IOP PUBLISHING BIOMEDICAL MATERIALS

Biomed. Mater. 5 (2010) 044109 (8pp) doi:10.1088/1748-6041/5/4/044109

The biological activities of (1,3)-(1,6)-β-D-glucan and porous electrospun PLGAmembranes containing β-glucan inhuman dermal fibroblasts and adiposetissue-derived stem cellsYeon I Woo1,2, Bong Joo Park1,3, Hye-Lee Kim1,2, Mi Hee Lee1,2,Jungsung Kim1,2, Young-Il Yang4, Jung Koo Kim3, Kazufumi Tsubaki5,Dong-Wook Han6 and Jong-Chul Park1,2,7

1 Department of Medical Engineering, Yonsei University College of Medicine, 134 Shinchon-dong,Seodaemun-gu, Seoul 120-752, Korea2 Brain Korea 21 Project for Medical Science, Yonsei University College of Medicine,134 Shinchon-dong, Seodaemun-gu, Seoul 120-752, Korea3 Department of Biomedical Engineering, College of Biomedical Science and Engineering,Inje University, Kimhae 621-749, Korea4 Department of Pathology, School of Medicine, Paik Institute for Clinical Research, Inje University,633-165 Gae-dong, Busan-jin-gu, Busan 614-735, Korea5 R&D division, Asahi Denka Co. Ltd, 7-2-35 Higashi-ogu, Arakawa-ku, Tokyo 116-8554, Japan6 Department of Nanomedical Engineering, College of Nanoscience & Nanotechnology,Pusan National University, geumjeong-gu, Busan 609-735, Korea

E-mail: parkjc@yuhs.ac

Received 27 February 2010Accepted for publication 12 July 2010Published 3 August 2010Online at stacks.iop.org/BMM/5/044109

AbstractIn this study, we investigated the possible roles of (1,3)-(1,6)-β-D-glucan (β-glucan) andporous electrospun poly-lactide-co-glycolide (PLGA) membranes containing β-glucan forskin wound healing, especially their effect on adult human dermal fibroblast (aHDF) andadipose tissue-derived stem cell (ADSC) activation, proliferation, migration, collagen gelcontraction and biological safety tests of the prepared membrane. This study demonstratedthat β-glucan and porous PLGA membranes containing β-glucan have enhanced the cellularresponses, proliferation and migration, of aHDFs and ADSCs and the result of a collagen gelcontraction assay also revealed that collagen gels contract strongly after 4 h post-gelationincubation with β-glucan. Furthermore, we confirmed that porous PLGA membranescontaining β-glucan are biologically safe for wound healing study. These results indicate thatthe porous PLGA membranes containing β-glucan interacted favorably with the membraneand the topical administration of β-glucan was useful in promoting wound healing. Therefore,our study suggests that β-glucan and porous PLGA membranes containing β-glucan may beuseful as a material for enhancing wound healing.

(Some figures in this article are in colour only in the electronic version)

7 Author to whom any correspondence should be addressed.

1748-6041/10/044109+08$30.00 1 © 2010 IOP Publishing Ltd Printed in the UK

Biomed. Mater. 5 (2010) 044109 Y I Woo et al

Introduction

β-glucans are a heterogeneous group of glucose polymersconsisting of β-(1,3)-linked β-D-glucopyranosyl units with aβ-(1,6)-linked side chain of varying distribution and lengththat mainly construct the outer cell walls of fungi and certainbacteria. They have been demonstrated to have immunestimulatory activity and to enhance wound healing especiallyby increasing macrophage infiltration into the injury sites andstimulating tissue granulation and re-epithelialization [1, 2].Several preclinical and clinical investigations have indicatedthe usefulness of β-glucans, a class of biological responsemodifier (BRM), in the acceleration of wound healing andinhibition of systemic inflammatory response syndrome andseptic shock [1–3]. These observations suggest that β-glucansand related immunomodulators may be useful adjuvant forhealing, particularly in burn wounds.

Kougias et al have reported the presence of at least twoglucan binding sites on normal human dermal fibroblastsother than on immunocytes like mammalian macrophages[4]. However, until now these kinds of investigation have notclarified the pathways of the effects of β-D-glucans on woundhealing in vitro, and also in vivo. Up to now, a few clinicalapplication cases have been reported, but there are no verifiedexplanations about how β-glucan affects wound healing [5].

There has been a recent surge of interest in the tissueengineering field to develop new materials/combinations forfabricating scaffolds with functional characteristics closelyresembling those of the tissues they aim to restore [6, 7]. Muchattention has been focused on designing implantable compliantbiodegradable polymer scaffolds that target acceleratingchronic wound repairs. Poly(D,L-lactic-co-glycolic acid)(PLGA) has been known as the most promising material anddiverse applications in tissue engineering have been found forits excellent biocompatibility, biostability and biodegradability[8, 9].

Electrospinning is a facile processing technique capable ofconverting polymers into inter-connected flexible nanofibrousstructures, and it is particularly appealing for fabricating largesheets of scaffolds suitable for dermal implantation [10].

In this study, we investigated the effects of β-glucanand the enhancement of cellular responses of adult humandermal fibroblasts (aHDFs) and adipose tissue-derived stemcells (ADSCs) on a porous electrospun PLGA membranecontaining β-glucan as an alternative therapeutic strategy fordamaged tissue.

Experimental details

(1,3)-(1,6)-β-D-glucan, cells and cell culture

Water-soluble (1,3)-(1,6)-β-D-glucan from Aureoubasidiumpullulans was obtained from Asahi Denka Co., Ltd (Tokyo,Japan) in a powder form (purity about 95%) as described inour previous study [11]. All reagents were purchased fromSigma-Aldrich Co. (St. Louis, MO) or GIBCO-BRL (GrandIsland, NY) unless otherwise noted.

aHDFs were obtained from Cambrex BioScienceWalkersville, Inc., and maintained in Dulbecco’s modified

Eagle’s medium (DMEM) containing 4.0 mM L-glutamine,1.5 g L−1 sodium bicarbonate, 4.5 g L−1 glucose, 1.0mM sodium pyruvate, 10% fetal bovine serum (FBS)and a 1% antibiotic antimycotic solution. ADSCswere isolated from freshly excised human subcutaneousfat tissue. Informed consent was obtained from eachdonor, and the study was approved by the Paik HospitalInstitutional Review Board. The isolated ADSCs weremaintained in DMEM/F12 containing 2.5 mM L-glutamine,1.2 g L−1 sodium bicarbonate, 10% FBS, 10 ng mL−1 hEGF,2 ng mL−1 hFGF and 1% antibiotic antimycotic solution.

Cell attachment, proliferation and migration by β-glucan

Cells, aHDFs and ADSCs, were plated in 24-well plates,treated with β-glucan and incubated for 4 h (with an initialcell density of 1.0 × 105 cells per well for an attachmentassay), 1, 3 and 5 days (with an initial cell density of 5.0 ×104 cells per a film for a proliferation assay). Afterthat, cells were incubated with 3-(4,5-dimethylthiazol-2-yl)-diphenyltetrazolium bromide (MTT) for 4 h and quantified bya colorimetric assay using a spectrophometer (Spectra Max340, Molecular Device Co., Sunnyvale, CA) at a wavelengthof 570 nm as previously described [12, 13].

A cell migration study using a previously describedmethod [12] attempted to evaluate the effect of β-glucanon the migration of aHDFs and ADSCs. Both cell typeswere plated at an initial density of 1.0 × 105 cells mL−1

in a 4-well chambered cover-glass slide and grown toconfluence overnight in the culture medium. Monolayers werewounded using a plastic micropipette tip, treated with β-glucan(1.0 mg mL−1) and incubated in a self-designed CO2 mini-incubator placed on a microscope stage for 36 h. Thewounded cells were visualized for migration of cells intothe denuded space by a charge-coupled device (CCD) cameraattached to the microscope (Olympus Optical Co. Ltd, Tokyo,Japan). The average speed of a single-cell migration wasanalyzed using the image-processing software MATLAB V7.0(MathWork Inc., USA).

Fibroblast-embedded collagen gel assay

To study the influence of wound healing on β-glucan-inducedcollagen gel contraction, fibroblast-embedded collagen gelwas prepared as previously described [14, 15]. Briefly, acollagen solution was prepared by mixing acid-soluble porcinetype I collagen (3 mg mL−1), fivefold concentrated DMEMand buffer solution (0.05 mol L−1 NaOH, 2.2% NaHCO3,200 mmol L−1 HEPES) at the ratio of 7:2:1:1, respectively.The collagen solution and aHDFs suspension were mixed (finalconcentration of type I collagen, 2.1 mg mL−1; final celldensity, 1.0 × 105 cells mL−1) in ice bath. One milliliter ofthe mixture (cell suspension of aHDFs in serum-free DMEMand collagen solution, final concentration, 1.0 × 105 cellsmL−1 and 2.1 mg mL−1 collagen) was plated in a 12-wellculture plate (Costar Corp., Cambridge, MA) and incubatedfor gelation at 37 ◦C for 30 min. One milliliter of serum-freeDMEM was then added on top of the gel to prevent the surfacefrom dehydrating. The gel was separated and floated from

2

Biomed. Mater. 5 (2010) 044109 Y I Woo et al

each well and then treated with the various concentrations ofβ-glucan after incubation for 12 h. After gel floating, themajor and minor axes of each gel sample were measured usingan electronic caliper (Mitutoyo Corp., Kawasaki, Japan), andthe NIH Image J software was used to quantify the areas ofthe collagen gels. The contraction of the gel was expressed inpercentages, with the surface area of the non-contracted stateserving as the norm (100%). Each value is equivalent to themean of the triplicate measurements.

Preparation and characterization of the porous PLGAmembrane and β-glucan grafting onto the scaffold

PLGA (75:25 (mol mol−1), MW 71 000, Alkermes,Inc., Cambridge, MA) membranes were prepared byelectrospinning, which is a facile processing technique capableof converting polymers into inter-connected flexible fibrousstructures and is particularly appealing for fabricating largesheets of scaffolds suitable for dermal implantation [16].

For the electrospun membranes, a series of PLGAsolutions was prepared by dissolving the copolymersin tetrahydrofuran:N,N dimethylformamide (8:2) at aconcentration of 20% (w/v), depending on the compositionof the copolymer. The polymer solution was delivered at aconstant flow rate (10 mL h−1) using an infusion pump to thestainless steel blunt-ended needle with an air gap between themetal collector and the needle tip of 20 cm at a driving voltageof 18 kV. β-glucan was grafted onto the surface of the PLGAmembrane as previously described [13]. In brief, the β-glucanwas dissolved in the phosphate buffer saline (PBS) solutionto form a 5 mg mL−1 solution. The PLGA membrane waspre-wetted with 70% ethanol for 5 min to be sterilized, andthen immersed in the β-glucan solution at 30 ◦C for 1 h. Themembrane was then dried.

Biological safety test of the porous PLGA membranecontaining β-glucan

Cellular toxicity test by the extract dilution method. Asdescribed by the document of International Standard ISO10993-5 [17], the MTT assay method was performed forcytotoxicity and the extracts of the PLGA membranescontaining β-glucan were prepared by shaking at 100 rpmfor 72 h at 37 ◦C and serially diluted by adding fresh DMEM(100%, 50%, 25%, 12.5%). For this study, the L-929 cells,mouse fibroblast cells (ATCC CCL 1, NCTC Clone 929, ofstrain L), were used. The cells were seeded into each well ofthe 24-well plate in triplicate and incubated at 37 ◦C for 24 hin order to obtain confluent monolayer’s of cells prior to use.All cultures were incubated for 24 h, under the same growthconditions. After incubation, each culture was stained with aMTT solution and lysed with a DMSO solution. Absorbancewas measured at 570 nm with an automatic microplate reader.The relative cell viability was expressed as a percentage of theoptical densities of the samples with the medium containingdiluted extracts to the optical densities of the samples grownin the fresh control medium.

MTT assays were used to estimate the cell attachmentand proliferation on PLGA membranes as described above.

The morphologies of cells grown on the PLGA membranecontaining β-glucan were observed after 5 days by a SEM(Hitachi S-800) at an accelerating voltage of 20 kV.

Intracutaneous (intradermal) reactivity test and sensitizationtest. Two biological tests in vivo were performed accordingto the document of International Standard ISO 10993-10 [18]and the method described by Magnusson et al [19].

For the intercutaneous reactivity test, a minimum of threehealthy adult albino rabbits (New Zealand White variety) ofeither sex were obtained from an approved supplier, traceablein Yonsei Medical Technology & Quality Evaluation Centerrecords. The animals were acclimated to the laboratory forat least 3 days. All animals weighed in excess of the 2.0 kgminimum ISO weight limit. The test article was extracted in0.9% sodium chloride and cotton oil with a ratio of 6 cm3 mLat 70 ◦C for 24 h. Each rabbit received three sequential0.2 mL intracutaneous injections on either side of the dorsalmid-line: test article extracts on the one side and concurrentvehicle controls on the other. Observations for erythema andedema were carried out at 24, 48 and 72 h after the injections.The primary dermal irritation index (PDII) was scored on a0–4 basis. Any adverse reactions at the test sites were alsonoted.

For the sensitization test, the method described byMagnusson was used to evaluate the allergic contactsensitization potential [19]. A guinea pig maximization testwas performed to evaluate the potential for delayed dermalcontact sensitization of the test article. A minimum of 20healthy adult albino guinea pigs of either sex were obtainedfrom an approved supplier, traceable in Yonsei MedicalTechnology & Quality Evaluation Center records. The animalswere acclimated to the laboratory for at least 5 days. The rangeof animal weights at first treatment was 300–500 g. A knownsensitizing agent, 1-chloro-2,4-dinitrobenzene (DNCB), wasused as the positive control. Ten test guinea pigs (per extract)were injected with the test article and Freund’s completeadjuvant (FCA), and five guinea pigs were injected with thecorresponding control blank and FCA (intradermal inductionphase). On day 6, the areas on the injection sites were treatedwith 10% sodium dodecyl sulfate (SDS). The day following theSDS treatment, the test animals were topically patched with theappropriate test extract and the control animals were patchedwith the corresponding control blank (topical induction phase).The patches were removed after 48 h of exposure. Following a2 week rest period, the test animals were topically patched ona previously untreated area with the appropriate test extract,while the same was done for the control animals using a controlblank (challenge phase). The patches were removed after 24h of exposure. The dermal patch sites were observed forerythema and edema 24 and 48 h after patch removal. Eachanimal was assessed for a sensitization response based upondermal scores.

Statistical analysis

All variables were tested in duplicate for each experiment,which was repeated twice. Quantitative data were expressed

3

Biomed. Mater. 5 (2010) 044109 Y I Woo et al

(A) (B)

(C) (D)

Figure 1. Effects of β-glucan on cellular responses, attachment and proliferation, of adult human dermal fibroblasts (aHDFs) and adiposetissue-derived mesenchymal stem cells (ADSCs). (A) Attachment test of aHDFs. (B) Proliferation test of aHDFs. (C) Attachment test ofADSCs. (D) Proliferation test of ADSCs. The results are shown as a mean ± standard deviation (n = 4) and the data were analyzed byStudent’s t-tests, and the values are significantly (p < 0.05) different from the non-treated control.

as a mean ± SD. Statistical comparisons were carried outwith a Student’s t-test. A value of p < 0.05 was consideredstatistically significant.

Results

Effects of β-glucan on cells attachment, proliferation and cellmigration

Figure 1 shows the effect of β-glucan on cellular responsessuch as attachment and proliferation. The β-glucan dose-dependently enhanced the proliferation of aHDFs for 5 daysand the proliferation rate was significantly (p < 0.05) higherin comparison with control (figure 1(B)). However, there wasno significant difference in attachment of aHDFs for 4 h(figure 1(A)). In ADSCs, β-glucan had no significantlydifferent effect on both the attachment and proliferation rateas shown in figures 1(C) and (D).

Cell movement is a combined effect of cell division andcell migration [20]. To examine the biological effects of β-glucan on the migration of aHDFs and ADSCs, wound healingmigration assays were performed as shown in figure 2. Themigration speed of cells in the β-glucan treated group wasfaster than in the non-treated group. Furthermore, the averagemigration speed of the β-glucan-treated (1 mg mL−1) aHDFwas 67.1 μm h−1, while that of the non-treated group was45.5 μm h−1 (figures 2(A) and (B)). In ADSCs, the cellmigration assay showed that β-glucan was able to induce themigration of cells, although it had no proliferative effect onADSCs as described above. The average migration speedof β-glucan-treated (1 mg mL−1) ADSCs was 86.3 μm h−1,while that of the non-treated group was 72.1 μm h−1 as shownin figures 2(C) and (D).

Collagen gel contraction assay

Three-dimensional collagen gels have been used as in vitrosystems for modeling of cellular activities during woundhealing [21]. This model system was adapted to assess thecapability of β-glucan to induce collagen gel contraction inconcert with cellular activities. These results are shown infigure 3. The collagen gels contracted strongly during 4 hincubation after the post-gelation treatment with β-glucan,whereas there was no obvious change in gel size in the control(figure 3(A)). The contracted collagen gel diameters aresummarized in figure 3(B). After incubation for 24 h, thepresence of the cell treated with β-glucan (1 and 2 mg mL−1)rendered collagen gel contraction to approximately 41.1 (p <

0.05) and 41.6% (p < 0.05) of its original size, respectively.In contrast, collagen gel without β-glucan contracted to72%.

Proliferation of cells on the β-glucan contained PLGAmembrane

To further analyze the ability of a scaffold to influence thegrowth of cells, aHDFs and ADSCs were cultured on β-glucan-containing PLGA membranes for 2 weeks and thecell viability was assayed with the MTT method at variouspoints along this time course (figure 4). The cell density onthe β-glucan-containing PLGA membrane was higher thanthat on the PLGA membrane without β-glucan in both cells.The cells in the membrane continued to proliferate up to theday 10. It is known as strong cell adhesion and spreadingon biomaterials [22]. Figure 5 shows the SEM images ofcell-free PLGA (figure 5(A)) and β-glucan-containing PLGA(figure 5(B)) and aHDFs on the PLGA (figure 5(C)) andβ-glucan-containing PLGA (figure 5(D)) membrane. Themorphologies of aHDF cultured on the PLGA membrane

4

Biomed. Mater. 5 (2010) 044109 Y I Woo et al

(A) (B)

(C) (D)

Figure 2. Effects of β-glucan on cell migration in adult human dermal fibroblasts (aHDFs) and adipose tissue-derived mesenchymal stemcells (ADSCs). (A) Real-time migration speed. (B) Average migration speed with and without β-glucan in aHDFs. (C) Real-time migrationspeed. (D) Average migration speed with and without β-glucan in ADSCs. Data are expressed as a mean ± standard deviation (n = 4) andanalyzed by Student’s t-tests. Statistical significance was considered as p < 0.1.

were spread morphology, whereas the cells on the β-glucan-containing PLGA membrane exhibited more spreadin the polygonal shape which is typical of the normal cellmorphology. The insets in figures 5(C) and (D) were magnifiedsections of the upper figures to observe the cells into the PLGAmembrane.

These characteristics are prerequisites for an effective re-epithelialization of artificial skin materials.

Biological safety test of the β-glucan-containing PLGAmembrane

Using the extract dilution method, the cytotoxicity of theβ-glucan-containing PLGA membrane was performed andthe test sample extract showed 91.3% of cell viability incomparison with the negative control as shown in figures 6(A)and (B). In the intracutaneous (intradermal) reactivity test,extracts of the β-glucan-containing PLGA membrane andnegative control solutions did not trigger irritation responsesat any circumstances as shown in table 1. No intracutaneousirritation was induced by the β-glucan-containing PLGAmembrane, and no toxic symptoms or abnormal behaviorssuch as convulsion, prostration or sign of biological reactivitywere observed at any injected sites with extracts or blankcontrols. Animals from all groups gained normal body weight

Table 1. The primary dermal irritation index (PDII) in rabbit for theintracutaneous reactivity test.

Sum of ObservationRabbit 24 h 48 h 72 h observation average

Total test scores 0 0 0 0 0/24Total control scores 0 0 0 0 0/24

Table 2. The dermal observation scoring by daily challengeobservations for the sensitization test.

Experimental 24 h 48 h

group Erythema Edema Erythema Edema Result

Positive control 1 0 2 1 PositiveNegative control 0 0 0 0 NegativeTest article 0 0 0 0 Negative

and appeared healthy at all times during the 72 h recoveryperiod. In the sensitization test, the test sample extract alsoshowed no significant evidence of causing delayed dermalcontact sensitization as shown in table 2.

Discussion

In order to evaluate the effects of β-glucan on wound healing,the cell proliferation and migration assays were carried out

5

Biomed. Mater. 5 (2010) 044109 Y I Woo et al

(A)

(B)

Figure 3. Effects of β-glucan on collagen gel contraction. (A) Representative photographs of collagen gel cultures of aHDFs incubated for4 and 24 h with the indicated concentration of β-glucan. (B) Time course and dose dependence of β-glucan-induced collagen gelcontraction. Cells were cultured in collagen gels for the indicated times in the absence or presence of β-glucan at concentrations of 0, 0.1,0.5, 1 and 2 mg mL−1, and the gels were monitored at 0, 4 and 24 h after gelation. Data are presented as means ± standard deviation ofrepresentative experiment. ∗p < 0.05 versus control (Student’s t-test).

with aHDFs and ADSCs, and the β-glucan-containing PLGAmembrane was evaluated by biological assays such as cellattachment, proliferation assays and biological safety tests.

In our studies, the cell proliferation was significantly(p < 0.05) enhanced by the β-glucan after 5 days and the β-glucan-treated group in a migration assay was faster than non-treated control. These results indicate that cells with impairedproliferative capacity were able to migrate by the β-D-glucan.This could be beneficial for healing of chronic dermal woundsby proliferation, as cell migration and proliferation are twomajor contributory factors enhancing the healing of chronicwounds [23].

The contraction of collagen gels by fibroblasts was firstreported by Bell et al and is a phenomenon useful in thestudy of cell to collagen interactions (collagen morphogenesis)[24, 25]. In this present study, we performedseveral experiments to study fibroblast-mediated collagengel contraction. These data suggest that β-glucan is

constitutionally well suited for application in dermal woundhealing.

Dermal fibroblasts play key roles in skin extracellularprotein turnover, extracellular matrix interaction, cell–cellcommunication, etc, which are closely related functions[26]. Prompting the wound healing process and identifyingfactors affecting wound healing after β-glucan exposure arecomplicated; therefore, it is challenging to develop newtherapeutic strategies to improve the cure rate. In the later partof the second stage of wound repair, new tissue formation,fibroblast migration and proliferation is an important event.This event is orchestrated by various growth factors producedby fibroblasts and other cutaneous cell types, includingplatelets, inflammatory cells, keratinocytes and epithelial cells[27].

The cell density on the β-glucan-containing PLGAmembrane was higher than that on the PLGA membranewithout β-glucan and biological safety was confirmed.The cell morphology and proliferation assay on the

6

Biomed. Mater. 5 (2010) 044109 Y I Woo et al

(A)

(B)

Figure 4. Proliferative effects of aHDFs (A) and ADSCs (B) onPLGA and β-glucan-containing PLGA membranes (∗p < 0.05versus the non-treated at the same time, analyzed by a Student’st-test, n = 4).

(A) (B)

(C ) (D)

Figure 5. SEM micrographs of cell-free PLGA (A) andβ-glucan-containing PLGA (B); aHDF cultured on PLGA (C) andβ-glucan-containing PLGA membrane (D). aHDFs were seeded at adensity of 1 × 104 cells and cultured for 10 days. After culture, thescaffolds were fixed for the SEM study. The insets are magnifiedsections of the upper figures to observe the cells into PLGAmembrane.

(A)

(B)

Figure 6. Cytotoxicity test of the β-glucan-containing PLGAmembrane. Relative cell viability of positive control, negativecontrol and test article extracts. (A) L-929 cells were monitored; (B)densitometric analysis of a MTT assay.

β-glucan-containing PLGA membranes showed that dermalfibroblasts well attached and moved in β-glucan-containingPLGA membranes without any cell adhesion facilitatingcomponent. It was distinctively different from the resultsof other studies describing various membranes in which cellsfailed to distribute uniformly in the matrices in conjunctionwith the majority of them attached only to the outersurfaces [28]. These results indicated that the β-glucan-containing PLGA membrane was non-toxic to cells and ithas good biocompatibility. Furthermore, the homogeneouscell distribution inside the porous β-glucan-containing PLGAmembrane resembles the normal dermal tissue, which is adesirable feature of dermal tissue engineered scaffolds aspreviously described [29].

In conclusion, our study suggests that the β-glucan andβ-glucan-containing porous PLGA membranes may be usefulas a material for enhancing the wound healing, because theycan promote cell proliferation, migration and collagen gelcontraction.

Acknowledgment

This study was supported by a grant of the Korea Healthcaretechnology R&D Project, Ministry of Health, Welfare andFamily Affairs, Republic of Korea (A084619).

References

[1] Portera C A, Love E J, Memore L, Zhang L, Muller A,Browder W and Williams D L 1997 Effect of macrophagestimulation on collagen biosynthesis in the healing woundAm. Surg. 63 125–31

7

Biomed. Mater. 5 (2010) 044109 Y I Woo et al

[2] Delatte S J, Evans J, Hebra A, Adamson W, Othersen H Band Tagge E P 2001 Effectiveness of beta glucan collagenfor treatment of partial thickness burns in childrenJ. Pediatr. Surg. 36 113–8

[3] Ross G D, Vetvicka V, Yan J, Xia Y and Vetvickova J 1999Therapeutic intervention with complement and beta-glucanin cancer Immunopharmacology 42 61–74

[4] Kougias P, Wei D, Rice P J, Ensley H E, Kalbfleisch J,Williams D L and Browder I W 2001 Normal humanfibroblasts express pattern recognition receptors for fungal(1→3)-β-D-glucans Infect. Immun. 69 3933–8

[5] Chen J and Seviour R 2007 Medicinal importance of fungalβ-(1→3),(1→6)-glucans Mycol. Res. 111 635–52

[6] Boland E D, Wnek G E, Simpson D G, Pawlowski K Jand Bowlin G L 2001 Tailoring tissue engineering scaffoldsusing electrostatic processing techniques: a study ofpoly(glycolic acid) electrospinning J. Macromol. Sci. Pure.Appl. Chem. A38 1231–43

[7] Matthews J A, Wnek G E, Simpson D G and Bowlin G L 2002Electrospinning of collagen nanofibers Biomacromolecules3 232–38

[8] Ramchandani M and Robinson D 1998 In vitro and in vivorelease of ciprofloxacin from PLGA 50:50 implantsJ. Control Release 54 167–75

[9] Gao J, Niklason L and Langer R 1999 Surface hydrolysis ofpoly(glycolic acid) meshes increases the seeding density ofvascular smooth muscle cells J. Biomed. Mater. Res.42 417–24

[10] Jiang H L, Fang D F, Hsiao B, Chu B and Chen W 2004Optimization and characterization of dextran membranesprepared by electrospinning Biomarcromolecules 5 326–33

[11] Lee D H, Han D-W, Park B J, Baek H S, Takatori K, Aihara M,Tsubaki K and Park J-C 2005 The influences of β-glucanassociated with BMP-7 on MC3T3-E1 proliferation andosteogenic differentiation Key Eng. Mater. 288–9 241–4

[12] Son H J, Bae H C, Kim H J, Lee D H, Han D-W and Park J-C2005 Effects of beta-glucan on proliferation and migrationof fibroblast Curr. Appl. Phys. 5 468–71

[13] Lee S G, An E, Lee J B, Park J-C, Shin J W and Kim J K 2007Enhanced cell affinity of poly(D,L-lactic-co-glycolic acid)(50/50) by plasma treatment with β-(1→3) (1→6)-glucanSurf. Coat. Technol. 201 5128–31

[14] Lee Y R, Oshita Y, Tsuboi R and Ogawa H 1996 Combinationof insulin-like growth factor (IGF)-1 and IGF-bindingprotein-1 promotes fibroblast-embedded collagen gelcontraction Endocrinology 137 5278–83

[15] Suhr K B, Tsuboi R and Ogawa H 2000Sphingosylphosphorylcholine stimulates contraction of

fibroblast-embedded collagen gel Br. J. Drematol.143 66–71

[16] Jiang H L, Fang D F, Hsiao B, Chu B and Chen W 2004Optimization and characterization of dextran membranesprepared by electrospinning Biomarcromolecules 5 326–33

[17] International Organization for Standardization 2009 Part 5:Tests for in vitro Cytotoxicity 3rd edn (Switzerland:International Organization for Standardization) ISO10993-5

[18] International Organization for Standardization 2002 Part 10:Tests for Irritation and Delayed-type Hypersensitivity2nd edn (Switzerland: International Organization forStandardization) 10993-10

[19] Magnusson B and Kligman A M 1969 The identification ofcontact allergens by animal assay: the guinea pigmaximization test J. Invest. Dermatol. 57 268–76

[20] Li W, Fan J, Chen M and Woodley D T 2004 Mechanisms ofhuman skin cell motility Histol. Histopathol. 19 1311–24

[21] Phillips J A, Vacanti C A and Bonassara L J 2003 Fibroblastsregulate contractile force independent of MMP activity in3D-collagen Biochem. Biophys. Res. Commun. 312 722–5

[22] Ratner B D, Hoffman A S, Schoen F J and Lemons J E 2004Biomaterials Science: An Introduction to Materials inMedicine 2nd edn (San Diego, CA: Academic) pp 260–81

[23] Clark RAF 1996 Wound repair: overview and generalconsiderations The Molecular and Cellular Biology ofWound Repair 2nd edn (New York: Plenum) pp 3–35

[24] Bell E, Ivarsson B and Merrill C 1979 Production of atissue-like structure by contraction of collagen lattices byhuman fibroblasts of different proliferative potential in vitroProc. Natl Acad. Sci. USA 76 1274–8

[25] Harris A K, Stopak D and Wild P 1981 Fibroblast traction as amechanism for collagen morphogenesis Nature290 249–51

[26] Lerman O Z, Galiano R D, Armour M, Levine J P and GurtnerG C 2003 Cellular dysfunction in the diabetic fibroblast:impairment in migration, vascular endothelial growth factorproduction, and response to hypoxia Am. J. Pathol. 162303–12

[27] Geoffrey C G, Sabine W, Yann B and Michael T L 2008Wound repair and regeneration Nature 453 314–21

[28] Wald H L, Sarakinos G, Lyman M D, Mikos A G, Vacanti J Pand Langer R 1993 Cell seeding in porous transplantationdevices Biomaterials 14 270–8

[29] Pan H, Jiang H and Chen W 2006 Interaction of dermalfibroblasts with electrospun composite polymer scaffoldsprepared from dextran and poly lactide-co-glycolideBiomaterials 27 3209–20

8