Supported Lipid Bilayer Composition Microarray Fabricated by Pattern

The microscope was also equipped with a UV lamp light source and a digital still camera (Nikon Coolpix 4500) to record the fluorescence microscope ima...
0 downloads 0 Views 3MB Size
Download PDF
LETTER pubs.acs.org/Langmuir

Supported Lipid Bilayer Composition Microarray Fabricated by Pattern-Guided Self-Spreading Kazuaki Furukawa* and Takashi Aiba NTT Basic Research Laboratories, NTT Corporation, 3-1 Morinosato Wakamiya, Atsugi, Kanagawa, Japan 243-0198

bS Supporting Information ABSTRACT: We report on the fabrication of a microarray of supported lipid bilayers (SLBs) with different chemical compositions and demonstrate its biosensing application. The technique utilizes the phenomenon of lipid self-spreading on a patterned surface, which offers complete positional selectivity for a supported lipid bilayer. We describe the fabrication of parallel 10μm-wide lines, each filled with an SLB with a unique composition, at 5 μm intervals. Structures obtained with our new technique are finer and more highly integrated than previously reported structures that employ the vesicle fusion technique on patterned surfaces. We also detected specific binding between biotin and streptavidin with high contrast, indicating that the microarray is valuable for biosensing applications.

’ INTRODUCTION Microarrays of biological molecules have become indispensable to modern high-throughput bioanalysis. The most successful example is a DNA chip equipped with 104 to 105 different DNA fragments fixed on a solid surface.1 3 The use of such microarrays is expanding to other biological molecules such as proteins,4 7 oligosaccharides,8,9 antigens, and antibodies.10,11 Of these, the protein chip has attracted a lot of attention as an important tool for studying membrane-associated proteins and cellular processes.12 However, the fixation of proteins on a solid surface poses delicate problems compared with that of DNA. This is due to the unavoidable and unfavorable transformation of proteins by, for instance, their modification for fixation and/or steric hindrance between the proteins and the surface. A microarray of supported lipid bilayers (SLBs) is expected to constitute a new platform on which to build protein microarrays.13 15 A number of proteins, especially membrane-associated proteins, function within or on cell membranes that have a fluid characteristic known as lateral diffusion. The SLB maintains the fluidity indispensable for the protein function.16 SLB microarrays have been fabricated on surfaces with a corral pattern using a vesicle fusion technique. The conventional vesicle fusion technique yields an SLB microarray where each array has an identical chemical composition17,18 or gradation array with two different compositions using a vesicle solution flow.19 However, the preferred microarray is one whose composition is completely different from that of neighboring corrals. This has been achieved by addressing each corral using a microcapillary tube containing a different liposome solution.20 The method is also important in that it showed the potential for realizing a significant increase in array integration. At the same time, it includes the difficulty intrinsic to the solution process, namely, that the integration r 2011 American Chemical Society

scale is limited by the droplet size. Recently, new techniques have been developed using the microfluidic system to fabricate largescale SLB composition microarrays for biomolecule sensing.21 However, it still seems difficult to overcome the size limit of the solution process. This letter describes a new technique for fabricating SLBcomposition microarrays that overcome the size limit imposed by the vesicle fusion-based technique. Our method utilizes the lipid self-spreading phenomenon,22,23 which is a self-organization process whereby lipid molecules form a lipid bilayer at a solid liquid interface, guided by a lithographic pattern on a solid surface. This is in accordance with our knowledge that the position and direction of self-spreading can be controlled with the surface pattern.24,25 The pattern design is versatile: SLB selfspreads along patterns that include curves 26 and nanoscale gaps27,28 and fills every corner of the patterns. It is possible to fabricate submicrometer-wide line patterns with our method, which has the potential to increase the integration density by 102 compared to previously reported methods. We also demonstrate a biosensing application of the microarray using biotin streptavidin binding.

’ EXPERIMENTAL SECTION L-R-Phosphatidylcholine (L-R-PC) extracted from egg yolk was purchased from Sigma-Aldrich. 1,2-Dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-benzoxadiazol-4-yl) (NBD-DHPE) and 1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-(cap-biotinyl) (biotinyl-DOPE) were purchased from Avanti Polar Lipid

Received: March 3, 2011 Revised: April 22, 2011 Published: May 23, 2011 7341

dx.doi.org/10.1021/la200808g | Langmuir 2011, 27, 7341–7344

Langmuir

Figure 1. Schematic illustration of a typical pattern design. The white and black areas are SiO2 and Au surfaces, respectively. Inc. Texas Red 1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine (TR-DHPE) and N-(6-chloro-7-hydroxycoumarin-3-carbonyl)-dimyristoylphosphatidyl-ethanolamine (CC2-DMPE) were purchased from Invitrogen. Texas Red-conjugated streptavidin (TR-SAv) was purchased from Vector Laboratories, Inc. Deionized water (Millipore, >18 MV) was used throughout the work. A micropattern was fabricated on a SiO2 wafer (Iiyama Precision Glass Co. Ltd.) by a conventional photolithographic technique using a TSMR-V90 (Tokyo Ohka Kogyo) photoresist. A metal pattern (approximately 5 nm Ti and 45 nm Au) was fabricated with a conventional lift-off process. The wafer was cut into 5  15 mm2 chips, which were washed consecutively with water, a H2O2/H2SO4 (1:4) mixture, water, NH4F, and water (5 min each). The chips were then fixed on slides prior to self-spreading experiments. A mixture of L-R-PC (99 mol %) and dye-conjugated lipid (1 mol %) was prepared according to a previously reported method. A chloroform solution of dye-conjugated lipid was mixed with L-R-PC to prepare solutions of L-R-PC containing 1 5 mol % of the dye-conjugated lipid. The chloroform was evaporated in a stream of nitrogen gas and dried in a vacuum overnight to yield sticky solids of lipid mixtures. A small amount of this solid was transferred using the tip of a glass capillary in contact with the surface inside the manually accessible rectangular area. A selfspreading lipid bilayer membrane was developed on the hydrophilic surface by immersing the device gently in phosphate-buffered saline (PBS) (0.1 M phosphate, 0.15 M NaCl, pH 6.9 7.2), which was prepared by dissolving a phosphate-buffered saline pack (Thermo Scientific) in deionized water. An Olympus BX51-FV300 equipped with 405, 488, and 543 nm laser light sources for excitation was used to obtain confocal laser scanning microscope images. We used a 430 460 nm filter with a 405 nm light source for the CC2 observations, a 505 525 nm filter with a 488 nm light source for the NBD observations, and a 610 nm high-pass filter with a 543 nm light source for the Texas Red observations. We chose as weak an excitation laser power as possible (0.1 0.3% of the full power of our system) to avoid unfavorable degradation of the dye fluorescence. We set the photomultiplier voltage and gain within ranges that avoided saturating the output signals. The space between the chip and the objective lens was filled with buffer solution throughout the observations, for which a Plan Apo 40  WLSM water-immersion objective lens (Olympus) was used unless otherwise mentioned. The microscope was also equipped with a UV lamp light source and a digital still camera (Nikon Coolpix 4500) to record the fluorescence microscope images. All observations were carried out at room temperature.

’ RESULTS AND DISCUSSION Figure 1 shows a typical designed pattern. Three rectangular macroscopic areas with approximate dimensions of 0.25 mm  0.5 mm are designed to allow the lipid source to be spotted manually. Each rectangular area is connected to three 10-μm-wide

LETTER

lines that form a microarray area of nine 10-μm-wide lines, each separated by a 5 μm space at the other end. A small amount of a lipid mixture is spotted on the inside surface of the manually accessible area under atmospheric conditions. The sample is then gently immersed in buffer solution to start the formation of an SLB by self-spreading. Because self-spreading occurs only on the SiO2 surface and not on the Au surface, SLB formation is perfectly guided by the surface pattern and the SLBs in different lines never mix with each other. Self-spreading stops if the SLBs reach the ends of the lines, and no further SLBs are formed. The chemical compositions of the SLBs are determined by the source mixture and basically maintain the source composition. The merit of our method is that SLBs with many different compositions can be formed in a common buffer solution without mixing with each other. This is a significant feature of our method compared with the vesicle fusion method, which uses different solutions for different SLB compositions.20 We first fabricated an SLB microarray of L-R-PC containing three different dye-conjugated lipid molecules: CC2-, NBD-, and Texas Red (TR)-conjugated lipids. Figure 2a shows a fluorescence microscope image of the initial sample under atmospheric conditions, showing that the lipid source spots yield fluorescence corresponding to the dyes: blue fluorescence from CC2, green from NBD, and red from TR. The sample was then immersed in PBS buffer and observed with a fluorescence microscope (Figure 2b d). The SLBs filled the macroscopic areas (Figure 2b) and spread further, guided by the 10-μm-wide line pattern (Figure 2c) until the SLB edges reached the ends of the lines. The SLB pattern consisting of parallel 10-μm-wide lines, each filled with an SLB with a unique composition, at 5 μm intervals was thus formed successfully as shown in Figure 2d. This is an advantage of our method because it might be difficult to fabricate SLB arrays with a 5 μm interval by using a solution process such as vesicle fusion. We believe that our method is advantageous to the further integration of array density because self-spreading occurs under identical solution conditions. In addition, the method is versatile in microarray design. We have demonstrated the preparation of several other composition microarrays (Figure S1 in the Supporting Information). The fabricated microarray always has the excess amount of lipid source left, which possibly disperses into the buffer solution to induce a vesicle-fusion-like process on the self-spread SLBs and forms a mixture of compositions. For microarray applications, it is necessary to prevent this process. We found that washing using deionized water is an effective way to remove the excess source and structures other than the SLB while retaining the SLB, the first layer directly supported on the solid surface (Figure S2 in the Supporting Information).29 The lateral diffusion characteristic remains after this washing process as confirmed by the fluorescence recovery after photobleaching technique. To demonstrate the validity of the microarray for biosensing applications, we fabricated a composition microarray, part of which contained biotin-conjugated lipid, and tested streptavidin detection using the microarray. Figure 3a shows a confocal laser scanning microscope image of the composition microarray consisting of three lines at the top (NBD-DHPE(1%)-L-R-PC(99%)), three lines in the middle (biotinyl-DOPE (1%)-L-R-PC(99%)), and three lines at the bottom (biotinyl-DOPE (1%)-NBD-DHPE(1%)-L-R-PC(98%)). The image in Figure 3a was taken 40 min after self-spreading started. The line patterns were filled by SLBs. Green fluorescence was observed from the top and bottom three lines because they contained NBD. The middle 7342

dx.doi.org/10.1021/la200808g |Langmuir 2011, 27, 7341–7344

Langmuir

LETTER

Figure 2. Fluorescence microscope image of a composition microarray of SLBs containing blue, green, and red fluorescent dyes taken with a digital still camera under UV excitation: (a) before self-spreading and taken using a UPlan Apo 10 objective lens under atmospheric conditions and (b d) during self-spreading in buffer solution.

Figure 3. Confocal laser scanning microscope images of the SLB microarray taken (a) before and (b) 90 min after TR-SAv addition. Top three lines: LR-PC with 1% NBD-DHPE. Middle three lines: L-R-PC with 1% biotinyl-DOPE. Bottom three lines: L-R-PC with 1% NBD-DHPE and 1% biotinylDOPE. The green and red colors are from NBD (505 525 nm fluorescence, under 488 nm excitation) and TR (over 610 nm fluorescence, under 543 nm excitation), respectively. Scale bar: 50 μm.

three lines did not contain dyes and were thus nonfluorescent. Therefore, there is no direct proof of SLB generation on these lines at this moment. To this sample we added TR-SAv and recorded time-lapse fluorescent images. The green and red fluorescent images together with their superimposition 90 min after the addition of TR-SAv are shown in Figure 3b. Green fluorescence is observed on the top three lines and the bottom three lines, corresponding to the initial green fluorescence. This indicates the stability of SLBs containing NBD-conjugated lipid. They did not peel off or break during the TR-SAv reaction process. Red fluorescence is observed in the middle and bottom three lines, corresponding to the SLBs containing biotin-conjugated lipids. This results from the specific binding between biotin and streptavidin. The red fluorescence from the middle three lines proves that an SLB containing biotinconjugated lipid was prepared by self-spreading, although no fluorescence was observed before the addition of TR-SAv. It should also be mentioned that little red fluorescence is observed from the top three lines or the Au surface, indicating that the nonspecific adsorption is limited, despite the fact that TR-SAv is present in the solution during the observations. Because the prevention of nonspecific adsorption is often a key topic in relation to biosensing devices, the high contrast obtained in Figure 3 is another advantage of our microarray. The average fluorescence intensity of the three areas is plotted in Figure 4 against time. Times t = 0 and 5400 correspond to Figure 3a,b, respectively. The fluorescence intensity from NBD is gradually decreased, which is most likely due to the dye degradation caused by repeated observations. The red fluorescence began to be observed at t exceeding 1000, and it reached a constant value within about 2000 s. The initial delay of the appearance of red fluorescence is due to the configuration of our apparatus. Because the working distance of the water-immersion

Figure 4. Time lapse observations of average fluorescence intensity. The green and red plots indicate green and red fluorescence, respectively. Time t is set at 0 when TR-SAv is added to the buffer solution.

lens is short at about 0.3 mm, it takes some time for added TRSAv to diffuse into the space between the lens and the microarray surface (Figure S3 in the Supporting Information). We also confirmed that the red fluorescence level from SLBs containing biotinyl-DOPE did not change greatly after we washed the microarray in deionized water, which shows that the desorption of TR-SAv-bound biotin does not readily occur. Thus, the saturation of red fluorescence is determined by the amount of TRSAv that can be bound to the prepared biotin surface. However, red fluorescence with a weak intensity is detected from the top three lines that do not contain biotinyl-DOPE. This could be attributed to the slight nonspecific absorption or TR-SAv dissolved in the buffer solution. As one of the indexes for evaluating the biosensing ability of our microarray, we employ the ratio of red fluorescence from the SLBs containing biotin to that from the SLB without biotin. A ratio of more than 3, which we believe to 7343

dx.doi.org/10.1021/la200808g |Langmuir 2011, 27, 7341–7344

Langmuir be sufficient for detecting molecules using fluorescence, was achieved after t = 3600 s, as seen in Figure 4. We have not investigated the causes of the nonspecific binding in further detail. However, regardless of the causes of the background, the contrast is at a good level for biosensing applications.

’ CONCLUSIONS In this letter, we described the development of a new method for fabricating an SLB composition microarray and demonstrated the validity of the microarrays for biosensing devices. The method overcomes the integration limit of the previously reported method based on the solution process. We showed the successful fabrication of 10-μm-wide parallel lines at 5 μm intervals. This highly integrative SLB microarray fabrication technique is advantageous for use with high-throughput device architecture. Because SLB is known to self-spread through a sub-100-nm gap,27 we can extend our technique to submicrometer arrays by using an electron beam lithography pattern that has the potential to increase the integration density by 102 compared with previously reported methods. We think that the composition submicroarrays can provide opportunities for previously impractical experiments. ’ ASSOCIATED CONTENT

bS

Supporting Information. Additional experimental results. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

LETTER

(10) Lee, H. J.; Nedelkov, D.; Corn, R. M. Anal. Chem. 2006, 78, 6504–6510. (11) Albertorio, F.; Daniel, S.; Cremer, P. S. J. Am. Chem. Soc. 2006, 128, 7168–7169. (12) Wong, L. S.; Khan, F.; Micklefield, J. Chem. Rev. 2009, 109, 4025–4053. (13) Sackmann, E. Science 1996, 271, 43–48. (14) Cornell, B. A.; Braach-Maksvytis, V. L. B.; King, L. G.; Osman, P. D. J.; Raguse, B.; Wieczorek, L.; Pace, R. J. Nature 1997, 387, 580–583. (15) Tanaka, M.; Sackmann, E. Nature 2005, 437, 656–663. (16) Salafsky, J.; Groves, J. T.; Boxer, S. G. Biochemistry 1996, 35, 14773–14781. (17) Groves, J. T.; Ulman, N.; Boxer, S. G. Science 1997, 275, 651–653. (18) Groves, J. T.; Boxer, S. G. Acc. Chem. Res. 2002, 35, 149–157. (19) Kam, L.; Boxer, S. G. J. Am. Chem. Soc. 2000, 122, 12901– 12902. (20) Cremer, P. S.; Yang, T. J. Am. Chem. Soc. 1999, 121, 8130–8131. (21) Smith, K. A.; Gale, B. K.; Conboy, J. C. Anal. Chem. 2008, 80, 7980–7987. (22) R€adler, J.; Strey, H.; Sackmann, E. Langmuir 1995, 11, 4539– 4548. (23) Nissen, J.; Gritsch, S.; Wiegand, G.; R€adler, J. Eur. Phys. J. B 1999, 10, 335–344. (24) Furukawa, K.; Nakashima, H.; Kashimura, Y.; Torimitsu, K. Lab Chip 2006, 6, 1001–1006. (25) Furukawa, K.; Nakashima, H.; Kashimura, Y.; Torimitsu, K. Langmuir 2008, 24, 921–926. (26) Furukawa, K.; Kashimura, Y. In Biosurfaces and Biointerfaces; Garrido, J. A., Johnston, E., Werner, C., Boland, T., Eds.; Materials Research Socicty Symposium Proceedings; Materials Research Society: Warrendale, PA, 2010; Vol. 1236, p SS03-03. (27) Kashimura, Y.; Durao, J.; Furukawa, K.; Torimitsu, K. Jpn. J. Appl. Phys. 2008, 47, 3248–3252. (28) Kashimura, Y.; Furukawa, K.; Torimitsu, K. Jpn. J. Appl. Phys. 2010, 49, 04DL15. (29) Furukawa, K.; Sumitomo, K.; Nakashima, H.; Kashimura, Y.; Torimitsu, K. Langmuir 2007, 23, 367–371.

’ ACKNOWLEDGMENT This work was supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (20310076). ’ REFERENCES (1) Schena, M.; Shalon, D.; Davis, R. W.; Brown, P. O. Science 1995, 270, 467–470. (2) Lashkari, D. A.; DeRisi, J. L.; McCusker, J. H.; Namath, A. F.; Gentile, C.; Hwang, S. Y.; Brown, P. O.; Davis, R. W. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 13057–13062. (3) Sassolas, A.; Leca-Bouvier, B. D.; Blum, L. J. Chem. Rev. 2008, 108, 109–139. (4) 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, 2101–2105. (5) Fang, Y.; Frutos, A. G.; Lahiri, J. J. Am. Chem. Soc. 2002, 124, 2394–2395. (6) Wilson, D. S.; Nock, S. Angew. Chem., Int. Ed. 2003, 42, 494–500. (7) Bertone, P.; Snyder, M. FEBS J. 2005, 272, 5400–5411. (8) Feizi, T.; Chai, W. G. Nat. Rev. Mol. Cell. Biol. 2004, 5, 582–588. (9) Blixt, O.; Head, S.; Mondala, T.; Scanlan, C.; Huflejt, M. E.; Alvarez, R.; Bryan, M. C.; Fazio, F.; Calarese, D.; Stevens, J.; Razi, N.; Stevens, D. J.; Skehel, J. J.; van Die, I.; Burton, D. R.; Wilson, I. A.; Cummings, R.; Bovin, N.; Wong, C. H.; Paulson, J. C. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 17033–17038. 7344

dx.doi.org/10.1021/la200808g |Langmuir 2011, 27, 7341–7344