Improving Protein Transfer Efficiency and Selectivity in Affinity Contact

ARTICLE pubs.acs.org/Langmuir

Improving Protein Transfer Efficiency and Selectivity in Affinity Contact Printing by Using UV-Modified Surfaces Chih-Hsin Chen and Kun-Lin Yang* Department of Chemical and Biomolecular Engineering, National University of Singapore, 4 Engineering Drive 4, Singapore 117576, Singapore

bS Supporting Information ABSTRACT: Affinity contact printing (RCP) is a technique that allows the selective capture of a target protein from solutions to a polymeric stamp decorated with an antibody, and then the target protein is printed onto a solid surface. The success of RCP critically relies on the precise control of protein
surface interactions. Here, we report a study on the effect of UV on the protein
surface interactions between protein and polydimethylsiloxane stamps and between protein and glass slides decorated with N,N-dimethyl-n-octadecyl-3-aminopropyltrimethoxysilyl chloride (DMOAP). Our results show that UV-modified surfaces can be used to improve the transfer efficiency and selectivity of proteins during RCP. For example, the protein transfer efficiency of human IgG onto a DMOAP-coated slide increases from 7.2% to 45.1% after the UV treatment. On the basis of these results, UV-modified surfaces were employed to develop a RCP system for protein detection. The detection limit of anti-IgG in this system is around 10 ng/mL, and the dynamic range is 4 orders of magnitude.

’ INTRODUCTION Affinity microcontact printing (RCP), which can be used to transfer proteins selectively from solutions to solid surfaces, has been considered as a potential method for protein detection.1
4 In a typical RCP, probe proteins are first immobilized on an elastomeric stamp, usually made of polydimethylsiloxane (PDMS), to capture target proteins from protein solutions. If a target protein recognizes a probe protein and binds to the probe protein (such as in the case of an antibody
antigen pair), a protein complex is formed on the stamp surface. Subsequently, the stamp is brought into conformal contact with a solid surface. When the PDMS stamp is peeled off from the solid surface, three different scenarios can be expected for the fate of the protein complex. First, if the interaction between the protein complex and the solid surface is stronger than the interaction between two protein pairs, the complex will break, allowing the transferring of the target protein to the solid surface (Scheme 1a). Therefore, detection of the target protein can be achieved simply by measuring the amount of proteins on the solid surface. In the second scenario, the interaction between the protein complex and the stamp surface is weaker than the interaction between two protein pairs. In this case, the entire protein complex will be transferred onto the solid surface (Scheme 1b). This is not desired because the lack of transferring specificity will lead to a false positive in protein detection. Lastly, if the interaction between protein complex and the solid surface is weaker than the interaction between probe and target proteins, the target proteins will not be transferred onto the solid surface (Scheme 1c) and result in false negative detection. Therefore, in the application of RCP for protein detection, precise control of the protein
surface interaction is very important. r 2011 American Chemical Society

In the past, control of interactions between proteins and a PDMS stamp is mostly achieved by chemical modifications of the PDMS stamp using organic silanes.4
10 For example, Renault et al.8 modified a PDMS stamp by using N-hydroxysulfosuccinimide (NHS) ester, which can react with the amine groups of protein mouse IgG. Thus, this protein is covalently immobilized on the stamp surface. After the binding of antimouse IgG to the surface-immobilized mouse IgG, antimouse IgG can be transferred selectively by using RCP. However, the surface modification requires several steps. First, the PDMS stamp has to be treated with oxygen plasma to create surface silanol groups. Then, the silanol groups are reacted with amine-terminated silanes to create an amine-modified surface. These amine groups further react with bifunctional cross-linkers to produce NHS ester terminal groups for covalent immobilization of the proteins. Because surface reaction is involved in each step, it is not easy to control the quality of the stamp. To control the interactions between proteins and solid surfaces (such that proteins can be transferred to the solid surface during RCP), solid surfaces are often coated with organosilanes, organothiols, or polymers.11
15 For example, Rozkiewicz et al.15 showed that aldehyde-terminated self-assembled monolayer (SAMs), which was obtained by the reaction between an amine-terminated SAMs and terephthaldialdehyde, can be used to facilitate the microcontact printing of proteins. The covalent interactions between proteins and aldehyde-terminated surfaces are more stable than noncovalent interactions. However, the wet process has some intrinsic limitations such as Received: February 10, 2011 Revised: March 16, 2011 Published: April 05, 2011 5427

dx.doi.org/10.1021/la200535c | Langmuir 2011, 27, 5427–5432

Langmuir Scheme 1. Schematic Illustration of Three Different Scenarios of RCP When the PDMS Stamp Is Peeled off from the Solid Surfacea

a

(a) Breaking the interaction between probe and target protein, only the target protein is transferred onto the solid surface. (b) Breaking the interaction between the stamp and protein complex, the entire protein complex is transferred onto the solid surface. (c) Breaking the interaction between protein complex and the stamp, the target protein remains on the stamp surface.

the ability to pattern the surface or to control the dosage of modification. Recently, it was demonstrated that solid surfaces can be modified by using UV.16
21 The advantages of using UV include shorter reaction time and the possibility for surface patterning. Besides, the dosage of surface modification can be adjusted simply by tuning the UV exposure time. For instance, Hao et al.16 showed that UV treatment is able to increase the wettability on the poly(ethylene terephthalate) (PET) surface. After UV treatment, carboxylic groups are formed only in the UV-exposed regions. In our past studies, we also found that PDMS can be easily modified by UV (254 nm) under ambient conditions.17 When microcontact printing was performed by using a UV-defined flat PDMS stamp, only the proteins immobilized on the PDMS surface without UV exposure will be transferred onto the glass slide while the proteins immobilized on the PDMS surface with UV exposure will remain on the stamp. Besides UV-modified PDMS surfaces, one of our previous studies has also shown that an inert hydrocarbon monolayer on silicon surfaces can also be easily activated by UV.22 When protein solutions are dispensed on the UV-defined hydrocarbon monolayer, protein patterns can be formed spontaneously on the area with UV exposure, which suggests that the solid surface modified with hydrocarbon monolayer also has strong interactions with proteins after UV exposure. Both of the results above demonstrated that UV modification can be used to increase the protein
surface interactions; nevertheless, the UV-modified surfaces has never been used in RCP. In this study, we study the effect of UV on the protein
surface interactions between protein and PDMS stamps and between protein and glass slides. On the basis of our past studies, we hypothesize that the UV treatment can strengthen the interactions between probe proteins and PDMS stamps and the interactions between the target proteins and glass slides. As shown in Scheme 1, both factors contribute positively to the success of applying RCP for protein detection. Furthermore, by using IgG and anti-IgG as a model protein, we compare the transfer efficiency and selectivity of RCP with and without using UV-treated surfaces.

’ EXPERIMENTAL SECTION

ARTICLE

human serum albumin (HSA), antihuman serum albumin (anti-HSA), and sodium borohydride (NaBH4) were purchased from Sigma Aldrich (Singapore). Cy3 monoreactive dye pack for tagging IgG and anti-HSA was purchased from GE Healthcare. A poly(dimethylsiloxane) (PDMS) kit, Sylgard 184, was purchased from Dow Corning. Nitrogen gas (purity 5N) was purchased from SOXAL. All aqueous solutions were prepared in deionized water with a resistance of 18.2 MΩ (Millipore). Preparation of DMOAP-Coated Slides. To clean the surface, glass slides were immersed in a 5% Decon-90 solution (a commercially available detergent) for 2 h, sonicated in DI water for 15 min, and rinsed thoroughly with DI water twice. After this, the slides were dried under a stream of nitrogen. The cleaned glass slides were immersed in an aqueous solution containing 0.1% (v/v) DMOAP for 5 min and then rinsed with copious amounts of DI water. DMOAP-coated slides were dried under a stream of nitrogen and heated in a 100 °C vacuum oven for 15 min. Preparation of PDMS Stamps. PDMS stamps were prepared by casting Sylgard 184 on a clean and flat silicon wafer. The stamps were then degassed in vacuum to remove air bubbles and cured at 100 °C for 3 h. To clean the PDMS stamps, a Soxhlet device was used to extract unreacted starting materials of PDMS into ethanol. Finally, the clean PDMS stamps were cured at 100 °C for 1 h to vaporize any ethanol trapped inside the PDMS stamp.

Surface Modification of PDMS Stamps and DMOAPCoated Slides by Using UV. The untreated surface was placed 1.5 cm below a UV pen lamp (254 nm, Sigma-Aldrich, model 11SC-1) in an open-ended glass tube. The exposure time for PDMS stamps and DMOAP-coated slides are 5 and 1 min, respectively. Preparation of Aldehyde-Terminated PDMS Stamps. Aldehyde-terminated PDMS stamps were obtained by coating (triethoxysilyl)butanal (TEA) on a PDMS surface. More details about the preparation steps can be found in our previous publication.23 Surface Reduction Test. Sodium borohydride was dissolved in carbonate buffer (0.1 M, pH = 10) to a final concentration of 50 mM. One microliter of the reducing solution was then dropped onto a UVmodified PDMS surface. After 2 h, the sample was rinsed with DI water and dried with nitrogen. Protein Immobilization on PDMS Stamps. IgG or Cy3
IgG (10 μg/mL in PBS buffer) were spotted on the surface of UV-modified PDMS stamps by using a micropipet. The volume of each spot was 300 nL. After immobilization under room temperature for 1 h, the stamp was cured at 50 °C for 30 min to remove the excess protein solutions. The protein immobilized PDMS stamp was then rinsed with PBS buffer (containing 0.1% Tween 20) and DI water and then dried under a stream of nitrogen. Affinity Contact Printing. The PDMS stamp with immobilized IgG was covered with Cy3
anti-IgG or Cy3
anti-HSA (50 μg/mL in 1 PBS buffer, containing 0.1% Tween 20) for 1 h. After that, it was rinsed with DI water and then dried under a stream of nitrogen. The protein-immobilized PDMS surface was brought into conformal contact with a DMOAP-coated slide. After 10 min of contacting, the stamp was peeled off from the surface, and the glass substrate was dried with nitrogen for subsequent analysis. Detection of Fluorescence Signals. The protein-printed glass slides were scanned with 40 μm spatial resolution by using a GenePix 4100A laser scanner (Molecular Devices). The PMT (photomultiplier tubes) setting was maintained at 500 for all analysis. All images were analyzed by using the GenePix Pro 6.1 software provided by the manufacturer. Determination of Transfer Efficiency of rCP. The transfer efficiency η of RCP was determined by the following equation

Materials. Glass slides were obtained from Fisher Scientific. N,NDimethyl-n-octadecyl-3-aminopropyltrimethoxysilyl chloride (DMOAP), Tween 20, human IgG (IgG), Cy3-labeled anti-human IgG (Cy3
anti-IgG),

η¼ 5428

ΓA  100% ΓB

ð1Þ

dx.doi.org/10.1021/la200535c |Langmuir 2011, 27, 5427–5432

Langmuir

Figure 1. Comparison of the fluorescence intensity on UV-modified, unmodified, and aldehyde-terminated PDMS stamps with immobilized Cy3
IgG before and after contact printing on a DMOAP-coated slide. It shows that Cy3
IgG immobilized tightly on the UV-modified and aldehyde-terminated PDMS stamp and does not detach from the stamp during contact printing. where ΓA is the protein surface density on a UV-modified DMOAP slide after RCP and ΓB is the protein surface density on the PDMS stamp before RCP. To quantify the surface density of proteins, protein solutions containing different concentrations of Cy3
IgG or Cy3
anti-IgG were spotted on UV-modified PDMS surface and UV-modified DMOAP-coated slide, respectively. These surfaces were dried at 50 °C to ensure that all proteins in the solutions were deposited on the surface. After this, fluorescence intensity of each spot was analyzed by using a fluorescence scanner to obtain a correlation between the total amount of proteins and the fluorescence intensity.

ARTICLE

Figure 2. Effect of reducing agents (NaBH4) on the immobilization of Cy3
IgG on a UV-modified PDMS stamp. The result indicates that aldehyde groups play a role in the immobilization of protein on the UVmodified surface.

’ RESULTS AND DISCUSSION

Figure 3. Fluorescence images of the DMOAP-coated slide printed by (a) UV-modified PDMS and (b) unmodified PDMS stamp with a circular region of immobilized IgG and immunobinding of Cy3
antiIgG. It shows the importance of using UV-modified PDMS to increase the selectivity of proteins transfer since nonspecifically immobilized proteins are not transferred onto the DMAOP-coated slide by RCP.

Effect of UV on PDMS and Protein
PDMS Interactions. To investigate the influence of UV on the surface properties of PDMS, we first immobilized Cy3
IgG on both UV-modified and unmodified PDMS stamp surfaces and used them for contact printing of Cy3
IgG on a DMOAP-coated slide. Figure 1 shows that for the UV-modified PDMS stamp, the fluorescence intensity on the stamp is negligible (within the experimental error) after the contact printing. This implies that Cy3
IgG is immobilized tightly on the stamp surface and does not detach from the surface during contact printing. In contrast, for the unmodified PDMS stamp, the fluorescence intensity decreases by 76.5% after the contact printing. This result suggests that proteins adsorbed on the unmodified PDMS can be transferred easily during contact printing. On the basis of these results, we hypothesize that UV treatment can lead to the generation of aldehydes on the stamp surface,24,25 and aldehydes can react with free amines of proteins to form imines that immobilize proteins on the surface. To test this hypothesis, we conducted two control experiments. First, we prepared an aldehyde-terminated PDMS stamp and repeated the microcontact printing of Cy3
IgG. Also shown in Figure 1, the fluorescence intensities before and after contact printing are similar. This result is consistent with our proposition that surface aldehyde groups cross-link with proteins and prevent them from being transferred during contact printing. Second, we dispensed a droplet of sodium borohydride (NaBH4) solution on surfaces of a UV-modified PDMS stamp. This reducing agent is known to reduce aldehydes to alcohols. Then, Cy3
IgG solution was dispensed on the stamp surface for

protein immobilization. The fluorescence image in Figure 2 shows that Cy3
IgG cannot be immobilized in the circular region in contact with NaBH4. This is good evidence showing that aldehydes play a role in the immobilization of protein on the UV-modified surface. rCP by Using UV-Modified PDMS Stamps. Next, we employed a UV-modified PDMS stamp to study the transfer of protein during RCP. First, we placed a droplet of solution containing IgG (nonfluorescent) on a UV-modified PDMS stamp (this procedure led to a circular region of covalently immobilized IgG on the stamp surface). Then, the stamp surface was covered with solution containing fluorescently labeled Cy3
anti-IgG. After rinsing and drying, the stamp was used for microcontact printing on a DMOAP-coated slide. The green fluorescence circle in Figure 3a shows the successful transfer of Cy3
anti-IgG (which binds to the immobilized IgG) from the stamp surface to the DMOAP-coated slide. Interestingly, the dark background in Figure 3a suggests that nonspecifically immobilized Cy3
anti-IgG did not transfer during microcontact printing. This result is consistent with our earlier conclusion that protein adsorbs strongly on UV-treated PDMS stamp and does not detach from the surface. For comparison, we repeated the same experiment by using an unmodified PDMS stamp. In this case, Figure 3b shows that both Cy3
anti-IgG binds to IgG and nonspecifically immobilized Cy3
anti-IgG can be transferred. These results, when combined, led us to conclude that UVtreatment can cause the selective transfer of antibodies that bind to surface immobilized antigens while prevent the transfer of 5429

dx.doi.org/10.1021/la200535c |Langmuir 2011, 27, 5427–5432

Langmuir

ARTICLE

nonspecifically immobilized proteins. This means that traditional antifouling strategies such as blocking surfaces with BSA is no longer be needed. Effect of UV on Protein Transfer Efficiency. Because UV treatment can lead to stronger interactions between proteins and PDMS, we hypothesize that similar effects also exist on other types of solid surfaces. Moreover, as shown in Scheme 1a, protein transfer efficiency strongly depends on the protein
surface interactions. In view of this, we investigated whether UV exposure on the DMOAP-coated slide can enhance the protein transfer efficiency in RCP. Figure 4a shows that, when a UVmodified DMOAP-coated slide is used, Cy3
anti-IgG can be printed onto the surface. The fluorescence intensity in Figure 4a is not very uniform, probably because the positions of the light source and the solid surface are fixed during UV-modification. If the sample is rotated under the UV, the surface aldehyde density would be more uniform. By using the fluorescence intensity on the surface and eq 1, we estimated that the transfer efficiency is 45.1%. In contrast, Figure 4b shows that Cy3
anti-IgG can also be printed onto the surface when unmodified DMOAP-coated slide is used; however, the transfer efficiency is only 7.2%. Because past studies have shown that the protein transfer efficiency during RCP is determined by surface properties,26,27 we prepared slides with different surface functionality and compared the protein transfer efficiency. As shown in Table 1, COOH- and CHO-terminated slides result in high protein transfer efficiency (35.1% and 39.6%, respectively). This result is consistent with our proposition that UV exposure leads to the generation of surface aldehyde groups, which enhance protein transfer during RCP. In contrast, both NH2-terminated and clean glass slides result in very low protein transfer efficiency (1.9% and 2.5%, respectively). We point out that the protein transfer efficiency in this study is not related to water contact angles on these surfaces. Surfaces with large and small water contact angles (e.g., DMOAP-coated and clean glass slide) can both lead to low protein transfer efficiency. Tan et al.27 reported that protein adsorbed on a PDMS stamp can be transferred onto solid surfaces only when the water contact angle of the surface is

Figure 4. Fluorescence images of (a) UV-modified DMOAP-coated slide and (b) unmodified DMOAP-coated slide printed by UV-modified PDMS stamp with a circular region of immobilized IgG and immunobinding of Cy3
anti-IgG. It clearly shows that the transfer efficiency of RCP increases when UV-modified DMOAP-coated slide is used.

below a threshold. Apparently, this trend is not observed in our experiment. Furthermore, the transfer efficiency is higher on the UV-modified DMOAP-coated slides (45.1%) than that on the CHO-terminated slides (39.6%). It is probably because strong hydrophobic interactions and tangling effects between the long hydrocarbon chain (C18) of DMOAP and proteins lead to very strong protein adsorption on the DMOAP-coated surface.28,29 These interactions, when combined with the additional aldehyde groups after UV treatment, cause the high transfer efficiency of RCP. Optimized UV Exposure Time. To determine the optimized UV-exposure time for RCP, DMOAP-coated slides were exposed to UV from 0 to 90 s. Figure 5 shows that the protein transfer efficiency increases with the increasing UV-exposure time. This result can be attributed to higher surface densities of CHO and COOH with longer exposure time. However, the transfer efficiency starts to decrease when the exposure time is more than 60 s. One possible explanation is the overoxidation of DMOAP by UV. Because there is only a monolayer of DMAOP coated on the surface, it can be easily removed from the surface after long UV exposure time. The results above show that 60 s of UV exposure is the optimized exposure time to activate the DMOAP-coated slide for maximum protein transfer efficiency. Protein Detection by Using rCP. On the basis of the experimental conditions mentioned above, we employed RCP to detect protein anti-IgG and anti-HSA, respectively, in buffer solution. To test the specificity, UV-modified PDMS stamps with a circular region of immobilized protein IgG and HSA were prepared. Next, the stamp was immersed into buffer solutions containing either Cy3
anti-IgG or Cy3
anti-HSA before RCP. The fluorescent image in Figure 6 shows that only Cy3
anti-IgG was printed on the surface when the stamp was immersed into Cy3
anti-IgG solutions, while only Cy3
anti-HSA was printed on the surface when the stamp was immersed into Cy3
antiHSA solutions. This result suggests that protein detection by

Figure 5. The effect of UV exposure time on DMOAP-coated slides on their transfer efficiency of RCP. It shows that the transfer efficiency of RCP increases as the UV exposure time is increased. However, it starts to drop when the exposure time is longer than 60 s.

Table 1. Water Contact Angles and rCP Transfer Efficiencies of Glass Slides with Various Surface Modifications

transfer efficiency (%) water contact angle, θ (deg)

unmodified DMOAP

UV-modified DMOAP

COOH- terminated

CHO- terminated

NH2- terminated

clean glass slide

7.2 84.6

45.1 75.3

35.1 35.2

39.6 26.7

1.9 49.9

2.5 2.4

5430

dx.doi.org/10.1021/la200535c |Langmuir 2011, 27, 5427–5432

Langmuir

ARTICLE

surface density of protein is too high. These results, when combined, lead us to conclude that RCP can be used to detect proteins in buffer solution. The detection limit of Cy3
anti-IgG is 10 ng/mL and the dynamic range is 4 orders of magnitude.

Figure 6. Fluorescence images of UV-modified DMOAP-coated slides after RCP. The stamp is an UV-modified PDMS stamp with circular regions of immobilized IgG (left) and HSA (left). The stamp was immersed into solutions containing (a) Cy3
anti-IgG and (b) Cy3
anti-HSA for immunobinding before RCP. It clearly shows that Cy3
anti-IgG and Cy3
anti-HSA can be captured by immobilized IgG and HSA, respectively, and then transferred to the UV-modified DMOAPcoated slide during RCP.

’ CONCLUSION In summary, we demonstrated that the interaction between IgG and PDMS stamp can be greatly enhanced by exposing the surface of the PDMS stamp to UV (254 nm). The strong interaction is probably caused by the generation of aldehyde groups on the surface of PDMS stamp after UV exposure, and that leads to covalent immobilization of IgG on the PDMS stamp. By using UV-modified PDMS stamps with immobilized IgG, only the specifically bound, but not the nonspecifically immobilized, Cy3
anti-IgG can be transferred to a DMOAPcoated slide by using RCP. We also demonstrated that UV can be used to increase the interaction between proteins and DMOAPcoated slides. When a UV-modified DMOAP-coated slide was used, RCP exhibits much higher transfer efficiency (45.1%) comparing to the result using unmodified DMOAP-coated slide (7.2%). The higher efficiency can be attributed to the enhanced protein
surface interactions when CHO and COOH groups are generated on the surface after UV exposure. Finally, it was found that Cy3
anti-IgG concentration in the buffer solutions is proportional to the fluorescence intensity on UV-modified DMOAP-coated slides, which is caused by the printed Cy3
anti-IgG. This feature allows us to use RCP to detect as low as 10 ng/mL of Cy3
anti-IgG in aqueous solution with a dynamic range of 4 orders. ’ ASSOCIATED CONTENT

bS

Supporting Information. Surface characterization by using XPS. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: þ65-6516-6614. E-mail: [email protected].

Figure 7. Fluorescence intensities on UV-modified DMOAP-coated slides after RCP. The amount of Cy3
anti-IgG transferred during RCP is proportional to the concentration of Cy3
anti-IgG until 102 μg/mL.

using RCP is specific. Next, we varied the Cy3
anti-IgG concentration in the buffer solution (from 1 mg/mL to 1 ng/mL) to investigate the detection limit of this system. As shown in Figure 7, the fluorescence intensity decreases with the decreasing Cy3
anti-IgG concentration from 100 μg/mL and 10 ng/mL. When the Cy3
anti-IgG concentration is 1 ng/mL, no fluorescence can be detected. This result clearly shows that the amount of Cy3
anti-IgG that can be transferred to the solid surface is proportional to the Cy3
anti-IgG concentration. However, when the Cy3
anti-IgG concentration is 1 mg/mL, we found that the fluorescence intensity on the UV-modified DMOAP-coated slide also decreases, even though the Cy3
antiIgG concentration increases. This result can possibly be attributed to the fluorescence quenching effect on the surface when the

’ ACKNOWLEDGMENT We would like to thank the Agency for Science, Technology and Research (A*STAR) in Singapore (Project 0821010027) for the funding. ’ REFERENCES (1) Jang, C. H.; Tingey, M. L.; Korpi, N. L.; Wiepz, G. J.; Schiller, J. H.; Bertics, P. J.; Abbott, N. L. J. Am. Chem. Soc. 2005, 127, 8912–8913. (2) Tingey, M. L.; Snodgrass, E. J.; Abbott, N. L. Adv. Mater. 2004, 16, 1331–1336. (3) Akbulut, O.; Yu, A. A.; Stellacci, F. Chem. Soc. Rev. 2010, 39, 30–37. (4) Bernard, A.; Fitzli, D.; Sonderegger, P.; Delamarche, E.; Michel, B.; Bosshard, H. R.; Biebuyck, H. Nat. Biotechnol. 2001, 19, 866–869. (5) Delamarche, E.; Geissler, M.; Bernard, A.; Wolf, H.; Michel, B.; Hilborn, J.; Donzel, C. Adv. Mater. 2001, 13, 1164–1167. (6) Tingey, M. L.; Wilyana, S.; Snodgrass, E. J.; Abbott, N. L. Langmuir 2004, 20, 6818–6826. (7) Delamarche, E.; Donzel, C.; Kamounah, F. S.; Wolf, H.; Geissler, M.; Stutz, R.; Schmidt-Winkel, P.; Michel, B.; Mathieu, H. J.; Schaumburg, K. Langmuir 2003, 19, 8749–8758. 5431

dx.doi.org/10.1021/la200535c |Langmuir 2011, 27, 5427–5432

Langmuir

ARTICLE

(8) Renault, J. P.; Bernard, A.; Juncker, D.; Michel, B.; Bosshard, H. R.; Delamarche, E. Angew. Chem., Int. Ed. 2002, 41, 2320–2323. (9) Xu, H. P.; Huskens, J. Chem. Eur. J. 2010, 16, 2342–2348. (10) Liu, Z. H.; Yi, Y.; Xu, H. P.; Zhang, X.; Ngo, T. H.; Smet, M. Adv. Mater. 2010, 22, 2689–2693. (11) Branch, D. W.; Wheeler, B. C.; Brewer, G. J.; Leckband, D. E. Biomaterials 2001, 22, 1035–1047. (12) Mendelsohn, A. D.; Bernards, D. A.; Lowe, R. D.; Desai, T. A. Langmuir 2010, 26, 9943–9949. (13) Engin, S.; Trouillet, V.; Franz, C. M.; Welle, A.; Bruns, M.; Wedlich, D. Langmuir 2010, 26, 6097–6101. (14) Embrechts, A.; Feng, C. L.; Mills, C. A.; Lee, M.; Bredebusch, I.; Schnekenburger, J.; Domschke, W.; Vancso, G. J.; Schonherr, H. Langmuir 2008, 24, 8841–8849. (15) Rozkiewicz, D. I.; Kraan, Y.; Werten, M. W. T.; de Wolf, F. A.; Subramaniam, V.; Ravoo, B. J.; Reinhoudt, D. N. Chem. Eur. J. 2006, 12, 6290–6297. (16) Hao, Z. X.; Chen, H. W.; Zhu, X. Y.; Li, J. M.; Liu, C. J. Chromatogr. A 2008, 1209, 246–252. (17) Xue, C. Y.; Chin, S. Y.; Khan, S. A.; Yang, K. L. Langmuir 2010, 26, 3739–3743. (18) Olah, A.; Hillborg, H.; Vancso, G. J. Appl. Surf. Sci. 2005, 239, 410–423. (19) Schnyder, B.; Lippert, T.; Kotz, R.; Wokaun, A.; Graubner, V. M.; Nuyken, O. Surf. Sci. 2003, 532, 1067–1071. (20) Hu, S. W.; Ren, X. Q.; Bachman, M.; Sims, C. E.; Li, G. P.; Allbritton, N. Anal. Chem. 2002, 74, 4117–4123. (21) Efimenko, K.; Wallace, W. E.; Genzer, J. J. Colloid Interface Sci. 2002, 254, 306–315. (22) Xue, C. Y.; Yang, K. L. J. Colloid Interface Sci. 2010, 344, 48–53. (23) Chen, C.-H.; Yang, K.-L. Analyst 2011, 136, 733–739. (24) Graubner, V. M.; Jordan, R.; Nuyken, O.; Schnyder, B.; Lippert, T.; Kotz, R.; Wokaun, A. Macromolecules 2004, 37, 5936–5943. (25) Song, J.; Tranchida, D.; Vancso, G. J. Macromolecules 2008, 41, 6757–6762. (26) Bernard, A.; Delamarche, E.; Schmid, H.; Michel, B.; Bosshard, H. R.; Biebuyck, H. Langmuir 1998, 14, 2225–2229. (27) Tan, J. L.; Tien, J.; Chen, C. S. Langmuir 2002, 18, 519–523. (28) Xue, C. Y.; Yang, K. L. Langmuir 2008, 24, 563–567. (29) Xue, C. Y.; Khan, S. A.; Yang, K. L. Adv. Mater. 2009, 21, 198–202.

5432

dx.doi.org/10.1021/la200535c |Langmuir 2011, 27, 5427–5432