Pr

Ma

b

c

a

ARRAA

KSGEC

1

tefiebAta

iBr[swBa

0d

International Journal of Biological Macromolecules 46 (2010) 544–550

Contents lists available at ScienceDirect

International Journal of Biological Macromolecules

journa l homepage: www.e lsev ier .com/ locate / i jb iomac

reparation of Thai silk fibroin/gelatin electrospun fiber mats for controlledelease applications

anunya Okhawilai a, Ratthapol Rangkupanb,c, Sorada Kanokpanonta, Siriporn Damrongsakkula,∗

Department of Chemical Engineering, Faculty of Engineering, Chulalongkorn University, PhayaThai Road, Phatumwan, Bangkok 10330, ThailandMetallurgy and Materials Science Research Institute, Chulalongkorn University, PhayaThai Road, Phatumwan, Bangkok 10330, ThailandCenter of Innovative Nanotechnology, Chulalongkorn University, PhayaThai Road, Phatumwan, Bangkok 10330, Thailand

r t i c l e i n f o

rticle history:eceived 28 December 2009eceived in revised form 18 February 2010ccepted 19 February 2010vailable online 26 February 2010

eywords:ilk fibroin

a b s t r a c t

This work studied the effect of preparation conditions for electrospun fiber mats of Thai silk fibroin/type Bgelatin (SF/GB) for controlled release applications. The increasing in applied voltage resulted in decreasedaverage fiber size and narrow fiber size distribution. An increasing in silk fibroin content in blendedsolution resulted in larger size of the obtained electrospun fibers. Smooth fibers could be produced fromSF/GB blended solution at weight blending ratios of 10/90, 20/80, 30/70, 40/60 and 50/50. The resultson in vitro biodegradation test showed that SF/GB 10/90 electrospun fiber mat was rapidly degradedin collagenase solution due to direct biodegradation of gelatin by collagenase. From in vitro controlled

elatinlectrospinningontrolled release

release of two active agents (azo-casein and methylene blue) from SF/GB blended fiber mats, it wassuggested that methylene blue could be adsorbed on the blended fiber mats, possibly due to attractiveinteraction of the positively charged molecules of methylene blue and negatively charged SF/GB fibermats. In contrast, the same charge of blended fiber mats and azo-casein would result in the repulsiveforce, resulting in continuous diffusion of azo-casein from blended fiber mats within 72 h. The resultsindicated that SF/GB electrospun fiber mats had a high potential to be applied in controlled release

applications.

. Introduction

Electrospinning has been recognized as a simple and versa-ile process to fabricate various polymeric and inorganic fibers bylectrostatic force [1]. The applications of electrospun biomaterialbers have been explored for vascular graft, wound dressing, tissuengineering scaffolds and drug delivery systems. These applicationsenefit from the high surface area of the fibers and high porosity [2].mong biomaterials used for medical applications, gelatin is one of

he attractive biopolymers since it is inexpensive, biocompatiblend biodegradable.

Gelatin is a natural biopolymer derived from collagen foundn animal tissues such as skin, muscle, and bone. Type A and

gelatin microparticles were reported to use in the controlledelease of bone morphogenetic protein-2 (BMP-2) by Patel et al.3]. They found that type B gelatin microparticles were able to

ustain the release of BMP-2 due to the attractive interaction,hich depended on the opposite electrostatic charge of gelatin andMP-2. In contrast, type A gelatin which showed positive charget physiological pH, produced the repulsive force and caused the

∗ Corresponding author. Tel.: +66 2 218 6862; fax: +66 2 218 6877.E-mail address: siriporn.d@chula.ac.th (S. Damrongsakkul).

141-8130/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.ijbiomac.2010.02.008

© 2010 Elsevier B.V. All rights reserved.

rapid release of BMP-2. The fabrication of gelatin into ultra-finefibers by electrospinning has been recently reported [4,5]. Jeer-atawatchai [5] showed that both type A and B gelatins in formicacid could be spun into fiber. The presence of bead was found inboth types of gelatin at low concentration (2.5–30%, w/v). The uni-form fiber was produced when concentration reached 40%(w/v).The obtained gelatin fiber mats were crosslinked via various meth-ods including chemical and physical crosslinking. Among variouschemical crosslinking techniques, spraying and direct soakingwith 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochlo-ride (EDC) solution slightly reduced the weight loss of fiber matwhile maintained the crosslinking percentage of each gelatin fibermat compared to the direct soaking in the same solution. However,the crosslinked electrospun fiber mat showed rapid degradationrate in vitro, which was not suitable for controlled release applica-tions. Alternative material that is of interest for controlled releaseapplications is the blended fiber mat of silk fibroin and gelatin. Thepresence of silk fibroin is expected to prolong the degradation rateof gelatin.

Silk fiber has been used as biological sature for decades. Recentlysilk fibroin from silkworm sources including Thai silkworm Bombyxmori (Nangnoi Srisaket 1), are considered as a potential biomaterialfor a range of tissue engineering applications due to its non-toxicityto cell, good mechanical properties and slow degradation rate

f Biolo

[tfid

Taobb

2

2

Qlpt

2

mwa2d6wfiw

2

2fmst

2

ietwtp0tlvt1(wlftoi

M. Okhawilai et al. / International Journal o

6–10]. Silk fibroin fiber mats were successfully prepared by elec-rospinning from various solution [11–15]. The uses of silk fibroinlm in model drug release, for example nerve growth factor andextran were reported [16,17].

This study aims to develop the electrospun fiber of blendedhai silk fibroin and gelatin for long-term use in controlled releasepplications. The morphology, physical and biological propertiesf the blended fiber mats were investigated. The controlled releaseehavior of active agents (azo-casein and methylene blue) from thelended Thai silk fibroin/gelatin fiber mats was investigated.

. Materials and methods

.1. Materials

Thai silk strand, B. mori (Nangnoi Srisaket 1) was supplied byueen Sirikit Seficulture Center, Nakornratchasima province, Thai-

and. Type B gelatin with an isoelectric point (IEP) of 4–5 wasurchased from Nitta Gelatin Inc., Japan. Other chemicals used inhis study were analytical grade.

.2. Preparation of regenerated Thai silk fibroin

Regenerated silk fibroin solution was prepared using theethod described by Kim et al. [18]. In brief, silkworm cocoonsere boiled for 20 min in an aqueous solution of 0.02 M Na2CO3,

nd then rinsed thoroughly with water. This process was repeatedtimes to degum the silk fiber then they were dried overnight. Theegummed silk fibroin was then dissolved in 9.3 M LiBr solutions at0 ◦C and dialyzed against deionized water for 2 days. The dialysateas centrifuged at 4 ◦C for 20 min to remove insoluble debris. Thenal concentration of silk fibroin solution was 6.5 wt.%. The solutionas further freeze-dried to obtain regenerated silk fibroin.

.3. Preparation of the blended solution

The blended solution of Thai silk fibroin and type B gelatin at0 wt.% was prepared by dissolving freeze-dried silk fibroin in 99%ormic acid. The desired amount of gelatin was then added. The

ixture was stirred for 3 h to obtain homogeneous solution. Theilk fibroin/gelatin weight blending ratios were varied from 10/90o 80/20.

.4. Electrospinning of Thai silk fibroin/gelatin blended solution

The homogeneous solution of silk fibroin/gelatin was filled upn a 5 ml syringe connected with a needle (0.55 mm inner diam-ter). A high voltage in the range from 15 to 25 kV was appliedo the droplet of injected solution. A grounded rotating collectorith aluminum foil attached to the drum was placed at 20 cm from

he needle tip. The collector was rotated slowly to minimize anyreferential alignment of the fiber. Solution flow rate was fixed at.2 ml/h. The suitable blending weight ratios were chosen to elec-rospin for 30 h to obtain fiber mats. The collected fiber mats wereeft overnight in a dessicator to ensure complete removal of sol-ent. Thai silk fibroin/type B gelatin electrospun fiber mats werehen crosslinked using the method of spraying with and soaking in- ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochlorideEDC) and N-hydroxysuccinimide (NHS) solution. The fiber matsere sprayed with 14 mM EDC and 55 mM NHS dissolved in abso-

ute ethanol and then left for 15 min. The spraying was repeatedor 5 times before the blended fiber mats were further soaked inhe same solution for 2 h. The spraying step was introduced inrder to pre-crosslink gelatin and reduce its solubility when soak-ng in EDC/NHS solution [5]. After that, the fiber mats were rinsed

gical Macromolecules 46 (2010) 544–550 545

in phosphate buffer saline (PBS) pH 7.4 and deionized water priorto air-dry.

2.5. Characterization of Thai silk fibroin/gelatin solutions andfiber mats

2.5.1. Viscosity and conductivityThe viscosity and conductivity of the solution were measured by

Brookshield viscometer at the spindle speed of 50 rpm and Tetra-con 325 conductivity meter, respectively. The temperature of thesolution was controlled at 27 ± 1 ◦C

2.5.2. MorphologyMorphology and size of Thai silk fibroin/type B gelatin elec-

trospun fibers were observed under scanning electron microscope(SEM), JEOL, JSM-6400. Prior to observation, Thai silk fibroin/typeB gelatin electrospun fiber mats were carefully fixed on stubs andgold-coated using a JEOL JFC-1100 sputtering device. The size ofthe electrospun fibers was determined from the SEM micrographsusing SemAfore 4.0 software. The average fiber size was calculatedfrom 100 random measurements

2.6. In vitro biodegradation

In vitro biodegradation of blended fiber mats was evaluated intwo different types of solution at pH 7.4, 37 ◦C: PBS and 0.01 U/ml ofcollagenase solution (1 mg = 2.69 units, Sigma, St. Louis, MO). Aftersilk fibroin/gelatin electrospun fiber mats were exposed in eachsolution for a certain time period, they were removed and washedwith deionized water. The degraded fiber mats were then dried at37 ◦C in an oven before weighing.

The result of in vitro biodegradation was presented as thepercentage of remaining weight of dried electrospun fiber mats,calculated as follows:

Remaining weight (%) = Wf

Wi× 100 (1)

Wi represented the initial weight of the electrospun fiber mats(g) and Wf was the final weight of the same fiber mats afterexposed in degradation solution. The values were expressed as themean ± standard deviation (n = 3).

2.7. In vitro controlled release of methylene blue and azo-casein

Methylene blue (373.91 g/mol) and azo-casein (30,000 kDa)were used as model compounds for positively and negativelycharged drugs, respectively. The compound solution was loadedinto fiber mats by adsorption onto fiber mats. The blended fibermats were centrifuged to obtain homogeneous distribution ofsolution and were left at room temperature overnight to allowfull adsorption. The electrospun fiber mats were immersed under1.5 ml of PBS (pH 7.4) containing 0.01 wt.% sodium azide at 37 ◦C.At each time period (0 h, 2 h, 12 h, 24 h and 72 h), 200 �l of buffersolution was removed and replaced with the equal volume of freshbuffer. The fiber mats were digested with 1.5 ml of papain solu-tion at 70 ◦C to obtain complete release of compound at the finalstage. The amount of methylene blue and azo-casein released inPBS was determined using a spectrophotometer at the maximumabsorbance of 600 nm and 405 nm, respectively. The effect of load-ing amount (5 mg and 10 mg of compound per g of fiber mats) wasinvestigated.

The percentage of cumulative release was calculated as follows:

Ci =t∑

i = 0

Mi (2)

5 f Biological Macromolecules 46 (2010) 544–550

C

Mwr

3

3s

itifisfichad

Fb

46 M. Okhawilai et al. / International Journal o

umulative release (%) = Ci

CT(3)

i represented the amount of active agents released at time i, Cias the cumulative release at time i, CT was the total cumulative

elease after electrospun fiber mats were digested.

. Results

.1. Physical characteristics of the blended silk fibroin/gelatinolution

Viscosity of Thai silk fibroin/type B gelatin (SF/GB) solutionncreased from 351 centipoises in a neat gelatin solution to 475 cen-ipoises for SF/GB with 60 wt.% of silk fibroin content as presentedn Fig. 1. Gelation occurred when blended solution containing silkbroin content more than 80 wt.%. From visual observation duringpinning process at all applied voltage, the SF/GB solution with silk

broin content lower than 50 wt.% was easier to spin with largeroverage area of fiber deposition compared to the solution withigher silk fibroin content. From the result of conductivity in Fig. 1,s the silk fibroin content increased from 20 to 80 wt.%, the con-uctivity of the solution lowered from 11.25 × 10−4 to 6.3 × 10−4 s.

ig. 2. SEM micrographs of Thai silk fibroin/type B gelatin (SF/GB) electrospun fiber matsar = 5 �m).

Fig. 1. Solution properties of Thai silk fibroin/type B gelatin blended solution (in99% formic acid) at various Thai silk fibroin contents (wt.%): (�) conductivity (leftscale) and (�) viscosity (right scale).

3.2. The effect of applied voltage on morphology and distributionof Thai silk fibroin/gelatin fiber size

Morphology of the SF/GB electrospun fiber mats at applied volt-age of 15 kV, 20 kV and 25 kV was demonstrated in Fig. 2. Evidently,

at various weight blending ratios: applied voltage of 15 kV, 20 kV and 25 kV (scale

M. Okhawilai et al. / International Journal of Biological Macromolecules 46 (2010) 544–550 547

(Cont

nficafir(twt

TT

Fig. 2.

on-woven fiber mat arranged randomly with nanometer-scaledber diameter. The smooth and continuous fiber with no beadsould be produced from all compositions of the blended solutiont three different applied voltages. The sizes of SF/GB electrospunber at various blending ratios and applied voltage were summa-

ized in Table 1. For the solution with higher silk fibroin contents60–80 wt.%), the applied voltage did not tend to affect the size ofhe fiber. This may be attributed to a change in fiber morphologyith more ribbon-like fiber formation at these high silk fibroin con-

ents, making it more difficult to compare fiber size with different

able 1hai silk fibroin/type B gelatin (SF/GB) fiber size at various weight blending ratios and ap

SF/GB blended ratios The size of SF/GB fiber (nm) at various applied volta

15 kV 20

Min–Max Average Mi

10/90 117–206 162 ± 24 520/80 157–404 280 ± 66 630/70 153–373 258 ± 60 540/60 194–437 304 ± 65 550/50 195–427 296 ± 63 560/40 210–408 325 ± 44 1370/30 245–505 389 ± 68 1880/20 395–735 598 ± 83 10

inued ).

morphology in nature. The result showed that, for the solution con-taining 50 wt.% silk fibroin content or less, higher applied voltagetended to produce smaller fiber size.

3.3. The effect of silk fibroin content on morphology and size

distribution of Thai silk fibroin/gelatin fiber

Fig. 3 and Table 1 clearly showed the effects of silk fibroin con-tents on the size and morphology of the obtained SF/GB fiber. Forall applied voltages, the average fiber size tended to increase as the

plied voltages.

ges

kV 25 kV

n–Max Average Min–Max Average

8–248 133 ± 36 79–174 118 ± 214–343 172 ± 61 109–238 167 ± 356–456 211 ± 75 140–267 195 ± 327–444 240 ± 67 137–322 221 ± 438–414 265 ± 78 148–373 245 ± 566–526 371 ± 82 277–529 406 ± 711–512 380 ± 80 220–572 376 ± 812–891 557 ± 133 391–765 577 ± 98

548 M. Okhawilai et al. / International Journal of Biological Macromolecules 46 (2010) 544–550

Ffiv

stcosrAl

3e

iiwbtrwSfil

3

tpfiSifiA

Fig. 5. Remaining weight of Thai silk fibroin/type B gelatin (SF/GB) electrospun fibermats after degraded in phosphate buffer saline solution at 37 ◦C over 14-day period(�) SF/GB 50/50, (�) SF/GB 30/70 and (�) SF/GB 10/90.

Fs

ig. 3. Average fiber size and the standard deviation of the size of Thai silkbroin/type B gelatin (SF/GB)electrospun fiber mats obtained from various appliedoltages (�) 15 kV, (�) 20 kV and (�) 25 kV.

ilk fibroin content in the solution increased. The fiber size distribu-ion, on the other hand, tended to be narrower at lower silk fibroinontent. The fiber morphology appeared to be smooth with no beadn string fiber observed. However, for the blended solution havingilk fibroin content more than 50 wt.%, some fibers appeared to beibbon-like, which were usually greater in size than round fiber.s a result, the average size and size distribution of fiber became

arger and broader, respectively.

.4. Morphology and weight loss of silk fibroin/gelatinlectrospun fiber mats after crosslinking

The morphology of the blended SF/GB fiber mats after crosslink-ng by spraying and soaking in EDC/NHS in ethanol was illustratedn Fig. 4. The morphology of post-treatment electrospun fiber mats

as found to differ from the original. Comparing among threelended ratios, the SF/GB fiber mat at 50/50 was less swollenhan the others. Its fibrous structure with interconnectivity stillemained. Whereas, the SF/GB fiber mat at 10/90 was most swollenithout interconnectivity. Additional results on the weight loss of

F/GB fiber mats at 50/50, 30/70 and 10/90 after crosslinking wereound to be 2.02%, 4.04% and 8.17%, respectively. The increasingn gelatin content resulted in slightly higher percentage of weightoss.

.5. In vitro biodegradation

In vitro biodegradation of Thai silk fibroin/type B gelatin elec-rospun fiber mats was investigated in PBS solution over 14-dayeriod. As seen in Fig. 5, differences in weight loss of the blended

ber mats could be observed along incubation time. The weight ofF/GB electrospun fiber mats slightly decreased when incubatedn PBS for a longer time, except the fiber mats at 10/90. SF/GBber mats at 10/90 degraded more rapidly than the other two.fter 14-day of incubation, the remaining weights of SF/GB fiber

ig. 4. SEM micrographs of Thai silk fibroin/type B gelatin (SF/GB) electrospun fiber mats aolution (scale bar = 1 �m).

Fig. 6. Remaining weight of Thai silk fibroin/type B gelatin (SF/GB) electrospun fibermats after exposed in 0.01 U/ml of collagenase solution at 37 ◦C (�) SF/GB 50/50, (�)SF/GB 30/70 and (�) SF/GB 10/90.

mats at 50/50, 30/70 and 10/90 were 90.87%, 96.97% and 40.16%,respectively. As collagenase naturally presented in body fluid, invitro biodegradation behavior of Thai silk fibroin/type B gelatinelectrospun fiber mats in collagenase solution at 37 ◦C was alsoinvestigated.

Fig. 6 illustrated the remaining weight of SF/GB electrospun fibermats at various weight blending ratios after degraded in collage-nase solution at 37 ◦C for 32 h period. After 2 h of incubation, theweight of SF/GB electrospun fiber mats at 50/50 was decreased by

10% while that of the other two fiber mats at the blending ratiosof 30/70 and 10/90 was decreased rapidly by 40% and 50%, respec-tively. Considering the time needed to achieve 50 wt.% weight lossas the half-life of SF/GB fiber mats, it was found that the half-life ofSF/GB fiber mats at 50/50, 30/70 and 10/90 weight blending ratios

t various weight blending ratios (50/50, 30/70, 10/90) after crosslinking in EDC/NHS

M. Okhawilai et al. / International Journal of Biological Macromolecules 46 (2010) 544–550 549

Fig. 7. In vitro release of azo-casein from Thai silk fibroin/type B gelatin (SF/GB)e(1

wfiwS2ogi

3

moawbwaocw253tt

icpbl51a2tioi4

Fig. 8. In vitro release of methylene blue from Thai silk fibroin/type B gelatin (SF/GB)

lectrospun fiber mats in phosphate buffer saline over 72 h period (�) SF/GB 50/50,�) SF/GB 30/70 and (�) SF/GB 10/90; close and open marker represented 5 mg and0 mg of azo-casein/g of fiber mats, respectively.

as 32 h, 10 h and 2 h, respectively. The results revealed that SF/GBber mats with higher gelatin content degraded faster than the onesith lower gelatin content. After 32 h of incubation, the weight of

F/GB fiber mats at 50/50, 30/70, and 10/90 decreased to 50.71%,5.30% and 11.90% of their initial weight. Also, it can be obviouslybserved that the weight loss of fiber mats after exposed to colla-enase for 32 h was approximately the amount of gelatin containedn blended fiber mats.

.6. In vitro controlled release of active compounds

The release profile of azo-casein from SF/GB electrospun fiberats was illustrated in Fig. 7. Rapid burst of azo-casein was

bserved in all samples in the first 2–12 h indicating its excessmount which not adsorbed tightly on the fiber mats. The releaseas plateau at 12–24 h before the continuous increase. At the

eginning (0 h), no release of azo-casein was observed from alleight blending ratios of the blended fiber mats containing 5 mg

nd 10 mg of azo-casein per g of fiber mats. Considering the releasef azo-casein from fiber mats with loading amount of 5 mg of azo-asein per g of fiber mats, 17.90%, 13.19% and 6.86% of azo-caseinere released from SF/GB 50/50, 30/70 and 10/90 fiber mats afterh of incubation, respectively. The release of azo-casein from SF/GB0/50, 30/70 and 10/90 blended fiber mats was found to increase to6.07%, 23.70% and 38.28%, respectively at 24 h of incubation. Afterhat, it was continuously released from three blended fiber mats,o approximately 85% at 72 h of incubation.

The release profile of methylene blue in PBS (pH 7.4) at 37 ◦C wasllustrated in Fig. 8. It could be observed that the blended fiber matsontaining high loading amount of methylene blue showed highercentage of methylene blue released. Considering the releaseehavior of methylene blue from SF/GB fiber mats containing low

oading amount of methylene blue, at the beginning (0 h), SF/GB0/50, 30/70, and 10/90 electrospun fiber mats released 5.14%,4.54% and 22.73% of methylene blue, respectively. The percent-ges of cumulative release of methylene blue were increased to9.69%, 60.53% and 81.51% within the first 2 h. This result impliedhe excess amount of methylene blue, which could not be adsorbed

nto fiber mats and could be released rapidly. At 24 h, the releasef methylene blue from SF/GB fiber mats at 50/50, 30/70 and 10/90ncreased slightly to 38.90%, 68.05% and 85.68% and sustained at4.67%, 73.11% and 88.71% at 72 h after incubation, respectively.

electrospun fiber in phosphate buffer saline over 72 h period (�) SF/GB 50/50, (�)SF/GB 30/70 and (�) SF/GB 10/90; close and open marker represented 5 mg and10 mg of methylene blue/g of fiber mats, respectively.

4. Discussion

In this study, electrospun of SF/GB fiber mats could be preparedfrom the blended solution in formic acid at different weight blend-ing ratios. Differences in weight blending ratio resulted in solutionhaving various viscosity and conductivity successively affecting themorphology of the obtained electrospun fibers. The fiber obtainedfrom electrospinning of the solution containing higher silk fibroincontent, of which the viscosity is greater and the conductivity islower, had larger average size and broader fiber size distribution.In general, the solution viscosity can be regarded as a primary forcethat restraint the elongation of the charged jet segment, while thecoulombic repulsion force from surface charges is the main drivingforce for the elongation of the jet segment. Higher viscosity solutionrequires more force to stretch or elongate the jet segment apart tothe same degree as those with lower viscosity. The conductivity, onthe contrary, can be viewed as a parameter that favors the elonga-tion of the jet. For a given solution dielectric constant and chargemobility, a solution with higher conductivity could have a highersurface charge density which causes a higher columbic repulsivebetween jet segment, resulting in a more elongation and a smallerof fiber size.

It is known that the structures of regenerated silk fibroin andgelatin are random coil and helix conformation, respectively [19],which are water soluble and mechanically weak. This can limit theirapplications. Crosslinking by spraying and soaking fiber mats inEDC/NHS dissolved in ethanol was an effective method to main-tain fibrous structure of as-spun fiber. In addition, the solubility ofmaterials was very little as noticed from low percentage of weightloss. The first spraying EDC/NHS in ethanol solution onto fibermats allowed the partially crosslinked gelatin molecules and thenthe treatment by soaking fiber mats into EDC/NHS solution coulddirectly crosslink the residue gelatin molecules [20].

The biodegradation behavior of three blended electrospunfiber mats was first examined in PBS. It was noticed that silkfibroin/gelatin electrospun fiber mats at 50/50 and 30/70 coulddegrade slower than the one at 10/90. Owing to silk fibroin afterexposed in alcohol is �-sheet conformation and hydrophobic [21],the weight loss of the blended fiber mats after biodegradation test

was considered to be the weight loss of gelatin content. Mandalet al. [22] reported that the presence of crystalline silk fibroinacted as water insoluble backbone hindering the complete degra-dation of the blended silk fibroin/gelatin fiber mats. Furthermore,

5 f Biolo

UowdfiwdmmotTaoSactls

aboo[fwatfi(cebarscsdcfabewftTalfitrbalbAPos

[

[[

[

[

[[

[

[

[

[

50 M. Okhawilai et al. / International Journal o

lubayram et al. [23] stated that both bulk and surface degradationf gelatin rapidly occurred because gelatin is highly hydrophilic,hich corresponded to this present work. To simulate the degra-ation behavior in human body, the degradation of three blendedber mats was further investigated in collagenase solution. Theeight loss of fiber mats implied that gelatin component wasirectly hydrolysed by collagenase, while silk fibroin componentight resist to collagenase hydrolysis, due to its �-sheet confor-ation and hydrophobicity. This result corresponded to the report

f Jeeratawatchai [5]. She has investigated the in vitro biodegrada-ion of type B gelatin electrospun fiber mat in collagenase solution.he fiber mat was found to remain about 20% of its initial weightfter 4 h of degradation. Comparing to the degradation behaviorf SF/GB fiber mats in this study, the remaining weight of theF/GB fiber mats at 10/90 was 26.88% after 4 h of degradationnd about 11.89% after 32 h of biodegradation. It could be con-luded that the blended SF/GB fiber mats degraded slower thanhe pure gelatin fiber mat. The addition of silk fibroin could pro-ong the biodegradation of SF/GB blended fiber mats in collagenaseolution.

An advantage of nanofibers favorable for controlled releasepplications is their high surface to volume ratio. Active agents cane loaded at higher doses compared to materials prepared fromther techniques. Various techniques used to load active agentsnto fiber mats including adsorption, embeding and encapsulation24]. Adsorption of active agents on the fiber mats has advantageor stability of the active agents. A small volume of active agentsas used to ensure all active agents adsorped onto fiber mats. Two

ctive agents including methylene blue and azo-casein were inves-igated. The release behavior of azo-casein from SF/GB electrospunber mats could be explained based on electrostatic charge. In PBSpH 7.4), azo-casein with an isoelectric point of 6 was negativelyharged, same as both silk fibroin and type B gelatin with an iso-lectric point of 3 and 4–5, respectively [10]. The repulsive forceetween azo-casein and materials would cause the rapid release ofzo-casein from electrospun fiber mats along a period of 72 h. Thisesulted in continuous diffusion of azo-casein from SF/GB electro-pun fiber mats. For the controlled release of methylene blue, theharges of methylene blue and blended fiber mats would be con-idered to understand the release behavior of methylene blue. Asescribed above, silk fibroin and type B gelatin carried negativeharge, while methylene carried positive charge, resulting in theormation of electrostatic attractive force between methylene bluend fiber mats. From Fig. 8, the result suggested that there was aurst release of methylene blue during the first 2 h, indicating thexcess loading of methylene blue on fiber mats. Furthermore, itas observed that, at t = 0, the amount of methylene blue released

rom the fiber mats with higher silk fibroin content was less thanhat released from the fiber mats with lower silk fibroin content.his result indicated that three SF/GB blended fiber mats coulddsorp different amount of methylene blue. IEP of silk fibroin wasower than that of type B gelatin, an attractive force between silkbroin and methylene blue could be stronger than that betweenype B gelatin and methylene blue, resulting in less methylene blueeleased from fiber mats with higher silk fibroin content at theeginning. The release profiles which plateaued after 12–72 h inll samples revealed that there was very little diffusion on methy-ene blue from SF/GB fiber mats. This could be explained by the

iodegradability of these SF/GB fiber mats in PBS presented in Fig. 5.s the degradation of all fiber mats during the first 72 h (3 days) inBS was very low (less than 10%), the adsorbed methylene bluento fiber mats could still be sustained. This corresponded to theustained release of BMP-2 by type B gelatin microparticles having

[[[

[

gical Macromolecules 46 (2010) 544–550

opposite electrostatic charge to BMP-2 [3]. BMP-2 was suggestedto be released when the degradation of gelatin occurred.

5. Conclusions

The blended Thai silk fibroin/type B gelatin electrospun fibermats were successfully fabricated in this work. The applied volt-age and the amount of silk fibroin contained were found to affectthe morphology of blended fiber mats. A decrease in average fiberdiameter resulted from increasing in applied voltage and decreas-ing in silk fibroin content. The round and smooth fiber couldbe obtained when silk fibroin was less than 60 wt.%. In addition,EDC/NHS solution was shown to be effectively crosslink SF/GB fibermats, resulting in a controllable biodegradation of fiber mats. Fromthe study on controlled release of active agents, SF/GB electrospunfiber mats were shown to be able to attract positively chargedmolecule such as methylene blue and repulse negatively chargedagent such as azo-casein. This finding could be employed to desireSF/GB electrospun fiber mats for specific controlled release appli-cations.

Acknowledgements

The financial supports from the Thailand Research Fund-MasterResearch Grants (TRF-MAG), the 90th Anniversary of Chula-longkorn University Fund (Ratchadaphiseksomphot EndowmentFund) and Center of Innovative Nanotechnology, ChulalongkornUniversity are gratefully acknowledged.

References

[1] J. Doshi, D. Reneker, J. Electrostat. 35 (1995) 151–160.[2] H. Jin, S. Fridrikh, G. Rutledge, D. Kaplan, Biomacromolecules 3 (2003)

1233–1239.[3] Z. Patel, M. Yamamoto, H. Ueda, Y. Tabata, A. Mikos, Acta Biomater. 4 (2008)

1126–1138.[4] P. Songchotikunpan, J. Tattiyakul, P. Supaphol, Biol. Macromol. 42 (2008)

247–255.[5] H. Jeeratawatchai, Master Thesis, Faculty of Engineering, Chulalongkorn Uni-

versity, 2008.[6] T. Ari, G. Freddi, R. Innocenti, M. Tsukada, J. Appl. Polym. Sci. 91 (2004)

2383–2390.[7] U.J. Kim, J. Park, C. Li, H.J. Jin, R. Valluzzi, D. Kaplan, Biomacromolecules 5 (2004)

786–792.[8] R. Horan, K. Antle, A. Collette, Y. Wang, J. Haung, J. Moreau, V. Vollach, D. Kaplan,

G. Altman, Biomaterials 26 (2005) 3385–3393.[9] N. Vachairoroj, J. Ratanavaraporn, S. Damrongsakkul, R. Pichyangkura, T.

Banaprasert, S. Kanokpanont, Int. J. Biol. Macromol. 45 (2009) 470–477.10] J. Chamchongkaset, S. Kanokpanont, D.L. Kaplan, S. Damrongsakkul, Adv. Mater.

Res. 8 (2008) 685–688.11] C. Chen, C. Chuanbao, M. Xilan, T. Yin, Z. Hesun, Polymer 47 (2006) 6322–6327.12] J. Ayutsede, M. Gandhi, S. Sukigara, M. Micklus, H. Chen, F. Ko, Polymer 46 (2005)

1625–1634.13] C. Meechaisue, P. Wutticharoenmongkol, R. Waraput, T. Hungjing, N. Ketbum-

rung, P. Pavasant, P. Supaphol, Biomed. Mater. 2 (2007) 181–188.14] L. Chunmer, V. Charu, J.J. Hyoung, K.J. Hyeon, D.L. Kaplan, Biomaterials 27 (2006)

3115–3124.15] S. Zhou, H. Peng, X. Yu, J. Phys. Chem. B 112 (2008) 11209–11216.16] S. Hofmann, C.T. Wong Po Foo, F. Rossetti, M. Textor, G. Vunjak-Novakovic, D.L.

Kaplan, H.P. Merkel, L. Meinel, J. Control Release 11 (2006) 219–227.17] L. Uebersax, M. Mattotti, M. Papaloizos, H. Merkle, B. Gander, L. Meinel, Bioma-

terials 28 (2007) 4449–4460.18] U.J. Kim, J. Park, H.J. Kim, M. Wada, D.L. Kaplan, Biomaterials 26 (2005)

2775–2785.19] P. Songchotikunpun, J. Tattiyakul, P. Supaphol, Biol. Macromol. 42 (2008)

247–255.20] A.J. Kuijpers, G.H.M. Engbers, J. Feijen, S.C. De Smedt, T.K.L. Meyvis, J. Demeester,

J. Krijgsveld, S.A.J. Zaat, J. Dankert, Macromolecules 32 (1999) 3325–3333.21] F. Zhang, B.Q. Zuo, H.X. Zhang, L. Bai, Polymer 50 (2008) 1–7.22] B. Mandal, J. Mann, S. Kundu, Eur. J. Pharm. Sci. 37 (2009) 160–171.23] K. Ulubayram, A. Cakar, P. Korkusuz, C. Ertan, N. Hasirci, Biomaterials 22 (2001)

1345–1356.24] T. Sill, H. Recum, Biomaterials 29 (2008) 1989–2006.