NANO LETTERS
Actomyosin-Driven Motility on Patterned Polyelectrolyte Mono- and Multilayers
2003 Vol. 3, No. 11 1505-1509
Jad A. Jaber,†,§ P. Bryant Chase,‡ and Joseph B. Schlenoff†,§,* Department of Chemistry and Biochemistry, Department of Biological Science and Program in Molecular Biophysics, and Center for Materials Research and Technology (MARTECH), The Florida State UniVersity, Tallahassee, Florida 32306 Received July 18, 2003; Revised Manuscript Received September 18, 2003
ABSTRACT Positive polyelectrolytes were investigated as new surface coatings for promoting in vitro actomyosin motility. Two surface arrangements were studied: a monolayer of the polyelectrolyte PAH, poly(allylamine hydrochloride), and multilayers consisting of 11−41 layers of alternating polypositive PAH/polynegative PSS (polystyrene sulfonate) electrolytes. For in vitro motility assays, rabbit skeletal muscle heavy meromyosin (HMM) was applied to the PAH surface of the polyelectrolyte mono/multilayer. Myosin-driven motion of actin filaments labeled with rhodamine− phalloidin was recorded at 30 °C using epifluorescence microscopy. Actin filaments were found to have a mean speed of 2.9 ± 0.08 µm/sec on the multilayer surface compared to 2.5 ± 0.06 µm/sec on the monolayer surface. Average filament length and speed increased when nonionic surfactant was added to HMM and ionic strength of the motility buffer increased, respectively. Microcontact printing with a waterinsoluble charged block copolymer on PAH produced patterned surfaces that restricted filament motion to PAH tracks.
Increasing demand for in situ characterization and quantification of samples in complex systems has stimulated the development of miniaturized chemical analysis systems1-4 that automatically perform multiple steps such as sampling, transport, separation, and detection. Crucial to these systems is the availability of nanomechanical devices that provide the necessary locomotive forces. Because production of nanoscale motors has proven challenging, a recent focus has been on adapting the highly efficient, naturally occurring nanoscale motor proteins5-7 kinesin and myosin, coupled with microtubules and actin filaments, respectively. In vitro interactions between actin and myosin, two major muscle proteins, powered by the hydrolysis of adenosine triphosphate (ATP), can produce movement and force in the same way they drive muscle contraction.8,10 The success of a device comprising the actomyosin system depends on the proper interfacing/immobilization of the proteins to synthetic nanomechanical components. Surfaces used should be biocompatible and patterned in a way that would allow ordered and controllable actomyosin, kinesin/microtubule11,12 interactions. Materials used to create these interfaces include nitrocellulose, photoresists13-15 such as methacrylate polymers; glass cleaned with KOH/ethanol and siliconized glass.16 Nanostructured surfaces having submicrometer grooves have been produced, using electron beam lithography and UV * Corresponding author. † Department of Chemistry and Biochemistry. ‡ Department of Biological Science and Program in Molecular Biophysics. § Center for Materials Research and Technology. 10.1021/nl034539h CCC: $25.00 Published on Web 10/18/2003
© 2003 American Chemical Society
photolithography, to restrict actomyosin motility to specified areas.13-15 Sufficiently narrow grooves constrain filament motion to a track and minimize the number of filaments that change direction. A simple alternative approach to the fabrication of patterned microstructures is the use of wet microcontact printing techniques.17-20 Here, we describe a novel approach to prepare actomyosin compatible surfaces, using polyelectrolyte multilayers (PEMUs). These surfaces are rugged, amorphous nanocomposites prepared by the layer-by-layer assembly method.21-22 They offer a wide range of compositional flexibility, permitting optimization of the surfacenanomotor interaction and are compatible with wet contact printing methods. A positively charged terminal layer allowed rabbit skeletal muscle HMM to bind and retain motor function. Microstructured channels for motility were created on these PEMUs by using poly(dimethyl siloxane) (PDMS) stamps and PEBSS [Poly(styrene sulfonate)-block-poly(ethylene-ran-butylene)-block-poly(styrene sulfonate)] a negatively charged hydrophobic polymer as the inking solution to produce barriers. Actin and Myosin. Fast skeletal muscle myosin is a wellcharacterized motor protein that is highly abundant in muscle tissue. Atomic resolution structures have been determined for the motor domain of myosin and also for actin, leading to a molecular model for the mechanism by which actomyosin converts chemical energy from MgATP hydrolysis into mechanical work.9 Myosin and actin were purified from rabbit back and leg muscles as previously described.33 HMM was prepared from myosin and affinity purified to remove
Figure 2. Actin filaments on a PAH terminated multilayer without (left panel) and with (right panel) addition of 0.6% Triton X-100 to HMM solution.
Figure 1. Structure of polyelectrolytes used.
“dead heads”.28,33 Myosin with a pI of 5.4 represents a good candidate for attachment on cationic surfaces. F-actin was labeled with rhodamine phalloidin (RhPh).28,33 Polyelectrolyte Surfaces. Electrostatic layer-by-layer assembly of the oppositely charged polymers (PAH and PSS) depicted in Figure 1 was employed for the preparation of uniform thin film coatings.22 Polyelectrolytes may be assembled on a variety of substrates such as glass coverslips for microscopy by alternating exposure to solutions of polycations and polyanions with an intermediate rinse in water or buffer. A very uniform ultrathin film (1-100 nm) is produced where the surface charge is dictated by the lastadded polyelectrolyte.16,21,23-26 PEMU Coating and Motility Assay. Microscope cover slips (18 × 18 × 0.15 mm) were cleaned in “piranha” (70% H2SO4(conc)/30% H2O2: caution, piranha is a strong oxidizer and should not be stored in closed containers). Poly(styrene sulfonic acid), PSS, (molecular weight 5 × 105) and poly(allylamine hydrochloride), PAH, (molecular weight 7 × 104) were used as the polyanion and polycation, respectively. Both solutions were prepared in imidazole buffer (0.025 M, pH)7.4). Polymer solution concentrations were 0.01 M (quoted with respect to the monomer repeat unit). A robotic platform (nanoStrata Inc.) exposed the cover glass alternately to the two polymer solutions for 5 min with three rinses of imidazole buffer between each lasting for 1 min. Rinse and polymer solution volumes were approximately 50 mL each. Surface compositions ranged from a monolayer of PAH to multilayers of up to 41 alternating layers of PAH and PSS with PAH always being the top layer. Prior to use, PEMUs were annealed in 1.5 M NaCl for 3 h to produce a smooth surface.27 Atomic force microscope (AFM) scans showed the RMS surface roughness of multilayers to decrease from 4.0 ( 0.3 to 1.0 ( 0.4 nm following this procedure. Cover slips were then placed in a flow cell and an in vitro actomyosin motility assay was performed, as previously described.28,33 1506
HMM prepared in imidazole buffer with a pH of 7.4 will have a net negative charge, thus its adsorption on the positively charged PAH surface is facilitated. Nonspecific protein binding sites were blocked with BSA after application of HMM. Assays were conducted with AB solution (25 mM KCl, 25 mM imidazole, 4 mM MgCl2, 1 mM EGTA, 1 mM DTT, pH 7.4). Motility buffer was AB plus 2 mM ATP, 0.3% methylcellulose, 16.7 mM glucose, 100 µg/mL glucose oxidase, 18 µg/mL catalase, and an additional 40 mM DTT plus modifications described below for specific experiments. Sliding RhPh-labeled actin filaments at 30 °C were recorded on videotape by epifluorescence microscopy. Movement of essentially all actin filaments occurred on both mono- and multilayers. Quantitative analysis of filament motion was difficult, however, because filaments were much shorter (average length 1.12 ( 0.12 µm, N ) 50) than typically observed with nitrocellulose-coated surfaces, suggesting that some feature of the surface was causing filaments to break up. The length of the actin filament is important not only because of detection limits, but also because it limits both the potential cargo that might be transported and the ability to control its movement. To minimize filament breakup, the nonionic surfactant Triton X-100, was added to HMM solution at a concentration of 0.6% when it was incubated with the PAH surface. The average sliding speed remained the same (2.7 ( 0.01 µm/s and 2.9 ( 0.08 µm/s without and with the addition of Triton X-100, respectively), but filament length was comparable to that obtained on a nitrocellulose surface with an average length of 7.5 ( 3.6 µm (Figure 2). Protein adsorption on multilayers was generally found to be favored on oppositely charged surfaces and suppressed, but never eliminated, on like-charged surfaces.39 Although binding of HMM to the different polyelectrolyte surfaces via electrostatic interaction seems a reasonable mechanism, there are many other types of interaction which may drive a protein to a surface. To illustrate the generality of HMM binding to surfaces, it was fluorescently labeled with tetramethylrhodamine-5-isothiocyanate. Fluorescence microscopy revealed that HMM adsorbs to the three surfaces investigated (Table 1), independent of charge. Binding of HMM, however, does not guarantee subsequent binding of actin, and bound actin may not exhibit motility. Binding and motility assays were performed with unlabeled HMM and (RhPh) labeled actin. As shown in Table 1, actin filaments Nano Lett., Vol. 3, No. 11, 2003
Table 1. Protein Binding Characteristics of Different Surfaces Used actin binding of binding of binding of actin motility surfaces actin filaments labeled HMM on HMMc PAH PSS PEBSS c
yesa yesb yesb
yes yes yes
yes yes no
yes no no
a Filaments irreversibly adhering. b Filaments attached from one end only. HMM on polyelectrolyte surface.
Figure 4. Effect of salt concentration of the buffer on speed. Triangles, squares, and diamonds represent the average speed of actin filaments on a PAH terminated PEMU (11 layers), nitrocellulose, and PAH monolayer, respectively.
Figure 3. Actin sliding speed (diamonds) and multilayer thickness (squares) versus number of polyelectrolyte layers.
were observed to bind HMM on the PAH and PSS but not the PEBSS surfaces. This implies that HMM bound to PEBSS is rendered inactive, perhaps due to the hydrophobic nature of PEBSS. When the motility buffer was added, actin sliding was noted only on the PAH surface. Sheared, completely adherent actin filaments were observed on the PSS surface. Hence, PEBSS can be used to produce actinrepelling features. Actin by itself binds to all three surfaces (Table 1) and, unexpectedly, attached from one end only to PSS and PEBSS. Monolayer vs Multilayer. The use of a multilayer permits a multicomposite, multifunction approach to integrating bionanomotors into a functional package. PEMUs may be designed with layers of active materials, such as enzymes, embedded in them and may have functions other than promoting protein adhesion, such as controlling permeability,29 electrical conductivity,30 sensors,31 nanoporosity,32 and swelling.27 The average speed of actin filaments was determined on the myosin functionalized PAH monolayer/ multilayer by computer analysis of digitized movies. Filament sliding speed was determined on PAH-terminated monolayers and PEMUs of up to 41 layers (Figure 3). We found an enhancement of speed for PEMUs versus monolayer, probably due to reduced interaction with the underlying glass. Filament sliding speed was faster on the multilayer surface, with a speed of 3.4 ( 0.03 µm s-1 (N ) 27) compared to 2.7 ( 0.1 µms-1 (N ) 22) on the monolayer surface. For reference, the sliding speed on a nitrocelluslose surface under our conditions was 3.3 ( 0.1µm s-1 (N ) 20). These measurements were done with 0.0145 M KCl in the motility buffer. Salt and Speed. Electrostatic interactions are important for formation of the actomyosin complex. Electrostatic interactions of these proteins with the underlying surface are also possible, although such nonspecific interactions should Nano Lett., Vol. 3, No. 11, 2003
have been blocked (by BSA and by sheared, unlabeled F-actin) during manufacture of the flow cell. Increasing the salt (KCl) concentration of the motility buffer weakens these interactions, as a result the actin filaments will slide more smoothly and their speed is enhanced.33,35 The motility buffer was prepared with a range of KCl concentrations. At low salt concentrations the speed of actin on the multilayer surface was comparable to that on a conventional nitrocellulose surface, whereas actin moved slightly more slowly on a PAH monolayer (Figure 4). As the salt concentration was increased, motion was faster on multilayer surfaces compared to nitrocellulose. However, at salt concentrations above 0.06 M, filaments dissociated from the surface and diffused into the motility buffer solution. Nicolau et al.13 reported a velocity of 4.03 µm s-1 on poly[(tert-butyl methacrylate)-co-(methyl methacrylate)] surface using 40 mM KCl in the motility buffer. At this salt concentration our multilayer surface allowed for a speed of 5.2 µm s-1. On the other hand Bunk et al.,15 using a MRL6000.1XP surface, reported a speed of 5.5 µm s-1 with salt concentrations of 75 mM in the motility buffer. Our data would extrapolate to a higher value, but at this concentration, our filaments dissociated from the multilayer surface. Surface Patterning. The fabrication of patterned surfaces using poly(dimethylsiloxane) elastomeric stamps was first used to transfer alkanethiols onto gold surfaces.17-19 Microcontact printing has recently been extended to create polymer patterned surfaces on multilayers20,34 (polymer-onpolymer stamping, or POPS). In POPS the surface of a stamp is inked with polymer and, after drying, the stamp is pressed on the top of the multilayer. A polyelectrolyte of opposite charge to the surface facilitates adhesion. PEBSS was used to create water insoluble walls to delineate channels with PAH on the bottom of the channel. The hydrophobic nature of the inking solution eliminated the need to oxidize the PDMS surface with O2 plasma to make it more wettable. The polymer (2.5 wt % in a mixture of ethanol, propanol, dichloroethane, and tetrahydrofuran) was applied to the surface of the stamp with a cotton swab. The stamp was then pressed for 7 s on the top of the multilayer. Figure 5 shows the well defined channels, with PEBSS walls > 100 nm high, that were obtained. Sliding actin filaments were 1507
used to limit motion of actin filaments (long and medium sizes) to one of two directions along the same axis. These novel surfaces for motility assays should be useful for constructing biomotor powered nanomechanical devices. Acknowledgment. The authors are grateful to Professor Seunghun Hong and Mr. Pradeep Manandhar for supplying the PDMS stamps, Mr. Tom Asbury for writing the computer program used to analyze actin filament speed, Dr. Justin R. Grubich and Nicolas Brunet for assistance with initial motility assays, and Shanedah Williams, Lori McFadden, and Alyson Barnes for providing myosin and actin. This work is sponsored by the Defense Advanced Research Projects Agency (DARPA) Defense Sciences Office (DSO) under the auspices of Dr. Alan Rudolph and Dr. Anantha Krishnan through the Space and Naval Warfare Systems Center (SPAWAR), San Diego contract No. N66001-02-C-8030. References
Figure 5. (A) and (C) are AFM scans and section analysis of the stamped multilayer surface showing a pattern with barriers of PEBSS 1.4 µm wide and 132 nm height. (B) Snapshot of fluorescent-labeled actin filaments sliding in channels (PAH) of 3.9 µm widths. The dotted arrows show the trajectories followed by two filaments. The dip in the center of the wall is due to the stamp features.
observed only within the channels; no actin filaments, moving or immobile, were observed on the PEBSS walls, suggesting that functional HMM bound only to PAH on the floor of the channels. Filaments were divided into three categories according to length. The “long” (longer than the channel width 9.3 ( 0.8 µm) and “medium” (approximately as long as the channel width 4.5 ( 0.5 µm) filaments remained inside the tracks with no change in lanes or direction (U-turns). Actin filaments rebounded efficiently off walls and remained inside the channels, consistent with measured persistence length of 18 µm for phalloidin stabilized F-actin.36-38 During an observation period of 26 s, a small proportion of filaments (11.5%, N ) 35) were observed to diffuse into solution, only when interacting with the walls. In the case of short filaments (shorter than the channel width 1.66 ( 0.15 µm), 40% of the filaments maintained the same direction (N ) 35). Conclusion. Polyelectrolyte surfaces utilizing poly(allylamine hydrochloride) and poly(styrene sulfonic acid) were proven to support actomyosin motility with little dependence on surface thickness. Motility on these surfaces was comparable to that obtained on nitrocellulose surfaces with appropriate modifications to the manufacturing process. The use of PEBSS to produce nanostructured channels on the multilayer surface was effective in restricting actomyosin interaction to the designated areas. PEBSS channels were 1508
(1) Culbertson, C. T.; Jacobson, S. C.; Ramsey, J. M. Anal. Chem. 2000, 72, 5814. (2) Harrison, D. J.; Manz, A.; Fan, Z. H.; Ludi, H.; Widmer, H. M. Anal. Chem. 1992, 64, 1926. (3) Ramsey, R. S.; Ramsey, J. M. Anal. Chem. 1997, 69, 1174. (4) Meng, Z. J.; Qi, S. Z.; Soper, S. A.; Limbach, P. A. Anal. Chem. 2001, 73, 1286. (5) Hess, H.; Vogel, V. ReV. Mol. Biotechnol. 2001, 82, 67. (6) Schmidt, J. J.; Xingqun, J.; Montemagno, C. D. Nano Lett. 2002, 2, 1229. (7) Soong, R. K.; Bachand, G. D.; Neves, H. P.; Olkhovets, A. G.; Craighead, H. G.; Montemagno, C. D. Science 2000, 290, 1555. (8) Bagshaw, C. R.; Muscle Contraction; Chapman and Hall: London, 1982. (9) Spudich, J. A. Nat. ReV. Mol. Cell Biol. 2001, 2, 387. (10) Toyoshima, Y. Y.; Toyoshima, C.; Spudich, J. A. Nature 1989, 341, 154. (11) Hess, H.; Clemmens, J.; Qin, D.; Howard, J.; Vogel, V. Nano Lett. 2001, 1, 235. (12) Limberis, L.; Magda, J. J.; Stewart, R. J. Nano Lett. 2001, 1, 277. (13) Nicolau, D. V.; Suzuki, H.; Mashiko, S.; Taguchi, T.; Yoshikawa, S. Biophys. J. 1999, 77, 1126. (14) Suziki, H.; Yamada, A.; Oiwa, K.; Nakayama, H.; Mashiko, S. Biophys. J. 1997, 72, 1997. (15) Bunk, R.; Klinth, J.; Montelius, L.; Nicholls, I.; Omling, P.; Ta˚gerud, S.; Månsson, A. Biochem. Biophys. Res. Commun. 2003, 301, 783. (16) Kishino, A.; Yanagida, T. Nature 1988, 334, 74. (17) Kumar, A.; Whitesides, G. M. Science 1994, 263, 60. (18) Kumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1994, 10, 1498. (19) Xia, Y.; Whitesides, G. M. J. Am. Chem. Soc. 1995, 117, 9576. (20) Jiang, X.; Zheng, H.; Gourdin, S.; Hammond, P. T. Langmuir 2002, 18, 2607. (21) Decher, G. Science 1997, 277, 1232. (22) Decher, G.; Schlenoff, J. B. Multilayer Thin Films: Sequential Assembly of Nanocomposite Materials; Wiley-VCH: New York, 2003. (23) Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153. (24) Decher, G.; Schmitt, J. Prog. Colloid Polym. Sci. 1992, 89, 160. (25) Farhat, T.; Yassin, G.; Dubas, S. T.; Schlenoff, J. B. Langmuir 1999, 15, 6621. (26) Schlenoff, J. B.; Farhat, T. Langmuir 2001, 17, 1184. (27) Dubas, S. T.; Schlenoff, J. B. Langmuir 2001, 17, 7725. (28) Kron, S. J.; Toyoshima, Y. Y.; Uyeda, T. Q. P.; Spudich, J. A. Methods Enzymol. 1991, 196, 399. (29) Levasalmi, J.; McCarthy, T. J. Macromolecules 1997, 30, 1752. (30) Cheung, J. H.; Fou, A. F.; Rubner, M. F. Thin Solid Films 1994, 244, 985. (31) Sun, Y.; Zhang, X.; Sun, C.; Wang, B.; Schen, J. Macromol. Chem. Phys. 1996, 197, 174. (32) Mendelsohn, J. D.; Barrett, C. J.; Chan, V. V.; Pal, A. J.; Mayes, A. M.; Rubner, M. F. Langmuir 2000, 16, 5017. Nano Lett., Vol. 3, No. 11, 2003
(33) Chase, P. B.; Chen, Y.; Kulin, K. L.; Daniel, T. L. Am. J. Physiol. Cell Physiol. 2000, 278, 1088. (34) Jiang, X.; Hammond, P. T. Langmuir 2000, 16, 8501. (35) Homsher, E.; Wang, F.; Sellers, J. R. Am. J. Physiol. 1992, 262, C723. (36) Gittes, F.; Mickey, B.; Nettleton, J.; Howard, J. J. Cell. Biol. 1993, 120, 934.
Nano Lett., Vol. 3, No. 11, 2003
(37) Isambert, H.; Venier, P.; Maggs, A. C.; Fattoum, A.; Kassab, R.; Pantaloni, D.; Carlier, M. F. J. Biol. Chem. 1995, 270, 11444. (38) Clemens, J.; Hess, H.; Howard, J.; Vogel, V. Langmuir 2003, 19, 1738. (39) Schlenoff, J. B.; Salloum, D. S.; Jaber, J. A.; Chase, P. B. Polym. Mater. Sci. Eng. 2003, 89, 75.
NL034539H
1509
