µL of 10 mM DTT with 200 µL of 10 µM anti-FITC antibody F(ab′)2 at room temperature. The purification of the antibody Fab′ fragments was carried out by centrifugation on a Microcon YM-30 column (Nihon Millipore Co., Tokyo, Japan) using 0.05 M phosphate buffer (pH 6.5, containing 5.5 mM EDTA). A SH-PEG-SH (5k)-modified gold surface was constructed in the SPR instrument by injecting 0.2 mM SH-PEG-SH (5k) in phosphate buffer solution (pH 7.4, 0.05 M, containing 1 M NaCl) for 30 min at a constant flow rate of 5 mL/min at 37 °C. To fabricate a conventional MeO-PEG-SH (2k)/MeO-PEG-SH (5k) mixed tetheredchain-modified gold surface, 0.2 mM MeO-PEG-SH (5k) in phosphate buffer solution (pH 7.4, 0.05 M, containing 1 M NaCl) was injected for 30 min at a constant flow rate of 5 mL/min at 37 °C, followed by the injection of 0.2 mM MeO-PEG-SH (2k) in phosphate buffer solution (pH 7.4, 0.05 M, containing 1 M NaCl) on the gold surface. Additionally, the shorter MeO-PEG-SH (2k) was also layered as a filler on the constructed surface thus prepared by repetitive injection. Each protein solution was flowed onto these PEG-modified gold sensor surfaces for 30 min at a flow rate of 5 µL/min at 25 °C. The SPR angle shift of the adsorbed protein on the PEG-graft surface was converted using the following relationship: 0.01° ) 8.8 ng/cm2. X-ray Photoelectron Spectroscopy Analysis of the MeO-PEGSH (2k)/SH-PEG-SH (5k) Mixed Tethered-Chain Surface. The X-ray photoelectron spectra of the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain gold surface and the conventional MeOPEG-SH (2k)/MeO-PEG-SH (5k) mixed tethered-chain gold surface were collected using an XPS spectrometer (ESCA-300; Scienta, Uppsala, Sweden) with a monochromatized Al KR X-ray source (1486.6 eV) at a power of 2.4 kW and a base pressure of 7.3 × 10-8 Pa in the analytical chamber. The analyzer slit width and the pass energy were 2.0 mm and 150 eV, respectively. Mixed-PEG tetheredchain surfaces were constructed on bare gold sensor chips (SIA kit Au: Biacore AB, Uppsala, Sweden). The MeO-PEG-SH (2k)/SHPEG-SH (5k) mixed tethered-chain surface on the gold surface was prepared for XPS analysis by consecutive treatment with 0.2 mM MeO-PEG-SH (2k) in phosphate buffer solution (pH 7.4, 0.05 M, containing 1 M NaCl) for 1 min at 37 °C and 0.1 mM SH-PEG-SH (5k) in phosphate buffer solution (pH 7.4, 0.05 M, containing 1 M NaCl) for 30 min at 37 °C. Additionally, the shorter MeO-PEG-SH (2k) was also layered as a filler on the constructed surfaces thus prepared by immersing the chip in a solution of MeO-PEG-SH (2k) in phosphate buffer twice. A conventional MeO-PEG-SH (2k)/MeOPEG-SH (5k) mixed tethered-chain surface for XPS analysis was also prepared by consecutive treatment with 0.1 mM MeO-PEG-SH

Figure 2. Plots of the normalized chain density of SH-PEG-SH (5k) immobilized on a MeO-PEG-SH (2k)-immobilized gold surface versus the normalized chain density of MeO-PEG-SH (2k) immobilized on a bare gold surface, as determined by SPR analysis.

Figure 3. Amount of BSA-maleimide (b) and BSA (O) adsorbed on a MeO-PEG-SH (2k)/SH-PRG-SH (5k) mixed tethered-chain surface having SH-PEG-SH (5k) with various chain densities.

(5k) in phosphate buffer solution (pH 7.4, 0.05 M, containing 1 M NaCl) for 30 min at room temperature and 0.2 mM MeO-PEG-SH (2k) in phosphate buffer solution (pH 7.4, 0.05 M, containing 1 M NaCl) for 30 min at 37 °C twice. All the samples were rinsed with water and dried overnight in vacuo. XPS measurements were carried out at an electron takeoff angle of 90°. Survey scans were performed to identify the C, S, and Au elements. Narrow scans of the C 1s, S 2p, and Au 4f levels were recorded, and data were collected from 25 scanning cycles. The spectrometer function was adjusted to give a value of 286.5 eV for the C 1s peak of the C-O bond. Selective Detection of Target Molecules on the Protein-Installed MeO-PEG-SH (2k)/SH-PEG-SH (5k) Mixed Tethered-Chain Surface. The selective detection of anti-BSA antibody on the BSAmaleimide-installed MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface was evaluated by SPR analysis. A solution of 1 µM anti-BSA antibody and 10 µM BSA in 0.05 M phosphate buffer (pH 7.4, containing 0.15 M NaCl) was flowed onto the prepared

Sulfanyl-Terminated PEG-Brushed Layer on a Gold Surface

Langmuir, Vol. 24, No. 17, 2008 9627

Figure 4. Amounts of various proteins (BSA-maleimide and BSA (A), FITC antibody Fab′ fragment and anti-FITC antibody (B)) adsorbed on the most reactive MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface (a), a SH-PEG-SH (5k)-modified surface (b), and a conventional MeO-PEG-SH (2k)/MeO-PEG-SH (5k) mixed tethered-chain surface (c), as determined by SPR analysis.

Figure 5. XPS S 2p spectra of the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain gold surface (A) and the conventional MeO-PEGSH (2k)/MeO-PEG-SH (5k) mixed tethered-chain gold surface (B).

BSA-maleimide-installed MeO-PEG-SH (2k)/SH-PEG-SH (5k) surface for 30 min at a flow rate of 5 µL/min at 25 °C. As a control, 1 µM anti-BSA antibody in 0.05 M phosphate buffer (pH 7.4, containing 0.15 M NaCl) was flowed onto a MeO-PEG-SH (2k)/ SH-PEG-SH (5k) mixed tethered-chain surface without BSAmaleimide. In the same way as mentioned above, the selective detection of FITC-BSA on the anti-FITC antibody Fab′ fragment-installed MeOPEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface was also evaluated by SPR analysis. A solution of 1 µM FITC-BSA and 1 µM anti-FITC antibody in 0.05 M phosphate buffer solution (pH 7.4, containing 0.15 M NaCl) was flowed onto the prepared antiFITC antibody Fab′ fragment-installed MeO-PEG-SH (2k)/SH-PEGSH (5k) surface for 30 min at a flow rate of 5 µL/min at 25 °C. As a control, 1 µM FITC-BSA in 0.05 M phosphate buffer (pH 7.4, containing 0.15 M NaCl) was flowed onto a MeO-PEG-SH (2k)/ SH-PEG-SH (5k) mixed tethered-chain surface without BSAmaleimide.

Results and Discussion Our strategy of constructing a sulfanyl-terminated PEG-brushed gold surface is based on the formation of a mixed-PEG brush by the combined use of the longer SH-PEG-SH (5k) and the

shorter MeO-PEG-SH (2k), where the preimmobilized shorterMeO-PEG-SH (2k) prevents loop formation in the longer SHPEG-SH (5k) on the gold surface, as shown in Figure 1. If a densely packed MeO-PEG (2k) layer is constructed on the gold surface, it prevents the immobilization of SH-PEG-SH (5k) on the surface, resulting in the construction of a less reactive PEG tethered-chain surface, whereas if the density of the preimmobilized MeO-PEG (2k) is too low, it may not prevent the formation of loops in the ensuing process of SH-PEG-SH (5k) immobilization. On the contrary, if the shorter MeO-PEG-SH (2k) immobilized on the gold surface has the appropriate chain density, the MeO-PEG-SH (2k) prevents loop formation in SH-PEG-SH (5k), and the most reactive sulfanyl-terminated mixed-PEG gold surface can be constructed. To validate this strategy, SPR and XPS analyses of the constructed PEG-modified surfaces were carried out. Figure 2 shows the plots of the normalized chain density of SH-PEG-SH (5k) immobilized on MeO-PEG-SH (2k) preimmobilized on the gold surface versus the chain density. The chain densities of SH-PEG-SH (5k) and MeO-PEG-SH (2k) on the gold surface were measured by the SPR sensor and standardized to the normalized densities using the equations

CD(p) ) ∆θA ⁄ MW

(1)

CD(N) ) CD(p) ⁄ CD(PEG-EG-SH(2k),∆θ)0.2°)

(2)

where CD(p) is the chain density of the immobilized polymer, ∆θ is the angle shift obtained from the SPR measurement, A is a conversion constant with which the angle shift can be changed into the mass of the polymer, and MW is the molecular weight of polymer. In this study, the normalized surface density (CD(N)) of SH-PEG-SH (5k) and MeO-PEG-SH (2k) was defined using eq 2, where the CD(p) was normalized by the CD(p) of MeOPEG-SH (2k) for which the SPR angle shift was 0.2°, which is the maximum SPR angle shift caused by the injection of SHPEG-SH under these experimental conditions. In the region of SPR angle shifts above 0.2°, almost no increase in the SPR angle shift was observed. In Figure 2, as the normalized chain density of the preimmobilized shorter MeO-PEG-SH (2k) on the surface decreased, the chain densities of the longer SH-PEG-SH (5k) increased. This means that the chain density of the long SHPEG-SH (5k) on the sensor chips could be controlled linearly by the amount of preimmobilized short MeO-PEG-SH (2k). Furthermore, to estimate the presence of free sulfanyl groups on the constructed surfaces, maleimide-installed BSA was reacted

9628 Langmuir, Vol. 24, No. 17, 2008

Yoshimoto et al.

Figure 6. (a) SPR sensorgrams caused by the injection of 1 µM anti-BSA antibody solution onto the BSA-maleimide-installed MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface (BSA-terminated-PEG surface) and the most reactive MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface (SH-terminated-PEG surface). As a control, 10 µM BSA solution was also injected onto the BSA-terminated-PEG surface. (b) SPR sensorgrams caused by the injection of 1 µM FITC-BSA solution on the anti-FITC antibody Fab′ fragment-installed MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface (Fab′-terminated-PEG surface) and the SH-terminated-PEG surface. A solution of 1 µM anti-FITC antibody was also injected onto the Fab′-terminated-PEG surface, as a control.

with the MeO-PEG-SH (2k)/SH-PEG-SH (5k) tethered-chain surface via a Michael reaction. Figure 3 shows the adsorbed amount of BSA and BSA-maleimide on MeO-PEG-SH (2k)/ SH-PEG-SH (5k) surfaces with various chain densities of SHPEG-SH (5k). In the case of BSA without the maleimide group, the adsorption amounts were very low (0.1-0.26 × 10-3 molecules/nm2) and there were no differences in the values across the whole region of the normalized chain density of SH-PEGSH (5k). On the contrary, in the case of BSA-maleimide, a large adsorption amount (1.4-1.8 × 10-3 molecules/nm2) was observed only in the region where the normalized chain density of SHPEG-SH (5k) was 0.4. These results clearly indicate that the conjugation of the free sulfanyl group with the maleimide group occurred on the MeO-PEG-SH (2k)/SH-PEG-SH (5k) surface, indicating the formation of sulfanyl-terminated PEG tethered chains on the gold surface. As mentioned above, in the region of high chain density of SH-PEG-SH (5k) (i.e., where the normalized chain density is higher than 0.43 × 10-3 molecules/ nm2), loop formation in SH-PEG-SH (5k) occurred on the surface, resulting in the decrease the number of free sulfanyl groups. To evaluate the performance of the most reactive MeO-PEGSH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface, the adsorption amounts of BSA-maleimide and FITC antibody Fab′ fragments on a SH-PEG-SH (5k)-modified gold surface and a conventional MeO-PEG-SH (2k)/MeO-PEG-SH (5k) mixed

tethered-chain gold surface were measured by SPR spectroscopy. The experimental results for the protein adsorption on these surfaces are summarized in Figure 4. Figure 4 clearly shows the selective binding of BSA-maleimide and FITC antibody Fab′ fragments to the constructed MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface. On the surface, very large adsorption amounts of BSA-maleimide and FITC antibody Fab′ fragments were observed, in sharp contrast to the nonspecific adsorption of BSA and anti-FITC antibody (Figure 4A-(a) and B-(a)). This result also supports the presence of sulfanyl groups on the MeO-PEG-SH (2k)/SH-PEG-SH (5k) surface, because the sulfanyl termini of the Fab’ fragments could react with the free sulfanyl groups on the MeO-PEG-SH (2k)/SH-PEG-SH (5k) surface easily, forming covalent disulfanyl bonds. In contrast, the conventional MeO-PEG-SH (2k)/MeO-PEG-SH (5k) mixed tethered-chain gold surface reported by us previously17 had a completely nonfouling character, and neither specific proteins nor free proteins showed very low adsorption (Figure 4, part A, bar graph c; part B, bar graph c). Interestingly, the most reactive MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface prevented nonspecific protein adsorption to the same degree as the MeO-PEG-SH (5k)/MeO-PEG-SH (2k) surface (white bars in Figure 4, part A, bar graph a; part B, bar graph a). In contrast, on the SH-PEG-SH-modified gold surface, the amounts of immobilized BSA-maleimide and antibody Fab′

Sulfanyl-Terminated PEG-Brushed Layer on a Gold Surface

fragments were decreased. Furthermore, the surface did not show a strongly nonfouling character and the adsorbed amount of BSAmaleimide was almost equal to that of BSA, whereas the adsorbed amount of FITC antibody Fab′ fragments was twice that of antiFITC antibody (Figure 4, part A, bar graph b; part B, bar graph b). This could be explained by loop formation in SH-PEG-SH (5k) on the gold surface, resulting in a decreased number of sulfanyl groups on the PEG surface and an increase in the nonspecific adsorption of proteins. These results suggest that a SH-PEG-SH (5k)-brushed layer formed on the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface, with MeOPEG-SH (2k) preventing loop formation. Therefore, the free sulfanyl groups at the end of the SH-PEG-SH (5k) on the surface could be utilized for the immobilization of specific biomolecules on the PEG tethered-chain surface. The presence of free sulfanyl groups on the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface was directly confirmed by XPS analysis. Figure 5 shows the XPS S2p spectra of the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tetheredchain surface and the conventional MeO-PEG-SH (2k)/MeOPEG-SH (5k) mixed tethered-chain surface. The S2p spectra of sulfanyl-terminated molecules chemisorbed on a gold substrate has a doublet structure due to the presence of the S2p3/2 and S2p1/2 peaks, and all the spectra can be fitted using a 2:1 peak area ratio and a 1.2 eV splitting.19c,d It has been reported that S 2p3/2 peaks were usually observed at a binding energy (BE) of about 161.9 eV, consistent with the sulfur atoms bound to the gold surface. On the other hand, S2p3/2 peaks of sulfur atoms unbound to the gold surface were observed at a BE of about 163.5 eV. In this study, the peaks of Au-bound sulfur at a BE of 161.6 eV were observed on both PEG-modified gold surfaces (Figure 5). However, no detectable intensity was present in the BE region above 164 eV on the conventional MeO-PEG-SH (2k)/MeOPEG-SH (5k) mixed tethered-chain surface (Figure 5 (B)), and the peaks of Au-unbound sulfur at a BE of 163.2 eV were observed on the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tetheredchain surface only (Figure 5A). These results strongly support the presence of free sulfanyl groups on the MeO-PEG-SH (2k)/ SH-PEG-SH (5k) mixed tethered-chain surface, which was also indicated by the experimental results of the SPR analysis based on specific protein conjugation. The performance of the protein-installed MeO-PEG-SH (2k)/ SH-PEG-SH (5k) mixed tethered-chain surface thus obtained was assessed by the detection of target proteins based on SPR analysis. The SPR sensorgrams caused by the injection of protein solutions onto the BSA-maleimide-installed surface and the FITC antibody Fab’ fragment-installed surface are shown in parts a

Langmuir, Vol. 24, No. 17, 2008 9629

and b of Figure 6, respectively. On the BSA-maleimide-installed surface, the SPR angle shift caused by the injection of 1 µM target protein, anti-BSA antibody, was about 10 times larger than that caused by the injection of 10 µM BSA. The MeOPEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface also prevented the adsorption of anti-BSA antibody at SPR angle shifts of less than 80°; therefore, the specific interaction between BSA and anti-BSA antibody was detectable on this proteininstalled mixed-PEG surface. Similarly, the specific detection of FITC-BSA was also accomplished by the FITC antibody Fab′ fragment-installed MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tethered-chain surface. The SPR angle shift caused by the injection of 1 µM target protein, FITC-BSA, on this surface was 6-10 times larger than that caused by 1 µM anti-FITC antibody solution on this surface. Additionally, almost no adsorption of FITCBSA on the MeO-PEG-SH (2k)/SH-PEG-SH (5k) mixed tetheredchain surface occurred.

Conclusions In this paper, we described a new strategy for the facile construction of a sulfanyl-terminated mixed-PEG-brushed layer, which was newly constructed by consecutive treatment with a short sulfanyl-ended semitelechelic PEG (MeO-PEG-SH) (2k) and a long sulfanyl-ended telechelic PEG (SH-PEG-SH) (5k). Under optimized experimental conditions, the MeO-PEG-SH (2k) prevented loop formation in SH-PEG-SH (5K) and formed a sulfanyl-terminated mixed-PEG tethered-chain layer on the gold surface. Interestingly, the constructed SH-PEG-SH (5k)/ MeO-PEG-SH (2k) layer showed a fairly low nonspecific protein adsorption, whereas the maleimide-installed protein and antibody Fab′ fragments were immobilized easily on the free sulfanyl groups of the constructed mixed-PEG layer. The proteins conjugated on the sulfanyl-terminated mixed-PEG surfaces thus obtained were confirmed to retain their activity of target protein recognition, accompanied by a highly nonfouling character. Accordingly, the sulfanyl-terminated PEG-brushed layer thus prepared can be regarded as a promising tool for the effective immobilization and detection of specific proteins. Acknowledgment. This research was partially supported by the Ministry of Education, Science, Sports and Culture, Grantin-Aid for Scientific Research (A), No. 18200033. Supporting Information Available: Size exclusion chromatogram and 1H NMR spectrum of SH-PEG-SH (5k) and precursors. This material is available free of charge via the Internet at http://pubs.acs.org. LA8012798