Flexible silk fibroin films modified by genipin and glycerol

Abstract

Silk fibroin (SF) films, modified by genipin (GP) and glycerol (Gl), with favourable mechanical properties, were obtained by a casting/solvent evaporation method. Simultaneously, the chemical, mechanical and structural properties of the films were examined and analyzed. Compared to uncrosslinked SF films, fibroin solubility of the modified SF (MSF) and Gl/MSF films in the warm water decreased dramatically from 46% to 15%, exhibiting good stability under a physiological environment. The best, valuable modified films with tensile strength of 18.0 MPa, breaking elongation of 171.1% and Young's modulus of 463.1 MPa were obtained when the GP and Gl content were both 20 wt% and relative to the amount of fibroin. The deformability of the MSF films augmented significantly by increasing the Gl concentration. Fourier transform infrared (FT-IR) results revealed that GP could react with SF macromolecules to form inter- and intra-molecular conjugated covalent bonds. Moreover, the FT-IR and X-ray diffraction (XRD) studies illustrated the GP induced conformational transition from random coil to β-sheet SF chains, yielding MSF and Gl/MSF films with enhanced stable thermal stability. The cytocompatibility of the MSF films were evaluated through MTT assay using L929 fibroblast. Compared to the SF films treated with 75% ethanol, the MSF films exhibited significant cytocompatibility, which was demonstrated by cell adhesion, proliferation and cell morphology. The intrinsic properties and biological results suggest that the MSF films may be potential candidate materials for wound dressing applications or tissue engineering strategies.

---

Introduction

Soluble silk fibroin (SF) solution can be obtained by treatment with reagents such as CaCl₂·EtOH·H₂O, Ca(NO₃)₂·MeOH·H₂O or aqueous LiBr¹ and purified by dialysis against deionized water. SF exhibits advantages such as excellent biocompatibility, thermal stability, oxygen permeability and low inflammatory response. Previous reports illustrated that the SF materials has been utilized for wound protection, drug delivery systems, peripheral nerve regeneration materials, blood vessel engineering, bone tissue engineering and organic electronics.^2–7 The SF molecules show the characteristics of a random coil conformation in an aqueous solution, with an average hydrodynamic radius of 139 nm. Its molecular weight in heavy chain and light chain forms is ∼300 kDa and ∼26 kDa, respectively.⁸ Normally, the SF films can be prepared by casting the SF solution onto different substrates such as polytetrafluoroethylene, polyethylene, polystyrene, or glass. These films can be dissolved easily in water or solvents without additional treatment. In general, stable SF films can be obtained by means of physical methods, including high-temperature processing, methanol/ethanol treatment and the use of ultraviolet radiation.⁹ Though these methods are useful in stabilizing the SF film against water, the β-sheet-rich films are generally rigid and brittle in the dry state, causing difficulty in practical applications. In order to improve both stability and mechanical properties of the SF films, chemical reagents can also be adopted by crosslinking or blending. Experimental processing is normally implemented by the participation of bi-functional reagents such as glutaraldehyde (GTA),¹⁰ carbodiimides¹¹ and polyethylene glycol diglycidyl ether (PEGDE).¹² Nevertheless, it is noticeable that chemical reagents would be probably toxic if they were released into a host through biodegradation. Accordingly, it is necessary to meet the increasing demand for a crosslink agent to form stable and biocompatible crosslinked products without added cytotoxicity problems. Genipin (GP) is a natural crosslinking reagent with lower cytotoxicity, which can be isolated from the fruits of the plant Gardenia jasminoides Ellis and is obtained from geniposide via enzymatic hydrolysis with β-glucosidase.¹³ A series of studies indicated that GP has been adopted as a reagent for repairing biological tissues¹⁴ and crosslinking biomaterials containing amino groups such as collagen, chitosan and gelatine.^15–20 These products possessed higher cytocompatibility and stability. On the basis of former studies, Gl, a type of small molecule polyalcohol, can be used as a plasticizer to improve properties of films and is non-toxic to the human body. The films blended with Gl modified the mechanical properties, especially the elongation at break.²¹ The Gl can reduce phase separation between silk and PVA in the blend.²² These earlier studies substantiated the observation that blending Gl directly was beneficial for improving the properties of fibroin films.

Various silk-based materials, such as film, sheet, and scaffold, have been prepared to satisfy tissue engineering applications in different ways. Unfortunately, these silk materials produced from regenerated fibroin solution lose flexibility in the dry state, which generates more practical difficulties in some applications. In this study, we utilized GP and Gl to prepare novel flexible, stable SF films with less cytotoxicity. These films have been characterized by their mechanical properties of tensile strength, elongation at break, and Young's modulus by mechanical test, solubility in aqueous media by measuring fibroin solubility, conformational structures by FT-IR and XRD, thermal properties by thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) and biocompatibility by MTT assay.

Experimental

Materials

Bombyx mori raw silk fibers were purchased from Zhejiang the Second Silk Co. Ltd. (China). GP was purchased from Wako Pure Chemical Industries, Ltd. (Japan), methyl thiazolyl tetrazolium (MTT) and dimethyl sulfoxide (DMSO) were purchased from Sigma Chemical Company (USA). L929 cells, Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS) and penicillin/streptomycin were purchased from Shanghai Pufei Biotechnology Co. Ltd, (China). All other chemicals used in this study were purchased from Shanghai Sinopharm Chemical Reagent Co. Ltd. (China) and used without further purification.

Preparation of regenerated silk fibroin solution

Silk fibroin solution was prepared using a chemical degumming method before dissolution and dialysis. Raw silk fibers were treated three times in 0.205 wt% Na₂CO₃ solutions at 98 ± 2 °C for 30 min to remove sericin. After being air-dried, the refined silks were dissolved in a ternary solvent CaCl₂ : CH₃CH₂OH : H₂O (mole ratio = 1 : 2 : 8) at 78 ± 2 °C for 1 h. The mixed solution was then dialyzed in deionized water for 4 days to obtain fibroin solution with concentration of about 3 wt%.

Preparation of silk fibroin films

The modified SF solution was prepared by adding the GP directly to the 3 wt% SF solution in various weight ratios of 0, 5, 10, 12.5, 15, 17.5, 20, 22.5, 25, 27.5, and 30% w/w GP/SF. The Gl groups were also prepared by mixing the 20% w/w GP/SF solution and Gl in various weight ratios of 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100% w/w Gl/SF. Then, 40 mL mixed solutions were gently shaked in the thermostat at 37 °C for 12 h, poured separately into polystyrene plates of 10 cm × 20 cm and dried at 60 °C for 2 h. All the films were removed from the mould after keeping them at room temperature for 12 h. SF films without GP and Gl were used as control materials and were prepared according to the process described above.

Mechanical tests

Stress–strain curves of strip shaped (120 mm × 5 mm, thickness around 50 μm) films equilibrated at constant relative humidity of 65% for 24 h were performed on an INSTRON Testing Machine 3365 at standard conditions (20 °C and 65% relative humidity), and a crosshead speed of 20 mm min⁻¹. The thickness of the samples was determined using a Mitutoya thickness equipment ID-C112BS. The Young's modulus, E, was determined with a Y391 Yarn Elasticity Tester. The elongation value was obtained after a drawing load of 0.735 N was exerted on the samples for 5 s. The experiment was performed under the following conditions: pretension of 0.2 N and gauge length of 50 mm. The initial tensile modulus was calculated according to the eqn (1),where E is initial tensile modulus (MPa), P is the drawing load (N), L₀ is the specimen gripping length at pretension (mm), S is the specimen cross-sectional area (mm²) and ΔL is the elongation value in 5 s (mm). The measurement was repeated three or more times and averaged.

Degree of crosslinking

The degree of crosslinking of the MSF films was determined by ninhydrin assay.²³ The films were weighed (50 mg, n = 3 per group) and heated with ninhydrin solution (2 wt%) at 100 °C for 20 min. The number of free amino groups in the sample was proportional to the absorbance of the solution. The absorbance of the resulting solution was recorded at a wavelength of 570 nm using a Bio-Tek Synergy HT microplate reader. Glycine solutions of various known concentrations were used as standards, and SF films without GP were used as control materials. We can calculate the degree of crosslinking, D (%), by following eqn (2):where (NH₂)_UC and (NH₂)_C represent the mole fraction of free NH₂ remaining in uncrosslinked and crosslinked samples, respectively.

Stability of fibroin films

The stability of fibroin in the modified films was assessed through measuring the weight loss of the films in warm water. Take 100 mg of films baked at 105 °C to a constant weight measure, and calculate the moisture content, w (%), 0 < w < 1. Then, under the same conditions, measure the weight of the material, m₁ (mg). Deionized water was injected into a tapered bottle with 1 : 100 g mL⁻¹ bath ratio and each material was placed in water. After shaking the materials at 37 °C for 24 h, the mixture was filtered, the remains of the films were dried at 105 °C and weighed, m₂ (mg). Fibroin solubility, S (%), in water was determined by eqn (3):

Characterization

FT-IR spectroscopy. The SF films were cut into micro-particles with radii less than 40 μm, and then the samples were made into KBr pellets. Spectra were obtained on a Nicolet Avatar-IR360 with the frequency ranging from 400 to 4000 cm⁻¹.

XRD. XRD was performed by a Rigaku D/Max-3C diffractometer with Cu-Kα radiation (λ = 0.15418 nm). The X-ray source was operated at 40 kV and 40 mA. Diffraction intensity was measured in reflection mode at a scanning rate of 2° min⁻¹ for 2θ = 5–40°.

TGA and DSC. TGA and DSC measurements were performed with a Perkin-Elmer DSC-7. The temperature was lowered to room temperature and increased to 500 °C at a heating rate of 10 °C min⁻¹. The sample weight was 3–4 mg. The open aluminium cell was swept with N₂ during the analysis.

Cell culture and MTT assay

L929 cells were maintained in a complete DMEM medium with 10% FBS and 1% penicillin/streptomycin as described earlier.²⁴ All the samples were cast on 24-well tissue culture plates and irradiated with an exposure dose of 20 kGy by a Co⁶⁰ irradiator followed by extensive washing with sterile PBS (pH 7.4) prior to cell seeding. Each sample was made in triplicate. Cells were trypsinized, counted, and plated at a density of 5 × 10⁴ cells per well into films and grown for 1, 3, 5, 7 and 9 days. The plates were transferred to a 37 °C and 5% CO₂ incubator in 95% humidity and the medium was changed every second day. The 24-well plate without materials and the uncrosslinked fibroin films treated with 75% ethanol were used as blank and positive controls, respectively. On specified days, cell viability was evaluated using MTT assay. In brief, 100 μL of MTT (5 mg mL⁻¹) diluted 1 : 10 times in PBS was added to each well followed by incubation for 4 h at 37 °C. After incubation, 1 mL of dimethyl sulfoxide was added to dissolve the blue formazan product formed and the absorbance (OD values) was measured at 595 nm using a Bio-Tek Synergy HT microplate reader.

Results and discussion

Mechanical properties

Tensile testing of the films provides parameters for the strength and elasticity of the films given by the parameters such as tensile strength, elongation at break and Young's modulus. Mechanical properties provide information about the inherent character of the films. A value of moderate tensile strength, low elongation at break, high Young's modulus suggests that a polymeric film is hard and brittle; a value of moderate tensile strength, high elongation at break, low Young's modulus suggests that a polymeric film is soft and tough, and a hard and tough polymeric film is characterized by high elongation at break, high Young's modulus and high tensile strength.²⁵ A strong and elastic film in nature could expand the range of application in biomaterials.

To elucidate the effect of concentration of GP and Gl on the physical properties of SF films, tensile measurements were carried out. Fig. 1 shows the tensile strength, elongation and Young's modulus at the break of MSF films with different GP content and Gl/MSF films with different Gl content in the dry state. The typical stress–strain curves of the MSF films with 20% GP and the Gl/MSF films with 20% Gl are embedded in Fig. 1(c) and (d), respectively. The tensile strength of the uncrosslinked SF films was lower than the MSF ones. The changes of mechanical tensile strength with different GP concentrations indicated that mechanical strength increased first and then decreased a little by increasing GP content in films. Conversely, the addition of GP enhanced Young's modulus gradually (Fig. 1(c)). The Young's modulus and tensile strength of the MSF films crosslinked by 12.5 wt% GP were 1078.5 ± 83.3 and 55.9 ± 3.4 MPa, respectively, which was tougher than the other films, but the films still exhibited brittleness in the dry state. Fig. 1(b) and (d) showed that the increasing Gl levels resulted in a sharp decrease in tensile strength, a rapid increase in the elongation at break and a decline in Young's modulus of the Gl/MSF films. At about 20 wt% Gl, the Young's modulus, tensile strength and elongation at break of Gl/MSF films were 463.1 ± 54.4, 18.0 ± 3.1 MPa and 171.1% ± 5.3%, respectively, which became more soft and maintained its tensile strength at the same time. These results showed the small amount of Gl made MSF film turn flexible from brittle. A suitable amount of GP added to the SF solution leads to chemical reaction between SF molecular chains.²³ While the covalent bond with cyclic structure connected fibroin chains in the system improved the regularity in the films, which results in increase of the tensile strength, this cannot change the brittle property in the films. However, excess rigid crosslinking sites made films fragile and prone to cracks and flaws. As a result, introduction of Gl could overcome the disadvantage of brittle materials. Many studies indicated that the characteristic property of other polysaccharide or protein films was improved by being plasticized with different hydrophilic compounds.^26,27 Since addition of Gl, the hydroxyl groups of which interact with SF molecular chains, replaces water molecules in silk fibroin chain hydration and leads to larger gaps among SF molecules.²¹ Therefore, the softness and elasticity of the films increase due to the movement of molecular chains becoming easier. Moreover, the increased Gl content enhanced the hydrophilic properties in the films, which could form more bound water rather than free water from the environment to augment the plasticity. However, the elongation of the Gl/MSF films decreased a little at more than 60% Gl, which may be ascribed to reduction of random coil structures in the films followed by a decrease in the area of hydration between Gl and SF. The results mentioned above indicated that adding GP at around 15 wt% and Gl at 20 wt% was effective in improving the mechanical properties. These MSF films with tough characteristics are found to be suitable for biomaterials applications.

Fig. 1 Effect of GP and Gl content on mechanical properties of MSF films (a) tensile strength and elongation with different GP content, (b) tensile strength and elongation with different Gl content, (c) Young's modulus with different GP content, and (d) Young's modulus with different Gl content.

Stability of fibroin films

The fibroin solubility of the MSF films and Gl/MSF films (with varying levels of GP and different Gl content) in warm water is shown in Fig. 2(a) and (b). Without GP, very high solubility of the uncrosslinked silk fibroin was observed (46%); with only 5% GP (MSF films), the solubility dropped dramatically to 19%, and the mass loss stabilized at 15% GP and above. The amount of fibroin solubility was found to be dependent on the GP content in the films. These results suggested that the GP greatly stabilized the films against aqueous solvent. Compared to the uncrosslinked SF films, fibroin solubility of the Gl/MSF films also dropped dramatically with addition of Gl. Furthermore, the addition of Gl did not have obvious influence on those films. Slight increase in solubility was observed in Fig. 2(b) when Gl content exceeded 20 wt%. The trend of increase may be ascribed to the weight loss of superfluous Gl, which cannot attach the SF molecular through hydrogen bonds. In order to test the formation of covalent bonds in fibroin films, the degree of crosslinking under different conditions was evaluated using ninhydrin assay (Fig. 2(c)). After 3 h of reaction, a color change in the solutions was observed from light yellow to light blue, indicating the reaction between SF chains and GP. With the extension of reaction time, the fibroin solution eventually turned dark blue and crosslinked films with maximum crosslinking were obtained, as shown in Fig. 2. Many studies describe how GP reacts with amino acids or proteins to form dark blue pigments associated with the oxygen-radical polymerization of GP.¹⁷ The higher crosslinking degree related with stable films is obtained when the GP content reaches 12.5 wt%. The crosslinking reaction markedly reduces the fibroin degree of solubility of the SF films. The MSF films with degree of crosslinking above 85.08% ± 2.85% exhibited stability under a simulated physiological environment.

Fig. 2 Fibroin solubility of (a) MSF films with GP, (b) Gl/MSF films with Gl, (c) degree of crosslinking with GP.

Conformational structures analysis

FT-IR spectroscopy could characterize the secondary structure of proteins by examining the absorptions in the range of 4000–500 cm⁻¹. FT-IR spectra of uncrosslinked SF and MSF films modified by GP and Gl in the range of 4000–500 cm⁻¹ and partial enlarged drawing are shown in Fig. 3. SF protein exists in three conformations, namely, random coil, Silk I (α-helical) and Silk II (β-sheet) conformations. The strong absorptions at 1645 cm⁻¹ for amide I (C=O stretching), 1542 cm⁻¹ for amide II (N–H deformation), 1235 cm⁻¹ for amide III (C–N stretching, C=O bending vibration) were observed from the uncrosslinked SF film group (black line), which indicated it contained random coil and α-helical conformations.^28,29 The spectra of the MSF films (10 wt% and 20 wt% GP) showed shifts in the absorption band of amide I from 1645 cm⁻¹ to lower wavenumber (1635 cm⁻¹) reflecting the silk in the β-sheet structure in the MSF films after crosslinking by GP.²⁹ The amide I and amide II bands shifted to 1623 cm⁻¹ and 1230 cm⁻¹, respectively, in the red line and the intensities of these shifted peaks increased proportionally, illustrating augmentation of the β-sheet structure. Moreover, there appeared new absorption bands at 1107 cm⁻¹ (–CHO) in all the MSF films, which further verified that new covalent bonds appeared among silk fibroin molecules.²³ Moreover, the absorption band appearing at 1235 cm⁻¹ (amide III) represents a mixed vibration of CO–N and N–H. Owing to decrease of the –NH₂ on the lysine after reaction, this absorption band and characteristic absorption bands at 1334 cm⁻¹, 1162 cm⁻¹, and 1065 cm⁻¹ decrease after the addition of GP. The SF and MSF films show a peak at 3400 cm⁻¹; upon blending with Gl, this band shifted to lower wavenumber. Displacement of the peak can be associated with an interaction between fibroin and Gl. Moreover, the intensity at 3400 cm⁻¹ decreased with addition of Gl, which can be explained as there had been an increase in the specified intermolecular hydrogen bonds between SF chains.

Fig. 3 FT-IR spectra of uncrosslinked SF and MSF films.

The GP crosslinking of MSF films might induce conformational changes due to the structural rearrangement of chains to form covalent bonds. Some studies confirmed that GP crosslinking of SF is followed by protein conformational changes.²³ The exact mechanism behind the chemical reaction between SF and GP has not been fully described, but it is similar to that observed for amino-group containing compounds¹⁷ wherein amino groups of SF interact with ester groups of GP, leading to the formation of secondary amide linkages. The predicted reaction mechanisms of SF crosslinked by genipin may be similar to previous reports¹⁷ and illustrated in Fig. 4. At first, the amino groups in the SF initiate nucleophilic attack at C-3 of the six-membered ring, resulting in the opening of the GP dihydropyran ring followed by formation of a nitrogen-iridoid intermediate. Then, GP–SF intermediates self-polymerization occurs by radical reaction of two amino-attached rings, which creates highly conjugated heterocyclic genipin derivatives. SF proteins contain amino acids lysine and arginine, which could react with GP. Although the fraction of these amino acids is very low, increasing the amount of GP ensures the reaction takes place. The FT-IR results confirm the crosslinking reaction. The basic sequence of the crystalline block is of a hydrophobic group –(Ala–Gly)_n– that adopts a β-sheet structure and results in higher hydrophobicity.³⁰ The introduction of the Gl enhanced hydrophobic interactions in silk fibroin molecules leads to further structure transformation from random coil to β-sheet. It may be concluded that crosslinking reactions taking place between SF molecular and GP induce conformational changes in the MSF films. Increasing covalent bonds and β-sheet structure gave the MSF films attractive mechanical properties and stability.

Fig. 4 Schematic of the predicted mechanism of the reaction.

The conformational transformation of MSF films was further investigated using X-ray diffraction curves of the different SF films samples (Fig. 5). The uncrosslinked SF film showed arc-shaped scattering at around 20° and weak diffraction at around 12.2°, corresponding to the random coil and α-helical structures, respectively. There appeared to be an obvious characteristic peak at 20.7° in the MSF films with GP content equal to 10 wt%, assigned to the β-sheet structure.³¹ With increasing GP content, the red and blue curves show strong and sharp diffraction at around 20.7° and weak diffraction at around 24.3°, illustrating that the content of β-sheet conformation increased. Gl/SF films showed a broad peak between 9.1° and 12.2°, which was attributed to the superposition of the two types of crystallization peak. Furthermore, compared to uncrosslinked SF films, the peak at 20.7° showed much stronger intensities. This result indicated that the crosslinking reaction could induce the formation of β-sheet conformations from random coil and/or α-helical structures, which was consistent with the results of FT-IR spectra.

Fig. 5 XRD diffraction curves of uncrosslinked SF and MSF films.

Thermal analysis

Thermal analysis could reflect the interaction among SF, GP, and Gl. Thermal stability of different types of films was investigated using TG weight loss and DSC curves. The TG weight loss curves and differential coefficient curves are shown in Fig. 6. The major weight loss temperature (T_m) started at 285 °C for the uncrosslinked SF films, and at 288, 292, 289 °C for the 10 wt% GP crosslinked SF films, 20 wt% GP crosslinked SF films and 20 wt% Gl modified SF films with 20 wt% GP, respectively. These results showed that the MSF films exhibited higher thermal stability than uncrosslinked SF films because the crosslinking reaction and the increase of β-sheet. The weight loss within 150–200 °C may originate from the loss of bound water molecules due to the addition of Gl.

Fig. 6 TG and DSC curves of uncrosslinked SF and MSF films.

Fig. 7 shows the DSC of the different SF film samples. The DSC curves for the uncrosslinked SF films show one large endothermic peak at around 283 °C (T_d), attributed to thermal degradation of fibroin.³² The uncrosslinked SF films (black line) was quite different from the other samples in the presence of an exotherm transition at 218 °C, which was ascribed to the strong molecular motion within transformation of the random coil to the β-sheet structure during the heating process.³² The thermal behavior of uncrosslinked SF films was typical of an amorphous SF with a random coil conformation, as previously shown by FT-IR results. The MSF films crosslinked by GP showed significant reduction in intensity of the exotherm peak, which indicates the changes in the films from random coil to β-sheet after GP crosslinking. The increase in the thermal stability, given by the increase in T_d, was related to the increase in the extent of covalent crosslink. This fact is another confirmation of the crosslinking reaction between GP and SF. A new endothermic peak at 191 °C appeared in the Gl/MSF films (green line), which may be caused by degradation of hydrogen bonded water molecules, which is consistent with the TG results. The thermal analysis results underline the increase of structural regularity and thermal stability in the MSF films.

Fig. 7 DSC curves of uncrosslinked SF and MSF films.

Biocompatibility of SF films

The cell viability and proliferation on films in different days are indicative of the cellular compatibility and appropriateness for tissue engineering applications.³³

In this study, MTT assay was carried out as a preliminary approach to assess the biocompatibility of SF films. Morphology of L929 cells on the films and MTT results of days 1, 3, 5, 7 and 9 in different films are shown in Fig. 8. Compared to the blank control, both uncrosslinked and modified SF films can support L929 cells attachment and proliferation, because fibroin polymers have been extensively considered as non inflammatory and highly biocompatible for various cell types.³⁴ The morphology of L929 cells attached to the films displayed obvious differences at 3 days, which exhibited scalene triangle shape, the same as the blank control. MTT results showed the number of cells on all the films increased as a function of culture time. More importantly, the cellular viability of cells in MSF films with 20 wt% GP did not show a significant decline compared to uncrosslinked SF films and blank control. The OD values of the uncrosslinked SF films and Gl/MSF films showed a significant difference; the Gl/MSF group had less cells than the pure one after culture on day 3 and 5; however, this significant difference disappeared after 7 days. The observed tiny difference in OD values was probably due to release of surplus Gl in the films. The result suggested that the modified films exhibited good cytocompatibility and supported growth of L929 cells as comparable to uncrosslinked SF films, thus providing a good substrate for cell culture.

Fig. 8 (A) Micrographs of L929 cells on blank plate, SF films, MSF films and Gl/MSF films for 1 and 3 days, (scale bar, 100 μm). (B) Cell viability of L929 cells cultured on blank plate, SF films, MSF films and Gl/MSF films at 1, 3, 5, 7, and 9 days. Each point represents the mean ± SD (n = 3).*Significant difference between two groups at P < 0.05.

Conclusions

The objective of this study was to prepare SF films with flexibility, stability and biocompatibility. By modifying with different amount of GP and Gl, the effect of GP and Gl on mechanical properties, secondary structure, thermal stability and biocompatibility was determined. The following conclusions can be drawn from the present investigations.

(1) Excellent mechanical properties of SF films can be obtained through tuning the amount of GP and Gl. When the concentrations of GP and Gl were both 20 wt%, a flexible silk film was achieved with the Young's modulus of 463.1 ± 54.4 MPa, tensile strength of 18.0 ± 3.1 MPa and elongation at break of 171.1% ± 5.3%.

(2) As a result of new covalent bonds formed and augmentation of the β-sheet structure in the MSF and Gl/MSF films after GP crosslinking, all the MSF films exhibited an obvious decrease of fibroin solubility in a simulated physiological environment and a slight increase of thermal stability. These MSF films also facilitate the attachment and proliferation of L929 cells.

Considering expanding applications of fibroin-based biomaterials, the present study provides an effective way to fabricate the flexible, stable and less cytotoxic MSF films in vitro, which would further propel the applications of various tissue engineering applications.

Acknowledgements

This study was supported by the HongKong, Macao and Taiwan Science & Technology Cooperation Program of China (No. 2015DFH30180), the Key Grant Project of Chinese Ministry of Education (No. 313041), the Fundamental Research Funds for the Central Universities (WUT:2014-VII-028) and the Project of Hubei Engineering University (Z2014009).

Notes and references

1. X. Chen, D. P. Knight, Z. Shao and F. Vollrath, Polymer, 2001, 42, 9969.
2. Y. Zhou, H. Yang, X. Liu, J. Mao, S. Gu and W. Xu, Int. J. Biol. Macromol., 2013, 53, 88.
3. K. Numata, S. Yamazaki and N. Naga, Biomacromolecules, 2012, 13, 1383.
4. S. Madduri, M. Papaloizos and B. Gander, Biomaterials, 2010, 31, 2323.
5. M. Lovett, C. Cannizzaro, L. Daheron, B. Messmer, G. Vunjak-Novakovic and D. L. Kaplan, Biomaterials, 2007, 28, 5271.
6. C. Li, C. Vepari, H. J. Jin, H. J. Kim and D. L. Kaplan, Biomaterials, 2006, 27, 3115.
7. Y. Liu, N. Qi, T. Song, M. Jia, Z. Xia, Z. Yuan and B. Sun, ACS Appl. Mater. Interfaces, 2014, 23, 20670.
8. I. C. Um, H. Y. Kweon, G. L. Kwang and Y. H. Park, Int. J. Biol. Macromol., 2003, 33, 203.
9. G. M. Nogueira, A. C. C. Rodas, A. Leite, C. Giles, O. Z. Higa, B. Polakiewicz and M. M. Beppu, Bioresour. Technol., 2010, 101, 8446.
10. H. S. Mansur, E. S. Costa, A. A. Mansur and E. F. Barbosa-stanciol, Mater. Sci. Eng., C, 2009, 29, 1574.
11. J. Y. Lai and Y. T. Li, Mater. Sci. Eng., C, 2010, 30, 677.
12. P. Moonsri, R. Watanesk, S. Watanesk, H. Niamsup and R. L. Deming, J. Appl. Polym. Sci., 2008, 108, 1402.
13. S. D. Tommaso, H. David, J. Gomar, F. Leroy and C. Adamo, RSC Adv., 2014, 4, 11029.
14. J. Qiu, J. Li, G. Wang, L. Zheng, N. Ren, H. Liu, W. Tang, H. Jiang and Y. Wang, ACS Appl. Mater. Interfaces, 2012, 5, 344.
15. Y. J. Hwang, J. Larsen, T. B. Krasieva and J. G. Lyubovitsky, ACS Appl. Mater. Interfaces, 2011, 3, 2579.
16. N. Iwasaki, Y. Kasahara, S. Yamane, T. Igarashi, A. Minami and S. I. Nisimura, Polymer, 2010, 3, 100.
17. H. M. Chen, O. Y. Wei, L. Y. Bisi, C. Martoni and S. Prakash, J. Biomed. Mater. Res., Part A, 2005, 7, 5917.
18. F. Wu, G. Meng, J. He, Y. Wu, F. Wu and Z. Gu, ACS Appl. Mater. Interfaces, 2014, 6, 10005.
19. D. M. Kirchmajer, C. A. Watson, M. Ranson and M. Panhuis, RSC Adv., 2013, 3, 1073.
20. C. T. Turo, P. Gentile, S. Saracino, V. Chiono, V. K. Nandagiri, G. Muzio, R. A. Canuto and G. Ciardell, Int. J. Biol. Macromol., 2011, 49, 700.
21. S. Lu, X. Wang, Q. Lu, X. Zhang, J. A. Kluge, N. Uppal, F. Omenetto and D. L. Kaplan, Biomacromolecules, 2009, 11, 143.
22. L. Dai, J. Li and E. Yamada, J. Appl. Polym. Sci., 2002, 86, 2342.
23. S. S. Silva, A. Motta, M. T. Rodrigues, A. F. Pinheiro, M. E. Gomes, J. F. Mano, R. L. Reis and C. Migliaresi, Biomacromolecules, 2008, 9, 2764.
24. B. B. Mandal and S. C. Kundu, Macromol. Biosci., 2008a, 8, 807.
25. K. K. Peh and C. F. Wong, J. Pharm. Pharm. Sci., 1999, 2, 53.
26. I. S. Arvanitoyannis and C. G. Biliaderis, Carbohydr. Polym., 1999, 38, 47.
27. I. S. Arvanitoyannis, A. Nakayama and S. Aiba, Carbohydr. Polym., 1998, 37, 371.
28. M. Li, M. Ogiso and N. Minoura, Biomaterials, 2003, 24, 357.
29. Q. Lu, X. Hu, J. Wang, A. Q. Kluge, S. Lu, P. Cebe and D. L. Kaplan, Acta Biomater., 2010, 6, 1380.
30. X. Wu, J. Hou, M. Li, J. Wang, D. L. Kaplan and S. Lu, Acta Biomater., 2012, 8, 2185.
31. T. Asakura, R. Tabeta and H. Saito, Macromolecules, 1985, 18, 1841.
32. X. Hu, D. L. Kaplan and P. Cebe, Macromolecules, 2006, 39, 6161.
33. J. Han, Z. Zhou, R. Yin, D. Yang and J. Nie, Int. J. Biol. Macromol., 2010, 46, 199.
34. T. Yucel, M. L. Lovett and D. L. Kaplan, J. Controlled Release, 2014, 190, 381.

---