Combined Effects of PEG Hydrogel Elasticity and Cell-Adhesive

Nov 25, 2013 - Combined Effects of PEG Hydrogel Elasticity and Cell-Adhesive ... indicate that cell adhesion, polarity, and associated migration persi...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/Biomac

Combined Effects of PEG Hydrogel Elasticity and Cell-Adhesive Coating on Fibroblast Adhesion and Persistent Migration Dimitris Missirlis* and Joachim P. Spatz Department of New Materials and Biosystems, Max Planck Institute for Intelligent Systems, Heisenbergstr. 3, 70569 Stuttgart, Germany Department of Biophysical Chemistry, University of Heidelberg, 69120 Heidelberg, Germany S Supporting Information *

ABSTRACT: The development and use of synthetic, crosslinked, macromolecular substrates with tunable elasticity has been instrumental in revealing the mechanisms by which cells sense and respond to their mechanical microenvironment. We here describe a hydrogel based on radical-free, cross-linked poly(ethylene glycol) to study the effects of both substrate elasticity and type of adhesive coating on fibroblast adhesion and migration. Hydrogel elasticity was controlled through the structure and concentration of branched precursors, which efficiently react via Michael-type addition to produce the polymer network. We found that cell spreading and focal adhesion characteristics are dependent on elasticity for all types of coatings (RGD peptide, fibronectin, vitronectin), albeit with significant differences in magnitude. Importantly, fibroblasts migrated slower but more persistently on stiffer hydrogels, with the effects being more pronounced on fibronectin-coated substrates. Therefore, our results validate the hydrogels presented in this study as suitable for future mechanosensing studies and indicate that cell adhesion, polarity, and associated migration persistence are tuned by substrate elasticity and biochemical properties.

1. INTRODUCTION

reinforce or disassemble, leading to the observation that on stiffer substrates FA size is amplified.1,11 Much of our current understanding on the effects of elasticity on cell adhesion and migration has relied on the use of synthetic, polymeric materials as substrates with tunable viscoelasticity and a reductionist approach to isolate the effects of elasticity.12 Pioneering work by Wang et al. established poly(acrylamide) (PAAm) hydrogels as protein-repellant, compliant substrata,3,13 replacing the more hydrophobic PDMS materials used previously.14 PAAm is now viewed as the standard material for use in mechanosensing studies, on which ECM protein are coupled to provide cell attachment points. However, recent evidence suggested that the selection of underlying material greatly influences cell mechanoresponse.4,15 It was proposed that the structure of the polymer network at the nanoscale affected the way cell adhesive elements were tethered to the polymer network, which in turn defined the forces acting on the substrate.15 Therefore, it becomes critical to examine mechano-response on substrates of different chemical composition and structure in order to unveil the influential parameters.

The mechanical microenvironment has emerged as a potent regulator of cellular functions. Cells sense and process existing physical stimuli into biochemical information that along with soluble signals of their milieu shape their behavior. A key physical property of the extracellular matrix (ECM) is its mechanical compliance, a property that can be characterized by viscoelasticity measurements.1,2 In vitro experimentation on compliant substrates has revealed that diverse cellular processes, including cell proliferation, migration, and differentiation, depend on substrate elasticity.2−5 In parallel, strong indications exist of a correlation between ECM mechanics and cell functions in vivo, despite the difficulties associated with isolating and quantifying individual factors within a complex network of interdependent relations.2,6,7 Adherent cells interrogate substrate elasticity through forces generated by actomyosin contractility and transmitted to the extracellular space via integrin-mediated anchoring junctions.8,9 The formation and evolution of these junctions is a dynamic process that involves a large number of proteins and interactions.10 Eventually, peripheral focal adhesions (FAs) assemble, which mechanically link the extracellular matrix to stress fibers, allowing force propagation across the cell membrane. According to the applied tension and the corresponding matrix (substrate) deformation adhesions © 2013 American Chemical Society

Received: October 3, 2013 Revised: November 22, 2013 Published: November 25, 2013 195

dx.doi.org/10.1021/bm4014827 | Biomacromolecules 2014, 15, 195−205

Biomacromolecules

Article

hydrophobic surfaces were used, gels were easily detached from the substrates, whereas in the other case, gels adhered to the silanized glass side. The cross-linking reaction was left to proceed at 37 °C for 2 h in humidified atmosphere. Cysteamine-containing hydrogels were prepared in a similar manner with the difference that cysteamine (1 mM in precursor solution) was added immediately after mixing PEGVS with PEGSH. Thiol Quantification. Thiol concentration was determined using the Ellman’s assay (Supporting Information) or an approach based on covalent fluorescent labeling described here. Swollen PEG gels inside wells of a 96-well plate were treated with Tris(2-carboxyethyl)phosphine hydrochloride (TCEP; 1 mM, pH 7) to reduce disulfide bonds. After a 15 min incubation, the TCEP solution was removed and the thiol-reactive dye oregon green 488 maleimide (1.0 mM) added. The reaction was left to proceed for 1 h at room temperature under constant shaking. Gels were washed thoroughly in Milli-Q water until the level of fluorescence in the solution on top of gels reached background values. Fluorescence intensity values of gels were measured using a plate reader (Tecan infinite M200) and the amount of immobilized dye was determined based on a calibration curve. Results are presented as percentage of initial thiols reacted, assuming 100% efficiency in the maleimide−thiol coupling reaction. Fibronectin/Vitronectin Chemisorption. FN or VN were covalently coupled onto hydrogels using the heterobifunctional linker sulfo-SANPAH based on previously published protocols.28 Briefly, gels were covered with a solution of sulfo-SANPAH (1.0 mg/mL) in HEPES buffer (100 mM, pH 7) and irradiated with UV light for 5 min, followed by exchange of the cross-linker solution and repetition of the illumination for 2 min. Gels were then washed thrice with distilled water, twice with PBS, and a 100 μg/mL fibronectin or vitronectin solution was applied on their surface. Following overnight incubation at 4 °C, gels were washed thrice with PBS and used within 5 days of preparation. Immunolabeling and laser scanning confocal microscopy (LSCM; Zeiss LSM 5) were used to compare the amounts of chemisorbed protein on PEG gels of different elasticity. Following protein immobilization, gels were incubated with primary antibodies for 1 h in a solution of bovine serum albumin (BSA; 1% w/v), washed thrice with the BSA solution, and incubated an additional hour with the secondary antibody. The mean fluorescence intensity from five different fields of view on the surface of the gels was calculated, using the same microscope settings. Mechanical Characterization. The Young’s modulus of hydrogels was measured using a Nano Wizard II scanning probe microscope (JPK Instruments AG) mounted on an optical microscope (Zeiss Axiovert 200). Cantilevers with 10 μm spherical borosilicate glass tips and bending spring constants between 1.5 and 2.4 N/m were used (sQube). The spring constant of each cantilever was measured before each set of measurements by the thermal noise calibration method. Measurements were performed in PBS at room temperature. A sample size of 10 force curves/position ×3−5 positions/gel ×3 gels was used. The Young’s modulus was determined based on the Hertz model for a spherical indenter using the software provided by JPK. A Poisson ratio of 0.5 was used according to Matsunaga et al., who studied a similar hydrogel system.29 The Hertz model assumes homogeneous, linear elastic materials and an indentation that is negligible in comparison to the sample thickness. Equilibrium swelling measurements were performed on hydrogel discs that were cast between two hydrophobic coverslips and therefore could be recovered after gelation. The swelling ratio Qw was determined as ms/mi, which is the mass ratio of gels in the swollen state (ms) and immediately postgelation (mi). Measurements were performed at room temperature and 37 °C. Cell Culture. A rat fibroblast cell line (REF52 WT) and REF52 stably transfected with paxillin fused to yellow fluorescent protein (REF52 PAX) were cultured as subconfluent monolayers in Dulbecco’s modified eagle’s medium (DMEM), supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37 °C and 5% CO2 in a humidified atmosphere.30 Cells were serum-starved overnight (12−16 h) prior to seeding, unless noted otherwise. Cells

Besides the choice of the bulk substrate, the type and concentration of surface-exposed, adhesive ligands are also known to influence the response of cells to substrate elasticity.4,16,17 An intriguing possibility is that cells utilize different integrins to sense the mechanics of the ECM18−22 and/or regulate surface integrin activation depending on substrate stiffness.23,24 In particular, the αvβ3 integrin has been proposed to be responsible for mechanotransduction and α5β1 to provide adhesion strength.21 Nevertheless, direct comparisons between various ligands on elastic substrata are sparse. The above highlight the need to extend the study of substrate elasticity effects on cells to different materials and examine them in combination with varying adhesive coatings, in order to reveal potential differences and unifying principles. In this work we have adapted a radical-free, Michael-type addition crosslinking polymerization of branched poly(ethylene glycol) (PEG) precursors to prepare hydrogels with an elasticity range of 4 to 70 kPa,25,26 which spans that of many soft tissues in our body.27 Hydrogels were functionalized with either short cell-adhesive peptides or the full-length ECM glycoproteins fibronectin (FN) and vitronectin (VN). Using these materials as substrates, we have interrogated fibroblast adhesion and migration under the combined effects of elasticity and cell adhesion ligand type and report on the commonalities and differences in cell response.

2. EXPERIMENTAL SECTION Materials. Peptides with sequences GCGWGRGDSPG (RGD) and GCGWGRDGSPG (RDG) were purchased from Chinatech Peptide Co. with >95% purity (one-letter amino acid code). Free thiol concentration was determined by the Ellman’s reagent test. Branched, 4-arm, nonfunctionalized poly(ethylene glycol) (PEGOH) and thiol end-functionalized poly(ethylene glycol) (PEGSH) were purchased from Jenkem Technology U.S.A., Inc. PEG−OH was end-functionalized with vinyl sulfone groups (PEGVS) by modifying a previously described protocol25 (Supporting Information). Lyophilized bovine plasma fibronectin (from amine-free buffer) was purchased from Life Technologies and dissolved in Milli-Q water at a concentration of 1.0 mg/mL. Human vitronectin, oregon green 488 maleimide, fluorescamine, 4′,6-diamidino-2-phenylindole (DAPI), and wheat germ agglutinin alexa fluor 488 conjugate (WGA-AF488) were purchased from Life Techologies. Sulfosuccinimidyl-6-[4′-azido-2′nitrophenylamino]hexanoate (sulfo-SANPAH) was obtained from Pierce. A list of the antibodies used is provided in the Supporting Information. All other reagents and solvents were purchased from Sigma-Aldrich. Hydrogel Preparation. PEG-based hydrogels were prepared via Michael-type addition, cross-linking reaction of PEGVS with PEGSH.25 The molecular weight (MW) of PEGVS and the polymer precursor concentration were varied; formulations are recorded as PEGVS(X)-PEGSH(5k)-Y, where X the MW of PEGVS (10k or 20k) and Y the wt% of PEG in the precursor solution. For peptide incorporation, cysteine-containing peptides (RGD or RDG) were reacted with a fraction of vinyl sulfones on PEGVS for 15 min in PBS 10 mM at room temperature, prior to cross-linking. The PEGVS solution was mixed with PEGSH at a molar VS:SH ratio of 1:1.05 and vortexed for 5 s. Gelation time was determined using the tube inversion method. At predefined time points after precursor mixing, glass vials were inverted and visually inspected: absence of flow for 15 s indicated gelation. In order to obtain flat surfaces, the precursor solution was immediately placed after mixing either between two hydrophobic glass slides (treated with Sigmacote) or between a glass surface silanized with (3-aminopropyl)triethoxysilane (APTES) and a hydrophobic glass slide. The two glass surfaces were separated by Teflon spacers of defined thickness (typically 120 μm). When two 196

dx.doi.org/10.1021/bm4014827 | Biomacromolecules 2014, 15, 195−205

Biomacromolecules

Article

Figure 1. Poly(ethylene glycol) (PEG) hydrogels based on Michael-type addition with tunable elasticity. (A) Schematic representation of hydrogel formation following cross-linking of 4-arm PEG precursors, end-functionalized with vinyl sulfones (PEGVS) or thiols (PEGSH). (B) Thiolcontaining small molecules are incorporated in the polymer network by prior reaction with PEGVS. (C) The Young’s modulus of gels was determined by AFM force spectroscopy and ranged from 4 to 70 kPa, depending on precursor structure and concentration; mean and standard deviation are presented. (D) Values of Young’s modulus and swelling ratio at 25 and 37 °C, for three different gel formulations, designated as “soft”, “medium”, and “stiff”. were detached using accutase treatment for 5 min, resuspended in serum-free DMEM, and centrifuged at 1500 rpm for 4 min. The cell pellet was resuspended in serum-free medium and cells seeded on hydrogels. Imaging. Phase contrast and epifluorescence imaging were performed using a Delta Vision system (Applied Precision Inc.) on an Olympus IX inverted microscope equipped with a cooled CCD camera. For high magnification images a 60×/1.3 NA (Olympus) oilimmersion objective was used and for time-lapse imaging a 10×/0.3 NA (Zeiss) objective. Live cell imaging was performed at 37 °C and 5% CO2. Epifluorescence microscopy was additionally performed using a Leica DM6000B upright microscope equipped with a CCD camera. A water immersion objective 40×/0.8 NA (Leica) was used. Cell Adhesion and Cell Density Measurements. REF52 WT (5 × 103 cells/cm2) in serum-free medium were incubated on top of gels for 1 h. Nonadherent cells were removed by aspiration and gels were placed on a rotating horizontal shaker (150 rpm) for 5 min to remove weakly bound cells. Gels were then washed twice with PBS, fixed in 4% paraformaldehyde for 15 min, and stained with 1.0 μg/mL DAPI. Cell density was calculated by counting fluorescent nuclei on a fixed surface area. Projected Cell Area. REF52 WT (2 × 103 cells/cm2) in serumfree medium were seeded on top of gels for 30 min, after which nonadherent cells were removed by aspiration and supplemented DMEM was added. At predetermined time points, cells were washed with PBS, fixed in 4% paraformaldehyde for 15 min and stained with 10 μg/mL WGA-AF488. Projected cell area was determined using the “Cell Outliner” plugin of ImageJ software (NIH) from fluorescence microscopy images with >100 cells analyzed/experiment and at least two independent experiments. Cell Proliferation Assay. Proliferation of REF52 WT (% of cells actively synthesizing DNA) was evaluated 24 and 48 h after seeding on hydrogels using a commercially available imaging kit from Life Technologies (Cat. No. C10337). Briefly, cells were incubated with 50 μM EdU (5-ethynyl-2′-deoxyuridine) in supplemented DMEM for 4 h and then fixed with 4% formaldehyde in PBS. EdU was stained using click chemistry according to instructions provided with the kit. Cell nuclei were counterstained with 10 μg/mL Hoechst 33342 for 1 h at room temperature. The percentage of EdU-positive cells was calculated by counting >200 cells in multiple fields on each gel. Immunofluorescence Microscopy. Cells (REF52 WT or REF52 PAX) on hydrogels were washed thrice with PBS and fixed with 4% paraformaldehyde for 15 min. Cells were then washed thrice with PBS,

permeabilized with 0.1% Triton X-100 for 15 min and washed thrice with PBS. Next, cells were incubated with 1 wt % bovine serum albumin (BSA) to block protein binding sites. Primary antibodies were diluted in 1% BSA and incubated with cells at room temperature for 1 h. Cells were washed thrice with 1% BSA and incubated with secondary antibodies for 1 h. Filamentous actin (F-actin) was labeled with Phalloidin-tetramethyl rhodamine B isothiocyanate (2.5 μg/mL) and nuclei with DAPI (1.0 μg/mL). Focal Adhesion Analysis. Focal adhesions were quantified from LSCM images of cells immunolabeled against phosphotyrosine, which provided the best signal-to-noise ratio from the antibodies tested. Images (2048 × 2048 pixels) were acquired using a 60×/NA 1.1 (LUMFI; Olympus) water immersion objective and analyzed using a short, custom-written macro in ImageJ, which is provided in the Supporting Information (Figure S1). A lower threshold of 0.5 μm2 was set to exclude small focal complexes and noise. Data from at least 20 cells/experiment and two independent experiments are presented. Single Cell Motility Assay. REF52 WT (2 × 103 cells/cm2) in serum-free medium were seeded on top of gels for 30 min, after which nonadherent cells were removed by aspiration and supplemented DMEM was added. Live-cell, time-lapse microscopy images were acquired every 10 min for 16 h, starting 5 h after cell seeding. Cell trajectories from time-lapse movies were obtained using the “manual tracking” plugin of ImageJ software. Only cells that (1) remained within the field of view during the entire time frame, (2) did not divide, and (3) were spread and apparently alive during the experiment, were analyzed. Speed was calculated as the total path length traveled by a cell divided by the time, and the directionality index was calculated by dividing the distance from the origin by the total path length. Statistical Analysis. All statistical analyses were performed using the software Prism (GraphPad Inc.). Experiments were statistically analyzed using the Tukey test, which compares all pairs of columns. Only statistical significant differences are presented in graphs with p values 350), (C) cell aspect ratio, and (D) FA area (n = 1000−4000) as a function of elasticity and protein coating. (B, D) Box plots with 1−99 percentile are presented and in (C), the mean and SEM are presented.

fold) compared to VN-coated ones (1.5-fold). On RGDfunctionalized gels, no significant differences were noted, even though the trends were analogous to the ECM-coated gels (Figure S12). In summary, our findings revealed that cell migration depends on both the type of coating and substrate stiffness, with a significant decrease in cell speed on stiff substrates for ECM-coated gels and further highlighted a previously unappreciated correlation with persistent migration.

appreciably from 6 to 24 h (Figure 2C). Comparing FN with VN, we noted that cell aspect ratio depended on the type of coating. Cells were more rounded on VN compared to FN, with the difference being statistically significant for medium elasticity gels. We next focused on individual adhesion characteristics as a function of both stiffness and ECM protein coating. Quantification of FAs revealed an increase of FA area with stiffness as anticipated1 (Figure 4D). Interestingly, FAs were significantly larger on VN compared to FN only for gels with intermediate stiffness. Fibroblast Migration Depends on a Combination of Substrate Elasticity and Coating. The combined effects of ECM coating and substrate elasticity on adhesion presented so far were examined on fixed cells providing only a static view of cells. The connection between cell adhesion and migration is widely recognized and accordingly, we anticipated differences in cell motility on substrates of differing elasticity and type of protein coating. In the absence of chemotactic signals, we observed slower migration speed on stiffer substrates, on both FN- and VN-coated gels; however the dependence was dissimilar in magnitude (Figure 5A). Cell speed decreased 33% on the stiffer FN-coated gels compared to a 16% decrease for VN-coated ones. Interestingly, when the directionality index was calculated, a marked dependence on substrate elasticity was noted suggesting that increased traction forces are required for persistent cell movement (Figure 5B). Similar to cell speed, the increase in persistence was larger for FN-coated substrates (2.2-

4. DISCUSSION The PEG-based hydrogels presented in this work fulfill the principal requirements as substrates for mechanobiology studies, including cytocompatibility, controllable mechanical properties and ligand density, versatility in functionalization, and optical transparency for microscopy examination. The range of mechanical properties possible with these materials covers that of many soft connective tissues measured ex vivo, where fibroblasts reside.2,27 They are therefore an attractive substitute to traditional PAAm and PDMS substrates. PEG is a popular polymer for the development of elastic substrates and many different modifications and cross-linking chemistries have been proposed toward hydrogel formation. The large majority of these rely on radical polymerization of vinyl moieties.38,51−55 Here, instead, we exploited Michael-type addition as an alternative cross-linking polymerization; despite its advantageous characteristics, it has received limited attention to produce substrates with tunable mechanics.39,56 The 201

dx.doi.org/10.1021/bm4014827 | Biomacromolecules 2014, 15, 195−205

Biomacromolecules

Article

reported. The different cross-linking chemistry in those studies further indicates that adhesion to PEG-based hydrogels was not specific to the type used in our study. The adhesion mechanism is not clear at present but most likely involves multiple, nonspecific weak interactions between the cell and the polymer network. It is important to note that PEG chains in the hydrogels presented here are end-linked to the matrix, which is in contrast to mobile, surface-tethered chains with free ends, typically used for antiadhesive purposes.63,64 Fibroblasts on bare PEG gels could not form organized focal adhesions or spread in the absence of adhesive ligands; in contrast, incorporated RGD peptides provided specific anchoring points. Cells recognized RGD ligands through their integrin receptors, spread and formed FAs above a threshold concentration. This threshold was estimated at 250 peptides/ μm2 assuming cells can detect peptides 5 nm inside the gel.23,36 This is in agreement to ligand densities accurately determined previously (between 190 and 280 peptides/μm2) that permit efficient spreading and focal adhesion formation for this type of cells30,32 as well as for other mesenchymal cells.65,66 This seemingly high concentration should be considered in view of the fact that FA formation is sensitive to local ligand density instead of the global density presented by the substrate.67 Fibroblasts exhibited an increase of their spreading on stiffer substrates, with similar trends observed on peptide- and ECM protein-functionalized gels. The increase in cell area occurred at an elasticity range (between 5 and 30 kPa) comparable to values reported in literature for PAAm or PEG gels.15,16,43,68−70 The commonality in trends across the different coatings noted in our study, including short peptides, suggests that under the given soluble microenvironment (10% fetal serum) cells sense and respond to mechanical properties of the underlying polymer matrix and not simply to differences in tethering as previously suggested for collagen-coated matrices.15 Interestingly, the above suggestion was derived through the observation (among others) that human keratinocytes and mesenchymal stem cells on PDMS did not respond to substrate elasticity in respect to cell spreading area.15 This finding is in line with a recent study performed on PDMS using human fibroblasts.11 It is not clear why cells on PDMS appear to react distinctly from those on PAAm gels and the gels reported here in respect to their stiffness-dependent spreading behavior.15,16,68−70 The mechanical feedback from end-tethered RGD peptides on soft versus stiff substrates in our study should be directly correlated to the underlying substrate mechanics, in contrast to collagen or other ECM proteins that may be anchored at multiple points along their surface. Therefore, our results suggest that cells respond to substrate elasticity and not only to differences in ligand tethering, at least for RGD-functionalized gels. We speculate that the hydrophobic PDMS permits adsorption of serum or cell-secreted proteins71 that could significantly alter cell spreading and override the mechanical signals of the substrate; the effect of different coatings has indeed been shown to greatly influence cell response to substrate elasticity.16,17 Our observations additionally corroborated a dependence of FA size and phosphorylation activity on substrate elasticity.1,3,44,45 Adhesions reinforce in order to sustain the force as cells generate higher tensions on stiffer substrates.41 By comparing different coatings, we noted that FA size was higher on RGD-functionalized gels followed by VN- and FN-coated gels. Significant differences between VN and FN were noted at intermediate stiffness values but were attenuated at softer and

Figure 5. Fibroblasts migrate slower but more persistently on stiffer gels. (A) Speed and (B) directionality index of REF52 WT cells on top of PEG hydrogels coated with either FN or VN as a function of elasticity.

chemistry is compatible with encapsulation of living cells and enables fabrication of multifunctional synthetic platforms for cell culture.57 Due to the nature of the cross-linking reaction, the resulting gels are homogeneous,29 which should result in a uniform ligand presentation and minimize cell-to-cell variability. Moreover, they permit a high degree of decoupling between mechanics and adhesion ligand density, which is not attainable with reconstituted ECM protein gels.58,59 On the other hand, it should be noted that these synthetic gels are amorphous in contrast to the fibrillar, proteinous networks of the ECM.60 As a consequence they do not exhibit nonlinear strain-stiffening behavior typical of biological gels.61 Our results challenge the widely accepted notion that PEG gels do not permit cell adhesion.38,43 We observed significant fibroblast adhesion in the absence of incorporated, substratebound, cell-adhesive ligands and serum proteins in the medium. Cells adhered even after gel treatment aimed to consume any postreaction, residual, reactive bonds, prolonged cell treatment with trypsin, or following integrin blocking using soluble RGD peptides. Previously, evidence for fibroblast adhesion (but not spreading) on photopolymerized PEG gels, nonfunctionalized62 or functionalized with nonadhesive RDG peptide,36 was 202

dx.doi.org/10.1021/bm4014827 | Biomacromolecules 2014, 15, 195−205

Biomacromolecules

Article

as versatile substrates for mechanosensing studies. Their elastic properties were controlled by precursor selection and concentration and they were readily functionalized with either short peptides or full-length ECM proteins. Fibroblasts adhered on hydrogels independent of elasticity or presence of celladhesive ligands; however, a dependence of cell adhesion on ligand type and density, in combination with substrate elasticity, was demonstrated. In particular, direct comparisons between the ECM proteins fibronectin and vitronectin revealed that fibroblasts spread more on FN-coated gels but formed larger adhesions on VN-coated ones at intermediate stiffness values. Most importantly, we demonstrated that fibroblast migration is additionally regulated by substrate elasticity for ECM-coated hydrogels, with cells being slower but more persistent on the stiffer substrates examined. Overall, we conclude that persistent migration of cells is sensitive to substrate mechanics and biochemistry and that the hydrogel platform presented here is a promising, universal tool for more detailed, future mechanosensing investigations.

stiffer matrices (Figure 4D). Interestingly, significant differences between the two coatings only at intermediate stiffness were also found for projected cell area, on a series of independent experiments (Figure 4B). This suggests the compelling possibility that substrate elasticity sensitizes cells to the underlying coating in a specific range of stiffness. FN and VN differ substantially in size and would therefore present different RGD concentrations, even on ideal, homogeneous monolayers. Moreover, the cell binding affinities and cell generated forces on RGD and FN coatings differ considerably.35,72 Finally, on protein-coated substrates using the sulfo-SANPAH chemistry, cells could detach proteins at a certain force,73 which would impose an artificial upper limit to force generation. The above limitations restrict our ability to quantitatively establish correlations to isolated physical parameters (e.g., adhesion strength or ligand density), despite our observations that ECM coatings influence cell phenotype. Future studies could address such limitations by substituting the protein immobilization chemistry and determining the amount of presented adhesive ligands, for example, by radiolabeling ECM proteins. The coatings utilized in this study aimed to differentially engage the α5β1 and αvβ3 integrins. Early studies demonstrated that cells utilize specifically the FN receptor α5β1 to bind FN and the VN receptor αvβ3 to bind VN.47,48 Thereupon, significant integrin cross-talk and the ability of αvβ3 to bind FN have been recognized.49,74,75 To better discriminate between the influence of two integrins, specific, peptidomimetic ligands were recently introduced76 and used to demonstrate increased traction forces on α5β1-binding ligands compared to αvβ3-binding ones;20 it would be of interest to test their effects on substrate mechanosensing in the future. Adhesion characteristics are intrinsically linked to the dynamic cell process of migration.77,78 Cell speed has been related to the density of adhesive ligands,79,80 adhesion strength35 and substrate elasticity.1,3,81 Previous studies on adherent cells have reported both a decrease in motility on stiffer substrates3 and a biphasic relationship, which was regulated by ligand density.81 Our observations revealed a decrease of cell speed with stiffness in accordance with the former findings, but we cannot exclude that on gels with lower elastic moduli a decrease in cell speed would occur. This dependence additionally appears to be cell specific, since substrate elasticity was shown to enhance motility on a hematopoietic cell line.82 Interestingly, the transition in cell speed and persistence occurred at higher elastic moduli compared to those on which differences on adhesion behavior, indicating a complex correlation between adhesion and motility. Our finding that persistence in single cell migration is higher on stiffer gels suggests that increased stiffness leads to polarization of certain intracellular signals. This agrees well with a recent study demonstrating a stiffness-dependent polarization of nonmuscle myosin IIB in mesenchymal stem cells.83 This effect should not be confounded with the process of durotaxis, where cells migrate from soft to stiff environments;5 in our studies the materials are homogeneous without elasticity gradients. It would be interesting to validate this original observation to different cell types and explore further the modest dependence noted on substrate coating.



ASSOCIATED CONTENT

S Supporting Information *

Supporting experimental methods, tables, and figures are provided including the list of used antibodies, gelation times (Table S1), thiol and cysteamine quantification results (Tables S2−S4, Figure S2), focal adhesion analysis (Figure S1), HPLC analysis of RGD peptide (Figure S3), comparison between ECM protein chemisorption on different gels (Figure S4), comparison of initial cell adhesion on different gels (Figures S5 and S11) or after integrin blocking with soluble RGD (Figure S6), additional immunofluorescence microscopy images on PEG gels (Figure S7 and S8), Young’s moduli of RGDfunctionalized gels (Figure S9), proliferation results on gels (Figure S10), and migration results on RGD-functionalized gels (Figure S12). This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Tel.: +49 6221 54 4969. E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Janosch Deeg for introduction to AFM measurements and Dr. Elisabetta Ada Cavalcanti-Adam for comments on the final manuscript. We acknowledge financial support from a Marie Curie IIF fellowship (to D.M.) and a CellNetworks Postdoctoral fellowship (to D.M.). The Max Planck Society is acknowledged for financial support. The work was part of the European Union Seventh Framework Program (FP7/2007-2013) under Grant Agreement No. NMP4-LA2009-229289 NanoII. This work is also part of the excellence cluster CellNetworks at the University of Heidelberg. J.P.S. is the Weston Visiting Professor at the Weizmann Institute of Science.



REFERENCES

(1) Guo, W.-H.; Frey, M. T.; Burnham, N. A.; Wang, Y.-L. Biophys. J. 2006, 90, 2213−2220. (2) Discher, D. E.; Janmey, P.; Wang, Y.-L. Science 2005, 310, 1139− 1143.

5. CONCLUSIONS The PEG-based hydrogels described in this study were synthesized by radical-free cross-linking chemistry and served 203

dx.doi.org/10.1021/bm4014827 | Biomacromolecules 2014, 15, 195−205

Biomacromolecules

Article

(3) Pelham, R. J.; Wang, Y. L. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 13661−13665. (4) Reilly, G. C.; Engler, A. J. J. Biomech. 2010, 43, 55−62. (5) Lo, C.-M.; Wang, H.-B.; Dembo, M.; Wang, Y.-L. Biophys. J. 2008, 79, 144−152. (6) Jaalouk, D. E.; Lammerding, J. Nat. Rev. Mol. Cell Biol. 2009, 10, 63−73. (7) Dufort, C. C.; Paszek, M. J.; Weaver, V. M. Nat. Rev. Mol. Cell Biol. 2011, 12, 308−319. (8) Roca-Cusachs, P.; Iskratsch, T.; Sheetz, M. P. J. Cell Sci. 2012, 125, 3025−3038. (9) Geiger, B.; Spatz, J. P.; Bershadsky, A. D. Nat. Rev. Mol. Cell Biol. 2009, 10, 21−33. (10) Zaidel-Bar, R.; Geiger, B. J. Cell Sci. 2010, 123, 1385−1388. (11) Prager-Khoutorsky, M.; Lichtenstein, A.; Krishnan, R.; Rajendran, K.; Mayo, A.; Kam, Z.; Geiger, B.; Bershadsky, A. D. Nat. Cell Biol. 2011, 13, 1457−1465. (12) Nemir, S.; West, J. L. Ann. Biomed. Eng. 2010, 38, 2−20. (13) Pelham, R. J.; Wang, Y. L. Mol. Biol. Cell 1999, 10, 935−945. (14) Harris, A. K.; Wild, P.; Stopak, D. Science 1980, 208, 177−179. (15) Trappmann, B.; Gautrot, J. E.; Connelly, J. T.; Strange, D. G. T.; Li, Y.; Oyen, M. L.; Stuart, M. A. C.; Boehm, H.; Li, B.; Vogel, V.; Spatz, J. P.; Watt, F. M.; Huck, W. T. S. Nat. Mater. 2012, 11, 1−8. (16) Rowlands, A. S.; George, P. A.; Cooper-White, J. J. Am. J. Physiol. 2008, 295, C1037−C1044. (17) Chopra, A.; Lin, V.; McCollough, A.; Atzet, S.; Prestwich, G. D.; Wechsler, A. S.; Murray, M. E.; Oake, S. A.; Kresh, J. Y.; Janmey, P. A. J. Biomech. 2012, 45, 824−831. (18) Schiller, H. B.; Hermann, M.-R.; Polleux, J.; Vignaud, T.; Zanivan, S.; Friedel, C. C.; Sun, Z.; Raducanu, A.; Gottschalk, K.-E.; Théry, M.; Mann, M.; Fässler, R. Nat. Cell Biol. 2013, 15, 625−636. (19) Lin, G. L.; Cohen, D. M.; Desai, R. A.; Breckenridge, M. T.; Gao, L.; Humphries, M. J.; Chen, C. S. FEBS Lett. 2013, 587, 763− 769. (20) Rahmouni, S.; Lindner, A.; Rechenmacher, F.; Neubauer, S.; Sobahi, T. R. A.; Kessler, H.; Cavalcanti-Adam, E. A.; Spatz, J. P. Adv. Mater. 2013, 25, 5869−5874. (21) Roca-Cusachs, P.; Gauthier, N. C.; del Rio, A.; Sheetz, M. P. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 16245−16250. (22) Jiang, G.; Huang, A. H.; Cai, Y.; Tanase, M.; Sheetz, M. P. Biophys. J. 2006, 90, 1804−1809. (23) Huebsch, N.; Arany, P. R.; Mao, A. S.; Shvartsman, D.; Ali, O. A.; Bencherif, S. A.; Rivera-Feliciano, J.; Mooney, D. J. Nat. Mater. 2010, 9, 518−526. (24) Du, J.; Chen, X.; Liang, X.; Zhang, G.; Xu, J.; He, L.; Zhan, Q.; Feng, X.-Q.; Chien, S.; Yang, C. Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 9466−9471. (25) Lutolf, M. P.; Hubbell, J. A. Biomacromolecules 2003, 4, 713− 722. (26) Lutolf, M. P.; Lauer-Fields, J. L.; Schmoekel, H. G.; Metters, A. T.; Weber, F. E.; Fields, G. B.; Hubbell, J. A. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 5413−5418. (27) Swift, J.; Ivanovska, I. L.; Buxboim, A.; Harada, T.; Dingal, P. C. D. P.; Pinter, J.; Pajerowski, J. D.; Spinler, K. R.; Shin, J. W.; Tewari, M.; Rehfeldt, F.; Speicher, D. W.; Discher, D. E. Science 2013, 341, 1240104−1240104. (28) Fischer, R. S.; Myers, K. A.; Gardel, M. L.; Waterman, C. M. Nat. Protoc. 2012, 7, 2056−2066. (29) Matsunaga, T.; Sakai, T.; Akagi, Y.; Chung, U.-I.; Shibayama, M. Macromolecules 2009, 42, 1344−1351. (30) Cavalcanti-Adam, E. A.; Volberg, T.; Micoulet, A.; Kessler, H.; Geiger, B.; Spatz, J. P. Biophys. J. 2007, 92, 2964−2974. (31) Anseth, K. S.; Bowman, C. N.; Brannon-Peppas, L. Biomaterials 1996, 17, 1647−1657. (32) Arnold, M.; Cavalcanti-Adam, E. A.; Glass, R.; Blümmel, J.; Eck, W.; Kantlehner, M.; Kessler, H.; Spatz, J. P. ChemPhysChem 2004, 5, 383−388. (33) Massia, S.; Hubbell, J. J. Cell Biol. 1991, 114, 1089−1100.

(34) Koo, L.; Irvine, D.; Mayes, A.; Lauffenburger, D.; Griffith, L. J. Cell Sci. 2002, 115, 1423−1433. (35) Rajagopalan, P.; Marganski, W. A.; Brown, X. Q.; Wong, J. Y. Biophys. J. 2004, 87, 2818−2827. (36) Hern, D. L.; Hubbell, J. A. J. Biomed. Mater. Res. 1998, 39, 266− 276. (37) Comisar, W. A.; Kazmers, N. H.; Mooney, D. J.; Linderman, J. J. Biomaterials 2007, 28, 4409−4417. (38) Zhu, J. Biomaterials 2010, 31, 4639−4656. (39) Zustiak, S. P.; Leach, J. B. Biomacromolecules 2010, 11, 1348− 1357. (40) Schwartz, M. A. Cold Spring Harbor Perspect. Biol. 2010, 2, a005066−a005066. (41) Balaban, N. Q.; Schwarz, U. S.; Riveline, D.; Goichberg, P.; Tzur, G.; Sabanay, I.; Mahalu, D.; Safran, S.; Bershadsky, A.; Addadi, L.; Geiger, B. Nat. Cell Biol. 2001, 3, 466−472. (42) Stricker, J.; Aratyn-Schaus, Y.; Oakes, P. W.; Gardel, M. L. Biophys. J. 2011, 100, 2883−2893. (43) Peyton, S. R.; Raub, C. B.; Keschrumrus, V. P.; Putnam, A. J. Biomaterials 2006, 27, 4881−4893. (44) Plotnikov, S. V.; Pasapera, A. M.; Sabass, B.; Waterman, C. M. Cell 2012, 151, 1513−1527. (45) Wei, W. C.; Lin, H. H.; Shen, M. R.; Tang, M. J. Am. J. Physiol. 2008, 295, C1579−C1589. (46) Klein, E. A.; Yin, L.; Kothapalli, D.; Castagnino, P.; Byfield, F. J.; Xu, T.; Levental, I.; Hawthorne, E.; Janmey, P. A.; Assoian, R. K. Curr. Biol. 2009, 19, 1511−1518. (47) Fath, K. R.; Edgell, C. J.; Burridge, K. J. Cell Sci. 1989, 92 (Pt 1), 67−75. (48) Singer, I. I.; Scott, S.; Kawka, D. W.; Kazazis, D. M.; Gailit, J.; Ruoslahti, E. J. Cell Biol. 1988, 106, 2171−2182. (49) Plow, E. F.; Haas, T. A.; Zhang, L.; Loftus, J.; Smith, J. W. J. Biol. Chem. 2000, 275, 21785−21788. (50) Malmström, J.; Lovmand, J.; Kristensen, S.; Sundh, M.; Duch, M.; Sutherland, D. S. Nano Lett. 2011, 11, 2264−2271. (51) Herrick, W. G.; Nguyen, T. V.; Sleiman, M.; McRae, S.; Emrick, T. S.; Peyton, S. R. Biomacromolecules 2013, 14, 2294−2304. (52) van Dijk, M.; van Nostrum, C. F.; Hennink, W. E.; Rijkers, D. T. S.; Liskamp, R. M. J. Biomacromolecules 2010, 11, 1608−1614. (53) Miller, J. S.; Shen, C. J.; Legant, W. R.; Baranski, J. D.; Blakely, B. L.; Chen, C. S. Biomaterials 2010, 31, 3736−3743. (54) Aydin, D.; Louban, I.; Perschmann, N.; Blümmel, J.; Lohmüller, T.; Cavalcanti-Adam, E. A.; Haas, T. L.; Walczak, H.; Kessler, H.; Fiammengo, R.; Spatz, J. P. Langmuir 2010, 26, 15472−15480. (55) Burdick, J. A.; Anseth, K. S. Biomaterials 2002, 23, 4315−4323. (56) Gilbert, P. M.; Havenstrite, K. L.; Magnusson, K. E. G.; Sacco, A.; Leonardi, N. A.; Kraft, P.; Nguyen, N. K.; Thrun, S.; Lutolf, M. P.; Blau, H. M. Science 2010, 329, 1078−1081. (57) Lutolf, M. P.; Hubbell, J. A. Nat. Biotechnol. 2005, 23, 47−55. (58) Grinnell, F.; Ho, C.-H. Biomaterials 2013, 34, 965−974. (59) Pathak, A.; Kumar, S. Integr. Biol. 2011, 3, 267. (60) Hynes, R. O. Science 2009, 326, 1216−1219. (61) Storm, C.; Pastore, J. J.; MacKintosh, F. C.; Lubensky, T. C.; Janmey, P. A. Nature 2005, 435, 191−194. (62) Schulte, V. A.; Díez, M.; Mö ller, M.; Lensen, M. C. Biomacromolecules 2009, 10, 2795−2801. (63) Blümmel, J.; Perschmann, N.; Aydin, D.; Drinjakovic, J.; Surrey, T.; López-García, M.; Kessler, H.; Spatz, J. P. Biomaterials 2007, 28, 4739−4747. (64) Groll, J.; Fiedler, J.; Engelhard, E.; Ameringer, T.; Tugulu, S.; Klok, H.-A.; Brenner, R. E.; Moeller, M. J. Biomed. Mater. Res. 2005, 74A, 607−617. (65) Schvartzman, M.; Palma, M.; Sable, J.; Abramson, J.; Hu, X.; Sheetz, M. P.; Wind, S. J. Nano Lett. 2011, 11, 1306−1312. (66) Frith, J. E.; Mills, R. J.; Cooper-White, J. J. J. Cell Sci. 2012, 1− 11. (67) Deeg, J. A.; Louban, I.; Aydin, D.; Selhuber-Unkel, C.; Kessler, H.; Spatz, J. P. Nano Lett. 2011, 11, 1469−1476. 204

dx.doi.org/10.1021/bm4014827 | Biomacromolecules 2014, 15, 195−205

Biomacromolecules

Article

(68) Yeung, T.; Georges, P. C.; Flanagan, L. A.; Marg, B.; Ortiz, M.; Funaki, M.; Zahir, N.; Ming, W.; Weaver, V.; Janmey, P. A. Cell Motil. Cytoskeleton 2004, 60, 24−34. (69) Solon, J.; Levental, I.; Sengupta, K.; Georges, P. C.; Janmey, P. A. Biophys. J. 2007, 93, 4453−4461. (70) Engler, A.; Bacakova, L.; Newman, C.; Hategan, A.; Griffin, M.; Discher, D. Biophys. J. 2004, 86, 617−628. (71) McDonald, J. C.; Duffy, D. C.; Anderson, J. R.; Chiu, D. T.; Wu, H.; Schueller, O. J.; Whitesides, G. M. Electrophoresis 2000, 21, 27−40. (72) Petrie, T. A.; Capadona, J. R.; Reyes, C. D.; Garcia, A. J. Biomaterials 2006, 27, 5459−5470. (73) Yip, A. K.; Iwasaki, K.; Ursekar, C.; Machiyama, H.; Saxena, M.; Chen, H.; Harada, I.; Chiam, K.-H.; Sawada, Y. Biophys. J. 2013, 104, 19−29. (74) Huveneers, S.; Danen, E. H. J. J. Cell Sci. 2009, 122, 1059−1069. (75) Morgan, M. R.; Byron, A.; Humphries, M. J.; Bass, M. D. IUBMB Life 2009, 61, 731−738. (76) Rechenmacher, F.; Neubauer, S.; Polleux, J.; Mas-Moruno, C.; De Simone, M.; Cavalcanti-Adam, E. A.; Spatz, J. P.; Fässler, R.; Kessler, H. Angew. Chem., Int. Ed. 2012, 52, 1572−1575. (77) Vicente-Manzanares, M.; Ma, X.; Adelstein, R. S.; Horwitz, A. R. Nat. Rev. Mol. Cell Biol. 2009, 10, 778−790. (78) Lauffenburger, D. A.; Horwitz, A. F. Cell 1996, 84, 359−369. (79) Wacker, B. K.; Alford, S. K.; Scott, E. A.; Thakur, M. D.; Longmore, G. D.; Elbert, D. L. Biophys. J. 2008, 94, 273−285. (80) Maheshwari, G.; Brown, G.; Lauffenburger, D. A.; Wells, A.; Griffith, L. G. J. Cell Sci. 2000, 113 (Pt 10), 1677−1686. (81) Peyton, S. R.; Putnam, A. J. J. Cell. Physiol. 2005, 204, 198−209. (82) Lee-Thedieck, C.; Rauch, N.; Fiammengo, R.; Klein, G.; Spatz, J. P. J. Cell Sci. 2012, 125, 3765−3775. (83) Raab, M.; Swift, J.; P Dingal, P. C. D.; Shah, P.; Shin, J. W.; Discher, D. E. J. Cell Biol. 2012, 199, 669−683.

205

dx.doi.org/10.1021/bm4014827 | Biomacromolecules 2014, 15, 195−205