Development of bioactive photocrosslinkable fibrous hydrogels

J. S. Stephens-Altus, P. Sundelacruz, M. L. Rowland, J. L. West

Department of Bioengineering, Rice University, 6500 Main Street, Houston, Texas

Received 10 June 2010; revised 3 December 2010; accepted 14 February 2011

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.33095

Abstract: Three-dimensional (3D) fibrous hydrogels were fab-

ricated by blending two photoactive polymers, poly(ethylene

glycol) diacrylate (PEGDA) and poly(vinyl alcohol) (PVA), and

the resulting solution was electrospun. PEGDA is a com-

monly used hydrogel material for tissue engineering applica-

tions since its interaction with cells can be tuned by

crosslinking in a variety of bioactive molecules including pep-

tides and proteins. The PVA in these materials aids in fiber

formation and stabilizes the fibrous network when hydrated.

The average dry fiber diameter in the hydrogels was 1.02 lmand upon swelling, the fiber diameter increased approxi-

mately six-fold. Fibers were stable under cell culture condi-

tions for up to 5 days. The adhesive ligand, RGDS, was

readily incorporated into the fibrous network via the conjuga-

tion of RGDS to PEG-monoacrylate which was then cross-

linked with the fibers. The bioactivity of the fibrous hydrogels

was compared with peptide-modified PEGDA-based hydro-

gels. The two hydrogel materials had similar cell adhesion

and viability. Cell morphology on the fibrous hydrogels was

dendritic showing a more in vivo like representation, as com-

pared to spread cell morphology on the PEGDA gels. The

ability to generate 3D fibrous architectures in hydrogel sys-

tems opens up new areas of investigation in cell-material

interactions and tissue formation. VC 2011 Wiley Periodicals, Inc.

J Biomed Mater Res Part A: 00A: 000–000, 2011.

Key Words: PEGDA, PVA, electrospinning, hydrogel, fibrous

architecture

INTRODUCTION

Hydrogels are a class of biomaterials widely used for tissueengineering, regenerative medicine, drug delivery, and to de-velop a more in depth knowledge of cell biology.1–5 Thesematerials are popular because of their biocompatibility, highwater content, and favorable diffusive properties.1–3 Hydro-gels can be divided into two main classes: natural and syn-thetic. Commonly used natural hydrogels include collagen,hyaluronic acid, alginate, and chitosan.1–3,6 The chemicalproperties of natural hydrogels are similar to native tissue;however their physical properties are usually much weaker.1,7

The use of synthetic materials is advantageous because itallows for the specific control over mechanical, chemical, andphysical properties and ensures reproducibility. Poly(ethyleneglycol) (PEG) and poly(vinyl alcohol) (PVA) are two examplesof synthetic hydrogel materials.3,8–11 PEG and PVA are of par-ticular interest because they have extremely low nonspecificprotein adsorption and therefore little or no cellular interac-tions2,8,9,11,12 in their unmodified state. This is a characteristicof PEG- and PVA-based hydrogels that has been taken advant-age of in the design of synthetic extracellular matrix (ECM)mimetic biomaterials. Towards the goal of designing a syn-thetic ECM, adhesion specific peptides and other bioactivemolecules have been incorporated into these hydrogelsystems via physical mixing or covalent crosslinking to tailorspecific biological features for a desired application.3,9,10

PEG-based hydrogels that mimic different features of tis-sue-specific ECM have been investigated extensively in thelast decade. Biomimetic PEG-based hydrogels can be thoughtof as a modular system in which a variety of parameters canbe varied independently or in combination without having tocontinuously redesign the system as a whole. PEG-basedhydrogels are formed by photoactively crosslinking the reac-tive end groups of polyethylene glycol diacrylate (PEGDA).3,11

Figure 1(A) illustrates the acrylation scheme used to createPEGDA from PEG. To form hydrogel materials, PEGDA iscrosslinked under long wavelength ultra violet (UV) or visiblelight in the presence of the appropriate photoinitiator, withcrosslinking conditions mild enough for the reaction to becarried out in the presence of cells.13–15 In addition, differentbiologically relevant features can be added to these hydrogelsby conjugating peptides or proteins to monoacrylated PEG[Fig. 1(B)] and incorporating them in the crosslinking reac-tion. For example, PEG hydrogels can be rendered cell-adhesive via incorporation of known adhesion ligands. Theadhesive ligands can be ubiquitous, such as the peptide Arg-Gly-Asp-Ser (RGDS), or cell type selective.10,16,17 Regulatorymolecules, such as growth factors and ephrins, can also beintegrated into the hydrogel matrix thereby regulating notonly cell adhesion but also cell function.18–25 In a similarmanner, the hydrogels can be modified with enzymaticallydegradable sequences to facilitate hydrogel degradation by

Correspondence to: J. L. West; e-mail: jwest@rice.edu

Contract grant sponsor: NIH-BRP

Contract grant sponsor: NSF graduate fellowship program

Contract grant sponsor: HHMI Professor’s Program

VC 2011 WILEY PERIODICALS, INC. 1

cellular proteases.2,15,20,26 Finally, hydrogel mechanical prop-erties can be tailored by varying either the molecular weight(MW) of PEG used to form the reactive monomers and/or theprepolymer solution concentration.14,16

Building on the utility of this ECM-mimetic hydrogel sys-tem, the goal of the current investigation was to develop ahydrogel matrix modeled after the three dimensional (3D)fibrous architecture of this natural milieu.7,27,28 The ECM isa complex structure that provides physical, chemical, andmechanical cues to direct cell migration, differentiation, cell-cell interactions, and hieratical tissue assembly.29–31 In tis-sue engineering and regenerative medicine, the synergisticeffect of materials and their 3D architecture has similarlybeen established.29,32 The use of 3D culture environments,for example, has been shown to greatly influence cell prolif-eration, migration, and differentiation in a variety of investi-gations to create in vivo-like cytoskeletal organization andgene expression.29,33–36

Electrospinning was used to create the fibrous hydrogelsdeveloped in this investigation. Briefly, electrospinning is aprocess that generates micron, submicron, and nanometerdiameter polymer fibers.37,38 An electric charge (5�30 kV)is applied to a polymer solution of the appropriate MW andviscosity. As the electric charge overcomes the surface ten-sion of the solution, a polymer jet is ejected from the solu-tion. The jet travels a set distance (5�30 cm) to a counterelectrode where it is collected in the form of fibers [Fig.2(B)]. To generate fibers from a PEGDA-based solution, ahigh MW polymer is added to create chain-chain entangle-ments and ensure fiber formation. This is a common tech-nique in electrospinning, where the high MW polymer isreferred to as a carrier.39–41 For example, poly(ethylene ox-ide) is a commonly used carrier that is later removed bywater extraction, often resulting in weakened fibers.40,42 In

this work, a photoactive PVA (130 kDa) has been incorpo-rated into the PEGDA solution to facilitate fiber formation[Fig. 2(A)]. In addition to serving as a carrier, PVA also aidsin long-term fiber stability. Each PVA polymer chain has�45 photoactive methacrylate groups along its backbone[Fig. 2(A)], which can crosslink with the PEGDA to form astable fibrous network in solution. Similar photoactive PVAhas been used in contact lenses, drug delivery systems, andtissue engineering constructs.8,9,11

In this investigation, we evaluated the ability to form fi-brous hydrogels of a PEG- and PVA-based composite and tomaintain this fibrous morphology under cell culture condi-tions. The fibrous hydrogels were compared to traditionalPEGDA-based hydrogels by examining the incorporation ofbioactive molecules and the effect of the fibrous architectureon cell morphology, attachment, and viability.

MATERIALS AND METHODS

Materials were obtained from Sigma-Aldrich unless other-wise noted.

Poly(ethylene glycol) diacrylate (PEGDA) synthesis andpurificationThe synthesis of photoactive PEGDA is illustrated in Figure1(A).3 Briefly, 0.1 mmol/mL PEG (MW 3.4 kDa or 10 kDa),0.4 mmol/mL acryloyl chloride, and 0.2 mmol/mL triethyl-amine were dissolved in anhydrous dichloromethane andstirred overnight under argon. The solution was washed in2M K2CO3 and separated into organic aqueous phases toremove the hydrochloric acid. The organic phase was driedwith magnesium sulfate (MgSO4) and the PEGDA was pre-cipitated in cold diethyl ether, filtered, and dried undervacuum.

FIGURE 1. Schematic of (A) PEGDA synthesis, (B) monoacrylate PEG-RGDS conjugation, and (C) PEGDA hydrogel polymerization. [Color figure

can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

2 STEPHENS-ALTUS ET AL. DEVELOPMENT OF BIOACTIVE PHOTOCROSSLINKABLE FIBROUS HYDROGELS

Monoacryloyl-PEG-RGDS synthesisThe peptide RGDS (American Peptide) and polymer acry-late-PEG-succinimidyl carboxymethyl (MW 3400, PEG-SCM,Laysan Bio.) were dissolved in dimethyl sulfoxide and mixedat a 1.2:1 ratio [Fig. 1(B)].19 Diisopropylamine was added ata 2:1 ratio to PEG-SCM and the reaction was run overnightat room temperature on a rocker plate. The resulting solu-tion was diluted with MilliQ water to double the volumeand then dialyzed to remove unconjugated reactants. Afterdialysis the acrylate-PEG-RGDS (PEG-RGDS) was lyophilizedand stored frozen until use.

Fluorescently labeled acrylate-PEG-RGDS synthesisA portion of the PEG-RGDS was fluorescently labeled withAlexa Fluor and used to evaluate the morphology of theswollen hydrogel fibers. PEG-RGDS and Alexa Fluor 546 car-boxylic acid, succinimidyl ester (Molecular Probes, Invitro-gen)19 were dissolved in 50 mM sodium bicarbonate buffer(pH 5) at �10 mole dye per mole PEG-RGDS and reactedfor 1 h at room temperature. The product was purified viadialysis and lyophilized.

PEGDA hydrogel polymerizationTwo sets of PEGDA-based hydrogels were investigated: a con-trol set containing only PEGDA, and a bioactive set of PEGDAplus PEG-RGDS. For all hydrogels, 10 kDA PEGDA was dis-solved in HEPES buffered saline (HBS, pH 7.4) to make a 10%(w/v) solution. In the bioactive sample set, 2.5 mM PEG-RGDS was also included. The photoinitiator, 2,2 dimethoxy-2-phenylacetophenone in 1-vinylpyrrolidone (acetophenone inNVP, 300 mg/mL), was added to each solution at 10 lL/mL,and the solutions were mixed and sterilized via filtration(0.2 lm). The PEGDA solutions were injected between twoglass slides that were separated by a 0.5 mm Teflon spacerand held together with clips. The gels were polymerized uponexposure to UV light (B-200SP UV lamp, UVP, 365 nm,10 mW/cm2) for 60 s. After crosslinking, the gels werepunched into 6.75 mm and 8 mm diameter samples for analy-sis of RGDS incorporation and cell seeding, respectively.

Electrospun hydrogelsPVA (MW 134,000) was provided by BioCure, Inc. (Norcross,GA) after a reaction with methacrylamidoacetaldhyde

FIGURE 2. Illustrates the electrospinning approach used to make the fibrous hydrogels; (A) idealized chemical structures of PVA and PEGDA, (B)

schematic of the electrospinning apparatus and the initial crosslinking steps post electrospinning, and (C) shows the functionalization with the

RGDS biomolecule and the final crosslinking step to crosslink the peptide to the fibers. [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com.]

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2011 VOL 00A, ISSUE 00 3

dimethyl acetal to generate photocrosslinkable side groups(degree of modification ¼ 0.45 Meg/g).8 See Figure 2(A) foridealized structure. The photoactive PVA was dissolved in70% ethanol to make a 5% (w/v) solution. The solutionwas vortexed for 5 min, sonicated for 20 min, and allowedto sit overnight for complete dissolution of the PVA. Next,the 3.4 kDa PEGDA was added at 30% (w/v), and the solu-tion was vortexed for 5 min and sonicated for 20 min todissolve the PEGDA. The photoinitiator, acetophenone inNVP (300 mg/mL), was added at 10 lL/mL, and the solu-tion was then transferred to a syringe with a 21-gauge nee-dle for electrospinning.

Two sets of electrospun fibrous hydrogels were investi-gated: the controls consisted of PVA and PEGDA, while bio-active hydrogels were formed of PVA and PEGDA and thenfurther functionalized with PEG-RGDS. The electrospinningset-up consisted of a syringe pump (Cole Palmer), a power

supply (Gamma High Voltage), and a collection plate orcounter electrode [Fig. 2(B)]. The PVA/PEGDA solution waselectrospun at 12 kV with a 20 cm gap between the needleand the counter electrode and a flow rate of 2.25 mL/h. Forevaluation of the hydrated fiber morphology, a trace amountof the fluorescently labeled PEG-RGDS was added to thepolymer solution prior to electrospinning.

After electrospinning, the fibrous membranes wentthrough a multi-step crosslinking process. The first crosslink-ing step was done immediately after electrospinning byexposing the membranes to UV for 60 s. The samples werethen sprayed with 70% ethanol and placed in the laminarflow hood under UV for 15 min to sterilize. Preliminary inves-tigations determined that including monoacrylated PEG-RGDS in the electrospinning solution at the concentrationsnecessary to elicit biological activity was actually detrimentalto the stability of the resulting fiber morphology. For this

FIGURE 3. Electrospun fiber morphology and fiber diameter. (A) SEM micrograph of the electrospun PVA-PEGDA fibers. Scale bar is 50 lm. (B)

Histogram showing the distribution of fiber diameters. [Color figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

FIGURE 4. Hydrated fiber morphology and fiber diameter. (A) micrograph of swollen electrospun PVA-PEGDA fibers. This is a projection from a

z-stack that is 40 lm in depth and 0.5 lm per slice. Scale bar 25 lm. (B) Histogram of the distribution of fiber diameters. [Color figure can be

viewed in the online issue, which is available at wileyonlinelibrary.com.]

4 STEPHENS-ALTUS ET AL. DEVELOPMENT OF BIOACTIVE PHOTOCROSSLINKABLE FIBROUS HYDROGELS

reason, PEG-RGDS was allowed to soak into the membranesafter the initial crosslinking step and then covalently immobi-lized within the matrix using a second UV exposure. For thebioactive sample set, a 2.5 mM PEG-RGDS solution plus 10lL/mL of the acetophenone photoinitiator was applied to theelectrospun membrane. In the control samples, sterile HBSwas used in place of the PEG-RGDS. During this functionaliza-tion process, the RGDS or HBS solution was applied to thefiber surface and allowed to penetrate the membrane for 3min before crosslinking for 60 s. Finally, the samples werepunched into 6.75 mm and 8 mm diameter disks for analysisof RGDS incorporation and cell seeding, respectively.

Scanning electron microscopyA FEI Quanta 400 high resolution field emission scanningelectron microscope (FESEM) was used to evaluate fibermorphology and fiber diameter of the as-spun fibroushydrogels. Samples were sputter coated (CRC-150 SputterCoater) with Au prior to imaging. All post image analysiswas done using Image J (NIH).

NinhydrinA ninhydrin assay was conducted to determine the amountof RGDS incorporated into the bioactive PEGDA gels andbioactive electrospun fibrous hydrogels (n ¼ 3). All PEGDAhydrogels and electrospun fibrous hydrogels, both controland bioactive, were soaked in sterile HBS at 37�C for 5 dayswith the HBS changed daily. To create the standard curves,control PEGDA gels and control electrospun membrane sam-ples were spiked with glycine at 0, 62.5, 125, 250, and 500lM. The same glycine solutions were added to empty vialsas an additional reference. All of the samples were lyophi-lized, and then subjected to acid hydrolysis in 6N HCl at150�C for 3 h under vacuum. After hydrolysis the HCl wasremoved via rotovap (Buchi) and the samples were reconsti-tuted in 0.1M sodium citrate, pH 5. The ninhydrin working

reagent was added and the samples were boiled for 15 min.After cooling, the samples were read in duplicate on a platereader (Bio Tek Instruments) at 570 nm.

Cell maintenanceNIH 3T3 fibroblasts (ATCC) were cultured in high glucoseDulbecco’s Modified Eagle Medium (DMEM, Gibco) with10% calf bovine serum (CBS, ATCC) and 1% L-Glutamine,penicillin, and streptomycin (GPS) at 37�C and 5% CO2.Fibroblasts were used between passages 6 and 10 with themedium changed every 3 days.

Cell viability, adhesion, and morphologyPrior to fibroblast seeding, the hydrogel and fibrous hydro-gel samples were soaked in sterile phosphate buffered sa-line (PBS, pH 7.4) at 37�C overnight. The PBS was changedthree times and afterwards the samples were soaked incomplete medium for 1 h. The medium was removed andfibroblasts were seeded at 5000 cells/cm2 in 24 well plates.Cells were evaluated for viability, adhesion, and morphologyas described below.

A fluorescent live/dead cell assay (Molecular Probes,Invitrogen), comprised of calcein AM (live) and ethidiumhomodimer (dead), was used to evaluate fibroblast viabilityafter the cells were allowed to attach to hydrogel samples(n ¼ 3) overnight. To label the cells, a solution of 2 lM cal-cein AM and 4 lM ethidium homodimer was added to thecomplete medium and enough solution was added to coverthe hydrogels in each well. Samples were incubated at 37�Cfor 15 min before imaging.

CellTiter Aqueous 96 (Promega) was used to evaluate cellattachement to the matrix (n ¼ 3). The cells were seeded onthe hydrogels and allowed to attach overnight. To determinethe number of cells on each sample, a ratio of 20 lL CellTiterper 100 lL of culture medium was added to each well. The

FIGURE 5. Cell adhesion to the PEGDA-based hydrogels verses fibrous hydrogels. There was no statistical difference in cell attachment between

the bioactive PEGDA hydrogels and bioactive PVA-PEGDA fibrous hydrogels. (* statistically different from all samples; # statistically different

from PEGDA þ RGDS and Electrospun þ RGDS, p < 0.05). [Color figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2011 VOL 00A, ISSUE 00 5

samples were then incubated at 37�C for 2 h and read on aplate reader (Bio Tek Instruments) at 570 nm.

To evaluate differences in morphology, fibroblasts wereseeded on the hydrogel samples (n ¼ 3) and then labeledwith DAPI and phalloidin to visualize nuclei and actin,respectively. This procedure was carried out after allowingthe fibroblasts to attach either overnight or for 5 days.Briefly, the samples were fixed in 4% paraformaldehyde for20 min, permeablized with 0.1% triton-x for 10 min, andblocked in a 1% bovine albumin serum (BSA) for 30 min.Samples were next incubated with a 1:40 dilution of Alexa-fluor 488-phalloidin (Molecular Probes, Invitrogen) for 45min followed by 300 nM DAPI (Molecular Probes, Invitro-gen) solution for 20 min. PBS washes were preformedbetween each of the steps in the staining process.

Fluorescence microscopyA Zeiss 5 Live confocal microscope and a Ziess Axiovert 135were used for fluorescent imaging of fiber morphology ofhydrated samples, cell viability, and cell morphology. Toevaluate hydrated fiber morphology, hydrogels spiked withfluorescently labeled PEG-RGDS were soaked in PBS over-

night at 37�C. The PBS was changed three times beforeimaging. To evaluate the cell morphology within the fibroushydrogels, z-stacks were obtained at 40� with a z spacingof 0.5 lm and compiled as projections. All post image analy-sis was done using Image J (NIH).

RESULTS AND DISCUSSION

Fiber formation, morphology, and stabilityMuch work has been done in the exploration of PEG-basedhydrogels as cell and tissue scaffolds, focusing primarily onthe bioinert properties of the base polymer material and theability to precisely engineer biofunctionality by covalentmodification. In effort to further develop these materials asECM-mimetics, this work sought to create a 3D fibrousarchitecture within the bioactive PEG hydrogel. To facilitatefiber formation, it is necessary to use a polymer solutionthat allows for chain-chain entanglements. In this study,fibers were formed by blending 3.4 kDa PEGDA with a 130kDa PVA capable of providing the chain-chain entangle-ments. In addition, the photoactive methacrylate groupsalong the backbone of the PVA polymer chain were able to

FIGURE 6. Cell Viability. Very few cells were present on control sample sets. Bioactive samples (containing PEG-RGDS) had greater cell numbers

in each case. The cell morphology of the fibroblasts on the bioactive electrospun sample is masked by the fibers. The cells are interpenetrated

within the fibrous membrane and therefore appear more rounded than cells on the flat bioactive hydrogels. The scale bar is 200 lm. [Color

figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

6 STEPHENS-ALTUS ET AL. DEVELOPMENT OF BIOACTIVE PHOTOCROSSLINKABLE FIBROUS HYDROGELS

crosslink with the acrylates on the PEGDA to form a stablefibrous network.

The as-spun fibers had an average diameter of 1.02 lm(Fig. 3), which was not altered by crosslinking. These fibershad a ribbon-shaped cross section as opposed to a roundedcross section. This phenomenon has been previouslyreported and is attributed to rapid solvent evaporation,which creates a skin on the fiber surface as the solvent inthe interior evaporates and atmospheric and electrostaticforces collapse the fiber.43,44

To evaluate the swollen or hydrated fiber morphologyand fiber stability, acrylated-PEG-RGDS-AlexaFluor 546 wasincluded in the electrospinning solution. Prior to imaging,the fluorescent electrospun fibrous hydrogels were soakedin PBS at 37�C for up to 5 days. Figure 4 shows a 3Dz-stack projection of the swollen fibers as well as the distri-bution of fiber diameters. After swelling, the fiber diametersincreased to an average of 7.1 lm, and the ribbon shape ofthe fibers persisted in the hydrated samples. In comparisonto pre-swelled materials, the hydrated samples had agreater distribution of fiber diameters, ranging from 3.35lm to over 13 lm.

Electrospinning of hydrogels has been previouslyreported, but few studies have investigated the swollenfibrous structure.45–49 Often membranes were swollen andthen dehydrated again prior to imaging, possibly impartingartifacts in the analysis. It is important to visualize thehydrated fibrous structure, as shown in this work, and todraw from those observations clues to the nature of thematerial’s interaction with cells. It is noted that the fibersgenerated in the these materials are larger than the fibrouscollagen and elastin components of the ECM, however furtherwork with cells of various types will be needed to determinethe conclusive effect of materials of this architecture.

RGDS incorporationThe ubiquitous adhesive ligand, RGDS, was reacted withmonacrylated PEG and then incorporated into the fibroushydrogels after a second UV exposure (Fig. 2). Similar meth-ods of covalent attachment have been employed in previousstudies with acrylate-based hydrogel systems.21,50,51 In pre-liminary studies, fluorescently labeled PEG-RGDS that wasallowed to soak in from the surface appeared to be evenlydistributed throughout the swollen fibrous matrix after

FIGURE 7. Cell morphology. The cells on both control sample sets demonstrate a rounded cell morphology. The cells on the bioactive PEGDA

hydrogels had a spread morphology and the cells on the bioactive electrospun samples show an extended dendritic morphology. Scale bar is

100 lm. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2011 VOL 00A, ISSUE 00 7

crosslinking. A ninhydrin assay was used to quantify theincorporation of RGDS in the different hydrogel materials.Based on this analysis, the amount of RGDS incorporatedinto the electrospun fibrous samples (163.13 lM) wasfound to be comparable to that in traditional PEGDA hydro-gels (177.69 lM). Since both the PEG and PVA hydrogelcomponents have been shown to be bioinert, similar levelsof RGDS in these materials indicate that they should possesscomparable bioactivity. As such, any divergent cellularresponse can be attributed to the differences in matrixarchitecture between the fibrous and traditional hydrogelsamples.

Cell viability, adhesion, and morphologyNIH 3T3 fibroblasts were seeded on the PEGDA-basedhydrogels and the fibrous hydrogels, both bioactive andnonbioactive, and were allowed to attach overnight to evalu-ate any differences in cell activity. The cell adhesion (Fig. 5)and cell viability (Fig. 6) were comparable between the bio-active PEGDA-based hydrogels and the bioactive fibroushydrogels. Fibroblasts on the bioactive PEGDA-based hydro-gels were well spread, covering most of the hydrogel sur-face. Similar numbers of cells were apparent on the bioac-tive fibrous hydrogels, but the cell morphology was maskedby the fibers. A few viable, but rounded cells were present

on the control fibrous hydrogels, which may be a result ofthe cells being physically entrapped in the fibers. The num-bers of adherent cells on the bioactive PEGDA-based hydro-gels and the fibrous hydrogels were statistically similar andstatistically greater than both non-bioactive controls.

Confocal microscopy was used to assess changes in cellmorphology influenced by PEGDA-based hydrogels and fi-brous hydrogels (Fig. 7). As seen in the viability studies,very few cells were present on control samples, and thecells that were present had a rounded morphology. The con-trol samples have little or no bioactivity (i.e., no RGDS wasincorporated) and therefore should elicit a low cellresponse as indicated by the rounded cell morphology.Fibroblasts on the bioactive hydrogels had a spread or flat-tened presentation. Cells on the bioactive fibrous hydrogelsshow an extended dendritic morphology, which is consistentwith fibroblasts seeded on 3D matrices.29,36,45,52,53 Thismorphology is thought to mimic that of the fibroblast’s invivo counterparts,52,53 a good indication that the RGDS pep-tide is providing the appropriate adhesion cues. It was alsoobserved that cells penetrated over 50 lm into the fibroushydrogels (Fig. 8), as opposed to the nonfibrous materials,where they remain associated with only the surface layer.Deeper imaging was prohibited by the scattering nature ofthe fibers.

FIGURE 8. Cell interactions with fibrous hydrogel. The cells interpenetrate the fibrous network. The projections were taken at 0.5 lm per slice.

(A) Day 1, total image depth is 32.8 lm. (B) Day 5, total image depth is 52.5 lm. (C) Day 5, total depth of image is 39.8 lm. Scale bar is 50 lm.

8 STEPHENS-ALTUS ET AL. DEVELOPMENT OF BIOACTIVE PHOTOCROSSLINKABLE FIBROUS HYDROGELS

The infiltration of cells into electrospun membranes haspreviously been an issue because small pore sizes have lim-ited cell migration.38,54,55 To address this issue, someresearchers have used leachable agents, such as polymersor salts, to increase the pore size within the membrane.56

These approaches have been effective; however the use ofleaching agents can reduce the mechanical integrity of themembranes. In this work, we demonstrated that the blend-ing of two hydrogel forming polymers by a combination ofelectrospinning and covalent crosslinking results in a bio-compatible material, which upon swelling, has an open po-rous network that does not compromise the integrity of thefibrous membrane. Although it is difficult to accurately mea-sure the porosity of a fibrous hydrogel, it is clear from 3Dimage projections that these hydrated networks allow forextensive cell infiltration, which is important in promotingmature cell scaffolds and tissue substitutes.

CONCLUSIONS

This investigation has demonstrated the ability to createphotoactive PVA/PEGDA fibrous hydrogels in which fibermorphology is maintained under cell culture conditions forat least 5 days. The ubiquitous adhesive peptide, RGDS, wascrosslinked into the fibrous hydrogels as a demonstration ofthe ability to render these materials bioactive. The concen-tration of RGDS incorporated in the fibrous materials wascomparable to that in traditional PEGDA-based hydrogels,indicating similar cell-adhesive properties. Cell viability andadhesion were similar for both materials. Interestingly, thefibroblasts displayed distinct morphological differences onthese two types of substrates. While cells were spread onthe PEGDA-based hydrogels, those on the fibrous hydrogelsshowed an extended in vivo-like morphology. These resultsindicate that modifying the hydrogel structure can influencethe behavior of attached cells in a manner similar to whatis expected from changing the biochemical nature of the ma-terial. The addition of a 3D fibrous architecture to the PEG-based hydrogel modular system allows another parameterof design for future investigations of materials that mimicthe functionality of the ECM.

REFERENCES1. Drury JL, Mooney DJ. Hydrogels for tissue engineering: Scaffold

design variables and applications. Biomaterials 2003;24:

4337–4351.

2. Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem Rev

2001;101:1869–1879.

3. Nguyen KT, West JL. Photopolymerizable hydrogels for tissue en-

gineering applications. Biomaterials 2002;23:4307–4314.

4. West JL, Hubbell JA. Photopolymerized hydrogel materials for

drug-delivery applications. React Polym 1995;25:139–147.

5. Vunjak-Novakovic G, Kaplan DL. Tissue engineering: The next

generation. Tissue Eng 2006;12:3261–3263.

6. Augst AD, Kong HJ, Mooney DJ. Alginate hydrogels as biomateri-

als. Macromol Biosci 2006;6:623–633.

7. Lutolf MP, Hubbell JA. Synthetic biomaterials as instructive

extracellular microenvironments for morphogenesis in tissue en-

gineering. Nat Biotechnol 2005;23:47–55.

8. Schmedlen KH, Masters KS, West JL. Photocrosslinkable polyvi-

nyl alcohol hydrogels that can be modified with cell adhesion

peptides for use in tissue engineering. Biomaterials 2002;23:

4325–4332.

9. Zajaczkowski MB, Cukierman E, Galbraith CG, Yamada KM. Cell-

matrix adhesions on poly(vinyl alcohol) hydrogels. Tissue Eng

2003;9:525–533.

10. Hern DL, Hubbell JA. Incorporation of adhesion peptides into

nonadhesive hydrogels useful for tissue resurfacing. J Biomed

Mater Res 1998;39:266–276.

11. Ifkovits JL, Burdick JA. Review: Photopolymerizable and degrad-

able biomaterials for tissue engineering applications. Tissue Eng

2007;13:2369–2385.

12. Martens P, Anseth KS. Characterization of hydrogels formed from

acrylate modified poly(vinyl alcohol) macromers. Polymer 2000;

41:7715–7722.

13. Bryant SJ, Anseth KS. The effects of scaffold thickness on tissue

engineered cartilage in photocrosslinked poly(ethylene oxide)

hydrogels. Biomaterials 2001;22:619–626.

14. Hahn MS, McHale MK, Wang E, Schmedlen RH, West JL. Physio-

logic pulsatile flow bioreactor conditioning of poly(ethylene gly-

col)-based tissue engineered vascular grafts. Ann Biomed Eng

2007;35:190–200.

15. Lee SH, Miller JS, Moon JJ, West JL. Proteolytically degradable

hydrogels with a fluorogenic substrate for studies of cellular pro-

teolytic activity and migration. Biotechnology Progress 2005;21:

1736–1741.

16. Peyton SR, Raub CB, Keschrumrus VP, Putnam AJ. The use of

poly(ethylene glycol) hydrogels to investigate the impact of ECM

chemistry and mechanics on smooth muscle cells. Biomaterials

2006;27:4881–4893.

17. Hahn MS, Miller JS, West JL. Three-dimensional biochemical and

biomechanical patterning of hydrogels for guiding cell behavior.

Adv Mater 2006;18:2679–2684.

18. Gonzalez AL, Gobin AS, West JL, McIntire LV, Smith CW. Integrin

interactions with immobilized peptides in polyethylene glycol dia-

crylate hydrogels. Tissue Eng 2004;10:1775–1786.

19. Leslie-Barbick JE, Moon JJ, West JL. Covalently-immobilized vas-

cular endothelial growth factor promotes endothelial cell tubulo-

genesis in poly(ethylene glycol) diacrylate hydrogels. J Biomater

Sci Polym Ed 2009;20:1763–1779.

20. Mann BK, Gobin AS, Tsai AT, Schmedlen RH, West JL. Smooth

muscle cell growth in photopolymerized hydrogels with cell adhe-

sive and proteolytically degradable domains: Synthetic ECM ana-

logs for tissue engineering. Biomaterials 2001;22:3045–3051.

21. Moon JJ, Hahn MS, Kim I, Nsiah BA, West JL. Micropatterning of

poly(ethylene glycol) diacrylate hydrogels with biomolecules to

regulate and guide endothelial morphogenesis. Tissue Eng Part A

2009;15:579–585.

22. Moon JJ, Lee S, West JL. Promotion of endothelial tubulogenesis

with EphrinA1 and EphB4 conjugated to synthetic hydrogels.

FASEB J 2006;20:A12–A12.

23. Moon JJ, Lee SH, Hahn MS, Nsiah BA, West JL. Regulation of en-

dothelial angiogenesis and vasculogenesis in synthetic poly(ethyl-

ene glycol) hydrogels modified with biomolecules. FASEB J 2007;

21:A748–A749.

24. Moon JJ, Lee SH, West JL. Synthetic biomimetic hydrogels incor-

porated with Ephrin-A1 for therapeutic angiogenesis. Biomacro-

molecules 2007;8:42–49.

25. Patel PN, Gobin AS, West JL, Patrick CW. Poly(ethylene glycol)

hydrogel system supports preadipocyte viability, adhesion, and

proliferation. Tissue Eng 2005;11:1498–1505.

26. Lee SH, Moon JJ, Miller JS, West JL. Poly(ethylene glycol) hydro-

gels conjugated with a collagenase-sensitive fluorogenic sub-

strate to visualize collagenase activity during three-dimensional

cell migration. Biomaterials 2007;28:3163–3170.

27. Ma PX, Zhang RY. Synthetic nano-scale fibrous extracellular ma-

trix. J Biomed Mater Res 1999;46:60–72.

28. Gelain F. Novel opportunities and challenges offered by nanobio-

materials in tissue engineering. Int J Nanomed 2008;3:415–424.

29. Cukierman E, Pankov R, Stevens DR, Yamada KM. Taking cell-ma-

trix adhesions to the third dimension. Science 2001;294:

1708–1712.

30. Stevens MM, George JH. Exploring and engineering the cell sur-

face interface. Science 2005;310:1135–1138.

31. Griffith LG, Naughton G. Tissue engineering—Current challenges

and expanding opportunities. Science 2002;295:1009–1014.

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2011 VOL 00A, ISSUE 00 9

32. Abbott A. Cell culture: Biology’s new dimension. Nature 2003;424:

870–872.

33. Flemming RG, Murphy CJ, Abrams GA, Goodman SL, Nealey PF.

Effects of synthetic micro- and nano-structured surfaces on cell

behavior. Biomaterials 1999;20:573–588.

34. Mattinez E, Engel E, Planell JA, Samitier J. Effects of artificial

micro- and nano-structured surfaces on cell behavior. Ann Anat

2009;191:126–135.

35. Engel E, Michiardi A, Navarro M, Lacroix D, Planell JA. Nanotech-

nology in regenerative medicine: The materials side. Trends Bio-

technol 2008;26:39–47.

36. Schindler M, Ahmed I, Kamal J, Nur-E-Kamal A, Grafe TH, Chung

HY, Meiners S. A synthetic nanofibrillar matrix promotes in vivo-

like organization and morphogenesis for cells in culture. Biomate-

rials 2005;26:5624–5631.

37. Rutledge GC, Fridrikh SV. Formation of fibers by electrospinning.

Adv Drug Deliv Rev 2007;59:1384–1391.

38. Pham QP, Sharma U, Mikos AG. Electrospinning of polymeric

nanofibers for tissue engineering applications: A review. Tissue

Eng 2006;12:1197–1211.

39. Casper CL, Yamaguchi N, Kiick KL, Rabolt JF. Functionalizing

electrospun fibers with biologically relevant macromolecules. Bio-

macromolecules 2005;6:1998–2007.

40. Tan AR, Ifkovits JL, Baker BM, Brey DM, Mauck RL, Burdick JA.

Electrospinning of photocrosslinked and degradable fibrous scaf-

folds. J Biomed Mater Res Part A 2008;87A:1034–1043.

41. Ifkovits JL, Devlin JJ, Eng G, Martens TP, Vunjak-Novakovic G,

Burdick JA. Biodegradable fibrous scaffolds with tunable proper-

ties formed from photo-cross-linkable poly(glycerol sebacate).

Acs Appl Mater Interfaces 2009;1:1878–1886.

42. Buttafoco L, Kolkman NG, Engbers-Buijtenhuijs P, Poot AA, Dijk-

stra PJ, Vermes I, Feijen J. Electrospinning of collagen and elastin

for tissue engineering applications. Biomaterials 2006;27:724–734.

43. Koombhongse S, Liu WX, Reneker DH. Flat polymer ribbons and

other shapes by electrospinning. J Polym Sci Part B: Polym Phys

2001;39:2598–2606.

44. Megelski S, Stephens JS, Chase DB, Rabolt JF. Micro- and nano-

structured surface morphology on electrospun polymer fibers.

Macromolecules 2002;35:8456–8466.

45. Ji Y, Ghosh K, Shu XZ, Li BQ, Sokolov JC, Prestwich GD, Clark

RAF, Rafailovich MH. Electrospun three-dimensional hyaluronic

acid nanofibrous scaffolds. Biomaterials 2006;27:3782–3792.

46. Kenawy ER, Layman JM, Watkins JR, Bowlin GL, Matthews JA,

Simpson DG, Wnek GE. Electrospinning of poly(ethylene-co-vinyl

alcohol) fibers. Biomaterials 2003;24:907–913.

47. Jin X, Hsieh YL. Anisotropic dimensional swelling of membranes

of ultrafine hydrogel fibers. Macromol Chem Phys 2005;206:

1745–1751.

48. Jin X, Hsieh YL. pH-responsive swelling behavior of poly(vinyl

alcohol)/poly(acrylic acid) bi-component fibrous hydrogel mem-

branes. Polymer 2005;46:5149–5160.

49. Li L, Hsieh YL. Ultra-fine polyelectrolyte fibers from electrospin-

ning of poly(acrylic acid). Polymer 2005;46:5133–5139.

50. Hahn MS, Taite LJ, Moon JJ, Rowland MC, Ruffino KA, West JL.

Photolithographic patterning of polyethylene glycol hydrogels.

Biomaterials 2006;27:2519–24.

51. Miller JS, Bethencourt MI, Hahn M, Lee TR, West JL. Laser-scan-

ning lithography (LSL) for the soft lithographic patterning of cell-

adhesive self-assembled monolayers. Biotechnol Bioeng 2006;93:

1060–1068.

52. Grinnell F. Fibroblast biology in three-dimensional collagen matri-

ces. Trends Cell Biol 2003;13:264–269.

53. Grinnell F, Ho CH, Tamariz E, Lee DJ, Skuta G. Dendritic fibro-

blasts in three-dimensional collagen matrices. Mol Biol Cell 2003;

14:384–395.

54. Liao S, Li BJ, Ma ZW, Wei H, Chan C, Ramakrishna S. Biomimetic

electrospun nanofibers for tissue regeneration. Biomed Mater

2006;1:R45–R53.

55. Venugopal J, Low S, Choon AT, Ramakrishna S. Interaction of

cells and nanofiber scaffolds in tissue engineering. J Biomed

Mater Res Part B: Appl Biomater 2008;84B:34–48.

56. Baker BM, Gee AO, Metter RB, Nathan AS, Marklein RA, Burdick

JA, Mauck RL. The potential to improve cell infiltration in com-

posite fiber-aligned electrospun scaffolds by the selective removal

of sacrificial fibers. Biomaterials 2008;29:2348–2358.

10 STEPHENS-ALTUS ET AL. DEVELOPMENT OF BIOACTIVE PHOTOCROSSLINKABLE FIBROUS HYDROGELS