Poly(vinylmethylsiloxane) Elastomer Networks as Functional ... · Published: March 01, 2011 r 2011...

Published: March 01, 2011

r 2011 American Chemical Society 1265 dx.doi.org/10.1021/bm101549y | Biomacromolecules 2011, 12, 1265–1271

ARTICLE

pubs.acs.org/Biomac

Poly(vinylmethylsiloxane) Elastomer Networks as FunctionalMaterials for Cell Adhesion and Migration StudiesShoeb Ahmed, Hyun-kwan Yang, Ali E. Ozcam, Kirill Efimenko, Michael C. Weiger, Jan Genzer,* and JasonM. Haugh*

Department of Chemical and Biomolecular Engineering, North Carolina State University, Raleigh, North Carolina 27695-7905, UnitedStates

’ INTRODUCTION

In humans and other multicellular organisms, cell adhesionand migration phenomena are fundamental to transient physio-logical processes such as development, the immune response,and wound repair. Productive cell movement is generallyachieved through a balance of distinct forces: adhesion to thesurrounding extracellular matrix (ECM), protrusion of the lead-ing membrane edge, and contraction of the cell body.1,2 Inconnective tissue cells such as fibroblasts, all of these subpro-cesses are controlled at least in part by the integrin family ofadhesion receptors, which act simultaneously as mechanicalelements linking the ECM to the cell’s actin cytoskeleton andas hubs for recruitment and activation of intracellular signalingmolecules that promote membrane protrusion.3,4 In particular,phosphoinositide 3-kinases (PI3Ks) and Rho-family GTPases(e.g., Rac and Cdc42) have been implicated most strongly intransducing integrin-mediated signaling to actuate polymeriza-tion of actin.5,6 Although it is appreciated that themechanical andbiochemical functions of integrins are hardly independent, themechanisms by which they are coupled are not yet fullyelucidated.7-9

The migration behaviors of fibroblasts and other motile cellsin culture depend on the density and composition of ECM

deposited on the underlying surface. In general, there is amonotonically increasing (but saturable) relationship betweenECM density and adhesion strength, determined by the numberof integrin-ECM bonds formed in the cell membrane-substratumcontact zone; however, cell migration speed is optimized atintermediate adhesiveness.10 Thus, provided that the ECMdensity at surface saturation exceeds the optimal level for thecell type in question, there will be a bell-shaped dependence ofmigration speed on ECM density or coating concentration. Amajor issue in such studies is the presentation of adhesive ligandson the surface, which is often not well characterized. Fibronectin,collagen, and other ECM proteins are large in size and readilyphysisorb on glass, plastic, and other materials. Although itprovides a facile means of immobilization, protein adsorptionis often accompanied by some degree of denaturation, such thatthe protein conformations on the surface are heterogeneous andvary according to the composition of the surface and the otherimmobilization conditions.11,12 Further, physisorbed proteinscan alter their conformations and positions on the surface withtime, which can be an issue especially when it is assumed that the

Received: December 21, 2010

ABSTRACT: Cell migration is central to physiological re-sponses to injury and infection and in the design of biomaterialimplants. The ability to tune the properties of adhesivematerialsand relate those properties in a quantitative way to the dynamicsof intracellular processes remains a definite challenge in themanipulation of cell migration. Here, we propose the use ofpoly(vinylmethylsiloxane) (PVMS) networks as novel substratafor cell adhesion and migration. These materials offer the abilityto tune independently chemical functionality and elastic mod-ulus. Importantly, PVMS networks are compatible with totalinternal reflection fluorescence (TIRF) microscopy, which is ideal for interrogating the cell-substratum interface; this lattercharacteristic presents a distinct advantage over polyacrylamide gels and other materials that swell with water. To demonstrate thesecapabilities, adhesive peptides containing the arginyl-glycyl-aspartic acid (RGD) tripeptide motif were successfully grafted to thesurface of PVMS network using a carboxyl-terminated thiol as a linker. Peptide-specific adhesion, spreading, and random migrationof NIH 3T3 mouse fibroblasts were characterized. These experiments show that a peptide containing the synergy sequence offibronectin (PHSRN) in addition to RGD promotes more productive cell migration without markedly enhancing cell adhesionstrength. Using TIRF microscopy, the dynamics of signal transduction through the phosphoinositide 3-kinase pathway weremonitored in cells as they migrated on peptide-grafted PVMS surfaces. This approach offers a promising avenue for studies ofdirected migration and mechanotransduction at the level of intracellular processes.

1266 dx.doi.org/10.1021/bm101549y |Biomacromolecules 2011, 12, 1265–1271

Biomacromolecules ARTICLE

protein has been deposited with a certain pattern or gradient.13,14

Alternatively, adhesive ligands may be attached covalently to thesurface via a number of protein conjugation chemistries; toensure a chemically homogeneous surface, minimal peptidesthat retain integrin-binding affinity such as those containingthe arginyl-glycyl-aspartic acid (RGD) sequence found infibronectin and other ECM proteins are typically used.15 Anissue with this approach is that other binding motifs found inECM proteins can affect the quality of the adhesive bond; infibronectin, the presence of a synergy sequence (PHSRN) alongwith RGD affects the rigidity of the bond16 and its ability tomediate certain intracellular signaling pathways.17 Other impor-tant considerations include the mechanical stiffness of thesubstratum material18,19 and the degree of clustering of theadhesive ligands.20-22 In addition, the incorporation of fluores-cent beads, which are displaced in response to cell-generatedtraction forces, allows for spatial mapping of those forces in thecontact zone.23-25

An integrated understanding of cell adhesion and migrationwill develop in conjunction with our ability to connect theaforementioned aspects of the cell-substratum interface withthe associated intracellular dynamics. A prominent approach forachieving this goal is total internal reflection fluorescence (TIRF)microscopy, in which the cell-substratum contact area and≈100 nm of the cytosol above it (i.e., < 10% of the cell volume)are selectively illuminated. With the development of geneticallyencoded, fluorescent protein biosensors, this approach has beenused to analyze the formation, maturation, and disassembly ofadhesions,26,27 actin and microtubule polymerization,28,29 andspatiotemporal dynamics of PI3K and Rac signaling30,31 inmigrating cells. A requirement for TIRF microscopy is a contrastin refractive index n between the substratum (typically glass, n =1.52) and the aqueous environment (n = 1.33).

Considering all of these factors, a material that is chemicallyand mechanically tunable and compatible with TIRF microscopywould be highly desirable for cell adhesion and migration studies.Whereas glass or glass with a thin coating is eminently compatiblewith TIRF, it is mechanically stiff; conversely, soft hydrogelsmostly contain water and therefore lack the optical properties forTIRF. Hydrophobic networks such as the commonly used poly-(dimethylsiloxane) (PDMS), with n > 1.4, should be compatiblewith TIRF; however, current methods do not allow for chemicalmodification of PDMS networks without compromising theirintegrity or mechanical properties. Here, we propose the use ofchemically reactive poly(vinylmethylsiloxane) (PVMS)-basednetworks as novel functional materials for cell migration studies.We demonstrate their compatibility with TIRF to analyze themigration and intracellular signaling dynamics of mouse fibro-blasts on surfaces grafted with different RGD-containing peptides.Future studies are envisioned wherein the mechanical complianceand surface topography of the material are also modulated.

’MATERIALS AND METHODS

Cell Culture and Reagents. Stable expression of the 30-phos-phoinositide-specific biosensor construct EGFP-AktPH (fusion of en-hanced green fluorescent protein and the Akt-1 pleckstrin homologydomain) in NIH 3T3 mouse fibroblasts (American Type CultureCollection) was achieved by retroviral infection and puromycin selec-tion, as previously described.30,31 The cells were subcultured usingDulbecco’s modified Eagle’s medium supplemented with 10% v/v fetalbovine serum (FBS) and 1% v/v penicillin/streptomycin/glutamate as

the growth medium (all from Invitrogen, Carlsbad, CA). Prior toexperiments, EGFP-AktPH-expressing cells were serum-starved for∼2.5 h, detached with a brief trypsin-EDTA treatment, then washedand resuspended in either low serum medium for cell adhesion/spreading measurements [phenol red-free Dulbecco’s modified Eagle’smedium supplemented with 1% v/v FBS, 1 mg/mL fatty acid-freebovine serum albumin (BSA), and 1% v/v penicillin/streptomycin/glutamate] or imaging buffer for TIRF microscopy (20 mMHEPES pH7.4, 125 mM NaCl, 5 mM KCl, 1.5 mMMgCl2, 1.5 mM CaCl2, 10 mMglucose, 1% v/v FBS, 2 mg/mL fatty acid-free BSA]. Human plasmafibronectin was obtained from BD Biosciences (San Jose, CA). The twocustom-designed peptides used in this study, CGRGDSY (RGD) andCSSPHSRNSGSGSGSGSGRGDSP (PHSRN-RGD), were obtainedfrom Peptide 2.0, Inc. (Chantilly, VA). 1-Ethyl-3-[3-dimethylaminopro-pyl]carbodiimide hydrochloride (EDC) and N-hydroxysulfosuccini-mide (Sulfo-NHS) were purchased from Thermo Scientific. Exceptwhere otherwise noted, all other reagents were obtained from Sigma-Aldrich.Preparation of Peptide-Grafted PVMS Substrates. The

following procedure is illustrated in Figure 1. Hydroxy-terminatedPVMS was synthesized using the methodology described elsewhere.32

The molecular weight of PVMS used in this study was 35 kDa, asdetermined by size exclusion chromatography (SEC). The siliconeelastomer networks were formed by cross-linking PVMS with poly-(vinylmethoxysiloxane) (PVMES) by means of a Sn-based catalyst.Films of the mixture,≈1 mm thick, were deposited by spin-coating onto2.5 � 2.5 cm2 glass slides and cured at 55 �C (Figure 1A). A siliconeelastomer sheet made of Sylgard-184 with a round opening in the centerwas placed over the PVMS network (Figure 1B). The opening wassubsequently filled with a solution of carboxyl-terminated thiol inmethanol and was sealed with a quartz slide (Figure 1C,D). Carboxyl-terminated thiols, HS-(CH2)n-COOH (hereafter denoted as Cn-COOH), where n = 2, 6, 8, or 10, were attached to the PVMS networkby thiol-ene addition reaction initiated with a pen-Ray 5.5 W lowpressure mercury arc lamp (Ultra-Violet Products, Upland, CA) withprimary output at 254 nm (Figure 1E). No photoinitiator was used, and

Figure 1. Schematic depicting the technological steps involved in thepreparation of peptide-coupled substrates. First, a film of poly-(vinylmethyl siloxane) (PVMS) is deposited onto a glass cover slide(A). The film is covered with a poly(dimethyl siloxane) network-based“well” (B) that is used to entrap a solution of a carboxy-terminated thiolin ethanol (C). After covering the “well” with a quartz slide (E), thesample is exposed to UV light (E) to graft the thiol to the vinyl grouppresent in PVMS. Carboxy-terminated thiol surfaces (F) are subse-quently activated by EDC/sulfo-NHS and exposed to a solution ofpeptides (G) that are chemically coupled to the support via peptidebonds (H).



the reaction time was varied. After the reaction, the film was washed withpure methanol for 1 day and dried.

The carboxylic acid in the COOH-terminus on the PVMS-thiolsubstrate was activated with Sulfo-NHS of 0.015 g (7 mM) and EDCof 0.010 g (5 mM) using MES buffer solution of 10 mL at pH 5-6,resulting in Sulfo-NHS ester (Figure 1F,G). The reaction timewas 1 day,after which the sample was washed with phosphate buffer, pH 7.2-7.5for 2 h. Finally, the peptide was chemically attached to the PVMS-sulfo-NHS ester by contacting the material with 10 mL peptide solution (300μg/mL RGD or 600 μg/mL PHSRN-RGD in phosphate buffer)(Figure 1H). After 1 day, the sample was washed copiously withmethanol and dried. A total of 2 h before each experiment, the substrateswere sterilized with 70% ethyl alcohol followed by preincubation inimaging buffer containing BSA to prevent nonspecific cell adhesion.Chemical Characterization. Fourier transform infrared spectros-

copy in the attenuated total reflection mode (FTIR-ATR) was used tocharacterize chemical changes using a Nicolet 6700 spectrometerequipped with germanium ATR crystal (Ge-ATR). All data wererecorded for 64 scans under extra dry air flux.

The chemical composition of samples was determined with a KratosAxis Ultra DLD X-ray photoelectron spectroscopy (XPS) instrumentusingmonochromated Al KR radiation with charge neutralization. Surveyand high-resolution spectra were collectedwith pass energies of 80 and 20eV, respectively, using both electrostatic and magnetic lenses. Elementalchemical compositions were determined from spectral regression usingVision and CasaXPS software. Peptide coverages on the surface wereestimated from the chemical composition data as detailed below.Total Internal Reflection Fluorescence (TIRF) Microscopy.

Ourprism-basedTIRFmicroscope has beendescribed in detail previously.33

EGFP was excited using a 60 mW, 488 nm line from a tunable wavelengthargon ion laser head (Melles Griot, Irvine, CA). A 20� water immersionobjective (Zeiss Achroplan, 0.5 NA) and 0.63X camera mount were used.Digital images, with 2� 2 binning, were acquired at 2-min intervals using aHamamatsu ORCA ER cooled CCD (Hamamatsu, Bridgewater, NJ), witha fixed exposure time� gain of 1000-1200 ms, and Metamorph software(Universal Imaging, West Chester, PA). The imaging buffer recipe is givenabove, and mineral oil was layered on top of the buffer to preventevaporation during experiments, which were performed at 37 �C.Image Analysis.During cell spreading experiments, phase contrast

images were taken every hour at two different locations of each sample. Acustom code implemented in MATLAB (Mathworks, Natick, MA) wasemployed to distinguish the cells from the rest of the area. Mean spreadarea per cell was calculated based on a manual count of the number ofcells in the field.

Segmentation of TIRF images was performed as follows. A customMATLAB code was used to read the image files as data matrices andprocess them using the k-means clustering method, which bins pixels byrelative intensity, assuming a specified number (k) of peaks in theintensity distribution. For our images, k = 4 was found to be optimum.The pixels in the acellular background (lowest intensity) are grouped inbin 1, whereas regions of intense fluorescence (hot spots) are found inbin 4. For each cell, masks were generated that define regions of interestcorresponding to the entire cell contact area (bins 2-4) and hot spotregions (bin 4 only). Cell centroid coordinates were determined for everyframe using another custom MATLAB code. Mean square displacement(MSD) was calculated for all cells from the centroid tracking data.

’RESULTS AND DISCUSSION

PVMS-based networks were prepared and chemically functio-nalized using carboxyl-terminated thiol as a linker, to which theadhesive peptides RGD and PHSRN-RGD (Scheme 1) wereattached by standard chemical cross-linking (Materials andMethods, Figure 1). Peptide coupling to the carboxythiol-modified PVMS substrate was verified by IR spectroscopy(Figure 2). The presence of C10-COOH in the substrate was

Figure 2. IR absorbance of bare PVMS (top), PVMS coupled with C10-COOH (middle), and PVMS-S-(CH2)10-COOH coupled with RGDpeptide (bottom).

Scheme 1. Structures of Peptide Molecules Used in This Work



confirmed by the presence of a strong COOHpeak at 1708 cm-1,and the coupling of peptide to the substrate is accompanied by areduction of the COOH signal and the concurrent appearance ofamide I and amide II bands present at 1641 and 1562 cm-1,respectively. The amide I band is associated primarily with theC-O stretching vibration and is related directly to the backboneconformation. Amide II results from the N-H bending vibrationand from the C-N stretching vibration. The presence of vinylpeaks observed after peptide attachment indicates that onlysurface-proximal vinyl groups were activated during the thiol-ene addition reaction.

To better quantify the thiol-ene addition reaction andpeptide coupling, we employed X-ray photoelectron spectrosco-py (XPS). Table 1 compares the coverages of C2-COOH andC10-COOH grafted to PVMS, as a function of reaction time, asestimated by XPS. Although the longest linker, C10-COOH,ultimately provides the greatest access of the carboxyl group for

peptide coupling from solution, the coverage of the smaller C2-COOH after thiol-ene reaction is much higher (by about 6-foldon average), consistent with differential access to reactive groupson the PVMS surface. The thiol-ene reaction time was limitedto 15 min in subsequent experiments to avoid UV-induceddamage to the PVMS network. The density of each peptidecoupled to C10-COOH-modified substrates was also estimatedby XPS. The area coverages of RGD and PHSRN-RGD peptideswere estimated to be 1.3 and 0.23% (or coverages of ≈17 and≈3% relative to the C10-COOH underlayer), respectively; asexpected, the coupling of the longer PHSRN-RGD peptide is lessefficient than that of RGD. Assuming an approximately sphericalshape and density of≈1 g/cm3 for each peptide molecule, thesecoverages correspond to estimated peptide densities of roughly 1� 104 and 1� 103 molecules/μm2, respectively, consistent withthe range of adhesive ligand densities found by others to besuitable for fibroblast adhesion and migration.21,34-36

To assess the biological activity of the peptide-grafted sub-strata, we allowed NIH 3T3 fibroblasts to settle, attach, andspread for up to 5 h. Figure 3 shows that PVMS and PVMS withC10-COOH linker only, with nonspecific adhesion passivated byphysisorbed BSA, do not support cell spreading. In contrast,PVMS with physisorbed fibronectin promotes rapid spreading asexpected. As compared with the linker control, both the RGD-and PHSRN-RGD-grafted surfaces supported peptide-specificfibroblast adhesion and spreading, which was nearly completewithin ≈3 h (Figure 3). Of the two surfaces, PHSRN-RGDsupported modestly higher extents of spreading, manifest as thefraction of cells that initiated spreading, despite having a lowerdensity of peptide on the surface as described above. The similar

Table 1. Concentration of C2-COOH and C10-COOHGrafted to PVMS as a Function of Grafting Time

UV exposure

time (min)

coverage of C2-COOH

on PVMS (%)

coverage of C10-COOH

on PVMS (%)

3 10.1 1.1

6 25.4 3.6

9 33.3 5.3

15 44.4 7.5

Figure 3. Adhesion of fibroblasts to PVMS substrates grafted withRGD-containing peptides. Phase contrast images indicate the extent ofcell spreading 1 and 5 h postplating. Each sample was imaged induplicate, and the results shown are representative of four independentexperiments. PVMS only and PVMS with C10-COOH linker serve asnegative controls, whereas PVMS with physisorbed fibronectin (10 μg/mL coating concentration) serves as a positive control (left). PVMSwithsurface-grafted peptides, RGD and PHSRN-RGD, support fibroblastadhesion and spreading, as shown by the accompanying kinetics of meanspread area per cell (right).

Figure 4. Synergy sequence of fibronectin enhances fibroblast migra-tion on PVMS substrates grafted with RGD-containing peptides.Populations of EGFP-AktPH-expressing fibroblasts were monitoredby TIRF microscopy as they migrated on PVMS with C10-COOHlinker and surface-grafted peptide, RGD (n = 25) or PHSRN-RGD (n =12). (a) The mean squared displacement (MSD) of the cell centroid,plotted versus the time interval between observations, was used toestimate the average cell speed, S, and persistence time, P, of eachpopulation. (b) The maximum displacements, dmax, achieved by in-dividual cells within 4 h, are shown as complementary cumulativedistribution functions for the two peptides.



extents of cell spreading suggest that the two peptide surfacesprovide for comparable adhesion strength, which depends notonly on the peptide density but also the affinity and complianceof peptide-integrin bonds.

Whereas both peptide-grafted surfaces support comparablekinetics of fibroblast adhesion and spreading, the surface graftedwith peptide containing the synergy sequence (PHSRN-RGD)was far superior in promoting random fibroblast migration(Figure 4). TIRF microscopy was used to monitor the morphol-ogy of the cell-substratum contact area and, thus, track itscentroid position as a function of time. A plot of mean-squareddisplacement (MSD) versus time interval, averaged over all cellcentroid paths using nonoverlapping intervals and fit using thecommon persistent random walk model,37 was used to estimatethe average cell speed (S) and persistence time (P) for the twocell populations (Figure 4A). As compared with cells on RGD,the cells migrating on PHSRN-RGD exhibited a > 2-fold higher Sand a shorter estimated P; these parameters tend to be antic-orrelated so as to maintain a consistent persistence length SP.38

At long times, the slope of MSD versus time is 2S2P (or 4 μ,where μ is the motility coefficient), and the value of this quantityis >2-fold higher for the PHSRN-RGD surface. Accordingly,when considering the movements of individual cells in eachpopulation, half of the cells on PHSRN-RGDmoved farther than38 μm in 4 h, compared with only 10 μm for RGD (Figure 4B).

It has been appreciated for some time that, for a given adhesiveligand, cell speed is optimized at an intermediate level ofadhesion strength.10,39 Given that the RGD peptide was graftedat a higher density than the PHSRN-RGD peptide, one mightargue that the lower speed fostered by RGD is attributable to the

adhesion strength being higher than optimal, whereas PHSRN-RGD is closer to the optimum conditions. In that case, one wouldexpect cells to initiate spreading earlier and achieve a higher levelof maximum spreading on RGD, which is not observed(Figure 3). If the adhesion strengths on the two surfaces are infact comparable, then it must be concluded that each adhesivebond formed with the PHSRN-RGD peptide is more efficientin transducing intracellular signals or mechanical forces thatpromote productive cell movement. This finding is in accordwith other evidence showing the ability of the fibronectin synergysite to complement the potency of RGD-mediated adhesioncomplexes.17

The stable cell line used in this study expresses a fluorescentfusion protein (EGFP-AktPH) that translocates to the plasmamembrane in response to intracellular signaling through thePI3K pathway. Thus, we simultaneously detect the spatiotem-poral dynamics of this pathway during cell migration by TIRFmicroscopy (Figure 5). Consistent with our previous studies offibroblast migration on fibronectin-coated glass,31 we found thatPI3K signaling is enriched in “hot spot” regions corresponding tomembrane protrusions. The numbers and relative fluorescenceintensities of these PI3K hot spots are not markedly different forcells on RGD versus PHSRN-RGD, despite the disparity inoverall cell movement during the 5 h time course. Accordingly, incells on RGD, hot spots tend to remain in roughly the samelocations during the experiment, whereas in cells on PHSRN-RGD, the hot spots move relative to the centroid and exhibitbranching associated with changes in the overall direction ofmigration (Figure 5). This apparent decoupling of PI3K signal-ing and cell motility is consistent with the results reported in

Figure 5. Live-cell TIRF microscopy reveals spatiotemporal dynamics of PI3K signaling in cell protrusions. Representative, EGFP-AktPH-expressingfibroblasts were monitored by TIRF microscopy as they migrated on RGD or PHSRN-RGD peptide; the fluorescence intensity (pseudocolored fromcool to warm) indicates the relative density of EGFP-AktPH translocation (PI3K signaling), and the arrows show the directions of cell centroidmovement at the indicated times (left). Image segmentation was used to identify hot spots of PI3K signaling in these cells, and their peripheral locations(angle relative to negative x-axis) during the time course are shown as a heat map, with the color code from cool to warm according to the relativefluorescence volume (area � intensity) of each hot spot. The corresponding protrusion heat map indicates areas of protrusion (warm colors) andretraction (cool colors), identified by comparing successive images and weighted according to area; the larger velocity scale (μm/min) for the cell onPHSRN-RGD is indicative of more rapid protrusion/retraction.



previous studies of fibroblast spreading, in which it was foundthat PI3K activation was not necessarily accompanied by hall-marks of integrin activation, such as tyrosine phosphorylation offocal adhesion kinase and paxillin.30 More germane to this work,these results demonstrate the suitability of PVMS-based net-works for studies involving TIRF microscopy.

’CONCLUSIONS

Contact with a solid matrix or support is a prerequisite forgenerating traction during cell movement, and therefore, syn-thetic materials with tunable properties have shown great pro-mise for studying cell adhesion and migration phenomena. Herewe propose the use of PVMS-based networks as chemicallyfunctional substrata that are especially advantageous relative tohydrogels because they do not swell with water (and thus changetheir mechanical properties) and because they offer compatibilitywith TIRF microscopy. We demonstrated surface grafting of twodifferent adhesive peptides containing the RGD tripeptide motifto the PVMS networks and, using TIRF microscopy, monitoredthemovements of living cells and the spatiotemporal dynamics ofintracellular signaling as they moved. This approach opens thedoor for more detailed studies relating the chemical properties ofthe substratum to the dynamics of molecular processes within thecell contact area, including assembly/disassembly of adhesioncomplexes and cytoskeletal structures and activation of varioussignaling pathways.

An important property of a soft material is the ability to tuneits stiffness. Evidence has been mounting for the past decade orso that mechanical compliance of the material affects thecoupling of adhesion and signaling processes and influences cellfate.18 Althoughmechanical properties of the PVMS substrata weused were not explored here, the elastic moduli of the surfaceswere certainly orders of magnitude above the soft-to-hardtransition as perceived by mammalian cells (≈10 kPa). We havepreviously reported approaches for modulating of PVMS net-work stiffness and making stiffness gradients in PVMS,40 and weenvision that these approaches might be modified so as to beuseful for studying cellular responses to so-called “durotactic”gradients.19,41 Another advantageous characteristic of siliconeelastomer networks for cell biological studies is the ability tomold topographical features.42,43 Importantly, silicone elastomernetworks maintain their shape in aqueous solution, unlike gelsthat swell with water. Hence, one may stably incorporatechannels, ridges, bumps, or dimples into or onto the materialwhile also taking advantage of chemical functionality and com-patibility with TIRF microscopy.

’AUTHOR INFORMATION

Corresponding Author*E-mail: [email protected]; [email protected].

’ACKNOWLEDGMENT

This work was supported by the Army Research Office/Defense Threat Reduction Agency (AB07CBT016).

’REFERENCES

(1) Lauffenburger, D. A.; Horwitz, A. F. Cell migration: a physicallyintegrated molecular process. Cell 1996, 84, 359–369.

(2) Schwartz, M. A.; Horwitz, A. F. Integrating adhesion, protrusion,and contraction during cell migration. Cell 2006, 125, 1361–1374.

(3) Hynes, R. O. Integrins: bidirectional, allosteric signaling ma-chines. Cell 2002, 110, 673–687.

(4) Wang, Y. Flux at focal adhesions: slippage clutch, mechanicalgauge, or signal depot. Sci. STKE 2007, 377, pe10. DOI: 10.1126/stke.3772007pe10.

(5) Ridley, A. J.; Schwartz, M. A.; Burridge, K.; Firtel, R. A.;Ginsberg, M.H.; Borisy, G.; Parsons, T. J.; Horwitz, A. R. Cell migration:integrating signals from front to back. Science 2003, 302, 1704–1709.

(6) Cain, R. J.; Ridley, A. J. Phosphoinositide 3-kinases in cellmigration. Biol. Cell 2009, 101, 13–29.

(7) Orr, A. W.; Helmke, B. P.; Blackman, B. R.; Schwartz, M. A.Mechanisms of mechanotransduction. Dev. Cell 2006, 10 (1), 11–20.

(8) Gupton, S. L.; Waterman-Storer, C. M. Spatiotemporal feedbackbetween actomyosin and focal-adhesion systems optimizes rapid cellmigration. Cell 2006, 125, 1361–1374.

(9) Giannone, G.; Dubin-Thaler, B. J.; Rossier, O.; Cai, Y. F.; Chaga,O.; Jiang, G. Y.; Beaver, W.; Dobereiner, H. G.; Freund, Y.; Borisy, G.;Sheetz, M. P. Lamellipodial actin mechanically links myosin activity withadhesion-site formation. Cell 2007, 128 (3), 561–575.

(10) Palecek, S. P.; Loftus, J. C.; Ginsberg, M. H.; Lauffenburger,D. A.; Horwitz, A. F. Integrin-ligand binding properties govern cellmigration speed through cell-substratum adhesiveness. Nature 1997,385, 537–540.

(11) Burmeister, J. S.; Vrany, J. D.; Reichert, W. M.; Truskey, G. A.Effect of fibronectin amount and conformation on the strength ofendothelial cell adhesion to HEMA/EMA copolymers. J. Biomed. Mater.Res. 1996, 30 (1), 13–22.

(12) Lee, M. H.; Ducheyne, P.; Lynch, L.; Boettiger, D.; Composto,R. J. Effect of biomaterial surface properties on fibronectin-alpha(5)beta-(1) integrin interaction and cellular attachment. Biomaterials 2006, 27(9), 1907–1916.

(13) Fang, F.; Szleifer, I. Kinetics and thermodynamics of proteinadsorption: A generalized molecular theoretical approach. Biophys. J.2001, 80 (6), 2568–2589.

(14) Roach, P.; Farrar, D.; Perry, C. C. Interpretation of proteinadsorption: Surface-induced conformational changes. J. Am. Chem. Soc.2005, 127 (22), 8168–8173.

(15) Hersel, U.; Dahmen, C.; Kessler, H. RGD modified polymers:biomaterials for stimulated cell adhesion and beyond. Biomaterials 2003,24 (24), 4385–4415.

(16) Garcia, A. J.; Schwarzbauer, J. E.; Boettiger, D. Distinct activa-tion states of R5β1 integrin show differential binding to RGD andsynergy domains of fibronectin. Biochemistry 2002, 41 (29), 9063–9069.

(17) Friedland, J. C.; Lee, M. H.; Boettiger, D. Mechanicallyactivated integrin switch controls R5β1 function. Science 2009,323, 642–644.

(18) Discher, D. E.; Janmey, P.; Wang, Y. L. Tissue cells feel andrespond to the stiffness of their substrate. Science 2005, 310, 1139–1143.

(19) Isenberg, B. C.; DiMilla, P. A.; Walker, M.; Kim, S.; Wong, J. Y.Vascular smooth muscle cell durotaxis depends on substrate stiffnessgradient strength. Biophys. J. 2009, 97 (5), 1313–1322.

(20) Maheshwari, G.; Brown, G.; Lauffenburger, D. A.; Wells, A.;Griffith, L. G. Cell adhesion and motility depend on nanoscale RGDclustering. J. Cell Sci. 2000, 113 (10), 1677–1686.

(21) Koo, L. Y.; Irvine, D. J.; Mayes, A. M.; Lauffenburger, D. A.;Griffith, L. G. Co-regulation of cell adhesion by nanoscale RGDorganizationand mechanical stimulus. J. Cell Sci. 2002, 115 (7), 1423–1433.

(22) Cavalcanti-Adam, E. A.; Volberg, T.; Micoulet, A.; Kessler, H.;Geiger, B.; Spatz, J. P. Cell spreading and focal adhesion dynamics areregulated by spacing of integrin ligands. Biophys. J. 2007, 92 (8),2964–2974.

(23) Dembo, M.; Wang, Y. Stresses at the cell-to-substrate interfaceduring locomotion of fibroblasts. Biophys. J. 1999, 76, 2307–2316.

(24) Sabass, B.; Gardel, M. L.; Waterman, C. M.; Schwarz, U. S.High-resolution traction force microscopy based on experimental andcomputational advances. Biophys. J. 2008, 94, 207–220.



(25) Fournier, M. F.; Sauser, R.; Ambrosi, D.; Meister, J. J.;Verkhovsky, A. B. Force transmission in migrating cells. J. Cell Biol.2010, 188, 287–297.(26) Huang, C.; Rajfur, Z.; Borchers, C.; Schaller, M. D.; Jacobson,

K. JNK phosphorylates paxillin and regulates cell migration. Nature2003, 424, 219–223.(27) Nayal, A.; Webb, D. J.; Brown, C. M.; Schaefer, E. M.; Vicente-

Manzanares, M.; Horwitz, A. R. Paxillin phosphorylation at Ser273localizes a GIT1-PIX-PAK complex and regulates adhesion and protru-sion dynamics. J. Cell Biol. 2006, 173, 587–599.(28) Krylyshkina, O.; Anderson, K. I.; Kaverina, I.; Upmann, I.;

Manstein, D. J.; Small, J. V.; Toomre, D. K. Nanometer targeting ofmicrotubules to focal adhesions. J. Cell Biol. 2003, 161, 853–859.(29) Bretschneider, T.; Diez, S.; Anderson, K.; Heuser, J.; Clarke,

M.; M€uller-Taubenberger, A.; K€ohler, J.; Gerisch, G. Dynamic actinpatterns and Arp2/3 assembly at the substrate-attached surface of motilecells. Curr. Biol. 2004, 14, 1–10.(30) Weiger, M. C.; Wang, C.-C.; Krajcovic, M.; Melvin, A. T.;

Rhoden, J. J.; Haugh, J. M. Spontaneous phosphoinositide 3-kinasesignaling dynamics drive fibroblast spreading and random migration. J.Cell Sci. 2009, 122, 313–323.(31) Weiger, M. C.; Ahmed, S.; Welf, E. S.; Haugh, J. M. Directional

persistence of cell migration coincides with stability of asymmetricintracellular signaling. Biophys. J. 2010, 98, 67–75.(32) Efimenko, K.; Crowe, J. A.; Manias, E.; Schwark, D.W.; Fischer,

D. A.; Genzer, J. Rapid formation of soft hydrophilic silicone elastomersurfaces. Polymer 2005, 46, 9329–9341.(33) Schneider, I. C.; Haugh, J. M. Quantitative elucidation of a

distinct spatial gradient-sensing mechanism in fibroblasts. J. Cell Biol.2005, 171 (5), 883–892.(34) Neff, J. A.; Tresco, P. A.; Caldwell, K. D. Surface modification

for controlled studies of cell-ligand interactions. Biomaterials 1999,20, 2377–1293.(35) Rajagopolan, P.; Marganski, W. A.; Brown, X. Q.; Wong, J. Y.

Direct comparison of the spread area, contractility, and migration ofbalb/c 3T3 fibroblasts adhered to fibronectin and RGD-modifiedsubstrata. Biophys. J. 2004, 87, 2818–2827.(36) Kuhlman, W.; Taniguchi, I.; Griffith, L. G.; Mayes, A. M.

Interplay between PEO tether length and ligand spacing governs cellspreading on RGD-modified PMMA-g-PEO comb copolymers. Bioma-cromolecules 2007, 8, 3206–3213.(37) Gail, M. H.; Boone, C.W. The locomotion of mouse fibroblasts

in tissue culture. Biophys. J. 1970, 10, 980–993.(38) Ware, M. F.; Wells, A.; Lauffenburger, D. A. Epidermal growth

factor alters fibroblast migration speed and directional persistencereciprocally and in a matrix-dependent manner. J. Cell Sci. 1998,111, 2423–2432.(39) DiMilla, P. A.; Stone, J. A.; Quinn, J. A.; Albelda, S. M.;

Lauffenburger, D. A. Maximal migration of human smooth muscle cellson fibronectin and type IV collagen occurs at an intermediate attach-ment strength. J. Cell Biol. 1993, 122, 729–737.(40) Crowe-Willoughby, J. A.; Weiger, K. L.; Ozcam, A. E.; Genzer,

J. Formation of silicone elastomer networks films with gradients inmodulus. Polymer 2010, 51, 763–773.(41) Lo, C. M.; Wang, H. B.; Dembo, M.; Wang, Y. L. Cell move-

ment is guided by the rigidity of the substrate. Biophys. J. 2000, 79 (1),144–152.(42) Diehl, K. A.; Foley, J. D.; Nealey, P. F.; Murphy, C. J. Nanoscale

topography modulates corneal epithelial cell migration. J. Biomed. Mater.Res., Part A 2005, 75 (3), 603–611.(43) Kim, D. H.; Han, K.; Gupta, K.; Kwon, K. W.; Suh, K. Y.;

Levchenko, A. Mechanosensitivity of fibroblast cell shape and move-ment to anisotropic substratum topography gradients. Biomaterials2009, 30 (29), 5433–5444.

Poly(vinylmethylsiloxane) Elastomer Networks as Functional ... · Published: March 01, 2011 r 2011...

Documents

Transcript of Poly(vinylmethylsiloxane) Elastomer Networks as Functional ... · Published: March 01, 2011 r 2011...