Effect of Mechanical Instability of Polymer Scaffolds on Cell Adhesion

Aug 6, 2013 - The adhesion of fibroblast on polymer bilayers composed of a glassy polystyrene (PS) prepared on top of a rubbery polyisoprene (PI) was ...
1 downloads 0 Views 299KB Size
Download PDF
Article pubs.acs.org/Langmuir

Effect of Mechanical Instability of Polymer Scaffolds on Cell Adhesion Shinichiro Shimomura,† Hisao Matsuno,*,† and Keiji Tanaka*,†,‡ †

Department of Applied Chemistry and ‡International Institute for Carbon-Neutral Energy Research (WPI-I2CNER), Kyushu University, Fukuoka 819-0395, Japan S Supporting Information *

ABSTRACT: The adhesion of fibroblast on polymer bilayers composed of a glassy polystyrene (PS) prepared on top of a rubbery polyisoprene (PI) was studied. Since the top PS layer is not build on a glassy, or firm, foundation, the system becomes mechanically unstable with decreasing thickness of the PS layer. When the PS film was thinner than 25 nm, the number of cells adhered to the surface decreased and the cells could not spread well. On a parallel experiment, the same cell adhesion behavior was observed on plasma-treated PS/PI bilayer films, where in this case, the surface was more hydrophilic than that of the intact films. In addition, the fluorescence microscopic observations revealed that the formation of F-actin filaments in fibroblasts attached to the thicker PS/PI bilayer films was greater than those using the thinner PS/PI bilayer films. On the other hand, the thickness dependence of the cell adhesion behavior was not observed for the PS monolayer films. Taking into account that the amount of adsorbed protein molecules evaluated by a quartz crystal microbalance method was independent of the PS layer thickness of the bilayer films, our results indicate that cells, unlike protein molecules, could sense a mechanical instability of the scaffold.

■

INTRODUCTION Cells typically attach to a polymer scaffold, often called an extracellular matrix (ECM), made from materials such as collagen, elastin, and proteoglycan, and/or interact with other cells in an animal tissue.1 The ECM plays an important role in providing support and anchorage for cells, segregating tissues from one another, and regulating intercellular communication. The elastic modulus of a typical ECM widely ranges from 0.1 to 100 kPa,2 and recent studies have revealed that these differences in the elastic modulus of ECM cause changes in cellular morphology and proliferation during cell growth process.3,4 In addition, these differing cell behaviors due to material stiffness could be observed not only for a natural ECM but also for artificial scaffolds made from synthetic polymers.5−11 For example, while mouse NIH3T3 fibroblasts which adhered to a stiffer polydimethylsiloxane surface with a modulus of approximately 800 kPa were well spread, their spreading was reduced for a softer surface with a modulus of approximately 200 kPa.7 Mouse myoblasts C2C12 behaved similarly during adhesion on pH-stimulus-responsive block copolymer hydrogels composed of poly(2-(diisopropylamino)ethyl methacrylate) and poly(2-(methacryloyloxy)ethyl phosphorylcholine) enabling a control of viscoelasticity.8 They exhibited pronounced stress fiber formation and elongation on increasing the hydrogel elastic modulus. Material stiffness also influences cellular differentiation. Naive mesenchymal stem cells (MSCs) were extremely sensitive to tissue level elastic scaffolds made from polyacrylamide gel coated with collagen.9 Such could be seen in a result where, while MSCs differentiated into neurocytes on softer matrices mimicking brain (0.1−1 kPa), they differentiated into myocytes on stiffer matrices mimicking muscle (8−17 kPa). In the case of © XXXX American Chemical Society

more rigid matrices mimicking collagenous bone (25−40 kPa), they differentiated into osteocytes. Also, the stiffness of polymer scaffolds affects cell migration. Cell movement toward a harder region of culture substrate surface from a softer one, known as mechanotaxis, could be observed in many types of cells.10,11 These results make it clear that cells have some mechano-sensing systems. For applications in biotechnology and biomedical fields, they are extremely important because synthetic polymer scaffolds have the potential to regulate cellular functions without the need for any liquid chemical factors such as external genes and cytokines. Actually, it has been reported that substrate-induced modulation of synchronized beating in cultured cardiomyocyte tissue on elasticitytunable substrates promoted myocardial conduction when the stiffness of the cell culture environment matched that of the cardiac cells.12 Despite a growing literature on the effect of bulk stiffness of polymer scaffolds on cell-surface interactions, less is known about the effect of mechanical properties in the surface region of polymer materials on cell adhesion. In general, molecular motion at polymer surfaces is much more enhanced than that in the internal bulk region.13 Such an enhanced mobility, which is consequently related to the mechanical properties, is a strong function of the depth from the outermost surface down to approximately 10 nm. Also, the mechanical properties of polymer materials composed of multi components depend on the configuration. For example, the surface storage modulus (E′) of a bilayer film composed of a glassy polystyrene (PS) coated on a rubbery polyisoprene (PI) decreased with Received: October 9, 2012

A

dx.doi.org/10.1021/la4027706 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

decreasing thickness of the PS layer.14 This is because, as the PS layer thickness decreases, the properties of the underlying PI layer start to be manifested through the top PS layer and eventually control the surface properties of the bilayer. We here describe such a polymer bilayer that is not build on a firm foundation, as a system exhibiting a mechanical instability. Using this concept, once the top PS layer becomes thinner than a threshold value, it can be said that the PS/PI bilayer system becomes mechanically unstable. In this study, we investigate the effect of mechanical instability for PS films on cell adhesion behaviors; PS is one of the general-purpose polymers commonly used for artificial culture substrates in biomedical fields. We here demonstrate that the cell adhesion can be affected by the instability derived from the internal region, or under layer, to which the cells were not directly attached.

■

second spin-casting process. Any residual toluene was removed prior to the second spin-casting process to ensure that the PS and PI will not swell or mix each other in the films. The obtained PS/PI films were dried under vacuum for 24 h to remove MEK. In this case, PS with a larger Mn of 235k was used to keep the outermost surface of PS films glassy. For the PS/PI films, AFM observation and contact angle measurement were made in the same manner described above. Some of the PS and PS/PI films were treated by plasma at room temperature for 2 s using a Harrick Plasma basic plasma cleaner (PDC-32G) because the surface of the intact PS is somewhat hydrophobic for cellular scaffold. Cell Adhesion Assay. The PS monolayer and the PS/PI bilayer films, which were thoroughly dried under vacuum to remove any residual organic solvent, were washed with a phosphate buffered saline (PBS) solution three times, and were subsequently placed on the base of 24-well culture dishes. The polymer scaffolds were immersed in cell culture media with serum-free normal Roswell Park Memorial Institute (RPMI) 1640 (Life Technologies, Corp.).16 Suspensions of mouse fibroblast L929 (RIKEN BRC Cell Bank) with 5.0 × 104 cells/well were seeded on to the scaffolds. The cultures were maintained at 310 K (37 °C) in a humidified atmosphere containing 5% CO2. The initial cell adhesion state and morphology after 4 h culturing were evaluated by phase-contrast microscopic observations. Observation of F-Actin Filament in Fibroblast. Fluorescence microscopy was used to characterize the morphology and formation of F-actin filament of cells seeded on the scaffolds. Cells were immobilized on the scaffolds prior to these observations. After removal of the culture medium, the samples were fixed with 4% paraformaldehyde for 30 min and washed with PBS at 310 K three times. In staining the actin filament formed in fibroblast, Alexa Fluor594 phalloidin was used. All cell nuclei were stained with 4′,6diamidino-2-phenylindole (DAPI). The excitation wavelengths were set to 581 and 358 nm for Alexa Fluor594 phalloidin and DAPI, respectively, and the fluorescence observations were performed at 609 and 491 nm, respectively. Protein Adsorption. The amounts of proteins adsorbed on the PS films were evaluated by a quartz crystal microbalance (QCM) method. Scaffold films were prepared by a spin-coating method on Au-coated quartz substrates with a fundamental frequency of 5 MHz. Here, typical serum proteins such as human serum albumin (HSA) and human immunoglobulin G (IgG) were chosen to evaluate protein adsorption on the scaffold surfaces. The concentration of protein was 0.1 mg/mL in both PBS solutions. Frequency shifts at the third overtone of the QCM substrates due to the mass increase on the polymer surfaces were recorded.

EXPERIMENTAL SECTION

Materials. Monodisperse PSs were purchased from Polymer Source, Inc. Monodisperse PI was synthesized by a living anionic polymerization using n-butyllithium as an initiator in toluene under an argon atmosphere. After the polymerization, the polymer was purified by reprecipitation in methanol, and then, the precipitate was dried under vacuum for 24 h at 323 K. Characterization of the PI was performed by proton nuclear magnetic resonance (1H NMR) spectroscopy and gel permeation chromatography (GPC). Table 1

Table 1. Characterization of PS and PI Used in This Study Polymer

Mn

Mw/Mn

surface state at 310 K

PS PS PS PS PS PI

6k 20.8k 56.5k 75k 235k 394k

1.06 1.07 1.07 1.05 1.05 1.04

rubbery rubbery glassy glassy glassy rubbery

summarizes the number-average molecular weight (Mn) and the molecular weight dispersity, Mw/Mn, where Mw is the weight-average molecular weight of samples determined by GPC using PS standards. Preparation and Characterization of Scaffold Films. Films of PS monolayer with various Mn’s were prepared on borosilicate cover glasses from toluene solutions by a spin-coating method. These were dried under vacuum for 24 h at room temperature. The thickness of the PS films (dPS) was adjusted to be approximately 200 nm. In our previous report, we revealed that the outermost surfaces in the films of PSs having smaller and larger Mn were in rubbery and glassy states, respectively, at room temperature, as shown in Table 1.15 Surface characterization of the scaffolds was performed by a conventional contact angle measurement using a Drop Master 500 (Kyowa Interface Science Co. Ltd.). Water droplet was used as a probe. The surface morphology of each film was examined by atomic force microscopy (AFM, Agilent 5500, Agilent Technologies, Inc.) using MAC mode at room temperature. The magnetically coated cantilever with a nominal spring constant of 2.8 N m−1 was used. The driven frequency used was 74.5 kHz, which was on the low-frequency side of the resonance. To avoid the possible deformation of the sample surface during the observation, the ratio of the set point value to the free amplitude of the cantilever was maintained at approximately 0.95 (5% damping of the amplitude of oscillation). In addition, PS/PI bilayer films were prepared in the same manner described in our previous report.14 At first, PI films with a thickness of 200 nm were spin-coated onto cover glasses from a toluene solution. The obtained PI films were dried under vacuum for 24 h to remove toluene. After that, PS films with various dPS’s ranging from 17 to 200 nm were spin-coated on the PI films from methyl ethyl ketone (MEK) solutions. It is noted that MEK is a nonsolvent for PI, thus, the PI films do not dissolve during the

■

RESULTS AND DISCUSSION Cell Adhesion for the PS Monolayer Films. First, to clarify the effect of surface modulus on cell adhesion, the number of adhering cells and their morphology on the PS scaffolds, which were spin-coated on glass substrates, were evaluated by microscopic observations. The thickness of the PS scaffolds were approximately 200 nm and sufficiently enough to avoid any thinning effects on the surface properties of the scaffolds.14 The root-mean-square surface roughness (RMS) for all the films was less than 0.20 nm, and any defects did not exist as shown in Figure S1 in the Supporting Information (SI), that is, the film surfaces were suitable as scaffolds for characterization of cell adhesion behavior. Our previous scanning viscoelasticity microscopic (SVM) measurement revealed the surface E′ of PS films.15 The surface E′ was dependent on Mn of PS. For example, in the case of polymers with Mn > ca. 40k, the surface was in a glassy state with an E′ of a few GPa at 293 K. On the other hand, in the case of a smaller Mn polymer, the E′ at the surface, being in a glass−rubber transition or a rubbery state, was less than 1 GPa. Figure 1a,b shows the phase contrast images of L929 fibroblasts adhered on the PS monolayer films B

dx.doi.org/10.1021/la4027706 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 1. Phase contrast images of L929 fibroblasts adhered on the PS monolayer films with Mn of (a) 6k and (b) 235k, respectively. Scale bars correspond to 50 μm. (c) Mn dependence of the number of cells adhered on the PS monolayer films. Error bars represent the standard deviations of the measurements.

Figure 2. Phase contrast images of L929 fibroblasts adhered on the PS/PI bilayer films with (a) 25 nm- and (b) 200 nm-dPS, respectively. Scale bars correspond to 50 μm. (c) Closed and open circles show the number of cells (N) on the PS/PI bilayer and PS monolayer films, respectively. Error bars represent the standard deviations of the measurements.

with Mn = 6k and 235k, respectively, after 4 h culturing in serum-free RPMI1640 medium. Although the surface of a PS film with Mn = 6k and 235k are in rubbery and glassy states at 310 K (37 °C),13 respectively, no morphological difference of cells adhered to the two films was discerned. Also, the number of adherent cells on both films was almost the same, as shown in Figure 1c. Thus, it can be claimed that the cell adhesion behavior on the PS films was insensitive to the Mn of PS. That is, the modulus at the outermost surface in these films supported on the rigid foundation of glasses may not be particularly important to the cell adhesion, provided that the surface E′ was on the order of a few hundred megapascals. Cell Adhesion for the PS/PI Bilayer Films. Second, PS/ PI (glassy/rubbery) bilayer films14 were used as scaffolds in order to gain access to the effects of mechanical instability originating from the underneath PI region on cell responses. Fixing the thickness of the underneath PI layer at 200 nm, which can be regarded as a bulk phase with the elastic modulus of a few MPa,17 the dPS with Mn of 235k was changed from 200 nm down to 17 nm. In general, the glass−rubber transition temperature (Tg) of thin PS films decreases with decreasing thickness once the films go below 50 nm.18 Although this is the case even for our system, the Tg of the upper PS layer should be much higher than room temperature.19 In other words, the upper PS layer in all the bilayer films was in a glassy state regardless of the dPS. Also, it was preconfirmed that the upper PS layer was completely homogeneous with RMS less than 0.20 nm, as shown in Figure S2. Thus, it would be safe to assume that the bilayer film surfaces were also sufficiently flat to the same level as PS monolayer films. It was also confirmed that any dewetting or phase separation of the PS films on the PI films did not occur. And also, we confirmed that contact angle values for PS/PI bilayer and PS monolayer films were almost the same (data shown in Figure S3a), that is, surface freeenergies were almost the same for the all scaffolds used here. Figure 2a,b shows the phase contrast images of L929 cells on PS/PI bilayers. The number of cells on these films became smaller in the dPS range, smaller than approximately 25 nm, as shown in Figure 2c. The surface E′ evaluated by SVM for PS/PI

bilayer films decreased with a decreasing thickness of the glassy PS layer,14 indicating that the surface became mechanically unstable with a decreasing PS layer thickness. The origin of the decrease in surface E′ with decreasing thickness here observed is based on not the softening of the upper PS layer but the manifestation of the mechanical properties of the bottom PI layer.20 Interestingly, the dPS dependence of cell adhesion on PS/PI bilayer films observed here appears to be coincident with that of the surface E′. Although the moduli of the PS/PI bilayer films were of the order of GPa, being higher than that of native ECM, it seems most likely that cells could sense a mechanical instability originating in the underlying PI layer at depth region of approximately 25 nm, even though cells could not directly contact this region. The cell adhesion test for a longer time was also made. The dPS-dependence of cell number after 24 h culturing was almost the same as that after 4 h culturing. This is shown in Figure S4. The spreading of the cells that adhered on the PS monolayer and the PS/PI bilayer films with different dPS was also evaluated. Figure 3 shows the abundance of projected area of L929 cells adhered on PS monolayer and PS/PI bilayer films with different dPS, respectively. The peak position and full width at half-maximum (fwhm) obtained from the normal distribution curves in Figure 3 are summarized in Table S1. Generally, when fibroblast cells like the mouse L929 used here are cultured under appropriate conditions, they attached to the surfaces and then spread. As shown in Figure 3a, cell spreading behaviors on the PS monolayer films were almost independent of dPS. On the other hand, with a decreasing dPS of the bilayer films, the abundance of the spread cells decreased, as shown in Figure 3b. These results again imply that the fibroblast cells could sense a change in the mechanical stability of the polymer surface. That is, if the upper stiff layer of scaffold is sufficiently thick, a cell could still attach and spread on the scaffold surface, despite it having a mechanically unstable layer lying underneath. Cytoskeleton Formation of Cells Adhered on the Bilayer Scaffolds. The generation of tension by actin stress C

dx.doi.org/10.1021/la4027706 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 5. Frequency shifts with standard deviations of the QCM due to protein adsorption on the PS/PI bilayer films with (a) 25 nm- and (b) 200 nm-dPS, respectively. Concentrations of HSA or IgG were 0.1 mg/mL in phosphate buffer solutions. Error bars represent the standard deviations of the measurements.

the same. This means that, unlike the behavior of cells, the interaction between the protein molecules and the PS/PI bilayer surfaces did not depend on the mechanical stability of the deeper region of the scaffolds. Generally, cells adhere to a surface via the preadsorption of proteins. However, the protein adsorption observed here was not affected by mechanical instability of the scaffolds. Thus, it is plausible that cells themselves could directly sense the mechanical instability of the scaffolds. Cell Adhesion for the Hydrophilic Scaffolds. The intact PS surface is relatively hydrophobic in comparison with commercially available scaffolds. Since the hydrophilicity, or hydrophobicity, should be one of the controlling factors for the cell adhesion, the scaffolds used were treated by plasma, leading to the hydrophilic surface. The water contact angle on the plasma-treated PS/PI bilayer and PS monolayer films was insensitive to the thickness of the PS layer. This trend was common for with and without the plasma treatment. Figure 6a,b shows the phase contrast images of L929 cells on the plasma-treated PS/PI bilayers with different thicknesses of PS layer. In the case of the plasma-treated films, the number of cells adhered both on monolayers and bilayers increased

Figure 3. Abundance of projected area of L929 fibroblasts adhered on (a) PS monolayer and (b) PS/PI bilayer films with different dPS, respectively. The normal distribution curves are superimposed. The number of cells for calculation of the abundance is three hundred for each sample.

fibers is one of the driving forces that affects cell morphology, orientation, and proliferation.1 To understand the structural response of cells attached to the scaffolds with different degrees of mechanical stability, the formation of F-actin filaments was examined by staining cells with fluorescent dyes after glutaraldehyde fixation. As shown in Figure 4, F-actin filaments

Figure 4. Fluorescent images of F-actin filaments (red) and nuclei (blue) of L929 fibroblasts cultured on PS/PI bilayer films with (a) 25 nm- and (b) 200 nm-dPS, respectively. Scale bars correspond to 20 μm.

formed sufficiently in the cells that adhered to the mechanically stable surface with a thicker upper PS layer. In contrast, this was not the case for the mechanically unstable surface having a thinner upper PS layer. Recently, anisotropic cell adhesion and spreading were reported for poly(lactic) acid (PLA) nanosheets coated on a patterned stainless steel substrate as cell scaffolds.21 When the film thickness of PLA became thinner, the cells adhered in accordance with the pattern of stiff material on the underlying metal substrate. Their finding is consistent with our claim here. Model Protein Adsorption on the PS/PI Bilayer Scaffolds. During cell adhesion processes, various types of proteins play an important role in subsequent cell adhesion. Here, we evaluated the adsorption amount of model proteins on the PS/PI bilayer films. Figure 5 shows the frequency shift values of QCM which represent the mass of HSA and IgG adsorbed on the substrates. In both proteins, the adsorption amounts on the thicker and thinner PS/PI bilayers were almost

Figure 6. Phase contrast images of L929 fibroblasts adhered on the plasma-treated PS/PI bilayer films with (a) 25 nm- and (b) 200 nmdPS, respectively. Scale bars correspond to 50 μm. (c) Closed and open circles refer to the number of cells (N) on the plasma-treated PS/PI bilayer and PS monolayer films, respectively. Error bars represent the standard deviations of the measurements. D

dx.doi.org/10.1021/la4027706 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

■

Article

CONCLUSION We have revealed the effects of the surface mechanical properties of the polymer scaffolds on fibroblast adhesion. Cells were insensitive to the difference in elastic modulus of the top surface as long as the surface E′ was in the range of GPa to MPa. On the contrary, cell adhesion behavior was affected by mechanical instability arising from the underlying region of the scaffold. A conclusive study on the mechanism responsible for this cellular behavior in conjunction with the data using different synthetic polymers will be reported in the near future. We believe that this fundamental knowledge is useful to develop novel highly functional scaffolds using synthetic polymers.

compared to that for the untreated scaffolds owing to the surface hydrophilicity. In addition, cells could well spread on the scaffolds after the plasma treatment. This is because an attractive force between cells and the plasma-treated surface might be more moderate than that between cells and the intact surface. Thus, cells attached on the plasma-treated surface easily transferee to the next spreading process. This can be seen for the bilayer film with a thick PS layer, where most of cells spread out, as shown in panel (b). On the other hand, in the case of the thinner-PS/PI film, small amounts of cells were still round, as shown in panel (a). The number of cells adhered on these scaffolds looks to be slightly decreased once the upper PS layer became thinner than approximately 25 nm, as shown in panel (c). A possible explanation for the decreased effect of the mechanical instability on the cell adhesion after the plasma treatment is that the effect of hydrophilicity is more dominant than the mechanical instability for the current PS/PI bilayer system. Although the effect of mechanical instability on the cell adhesion was not apparently striking after the plasma treatment in terms of the number of cells adhered, it is more evident considering other parameters like in the shape of cells adhered. Figure 7 shows the abundance of projected area of L929 cells

■

ASSOCIATED CONTENT

S Supporting Information *

See the Supporting Information (SI) for detailed polymer synthesis, and preparation and characterization of the scaffold films. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*(H.M.) E-mail [email protected]. (K.T.) FAX: +81-92-802-2880; TEL: +81-92-802-2878; E-mail: k-tanaka@ cstf.kyushu-u.ac.jp. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research was partly supported by the Scientific Research on Innovative Areas ‘‘Molecular Soft-Interface Science’’ (No. 23106716) program and by a Grant-in-Aid for Scientific Research (B) (No. 24350061) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. We deeply thank Dr. Tomoyasu Hirai of Kyushu University for his fruitful discussion, especially, on the synthesis of PI by using an anionic polymerization method.

Figure 7. Abundance of projected area of L929 fibroblasts adhered on plasma-treated (a) PS monolayer and (b) PS/PI bilayer films with different dPS, respectively. The normal distribution curves are superimposed. The number of cells for calculation of the abundance is three hundred for each sample.

■

ABBREVIATIONS PS, polystyrene; PI, polyisoprene; ECM, extracellular matrix; MSC, mesenchymal stem cells; E′, storage modulus; GPC, gel permeation chromatography; Mn, number-average molecular weight; Mw, weight-average molecular weight; dPS, thickness of the layer of PS; AFM, atomic force microscopy; MEK, methyl ethyl ketone; PBS, phosphate buffered saline; DAPI, 4′,6diamidino-2-phenylindole; QCM, quartz-crystal microbalance; HSA, human serum albumin; IgG, immunoglobulin G; RMS, root-mean-square surface roughness; SVM, scanning viscoelasticity microscopic; Tg, glass−rubber transition temperature; PLA, poly(lactic) acid

adhered on the PS monolayer and PS/PI bilayer films with different dPS. The peak position and fwhm obtained from the normal distribution curves in Figure 7 are summarized in Table S2. All samples here used were treated by plasma. In the case of the PS monolayer scaffolds, two histograms for the thinner and thicker PS films were comparable, meaning that the cell spreading was again insensitive to the thickness of the PS films directly coated on glass substrates. On the other hand, in the case of the PS/PI bilayer films, the cell spreading depended on the dPS value. Abundance of cells with a smaller projected area (i.e., immature cells) was greater on the thinner PS/PI scaffolds than on the thicker PS/PI scaffolds. Based on the histogram data, it can be concluded that the cell spreading on the scaffolds was related to the mechanical instability originated from a deeper region even for the hydrophilic polymer surfaces.

■

REFERENCES

(1) Dubiel, E. A.; Martin, Y.; Vermette, P. Bridging the gap between physicochemistry and interpretation prevalent in cell-surface interactions. Chem. Rev. 2011, 111, 2900−2936. (2) Bao, G.; Suresh, S. Cell and molecular mechanics of biological materials. Nat. Mater. 2003, 2, 715−725.

E

dx.doi.org/10.1021/la4027706 | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

standing poly(L-lactic acid) nanosheets. Langmuir 2011, 23, 13173− 13182.

(3) Stamenovic, D.; Wang, N. J. Cellular responses to mechanical stress, Invited Review: Engineering approaches to cytoskeletal mechanics. Appl. Physiol. 2000, 89, 2085−2090. (4) Maguire, P.; Kilpatrick, J. I.; Kelly, G.; Prendergast, P. J.; Campbell, V. A.; O’Connell, B. C.; Jarvis, S. P. Direct mechanical measurement of geodesic structures in rat mesenchymal stem cells. HFSP J. 2007, 1, 181−191. (5) Pelham, R. J., Jr.; Wang, Y.-L. Cell locomotion and focal adhesions are regulated by substrate flexibility. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 13661−13665. (6) Discher, D. E.; Janmey, P.; Wang, Y. L. Tissue cells feel and respond to the stiffness of their substrate. Science 2005, 310, 1139− 1143. (7) Chou, S.-Y.; Cheng, C.-M.; LeDuc, P. R. Composite polymer systems with control of local substrate elasticity and their effect on cytoskeletal and morphological characteristics of adherent cells. Biomaterials 2009, 30, 3136−3142. (8) Yoshikawa, Y. H.; Rossetti, F. F.; Kaufmann, S.; Kaind, T.; Madsen, J.; Engel, U.; Lewis, A. L.; Armes, S. P.; Tanaka, M. Quantitative evaluation of mechanosensing of cells on dynamically tunable hydrogels. J. Am. Chem. Soc. 2011, 133, 1367−1374. (9) Engler, A. J.; Sen, S.; Sweeney, H. L.; Discher, D. E. Matrix elasticity directs stem cell lineage specification. Cell 2006, 126, 677− 689. (10) Lo, C.-M.; Wang, H.-B.; Dembo, M.; Wang, Y.-L. Cell movement is guided by the rigidity of the substrate. Biophys. J. 2000, 79, 144−152. (11) Kidoaki, S.; Matsuda, T. Microelastic gradient gelatinous gels to induce cellular mechanotaxis. J. Biotechnol. 2008, 133, 225−235. (12) Hörning, M.; Kidoaki, S.; Kawano, T.; Yoshikawa, K. Rigidity matching between cells and the extracellular matrix leads to the stabilization of cardiac conduction. Biophys. J. 2012, 102, 379−387. (13) Tanaka, K.; Takahara, A.; Kajiyama, T. Rheological analysis of surface relaxation process of monodisperse polystyrene films. Macromolecules 2000, 33, 7588−7593. (14) Satomi, N.; Tanaka, K.; Takahara, A.; Kajiyama, T. Effect of internal bulk phase on surface viscoelastic properties by scanning probe microscopy. Macromolecules 2001, 34, 6420−6423. (15) Tanaka, K.; Taura, A.; Ge, S.; Takahara, A.; Kajiyama, T. Molecular weight dependence of surface dynamic viscoelastic properties for the monodisperse polystyrene film. Macromolecules 1996, 29, 3040−3042. (16) Serizawa, T.; Yamashita, K.; Akashi, M. Cell-adhesive and bloodcoagulant properties of ultrathin poly(methyl methacrylate) stereocomplex films. J. Biomater. Sci. Polym. Ed. 2004, 15, 511−526. (17) Pyne, A. R.; Rheology of Elastomers; Pergamon Press: London, 1958. (18) Karim, A.; Kumar, S. Polymer Surfaces, Interfaces and Thin Films; World Scientific: Singapore, 2000. (19) The surface Tg of the upper PS layer might be a function of thickness. However, the results shown in Figure 1 did not show any possibility that the Tg at the outermost surface in the PS layer impacts on the cell adhesion behavior. (20) The mechanical instability that we claim here originates from the underneath rubbery PI layer. Although the upper PS layer of the bilayer film was mechanically stiff under such conditions, the surface E′ of the system could be regulated by the stiffness of the underneath layer. This is like the mechanical response of the glassy, namely stiff, PS layer floated on the water reservoir. When the PS layer is sufficiently thick enough, the mechanical response is elastic. On the contrary, the mechanical response is viscous for thin and ultrathin PS layers. Our SVM measurement revealed an effect of mechanical response of the underneath PI layer on the surface E′ in the PS/PI bilayer. In our PS/PI bilayer system, the surface E′ decreased with decreasing the PS layer thickness once the upper PS layer became thinner than 70 nm at room temperature. (21) Fujie, T.; Ricotti, L.; Desii, A.; Menciassi, A.; Dario, P.; Mattoli, V. Evaluation of substrata effect on cell adhesion properties using free F

dx.doi.org/10.1021/la4027706 | Langmuir XXXX, XXX, XXX−XXX