Composite of porous starch-silk fibroin nanofiber-calcium phosphate

Author's Accepted Manuscript

Composite of porous starch-silk fibroin nanofibercalcium phosphate for bone regeneration Zhina Hadisi, Jhamak Nourmohammadi, Javad Mohammadi

www.elsevier.com/locate/ceramint

PII: DOI: Reference:

S0272-8842(15)00939-6 http://dx.doi.org/10.1016/j.ceramint.2015.05.010 CERI10587

To appear in:

Ceramics International

Received date: Revised date: Accepted date:

12 March 2015 23 April 2015 4 May 2015

Cite this article as: Zhina Hadisi, Jhamak Nourmohammadi, Javad Mohammadi, Composite of porous starch-silk fibroin nanofiber-calcium phosphate for bone regeneration, Ceramics International, http://dx.doi.org/10.1016/j.ceramint.2015.05.010 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Composite of porous starch-silk fibroin nanofiber-calcium phosphate for bone regeneration Zhina Hadisi1, Jhamak Nourmohammadi1,*, Javad Mohammadi1 1- Department of Biomedical Engineering, Faculty of New Sciences and Technologies University of Tehran, P.O. Box 14395-1561, Tehran, Iran. *Corresponding author: Jhamak Nourmohammadi; PhD. Faculty of New Sciences and Technologies, University of Tehran, P.O. Box: 14395-1561,Tehran, Iran. Tel: +98-21-66118560 Email: [email protected]

Abstract New bioactive nanobiocomposite scaffolds based on silk fibroin nanofiber-porous starch are presented for potential bone tissue regeneration. The silk fibroin nanofibers were fabricated directly via wet electrospinning using methanol coagulation bath and then the chopped electrospun nanofibers were incorporated into the starch matrix, followed by particulate leaching and freeze-drying. To achieve bioactivity, the calcium phosphate was then deposited throughout the fabricated scaffolds via alternate soaking in saturated calcium and phosphate solutions at 37°C. The morphology, structure, swelling, and calcium phosphate forming ability of the scaffolds were evaluated and the results indicated that addition of silk fibroin nanofibers into the starch matrix reduced the mean pore size, porosity, and water uptake of the fabricated scaffolds. Moreover, the deposited calcium phosphate layer consists of both brushite and apatitic calcium phosphate. The highest amount of formed calcium phosphate is evident in the starch matrix and increasing the amount of silk fibroin nanofibers decreases calcium phosphate formation. Cell culture experiments with osteoblast-like cells (MG63) on the scaffolds coated with calcium phosphate demonstrated that incorporation of SF nanofibers into the starch hydrogel improves cell viability, proliferation, and attachment.

Keywords:

Porous starch; silk fibroin;

nanobiocomposite; bone regeneration.

wet electrospinning; calcium phosphate

deposition;

Composite of silk fibroin nanofiber-porous starch-calcium phosphate for bone regeneration

Abstract New bioactive nanobiocomposite scaffolds based on silk fibroin nanofiber-porous starch are presented for potential bone tissue regeneration. The silk fibroin nanofibers were fabricated directly via wet electrospinning using methanol coagulation bath and then the chopped electrospun nanofibers were incorporated into the starch matrix, followed by particulate leaching and freeze-drying. To achieve bioactivity, the calcium phosphate was then deposited throughout the fabricated scaffolds via alternate soaking in saturated calcium and phosphate solutions at 37°C. The morphology, structure, swelling, and calcium phosphate forming ability of the scaffolds were evaluated and the results indicated that addition of silk fibroin nanofibers into the starch matrix reduced the mean pore size, porosity, and water uptake of the fabricated scaffolds. Moreover, the deposited calcium phosphate layer consists of both brushite and apatitic calcium phosphate. The highest amount of formed calcium phosphate is evident in the starch matrix and increasing the amount of silk fibroin nanofibers decreases calcium phosphate formation. Cell culture experiments with osteoblast-like cells (MG63) on the scaffolds coated with calcium phosphate demonstrated that incorporation of SF nanofibers into the starch hydrogel improves cell viability, proliferation, and attachment.

Keywords:

Porous starch; silk fibroin;

wet electrospinning; calcium phosphate

deposition;

nanobiocomposite; bone regeneration.

1. Introduction Despite the progresses achieved in various fields of orthopedics, large bone defect therapy has remained a challenge. Over the past few decades bone tissue engineering (BTE) has been developed in order to regenerate and repair damaged sites with biological substitutes [1]. Beside the need for osteogenesis cells

and signaling agents, scaffolds play an important role in formation of neo-bone in BTE. The scaffold should consist of biocompatible and biodegradable materials with interconnected pores to increase oxygen permeability and nutrient supply to deeper areas. Moreover, osteoconductive properties of scaffolds facilitate osteoblast cells spread, growth and new bone generation [2]. Various synthetic and natural porous polymeric materials have been used extensively for bone tissue regeneration. Natural biodegradable polymers including polysaccharides and proteins are often used due to their cytocompatibility and similarity to extracellular matrix (ECM) of native tissues [3]. Among different kinds of natural polymers, starch is the most promising material in the field of biomedical engineering, since it is biocompatible, cost-effective, and widely available. However, the major disadvantages of the starch-based materials using starch taken alone for BTE purposes are their lack of processability and high water sensitivity [4]. In this regard, in order to enhance starch matrix performance in biomedical engineering and biotechnology fields, there is a growing trend towards incorporating natural polymeric fibers into the starch matrix, including bacterial cellulose [5], plant cellulose [6], and bamboo [7]. In addition to these approaches, fiber-reinforced composites have gained much popularity in recent years for BTE scaffolding due to its similarity to bone structure [8]. Nano-scale fibers have been suggested to be effective reinforcing agents because of their resemblance to the fibrous structures of bone tissue ECM and large surface area, which enhance cell attachment, proliferation, and growth [9]. In the past few years, electrospinning has been considered as an effective method for producing various polymeric nanofibers [10]. Thus, designing porous starch-based composite by incorporating electrospun biopolymer nanofibers appears to be a promising approach to improve its properties to mimic the bone ECM structure. Silk Fibroin (SF) is a kind of fibrous biopolymer with slow degradation rate, which shows remarkable biological and mechanical properties in both in vitro and in vivo conditions. It has been proven that the slow degradation rate of SF materials along with the high oxygen permeability make them suitable for slow regenerating of growing tissue such as bone [11]. Regarding these features, various forms of SF such as nanofibers are widely studied for bone tissue scaffolding [12]. Based on our knowledge, no previous study tested effects of SF nanofibers addition on the starch-based hydrogels. The further requirement for starch-based BTE scaffolds is the ability to bond with surrounding tissue. However, homogenous distribution of ceramic fillers in polymer matrix is challenging and weak binding

between ceramics and polymer have been reported. Coating of polymeric surface with various calcium phosphates (CaPs), especially apatite, is a common approach in order to achieve bioactive BTE scaffolds [13]. T. Taguchi et.al. [14] proposed that alternate soaking process is a simple and feasible method for homogenous coating of CaPs in three-dimentional porous polymeric structure due to its ability to control the thickness of formed apatite layer through changing reaction cycles. This study aims to develop and study a set of new bioactive composite scaffolds from silk nanofiber-starch hydrogel for bone regeneration applications. Accordingly, the SF nanofibers were first prepared by wet electrospinning using methanol bath and then added to starch matrix by using glutaraldehyde as a crosslinker. Afterwards, CaP was deposited throughout the fabricated porous composites via alternate soaking in saturated calcium and phosphate solutions at 37°C. Various techniques such as Scanning Electron Microscopy (SEM), X-ray Diffraction (XRD), and IR spectroscopy were used to analyze the structure and morphology of fabricated composites and the deposited CaP. Besides structural characterization, the biological response of MG63 osteoblast-like cells cultivated on the surface of starchbased scaffolds coated with CaP was evaluated. 2. Experimental 2.1. Materials Silk cocoons from B. mori were generously provided by the Iranian silkworm research center. Besides, lithium bromide (LiBr; 746479, Sigma, Saint Louis, USA), sodium carbonate (Na2CO3; 106392, Merck, Germany), methanol (822283, Merck, Germany), cellulose dialysis tube (12 KDa, MWCO, Sigma, Saint Louis, USA), glycine (104169, Merck, Germany), glutaraldehyde 25% (820603, Merck, Germany), isopropanol (818766, Merck, Germany), hydrochloric acid fuming 37% (HCl; 100317, Merck, Germany), absolute ethanol 99.7% (Hamoon Teb Markazi, Zarandieh, Iran), formic acid (100264, Merck, Germany), and potato starch (Golha, Iran) were purchased. 2.2. Preparation and characterization of SF electrospun nanofiber SF was extracted from B. mori cocoons based on previous reports [15]. Briefly, B. mori cocoons were boiled in 0.02 M aqueous Na2CO3 solution for 1 h to remove sericin, washed with deionized water and dried at 37°C overnight. The degummed silk was dissolved in 9.3 M LiBr at 60 °C for 4 hours, dialyzed in a cellulose tube against distilled water for 72 hours and then lyophilized (MC4L, UNICRYO, Freeze-dryer,

Germany). The prepared SF sponge was then dissolved in formic acid under stirring for 3 hours at room temperature to yield 13 wt% SF solution. Subsequently, the obtained solution was filled in 1 ml syringe with 22 G blunted stainless steel needle. Electrospinning (Fanavaran Nano-Meghyas, Iran) was carried out at 20 kV with a constant flow rate of 0.25 ml.h-1. The electrospun fibers were collected in the methanol coagulation bath for direct recovery of ȕ-sheets. The methanol bath was placed 13 cm under the needle tip. To determine morphology and size, the fiber mat was attached on an aluminum stub, sputter coated with gold, and imaged with a Scanning Electron Microscope (SEM; Essen Philips XL 30). SEM images from approximately 50 random fibers were analyzed to determine the average fiber diameter and their distribution using the image analysis software (ImageJ; National Institutes of Health, Bethesda, Maryland, USA). Fourier Transform Infrared Spectroscopy in the Attenuated Total Reflectance mode (FTIR-ATR, Equimo55 bruker FTIR spectrometer) was used to examine the SF structural changes during electrospinning in methanol coagulation bath. The quantitative analysis of amide I absorption band within the 1600-1700 cm-1 region [16, 17], was done using curve-fitting and de-convolution methods to determine the amount of Į-helix, ȕ-sheet, and random coil elements in each spectrum. 2.3. Preparation and characterization of SF nanofiber-starch composite scaffolds The three-dimentional (3D) porous nano-biocomposites from pure starch and SF nanofibers were prepared via particulate leaching and freeze drying methods. Potato starch (12 g) and different amounts of chopped electrospun SF nanofibers were added to 100 ml distilled water containing 75 µl ml-1 glutaraldehyde and stirred thoroughly for 20 minutes. Starch and SF nanofibers were mixed with different weight ratios of 100/0, 95/5, 90/10, and 85/15. Afterwards, the pH of the solution was adjusted to 2 by adding drops of 1 N HCl to the SF nanofibers-starch suspension and heating it with continuous stirring at 80oC. Prior to gelatinization, 65 wt% sugar (150-350 µm) was added to the paste, stirred carefully and then poured into 24-well tissue culture polystyrene plates. The casted mixture was cooled in refrigerator at 4°C, frozen at -20 o C for 12 hours and then freeze-dried at -55 o C for 48 hours. In order to leach out the sugar, prepared samples were soaked in isopropanol (70%) for 4 days. Next, the samples were rinsed with 1% glycine solution to remove uncrosslinked glutaraldehyde and then washed several times with distilled water. The structural analysis of each scaffold after crosslinking by glutaraldehyde was investigated by FTIR-ATR. The internal morphology of starch-based scaffolds was observed by SEM. Using Image J software, the size

of the pores in each sample, the total area of the pores in each cross section (AP), and the total area of each cross section (AT) were measured. The measurements were carried out from the five SEM images of each sample (n=5). Next, the porosity percentage of the fabricated scaffolds was calculated using the following equation [18]:

Porosity (%) =

AP AT

Eq. (1)

As mentioned in the previous study, the swelling ratio and porosity of hydrogels have an effect on the amount of biomimetic CaP deposition [14]. For determination the swelling ratio (SR), three disks (n=3) of each composition with the initial weight of (Wo) were soaked in deionized water (T=37 oC) for several hours and allowed to completely swell. Afterwards, the excess water was removed by filter paper and then weight (Ws). The SR (%) was measured based on the following equation [14]:

SR(%) =

Ws − Wo × 100 Wo

Eq. (2)

2.4. Deposition of CaP on/in starch-based scaffolds and characterization CaP deposition throughout the fabricated scaffolds was evaluated based on alternate dipping process [14]. Briefly, the cylindrical samples were dipped into 10 ml calcium-rich solution (0.5 M CaCl2, pH= 7.4) at 37°C for 1 hour, washed rapidly with deionized water, and immediately dipped into 10 ml phosphate-rich solution (0.3 M Na2HPO4, pH= 8.5) at 37°C for another 1 hour. After repeating each cycle for 3 times, the samples were rinsed thoroughly by deionized water, vacuum-dried and kept in desiccator before analysis. The code and composition of the fabricated starch-based scaffolds before and after CaP deposition are presented in Table 1. The morphology of formed CaP layer throughout each sample was observed by SEM. Energy Dispersive Spectroscopy (EDS) in SEM was used to determine the Ca/P ratio of deposited CaP. The phase composition of deposited CaP was determined by X’Pert Pro MPD X-ray diffractometer using (CukĮ radiation, 40 kV and 30 mA). 2ș angles ranged from 0o to 40o using 0.02o step and 20s counting time. Moreover, the changes in chemical groups of each scaffold after CaP formation were determined by FTIRATR. Changes in pH value of the calcium and phosphate rich solutions after each immersion were

measured using a pH meter (ȍMetrohm 827). In addition, the amount of CaP deposited (CP-D) in each sample was determined by Eq. (3) [14]:

§ Wo · CP-D(mg/gel)=W1 - ¨ ¸ © 1+SR ¹

Eq. (3)

W1 and W2 are the weights of freeze-dried and wet sample after alternate soaking process respectively. SR is the swelling ratio of each sample, which was calculated based on Eq. (2). All measurements were carried out for three samples (n=3). 2.5. Cell culturing on the scaffolds coated with CaP Human osteoblast-like cells, MG63, from National Cell Bank of Iran (NCBI; Pasteur institute) were maintained in Dulbecco’s Modification of Eagles Medium (DMEM) containing 10% Fetal Bovine Serum (FBS; Gibco, Renfrewshire, Scotland), 100 U/mL penicillin, and 100 ȝg/mL streptomycin at 37 °C in humidified incubator with 5% CO2. Triplicate specimens (n=3) of each specimen were put into petri dishes, disinfected via interval immersion in ethanol and Phosphate Buffered Saline (PBS) for 5 times, and then washed twice with culture medium before cell seeding. The indirect 3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium-bromide (MTT, Sigma, Saint Louis, USA) assay based on extraction method was used to evaluate the relative viability or cell growth (%) as follows [19]: Eq. (4) ODs is the average optical density of each sample and ODc is the average optical density of the control group. Accordingly, culture medium was added to the pre-weighed samples (200 µl mg-1) in 24-well plates and then incubated at 37oC for 3 and 7 days. 24-well plates were also kept under the same condition as a control group. Ten thousand cells/ml of culture medium were inserted into 96-well plates and kept at 37oC in a humidified incubator with 5% CO2. After 24 hours of culturing, 100 µl of culture medium was refreshed with 3 and 7-day extracts. The medium was removed after another 24 hours and 100 ȝL MTT solution (0.5 mg/mL) in PBS was added to each well and then incubated at 37ºC for 4 hours. After incubation, 50 µl DMSO (D2650, Sigma, Saint Louis, USA) was added per well to dissolve the purple formazan and mixed thoroughly by pipetting; the solution was then incubated at 37°C for 10 minutes. Afterwards, the

absorbance of each well was measured using an ELISA Reader (Stat Fax-2100; GMI, Inc., Miami, FL, USA) with wavelength of 545 nm and then normalized by control group. Moreover, the morphology of 7-day cultured MG63 cells (5¯104 cells/cm2) was accomplished using SEM. Samples were mounted on an aluminum stub and sputter coated with gold for imaging. The cells were fixed with 4% (v/v) glutaraldehyde solution in PBS at 4ºC for 30 minutes, dehydrated in a series of ethanol/distilled water solution (10% ethanol increments; each step 10 min), and finally dried at room temperature. 2.6. Statistical analysis All data was expressed as mean ± standard deviation (SD). A one-way analysis of variance was used to compare any significant difference. A p< 0.05 was considered statistically meaningful. 3. Results and discussion 3.1. Structure and morphology of SF nanofiber Fig. 1 shows the SEM image and fiber diameter distributions of SF nanofibers fabricated by wet electrospinning system in methanol bath. The SEM image consisted of beadless fibers with average diameter of 277 ± 77 µm, presenting a diameter size between 152 µm and 484 µm. The changes in the IR spectra, the results of deconvolution analysis in amide I band, and the relative content of secondary structure in SF before and after electrospinning in methanol coagulation bath are shown in Fig. 2 and Table 2, respectively. Regenerated SF consists of three kinds of protein structure; ȕsheet, Į-helix, and random coil. The type of secondary structure in proteins can be determined based on the absorption associated with amide I (C=O stretching), amide II (N-H bending), and amide III (N-H bending and C-N stretching) in IR spectra [16, 17]. As shown in Fig. 2a, prior to electrospinning the peaks attributed to the amide I, amide II and amide III are located at 1650, 1533 and 1235 cm-1, respectively [20]. The presence of theses peaks indicates that SF is mainly in random coil and Į-helix conformations before electrospinning in methanol; the percentage of calculated ȕ-sheet is 29.28% (Fig. 2b and Table 2). Following the electrospinning in methanol bath (Fig. 2a), the bands related to amide I, amide II, and amide III in the IR spectrum were shifted to 1629, 1521, and 1261 cm-1, respectively. Also the quantitative proportion of ȕ-sheet was increased from 29.28% to 53.84% (Fig. 2c and Table 2). This suggests that the

random coil to ȕ-sheet conformational transition has happened during electrospinning in methanol bath, which was in agreement with the results reported by the others [16, 21]. 3.2. Characterization of SF nanofiber-starch composite scaffolds Fig. 3 depicts the IR spectra of each fabricated scaffold after cross-linking with glutaraldehyde. As shown in Fig. 3, the IR spectra consists of several discernible peaks at 3350 cm-1 (hydrogen bonded hydroxyl group), 2952 and 2896 cm-1 (asymmetric and symmetric stretching of C-H), 1645 cm-1 (H-O-H bending), 1532 cm-1(amide II; N-H bend in plane and C-N stretch), 1450 and 1380 cm-1 (angular deformation of CH), 1156, 1083, and 1023 cm-1 (C–O stretching), and 916, 889 and 765 cm-1 (entire anhydroglucose ring stretching vibrations) [22-24]. Importantly, the peaks related to the free aldehyde groups of glutaraldehyde at 1720-1740 cm-1 [24] were not observed in the IR spectra, suggesting that the glutaraldehyde covalently cross-linked the samples. Fig. 4 shows the SEM images of starch hydrogel before and after incorporation of various amounts of elctrospun SF nanofibers. Moreover, the mean pore size as well as the porosity percentage of each sample is summarized in Fig. 5. Pure starch hydrogel consisted of elongated pores with the mean pore size of 206 ± 23 µm. As the SEM images of composite samples depict (Fig. 4b-d), the SF nanofibers were interspersed within the starch matrix and got stuck inside the pores. Therefore, the mean sizes of the pores (Fig. 5a) reduced to 121.365 ± 19 in ST-5SF, 113.81 ± 10.1 in ST-10SF, and 93.856 ± 9 in ST-15SF with increasing SF nanofibers content. As shown in Fig. 5b, the calculated porosity of ST sample was 73.19 ± 1.64 %, whereas the amount of calculated porosity of ST-5SF, ST-10SF, and ST-15SF were 67.14 ± 2.51 %, 62.36 ± 0.60 %, and 60.27 ± 0.46 %, respectively. As can be seen from Fig. 5, higher quantity of SF nanofibers decreased both mean pore size and porosity percentage of the composite scaffolds significantly. Fig. 6 illustrates the changes in swelling ratio of the starch matrix by adding various amounts of SF nanofibers. Of all, pure starch hydrogel (ST) showed the highest water absorption (69.1 ± 2.7 %). Within composites, water uptake decreased to 60.7 ± 1.3 % in ST-5SF, 49.5 ± 0.9 % in ST-10SF, and 42.05 ± 0.6 % in ST-15SF with increasing SF nanofibers content. Lowering the pore size and the porosity percentage in the composite samples as SF concentration increases along with hydrophobic nature of SF nanofibers may cause this reduction. 3.3. Characterization of CaP coated scaffolds

The cross-sectional views of each starch-based scaffold after alternate dipping process are demonstrated in Fig. 7. It can be seen that, the plate-like precipitates as well as the micron-sized spherical aggregates have covered the surface of samples. According to the EDS analysis (Fig. 7), the Ca/P ratio of the aggregates was between 1.6-1.8, which is related to the apatitic CaP. Moreover, the Ca/P ratio for plate-like crystals is close to 1, suggesting that brushite crystals were also deposited throughout the scaffolds. The morphology of DCPD crystals and apatitic spherulites deposited on the fabricated samples resemble those supplied by B. Mavis et.al. [25], when a biomimetic CaP layer was formed on polycaprolactone nanofibers. Fig. 8a and 8b illustrate the characteristic XRD patterns and the IR spectra of fabricated scaffolds coated with CaP, respectively. As it can be seen in Fig. 8a, the appearance of typical diffraction peaks at 11.6o, 20.8o, and 29.18o illustrates the formation of brushite crystals (DCPD; JCPDS card No. 11-0293). Moreover, apatitic CaP (JCPDS card No. 84-1998) is indicated by the appearance of weak peaks at 26o and 32o. The XRD results support the SEM and EDS data. The changes in the functional groups of fabricated scaffolds after CaP deposition are shown in Fig. 8b. It is evident that the O-H stretching vibrations of water molecules appeared at 3539, 3485, 3278, and 3178 cm-1. The peaks centered at 1640 and 1226 cm-1 are assigned to H2O and in-plane P-O-H bending vibration, respectively. The P-O stretching vibrations were observed at 1134, 1057, and 987 cm-1. In addition, the P-O(H) stretching vibration was also detected at 876 cm-1. Another band at 655 cm-1 is attributed to P-O bending of phosphate groups [26, 27]. The formation of CaP layer throughout the fabricated scaffolds is due to the presence of hydroxyl groups in starch hydrogel as well as carboxyl groups in SF nanofiber. As mentioned in previous studies [28, 29], these functional groups are negatively charged in physiological condition (pH= 7.4, T= 37oC) and act as inductive sites for CaP nucleation. Therefore, heterogeneous nucleation began through bonding of calcium ions with these negatively charged functional groups during initial soaking in calcium rich solution (pH=7.4). Subsequent soaking in phosphate rich solution resulted in CaP nuclei formation, followed by growth throughout the scaffolds during several soaking processes. As indicated by XRD data (Fig. 8a), the fabricated scaffolds are able to induce brushite along with apatitic CaP. As mentioned by previous reports [30-32], phases such as amorphous calcium phosphate (ACP), brushite (DCPD), octacalcium phosphate (OCP), and calcium deficient hydroxyapatite (CDHA) are mainly precipitate in aqueous solution at low temperature. Various factors such as charged surface, type of functional groups, pH, and supersaturation

degree of aqueous medium (IP/K0) affect the nature and crystallinity of the deposited CaP on the surface of materials [33-35]. It has been reported that brushite precipitates in aqueous solution containing Ca2+ and HPO42- in pH values between 2 to 6.5 [27, 36]. In this study, the pH value of the CaCl2 solution decreased from 7.4 to 5.02, presumably due to the ionization of hydroxyl and carboxyl groups. Therefore, the reduction in pH value after first dipping cycle in CaCl2 should be the reason for brushite precipitation; this is a primary phase that can convert to stable apatitic CaP [25]. Table 3 shows the amount of CaP deposited throughout the fabricated starch-based scaffolds. While the highest amount of deposited CaP is evident in the starch matrix, increasing the amount of SF nanofibers decreases CaP formation. Aside from negatively charged functional groups, the swelling ratio of hydrogels plays a critical role in CaP formation [14, 37]. Therefore, the decreased swelling ratio of fabricated nanocomposites by adding more SF nanofibers (shown in Fig. 6) might reduce CaP forming ability of the scaffolds. 3.4. Cell culturing Fig. 9 displays the results of the cell viability assay of all samples in comparison with the control after exposed to the 3 and 7 days extracts. It can be seen that except for pure starch hydrogel (ST-CP), in samples consisting SF nanofibers (ST-SF-CP), cell viability is significantly higher than that of control samples (p90% at different days. Such results indicate the biocompatibility of the fabricated starchbase scaffolds. The SEM images of MG63 cells cultured on fabricated scaffolds after 7 days are shown in Fig. 10. The cells were adhered on the surface of all scaffolds, as indicated in Fig. 10. However, there are noticeable differences in the cells morphology of pure starch hydrogel in comparison with composite samples containing SF nanofibers. Cells displayed a rounded morphology with a small filopodia on the pure starch scaffold’s surface. However, the cells are well spread and anchored on composite scaffolds, in which the surface is covered with a monolayer of cells. Both SEM images and MTT results can prove the potential role of SF nanofibers in cell attachment and growth due to their high surface area. This is in agreement with previous reports carried out by H. Jin et.al [38], that the adhesion, spreading, and growth of human bone marrow stromal cells are enhanced on the electrospun SF nanofiber mat with the mean fiber diameter of 700±50 nm.

4. Conclusion This study offers a new composite scaffold based on silk fibroin nanofiber-porous starch-calcium phosphate for potential bone tissue regeneration. Wet electrospun SF nanofibers (β-sheet= 53.84%,

d = 277 ± 77 µm) were embedded in a porous starch matrix. As the amount of SF nanofiber increased, the average pore size, porosity, and swelling ratio decreased in the composites. The mineralization of CaP throughout the scaffolds occurred after soaking in calcium and phosphate rich solutions, alternately (3 times repeating). The deposited layer consists of both brushite and apatitic CaP, as indicated by SEM, XRD, and IR analysis. The highest amount of deposited CaP is related to the starch matrix and increasing the amount of SF nanofibers decreases CaP formation. The viability of osteoblast-like cells (MG63) exposed to the composites’ extracts was significantly higher than that of the pure starch and control group. In addition, the cells are well spread and anchored on the scaffolds with SF nanofibers, in which the surface is covered with a monolayer of cells. Acknowledgments The corresponding author would like to acknowledge the financial support from the Iran National Science Foundation (Project No: 93027760), and Faculty of New Sciences and Technologies (1393/06/29), University of Tehran, Iran.

Reference: [1] A.R. Shrivats, M.C. McDermott, J.O. Hollinger, Bone tissue engineering: state of the union. Drug discovery today 19 (2014) 781-786. [2] Y. Liu, J. Lim, S.-H. Teoh, Review: development of clinically relevant scaffolds for vascularised bone tissue engineering. Biotechnology advances 31 (2013) 688-705. [3] N.A. Peppas, J.Z. Hilt, A. Khademhosseini, R. Langer, Hydrogels in biology and medicine: from molecular principles to bionanotechnology. Advanced Materials 18 (2006) 1345-1360. [4] H. Ismail, M. Irani, Z. Ahmad, Starch-based hydrogels: present status and applications. International Journal of Polymeric Materials and Polymeric Biomaterials 62 (2013) 411-420.

[5] I.M. Martins, S.P. Magina, L. Oliveira, C.S. Freire, A.J. Silvestre, C.P. Neto, A. Gandini, New biocomposites based on thermoplastic starch and bacterial cellulose. Composites Science and Technology 69 (2009) 2163-2168. [6] B. Nasri-Nasrabadi, M. Mehrasa, M. Rafienia, S. Bonakdar, T. Behzad, S. Gavanji, Porous starch/cellulose nanofibers composite prepared by salt leaching technique for tissue engineering. Carbohydrate Polymers 108 (2014) 232-238. [7] D. Liu, T. Zhong, P.R. Chang, K. Li, Q. Wu, Starch composites reinforced by bamboo cellulosic crystals. Bioresource Technology 101 (2010) 2529-2536. [8] S. Mobini, M. Solati-Hashjin, H. Peirovi, N.A.A. Osman, M. Gholipourmalekabadi, M. Barati, A. Samadikuchaksaraei, Bioactivity and Biocompatibility Studies on Silk-Based Scaffold for Bone Tissue Engineering. Journal of Medical and Biological Engineering 33 (2013) 207-213. [9] W. Liu, S. Thomopoulos, Y. Xia, Electrospun nanofibers for regenerative medicine. Advanced healthcare materials 1 (2012) 10-25. [10] R. Khajavi, M. Abbasipour, Electrospinning as a versatile method for fabricating coreshell, hollow and porous nanofibers. Scientia Iranica 19 (2012) 2029-2034. [11] C. Vepari, D.L. Kaplan, Silk as a biomaterial. Progress in polymer science 32 (2007) 991-1007. [12] C. Li, C. Vepari, H.-J. Jin, H.J. Kim, D.L. Kaplan, Electrospun silk-BMP-2 scaffolds for bone tissue engineering. Biomaterials 27 (2006) 3115-3124. [13] K. Gkioni, S.C. Leeuwenburgh, T.E. Douglas, A.G. Mikos, J.A. Jansen, Mineralization of hydrogels for bone regeneration. Tissue Engineering Part B: Reviews 16 (2010) 577-585. [14] T. Taguchi, A. Kishida, M. Akashi, Apatite formation on/in hydrogel matrices using an alternate soaking process: II. Effect of swelling ratios of poly (vinyl alcohol) hydrogel matrices on apatite formation. Journal of Biomaterials Science, Polymer Edition 10 (1999) 331-339. [15] D.N. Rockwood, R.C. Preda, T. Yücel, X. Wang, M.L. Lovett, D.L. Kaplan, Materials fabrication from Bombyx mori silk fibroin. Nature protocols 6 (2011) 1612-1631. [16] R.I. Litvinov, D.A. Faizullin, Y.F. Zuev, J.W. Weisel, The Į-helix to ȕ-sheet transition in stretched and compressed hydrated fibrin clots. Biophysical journal 103 (2012) 1020-1027.

[17] P. Garidel, H. Schott, Fourier-Transform Midinfrared Spectroscopy for Analysis and Screening of Liquid Protein Formulations. BioProcess International (2006). [18] N. Annabi, A. Fathi, S.M. Mithieux, A.S. Weiss, F. Dehghani, Fabrication of porous PCL/elastin composite scaffolds for tissue engineering applications. The Journal of Supercritical Fluids 59 (2011) 157167. [19] F. Mirahmadi, M. Tafazzoli-Shadpour, M.A. Shokrgozar, S. Bonakdar, Enhanced mechanical properties of thermosensitive chitosan hydrogel by silk fibers for cartilage tissue engineering. Materials Science and Engineering: C 33 (2013) 4786-4794. [20] J. Shao, J. Zheng, J. Liu, C. Carr, Fourier transform Raman and Fourier transform infrared spectroscopy studies of silk fibroin. Journal of applied polymer science 96 (2005) 1999-2004. [21] C. Yan, B. Yang, Z. Yu, Methanol-induced conformation transition of gland fibroin monitored by FTIR spectroscopy and terahertz spectroscopy. Analyst 139 (2014) 1967-1972. [22] I. Capron, P. Robert, P. Colonna, M. Brogly, V. Planchot, Starch in rubbery and glassy states by FTIR spectroscopy. Carbohydrate Polymers 68 (2007) 249-259. [23] O. Sevenou, S. Hill, I. Farhat, J. Mitchell, Organisation of the external region of the starch granule as determined by infrared spectroscopy. International Journal of Biological Macromolecules 31 (2002) 79-85. [24] G. Socrates, Infrared and Raman characteristic group frequencies: tables and charts, John Wiley &Sons, 2004. [25] B. Mavis, T.T. Demirtaú, M. Gümüúderelio÷lu, G. Gündüz, Ü. Çolak, Synthesis, characterization and osteoblastic activity of polycaprolactone nanofibers coated with biomimetic calcium phosphate. Acta biomaterialia 5 (2009) 3098-3111. [26] I. Hofmann, L. Müller, P. Greil, F.A. Müller, Calcium phosphate nucleation on cellulose fabrics. Surface and Coatings Technology 201 (2006) 2392-2398. [27] S. Mandel, A.C. Tas, Brushite (CaHPO 4· 2H2O) to octacalcium phosphate (Ca8 (HPO4) 2 (PO4) 4· 5H2 O) transformation in DMEM solutions at 36.5°C. Materials Science and Engineering: C 30 (2010) 245-254. [28] M.P. Gashti, M. Stir, M. Bourquin, J.R. Hulliger, Mineralization of Calcium Phosphate Crystals in Starch Template Inducing a Brushite Kidney Stone Biomimetic Composite. Crystal Growth & Design 13 (2013) 2166-2173.

[29] Ö. Malay, O. Bayraktar, A. Batıgün, Complex coacervation of silk fibroin and hyaluronic acid. International Journal of Biological Macromolecules 40 (2007) 387-393. [30] X. Lu, Y. Leng, Theoretical analysis of calcium phosphate precipitation in simulated body fluid. Biomaterials 26 (2005) 1097-1108. [31] A.C. Tas, S.B. Bhaduri, Chemical Processing of CaHPO4·2H2O. Journal of the American Ceramic Society 87 (2004) 2195-2200. [32] J. Nourmohammadi, S. Sadrnezhaad, A. Behnamghader, In vitro bioactivity of novel cured ionomer cement based on iron oxide. Ceramics International 36 (2010) 1645-1651. [33] F. Barrere, P. Layrolle, C. Van Blitterswijk, K. De Groot, Biomimetic coatings on titanium: a crystal growth study of octacalcium phosphate. Journal of Materials Science: Materials in Medicine 12 (2001) 529-534. [34] J. Nourmohammadi, S. Sadrnezhaad, A.B. Ghader, Bone-like apatite layer formation on the new resinmodified glass-ionomer cement. Journal of Materials Science: Materials in Medicine 19 (2008) 3507-3514. [35] M.S.A. Johnsson, G.H. Nancollas, The role of brushite and octacalcium phosphate in apatite formation. Critical Reviews in Oral Biology & Medicine 3 (1992) 61-82. [36] M.A. Miller, M.R. Kendall, M.K. Jain, P.R. Larson, A.S. Madden, A.C. Tas, Testing of brushite (CaHPO4· 2H2O) in synthetic biomineralization solutions and in Situ crystallization of crushite microǦgranules. Journal of the American Ceramic Society 95 (2012) 2178-2188. [37] J.Y. Lai, Y.T. Li, C.H. Cho, T.C. Yu, Nanoscale modification of porous gelatin scaffolds with chondroitin sulfate for corneal stromal tissue engineering. International Journal of Nanomedicine 7 (2012) 1101-1114. [38] H.J. Jin, J. Chen, V. Karageorgiou, G.H. Altman, D.L. Kaplan, Human bone marrow stromal cell responses on electrospun silk fibroin mats. Biomaterials 25 (2004) 1039-1047.

Figure Captions Fig.1. SEM image and fiber diameter distribution of wet electrospun SF nanofibers. Fig.2. (a) ATR spectra recorded at room temperature of SF before and after wet electrospinning process (b) the results of de-convolution analysis on the amide I band before electrospinning , and (c) the results of deconvolution analysis on the amide I band after wet electrospinning in methanol bath. Fig.3. ATR spectra recorded at room temperature of starch-based scaffolds after crosslinking with glutaraldehyde. Fig.4. SEM images of starch-based scaffolds before and after incorporation of SF nanofibers (a) ST (b) ST5SF (c) ST-10SF, and (d) ST-15SF. Fig.5. (a) The average pore size, and (b) porosity of the fabricated scaffolds. Fig.6. Swelling ratio of the various starch-based scaffolds. Fig.7. SEM images of fabricated scaffolds coated with CaP (a) ST-CP (b) ST-5SF-CP (c) ST-10SF-CP, and (d) ST-15SF-CP. Fig.8. The (a) XRD patterns (b) ATR spectra of each fabricated scaffold after CaP deposition. Fig.9. The viability of MG63 cells after exposed to 3 and 7 days extracts. Fig.10. The morphology of osteoblast-like cells (MG63) after 7 days of culturing (a) ST-CP (b) ST-5SF-CP (c) ST-10 SF-CP, and (d) ST-15 SF-CP. .

Scaffold samples Before CaP deposition After CaP deposition ST ST-CP ST-5SF ST-5SF-CP ST-10SF ST-10SF-CP ST-15SF ST-15SF-CP

SF nanofibers content (%) 0 5 10 15

Table1. The code and composition of each fabricated composite scaffolds

Table 2. Relative content of secondary structure in SF before and after electrospinning in methanol coagulation bath

Secondary protein structures ȕ-sheet Į-helix Random coil

Samples (%) Prior to electrospinning After electrospinning (methanol free) (methanol bath) 29.28 53.84 29.91 17.68 23.98 19.33

Samples

ST

ST-5SF

ST-10SF

ST-15SF

CaP deposited (mg/gel)

4.88±0.21

3.57±0.18

3.35±0.07

3.08±0.05

ȕ-turns

16.83

9.15

Table 3. The amount of CaP deposited throughout the scaffolds during alternate soaking process

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure6

Figure7

Figure 8

Figure9

Figure 10