1. Introduction

It is well known that natural bone is a typical inorganic/organic composite comprising approximately 80 % HA and 20% collagen matrix. It is also known that the bone shows a porous structure, containing multiple levels of organization where HA crystals in nanolevel scale are embedded in collagen matrix.

The biomimetical strategy, based on the idea of mimicking natural bones composition offers great promise for the design and fabrication of biomaterials for healing bone defects. Therefore, growing interest has been focused on the integration of HA ceramic with bio-organic polymers, such as collagen[1], gelatine[2], alginate[3], polylactic acid [4], polyamide[5] and silk fibroin (SF)[6-14]. Among them SF is of practical interest due to its excellent intrinsic properties utilizable in the biotechnological and biomedical fields. It is known that SF is composed from 17 aminoacids. It has been proved to be good biocompatible material and it has been successfully used for various medical applications. X. D. Kong et al.,[6] demonstrated that silk fibres could induce HA nucleation on the surfaces proteins in simulated body fluid (SBF) solution at pH=8 at room temperature. FTIR results proved that the HA crystals are CO₃HA. On the other hand they also detected that there are strong chemical interactions between HA and SF protein. The presence of these bonds can be derived from the blue shift of amide II bond. For this reason, a co-precipitation method was often used to synthesize silk powder/HAnanocomposites[6-14]. For instance, L. Wang et al.[9] deposited HA on SF microspheres thorough the co-precipitation method. Others developed a HA/SF composite, where SF was modified with organic compound[15]. In other article, R. Kino et al.[16] reported that regenerated SF films containing calcium were coated with HA after soaking in 1.5 SBF solution for 6 hours. Y. Li et al.[17] proved that SF can induce HA deposition at 37℃ in 1.5 SBF. The effect of pH and initial Ca²⁺-H₂PO₄^- concentration on the crystal growth of HA in the presence of SF is studied by Y. Ren et al.[12]. The authors evaluate the effect of pH and initial Ca-P concentration on the products of SF mineralization. They also demonstrated that that pH=7 promotes the transition of dicalcium phosphate dihydrate (DCPD) to HA. X. Kong et al. observed that SF can accelerate the phase transmittance of amorphous calcium phosphate to the crystal hybrid between DCPD and HA[18]. L. Li et al.[19] prepared SF/Calcium Phosphate (CP) composite by adding the different amount of Na₂SiO₃ to assess the effect of silicon on the HA formation in the composite. Based on the obtained results they proved that the addition of silicon could accelerate the HA crystal formation. SF/Wollastonite composite scaffolds were prepared by H. Zhu et al.[20] via freeze-drying method. The obtained results showed that the composite scaffold was in vitro bioactive, because it induced the formation of CO₃HA on the surface after soaking in SBF for 5 days. Four type of hybrid films were fabricated from N. Vachiraroj et al.[21]. These materials include gelatin conjugated SF, HA/SF, HA/Gelatin conjugated SF and HA/Chitosan. The synthesized films were investigated in terms of in vitro adhesion, proliferation and oestrogenic differentiation with preosteoblast cell lines (MC3T3-E1) and rat bone marrow derived stem cells (MC3). The authors proved that MC3T3-E1 showed higher proliferation rate.

In our previous works, we prepared some in vitro bioactive ceramics in the CaO-SiO₂-P₂O₅[22-24], and CaO-SiO₂-P₂O₅-MgO systems[25] through different ways. Furthermore, we have synthesized in vitro bioactive composite materials in the presence of collagen[26-28] and gelatin[29, 30] with the prepared ceramics powders.

The purpose of the present article is preparing new bioactive composite by addition of SF with calcium phosphate silicate (CPS) ceramic and evaluating its bioactivity in 1.5 SBF solutions.

2. Experimental Part

CPS ceramic powder in the system 15CaO-6SiO₂-0.5P₂O₅, as an inorganic part of the composites, has been synthesized via polystep sol-gel method. The synthesized procedure and structure evolution is documented in Ref.[23].

SF without sericin was dissolved in 9M LiBr at 60^oC for 4 hours and then dialyzed against with distilled water at room temperature for 3 days using cellulose membrane (MWCO 12000 Da) for 72 hours to remove the salts[21]. The final concentration of SF solution was 3 wt. %, which was determined in accordance with[20]. SF/CPS composites were prepared by mixing method without binding agents. Briefly, a certain amount of CPS powder was added into the SF solution under stirring for 4 hours in order to disperse the CPS particles uniformly. The obtained mixture was poured into plastic cup. The mixture was dried in the microwave oven for 10 minutes. In the present work, three kind of samples were made with the weight ratio of SF:CPS of 75:25, 50:50 and 25:75 (in wt. %), described as 75S25C, 50S50C and 25S75C.

In vitro bioactivity of the prepared composites with different weight ratio of SF and SPS were evaluated by examining the apatite formation on their surfaces in 1.5 SBF solutions. 1.5 SBF solution was prepared from reagents as follows: NaCl=11.9925 g, NaHCO₃=0.5295 g, KCl=0.3360 g, K₂HPO₄.3H₂O=0.3420g, MgCl₂.6H₂O=0.4575g, CaCl₂.2H₂O=0.5520 g, Na₂SO₄=0.1065 g and buffering at pH=7.4 at 36.5℃ with TRIS=9.0075 and 1M HCl in distilled water. The synthesized composite materials were immersed in 1.5 SBF at human body temperature (36.6℃) in polyethylene bottles, in static conditions for 12 days. The samples were removed from the fluid after soaking, and gently rinsed with distilled water, then dried at 37℃ for 6 hours. To prevent a homogeneous apatite formation in the solution, it is known that 1.5 SBF needed to buffered at 7.25, as it is highly supersaturated and consequently unstable[31].

The structure and the in vitro bioactivity of the synthesized SC composites were monitored by FTIR, ICP-AES and SEM.

FTIR transmission spectra were recordered by using a Brucker Tensor 27 Spectrometer with scanner velocity 10 kHz. Diluted pellets of KBr were prepared by mixing of ~1 mg of the samples with 300 mg KBr. Transmission spectra were recordered using MCT detector with 64 scans and 1 cm^-1 resolution.

The samples were immersed in sealed test tubes containing 10 ml of 1.5 SBF solution for 12 days. The experiment was maintained at 37℃ in sterile polyethylene containers in static conditions. After immersion in 1.5 SBF for 12 days the samples were retrieved, gently rinsed with water, and dried at 37℃ for 6 hours. The ionic concentrations of Ca, P and Si of these solutions were determined by Inductively Coupled Plasma Atomic Emission Spectroscopy (ICP-AES), Iris 1000, Thermo Elemental, USA.

Morphological studies of the obtained composite materials before and after in vitro test for 12 days have been done by using SEM. The samples were gold-sputter coated and viewed in the secondary electron mode with a field emission gun. Scanning electron Microscope (Philips 515) operated at an accelerating voltage of 2.5 kV.

3. Results and Discussion

FTIR spectra of SF and SC composites before in vitro test are shown in Figure 1 in the spectrum range 2000-400 cm^-1.

SF proteins exists in three conformations, namely random coil, silk I (α-form) and silk II (β-sheet). The absorption band posited at 1644 cm^-1 for SF can be ascribed to the amide I mode, associated with random coil conformation occurring in the range of 1640-1650 cm^-1, 1534 cm^-1 for amide II in the random coil and the β-sheet conformations and 1240 cm^-1 for amide III[19, 20, 31]. From the depicted in Fig. 1, a FTIR spectrum, we can assume that both random coil and β-sheets conformation are presented in the structure after microwave treatment of SF. On the other hand, three spectra of the composites with different wt. % of the SF and CPS (Fig. 1, b, c and d) show the peak shift from 1644 to 1638 cm^-1 (for 75S25C sample), 1640 cm^-1 (for 50S50C sample) and 1641 cm^-1 (for 25S75C sample). In the case of 75S25C (fig. 1, b) the observed “red shift” could be ascribed to the conformational transition from random coil to β-sheet, i. e. antiparallel β-sheet conformation[32]. L. Li et al. reported that the same result can be observed, when the composite material synthesized through biomimetic way[19]. There is another possible interpretation, from our point of view : the depicted “red shift” can be assigned to the C=O: group which may interact with Ca²⁺ from partially dissolved CPS ceramic powder in SF solution under the preparation conditions of the composites. The same results can be observed for amide II and amide III region. New bands at 714-430 cm^-1 appear in addition to the characteristic amide I, II and III absorption bands, after preparation of the samples, were assigned to the presence of CPS ceramic[23].

Figure 1. FTIR spectra of SF (a), 75S25C (b), 50S50C (c) and 25S75C (d) before in vitro test

From the presented FTIR data we can see that when SF solutions was mixed with CPS ceramic powder and undergo microwave treatment, further conformational changes may occur, with a further transition of random coils to a more stable, antiparallel β-sheet conformation throughout intermediate forms, such as β-turns and β-blends.

Figure 2. Curve-fitting results of the FTIR spectra in the amide I region for the SF (a), 75S25C (b), 50S50C (c), and 25S75C (d)

FTIR spectra of SF and SF/CPS composites were examined more carefully in the amide I region. The amide I spectral region was resolved into 3 of 4 main components for SF and for composites, assuming a Gaussian behavior for each individual process. The curve-fitting results are shown in Figure 2 and Table 1.

Table 1. Curve-fitting results of ‘‘amide I’’ regions In FTIR spectra of the SF and SF/CPS composites

We can see from the presented in Figure 2 and Table 1 curve-fitting data, that in the sample with 25 wt. % CPS ceramic powder, the β-structure fraction significantly increased in parallel with pure SF. Furthermore, the fractions of β-structure significantly decreased, but the random coil fraction slightly increased, compared with pure SF, with increasing of CPS ceramic powder to 50 and 75 wt. %. Finally, the fraction of β-turns and blends does not significantly change comparing to SF.

We can conclude that significant changes in the percentage of β-structure, random coil and β-turns and blends are observed among the pure SF and SF/CPS composites, i. e. the addition of different weight of CPS ceramic has a significant effect on the secondary structure of SF. These observations are needed for future investigations.

SEM micrographs of the prepared composites with different weight ratio of the two components are shown in Figure 3.

Figure 3. SEM of 25S75C (a), 50S50C (b) and 75S25C (c) before in vitro test in 1.5 SBF

SEM micrographs of the synthesized composites before soaking in 1.5 SBF for various periods of time show that the original materials present textural differences. While 25S75C (Fig. 3, a) and 75S25C (Fig. 3, c) show a heterogeneous distribution of a ceramic powder over fibroin, 50S50C (Fig. 3, b) presents a more homogeneous surface when the organic and inorganic parts of the composite are equal. The fiber structure of fibroin was visibly recognized in the 25S75C and 50S50C samples (Fig. 3 a and b). When the quantity of fibroin increased to the 75 wt. % (Fig. 6, c) the SEM image shows the different textural future. It can be seen, the prepared sample consisting large individual fibroin lamella with a ceramic powder on the surface. The observed lamella in the 75S25C sample is a function of water residence time in the process of microwave drying of the prepared composite material. From general point of view, the synthesized fibroin-calcium phosphate silicate composites without any previous treatment, dried in microwave oven (Fig. 3, a-c) exhibit a smoother aggregate surface. This smooth surface is explained by the migration of fine ceramic particles and their coalescence on the fibroin surface, driven by different mechanisms of drying. As it can be read in E. Pecoraro et al.[33], the solvent activation energy (in our case water) can cause aggregate collapse which can give polycondensation ceramic particles and smoother surface of the prepared organic/inorganic composites.

The FTIR spectra of the synthesized SF/CPS composites after in vitro test in 1.5 SBF for 12 days are presented in Figure 4

Figure 4. FTIR spectra of the SF/CPS composites: 75S25C (a), 50S50C (b) and 25S75C (c) after in vitro test in 1.5 SBF for 12 days in static conditions

The very important vibration of the PO₄^3- bands in the depicted FTIR spectra (▲) could be distinguished in the three main regions: (i) the peaks with very small intensity, centered at 1040 cm^-1 (Fig. 4, b and c) are associated with asymmetric stretch vibration[34]. The presence of these peaks related with stretching vibrations of PO₄^3- in apatite crystalline phases[35], (ii) a weak band, posited at ~960 cm^-1 (Fig.4 a, b and c) corresponding to the symmetric stretching[34], and (iii) the well resolved triplet at 600, 560 and 470 cm^-1 could be ascribed to the asymmetric modes. On the other hand, the peak at ~1040 cm^-1 and the peaks at ~796 cm^-1 (Fig. 4 b and c) could be ascribed due to the both of the PO₄^3- and Si-O-Si ( ♦ ) band[36]. Furthermore, the peaks at 600 and 560 cm^-1 are assigned to O-Si-O ( ● ) vibration[37]. In addition the bands in the region 1560-1400 cm^-1 and the band at ~875 cm^-1 could be assigned to the presence of CO₃^2- ( ■ ) vibrations[22, 23, 25, 27-30].

More detailed information about the type of substitution in CO₃HA, formed on the surface of the SF/CPS samples, after 12 days of immersion in 1.5 SBF, was given by curve-fitting in the region 1400-1600 cm^-1 (Figure 5)

From the depicted in Figure 5 results, it can be seen that the bands, posited at 1409, 1417, 1446, 1452, 1454, 1457, 1470, 1521, 1523, 1526, 1538 and 1539 cm^-1 could be ascribed to the presence of B-type CO₃HA, i. e. part of PO₄^3- replaced by CO₃^2-[13, 28, 38-40]. The bands, centered at 1500, 1508, 1514, 1516, 1550 and 1570 cm^-1 could be assigned to A-type substitution in which CO₃^2-→OH^-[38, 41, 42]. Furthermore, there are some bands, posited at 1422, 1436 and 1547 cm^-1. They could be denoted as indicative bands for A/B-type CO₃HA in accordance with[43].

Figure 5. Curve-fitting spectra of the type of CO3HA, formed on the surface of SF/CPS composites: 75S25C (a), 50S50C (b) and 25S75C (c) after in vitro test in 1.5 SBF for 12 days in static conditions

The curve-fitting results including peak area (%) are shown in Table 2.

Table 2. Curve-fitting results of CO32- regions in the SF/CPS composites after in vitro test for 12 days in 1.5 SBF solutions in static condition

We can conclude, from the depicted in Table 2 results, that B-type CO₃HA (bioapatite) was preferentially observed, when CPS ceramics was 75 wt. % into the composite.

Figure 6 shows SEM images of the surfaces for studied composite materials with different weight ratio of SF and CPS, after soaking in 1.5 SBF solution for 12 days in static conditions

It is observed, from depicted SEM results, that the surface of the immersed samples fully covered from the particles with different morphology. The observed spherical particles seem have crackles at the beginning, but actually they have not. We can see, from the presented at Fig. 6 a, b and c SEM data, that the big spherical particles and small spheres with diameter up to ~2 µm are composed.

Figure 6. SEM images for 75S25C (a), 50S50C (b) and 25S75C (c) after soaking in 1.5 SBF solution for 12 days

Figure 7 presents The ICP-AES data for the evaluation of the ionic concentration of Ca, Si and P in the 1.5 SBF solution after soaking of the SF/CPS composites for 12 days.

Figure 7. Changes in Ca, Si and P concentration for the 1.5 SBF solution (sample 1) and SF/CPS composites: 75S25C (sample 2), 50S50C (sample 3) and 25S75C (sample 4) after in vitro test in 1.5 SBF for 12 days

It can be seen from the presented Fig. 7 that the Ca concentration for the samples 2-4 slightly increased, compared with the Ca concentration into the initial 1.5 SBF solution. The concentrations of Ca are in accordance with the weight ratio of the CPS ceramic powder in the prepared SF/CPS samples. We could see from the presented Fig.7, that the Si content in 1.5 SBF solutions increased up to 66 mg/l (sample 4). In accordance to our preliminary data[28], we can conclude that silicon containing CO₃HA (Si-CO₃HA) may be formed on the surface of the immersed samples for 12 days. Finally, the P concentration slightly decreased from 45 mg/l for the 1.5 SBF solution (sample 1) to 2 mg/l (sample 2) and 0 mg/l (samples 3 and 4). The obtained ICP-OES data indicates that the outset of HA precipitation is similar to the three composites, although the stronger one better resolved FTIR ν₂ PO₄^3- vibrations, posited at 560-567 cm^-1. The presence of these bands suggests that the HA layer possesses superior structural order on the composites after soaking. The presented ICP-AES data is in a good agreement with FTIR and SEM analysis of the immersed samples.

4. Conclusions

The purposes of the presented article were to prepare and to evaluate in vitro bioactivity of composites between SF and CPS ceramic powder. The SF, denoted by S, was mixed with CPS ceramic powder, denoted by C, to prepare 75S25C, 50S50C and 25S75C composites. The obtained mixtures were treated in microwave oven to evaporate the water solvent. After drying, the prepared composites were soaked in 1.5 SBF for 12 days in static conditions to evaluate their in vitro bioactivity. The obtained composites were characterized by FTIR and SEM. The ionic concentration of Ca, Si and P were examined by ICP-OES technique.

FTIR data of the composites before in vitro test proved that the “red shift” of amide II band becomes larger in comparison to amide I and II. These findings suggest that the chemical interaction exists between SF and CPS ceramic powder. After soaking of the synthesized samples in 1.5 SBF for 12 days FTIR results proved that B-, A- and A/B-type CO₃HA are deposited on the surface. ICP-AES measurements have shown that the rates of Ca realize are ordered as follows: 25S75C>50S50C>75S25C. For the three samples, the Si concentration increased and P concentration decreased after immersion. Based on these results we suppose that Si-CO₃HA may be formed on the composite surface. SEM of the immersed samples proved that there is high concentration of HA phase on the surface.