Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/269172613

Lithography-freefabricationofreconfigurablesubstratetopographyforcontactguidance

ARTICLEinBIOMATERIALS·JANUARY2015

ImpactFactor:8.56·DOI:10.1016/j.biomaterials.2014.10.078·Source:PubMed

READS

108

5AUTHORS,INCLUDING:

PitiratPholpabu

CarnegieMellonUniversity

2PUBLICATIONS10CITATIONS

SEEPROFILE

HaoshengWu

CarnegieMellonUniversity

10PUBLICATIONS35CITATIONS

SEEPROFILE

ChristopherJBettinger

CarnegieMellonUniversity

64PUBLICATIONS2,743CITATIONS

SEEPROFILE

Allin-textreferencesunderlinedinbluearelinkedtopublicationsonResearchGate,

lettingyouaccessandreadthemimmediately.

Availablefrom:ChristopherJBettinger

Retrievedon:03February2016

lable at ScienceDirect

Biomaterials 39 (2015) 164e172

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomateria ls

Lithography-free fabrication of reconfigurable substrate topographyfor contact guidance

Pitirat Pholpabu a, Stephen Kustra a, Haosheng Wu b, Aditya Balasubramanian b,Christopher J. Bettinger a, b, *

a Department of Biomedical Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh 15213, PA, USAb Department of Materials Science and Engineering, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh 15213, PA, USA

a r t i c l e i n f o

Article history:Received 16 September 2014Accepted 27 October 2014Available online 20 November 2014

Keywords:TopographyContact guidanceMicrofabricationMicrostructure

* Corresponding author. Department of BiomedicalUniversity, 5000 Forbes Avenue, Pittsburgh 15213, PA

E-mail address: cjbetti@gmail.com (C.J. Bettinger)

http://dx.doi.org/10.1016/j.biomaterials.2014.10.0780142-9612/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Mammalian cells detect and respond to topographical cues presented in natural and synthetic bio-materials both in vivo and in vitro. Micro- and nano-structures influence the adhesion, morphology,proliferation, migration, and differentiation of many phenotypes. Although the mechanisms that un-derpin celletopography interactions remain elusive, synthetic substrates with well-defined micro- andnano-structures are important tools to elucidate the origin of these responses. Substrates with recon-figurable topography are desirable because programmable cues can be harmonized with dynamiccellular responses. Here we present a lithography-free fabrication technique that can reversibly presenttopographical cues using an actuation mechanism that minimizes the confounding effects of appliedstimuli. This method utilizes strain-induced buckling instabilities in bilayer substrate materials with rigiduniform silicon oxide membranes that are thermally deposited on elastomeric substrates. The resultingsurfaces are capable of reversible of substrates between three distinct states: flat substrates(A ¼ 1.53 ± 0.55 nm; Rms ¼ 0.317 ± 0.048 nm); parallel wavy grating arrays (Ak ¼ 483.6 ± 7.8 nm; lk ¼4.78 ± 0.16 mm); perpendicular wavy grating arrays (A⊥ ¼ 429.3 ± 5.8 nm; l⊥ ¼ 4.95 ± 0.36 mm). Thecytoskeleton dynamics of 3T3 fibroblasts in response to these surfaces was measured using opticalmicroscopy. Fibroblasts cultured on dynamic substrates that are switched from flat to topographic fea-tures (FLAT-WAVY) exhibit a robust and rapid change in gross morphology as measured by a reduction incircularity from 0.30 ± 0.13 to 0.15 ± 0.08 after 5 min. Conversely, dynamic substrate sequences of FLAT-WAVY-FLAT do not significantly alter the gross steady-state morphology. Taken together, substrates thatpresent topographic structures reversibly can elucidate dynamic aspects of celletopography interactions.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Mammalian cells can detect the topography of biomaterials inboth natural and synthetic microenvironments [1e5]. Topographyplays an important role in determining the collective cell behaviorin many complex biological processes in development, woundhealing, and tissue regeneration [1,4,6e11]. Topographical cuescontrol fundamental cellular functions including adhesion, migra-tion, proliferation, and differentiation [12e16]. Most phenomeno-logical studies that correlate cell function with feature geometryand size employ substrates with static structures [7,17e25]. Staticsurfaces can extract many complex mechanisms that underpin

Engineering, Carnegie Mellon, USA..

cellematerials interactions such as contact guidance. However,there is a limit to the insight that can be gained by interrogatingdynamic systems with static cues. Substrates that present topog-raphy with spatiotemporal control are advantageous in studyingcellematerials interactions including contact guidance. They canpotentially decouple contact guidance responses from other con-founding processes such as cell attachment and spreading [5,26].

Controlled presentation of topographical cues has improvedthrough recent advances in stimuli-responsive materials [27,28]precise delivery of stimuli such as temperature changes, light, ormechanical strain [1,4,10,16,20,28,29]. This strategy has been usedfor dynamic microstructure presentation to study cellematerialsinteractions in numerous contexts [1,3,5,30]. Programmabletopography can also be engineered using stimuli-responsive poly-mers including those that respond to cues such as electric fields,temperature changes, and enzymes [3,5,27,31,32].

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172 165

The introduction of external stimuli may affect baseline meta-bolism, viability, or proliferation of cell populations [33]. Forexample, changes in temperature, the presence of enzymes, orirradiated light can impact basal cell function [3,33]. Mechanicalstimuli via direct application of strain is advantageous for topo-graphic feature formation because it is rapid, robust, and facile [1,5].Most currently available methods that use strain-induced topog-raphy require uniaxial strains of greater than 10% [1,26,34].Mammalian cells can detect strains of the underlying substrate assmall as 3.5% and respond to these stains by altering theirmorphology [35,36]. Therefore, inducing topographic features us-ing mechanical strains >3.5% may convolve contact guidance withresponses to substrate deformation. Strain-induced topographywill ideally utilizemechanical stimuli that are below the lower limitof detection for mammalian cells. Furthermore, mechanical stimulican be coupled with materials that can reversibly present homo-geneous topographical cues in various orientations. Herein wereport a lithography-free fabrication technique that is capable ofproducing strain-induced topography using sub-threshold uniaxialstrains. Cytoskeleton morphodynamics in fibroblasts are measuredusing these model surfaces.

2. Materials & methods

2.1. Fabrication and characterization of programmable dynamic topography

Elastomeric substrates were prepared using polydimethylsiloxane (PDMS, Syl-gard 184, Dow Corning, Midland, MI USA) cured in a 10:1 ratio at 75 �C. RectangularPDMS coupons (W� L� H¼ 1.5 cm � 3.5 cm � 600 mm) were mounted in a customfixation device and strained to the desired amount. Pre-strained substrates werecoatedwith 100 nm of silicon oxide (SiOx) deposited by thermal evaporation. Briefly,silicon monoxide (SiO) was thermally deposited at a rate of 1 Å/sec at 10�6 Torr(NexDep, Angstrom Engineering, Kitchener, ON Canada). All substrates were pro-cessed in an identical manner with the exception that the thickness of the bilayermembrane was set at either 10 or 100 nm. Dynamic topography sequences wereswitched between states in less than 3 s. Microstructures of bilayer membraneswere characterized using optical microscopy (Olympus BH2 microscope, OlympusAmerica Inc., Center Valley, PA USA) and atomic force microscopy (AFM, Dimension3100 SPM, Veeco, Plainville, NY USA). Fourier transforms of optical micrographswere prepared using ImageJ (National Institute of Health, USA, available at http://rsb.info.nih.gov/ij).

2.2. Fibroblast culture and imaging

All cell culture supplies were purchased from Invitrogen (Carlsbad, CA USA)unless otherwise stated. Bi-layer membranes were sterilized in ethanol (70% v/v)and irradiated with UV for 30 min. Bi-layer membranes were incubated with RGDsolution (50 mg/cm2) for 40 min and rinsed with 3x PBS. NIH 3T3 fibroblasts (ATCC,Monassas, VA USA) were seeded at densities of 25,000 cells/cm2 and incubated inDMEM medium supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S) at 37 �C. Live cell images were performed by culturing cells for24 h prior to dynamic topography sequences. Cells were cultured in en environ-mentally controlled chamber set at 37 �C with >90% relative humidity and 5% CO2

(BioImager, Mississauga, ON Canada). FB were imaged in phase contrast andanalyzed using NIH ImageJ to measure morphology and relative orientation. Cellcircularity was calculated using the following expression.

Ccell ¼4pAproj

P2Eqn. (1)

where Aproj and P are the projected surface area and perimeter of the cell, respec-tively. The axial ratio Raxis is calculated from the length of the major axis divided bythe length of the minor axis (Figure S1) [37].

Cells were fixed in 4% formaldehyde for 20 min, stained using 20 mL of AlexaFluor® 488 Phalloidin (200 U/mL), and counterstained with SlowFade® Gold Anti-fade Reagent with DAPI. Phase contrast and fluorescent images were recorded usingan EvosFLmicroscope (AdvancedMicroscopy Group, Bothell, WAUSA). FB on a staticsubstrate were also imaged using confocal fluorescence microscopy (LSM 510 METADuoScan, Carl Zeiss, Heidelberg Germany) and scanning electron microscopy (SEM)(PhilipsXL-30 FEG, FEI, Hillsboro, OR, USA). Cells dedicated for SEM imaging werefixed in 4% formaldehyde for 20 min, washed with distilled water for 3 times, anddehydrated in a series of mixtures containing ethanol and hexamethyldisilazane(HDMS), as previously described [38]. Dehydrated samples were coatedwith 4 nm ofplatinum prior to imaging (Emtech K575X, Quorum Technologies, Guelph, ON,Canada).

2.3. Statistical methods

Cell morphodynamic measurements were based on at least 100 cells per eachdata point and all experiments were repeated in triplicate. All data is shown asmean ± s. e. m. unless otherwise stated. A Student's t-test with p-value of 0.05 wasperformed to consider significant difference of two groups with statistical signifi-cance set at p-values of <0.05. One-way ANOVA with Tukey post-hoc criterion wasused to assess the significance across more than 2 groups.

3. Results and discussion

3.1. Microstructural characterization of dynamic topography

Strain-induced feature formation is a convenient strategy forrapid programmable presentation of topographic cues. Orderedbuckling is an energy-relief mechanism that occurs when a thinrigid membrane on an elastomeric substrate is compressed[39,40]. Releasing the pre-strain of bilayer substrates producesgrating arrays composed of ridge-groove features with short-rangeorder (Fig. 1) [41]. This phenomenon has been used as a non-conventional microfabrication technique [42]. Rigid silicon oxide(SiO2) membranes on polydimethylsiloxane (PDMS) substrates arecommonly fabricated by exposing PDMS to O2 plasma [43,44].PDMS-SiO2 bilayers fabricated in this manner typically requirelarge uniaxial strains (ε > 10%) to create grating arrays [43,45]. Onepossible explanation is that the minimum thickness of SiO2membranes processed using O2 plasma is large because a signifi-cant depth of the PDMS must be converted into SiO2 before astrain-sensitive percolating network of oxide structures is formedwithin the substrate. The alternative method described hereinuses thermal evaporation of SiO to deposit homogeneous SiOx(with 1.5 < x < 2) film apical to underlying PDMS substrates [46].This approach reduces the effective critical strain required forfeature formation via compression. Although the exact composi-tion of SiOx rigid membranes depends upon the deposition pro-cedure, the composition will hereby be referred to as SiO2 forsimplification.

The morphologies of PDMS-SiO2 bilayers in flat (F) and wavy(W⊥) configurations are shown in Fig. 2. Cycling the strain formsmicrostructures reversibly as indicated by optical and scanningprobe microscopy. AFM images of bilayers in flat configuration(ε ¼ 0%) contain features with characteristic peak-to-trough am-plitudes of approximately 2 nm. This feature height is an order ofmagnitude smaller than the minimum detection limit formammalian cells [20,47]. The size of the microstructure of thegrating is a strong function of the intensive mechanical propertiesof the two materials: the thickness of the membrane hf, and theamount of pre-strain εpre [48,49]. The anticipated values for averagepeak-to-through amplitude A0 and wavelength of the features l0 ingrating arrays with net compressive strains can be predicted usingthe following relationships:

l0 ¼ 2phf

�Ef3Es

�1 =

3

Eqn. (2)

A0 ¼ hf

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiεpre

εc� 1

rEqn. (3)

The parameters in each equation are as follows: hf is the thick-ness of the rigid membrane; Ef and Es are Young's moduli of sub-strate and rigid membrane, respectively; εpre is amount of uniaxialpre-strain; εc is critical buckling strain given by Ref. [49]:

εc ¼ 14

3EsEf

!2 =

3

Eqn. (4)

Fig. 1. Fabrication scheme and actuation of reconfigurable dynamic topography. a) The fabrication flow is shown: (i) PDMS substrates are strained to a prescribed amount and (ii)SiO2 membranes are deposited on the surface using thermal evaporation. (iii) The pre-strain is released to create wavy grating arrays via spontaneous buckling. b) Programmabletopography consists of three discrete strain-dependent states that depend on the difference between the pre-strain and applied strain Dε: Dε < 0% produces perpendicular wavygratings ðW⊥Þ; Dε ¼ 0% produces flat substrates ðFÞ; Dε > 0% produces parallel wavy gratings ðWkÞ. These states are shown sequentially from top to bottom as the value of Dεbecomes more positive.

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172166

A buckling pattern is established only when the applied strainexceeds εc. Values for l0 are predicted to be largely independent ofthe amount of pre-strain. However, experimentally observed valuesof l0 are reduced for uniaxial strains greater than 10% [49e51].Relationships that describe feature sized formed via spontaneousbuckling are valid if certain criteria are satisfied. The intensiveproperties and extensive geometry of the film must be such thatEf >> Ef and hf << ts where ts is the substrate thickness. The net

Fig. 2. PDMS coupons are pre-strained to εpre ¼ þ3% and SiO2 membranes of thickness hf areto characterization. SiO2 membranes with hf ¼ 100 nm generate flat ðFÞ surfaces, while SiO

strain Dε is defined as the difference between the resulting appliedstrain εapp and the pre-strain via Dε ¼ εapp e εpre. Feature formationeither by compression or tension (jεappj> εpre) yields grating fea-tures with peak-to-trough amplitudes of A > 400 nm. Topographicfeatures produced from pre-strained PDMS-SiO2 bilayer substratesin this study were consistent with constitutive relationships inEqns. (2)e(4). Releasing the pre-strain produces a net compressivestrain εapp < εpre and a uniaxial grating array that is oriented

deposited on elastomeric substrates. The pre-strain is released such that Dε ¼ �3% prior2 membranes with hf ¼ 100 nm generate substrates with perpendicular gratings ðW⊥Þ.

Fig. 3. PDMS-SiO2 bilayer substrates produce reconfigurable programmable topography. PDMS-SiO2 bilayer substrates are pre-strained to εpre ¼ 9% and coated with SiO2 layers ofhf ¼ 100 nm a) Wavelengths and amplitudes of the surface features are measured as a function of Dε ¼ εapp e εpre. Values of Dε < 0%, Dε ¼ 0% and Dε > 0% generate perpendicularwavy gratings ðW⊥Þ, featureless flat substrates ðFÞ, and parallel wavy gratings ðWkÞ features, respectively. b) Fast Fourier Transforms (FFT) of PDMS-SiO2 bilayer substrates with thefor �9% < Dε < þ9%. The intensity and relative orientation of FFT signals confirm the presence (absence) of distinguishable features and the orientation of wavy grating arrays in thefollowing discrete states: W⊥ , F , and Wk .

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172 167

orthogonally to the axis of the applied strain. Orthogonal featureshave an average amplitude of A⊥ ¼ 429.3 ± 5.8 nm and l⊥ ¼4.95 ± 0.36 mm (Fig. 2). Applying a strain εapp such that Dε ¼ 0%abolishes the grating. Tensile strains ofDε¼ εappe εpre> 3% producegrating arrays parallel to the axis of applied strain. Parallel featureshave an average amplitude of Ak ¼ 483.6 ± 7.8 nm and lk ¼4.78 ± 0.16 mm. Features emerge for strains as small as Dε ¼ jεapp eεprej ¼ 3%. Fast Fourier Transforms (FFT) of optical micrographs areshown in Fig. 3b. The intensity and orientation of these dataconfirm the presence (absence) and orientation of grating arraysunder compression, tension, and zero-strain states. The feature

Fig. 4. Fibroblast morphology and orientation on static flat (F) and static wavy gratings (Wcrographs suggest that fibroblasts were randomly oriented on static flat (F) substrates and aActin and nuclei are shown in green and blue, respectively. (For interpretation of the referarticle.)

amplitude A is also a strong function of the SiO2 membrane thick-ness hf. Optical and scanning probe micrographs of PDMS-SiO2bilayer substrates with hf ¼ 10 nm, pre-strain of εpre ¼ þ3%, and anapplied strain of Dε ¼ �3% produce random isotropic features withno short-range order and an Rms roughness of 0.317 ± 0.048 nm(Fig. 2). Strain-dependent topography can be cycled without anyobserved hysteresis in feature size. “Strain cycles could likely berepeated dozens of times without any fatigue or fracture of the SiO2membrane or mechanical failure of the PDMS substrate.” PDMS-SiO2 bilayer substrates formed by thermal evaporation of SiO2membranes offer additional advantages for use in dynamic

). Fibroblasts are fixed on PDMS-SiO2 bilayers for 24 h prior to characterization. Mi-ligned to wavy gratings (W). The direction of the grating is indicated by the red arrows.ences to colour in this figure legend, the reader is referred to the web version of this

Fig. 5. Quantification of fibroblast morphology and orientation on substrates with static flat (F) and static perpendicular gratings ðW⊥Þ. Values of circularity Ccell, projected area Aproj,axial ratio Raxis, and alignment angle qalign confirm that fibroblasts exhibit reduced spreading and increased alignment and orientation when cultured on W⊥ substrates compared toF substrates (***p < 0.001).

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172168

presentation of substrate topography. First, the minimum criticalstrain εc is reduced, which minimizes crosstalk from mechanicalstimuli during quantification of cytoskeleton remodeling [52].Second, small strains can alter the surface topography betweenthree discrete states: flat (F), orthogonal wavy ðW⊥Þ, and parallelwavy ðWkÞ. The precise selection of the SiO2 membrane permitsfacile fabrication of control substrates that are identical to dynamictopography substrates in terms of surface chemistry and appliedstrain. Lastly, topography can be presented both rapidly (~5 s) andreversibly without indirectly impacting feature geometry.

3.2. Morphological responses of fibroblasts to static topography

FB cultured on static flat (Static F) exhibit a more roundedmorphology and random spreading compared to the aligned mor-phologies of fibroblasts cultured on static grating arrays (Wavy W)as assessed by fluorescent microscopy (Fig. 4). Four parameterswere used to quantify FB morphology: circularity Ccell, ratio of

Fig. 6. Transient response of isolated fibroblast in response to PDMS-SiO2 bilayers cycled thachieve the following configurations: The strain is sequentially and quickly applied acrossDε ¼ þ3% produces perpendicular wavy grating arrays ðW⊥Þ; Dε ¼ �3% produces wavy grasubstrate topographies does not induce observable morphological changes in FB over the s

major-to-minor axes Raxis, projected area Aproj, and angle of majoraxis to grating direction qalign [37,53]. FB cultured on flat substratesexhibit a larger average cell circularity (0.30 ± 0.13 versus0.15 ± 0.08; p < 0.001) and a higher average angle between themajor axis of the cell and the features compared to topographicsubstrates (43.25� ± 23.40� versus 15.22� ± 13.10�; ***p < 0.001)(Fig. 5). These results suggest that static surfaces with gratingtopography can recapitulate the highly conserved morphologicalresponse of mammalian cells to grating features that is associatedwith contact guidance [54]. These measurements generate abaseline to compare the evolution of FB cytoskeleton morphologyon substrates with dynamic topography.

3.3. Cell culture on substrates with programmable topography

Many mammalian cell phenotypes respond to dynamic topo-graphical features through cytoskeleton reorganization [26]. Thestrain detection threshold for mammalian cells in response to

rough discrete states of substrate topography. The applied strain is applied rapidly tothe following values: Dε ¼ 0% generates flat substrates ðFÞ with no grating features;ting arrays that are parallel to the axis of applied strain ðWkÞ. The rapid switching ofpan of 30 s.

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172 169

uniaxial stretching is εcycle ¼ 3.5% as inferred by gross morpho-logical characterization [35,36]. This threshold provides a bench-mark to limit the maximum applies strains to jDεj < 3.5%.Nevertheless, applying strain in sub-threshold regimes may stillimpact other as of yet unknown aspects of cellematerials in-teractions. Therefore, several substrate topography sequences wereutilized to assess the potential impact of strains in the sub-threshold regime. Static substrates that are either Flat (F) or Wavy(W) are mapped to corresponding dynamic substrates with thefollowing sequences: FW⊥F and FW⊥. Substrates with FW⊥F se-quences are critical control conditions to measure the potential

0.4

0.3

0.2

0.1

5

4

3

2

-1 0 1 2 3 4 5 6

Circ

ular

ity, C

cell

(uni

tless

)A

xial

ratio

, Rax

is (u

nitle

ss)

Time (hrs)

-1 0 1 2 3 4 5 6Time (hrs)

F vs. FW┴FFW┴vs. FW┴F

F vs. FW┴

Circularity 0.1h 1h 2h 4h 6h

Axial ratio 0.1h 1h 2h 4h 6h

*** ***

*** ******

***

- -- -

--

***

***-

F vs. FW┴FFW┴vs. FW┴F

F vs. FW┴ ** ***

*** ****

***

- -- -

--

***

***-

Fig. 7. Cytoskeleton morphodynamics for FB cultured on substrates over the course of 6 h(FW⊥), and dynamic flat to wavy grating back to flat (FW⊥F). Results of one-way ANOVA w

impact of the transient strain on the downstream cytoskeletonmorphodynamics. Furthermore, a direct comparison between se-quences of FW⊥F and FW⊥ can isolate the direct impact of dynamictopography presentation on cytoskeleton morphology. FBmorphology was largely preserved immediately after appliedstrains ofDε¼ ±3% (Fig. 6). The orthogonal and parallel axes refer tothe relative orientation of the grating features to the direction ofapplied strain. The morphology of FB cultured on PDMS substratescoated with 10 nm SiO2 membranes was measured for t � 6 h(Fig. 7). PDMS substrates with SiO2 membranes of hf ¼ 10 nm,εpre ¼ þ3%, and Dε ¼ �3% exhibit random gratings with feature

Ang

le, θ

alig

n (d

eg.)

Are

a, A

proj

(μm

2 )

-1 0 1 2 3 4 5 6Time (hrs)

-1 0 1 2 3 4 5 6Time (hrs)

1400

1200

1000

800

600

400

60

50

40

30

20

10

Flat (F)

Flat to Wavy (FW┴)

Flat to Wavy to Flat (FW┴F)

Area 0.1h 1h 2h 4h 6h

Angle 0.1h 1h 2h 4h 6h

F vs. FW┴FFW┴vs. FW┴F

F vs. FW┴ - **

*** ***-

***

*** ****** **

*-

***

-***

F vs. FW┴FFW┴vs. FW┴F

F vs. FW┴ - ***

- **-

***

*** ***- -

******

***-

***

with the following sequences are shown: Static flat (F), dynamic flat to wavy gratingith Tukey post-hoc tests are summarized in the table.

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172170

sizes of A ¼ 1.53 ± 0.55 nm and l ¼ 1.8 ± 0.1 mm. These features donot significantly alter FB morphology via contact guidance mech-anisms since the feature heights are smaller than 35 nm, the pre-viously reported detection limit for contact guidance [55]. PDMSsubstrates with 10 nm SiO2 membranes serve as valuable controlmaterials since the strain and chemistry of the substrate are iden-tical to PDMS-SiO2 bilayers that with thicker SiO2 membranes.Maintaining constant material properties in dynamic substratescan be challenging [5]. Thin film deposition and precise strainsachieve dynamic programmable topography while preservingmany of the other physicochemical aspects of the cellularmicroenvironment.

After 24 h of cell seeding (t ¼ 0 h), the surface topography isswitched either from flat to wavy ðFW⊥Þ or flat to wavy to flatðFW⊥FÞ in rapid succession. The dynamic presentation of substratetopography induces alterations in FB morphology that initiatewithin 6 min and persist for t > 6 h (Fig. 7). The circularity and ratioof major/minor axes of the FW⊥ both exhibit a significant changewithin 5 min of substrate actuation (**p < 0.01), while the averageprojected area A andmajor axis angle exhibit significant differenceswithin 1 h (**p < 0.001). Substrates with FW⊥F programs presentgratings briefly (5 ± 3 s). FB cultured on FW⊥F programs exhibitcytoskeleton dynamics that are substantially different compared toStatic F substrates. FW⊥F sequences induce transient alterations inthe following parameters that are associated with elongated mor-phologies: circularity Ccell, projected area Aproj, and axial ratio Raxis.Values for Ccell and Raxis in FB populations cultured on substrateswith FW⊥F sequences converge to values observed in cells on StaticF substrates for t � 2 h. These results suggest that sub-thresholdstrains induce transient fluctuations in morphological response.Temporary presentation of topographical cues may engage mem-ory mechanisms in mammalian cells that have been previouslyreported in the context of other types of dynamic cellematerialsinteractions [56].

Substrates with FW⊥ sequences yield morphological changes inFB that are internally consistent with other observations within thisstudy. Substrate actuation via compressive strains is comparablebetween FW⊥ and FW⊥F sequences. FB cultured on substrates withFW⊥ sequences adopt a temporary rounded morphology that is

Fig. 8. Morphodynamics of a single FB in response to substrate perturbations. Single cells arPhase contrast micrographs are processed using NIH ImageJ to depict changes of cellular shaperpendicular wavy grating (FW⊥); (b) flat to perpendicular wavy grating back to flat subs

manifested by a transient decrease in Aproj for t¼ 1 h (***p < 0.001),a decrease in Ccell at t ¼ 0.1 h (***p < 0.001), and an increase in Raxisat t ¼ 0.1 h (**p < 0.01) compared to Static F substrates (Fig. 7). Themorphology of FB cultured on substrates with FW⊥ sequences fort� 6 h compared to Static F substrates approximate that of StaticWcompared to Static F. These results suggest that steady-state FBmorphology is achieved within 6 h.

3.4. Single cell sensing on programmable substrates

Programmable substrates can measure the cytoskeletonmorphology of individual FB to transiently altered strain-inducedtopography. Isolated FB alter their shape and orientation fort � 1 h. Single cell morphological responses observed in FB pop-ulations cultured on substrates with FW⊥ and FW⊥F sequences areshown in Fig. 8. FB alignment and elongation occurs for t < 2 h forFW⊥ sequences. FB cultured on substrates with FW⊥F sequencesextend lamellipodia in seemingly random directions within t < 2 h.Single-cell morphodynamics are consistent with morphologicalobservations made in FB populations. A notable observation is thatboth FW⊥ and FW⊥F sequences induce an initial temporary phase ofcell rounding immediately after substrate actuation. This observa-tion is curious because the minimum apparent strain detectionthreshold during cyclic stretching is εcycle ¼ 3.5%. Furthermore, thestrains applied in this system are approximately 1 order ofmagnitude smaller than other experiments that employ dynamicstrain. Typical cyclic strain experiments require uniaxial substratestrains of εcycle > 10% [57]. Furthermore, the impact of mechanicalstress on cytoskeleton dynamics typically results in strains of g

~20% [58,59]. It is unlikely that the application of transient strainswill directly impact binding between integrins and adhesion-promoting peptides because of the strength of these bonds [60].Typical dissociation constants for this specific kind of recep-toreligand interaction is on the order of KD ¼ 10�9 M, which cor-responds to a characteristic binding distance of r0 ¼ 3 nm [60].Considering the characteristic dimension of integrins is ~10 nm[61], it is likely that 3% strains will not dissociate integrin-peptidecoupling. Dynamic topography presentation may, however, alterother downstream effectors of cellematrix interactions. Transient

e captured before and after topographical switching, indicated at zero time point (:00).pe and orientation. Substrates are programmed with the following sequences: (a) flat totrates (FW⊥F).

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172 171

substrate strains may disrupt focal adhesions or cytoskeleton pro-teins that induce an altered state in which the cell rapidly in-terrogates the local chemical and topographicalmicroenvironments by forming nascent integrins [62]. This topic isthe subject of ongoing research.

4. Conclusions

We describe a lithography-free approach to fabricate dynamicsubstrate topography for measuring morphodynamics of cellsduring contact guidance. Dynamic substrates are advantageousbecause they present ordered topographical cues in a manner thatdecouples potential contributions from other stimuli includingsubstrate chemistry and large mechanical strains. Homogeneoussurface chemistry (SiO2) and small required extensive strains (~3%)are key features of the approach described herein. Although thesubstrate cannot completely eliminate the effect of external me-chanical stimuli on in vitro celletopography interactions, usingsmall strains minimizes the impact of the applied stimulus. Thiswork represents a strategy to better isolate the effect of dynamicprogrammable topography on living cells. Furthermore, this tech-nique permits rapid and reversible presentation of topographiccues with precise temporal precision. Dynamic substrate topog-raphy afforded by the strategy described herein could be furtherapplied to studying signal transduction pathways that original fromtransient cellematrix interactions and propagate to downstreampathways that control cytoskeleton reorganization. Thus, pro-grammable substrate topography is a promising strategy to eluci-date cytoskeleton remodeling dynamics in many cell phenotypes inthe context of cellebiomaterials interactions.

Acknowledgement

The authors would like to acknowledge the following people fortheir help: Stephanie Chang and Prof. Yu-li Wang for providing theNIH 3T3 cells; Patrick Boyer and Prof. Kris Dahl for providing theliquid nitrogen storage tank; Haibing Teng, Margaret Fuhrman, andAlan Waggoner (Director) of MBIC/CMU for use of the facilities andtechnical assistance. The authors would also like to acknowledgefunding provided by the National Institutes of Health (NIBIBR21RB015165).

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.biomaterials.2014.10.078.

References

[1] Cheng C-M, Steward RL, LeDuc PR. Probing cell structure by controlling themechanical environment with cell-substrate interactions. J Biomech 2009;42:187e92. http://dx.doi.org/10.1016/j.jbiomech.2008.10.014.

[2] Dalby MJ, Riehle MO, Sutherland DS, Agheli H, Curtis ASG. Changes in fibro-blast morphology in response to nano-columns produced by colloidallithography. Biomaterials 2004;25:5415e22.

[3] Davis KA, Burke KA, Mather PT, Henderson JH. Dynamic cell behavior on shapememory polymer substrates. Biomaterials 2011;32:2285e93. http://dx.doi.org/10.1016/j.biomaterials.2010.12.006.

[4] Guvendiren M, Burdick JA. Engineering synthetic hydrogel microenviron-ments to instruct stem cells. Curr Opin Biotechnol 2013;24:841e6. http://dx.doi.org/10.1016/j.copbio.2013.03.009.

[5] Kim J, Hayward RC. Mimicking dynamic in vivo environments with stimuli-responsive materials for cell culture. Trends Biotechnol 2012;30:426e39.http://dx.doi.org/10.1016/j.tibtech.2012.04.003.

[6] Spivey EC, Khaing ZZ, Shear JB, Schmidt CE. The fundamental role of subcel-lular topography in peripheral nerve repair therapies. Biomaterials 2012;33:4264e76. http://dx.doi.org/10.1016/j.biomaterials.2012.02.043.

[7] Nikkhah M, Edalat F, Manoucheri S, Khademhosseini A. Engineering micro-scale topographies to control the cellesubstrate interface. Biomaterials2012;33:5230e46.

[8] Kim HN, Jiao A, Hwang NS, Kim MS, Kang DH, Kim D-H, et al. Nano-topography-guided tissue engineering and regenerative medicine. Adv DrugDeliv Rev 2013;65:536e58. http://dx.doi.org/10.1016/j.addr.2012.07.014.

[9] Franco D, Milde F, Klingauf M, Orsenigo F, Dejana E, Poulikakos D, et al.Accelerated endothelial wound healing on microstructured substrates underflow. Biomaterials 2013;34:1488e97. http://dx.doi.org/10.1016/j.biomaterials.2012.10.007.

[10] Leclerc A, Tremblay D, Hadjiantoniou S, Bukoreshtliev NV, Rogowski JL,Godin M, et al. Three dimensional spatial separation of cells in response tomicrotopography. Biomaterials 2013;34:8097e104. http://dx.doi.org/10.1016/j.biomaterials.2013.07.047.

[11] Tzvetkova-Chevolleau T, St�ephanou A, Fuard D, Ohayon J, Schiavone P,Tracqui P. The motility of normal and cancer cells in response to the combinedinfluence of the substrate rigidity and anisotropic microstructure. Biomaterials2008;29:1541e51. http://dx.doi.org/10.1016/j.biomaterials.2007.12.016.

[12] Kolind K, Kraft D, Bøggild T, Duch M, Lovmand J, Pedersen FS, et al. Control ofproliferation and osteogenic differentiation of human dental-pulp-derivedstem cells by distinct surface structures. Acta Biomater 2014;10:641e50.http://dx.doi.org/10.1016/j.actbio.2013.11.006.

[13] Han L, Mao Z, Wu J, Guo Y, Ren T, Gao C. Unidirectional migration of singlesmooth muscle cells under the synergetic effects of gradient swelling cue andparallel groove patterns. Colloids Surf B Biointerfaces 2013;111C:1e6. http://dx.doi.org/10.1016/j.colsurfb.2013.05.011.

[14] Sheets K, Wunsch S, Ng C, Nain AS. Shape-dependent cell migration and focaladhesion organization on suspended and aligned nanofiber scaffolds. ActaBiomater 2013;9:7169e77. http://dx.doi.org/10.1016/j.actbio.2013.03.042.

[15] Seo CH, Jeong H, Furukawa KS, Suzuki Y, Ushida T. The switching of focaladhesion maturation sites and actin filament activation for MSCs by topog-raphy of well-defined micropatterned surfaces. Biomaterials 2013;34:1764e71. http://dx.doi.org/10.1016/j.biomaterials.2012.11.031.

[16] Ferrari A, Cecchini M, Serresi M, Faraci P, Pisignano D, Beltram F. Neuronalpolarity selection by topography-induced focal adhesion control. Biomaterials2010;31:4682e94. http://dx.doi.org/10.1016/j.biomaterials.2010.02.032.

[17] Walboomers XF, Monaghan W, Curtis AS, Jansen JA. Attachment of fibroblastson smooth and microgrooved polystyrene. J Biomed Mater Res 1999;46:212e20. http://dx.doi.org/10.1002/(SICI)1097-4636(199908)46. 2<212::AID-JBM10>3.0.CO;2-Y.

[18] Koo S, Muhammad R, Peh GSL, Mehta JS, Yim EKF. Micro- and nano-topography with extracellular matrix coating modulate human cornealendothelial cell behavior. Acta Biomater 2014;10:1975e84. http://dx.doi.org/10.1016/j.actbio.2014.01.015.

[19] Zhao G, Raines AL, Wieland M, Schwartz Z, Boyan BD. Requirement for bothmicron- and submicron scale structure for synergistic responses of osteoblaststo substrate surface energy and topography. Biomaterials 2007;28:2821e9.http://dx.doi.org/10.1016/j.biomaterials.2007.02.024.

[20] Biela SA, Su Y, Spatz JP, Kemkemer R. Different sensitivity of human endo-thelial cells, smooth muscle cells and fibroblasts to topography in the nano-micro range. Acta Biomater 2009;5:2460e6. http://dx.doi.org/10.1016/j.actbio.2009.04.003.

[21] Flemming RG, Murphy CJ, Abrams GA, Goodman SL, Nealey PF. Effects ofsynthetic micro- and nano-structured surfaces on cell behavior. Biomaterials1999;20:573e88.

[22] Lü D, Luo C, Zhang C, Li Z, Long M. Differential regulation of morphology andstemness of mouse embryonic stem cells by substrate stiffness and topog-raphy. Biomaterials 2014;35:3945e55. http://dx.doi.org/10.1016/j.biomaterials.2014.01.066.

[23] Davidson P, Bigerelle M, Bounichane B, Giazzon M, Anselme K. Definition of asimple statistical parameter for the quantification of orientation in two di-mensions: application to cells on grooves of nanometric depths. Acta Bio-mater 2010;6:2590e8. http://dx.doi.org/10.1016/j.actbio.2010.01.038.

[24] Watari S, Hayashi K, Wood JA, Russell P, Nealey PF, Murphy CJ, et al. Modu-lation of osteogenic differentiation in hMSCs cells by submicrontopographically-patterned ridges and grooves. Biomaterials 2012;33:128e36.http://dx.doi.org/10.1016/j.biomaterials.2011.09.058.

[25] Chua JS, Chng C-P, Moe AAK, Tann JY, Goh ELK, Chiam K-H, et al. Extendingneurites sense the depth of the underlying topography during neuronal dif-ferentiation and contact guidance. Biomaterials 2014;35:7750e61. http://dx.doi.org/10.1016/j.biomaterials.2014.06.008.

[26] Lam MT, Clem WC, Takayama S. Reversible on-demand cell alignment usingreconfigurable microtopography. Biomaterials 2008;29:1705e12. http://dx.doi.org/10.1016/j.biomaterials.2007.12.010.

[27] Yang P, Baker RM, Henderson JH, Mather PT. In vitro wrinkle formation viashape memory dynamically aligns adherent cells. Soft Matter 2013;9:4705.http://dx.doi.org/10.1039/c3sm00024a.

[28] Kloxin AM, Tibbitt MW, Anseth KS. Synthesis of photodegradable hydrogels asdynamically tunable cell culture platforms. Nat Protoc 2010;5:1867e87.http://dx.doi.org/10.1038/nprot.2010.139.

[29] Straley KS, Heilshorn SC. Dynamic, 3D-Pattern formation within Enzyme-responsive hydrogels. Adv Mater 2009;21:4148e52. http://dx.doi.org/10.1002/adma.200901865.

[30] Hui EE, Bhatia SN. Micromechanical control of cellecell interactions. Proc NatlAcad Sci 2007;104:5722e6. http://dx.doi.org/10.1073/pnas.0608660104.

[31] Lamb BM, Yousaf MN. Redox-switchable surface for controlling peptidestructure. J Am Chem Soc 2011;133:8870e3. http://dx.doi.org/10.1021/ja203198y.

P. Pholpabu et al. / Biomaterials 39 (2015) 164e172172

[32] Idota N, Tsukahara T, Sato K, Okano T, Kitamori T. The use of electron beamlithographic graft-polymerization on thermoresponsive polymers for regu-lating the directionality of cell attachment and detachment. Biomaterials2009;30:2095e101. http://dx.doi.org/10.1016/j.biomaterials.2008.12.058.

[33] Wong JY, Leach JB, Brown XQ. Balance of chemistry, topography, and me-chanics at the cellebiomaterial interface: issues and challenges for assessingthe role of substrate mechanics on cell response. Surf Sci 2004;570:119e33.http://dx.doi.org/10.1016/j.susc.2004.06.186.

[34] Kaunas R, Nguyen P, Usami S, Chien S. Cooperative effects of Rho and me-chanical stretch on stress fiber organization. Proc Natl Acad Sci U S A2005;102:15895e900. http://dx.doi.org/10.1073/pnas.0506041102.

[35] Winston FK, Macarak EJ, Gorfien SF, Thibault LE. A system to reproduce andquantify the biomechanical environment of the cell. J Appl Physiol BethesdaMd 1985;1989(67):397e405.

[36] Song G, Ju Y, Shen X, Luo Q, Shi Y, Qin J. Mechanical stretch promotes pro-liferation of rat bone marrow mesenchymal stem cells. Colloids Surf B Bio-interfaces 2007;58:271e7. http://dx.doi.org/10.1016/j.colsurfb.2007.04.001.

[37] Den Braber ET, de Ruijter JE, Ginsel LA, von Recum AF, Jansen JA. Quantitativeanalysis of fibroblast morphology on microgrooved surfaces with variousgroove and ridge dimensions. Biomaterials 1996;17:2037e44. http://dx.doi.org/10.1016/0142-9612(96)00032-4.

[38] Slízov�a D, Krs O, Pospísilov�a B. Alternative method of rapid drying vascularspecimens for scanning electron microscopy. J Endovasc Ther off J Int SocEndovasc Spec 2003;10:285e7. http://dx.doi.org/10.1583/1545-1550(2003)010. <0285:AMORDV>2.0.CO;2.

[39] Bowden N, Brittain S, Evans AG, Hutchinson JW, Whitesides GM. Spontaneousformation of ordered structures in thin films of metals supported on anelastomeric polymer393; 1998. p. 146e9. http://dx.doi.org/10.1038/30193.

[40] Box F, Bowman R, Mullin T. Dynamic compression of elastic and plasticcellular solids. Appl Phys Lett 2013;103:151909. http://dx.doi.org/10.1063/1.4824845.

[41] Wu H, Kustra S, Gates EM, Bettinger CJ. Topographic substrates as strain relieffeatures in stretchable organic thin film transistors. Org Electron 2013;14:1636e42. http://dx.doi.org/10.1016/j.orgel.2013.02.037.

[42] Quero JM, Perdigones F, Aracil C. Smart sensors and mems. Elsevier; 2014.http://dx.doi.org/10.1533/9780857099297.2.281.

[43] Yu C, Jiang H. Forming wrinkled stiff films on polymeric substrates at roomtemperature for stretchable interconnects applications. Thin Solid Films2010;519:818e22. http://dx.doi.org/10.1016/j.tsf.2010.08.106.

[44] B�efahy S, Lipnik P, Pardoen T, Nascimento C, Patris B, Bertrand P, et al.Thickness and elastic modulus of plasma treated PDMS silica-like surfacelayer. Langmuir ACS J Surf Colloids 2010;26:3372e5. http://dx.doi.org/10.1021/la903154y.

[45] Shih T-K, Ho J-R, Liao H-Y, Chen C-F, Liu C-Y. Fabrication of optical gratings byshrinkage of a rubber material. Thin Solid Films 2008;516:5339e43. http://dx.doi.org/10.1016/j.tsf.2007.07.092.

[46] Barranco A, Yubero F, Espin�os JP, Groening P, Gonz�alez-Elipe AR. Electronicstate characterization of SiOx thin films prepared by evaporation. J Appl Phys2005;97:113714. http://dx.doi.org/10.1063/1.1927278.

[47] Dalby MJ, Riehle MO, Johnstone H, Affrossman S, Curtis ASG. Investigating thelimits of filopodial sensing: a brief report using SEM to image the interaction

between 10 nm high nano-topography and fibroblast filopodia. Cell Biol Int2004;28:229e36. http://dx.doi.org/10.1016/j.cellbi.2003.12.004.

[48] Ohzono T, Monobe H. Microwrinkles: shape-tunability and applications.J Colloid Interface Sci 2012;368:1e8. http://dx.doi.org/10.1016/j.jcis.2011.11.075.

[49] Jiang H, Khang D-Y, Song J, Sun Y, Huang Y, Rogers JA. Finite deformationmechanics in buckled thin films on compliant supports. Proc Natl Acad Sci U SA 2007;104:15607e12. http://dx.doi.org/10.1073/pnas.0702927104.

[50] Song J, Jiang H, Huang Y, Rogers JA. Mechanics of stretchable inorganic elec-tronic materials. J Vac Sci Technol A 2009;27:1107e25. http://dx.doi.org/10.1116/1.3168555.

[51] Harrison C, Stafford CM, Zhang W, Karim A. Sinusoidal phase grating createdby a tunably buckled surface. Appl Phys Lett 2004;85:4016e8. http://dx.doi.org/10.1063/1.1809281.

[52] Houtchens GR, Foster MD, Desai TA, Morgan EF, Wong JY. Combined effects ofmicrotopography and cyclic strain on vascular smooth muscle cell orientation.J Biomech 2008;41:762e9. http://dx.doi.org/10.1016/j.jbiomech.2007.11.027.

[53] Hakkinen KM, Harunaga JS, Doyle AD, Yamada KM. Direct comparisons of themorphology, migration, cell adhesions, and actin cytoskeleton of fibroblasts infour different three-dimensional extracellular matrices. Tissue Eng Part A2011;17:713e24. http://dx.doi.org/10.1089/ten.TEA.2010.0273.

[54] Van Delft FCMJM, van den Heuvel FC, Loesberg WA, te Riet J, Sch€on P,Figdor CG, et al. Manufacturing substrate nano-grooves for studying cellalignment and adhesion. Microelectron Eng 2008;85:1362e6. http://dx.doi.org/10.1016/j.mee.2008.01.028.

[55] Dalby MJ, Riehle MO, Johnstone HJH, Affrossman S, Curtis ASG. Polymer-demixed nanotopography: control of fibroblast spreading and proliferation.Tissue Eng 2002;8:1099e108. http://dx.doi.org/10.1089/107632702320934191.

[56] Yang C, Tibbitt MW, Basta L, Anseth KS. Mechanical memory and dosing in-fluence stem cell fate. Nat Mater 2014;13:645e52. http://dx.doi.org/10.1038/nmat3889.

[57] Morioka M, Parameswaran H, Naruse K, Kondo M, Sokabe M, Hasegawa Y,et al. Microtubule dynamics regulate cyclic stretch-induced cell alignment inhuman Airway smooth muscle cells. PLoS One 2011;6. http://dx.doi.org/10.1371/journal.pone.0026384. e26384.

[58] Zhu X, Mills KL, Peters PR, Bahng JH, Liu EH, Shim J, et al. Fabrication ofreconfigurable protein matrices by cracking. Nat Mater 2005;4:403e6. http://dx.doi.org/10.1038/nmat1365.

[59] Ehrlicher AJ, Nakamura F, Hartwig JH, Weitz DA, Stossel TP. Mechanical strainin actin networks regulates FilGAP and integrin binding to filamin A. Nature2011;478:260e3. http://dx.doi.org/10.1038/nature10430.

[60] Lauffenburger DA. Receptors: models for binding, trafficking, and signaling.New York: Oxford University Press; 1996.

[61] Comisar WA, Mooney DJ, Linderman JJ. Integrin organization: linking adhe-sion ligand nanopatterns with altered cell responses. J Theor Biol 2011;274:120e30. http://dx.doi.org/10.1016/j.jtbi.2011.01.007.

[62] Beningo KA, Dembo M, Kaverina I, Small JV, Wang Y. Nascent focal adhesionsare responsible for the generation of Strong propulsive forces in migratingfibroblasts. J Cell Biol 2001;153:881e8. http://dx.doi.org/10.1083/jcb.153.4.881.