lable at ScienceDirect

Polymer Degradation and Stability 109 (2014) 226e232

Contents lists avai

Polymer Degradation and Stability

journal homepage: www.elsevier .com/locate/polydegstab

Regulating the degradation rate of silk fibroin films through changingthe genipin crosslinking degree

Renchuan You, Yamei Xu, Guiyang Liu, Yu Liu, Xiufang Li, Mingzhong Li*

National Engineering Laboratory for Modern Silk, College of Textile and Clothing Engineering, Soochow University, No. 199 Ren'ai Road, Industrial Park,Suzhou 215123, China

a r t i c l e i n f o

Article history:Received 17 April 2014Received in revised form20 July 2014Accepted 27 July 2014Available online 6 August 2014

Keywords:Silk fibroinDegradationGenipinCrosslinking

* Corresponding author. Tel.: þ86 512 6706 1150; fE-mail addresses: mzli@suda.edu.cn, mmzzllii@16

http://dx.doi.org/10.1016/j.polymdegradstab.2014.07.00141-3910/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

Degradation behavior is a key fundamental topic in the field of silk-based biomaterials. In this study,genipin-crosslinked silk fibroin films with varying crosslinking degrees were generated to investigate theeffect of the crosslinking degree on the degradation behavior. Higher crosslinking degrees resulted inincreased inter-intramolecular network crosslinking density through covalent bonds, which restrictedenzymatic attacks and the release of enzyme-degraded polypeptides from silk fibroin molecular net-works, providing greater resistance to enzyme degradation. Furthermore, genipin crosslinking inducedthe conformational transition from random coil to b-sheet due to the structural rearrangement of thechains to form covalent bonds. High b-sheet content contributes to a more crystalline structure that alsoenhances resistance to enzyme degradation. Consequently, changing the crosslinking degree had asubstantial impact on the degradation rate. The high-crosslinking films (with a crosslinking degreegreater than 90%) showed a similar degradability to 75% ethanol-treated films, which were barelydegraded by collagenase IA and subcutaneous implantation in SD rats. However, decreasing the cross-linking degree from 90% to 78% significantly increased the degradation ratio from 4 wt% to 18 wt% after28 days in vivo degradation, respectively. In particular, the degradability was strongly correlated to thecrosslinking degree in the low-crosslinking (lower than 90%) films. These results reveal that the cross-linking process can efficiently control molecular structures and regulate the crystallization of silk fibroinmaterials, in turn providing control of the degradation rate.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The desired scaffold degradation rate depends on the specifictissue engineering application. Ideally, the rate of scaffold degra-dation should mirror the rate of new tissue regeneration [1]. Silkfibroin (SF) is a promising biomaterial for tissue engineering andregenerative medicine due to its abundance, biocompatibility,mechanical robustness, and tunable degradation properties [2,3].The degradability of silk biomaterials depends on morphologicalfeatures, processing modes and b-sheet content [4e10]. The porestructure of silk materials and the molecular weight of regeneratedSF affect the degradation rate [4e6]; however, these profiles pro-vide only a narrow range of degradation regulation. A wider rangedegradation rates and tighter control are desirable for varied tissueregeneration needs; for example, a scaffold for bone repair requires

ax: þ86 512 6724 6786.3.com (M. Li).

29

a slow degradation rate, while rapid degradation is required fordermal tissue repair.

Natural silk fibers must be regenerated and processed intodifferent products such as films, nanofibers and porous scaffolds forvarious biomedical applications. In general, the regenerated SFmaterials are water-soluble and should be treated to induce water-resistant crystallization and chemical cross-linking. To induce b-sheet crystallization, several methods are used in SF biomaterials,such as organic solvent immersion and water vapor annealing[11e13], and the most commonmethod based on structural changeis the induction of methanol or ethanol immersion. The structuraltransition from random coil to the b-sheet results in aqueousinsolubility after treatment; however, a significant increase in b-sheet structure leads to lower degradation rates [7e9]. Further-more, these SF materials tend to be stiff and brittle in the dry state[14]. To maintain water-insolubility and improve the mechanicalproperties such as flexibility, chemical crosslinking reagents suchas genipin, polyethylene glycol diglycidyl ether (PEG-DE) and 1-ethyl-3-(3-dimethyl-aminopropyl) carbodiimide hydrochloride

R. You et al. / Polymer Degradation and Stability 109 (2014) 226e232 227

(EDC) are widely used to obtain water-insoluble SF materials fortissue engineering purposes [15e19]. However, none of thesestudies examine the effect of the crosslinking degree on degrada-tion rates in great detail.

Genipin is an effective natural crosslinking agent that can reactspontaneously with amino acids or proteins [20]. It has been re-ported that genipin-crosslinked biological tissues and biopolymersshow good mechanical properties and significantly reduced cyto-toxicity compared to synthetic crosslinking agents such as glutar-aldehyde and epoxy compounds [15,21]. Genipin has been used tocrosslink SF-based biomaterials such as films [21], porous scaffolds[15,16] and nanofibers [22,23]. In this study, SF films with varyingcrosslinking degrees were prepared by changing the genipin ratioto examine the impact of the crosslinking degree on the degrada-tion rate. These SF films had varied crosslinking densities and b-sheet contents, which are very useful in regulating the degradationrate. This method provides new approaches to modulate thedegradation rate of SF materials for various tissue repair needs.

2. Materials and methods

2.1. Preparation of regenerated silk fibroin films

A SF solution was prepared as described previously [4]. Briefly,Bombyx mori raw silk fibers (Huzhou, China) were boiled for 30minin an aqueous solution of 0.05% Na2CO3 and then rinsed in distilledwater to remove sericin. The extracted fibers were subsequently airdried and dissolved in CaCl2:CH3CH2OH:H2O solution (molar ratio1:2:8) at 72 �C ± 2 �C for 1 h. This solution was dialyzed againstdistilled water (MWCO 9e12 kDa) in cellulose tubes for 4 days. Theresulting SF solution was stored at 4 �C after filtration. To generategenipin-crosslinked SF films with varying crosslinking degrees,genipin (SigmaeAldrich) was added to the SF solution at 5%, 10%,20% and 30% of the SF weight in solution, respectively. Subse-quently, the mixture solutions were stirred slowly for 12 h at 37 �C.Next, 40 mL of solutions were cast into polyethylene plates at60 �C ± 2 �C for 2 h, and then further dried at room temperature for24 h to obtain genipin-crosslinked films. The pure SF films wereprepared at the same conditions and became insoluble aftertreatment with 75% ethanol for 2 h.

2.2. Determination of crosslinking extent

The crosslinking degree was determined by the ninhydrin assayaccording to a previously reported method [16]. The films wereweighed (0.05 g, n ¼ 3 per group) and immersed in 1.5 mL distilledwater for 1 h. Subsequently, a 450 mL 0.1% ninhydrin (Sigma-eAldrich) solution was added to each sample and heated at 100 �Cfor 20 min. The number of free amino groups in the sample wasproportional to the absorbance of the solution. The absorbance wasrecorded at 450 nm using a microplate reader (Bio-Tek Synergy HT,USA). Glycine solutions of various known concentrations were usedas standards. The crosslinking degree was expressed as the per-centage of reacted free amine number relative to the initial freeamine number.

2.3. In vitro enzymatic degradation

The SF films were weighed and incubated at 37 �C in PBS solu-tion (PBS; 0.05 M, pH 7.4) containing 1.0 U/mL Collagenase IA (fromClostridium histolyticum, EC 3.4.24.3, SigmaeAldrich). The samples(n ¼ 3 per time point) were incubated in enzyme solution (bathratio 1:50) for 1, 3, 6, 12, 18 and 30 days under slow shaking and inPBS under otherwise identical conditions as a control. The degra-dation solution was replaced with a fresh enzyme solution every 3

days. At the designated time points, the degradation products andremains were collected for analysis. The remaining samples wererinsed in deionized water and then dried at 105 �C to constantweight. Quantitative changes were expressed as the percentage ofweight retained relative to the initial dry weight.

2.4. The morphological and structural change

The surface morphologies of SF films after degradation for 0, 18and 30 days were observed by scanning electron microscopy (SEM;S-4800, Hitachi, Japan). For the molecular conformation measure-ments, Fourier-transform infrared (FTIR) spectroscopy analysis wasperformed using a Nicolet 5700 spectrometer (Thermo Scientific,USA). The secondary structure content of the SF films wasmeasured as described previously [4,24]. Briefly, Fourier self-deconvolution of the amide I region (1595e1705 cm�1) was per-formed using Opus 6.5 software (Bruker, Germany), and the Fourierself-deconvolution spectra were curve-fitted to measure the rela-tive areas of the amide I region components. Furthermore, X-raydiffraction (XRD) was performed to investigate the changes in thecrystal structure of the samples using an X-ray diffractometer(X'Pert-Pro MPD, PANalytical B.V. Holland) with Cu Ka radiation at40 kV and 30 mA and a scanning rate of 0.6/min.

2.5. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis(SDS-PAGE)

The degradation products at 6 days and 30 days were examinedusing SDS-PAGE. The samples were run on 5% and 8% poly-acrylamide gel in running buffer (0.25 M TriseHCl, 10% SDS, 0.5%bromophenol blue, 50% glycerol and 5% 2-mercaptoe-thanol, pH8.3). The stacking gel contained 5% acrylamide, 0.1% ammoniumpersulfate and 0.1% SDS in 1.0 M TriseHCl buffer (pH 6.8), and theseparating gel contained 8% or 10% acrylamide, 0.1% ammoniumpersulfate and 0.1% SDS in 1.5 M TriseHCl buffer (pH 8.8). Pre-stained protein served as the MW markers.

2.6. In vivo degradation

To study the relationship between the in vivo degradation andthe crosslinking degree, the 78%-, 90%- and 94%-degree-cross-linking films were implanted in SpragueeDawley (SD) rats, and the75% ethanol-treated films were used as controls. The animal ex-periments were in accordance with the Management Ordinance ofExperimental Animals of China ([2001] No. 545) and wereapproved by the Jiangsu Province in experimental animal man-agement rules ([2008] No. 26). The animals were divided into fourgroups of 5. The SF films were cut into 20 � 20 mm pieces andsubcutaneously implanted in the back of male SD rats (180e200 g,SPF grade). Pentobarbital sodium (30e60 mg/kg body weight) wasadministered pre-surgically. After shaving and disinfection, twoperpendicular incisions were created on the back of the rats, andthen a blunt probe was inserted through the incision to form asubcutaneous implantation space (approximately20 mm � 20 mm). The samples were implanted into the subcu-taneous space. The wounds were closed with 6-0 silk sutures andcovered by Vaseline carbasus and dry carbasus. At 28 days, thespecimens were harvested. A portion of harvested samples wereimmediately fixed in 4% formaldehyde in PBS at room temperatureand embedded in paraffin to cut tissue sections for hematoxylinand eosin (H & E) staining and optical microscope observation(Olympus BH-2, Japan). The other portion of harvested specimen(n ¼ 3 per group) were explanted and freed from the surroundingtissue. The specimenwere rinsed in deionized water and then driedat 105 �C to constant weight. The degradation ratio was expressed

R. You et al. / Polymer Degradation and Stability 109 (2014) 226e232228

as the percentage of reduced weight to the initial dry weight. Sta-tistical comparisons were performed using SPSS version 16.0 soft-ware (SPSS Inc., Chicago, Illinois), and p < 0.05 was considered to bestatistically significant.

3. Results

3.1. Crosslinking degree evaluation

The goal was to generate a series of SF films with a graded seriesof crosslinking degree using genipin. Fig. 1 shows the crosslinkingdegrees of silk fibroin films fixed with genipin at different ratios.The ninhydrin assay indicated that increasing genipin additionfrom 5 wt% to 10 wt% and 20 wt% significantly increased thecrosslinking degree from 68% to 78% and then to 90%, respectively.Furthermore, the crosslinking degree only increased to 94% whengenipin content increased from 20 wt% to 30 wt%, which suggestedthat the genipin ratio range from 5 wt% to 20 wt% tended to be themost important region to generate silk fibroin films with signifi-cantly different crosslinking degrees. The 68%- and 78%-degree-crosslinking films were classified as ‘low-crosslinking films’,whereas the 90%- and 94%-degree-crosslinking films were deemed‘high-crosslinking films’ in this study.

3.2. Morphological change

The morphological changes induced by enzymatic incubationwere studied by scanning electron microscopy (SEM). The surfaceof genipin-crosslinked SF films is smooth after degradation(Fig. 2(AeD)-0). Following exposure to enzyme solutions, thesurface morphology of the low-crosslinking films was drasticallyinfluenced by the enzymatic degradation. The films showed ahigher degree of surface roughness after 18 days of degradation(Fig. 2(A and B)-18). With further degradation, the 68%-degree-crosslinking film was drastically corroded at 30 days (Fig. 2(A-30)); meanwhile, some cracks and cavities caused by enzymaticdegradation were observed on the surface of the 78%-degree-crosslinking film (Fig. 2(B-30)). In contrast, no evident erosionoccurred on the high-crosslinking films incubated in CollagenaseIA for 30 days (Fig. 2(C and D)-30). The results indicated that low-crosslinking SF films were susceptible to enzymatic degradation,and the degradation level depended on the crosslinking degree.

Fig. 1. Effect of weight percentage of genipin on the crosslinking degree of silk fibroinfilms.

3.3. Mass loss

Enzyme degradation of the genipin-crosslinked SF films atdifferent degrees of crosslinking was studied in vitro. An uncros-slinked sample (pure SF film) and a 75% ethanol-treated film wereused as controls. Fig. 3 shows the mass loss of the silk fibroin filmsover the enzymatic degradation time. The untreated film can befully dissolved and degraded in 6 days. The mass loss of the low-crosslinking films significantly increased with the enzymaticdegradation time. After 30 days of degradation, themass remains ofthe 68%- and 78%-degree-crosslinking films were 13.27% and36.36%, respectively. However, the high-crosslinking films onlyshowed slight degradation of 6.95% and 5.28%, respectively, whichwere similar to the 6.45% of ethanol-treated films. These resultsshowed that the extent of mass loss depended significantly on thecrosslinking degree of SF films, and the enzymatic degradability ofSF films was slight when the crosslinking degree was above 90%.

3.4. Structural changes during in vitro degradation

The FTIR spectra were obtained to determine the conforma-tional transformation in the SF films after degradation. As shown inFig. 4a, the low-crosslinking films (A-0 and B-0) showed obviouspeaks at 1645 cm�1 (random coil, amide I), 1540 cm�1 (weak, a-helix, amide II) and 1236 cm�1 (random coil) before degradation,which were similar to the uncrosslinked film (line E). However, thehigh crosslinking films (C-0 and D-0) showed an obvious peak at1633 cm�1 in amide I, indicating the structural transition fromrandom coil to b-sheet in SF films with the increased crosslinkingdegree. After degradation for 30 days, the peaks at 1645 cm�1

(random coil) disappeared, whereas the strong peaks at 1627 cm�1

(b-sheet) appeared (Fig. 4a(AeD)-30). The same trend in structuralchange was also found in the amide Ⅲ region: the peak at1236 cm�1 (random coil) decreased, whereas peaks at 1268 cm�1

(a-helix) and 1228 cm�1 (b-sheet) appeared. The results indicatedthat the secondary structures of the SF films were mainly a-helixand b-sheet after degradation. Furthermore, the change of crystalstructure in the SF films was determined by XRD (Fig. 4b). Theuncrosslinked film (line E) and low-crosslinking films (A-0 and B-0)showed a scattering diffraction peak at 21.5�, whereas the 90%- and94%-degree-crosslinking films (C-0 and D-0) showed significantpeaks at approximately 19.8�. The results demonstrated that thestructure of SF films transformed from amorphous to crystalstructure with the increase in crosslinking degree. After degrada-tion for 30 days, the typical peaks at 20.4� (silk II), 24.3� (silk II) and31.6� (silk II) appeared (Fig. 4b(AeD)-30), indicating that theamorphous regions were preferentially degraded.

As shown in Fig. 5, the result of the fit covering the amide Iregion shows the contents of the b-sheet structures in SF films. Theb-sheet contents of the 68%-, 78%-, 90%- and 94%-degree-cross-linking films were 15.6%, 17.9%, 27.9% and 30.8% before degradation,respectively. These results showed that the b-sheet contentsincreased with increasing crosslinking degree. After 30 days ofdegradation, the b-sheet contents increased to 40.5%, 39.1%, 34.1%and 34.6%, respectively.

3.5. The molecular weight distribution of the degradation products

The molecular weight (MW) distribution of the degradationproducts at 6 and 30 days was determined by SDS-PAGE (Fig. 6).After 6 days of degradation (Fig. 6a), all samples showed multiplebands at 40e350 kDa. A 75 kDa band appeared in the lanes B, C andD, whereas a sequential band at 60e85 kDa appeared in lane A.Moreover, the bands in lanes A, B and C were stronger than in laneD, especially in the region above 200 kDa. As shown in Fig. 6b,

Fig. 2. SEM images of the silk fibroin films incubated in the Collagenase IA solution for 0, 18 and 30 days: (A) the 68%-degree-crosslinking films, (B) the 78%-degree-crosslinkingfilms, (C) the 90%-degree-crosslinking films, (D) the 94%-degree-crosslinking films. Scale bars: (A-0, B-0, C-0, D-0, A-18, B-18, C-18, D-18) 5 mm, (A-30, B-30, C-30, D-30) 10 mm.

Fig. 3. Enzymatic degradation of different crosslinking films: (A) the 68%-degree-crosslinking films, (-) the 78%-degree-crosslinking films, (△) the 90%-degree-cross-linking films, (,) the 94%-degree-crosslinking films, (C) the untreated films, (B) theethanol-treated films.

R. You et al. / Polymer Degradation and Stability 109 (2014) 226e232 229

sequential bands at 95e350 kDa were observed in all samples afterdegradation for 30 days, and a significant band appeared at 110 kDa.These results indicated that the SF films were degraded into poly-peptides by Collagenase IA and then released into the enzymesolutions.

3.6. Quantitative changes after in vivo degradation

The SF films were implanted subcutaneously into back of SD ratsfor 28 days. H & E staining images (Fig. 7a) showed that the low-crosslinking films (78%-degree-crosslinking) exhibited evidentdegradation, where numerous cracks and some fractures wereobserved (Fig. 7a, short arrows). The 90%-degree-crosslinking filmshowed slight degradation, where some cavities appeared (Fig. 7b,short arrows). In contrast, the degradation of the 94%-degree-crosslinking and ethanol-treated films was negligible, and thesefilms still maintained structural integrity (Fig. 7c and d). Addi-tionally, the quantitative changes after degradationwere calculated(Fig. 7e). The degradation ratio of the 78%-degree-crosslinking filmswas 18%, whereas the mass loss of the high-crosslinking films (90%-and 94%-degree-crosslinking) and the ethanol treated films were4%, 2.7%, and 2.8%, respectively.

Fig. 4. (a) FTIR spectra of different crosslinking films after enzymatic degradation for 0 and 30 days. (b) XRD data of different crosslinking films after enzymatic degradation for0 and 30 days. (A) The 68%-degree-crosslinking films, (B) the 78%-degree-crosslinking films, (C) the 90%-degree-crosslinking films, (D) the 94%-degree-crosslinking films, (E) theuncrosslinked films.

R. You et al. / Polymer Degradation and Stability 109 (2014) 226e232230

4. Discussion

To construct efficient tissue engineering scaffolds, their degra-dation rate needs to match the rate of tissue regeneration. Thedegradability of silk materials has been demonstrated in numerousin vitro and animal models [4e7,25e28]; the results revealed thatthe regenerated SF materials are biodegradable. The SDS-PAGE re-sults showed that the genipin-crosslinked SF films were initiallydegraded by Collagenase IA into polypeptides and released into theenzyme solution. The FTIR spectra and XRD confirmed that theamorphous regions were preferentially degraded by Collagenase IA,leading to an increase of b-sheet content after degradation. Previ-ous work has illustrated that the abundant -Xaa-Gly- sequence inthe non-b-sheet region of SF materials results in the rapid degra-dation of amorphous regions by Collagenase IA [4]. Furthermore,the degradation level of the SF films of low crosslinking degree

Fig. 5. Change of the b-sheet content in silk fibroin films with varying crosslinkingdegree after enzyme degradation for 30 days.

(68%- and 78%-degree-crosslinking) was compared with thedegradation level of the SF films of high crosslinking degree (90%-and 94%-degree-crosslinking). The surface roughness properties ofthe low-crosslinking SF films increased to a greater extent, whereasthe high-crosslinking films only showed a slight roughness. TheSDS-PAGE results also suggested that low-crosslinking films weremore readily degraded into polypeptides.

The genipin crosslinking mechanism of the amino-group con-taining compounds has been fully described. The nucleophilic at-tacks of the amino groups in the SF molecule initiate the opening ofthe genipin dihydropyran ring, and then the genipin-SF monomersself-polymerize by the radical reaction of two amino-attached openrings, leading to the formation of inter-intramolecular crosslinkingin SF molecules [15,21]. Covalent crosslinking yields a macromo-lecular network through the intermolecular and intramolecularcovalent bonds, limited swelling ratio and water content of proteinpolymer materials [29]. Moreover, the network properties can beeasily tailored by the concentration of the dissolved polymer andthe amount of crosslinking agent [30]. Higher crosslinking densityreduces the swelling ratio and resists enzymatic attacks due todecreased space between the macromolecular chains. Additionally,higher crosslinking density restricts the release of enzyme-degraded polypeptides from SF molecular networks. Therefore,increased crosslinking leads to a greater resistance to enzymedegradation. The XRD results indicated that amorphous contentdecreased due to the increased crosslinking degree, whereas crys-talline structure increased. The FTIR spectra also clearly showedthat the secondary structure of SF films transformed from randomcoil to b-sheet with increasing crosslinking degree, indicating thatgenipin crosslinking induced the conformational transition fromrandom coil to b-sheet through the structural rearrangement of SFmolecular chains to form covalent bonds. The amorphous regionsin SF films were readily degraded. In particular, the degradationrate of SF materials is correlated with the b-sheet structure content,where lower b-sheet content contributes to a less crystallinestructure that allows easier enzyme degradation [7,8]. The decon-volution of FTIR spectra in amide I showed that the b-sheet contentwas significantly correlated with the crosslinking degree. The b-sheet contents of the SF films increased from 15.6% to 30.8% as thecrosslinking degree increased from 68% to 94%. The result

Fig. 6. The SDS-PAGE of the degradation products from different films: (a) 6 days degradation, (b) 30 days degradation. (A) The 68%-degree-crosslinking films, (B) the 78%-degree-crosslinking films, (C) the 90%-degree-crosslinking films, (D) the 94%-degree-crosslinking films, (m) blank sample, (M) the molecular weight markers. The concentration of stackinggel was 5%. The concentration of the separating gel: (a) 10%, (b) 8%.

R. You et al. / Polymer Degradation and Stability 109 (2014) 226e232 231

suggested that the degradation rate of SF films could be furthermodulated based on the crystal structure through changing thecrosslinking degree.

In vitro degradation showed that the mass loss of SF films wassignificantly influenced by the crosslinking degree, especially thelow-crosslinking films. As shown in Figs. 3 and 5, the degradationrate followed a clear trend after exposure to Collagenase IA solu-tion: with higher crosslinking degree, a denser crosslinkingnetwork and more b-sheet crystals were formed in the SF films,which resulted in a slower degradation. After 30 days of degrada-tion, the remaining mass of the 68%- and 78%-degree-crosslinkingfilms were 13.27% and 36.36%, respectively, whereas the high-(larger than 90%) crosslinking films merely showed slight masslosses of 6.95% and 5.28%, which were similar to the results forethanol-treated films. Subcutaneous implantation is a commonlyused model to investigate in vivo degradation. The extracellularmatrixes of the dermis tissue contain abundant Collagenase IA,

Fig. 7. (a)e(d): H & E stains of the silk fibroin films after subcutaneous implantation in ratsfilms, (c) the 94%-degree-crosslinking films, (d) the 75% ethanol treated films. The longbars ¼ 100 mm. (e) The degradation percentages of the SF films (n ¼ 3, *p < 0.05, **p < 0.0

which is close to our in vitro model [4]. The trend of the in vivodegradation level in the SF films was consistent with in vitrodegradation. The low-crosslinking films (78%-degree-crosslinking)showed an 18% degradation ratio, whereas the mass loss of thehigh-crosslinking films was 4% and 2.7% for the 90%- and 94%-de-gree-crosslinking films, respectively. These results demonstratedthat the high-crosslinking films (higher than 90%) and ethanol-treated films were barely degraded by Collagenase IA and subcu-taneous implantation, whereas the degradation rate was stronglycorrelated to the crosslinking degree in the low-crosslinking films(lower than 90%-degree-crosslinking). Therefore, to improve thedegradation rate of SF materials to match the rate of tissue regen-eration, the degradation rate of SF equivalents could be regulatedthrough changing the crosslinking degree. This novel processingmethod provides a promising way to tune the degradation prop-erties of SF materials to match different tissue remodeling rates intissue repairs.

for 28 days for (a) the 78%-degree-crosslinking films, (b) the 90%-degree-crosslinkingarrows mark the undegraded films, and short arrows mark degraded sites. Scale

1, error bars indicate SD).

R. You et al. / Polymer Degradation and Stability 109 (2014) 226e232232

5. Conclusions

The effect of crosslinking degree on the SF materials was studiedbased on filmswith different crosslinking degrees. It was found thatthe degradation rate of SF films could be efficiently regulated viachanging the crosslinking degree. Higher crosslinking densityresisted enzymatic attacks and restricted the release of enzyme-degraded polypeptides from macromolecular networks, leadingto greater resistance to enzyme degradation. Furthermore, genipincrosslinking induced the conformational transition from randomcoil to b-sheet due to the rearrangement of chains to form covalentbonds, where increased b-sheet content lead to a lower degrada-tion rate. The change in crosslinking degree provided a wider rangeof regulation for the degradation rate of SF films. Thus, the SFbiomaterials with more controllable degradation properties couldbe prepared to match tissue regeneration rates for tissue repairsand regenerative medicine needs.

Acknowledgments

This work was supported by the National Nature ScienceFoundation of China (31370968), Nature Science Foundation ofJiangsu Province (BK20131177), College Natural Science ResearchProject of Jiangsu Province (12KJA430003) and Priority AcademicProgram Development of Jiangsu Higher Education Institutions.

References

[1] Drury J, Mooney D. Hydrogels for tissue engineering: scaffold design variablesand applications. Biomaterials 2003;24(24):4337e51.

[2] Shang K, Rnjak-Kovacina J, Lin Y, Hayden RS, Tao H, Kaplan DL. Acceleratedin vitro degradation of optically clear low b-Sheet silk films by enzyme-mediated pretreatment. Transl Vis Sci Tech 2013;2(3).

[3] Kundu B, Rajkhowa R, Kundu SC, Wang X. Silk fibroin biomaterials for tissueregeneration. Adv Drug Deliv Rev 2013;65(4):457e70.

[4] You R, Zhang Y, Liu Y, Liu G, Li M. The degradation behavior of silk fibroinderived from different ionic liquid solvents. Nat Sci 2013;5(6A):10e9.

[5] Wang Y, Rudym DD, Walsh A, Abrahamsen L, Kim HJ, Kim HS, et al. In vivodegradation of three-dimensional silk fibroin scaffolds. Biomaterials2008;29(24e25):3415e28.

[6] Makaya K, Terada S, Ohgo K, Asakura T. Comparative study of silk fibroinporous scaffolds derived from salt/water and sucrose/hexafluoroisopropanolin cartilage formation. J Biosci Bioeng 2009;108(1):68e75.

[7] Numata K, Cebe P, Kaplan DL. Mechanism of enzymatic degradation of beta-sheet crystals. Biomaterials 2010;31(10):2926e33.

[8] Hu Y, Zhang Q, You R, Wang L, Li M. The relationship between secondarystructure and biodegradation behavior of silk fibroin scaffolds. Adv Mater SciEng 2012. 2012, Article ID 185905.

[9] Kojthung A, Meesilpa P, Sudatis B, Treeratanapiboon L, Udomsangpetch R,Oonkhanond B. Effects of gamma radiation on biodegradation of Bombyx morisilk fibroin. Int Biodeterior Biodegr 2008;62(4):487e90.

[10] Zhao C, Wu X, Zhang Q, Yan S, Li M. Enzymatic degradation of Antheraeapernyi silk fibroin 3D scaffolds and fibers. Int J Biol Macromol 2011;48:249e55.

[11] Li M, Tao W, Kuga S, Nishiyama Y. Controlling molecular conformation ofregenerated wild silk fibroin by aqueous ethanol treatment. Polym AdvTechnol 2003;14:694e8.

[12] Seib FP, Maitz MF, Hu X, Werner C, Kaplan DL. Impact of processing param-eters on the haemocompatibility of Bombyx mori silk films. Biomaterials2012;33(4):1017e23.

[13] Hu X, Shmelev K, Sun L, Gil ES, Park SH, Cebe P, et al. Regulation of silk ma-terial structure by temperature-controlled water vapor annealing. Bio-macromolecules 2011;12(5):1686e96.

[14] Lu S, Wang X, Lu Q, Zhang X, Kluge JA, Uppal N, et al. Insoluble and flexible silkfilms containing glycerol. Biomacromolecules 2009;11(1):143e50.

[15] Vasconcelos A, Gomes AC, Cavaco-Paulo A. Novel silk fibroin/elastin wounddressings. Acta Biomater 2012;8(8):3049e60.

[16] Silva SS, Motta A, Rodrigues MT, Pinheiro AFM, Gomes ME, Mano JF, et al.Novel genipin-cross-linked chitosan/silk fibroin sponges for cartilage engi-neering strategies. Biomacromolecules 2008;9(10):2764e74.

[17] Wang J, Wei Y, Yi H, Liu Z, Sun D, Zhao H. Cytocompatibility of a silk fibrointubular scaffold. Mater Sci Eng C 2014;34:429e36.

[18] Yan S, Zhang Q, Wang J, Liu Y, Lu S, Li M, et al. Silk fibroin/chondroitin sulfate/hyaluronic acid ternary scaffolds for dermal tissue reconstruction. Acta Bio-mater 2013;9(6):6771e82.

[19] Zhang Q, Zhao Y, Yan S, Yang Y, Zhao H, Li M, et al. Preparation of uniaxialmultichannel silk fibroin scaffolds for guiding primary neurons. Acta Biomater2012;8:2628e38.

[20] Jin J, Song M, Hourston DJ. Novel chitosan-based films cross-linked by genipinwith improved physical properties. Biomacromolecules 2004;5(1):162e8.

[21] Wang L, Wang Y, Qu J, Hu Y, You R, Li M. The cytocompatibility of genipin-crosslinked silk fibroin films. J Biomater Nanobiotechnol 2013;4(3):213e21.

[22] Silva SS, Maniglio D, Motta A, Mano JF, Reis RL, Migliaresi C. Genipin-modifiedsilk fibroin nanometric nets. Macromol Biosci 2008;8(8):766e74.

[23] Zhang K, Qian Y, Wang H, Fan H, Huang C, Yin A, et al. Genipin-crosslinked silkfibroin/hydroxybutyl chitosan nanofibrous scaffolds for tissue-engineeringapplication. J Biomed Mater Res Part A 2010;95(3):870e81.

[24] Hu X, Kaplan DL, Cebe P. Determining beta-sheet crystallinity in fibrousproteins by thermal analysis and infrared spectroscopy. Macromolecules2006;39(18):6161e70.

[25] Li M, Ogiso M, Minoura N. Enzymatic degradation behavior of porous silkfibroin sheets. Biomaterials 2003;24(2):357e65.

[26] Lu Q, Zhang B, Li M, Zuo B, Kaplan DL, Huang Y, et al. Degradation mechanismand control of silk fibroin. Biomacromolecules 2011;12(4):1080e6.

[27] Zhou J, Cao C, Ma X, Hu L, Chen L, Wang C. In vitro and in vivo degradationbehavior of aqueous-derived electrospun silk fibroin scaffolds. Polym DegradStab 2010;95(9):1679e85.

[28] Yang Y, Zhao Y, Gu Y, Yan X, Liu J, Ding F, et al. Degradation behaviors of nerveguidance conduits made up of silk fibroin in vitro and in vivo. Polym DegradStab 2009;94(12):2213e20.

[29] Martinez AW, Caves JM, Ravi S, Li W, Chaikof EL. Effects of crosslinking on themechanical properties, drug release and cytocompatibility of protein poly-mers. Acta Biomater 2014;10(1):26e33.

[30] Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydro-gels. Adv Drug Deliv Rev 2012;64:223e36.