在聚多钯胺薄膜上研究不同浓度的生长因子

Mussel-Inspired Immobilization of Vascular Endothelial Growth Factor(VEGF)for Enhanced Endothelialization of Vascular Grafts Young Min Shin,?Yu Bin Lee,?Seok Joo Kim,?Jae Kyeong Kang,§Jong-Chul Park,§Wonhee Jang,∥and Heungsoo Shin*,?,?,§

?Department of Bioengineering,?Institute for Bioengineering and Biopharmaceutical Research,and§Institute of Aging Society, Hanyang University,17Haengdang-dong,Seongdong-gu,Seoul133-791,Korea

§Department of Medical Engineering,BK21Project for Medical Science,Yonsei University College of Medicine,134Shinchon-dong, Seodaemun-ku,Seoul120-752,Korea

∥Department of Life Science,Dongguk University,Seoul,100-715,Korea

and X-ray photoelectron spectroscopy.Immobilization of VEGF

±0.4and197.4±19.7ng/cm2for DPv20and DPv200?lms,

VEGF revealed that the?uorescence intensity increased as a function

1.INTRODUCTION

The endothelial layer is important for maintaining healthy blood vessel conditions because it is involved in prevention of excessive tissue ingrowth(intimal hyperplasia)and thrombo-genesis.1For patients with vascular diseases,many prosthetic vascular grafts produced from biocompatible polymers such as expanded poly(tetra?uoroethylene)(ePTFE),poly(ethylene terephthalate)(PET),polycaprolactone(PCL),and polyur-ethane(PU)have been used in clinical settings,but they often have signi?cantly low patency rates for replacement of small diameter(<5to6mm)blood vessels.2,3This is mainly attributed to limited endothelialization on the surfaces of vascular grafts,which is then unable to suppress thrombosis and/or intimal hyperplasia e?ectively.

To improve endothelialization,we can modify the surface of vascular graft materials with functional groups or bioactive macromolecules.It was reported that functionalization of carboxyl or amine groups on the surface of PU-based grafts enhanced adhesion and proliferation of endothelial cells (ECs)4,5and ePTFE vascular grafts covalently immobilized with O-carboxymethylchitosan or heparin signi?cantly en-hanced blood compatibility.6,7Extracellular matrix(ECM) proteins,peptides,and growth factors have been actively investigated to improve cell compatibility of synthetic vascular grafts.8?10Collagen,RGD(Arg-Gly-Asp)sequence-containing peptide,vascular endothelial growth factor(VEGF),basic ?broblast growth factor(bFGF),and transforming growth factor beta(TGF-b)have been used to functionalize the surface of vascular prostheses.11?15

Identi?cation of the method to functionalize each bioactive molecule is of great signi?cance.Physical adsorption is a simple process;however,simply adsorbed molecules are often released with a burst in the initial stage,and thus their prolonged e?ect may be quickly diminished.Chemical conjugation may be more e?ective for prolonged periods of time,allowing successive regulation of cell behaviors.16Various techniques such as

Received:February5,2012

Revised:May14,2012

Published:May22,2012

surface etching,plasma treatment,and gamma-ray irradiation have been studied for covalent coupling of proteins including

ECM proteins and growth factors to the polymeric graft materials.17?21Although biofunctionalized surface can be simply developed by chemical conjugation,surface degradation and polymeric chain incision during the processes may cause alteration in mechanical properties.Denaturation of bioactive molecules by toxic chemicals and multiple reaction steps may be an additional issue to be addressed.

Recently,there have been reports on facile surface modi ?cation methods using dopamine that can be readily polymerized on the surface of various materials such as titanium,polydimethylsiloxane (PDMS),polystyrene (PS),polyethylene (PE),PTFE,and silicon rubber under slightly basic conditions (pH 8.5).22The process is referred to as mussel-inspired coating because dopamine has key functional groups that are responsible for mussel ’s adhesive versatility,cathecol (from 3,4-dihydroxy-L -phenylalanine (DOPA))and primany amines (from lysine).The deposited polydopamine layer is stable and enables the conjugation of molecules containing primary amine or thiol groups via imine formation or Michael addition reaction.For example,trypsin was immobilized on the polydopamine-coated cellulose substrates that maintained its enzymatic activity in a time-and concentration-dependent manner.23Other biomolecules such as bovine serum albumin and polylysine were immobilized on a polydopamine-deposited surface and enhanced hepatocyte binding a ?nity and produced a more highly organized neuronal network,respectively.24,25Inspired from these previous reports,in this study,we present a simple method of polydopamine-mediated immobilization of growth factors on the surface of polymeric vascular graft materials for accelerated endothelial-ization.In particular,we immobilized VEGF onto a model elastomeric material,poly(L -lactide-co -ε-caprolactone)(PLCL),and characterized surface properties and the distribution of immobilized growth factors.In addition,we investigated the e ?ect of immobilized growth factors on adhesion,spreading,migration,proliferation,and expression of CD 31marker of human umbilical vein endothelial cells (HUVECs).Finally,the versatility of the process was recon ?rmed by using another mitogenic growth factor,bFGF.

2.MATERIALS AND METHODS

2.1.Materials.3,4-Dihydroxyphenethylamine was obtained from Sigma (St.Louis,MO),chloroform from Junsei (Tokyo,Japan),endothelial cell growth medium-2(EGM-2)from Lonza (Basel,Switzerland),and Tris-HCl from Shelton Scienti ?c,(Peosta,IA).Fluorescent probes such as rhodamine-phalloidin,Hoechst 33258,and Alexa Fluor 488rabbit antimouse IgG were obtained from Molecular Probes (Eugene,OR),and a Milli-Q Plus System (Millipore,Billerica,MA)was used to produce ultrapure water.All other chemicals and solvents were of analytical grade and used without further puri ?cation.2.2.Immobilization of Growth Factors on PLCL Films Using Polydopamine.PLCL was dissolved in chloroform (10wt %)and poured onto a glass plate walled with aluminum tape and covered with aluminum foil.To evaporate the solvent,we placed the glass plate under ambient conditions for 24h and thoroughly dried it in a vacuum oven at 40°C for 72h.Polydopamine was deposited on the surface of PLCL ?lms as previous described.26In brief,the substrates were wetted with 70%ethanol and distilled water (DW),quickly immersed in the dopamine solution (2mg/mL in 10mM Tris-HCl bu ?er,pH 8.5),and then gently shaken on the rocker for 30min at room temperature.Following the reaction,the substrates were vigorously washed with DW three times to remove heterogeneously formed polydopamine,and the resulting samples were stabilized at 40°C for 2h.For immobilization of VEGF,the polydopamine-deposited ?lms were immersed and shaken in solutions containing growth factors at di ?erent concentrations for 12h at 37°C.We ?xed the volume of VEGF solution to 300μL for the reaction and prepared three di ?erent VEGF concentrations (132,1320,5280ng/mL).DP denotes the ?lm with deposited polydopamine,and DPv20,DPv200,and DPv800represent DP treated with 132,1320,and 5280ng/mL of VEGF solution,respectively.The schematic diagram of surface modi ?cation is illustrated in Figure 1.

2.3.Characterization of Surface Properties.Morphology and surface roughness of the prepared ?lms were visualized using scanning electron microscopy (SEM,JSM-6701F,JEOL,Tokyo,Japan)and atomic force microscopy (AFM,N8NEOS,Bruker Nano-GmbH,Bremen,Germany)using the oscillation mode on a Tap190Al-G (budget sensors)probe.Surface wettability was examined by observing the water contact angle on the ?lms (Pheonix 300,Surface Electro Optics,Suwon,Korea).All ?lms were ?xed on the slide glass,10μL of water was dropped on the ?lms,and the images of the droplet on the ?lms were captured 5s after dropping.The water contact angle was measured using Image-Pro Plus (MediaCybernetics,Bethesda,MD).The surface atomic composition was analyzed with X-ray photo-electron spectroscopy (XPS).The ESCA LAB 220I (Thermo VG Scienti ?c,Waltham,MA)spectrometer was used with a magnesium anode source producing Mg ?K (1253.6eV photons)X-rays with the pass energy of 20eV for high-resolution narrow scans.

Figure 1.Schematic diagram of surface modi ?cation with polydopamine coating method.

2.4.Imaging and Measurement of Immobilized VEGF.For the direct imaging of the immobilized VEGF,various amounts of VEGF were immobilized onto the polydopamine-deposited ?lms,which were blocked with 5%BSA solution for 12h and incubated with biotin-conjugated anti-VEGF (25μg/mL,Peprotech,Rocky Hill,NJ)for 60min at 37°C.The ?lms were then incubated with 1:100FITC-conjugated streptavidin (BD bioscience,San Jose,CA)for 1h at 37°C.After the ?lms were washed three times with DW,the samples were mounted on glass slides with Vectashield mounting medium (Vector Laboratory,Peterborough,U.K.).The immuno ?uorescence images and relative ?uorescence units (RFUs)were obtained using a ?uorescent microscope (TE2000,Nikon,Tokyo,Japan).To quantify the amounts of immobilized VEGF on the ?lms,we measured the concentration of remaining VEGF solution after immobilization,as recommended by the manufacturer (Peprotech,Rocky Hill,NJ),and also remeasured VEGF released from the ?lm after washing three times.

2.5.Culture of HUVECs and Cell Adhesion Study.HUVECs (CC-2519,Lonza Group,Basel,Switzerland)were subcultured in EGM-2under standard culture conditions (37°C with 5%CO 2,95%humidity),and EBM-2basal medium (without supplement of VEGF,bFGF,EGF,and IGF)was used in all experiments for evaluating cell behavior.The medium was changed every 2days.For cell adhesion study,HUVECs were seeded on the ?lms at a density of

3.2×103cells/cm 2and cultured for 24h;then,the adherent cells were ?xed using 3.7%paraformaldehyde.They were then permeabilized with cytoskeletal bu ?er solution (10.3g sucrose,0.292g NaCl,0.06g MgCl 2,0.476g HEPES bu ?er,0.5mL Triton X-100,in 100mL of water,pH 7.2)for 10min at 4°C and then blocked with 10%FBS in PBS for 1h at 37°C.The ?xed cells were incubated with antipaxilin (1:200,BD Science,NJ,USA)for 60min at 37°C and subsequently incubated with 1:100Alexa Fluor 488rabbit antimouse IgG,1:200rhodamine-phalloidin,and 1:5000Hoechst 33258for 1h at 37°C to visualize focal contact formation.After washing with DW,the samples were mounted on glass slides with Vectashield mounting medium (Vector Laboratory).The immuno ?uorescence images were obtained using a ?uorescent microscope (TE2000,Nikon).The number of adherent cells was measured by counting the number of nuclei at the projected area (magni ?cation 100×),and stained images were obtained from ?ve di ?erent areas (center,left,right,top,and bottom)per sample (n =3).

2.6.HUVEC Migration Assay Using Time-Lapse Microscopy.Migration of HUVECs was investigated with a wound closure model using time-lapse microscopy.In brief,presterilized samples were mounted on the coverslip,a silicon culture insert (Ibidi,Munich,Germany)was placed on the samples,and then 100μL of suspended

cells (1×104cells/well)was seeded into each well and subsequently incubated for 24h.After incubation,the silicon inserts were removed,and the appropriate volume of fresh medium was added to the samples,which were directly transferred to the inverted microscope (Olympus Optical,Tokyo,Japan)equipped with a microculture system.The images of cell colonies were captured using a change-coupled device (CCD)camera (Electric Biomedical,Osaka,Japan)at 5min intervals for up to 15h.Cell migration distance was calculated from the obtained images with manual tracking and analyzed using chemotaxis and a migration tool plug-in (v.1.01,Ibidi)in the ImageJ software (1.37v by W.Rasband,National Institutes of Health,Baltimore,MD).The accuracy of the experiment is within 1.8μm/pixel.Average migration speed was determined by dividing the migration distance of the centroid between two time points over each time interval (15cells/group).

2.7.Cell Proliferation and Immuno ?uorescence Staining for CD 31.Initially,proliferation of HUVECs for over 3days was assessed by BrdU incorporation assay.The cells cultured on the ?lms for 3days were pulsed with 10μM of BrdU (Sigma Aldrich,St.Louis,MO)for 24h.Following the pulsing,the cells were washed with PBS and then ?xed with

3.7%paraformaldehyde.DNA in the cells was denatured with 4N HCl containing 1%Triton X-100for 10min and neutralized by repeated washing with PBS.After blocking with 10%FBS in PBS for 30min,the cells were incubated with anti-BrdU (1:100)(Sigma Aldrich)and then reincubated with Alexa ?uor 488rabbit antimouse IgG (1:50)and Hoechst 33258(1:1000)for 1h at 37°C.The ?uorescence images were obtained using ?uorescence microscopy,and images were analyzed with Image-Pro Plus.For long-term proliferation assay,2×104of HUVECs were seeded on fabricated samples (size =1.99cm 2_24well)with EGM-2media with no supplement of growth factors and cultured for over 7days.The media was changed every day.To observe the di ?erence of proliferation on each sample group,cell images were obtained at 1,3,5,and 7days (n =3).Each sample was observed with ?ve separate ?elds (middle,top,bottom,left,right).Data were normalized with values of 1day for each group for calculation of relative cell number.

For the investigation of cell ?cell junction,8×104cells were seeded on the sample and cultured for 3days with EGM-2media with no supplement of growth factors.On the day of analysis,each sample was gently washed with PBS and treated with 4%paraformaldehyde for 20min to ?x cells.Cells were then permeablized and incubated with blocking bu ?er (5%BSA,0.1%Tween-20in PBS)for 1h in at 37°C.For immunostaining,CD 31antibody (Cell Signaling Technology,Danvers,MA)solution (1:100)was treated to the samples (1h,37°C),and Alexa Fluor 488rabbit antimouse IgG (1:100),rhodamine-phalloidin (1:200),and Hoechst 33258(1:10000)-containing solution were then used for visualization of nuclei,f-actin and CD 31(1h,37

Figure 2.Surface morphology and roughness.(a)Scanning electron microscopy (SEM)and (b)atomic force microscopy (AFM)images of modi ?ed ?lms.(Scale bar =500μm).

°C),respectively.The samples were then mounted on slide glass with Vectashield mounting medium (Vector Laboratory)and immuno-?uorescence images were obtained through confocal microscope (Nikon).

2.8.Statistical Analysis.All data are presented as mean ±standard deviation (SD)for n =

3.Statistical signi ?cance was assessed with ANOVA and Student ’s t test (p <0.05).

3.RESULTS AND DISCUSSION

3.1.SEM and AFM Images.The surface morphology and

roughness were observed with SEM and AFM after modi ?cation.Dopamine is known to be polymerized under slightly alkaline conditions (pH 8.5),and polymerized dopamine can be spontaneously deposited onto many types of substrates including metal and polymers.22As shown in Figure 2a,SEM investigations revealed the di ?erence in the surface morphology between PLCL,DP,and DPv200.PLCL ?lm showed rough surface in which nanoscaled ?brous texture was randomly generated,possibly due to evaporation of volatile solvent under an ambient condition.Polydopamine coating for 30min dramatically changed the surface morphology of PLCL ?lm,and nanoscaled ?brous texture almost disappeared.On the deposited polydopamine layer,partial aggregates of polymer-ized dopamine were also found.In our previous work,we found that homogeneous deposition of polydopamine can be achieved within few minutes that covered throughout the surface of polymeric ?lm.26As the coating time increased,the thickness of polydopamine layer seemed to be thickened by multiple depositions of polydopamine layers.During prolonged coating process,we observed that polymerized dopamine changed surface morphology (appearance of nanosized aggregates or lumps).The same phenomena have been reported in many literatures studying polydopamine coating on various types of material surfaces.24,27,28Immobilization of VEGF on the polydopamine-deposited PLCL ?lms slightly changed the surface morphology,in which additional deposition of molecules with several tens of nanometers newly appeared.The surface roughness among PLCL,DP,and DPv200samples was more clearly distinguished in AFM images,as shown in Figure 2b.The PLCL ?lm demonstrated microsize scale undulation,which was relatively ?attened by coating with polydopamine.On the DPv200?lm,the relative roughness on the ?at surface signi ?cantly increased,which seemed to result from the secondary immobilization of VEGF with 3D structure.Previously,Zhou et al.reported that immobilization of proteins on the charged surface exhibited distinct 3D protein nanostructure by electrostatic immobilization and biospeci ?c interaction (antigen ?antibody binding),resulting in increased surface roughness.29In addition,polydopamine-coated dia-mond-like carbon (DLC)?lm demonstrated an ad-layer consisting of nanosized particles and increased surface roughness after BSA immobilization on nanoscale.27Therefore,polydopamine was successfully coated on the ?lm partially exhibiting polydopamine aggregates,and the surface roughness was subsequently increased as VEGF was immobilized.

3.2.Surface Wettability.The surface wettability of polydopamine-coated substrates has been previously reported.For example,the water contact angle of clean glass (11.1±1.1°)increased to 4

4.0±1.8°after 24h of polydopamine coating,and polydopamine-coated PVDF ?lm showed a dramatic decrease in water contact angle.30In good agreement with previous reports,the water contact angle on the PLCL ?lm coated with polydopamine for 30min was signi ?cantly

decreased from 75.1±2.1to 65.3±2.0°(Figure 3).In addition,immobilization of VEGF on the polydopamine-coated

?lm subsequently decreased the water contact angle as a function of VEGF concentration (58.9±1.6and 52.4±2.4°for reaction with 120and 1200ng/mL of VEGF,respectively).These ?ndings may be due to the combined e ?ect of hydrophilicity of polydopamine and VEGF.The representative water droplet images are displayed in Figure 3.

3.3.X-ray Photoelectron Spectroscopy.The chemical composition of the modi ?ed ?lms was characterized with XPS.As shown in Figure 4a,PLCL ?lms exhibited only carbon and oxygen peaks as the main atomic elements,whereas a nitrogen peak was newly recorded at 399eV in the XPS spectra for DP and DPv200?lms.The intensity of nitrogen signal in the high-resolution narrow spectrum showed an increase in the DPv200?lms compared with the DP ?lms,indicating successful immobilization of VEGF.An evident change in carbon bond composition observed in the high-resolution narrow carbon spectra (C 1s)more clearly supported these conclusions (Figure 4b).Peaks at 288.6eV (C O),286.6eV (C ?O),and 28

4.6eV (C ?H)were detected on the PLCL ?lms,which are attributed to the presence of esters in the main backbone of the polymer,whereas a broad peak C ?N bond (286.0eV)was recorded on both DP and DPv200?lms,indicating the presence of polydopamine and VEGF.The intensity of the C ?N bond increased as VEGF was further immobilized (DPv200group),suggesting that protein molecules consisting of numerous peptide bonds contributed to the signal intensity (the N/C ratio increased from 0.07to 0.13).The polydop-amine layer and immobilization of VEGF was recon ?rmed by the appearance of nitrogen peaks (N 1s)and their enhanced intensity,as shown in Figure 4c.Although minimally detected,a sulfur signal appeared only in the DPv200?lm,which is due to the presence of sulfur atoms in the 3D structure of the proteins as part of disul ?de bonds or the amino acid cysteine (Figure

Figure 3.Surface wettability.Water contact angle and representative images of water droplet on the ?lms.Asterisk “*”indicates statistical signi ?cance between groups (p <0.05).

4d).Consistent with our results,Ku et al.demonstrated that the composition of the carbon peak and surface tension was changed by further immobilized protein when polydopamine-coated substrates were reacted with serum protein solution.31These results suggest that VEGF was immobilized on the polydopamine-coated ?lms without further chemical reaction.3.4.Imaging and Quanti ?cation of Immobilized VEGF.Although several reports have demonstrated secondary immobilization of bioactive molecules including proteins,

peptide,and enzymes on the polydopamine-coated substrates,the concentration and bioactivity have not yet been fully characterized.23,30,32Activation of target cells by growth factors relies on the appropriate presentation of 3-D protein structure and concentration.To investigate the presence of immobilized VEGF on the ?lms,we exploited the basic principle of ELISA,in which antibody against VEGF was attached to the immobilized VEGF that was subsequently probed with a ?uorescent dye.As shown in Figure 5a,the green ?uorescence was partially observed on the ?lm (DPv20),and the ?uorescence was intensi ?ed as the amount of VEGF (DPv200and DPv800)for immobilization increased.The distribution of strong ?uorescence (bright green spot)was homogeneously observed throughout the surface areas of the samples,indicating that VEGF was successfully immobilized on the polydopamine-coated surface in a concentration-dependent manner.RFUs increased from 7.4±0.7units for DPv20to 15.9±1.8units and 25.7±1.1units for DPv200and DPv800,respectively (Figure 5a).Previously,the conjugation yield of VEGF on the collagen sca ?olds via EDC/NHS chemistry was approximately 3to 5%,33and we also obtained approximately 10to 20%peptide conjugation yield using the same chemistry.17In contrast,Poh et al.reported that ～52%of

VEGF was immobilized on the polydopamine-coated titanium substrates (26.0±2.5ng/cm 2),suggesting that conjugation of biomacromolecules inspired by polydopamine may be more e ?cient.34Consistent with the ?uorescence images,the amount of VEGF on the polydopamine-deposited ?lm,indirectly measured by ELISA,was dependent on the concentration of VEGF,which was 19.8±0.4and 197.4±19.7ng/cm 2for DPv20and DPv200?lms,respectively (Figure 5b).Although it is di ?cult to compare our results with previous ones due to the use of di ?erent materials and coating time,it should be noted that secondary ligation of VEGF via deposited polydopamine layer is very e ?ective under our experimental conditions.Furthermore,the VEGF immobilized on the surface was not released or detached from the ?lms during repeated vigorous washing.Therefore,these results,in addition to the changes in wettability and atomic composition,clearly indicate that VEGF was successfully immobilized on the polydopamine-coated substrates in a concentration-dependent manner.

3.5.Adhesion and Spreading of HUVECs on the Films.The formation of healthy endothelium is highly desirable for developing vascular grafts,and thus adhesion of EC should be regulated.Previous reports demonstrated that polydopamine coating enhance adhesion and survival of the cells,which appears to be regulated by adsorbed serum proteins.For example,adhesion of human dermal microvascular endothelial cells (HDMECs)and PC12cells was enhanced on the polydopamine-coated substrates,31and the viability of HUVECs was also improved on the polydopamine-coated PCL nano ?bers.35To examine the function of the immobilized VEGF on the polydopamine coated ?lms,the ?rst experiment

Figure 4.Surface atomic composition of modi ?ed ?lms.(a)XPS wide spectrum of the ?lms,(b)high-resolution spectrum of carbon peak,(c)high-resolution spectrum of nitrogen peaks (N 1s),and (d)high-resolution spectrum of sulfur peak (S 2p).

Figure 5.Qualitative and quantitative analyses of VEGF immobiliza-tion.(a)RFU results from ?uorescent-labeled VEGF on the ?lms (scale bar:50μm)and (b)immobilized amounts of VEGF on the ?lms quanti ?ed by ELISA.

was observation of adherent morphology and measurement of adherent cell number after ?uorescence staining (Figure 6).PLCL is highly hydrophobic polymer,and we reported that adhesion of hMSC on the PLCL ?lm was limited.36Consistent with previous results,on the PLCL ?lm without modi ?cation,

HUVECs showed limited spreading,and intracellular paxillin and F-actin were poorly developed.The number of HUVECs attached to the substrates was signi ?cantly increased from 30.8±3.9to 40.5±11.3when HUVECs were cultured on DP,indicating that polydopamine deposition enhanced cell adhesion,which may be due to either enhanced protein adsorption or direct electrostatic interactions with cells or by both of them,as previously reported.31,35As cell adhesion was enhanced,HUVECs were more spread with visible presentation of more mature F-actin intracellular stress ?bers as well as focal contacts.Whereas polydopamine coating on the PLCL ?lm enhanced cell attachment,the e ?ect by VEGF immobilization was marginal.The number of adherent cells on the DPv20and DPv200?lms was signi ?cantly greater than that on the PLCL ?lm;however,it was comparable to that on the DP ?lm.VEGF is known to stimulate cellular responses by binding to and subsequent phosphorylation of tyrosine kinase receptors in cell membrane.Nonetheless,there have been several reports demonstrating VEGF-modulated adhesion of HUVECs by immobilized or soluble VEGF.For example,collagen sca ?olds immobilized with VEGF (97.2±8.0ng)showed an approximately two times greater viability of ECs as compared with nonmodi ?ed sca ?olds,33and adhesion of HUVECs was regulated by absence/presence of VEGF (10ng/mL).37However,the primary function of VEGF is involved in mitogenesis and migration of ECs rather than cell adhesion.Additionally,cell adhesion appeared to be primarily a ?ected by polydopamine deposition,which may have masked any potential in ?uence of VEGF on adhesion of HUVECs.

3.6.Migration of HUVECs on Films.We then assessed the bioactivity of the immobilized VEGF by measuring migration and proliferation behavior of HUVECs on the modi ?ed substrates.Time-lapse images of wound recovery by migrating HUVECs on the substrates were obtained during 15h of culture,and representative images after 12h are shown in Figure 7a.We found that each satellite cell (indicated with a white circle)migrated from the defects,occupied the void space,and ?nally,covered ～10%of the wound area after 12h (migrating cell area is indicated with white dashed lines).On DP and DPv200?lms,the majority of migrating cells moved toward the center from the outermost edge of the defect while maintaining their cell ?cell interactions,although there was no precise evidence that immobilized VEGF on the dopamine-deposited substrate modulated overall morphological change in HUVEC migration.The migration of HUVECs on the PLCL ?lm was limited,and more individual cells were sparsely distributed in the defect.Next,we calculated the migration speed using manual tracking and a chemotaxis/migration tool.Previously,Z.Yin et al.reported that the migration speed of HUVECs induced by VEGF is ～1μm/min in complete culture medium (10ng/mL VEGF)and 0.3μm/min under the VEGF-free culture condition,indicating that VEGF regulates the migration speed of HUVECs.37As shown in Figure 7a,the migration speed of HUVECs on the PLCL ?lms,including independently moving cells,was 27.4± 5.0μm/h,which increased to 29.8±3.1μm/h on the DP ?lm.Because adhesion of HUVECs is regulated by the coated polydopamine,these results seem to result from interactions with adsorbed serum proteins on the DP ?lm.The maximum migration speed of HUVECs was recorded on the DPv200?lms (32.5±2.4μm/h),indicating that VEGF is active on the surface and potentially accelerated migration.Although the migration speed on DPv200?lms was statistically greater than that on PLCL and

Figure 6.Adhesion morphology and focal adhesion formation of HUVECs cultured on the VEGF-immobilized ?lms.Immuno ?uorescence staining for paxillin (green)and F-actin (red)(scale bar:50μm)and focal adhesions formed by paxillin were pointed by white arrows.Number of adherent cells per projected area on the ?lms.Asterisk “*”indicates statistical signi ?cance between the groups (p <0.05).

DP (p <0.012),it may be too early to make clear conclusion that VEGF immobilized on polydopamine-coated substrates is

bioactive due to accuracy of the experiment (resolution we acquired from the experiment was 1.8μm/pixel).The enhanced migration may be due to the presence of either VEGF or polydopamine-coating.HUVEC migration requires complex signal transduction such as ERK signaling.18,34In either case,active migration of HUVECs on the substrates can lead to faster recovery of defects,and,consequently,alternative proliferation can guide the generation of endothelium with appropriate cell ?cell contacts to maintain intensive regulation of vessel health.1

3.7.Proliferation and Immuno ?uorescence Staining of HUVECs on Films.We then measured proliferation of HUVECs using a BrdU incorporation assay.There are many reports that soluble or immobilized VEGF facilitates the proliferation of HUVECs by binding with its receptor.For example,VEGF-immobilized titanium implants via polydop-amine coating signi ?cantly enhanced the proliferation of HDMECs for longer than 14days,34and heparinized chitosan/PCL nano ?bers-incorporated with VEGF also en-hanced HUVEC proliferation up to approximately two times greater than that of the same sca ?olds without VEGF for 72h of culture.38As shown in Figure 7b,25.3±10.1%of BrdU-positive cells were observed on the PLCL ?lms,which was consistent with adhesion,spreading,and migration results,indicating that hydrophobicity and absence of cell-interactive cues of PLCL ?lms suppressed metabolic activities of HUVECs.When the ?lms were modi ?ed with polydopamine and 20and 200ng of VEGF,the BrdU incorporation ratio was signi ?cantly increased to 50.3±10.0,5

4.5±

5.9,and 64.9±9.2%,respectively,which is approximately two times greater than that on the PLCL ?lms.These results again suggest that VEGF immobilized on the ?lms can stably regulate proliferation.We then reprepared ?lms immobilized with bFGF,another type of mitogenic growth factor for HUVECs,and evaluated proliferation of HUVECs.Consistent with VEGF results,bFGF immobilized on the ?lms (DPbf200)enhanced proliferation (68.4±8.4%)(Figure 7c).

Although DPv200signi ?cantly enhanced proliferation of HUVECs,the di ?erence between DP and DPv200appeared to be marginal for 3days of culture.Therefore,we extended our culture period and measured the number of cells on each substrate.As shown in Figure 8a,the relative cell number was continuously reduced on PLCL during 7days of culture.On DP,the proliferation was slightly increased over 5days,and no further increase was observed.However,the relative cell number on DPv200was signi ?cantly increased 1.63±0.02at

Figure 7.Migration and proliferation of HUVECs on the ?lms.(a)Migration velocity of HUVECs for 15h of incubation and representative images of migrated cells after 12h.p values are calculated to propose statistical comparison between the groups,and the values are presented.The black and white dashed lines indicate initial wound area and cell migrated area,respectively.Proliferation of HUVECs on the ?lms immobilized with (b)VEGF and (c)bFGF was analyzed through BrdU incorporation assay.Asterisk “*”that is indicated in panels b and c means statistical signi ?cance between the groups (p <0.05).

day7as compared with the number of cells on day1,indicating that VEGF is actively modulating proliferation of HUVECs on DPv200for long-term cultivation.We then further examined the biological activity of VEGF by staining platelet/EC adhesion molecule CD31that is involved in maintenance of integrated vascular structure.As shown in Figure8b,HUVECs cultured on DPv200presented a strong and mature gap junction with intensive staining at intermittent of cell to cell adhesion after3days of incubation.Although HUVECs on DP are in close contact to each other,a limited level of cell?cell communication was observed.Collectively,these results suggest that VEGF as immobilized on the polydopamine layer is biologically active to modulate proliferation,migration, and cell?cell signaling.

4.CONCLUSIONS

Endothelialization on vascular grafts is highly desired for successful replacement of damaged vessels and therefore is of importance to enhance adhesion,migration,and proliferation of ECs.In this study,we introduced an easy process to immobilize growth factors on the surfaces of elastomeric polymer,PLCL inspired by polydopamine coating.The surface roughness of PLCL?lms increased after polydopamine coating and further changed with immobilization of VEGF,suggesting successful introduction of VEGF.Surface wettability was a?ected by polydopamine coating,and immobilization of VEGF further decreased the water contact angle in a VEGF concentration-dependent manner.The immobilized amount of VEGF on the polydopamine-coated surface was also dependent on the concentration of reacted VEGF,which was con?rmed by ?uorescence labeling and ELISA.Adhesion and spreading of HUVECs on the?lms was primarily a?ected by deposition of polydopamine layer,which may have masked the e?ect of VEGF,or the immobilized VEGF had marginal e?ect on cell adhesion.On the contrary,the immobilized VEGF signi?cantly enhanced proliferation and improved migration of HUVECs, suggesting that they maintained biological activity following immobilization setup.In addition,the expression and local-ization CD31in HUVECs was positively controlled by VEGF on the surface.We also demonstrated that the same process can be applied to immobilization of bFGF to render vascular graft materials to modulate EC activities actively.In conclusion,the combination of polydopamine and growth factors presents a strategy for controlling cell behavior on the surfaces of vascular graft materials,and this approach may be used for other biomaterial surfaces in tissue engineering sca?olds or cell

culture support.

■AUTHOR INFORMATION

Corresponding Author

*Tel:+82-2-2220-2346.Fax:+82-2-2298-2346.E-mail:hshin@ hanyang.ac.kr.

Notes

The authors declare no competing?nancial interest.■ACKNOWLEDGMENTS

This work was supported by National Research Foundation (NRF)of Korea Grant funded by the Korean Government

(MEST)(2011-0015222)(2011-0005550)

■REFERENCES

(1)de Mel, A.;Jell,G.;Stevens,M.M.;Seifalian, A.M. Biomacromolecules2008,9,2969?2979.

(2)Ma,Z.;He,W.;Yong,T.;Ramakrishna,S.Tissue Eng.2005,11, 1149?1158.

(3)Patel,S.;Tsang,J.;Harbers,G.M.;Healy,K.E.;Li,S.J.Biomed. Mater.Res.,Part A2007,83A,423?433.

(4)Kaibara,M.;Takahashi,A.;Kurotobi,K.;Suzuki,Y.Colloids Surf., B2000,19,209?217.

(5)Zhu,Y.;Gao,C.;He,T.;Shen,J.Biomaterials2004,25,423?430.

(6)Begovac,P.C.;Thomson,R.C.;Fisher,J.L.;Hughson,A.; Ga l lhagen,A.Eur.J.Vasc.Endovasc.Surg.2003,25,432?437.

(7)Zhu,A.;Zhang,M.;Zhang,Z.Polym.Int.2004,53,15?19.

(8)Xia,J.H.;Xie,A.N.;Zhang,K.L.;Xu,L.;Zheng,X.Y.Chin. Med.J.(Beijing,China,Engl.Ed.)2006,119,117?121.

(9)Zhu,Y.;Gao,C.;Liu,X.;He,T.;Shen,J.Tissue Eng.2004,10, 53?61.

(10)Zhu,Y.;Gao,C.;Liu,X.;Shen,J.Biomacromolecules2002,3, 1312?1319.

(11)He,W.;Yong,T.;Teo,W.E.;Ma,Z.;Ramakrishna,S.Tissue Eng.2005,11,1574?1588.

(12)Xia,Y.;Boey,F.;Venkatraman,S.S.Biointerphases2010,5, FA32?FA40.

(13)Cheng,Y.;Liu,Y.F.;Zhang,J.L.;Li,T.M.;Zhao,N.World J. Gastroenterol.2007,13,2862?2866.

(14)Takemoto,S.;Morimoto,N.;Kimura,Y.;Taira,T.;Kitagawa, T.;Tomihata,K.;Tabata,Y.;Suzuki,S.Tissue Eng Part A2008,14, 1629?1638.

(15)Yao,E.H.;Fukuda,N.;Ueno,T.;Matsuda,H.;Nagase,H.; Matsumoto,Y.;Sugiyama,H.;Matsumoto,K.Cardiovasc.Res.2009, 81,797?804.

(16)Ito,Y.Soft Matter2008,4,46?56.

(17)Shin,Y.;Shin,H.;Lim,Y.Macromol Res.2010,18,472?481.

(18)Wu,J.;Zeng,F.;Huang,X.-P.;Chung,J.C.Y.;Konecny,F.; Weisel,R.D.;Li,R.-K.Biomaterials2011,32,579?586.

Figure8.Cell proliferation assay for long-term culture and CD31 immuno?uorescence staining.(a)Relative cell number for over7days and representative images of HUVECs cultured on each substrate for7 days.“*”,“$”,and“#”indicate statistical signi?cance as compared with the value at1,3,and5days,respectively.(b)CD31 immuno?uorescence staining was performed to con?rm the e?ect of VEGF on maturation of gap junction.Nuclei and F-actin were stained with blue and red dye,and CD31is indicated with green dye.(Scale bar=100μm).

(19)Grondahl,L.;Chandler-Temple,A.;Trau,M.Biomacromolecules 2005,6,2197?2203.

(20)Park,G.E.;Pattison,M.A.;Park,K.;Webster,T.J.Biomaterials 2005,26,3075?3082.

(21)Park,H.;Lee,K.Y.;Lee,S.J.;Park,K.E.;Park,W.H. Macromol.Res.2007,15,238?243.

(22)Lee,H.;Dellatore,S.M.;Miller,W.M.;Messersmith,P.B. Science2007,318,426?430.

(23)Lee,H.;Rho,J.;Messersmith,P.B.Adv.Mater.2009,21,431?434.

(24)Zhu,B.C.;Edmondson,S.Polymer2011,52,2141?2149.

(25)Kang,K.;Choi,I.S.;Nam,Y.Biomaterials2011,32,6374?6380.

(26)Shin,Y.M.;Lee,Y.B.;Shin,H.Colloids Surf.,B2011,87,79?87.

(27)Tao,C.;Yang,S.;Zhang,J.;Wang,J.Appl.Surf.Sci.2009,256, 294?297.

(28)Muller,M.;Kessler,https://www.sodocs.net/doc/c811608718.html,ngmuir2011,27,12499?12505.

(29)Zhou;Wang,X.;Birch,L.;Rayment,T.;Abell,https://www.sodocs.net/doc/c811608718.html,ngmuir 2003,19,10557?10562.

(30)Zhu,L.P.;Yu,J.Z.;Xu,Y.Y.;Xi,Z.Y.;Zhu,B.K.Colloids Surf., B2009,69,152?155.

(31)Ku,S.H.;Ryu,J.;Hong,S.K.;Lee,H.;Park,C.B.Biomaterials 2010,31,2535?2541.

(32)Jiang,J.-H.;Zhu,L.-P.;Li,X.-L.;Xu,Y.-Y.;Zhu,B.-K.J.Membr. Sci.2010,364,194?202.

(33)Miyagi,Y.;Chiu,L.L.;Cimini,M.;Weisel,R.D.;Radisic,M.; Li,R.K.Biomaterials2011,32,1280?1290.

(34)Poh,C.K.;Shi,Z.;Lim,T.Y.;Neoh,K.G.;Wang,W. Biomaterials2010,31,1578?1585.

(35)Ku,S.H.;Park,C.B.Biomaterials2010,31,9431?9437.

(36)Shin,Y.M.;Kim,K.S.;Lim,Y.M.;Nho,Y.C.;Shin,H. Biomacromolecules2008,9,1772?1781.

(37)Yin,Z.;Noren,D.;Wang,C.J.;Hang,R.;Levchenko,A.Mol. Syst.Biol.2008,4.

(38)Du,F.;Wang,H.;Zhao,W.;Li,D.;Kong,D.;Yang,J.;Zhang, Y.Biomaterials2012,33,762?770.