- Email: [email protected]
45S5 Bioglass® scaffolds for MC3T3-E1 cell stimulation and drug release
Materials Science and Engineering C 56 (2015) 473–480
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Multifunctional chitosan/polyvinyl pyrrolidone/45S5 Bioglass® scaffolds for MC3T3-E1 cell stimulation and drug release Qingqing Yao a, Wei Li b, Shanshan Yu a, Liwei Ma a, Dayong Jin c,d, Aldo R. Boccaccini b,⁎, Yong Liu a,d,⁎⁎ a
Institute of Advanced Materials for Nano-Bio Applications, School of Ophthalmology & Optometry, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou, Zhejiang 325027, China Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, Cauerstrasse 6, Erlangen 91058, Germany Institute for Biomedical Materials and Devices, Faculty of Science, University of Technology Sydney, NSW 2007, Australia d Advanced Cytometry Labs, ARC Center of Excellence for Nanoscale BioPhotonics, Macquarie University, Sydney, NSW 2109, Australia b c
a r t i c l e
i n f o
Article history: Received 23 January 2015 Received in revised form 12 May 2015 Accepted 23 June 2015 Available online 9 July 2015 Keywords: Chitosan Polyvinyl pyrrolidone 45S5 Bioglass® Scaffolds Drug release MC3T3-E1 cells
a b s t r a c t Novel chitosan–polyvinyl pyrrolidone/45S5 Bioglass® (CS-PVP/BG) scaffolds were prepared via foam replication and chemical cross-linking techniques. The pristine BG, CS-PVP coated BG and genipin cross-linked CS-PVP/BG (G-CS-PVP/BG) scaffolds were synthesized and characterized in terms of chemical composition, physical structure and morphology respectively. Resistance to enzymatic degradation of the scaffold is improved significantly with the use of genipin cross-linked CS-PVP. The bio-effects of scaffolds on MC3T3-E1 osteoblast-like cells were evaluated by studying cell viability, adhesion and proliferation. The CCK-8 assay shows that cell viability on the resulting G-CS-PVP/BG scaffold is improved obviously after cross-linking of genipin. Cell skeleton images exhibit that well-stretched F-actin bundles are obtained on the G-CS-PVP/BG scaffold. SEM results present significant improvement on the cell adhesion and proliferation for cells cultured on the G-CS-PVP/BG scaffold. The drug release performance on the as-synthesized scaffold was studied in a phosphate buffered saline (PBS) solution. Vancomycin is found to be released in burst fashion within 24 h from the pristine BG scaffold, however, the release period from the G-CS-PVP/BG scaffold is enhanced to 7 days, indicating improved drug release properties of the G-CS-PVP/BG scaffold. Our results suggest that the G-CS-PVP/BG scaffolds possess promising physicochemical properties, sustained drug release capability and good biocompatibility for MC3T3-E1 cells' proliferation and adhesion, suggesting their potential applications in areas such as MC3T3-E1 cell stimulation and bone tissue engineering. © 2015 Published by Elsevier B.V.
1. Introduction Bone defects are usually arisen from trauma, infection, tumor, osteoporotic fracture, osteonecrosis or congenital deformity [1,2]. Traditional treatments e. g. autograft and allograft transplants have been seen as the effective methods for bone repair. These techniques, however, are suffering from many drawbacks such as donor site morbidity and graft rejection [3–5]. Recently, tissue engineering techniques with the use of scaffold-based approaches have been attracting increasing attention [3,4]. 3D scaffolds can provide numerous advantages for diverse applications such as suitable environment for cell attachment, proliferation, and differentiation [5], and even as the carrier for drug delivery systems [6].
⁎ Corresponding author. ⁎⁎ Correspondence to: Y. Liu, Institute of Advanced Materials for Nano-Bio Applications, School of Ophthalmology & Optometry, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou, Zhejiang 325027, China. E-mail addresses: [email protected] (A.R. Boccaccini), [email protected] (Y. Liu).
http://dx.doi.org/10.1016/j.msec.2015.06.046 0928-4931/© 2015 Published by Elsevier B.V.
45S5 Bioglass® (BG) has been intensively studied as the alternative for bone tissue engineering considering its ability to integrate with bone [7,8]. BG can interact with the surrounding tissue milieu and serve as a substrate for osteogenic stem cells' attachment and differentiation [9]. However, the porous BG scaffolds always suffer from relatively low mechanical strength and high brittleness [10]. Incorporation of BG with particular polymers (e.g. chitosan based system) has been seen as an efficient way to improve the properties of scaffolds [11–14]. Chitosan (CS), a linear polysaccharide composed of glucosamine and N-acetyl glucosamine units linked by β(1–4)-glycosidic bonds and deacetylated from chitin [15], has been well known for its unique properties including biodegradability, biocompatibility, non-toxicity, hydrophilicity and nonimmunogenicity. Various techniques, including the cross-linking, blending, and chemical modification, have been developed to improve the mechanical properties of CS-based materials [15,16]. For example, polyvinylpyrrolidone (PVP), a hydrophilic polymer excipient, can form good miscibility with CS in diluted acidic aqueous solutions [17,18]. PVP can also enhance the thermodynamic stability in a binary solution. To prevent CS from enzymatic degradation and retaining the stability of CS under wet conditions, the cross-linking methods are usually applied to anchor the amine groups on the polymeric chains of CS [19]. Genipin
474
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
is obtained from geniposide, which may be isolated from the fruits of Gardenia jasminoides Ellis [20]. It has been found that genipin is an excellent natural cross-linker for CS, since genipin is less cytotoxic than glutaraldehyde and many other commonly used synthetic crosslinking reagents [21–23]. In this study, a novel BG-based scaffold system was developed, which incorporates a genipin cross-linked CS-PVP coating on the BG foam (G-CS-PVP/BG) fabricated by the replication method. Genipin was introduced to improve the mechanical strength and degradation properties of the CS-PVP/BG scaffold. The physicochemical properties of three types of BG scaffolds were characterized. In addition, the possibility of using the scaffold as a drug release carrier was discussed. The release behaviors of vancomycin from the pristine BG, CS-PVP/BG and G-CS-PVP/BG were quantitatively studied in PBS solution for 7 days. Preliminary cytotoxicity of the asprepared scaffolds was evaluated against MC3T3-E1 cells.
1% genipin solution was added into the CS-PVP solution at a ratio of 1:20 (v/v). The as-prepared BG scaffold was then immersed in the CS-PVP solution for 10 min at room temperature. The samples were subsequently kept for 24 h at room temperature. The G-CS-PVP/BG scaffold was thus obtained after being dried at 37 °C for 24 h. 2.3. Physicochemical characterization High porosity of the BG scaffold was determined by their mass and dimensions according to Eq. (1). In Eq. (1), ρsolid = 2.7 g/cm3 is the density of solid BG, W1 is the mass of the scaffold, and V1 is the volume of the scaffold (which was determined by measuring the scaffold dimensions using digital calipers). P scaffold ¼
1−
2. Materials and methods 2.1. The 45S5 Bioglass®-based scaffolds BG scaffolds with interconnected pores were prepared by the foam replication method, following a procedure as reported elsewhere [24]. As schematically shown in Fig. 1(a), 0.8 g polyvinyl alcohol (PVA) (Merck, Germany) was dissolved in 10 mL distilled water and heated to 80 °C until a homogeneous solution was obtained. 50 wt.% commercial 45S5 Bioglass® power (particle size ~ 5 μm) was added to the PVA solution. Polyurethane (PU) foams (45 ppi, Eurofoam GmbH, Wiesbaden, Germany) with dimension of 10 × 10 × 10 mm3 were immersed in the BG slurry. Excess slurry was squeezed out and dried at room temperature overnight. The above steps were repeated twice. The green bodies were subsequently sintered to prepare the BG scaffold. The sintering schedule is shown in Fig. 1(b) which contains an intermediate step remaining at 400 °C for 1 h to remove the PU foam template, followed by increasing the temperature to 1000 °C and remaining for 2 h to sinter BG. 2.2. The CS-PVP/BG and G-CS-PVP/BG scaffolds 0.2 g chitosan (CS, Sigma, USA) and 0.1 g polyvinylpyrrolidone (PVP, Aladdin, China) were dissolved in 10 ml 1% acetic acid solution until a homogeneous polymer solution was obtained. The as-synthesized BG scaffold was immersed in the CS-PVP solution for 10 min at room temperature. The resulting CS-PVP/BG scaffold was consequently obtained after being dried at room temperature over 24 h.
W1 V1 ρsolid
100%:
ð1Þ
Composition of the scaffold was measured using Fourier transform infrared spectroscopy (FT-IR, Nicolet, USA). The spectra were collected in the 4000–400 cm−1 range with a resolution of 32 cm−1. The crystalline composition of the as-synthesized scaffold was determined using X-ray diffraction (XRD) analysis (D8 Advance, Bruker AXS, Germany). Data were obtained over a 2θ range of 5°–70°. The cross-linking degree of the G-CS-PVP/BG scaffold was obtained using the ninhydrin assay, which was defined as the percentage of free amino groups in the cross-linked chitosan [21]. Typically, 5 mg CS-PVP/BG and G-CS-PVP/BG scaffolds were added into 1 mL ninhydrin solution and heated at 90 °C for 15 min. After cooling to the room temperature, 8 mL 50% ethyl alcohol was added to the above solution. The optical absorbance of the solution was recorded by a spectrophotometer (Cary 100 UV–VIS, Agilent Technologies, USA) while glycine at various concentrations was used as the standard. The microstructure and surface morphology of the scaffold were characterized using scanning electron microscopy (SEM) (Zeiss Leica, Germany). Samples were gold-sputtered and observed at an accelerating voltage of 20 kV. 2.4. In vitro enzymatic degradation The in vitro biodegradation of the CS-PVP/BG and the G-CS-PVP/BG scaffold was investigated by lysozyme (Sigma, Canada) with an activity of 40,000 U · mg−1 in solid form. The samples were weighed and then sterilized by UV light. The sterilized samples were immersed in 10 mL PBS solution at 37 °C with 1000 U · mg−1 of lysozyme. Each sample was incubated in an orbital shaker at a speed of 100 rpm under dark. The incubation solution was changed every day. At predetermined time intervals, samples were removed from the incubation solution and washed with deionized water, and then dried at 37 °C for 24 h. The weight loss caused by the enzymatic degradation was calculated according to Eq. (2), where W0 is the initial mass and W1 is the mass after soaking in PBS at a given time: Weight loss ¼
W0 −W1 100: W0
ð2Þ
2.5. In vitro drug release
Fig. 1. (a) Schematic fabrication of the Bioglass® scaffolds. (b) Heat treatment process used for burning-out the PU foam and sintering the Bioglass® green bodies to fabricate the Bioglass®-based scaffolds.
Vancomycin was loaded into the BG scaffold, CS-PVP/BG scaffold and G-CS-PVP/BG scaffold respectively. Three samples of each group were immersed in 10 mL PBS solution at 37 °C. Each sample was incubated in an orbital shaker at a speed of 100 rpm. The drug release kinetics in PBS was determined using a UV spectrophotometer at a wavelength of 280 nm. A calibration curve was obtained with vancomycin concentrations in the range of 0.001–0.28 mg/mL. It was observed that the
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
calibration curve follows Beer's law: A = 4.837 C, where A is the absorbance and C is the concentration of vancomycin. 2.6. Cell culture MC3T3-E1 cells were cultured in a modified Eagle's minimum essential medium (Gibco, Invitrogen, USA), with a volume fraction of 10% fetal bovine serum(Gibco, Invitrogen, USA) under a humidified atmosphere of 5% CO2 at 37 °C. Medium was changed every two or three days. The samples were immersed in the extracting media at 100 mg/mL and incubated for 24 h in a cell incubator (37 °C and 5% CO2) for preparation of the extracts. 2.6.1. Cell proliferation Cell proliferation was quantitatively analyzed using cell counting kit-8 (CCK-8; Beyondtime Bio-Tech, China). Cells were seeded into 96well plates at 2000 cells/well and incubated for 24 h. Medium from each well was subsequently removed and replaced with 200 μL extraction medium from one of the specimens. Cells were cultured for 1, 3 and 5 days, respectively. 20 μL CCK-8 reagent was subsequently added per well, followed by another 2 h incubation. The absorbance of produced WST-8formazan at 450 nm was measured by a microplate reader (Model 680, Bio-Rad Laboratories). Analysis was performed in triplicate. The results were expressed as percentages of the mean absorbance (optical density) of treated vs. controls. The mean optical density of the control (the BG) was set to represent 100% viability. 2.6.2. Live/dead staining The scaffold was placed into a 24-well flat culture plate, followed by immersing in α-MEM for 4 h in a cell incubator. 1 × 105 cells were then seeded into each well and cultured for 4 days. The samples were subsequently washed with PBS solution for 3 times. Cells were labeled with a freshly made solution of dyes taken from a final concentration of 10 μM calcein AM [4′5′-bis (N′N-bis(carboxymethyl) aminomethyl fluorescein acetoxymethyl ester)] and 15 μMPI (propidium iodide) in phosphate buffer solution (PBS). After being stained for 15 min at 37 °C, the samples were washed with PBS three times and characterized using a fluorescence microscope (IX 2-UCB, Olympus, Germany). 2.6.3. Cell skeleton and morphology The cytoskeleton organization of MC3T3-E1 cells grown on extracts of composites was analyzed using Filamentous actin (F-actin) staining. MC3T3-E1 cells were cultured at a seeding density of 1 × 104 cells/mL for 1 day and 3 days, respectively. Cells were subsequently fixed with 4% paraformaldehyde for 10 min, permeabilized with 0.1% TritonX100 for 5 min, blocked with 1% bovine serum albumin for 20 min, stained with Rhodamine Phalloidin for 20 min and stained with DAPI for 5 min in the dark. The stained MC3T3-E1 cells were observed under laser confocal microsopy (LSM 710, Zeiss, Germany). For cell morphology characterization, the scaffold was placed into a 24-well flat culture plate, and was immersed in α-MEM for 4 h in a cell incubator, then 4 × 104 cells were seeded into each well and cultured for 4 days and 7 days. After culturing, the samples were washed twice with PBS solution and fixed in 2.5% glutaraldehyde for 3 h at 4 °C. The specimens were subsequently dehydrated in ethanol solutions of various concentrations (30, 50, 70, 90 and 100%) for 15 min respectively prior to be freeze-dried. After being completely dried, morphology of the cells was observed by SEM. 3. Results 3.1. Porosity of the scaffold and cross-linking degree Porosity of the BG scaffold is found to be 96% according to Eq. (1). The sintered BG scaffold shows interconnected macroporous structure
475
with pore diameters in the range of 200–700 μm, which is promising for cell attachment and adhesion [2,25]. SEM images of the as-fabricated scaffolds are shown in Fig. 2. The sintered BG scaffold exhibits well-defined porous structure (Fig. 2(a) and (b)). After introduction of CS-PVP to the BG scaffold, pores are partially covered while the highly porous interconnected structure is still visible (Fig. 2(c) and (d)). The porous structure of the scaffold remains well after genipin cross-linking (Fig. 2(e) and (f)). Successful cross-linking of the genipin on the CS-PVP/BG scaffold is confirmed by the color changing from white to blue after cross-linking. The cross-linking degree is found to be 19.6%, determined by the ninhydrin assay. 3.2. Physicochemical characterization Fig. 3(A) shows the FTIR spectra of the BG, the CS-PVP/BG and the G-CS-PVP/BG, respectively. Characteristic peaks of BG (sodium calcium silicate [Na2Ca2Si3O9]) at 1090 cm− 1, 624 cm− 1, 529 cm− 1 and 470 cm−1 are dominant in all samples. After incorporation of CS-PVP with the BG, bands related to the presence of PVP and CS are observed both at the spectrum of CS-PVP/BG and that of G-CS-PVP/BG (Fig. 3A(b), (c)). Peaks due to absorption bands of PVP are found at 1650 cm− 1 (C_O stretching), 1420 cm− 1 (C–H bending vibration), and 1290 cm−1(C–N stretching) respectively while three characteristic peaks of CS are identified at 3450 cm− 1 (hydroxyl stretching), 1650 cm−1 (amide I) and 1560 cm− 1 (amide II), respectively. When the CS-PVP/BG was cross-linked with genipin, the peak associated with amide I shifts from 1654 to 1647 cm− 1 while the peak arising from amide II shifts from 1567 to 1559 cm− 1 (Fig. 3A(b) and (c)), confirming the reaction between amino groups of CS and ester groups of genipin [26], and formation of hydrogen bonding between CS and genipin [27,28]. Fig. 3B shows XRD patterns of various scaffolds. Compared with the standard PDF 75–1687, both the angular position and peak intensity of the sintered BG are seen to match the crystalline phase of sodium calcium silicate [Na2Ca2Si3O9] [29] (Fig. 3B(a)). The characteristic peaks of CS and PVP are not found in the spectrum of either the CS-PVP/BG scaffold or the G-CS-PVP/BG scaffold, but only the characteristic peaks of Na2Ca2Si3O9 are observed in Fig. 3B(b) and (c). The results demonstrate that no recrystallization of CS and PVP has occurred during the coating process. For better understanding the influence of extracts from the scaffolds on the biocompatibility, we have measured pH values of the extracts from different scaffolds. Before the extract preparation, pH values of the 45S5 BG, the CS-PVP/BG and the G-CS-PVP/BG were controlled by soaking the scaffolds in DI water over 3 days while DI water was replaced every 8 h. 100 mg/ml samples were subsequently immersed in the extracting media and incubated for 24 h in a cell incubator (37 °C and 5% CO2) to get the extracts. pH values of three extraction media are higher than 7 (Fig. 4) due to alkalinity of the 45S5 BG. The 45S5 BG extract shows the highest pH value while the G-CS-PVP/BG extract exhibits the lowest pH value. 3.3. In vitro enzymatic degradation Fig. 5 presents the weight loss ratios of the CS-PVP/BG compared with the G-CS-PVP/BG scaffold over 2 and 7 days' lysozyme incubation respectively. After 2 days, the weight loss ratio of the CS-PVP/BG scaffold (39.3%) is much higher than that of the G-CS-PVP/BG scaffold (32.6%) (P b 0.01). This may be attributed to degradation of water-soluble PVP and partial degradation of CS in PBS solution at 37 °C. After 7 days' incubation, the CS-PVP/BG scaffold shows a significantly increased weight loss (46.9%) compared to 35.3% weight loss of the G-CS-PVP/BG scaffold (35.3%, P b 0.05), suggesting significantly enhanced stability of the scaffold after cross-linking. It has been reported that significant degradation of CS in lysozyme is arisen from the binding of N-acetylglucosamine
476
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Fig. 2. SEM images of (a, b) the BG, (c, d) the CS-PVP/BG, (e, f) the G-CS-PVP/BG scaffold.
residues to the active sites of lysozyme [30]. The chemical cross-linking between CS and genipin, however, is expected to form a cyclic covalentcross-linked structure, which is beneficial for suppressing degradation of CS in lysozyme [21].
Fig. 3. (A) FT-IR spectra of (a) the BG scaffold, (b) the CS-PVP/BG scaffold, and (c) the G-CS-PVP/BG scaffold. (B) XRD patterns of (a) the BG, (b) the CS-PVP/BG and (c) the G-CS-PVP/BG. Characteristic peaks of Na2Ca2Si3O9 phase are marked by (▼).
3.4. Cytotoxicity Biocompatibility of the scaffold is essential for its application in tissue engineering. Cell viabilities of the as-synthesized scaffolds were thus evaluated using the CCK-8 assay. The pristine BG scaffold was used as the control. As shown in Fig. 6, the CS-PVP/BG scaffold shows maximum toxicity with MC3T3-E1 cells when compared with both the BG and the G-CS-PVP/BG. The cell viability at the G-CS-PVP/BG is
Fig. 4. pH values of the BG extract, the CS-PVP/BG extract and the G-CS-PVP/BG extract by soaking the scaffolds in the extracting media for 24 h (37 °C and 5% CO2).
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Fig. 5. Weight loss of the CS-PVP/BG scaffold and the G-CS-PVP/BG scaffold immersed in lysozyme solution for 2 and 7 days (*P b 0.05; **P b 0.01).
comparable to that at the control, suggesting excellent biocompatibility of the resulting G-CS-PVP/BG scaffold. 3.5. Fluorescence staining Live/dead cell staining results on the as-synthesized scaffolds are shown in Fig. 7. Cells are homogeneously distributed on the struts of all types of scaffolds after 4 days' culture. Few dead cells are observed at the G-CS-PVP/BG scaffold, suggesting the excellent biocompatibility of the scaffold. It is interesting to find out that cells can penetrate into all scaffolds, confirming that highly porous interconnected structures are attractive for cell attachment and adhesion. Live/dead cell staining was also used to count the live cells cultured with the three types of extracts after 1, 3 and 5 days respectively. As shown in Fig. S1, cell numbers obtained with the G-CS-PVP/BG extract (Fig. S1(c), (f), and (i)) are comparable to those observed with the 45S5 BG extract (Fig. S1(a), (d), and (g)) while significant reduction in cell numbers are identified with the CS-PVP/BG (Fig. S1(b), (e), (h)). Live/dead fluorescence staining results are well consistent with the CCK-8 results. 3.6. Cell skeleton and morphology Cytoskeletons of cells were stained green with Rhodamine Phalloidin while nuclei of cells were stained blue with Hoechst 33258. Results were identified using LSCM images which revealed the formation of the actin cytoskeleton, providing information of the structural framework and participates in cell migration [31]. There are bright
477
and well-pronounced elongated filopodia on the edges of cells on the all scaffolds after 1 day (Fig. 8(a), (c), and (e)). Almost all cells are grouped and exhibit elongated morphology on the day 4 (Fig. 8(b), (d), and (f)). The interconnection between surrounding cells is evident and the presence of some membrane channels between adjacent cells is apparent. Cells containing well-stretched F-actin bundles are observed obviously. These actin-rich structures known as the tunneling nanotubes, could mediate the intercellular transfer of organelles, plasma membrane components and cytoplasmic molecules [32,33]. Fluorescent cell skeletons were further studied on the cells cultured with the scaffold extracts. As shown in Fig. S2, more cells are found with the 45S5 BG extract and the G-CS-PVP/BG extract when compared to cells incubated with the CS-PVP/BG extract. Particularly, the cells with the G-CS-PVP/BG show much better stretched morphology. On the other hand, polymers (CS and PVP) dissolved in the culture medium might also influence the cell viability. The cross-linking of CS allows fewer ions (from 45S5 BG) and polymers (from CS and PVP) to be released into the extracting media, leading to similar cell viability obtained with the G-CS-PVP/BG extract and the 45S5 BG extract. SEM images of pre-osteoblasts MC3T3-E1 cells seeded scaffolds after 4 and 7 days are shown in Fig. 9. After 4 days' culture, cells on all scaffolds are fully attached on the surface and extend their cytoplasm in the form of thin and long fibrils from the leading edges. Cells cultured on the BG scaffold (Fig. 9(a)) and the G-CS-PVP/BG (Fig. 9(e)) scaffold present polygonal morphology and expand relatively better than those on the CS-PVP/BG scaffold (Fig. 9(c)), which may be attributed to the relatively high toxicity of the CS-PVP/BG scaffold compared to other two scaffolds as we found from the CCK-8 assay. After 7 days' incubation, cells cultured on the BG scaffold and the CS-PVP/BG scaffold are grouped and exhibit similar morphology. Well interconnection between cells is observed. SEM results confirm that biocompatibility of the G-CS-PVP/BG scaffold is improved by cross-linking. The resulting G-CS-PVP/BG scaffold exhibits superb properties for promoting the spreading and proliferation of MC3T3-E1 cells. 3.7. In vitro drug release For purpose of developing novel scaffolds for multifunctional applications, we further investigated in vitro drug release from the as-synthesized scaffolds. The release level of vancomycin from the as-prepared scaffold was determined based on the total amount of vancomycin incorporated in the corresponding scaffold. Cumulative release results in PBS solution are shown in Fig. 10. There are two release stages observed at the scaffolds without genipin. The release curve on the G-CS-PVP/BG scaffold, however, exhibits three stages. At the first stage, a rapid drug release is observed for all samples, which is due to the dissolution of the surface attached vancomycin. The G-CS-PVP/BG scaffold shows the lowest initial burst release. Secondarily, the release at the BG scaffold completes within 24 h, which is in consistent with our previous work [34]. After introduction of CS-PVP onto BG, the release is improved to 48 h until a plateau is reached. Further cross-linkage of genipin to the CS-PVP/BG provides additional advantages for controlled and stained release. The drug entrapped at the G-CS-PVP/BG scaffold is released in a controlled manner over a 7-day period, which may be attributed to the cross-linking of CS with genipin. 4. Discussion
Fig. 6. CCK-8 counts of cells cultured on: the BG, the CS-PVP/BG and the G-CS-PVP/BG scaffolds. Error bars represent means ± SD for n = 4 (*P b 0.05; **P b 0.01).
Due to the unique biocompatibility and poor mechanic property of the BG, numerous natural and synthetic polymers have been used to increase the stiffness of BG scaffolds [29,34–38]. Hydrophilic polymers usually suffer from rapid degradation while hydrophobic nature of hydrophobic polymers always hinders cells' adhesion and expansion. In this work, we used genipin as the cross-linking reagent to overcome degradation issues of the CS-PVP/BG scaffold. Introduction of genipin
478
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Fig. 7. Fluorescence staining of cells cultured on: (a) the BG, (b) the CS-PVP/BG and (c) the G-CS-PVP BG scaffolds for 4 days.
remains the highly porous interconnected structures which are attractive for cell seeding and proliferation. The CCK-8 results show that G-CS-PVP coating on the BG scaffold does not affect cell viability and proliferation, suggesting that presence of genipin causes no negative effect on mitochondrial functions but enhanced biocompatibility of the scaffold. These results are consistent with the cell morphology results which show that increased cell
morphology on the CS-PVP/45S5 BG scaffold after cross-linking. These results were further evident by cell skeleton observations. MC3T3-E1 cells cultured on the resulting scaffold exhibit a favorable F-actin distribution. Enhanced cell proliferation is obtained after genipin cross-linking with the CS-PVP/BG scaffold. This may be due to presence of hydroxyl groups at genipin which enhance the hydrophilicity of the scaffold and facilitate cells' adhesion, dispersion and proliferation.
Fig. 8. Fluorescence images of cell skeletons on: (a, b) the BG, (c, d) the CS-PVP/BG, and (e, f) the G-CS-PVP/BG scaffolds. (a, c, e) Cells were cultured for 1 day, and (b, d, f) cells were incubated for 3 days.
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
479
Fig. 9. SEM micrographs of cell morphology on: (a, b) the 45S5 BG scaffold, (c, d) the CS-PVP/45S5 BG scaffold and (e, f) the G-CS-PVP/45S5 BG scaffold.
In vitro drug release results suggest that vancomycin entrapped in the resulting G-CS-PVP/BG scaffold shows a better controlled stained release kinetics than the other scaffolds. This result could be attributed to two reasons. Firstly, both CS and PVP are hydrophilic polymers and the degradation rate is quick high in PBS solution at 37 °C. But genipin cross-linking decreases the degradation of polymers. Secondly, the
cross-linking degree of CS is low in the acidic solution (19.6%). It is thus harder for genipin to react with protonated amino groups in CS via nucleophilic attack [2]. 5. Conclusions Novel G-CS-PVP/BG scaffolds were prepared using genipin as the cross-linker of the chitosan coating. The physicochemical properties of the scaffolds were investigated. Introduction of G-CS-PVP onto the BG scaffolds has improved resistance to enzymatic degradation of the scaffolds while remained good biocompatibility and unique porous structures of the BG scaffolds for good adhesion and proliferation of MC3T3-E1 cells. Cells on the scaffolds show well-stretched F-actin bundles after 4 days' incubation. Furthermore, the antibiotic vancomycin was entrapped into the BG scaffolds via CS-PVP coating. Vancomycin in the G-CS-PVP/BG scaffolds shows significantly improved release period of 7 days from 24 h of the BG scaffold. Our preliminary results suggest that the as-prepared G-CS-PVP/BG scaffolds are promising for tissue engineering applications considering their excellent bioactivity and in-situ antibiotic releasing function. Acknowlegments
Fig. 10. In vitro vancomycin release behaviors from the BG scaffold, the CS-PVP/BG scaffold and the G-CS-PVP/BG scaffold.
Financial support for this work from the Chinese National Nature Science Foundation (21374081, 51433005), and Zhejiang National Nature Science Foundations (LQ15H180003, Y13H180013) are acknowledged.
480
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.06.046. References [1] M. Yazdimamaghani, D. Vashaee, S. Assefa, K.J. Walker, S.V. Madihally, G.A. Köler, L. Tayebi, J. Biomed. Nanotechnol. 10 (2014) 911–931. [2] C. Vitale-Brovarone, E. Verne', L. Robiglio, P. Appendino, F. Bassi, G. Martinasso, G. Muzio, R. Canuto, Acta Biomater. 3 (2007) 199–208. [3] K.K. Mallick, J. Winnett, J. Am. Ceram. Soc. 95 (2012) 2680–2686. [4] S. Bose, M. Roy, A. Bandyopadhyay, Trends Biotechnol. 30 (2012) 546–554. [5] Y. Liu, K. Gilmore, J. Chen, V. Misoska, G.G. Wallace, Chem. Mater. 19 (2007) 2721–2723. [6] K.T. Shalumon, J.P. Chen, Curr. Pharm. Des. 21 (2015) 1979–1990. [7] M.N. Rahaman, D.E. Day, B.S. Bal, Q. Fu, S.B. Jung, L.F. Bonewald, A.P. Tomsia, Acta Biomater. 7 (2011) 2355–2373. [8] G.G. Niederauer, M.A. Slivka, N.C. Leatherbury, D.L. Korvick, H.H. Harroff, W.C. Ehler, C.J. Dunn, K. Kieswetter, Biomaterials 21 (2000) 2561–2574. [9] P. Ducheyne, Q. Qiu, Biomaterials 20 (1999) 2287–2303. [10] S. Abdollahi, A.C.C. Ma, M. Cerruti, Langmuir (2013) 1466–1474. [11] A. Mohammadkhah, L.M. Marquardt, S.E. Sakiyama-Elbert, D.E. Day, A.B. Harkins, Mater. Sci. Eng. C 49 (2015) 632–639. [12] C. Wu, Y. Ramaswamy, P. Boughton, H. Zreiqat, Acta Biomater. 4 (2008) 343–353. [13] W. Li, H. Wang, Y. Ding, E.C. Scheithauer, O.-M. Goudouri, A. Grünewald, R. Detsch, S. Agarwal, A.R. Boccaccini, J. Mater. Chem. B 3 (2015) 3367–3378. [14] A. Di Martino, M. Sittinger, M.V. Risbud, Biomaterials 26 (2005) 5983–5990. [15] N.M. Julkapli, H.M. Akil, Z. Ahmad, Compos. Interfaces 18 (2011) 449–507. [16] R.A.A. Muzzarelli, Carbohydr. Polym. 77 (2009) 1–9.
[17] D.Y. Zuo, Y. Wang, W.L. Xu, H.T. Liu, Adv. Polym. Technol. 31 (2012) 310–318. [18] D. Yu, X. Wang, X. Li, E. Chian, Y. Li, Y. Liao, Acta Biomater. 9 (2013) 5665–5672. [19] M.F. Butler, Y.F. Ng, P.D.A. Pudney, J. Polym. Sci., Polym. Chem. Ed. 41 (2003) 3941–3953. [20] M. Mekhail, K. Jahan, M. Tabrizian, Carbohydr. Polym. 108 (2014) 91–98. [21] F.L. Mi, H.W. Sung, S.S. Shyu, C.C. Su, C.K. Peng, Polymer 44 (2003) 6521–6530. [22] T.H. Tsai, J. Westly, T.F. Lee, C.F. Chen, J. Liq. Chromatogr. 17 (1994) 2199–2205. [23] L. Bi, Z. Cao, Y. Hu, Y. Song, L. Yu, B. Yang, J. Mu, Z. Huang, Y. Han, J. Mater. Sci. Mater. Med. 22 (2011) 51–62. [24] Q. Chen, I.D. Thompson, A.R. Boccaccini, Biomaterials 27 (2006) 2414–2425. [25] V. Karageorgiou, D. Kaplan, Biomaterials 26 (2005) 5474–5491. [26] F.L. Mi, Biomacromolecules 6 (2005) 975–987. [27] F.L. Mi, H.W. Sung, S.S. Shyu, J. Appl. Polym. Sci. 81 (2014) 1700–1711. [28] C.D. Ji, N. Annabi, A. Khademhosseini, F. Dehghani, Acta Biomater. 7 (2011) 1653–1664. [29] Q. Yao, P. Nooeaid, R. Detsch, J.A. Roether, Y.M. Dong, O.M. Goudouri, D.W. Schubert, A.R. Boccaccini, J. Biomed. Mater. Res. A 102 (2014) 4510–4518. [30] B.S. Liu, T.B. Huang, Macromol. Biosci. 8 (2008) 932–941. [31] T. Jiang, Z. Zhang, Y. Zhou, Y. Liu, Z. Wang, H. Tong, X. Shen, Y. Wang, Biomacromolecules 11 (2010) 1254–1260. [32] A. Rustom, R. Saffrich, I. Markovic, P. Walther, H.H. Gerdes, Science 303 (2004) 1007–1010. [33] H.H. Gerdes, R.N. Carvalho, Curr. Opin. Cell Biol. 20 (2008) 470–475. [34] Q. Yao, P. Nooeaid, J.A. Roether, Y. Dong, Q. Zhang, A.R. Boccaccini, Ceram. Int. 39 (2013) 7517–7522. [35] W. Li, M.-I. Pastrama, Y. Ding, K. Zheng, C. Hellmich, A.R. Boccaccini, J. Mech. Behav. Biomed. Mater. 40 (2014) 85–94. [37] M. Montazeri, S. Karbasi, M.R. Foroughi, A. Monshi, R. Ebrahimi-Kahrizsangi, J. Mater. Sci. Mater. Med. 26 (2015) 62. [38] G. Yang, X. Yang, L. Zhang, M. Lin, X. Sun, X. Chen, Z. Gou, Mater. Lett. 75 (2012) 80–83.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Multifunctional chitosan/polyvinyl pyrrolidone/45S5 Bioglass® scaffolds for MC3T3-E1 cell stimulation and drug release Qingqing Yao a, Wei Li b, Shanshan Yu a, Liwei Ma a, Dayong Jin c,d, Aldo R. Boccaccini b,⁎, Yong Liu a,d,⁎⁎ a
Institute of Advanced Materials for Nano-Bio Applications, School of Ophthalmology & Optometry, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou, Zhejiang 325027, China Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, Cauerstrasse 6, Erlangen 91058, Germany Institute for Biomedical Materials and Devices, Faculty of Science, University of Technology Sydney, NSW 2007, Australia d Advanced Cytometry Labs, ARC Center of Excellence for Nanoscale BioPhotonics, Macquarie University, Sydney, NSW 2109, Australia b c
a r t i c l e
i n f o
Article history: Received 23 January 2015 Received in revised form 12 May 2015 Accepted 23 June 2015 Available online 9 July 2015 Keywords: Chitosan Polyvinyl pyrrolidone 45S5 Bioglass® Scaffolds Drug release MC3T3-E1 cells
a b s t r a c t Novel chitosan–polyvinyl pyrrolidone/45S5 Bioglass® (CS-PVP/BG) scaffolds were prepared via foam replication and chemical cross-linking techniques. The pristine BG, CS-PVP coated BG and genipin cross-linked CS-PVP/BG (G-CS-PVP/BG) scaffolds were synthesized and characterized in terms of chemical composition, physical structure and morphology respectively. Resistance to enzymatic degradation of the scaffold is improved significantly with the use of genipin cross-linked CS-PVP. The bio-effects of scaffolds on MC3T3-E1 osteoblast-like cells were evaluated by studying cell viability, adhesion and proliferation. The CCK-8 assay shows that cell viability on the resulting G-CS-PVP/BG scaffold is improved obviously after cross-linking of genipin. Cell skeleton images exhibit that well-stretched F-actin bundles are obtained on the G-CS-PVP/BG scaffold. SEM results present significant improvement on the cell adhesion and proliferation for cells cultured on the G-CS-PVP/BG scaffold. The drug release performance on the as-synthesized scaffold was studied in a phosphate buffered saline (PBS) solution. Vancomycin is found to be released in burst fashion within 24 h from the pristine BG scaffold, however, the release period from the G-CS-PVP/BG scaffold is enhanced to 7 days, indicating improved drug release properties of the G-CS-PVP/BG scaffold. Our results suggest that the G-CS-PVP/BG scaffolds possess promising physicochemical properties, sustained drug release capability and good biocompatibility for MC3T3-E1 cells' proliferation and adhesion, suggesting their potential applications in areas such as MC3T3-E1 cell stimulation and bone tissue engineering. © 2015 Published by Elsevier B.V.
1. Introduction Bone defects are usually arisen from trauma, infection, tumor, osteoporotic fracture, osteonecrosis or congenital deformity [1,2]. Traditional treatments e. g. autograft and allograft transplants have been seen as the effective methods for bone repair. These techniques, however, are suffering from many drawbacks such as donor site morbidity and graft rejection [3–5]. Recently, tissue engineering techniques with the use of scaffold-based approaches have been attracting increasing attention [3,4]. 3D scaffolds can provide numerous advantages for diverse applications such as suitable environment for cell attachment, proliferation, and differentiation [5], and even as the carrier for drug delivery systems [6].
⁎ Corresponding author. ⁎⁎ Correspondence to: Y. Liu, Institute of Advanced Materials for Nano-Bio Applications, School of Ophthalmology & Optometry, Wenzhou Medical University, 270 Xueyuan Xi Road, Wenzhou, Zhejiang 325027, China. E-mail addresses: [email protected] (A.R. Boccaccini), [email protected] (Y. Liu).
http://dx.doi.org/10.1016/j.msec.2015.06.046 0928-4931/© 2015 Published by Elsevier B.V.
45S5 Bioglass® (BG) has been intensively studied as the alternative for bone tissue engineering considering its ability to integrate with bone [7,8]. BG can interact with the surrounding tissue milieu and serve as a substrate for osteogenic stem cells' attachment and differentiation [9]. However, the porous BG scaffolds always suffer from relatively low mechanical strength and high brittleness [10]. Incorporation of BG with particular polymers (e.g. chitosan based system) has been seen as an efficient way to improve the properties of scaffolds [11–14]. Chitosan (CS), a linear polysaccharide composed of glucosamine and N-acetyl glucosamine units linked by β(1–4)-glycosidic bonds and deacetylated from chitin [15], has been well known for its unique properties including biodegradability, biocompatibility, non-toxicity, hydrophilicity and nonimmunogenicity. Various techniques, including the cross-linking, blending, and chemical modification, have been developed to improve the mechanical properties of CS-based materials [15,16]. For example, polyvinylpyrrolidone (PVP), a hydrophilic polymer excipient, can form good miscibility with CS in diluted acidic aqueous solutions [17,18]. PVP can also enhance the thermodynamic stability in a binary solution. To prevent CS from enzymatic degradation and retaining the stability of CS under wet conditions, the cross-linking methods are usually applied to anchor the amine groups on the polymeric chains of CS [19]. Genipin
474
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
is obtained from geniposide, which may be isolated from the fruits of Gardenia jasminoides Ellis [20]. It has been found that genipin is an excellent natural cross-linker for CS, since genipin is less cytotoxic than glutaraldehyde and many other commonly used synthetic crosslinking reagents [21–23]. In this study, a novel BG-based scaffold system was developed, which incorporates a genipin cross-linked CS-PVP coating on the BG foam (G-CS-PVP/BG) fabricated by the replication method. Genipin was introduced to improve the mechanical strength and degradation properties of the CS-PVP/BG scaffold. The physicochemical properties of three types of BG scaffolds were characterized. In addition, the possibility of using the scaffold as a drug release carrier was discussed. The release behaviors of vancomycin from the pristine BG, CS-PVP/BG and G-CS-PVP/BG were quantitatively studied in PBS solution for 7 days. Preliminary cytotoxicity of the asprepared scaffolds was evaluated against MC3T3-E1 cells.
1% genipin solution was added into the CS-PVP solution at a ratio of 1:20 (v/v). The as-prepared BG scaffold was then immersed in the CS-PVP solution for 10 min at room temperature. The samples were subsequently kept for 24 h at room temperature. The G-CS-PVP/BG scaffold was thus obtained after being dried at 37 °C for 24 h. 2.3. Physicochemical characterization High porosity of the BG scaffold was determined by their mass and dimensions according to Eq. (1). In Eq. (1), ρsolid = 2.7 g/cm3 is the density of solid BG, W1 is the mass of the scaffold, and V1 is the volume of the scaffold (which was determined by measuring the scaffold dimensions using digital calipers). P scaffold ¼
1−
2. Materials and methods 2.1. The 45S5 Bioglass®-based scaffolds BG scaffolds with interconnected pores were prepared by the foam replication method, following a procedure as reported elsewhere [24]. As schematically shown in Fig. 1(a), 0.8 g polyvinyl alcohol (PVA) (Merck, Germany) was dissolved in 10 mL distilled water and heated to 80 °C until a homogeneous solution was obtained. 50 wt.% commercial 45S5 Bioglass® power (particle size ~ 5 μm) was added to the PVA solution. Polyurethane (PU) foams (45 ppi, Eurofoam GmbH, Wiesbaden, Germany) with dimension of 10 × 10 × 10 mm3 were immersed in the BG slurry. Excess slurry was squeezed out and dried at room temperature overnight. The above steps were repeated twice. The green bodies were subsequently sintered to prepare the BG scaffold. The sintering schedule is shown in Fig. 1(b) which contains an intermediate step remaining at 400 °C for 1 h to remove the PU foam template, followed by increasing the temperature to 1000 °C and remaining for 2 h to sinter BG. 2.2. The CS-PVP/BG and G-CS-PVP/BG scaffolds 0.2 g chitosan (CS, Sigma, USA) and 0.1 g polyvinylpyrrolidone (PVP, Aladdin, China) were dissolved in 10 ml 1% acetic acid solution until a homogeneous polymer solution was obtained. The as-synthesized BG scaffold was immersed in the CS-PVP solution for 10 min at room temperature. The resulting CS-PVP/BG scaffold was consequently obtained after being dried at room temperature over 24 h.
W1 V1 ρsolid
100%:
ð1Þ
Composition of the scaffold was measured using Fourier transform infrared spectroscopy (FT-IR, Nicolet, USA). The spectra were collected in the 4000–400 cm−1 range with a resolution of 32 cm−1. The crystalline composition of the as-synthesized scaffold was determined using X-ray diffraction (XRD) analysis (D8 Advance, Bruker AXS, Germany). Data were obtained over a 2θ range of 5°–70°. The cross-linking degree of the G-CS-PVP/BG scaffold was obtained using the ninhydrin assay, which was defined as the percentage of free amino groups in the cross-linked chitosan [21]. Typically, 5 mg CS-PVP/BG and G-CS-PVP/BG scaffolds were added into 1 mL ninhydrin solution and heated at 90 °C for 15 min. After cooling to the room temperature, 8 mL 50% ethyl alcohol was added to the above solution. The optical absorbance of the solution was recorded by a spectrophotometer (Cary 100 UV–VIS, Agilent Technologies, USA) while glycine at various concentrations was used as the standard. The microstructure and surface morphology of the scaffold were characterized using scanning electron microscopy (SEM) (Zeiss Leica, Germany). Samples were gold-sputtered and observed at an accelerating voltage of 20 kV. 2.4. In vitro enzymatic degradation The in vitro biodegradation of the CS-PVP/BG and the G-CS-PVP/BG scaffold was investigated by lysozyme (Sigma, Canada) with an activity of 40,000 U · mg−1 in solid form. The samples were weighed and then sterilized by UV light. The sterilized samples were immersed in 10 mL PBS solution at 37 °C with 1000 U · mg−1 of lysozyme. Each sample was incubated in an orbital shaker at a speed of 100 rpm under dark. The incubation solution was changed every day. At predetermined time intervals, samples were removed from the incubation solution and washed with deionized water, and then dried at 37 °C for 24 h. The weight loss caused by the enzymatic degradation was calculated according to Eq. (2), where W0 is the initial mass and W1 is the mass after soaking in PBS at a given time: Weight loss ¼
W0 −W1 100: W0
ð2Þ
2.5. In vitro drug release
Fig. 1. (a) Schematic fabrication of the Bioglass® scaffolds. (b) Heat treatment process used for burning-out the PU foam and sintering the Bioglass® green bodies to fabricate the Bioglass®-based scaffolds.
Vancomycin was loaded into the BG scaffold, CS-PVP/BG scaffold and G-CS-PVP/BG scaffold respectively. Three samples of each group were immersed in 10 mL PBS solution at 37 °C. Each sample was incubated in an orbital shaker at a speed of 100 rpm. The drug release kinetics in PBS was determined using a UV spectrophotometer at a wavelength of 280 nm. A calibration curve was obtained with vancomycin concentrations in the range of 0.001–0.28 mg/mL. It was observed that the
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
calibration curve follows Beer's law: A = 4.837 C, where A is the absorbance and C is the concentration of vancomycin. 2.6. Cell culture MC3T3-E1 cells were cultured in a modified Eagle's minimum essential medium (Gibco, Invitrogen, USA), with a volume fraction of 10% fetal bovine serum(Gibco, Invitrogen, USA) under a humidified atmosphere of 5% CO2 at 37 °C. Medium was changed every two or three days. The samples were immersed in the extracting media at 100 mg/mL and incubated for 24 h in a cell incubator (37 °C and 5% CO2) for preparation of the extracts. 2.6.1. Cell proliferation Cell proliferation was quantitatively analyzed using cell counting kit-8 (CCK-8; Beyondtime Bio-Tech, China). Cells were seeded into 96well plates at 2000 cells/well and incubated for 24 h. Medium from each well was subsequently removed and replaced with 200 μL extraction medium from one of the specimens. Cells were cultured for 1, 3 and 5 days, respectively. 20 μL CCK-8 reagent was subsequently added per well, followed by another 2 h incubation. The absorbance of produced WST-8formazan at 450 nm was measured by a microplate reader (Model 680, Bio-Rad Laboratories). Analysis was performed in triplicate. The results were expressed as percentages of the mean absorbance (optical density) of treated vs. controls. The mean optical density of the control (the BG) was set to represent 100% viability. 2.6.2. Live/dead staining The scaffold was placed into a 24-well flat culture plate, followed by immersing in α-MEM for 4 h in a cell incubator. 1 × 105 cells were then seeded into each well and cultured for 4 days. The samples were subsequently washed with PBS solution for 3 times. Cells were labeled with a freshly made solution of dyes taken from a final concentration of 10 μM calcein AM [4′5′-bis (N′N-bis(carboxymethyl) aminomethyl fluorescein acetoxymethyl ester)] and 15 μMPI (propidium iodide) in phosphate buffer solution (PBS). After being stained for 15 min at 37 °C, the samples were washed with PBS three times and characterized using a fluorescence microscope (IX 2-UCB, Olympus, Germany). 2.6.3. Cell skeleton and morphology The cytoskeleton organization of MC3T3-E1 cells grown on extracts of composites was analyzed using Filamentous actin (F-actin) staining. MC3T3-E1 cells were cultured at a seeding density of 1 × 104 cells/mL for 1 day and 3 days, respectively. Cells were subsequently fixed with 4% paraformaldehyde for 10 min, permeabilized with 0.1% TritonX100 for 5 min, blocked with 1% bovine serum albumin for 20 min, stained with Rhodamine Phalloidin for 20 min and stained with DAPI for 5 min in the dark. The stained MC3T3-E1 cells were observed under laser confocal microsopy (LSM 710, Zeiss, Germany). For cell morphology characterization, the scaffold was placed into a 24-well flat culture plate, and was immersed in α-MEM for 4 h in a cell incubator, then 4 × 104 cells were seeded into each well and cultured for 4 days and 7 days. After culturing, the samples were washed twice with PBS solution and fixed in 2.5% glutaraldehyde for 3 h at 4 °C. The specimens were subsequently dehydrated in ethanol solutions of various concentrations (30, 50, 70, 90 and 100%) for 15 min respectively prior to be freeze-dried. After being completely dried, morphology of the cells was observed by SEM. 3. Results 3.1. Porosity of the scaffold and cross-linking degree Porosity of the BG scaffold is found to be 96% according to Eq. (1). The sintered BG scaffold shows interconnected macroporous structure
475
with pore diameters in the range of 200–700 μm, which is promising for cell attachment and adhesion [2,25]. SEM images of the as-fabricated scaffolds are shown in Fig. 2. The sintered BG scaffold exhibits well-defined porous structure (Fig. 2(a) and (b)). After introduction of CS-PVP to the BG scaffold, pores are partially covered while the highly porous interconnected structure is still visible (Fig. 2(c) and (d)). The porous structure of the scaffold remains well after genipin cross-linking (Fig. 2(e) and (f)). Successful cross-linking of the genipin on the CS-PVP/BG scaffold is confirmed by the color changing from white to blue after cross-linking. The cross-linking degree is found to be 19.6%, determined by the ninhydrin assay. 3.2. Physicochemical characterization Fig. 3(A) shows the FTIR spectra of the BG, the CS-PVP/BG and the G-CS-PVP/BG, respectively. Characteristic peaks of BG (sodium calcium silicate [Na2Ca2Si3O9]) at 1090 cm− 1, 624 cm− 1, 529 cm− 1 and 470 cm−1 are dominant in all samples. After incorporation of CS-PVP with the BG, bands related to the presence of PVP and CS are observed both at the spectrum of CS-PVP/BG and that of G-CS-PVP/BG (Fig. 3A(b), (c)). Peaks due to absorption bands of PVP are found at 1650 cm− 1 (C_O stretching), 1420 cm− 1 (C–H bending vibration), and 1290 cm−1(C–N stretching) respectively while three characteristic peaks of CS are identified at 3450 cm− 1 (hydroxyl stretching), 1650 cm−1 (amide I) and 1560 cm− 1 (amide II), respectively. When the CS-PVP/BG was cross-linked with genipin, the peak associated with amide I shifts from 1654 to 1647 cm− 1 while the peak arising from amide II shifts from 1567 to 1559 cm− 1 (Fig. 3A(b) and (c)), confirming the reaction between amino groups of CS and ester groups of genipin [26], and formation of hydrogen bonding between CS and genipin [27,28]. Fig. 3B shows XRD patterns of various scaffolds. Compared with the standard PDF 75–1687, both the angular position and peak intensity of the sintered BG are seen to match the crystalline phase of sodium calcium silicate [Na2Ca2Si3O9] [29] (Fig. 3B(a)). The characteristic peaks of CS and PVP are not found in the spectrum of either the CS-PVP/BG scaffold or the G-CS-PVP/BG scaffold, but only the characteristic peaks of Na2Ca2Si3O9 are observed in Fig. 3B(b) and (c). The results demonstrate that no recrystallization of CS and PVP has occurred during the coating process. For better understanding the influence of extracts from the scaffolds on the biocompatibility, we have measured pH values of the extracts from different scaffolds. Before the extract preparation, pH values of the 45S5 BG, the CS-PVP/BG and the G-CS-PVP/BG were controlled by soaking the scaffolds in DI water over 3 days while DI water was replaced every 8 h. 100 mg/ml samples were subsequently immersed in the extracting media and incubated for 24 h in a cell incubator (37 °C and 5% CO2) to get the extracts. pH values of three extraction media are higher than 7 (Fig. 4) due to alkalinity of the 45S5 BG. The 45S5 BG extract shows the highest pH value while the G-CS-PVP/BG extract exhibits the lowest pH value. 3.3. In vitro enzymatic degradation Fig. 5 presents the weight loss ratios of the CS-PVP/BG compared with the G-CS-PVP/BG scaffold over 2 and 7 days' lysozyme incubation respectively. After 2 days, the weight loss ratio of the CS-PVP/BG scaffold (39.3%) is much higher than that of the G-CS-PVP/BG scaffold (32.6%) (P b 0.01). This may be attributed to degradation of water-soluble PVP and partial degradation of CS in PBS solution at 37 °C. After 7 days' incubation, the CS-PVP/BG scaffold shows a significantly increased weight loss (46.9%) compared to 35.3% weight loss of the G-CS-PVP/BG scaffold (35.3%, P b 0.05), suggesting significantly enhanced stability of the scaffold after cross-linking. It has been reported that significant degradation of CS in lysozyme is arisen from the binding of N-acetylglucosamine
476
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Fig. 2. SEM images of (a, b) the BG, (c, d) the CS-PVP/BG, (e, f) the G-CS-PVP/BG scaffold.
residues to the active sites of lysozyme [30]. The chemical cross-linking between CS and genipin, however, is expected to form a cyclic covalentcross-linked structure, which is beneficial for suppressing degradation of CS in lysozyme [21].
Fig. 3. (A) FT-IR spectra of (a) the BG scaffold, (b) the CS-PVP/BG scaffold, and (c) the G-CS-PVP/BG scaffold. (B) XRD patterns of (a) the BG, (b) the CS-PVP/BG and (c) the G-CS-PVP/BG. Characteristic peaks of Na2Ca2Si3O9 phase are marked by (▼).
3.4. Cytotoxicity Biocompatibility of the scaffold is essential for its application in tissue engineering. Cell viabilities of the as-synthesized scaffolds were thus evaluated using the CCK-8 assay. The pristine BG scaffold was used as the control. As shown in Fig. 6, the CS-PVP/BG scaffold shows maximum toxicity with MC3T3-E1 cells when compared with both the BG and the G-CS-PVP/BG. The cell viability at the G-CS-PVP/BG is
Fig. 4. pH values of the BG extract, the CS-PVP/BG extract and the G-CS-PVP/BG extract by soaking the scaffolds in the extracting media for 24 h (37 °C and 5% CO2).
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Fig. 5. Weight loss of the CS-PVP/BG scaffold and the G-CS-PVP/BG scaffold immersed in lysozyme solution for 2 and 7 days (*P b 0.05; **P b 0.01).
comparable to that at the control, suggesting excellent biocompatibility of the resulting G-CS-PVP/BG scaffold. 3.5. Fluorescence staining Live/dead cell staining results on the as-synthesized scaffolds are shown in Fig. 7. Cells are homogeneously distributed on the struts of all types of scaffolds after 4 days' culture. Few dead cells are observed at the G-CS-PVP/BG scaffold, suggesting the excellent biocompatibility of the scaffold. It is interesting to find out that cells can penetrate into all scaffolds, confirming that highly porous interconnected structures are attractive for cell attachment and adhesion. Live/dead cell staining was also used to count the live cells cultured with the three types of extracts after 1, 3 and 5 days respectively. As shown in Fig. S1, cell numbers obtained with the G-CS-PVP/BG extract (Fig. S1(c), (f), and (i)) are comparable to those observed with the 45S5 BG extract (Fig. S1(a), (d), and (g)) while significant reduction in cell numbers are identified with the CS-PVP/BG (Fig. S1(b), (e), (h)). Live/dead fluorescence staining results are well consistent with the CCK-8 results. 3.6. Cell skeleton and morphology Cytoskeletons of cells were stained green with Rhodamine Phalloidin while nuclei of cells were stained blue with Hoechst 33258. Results were identified using LSCM images which revealed the formation of the actin cytoskeleton, providing information of the structural framework and participates in cell migration [31]. There are bright
477
and well-pronounced elongated filopodia on the edges of cells on the all scaffolds after 1 day (Fig. 8(a), (c), and (e)). Almost all cells are grouped and exhibit elongated morphology on the day 4 (Fig. 8(b), (d), and (f)). The interconnection between surrounding cells is evident and the presence of some membrane channels between adjacent cells is apparent. Cells containing well-stretched F-actin bundles are observed obviously. These actin-rich structures known as the tunneling nanotubes, could mediate the intercellular transfer of organelles, plasma membrane components and cytoplasmic molecules [32,33]. Fluorescent cell skeletons were further studied on the cells cultured with the scaffold extracts. As shown in Fig. S2, more cells are found with the 45S5 BG extract and the G-CS-PVP/BG extract when compared to cells incubated with the CS-PVP/BG extract. Particularly, the cells with the G-CS-PVP/BG show much better stretched morphology. On the other hand, polymers (CS and PVP) dissolved in the culture medium might also influence the cell viability. The cross-linking of CS allows fewer ions (from 45S5 BG) and polymers (from CS and PVP) to be released into the extracting media, leading to similar cell viability obtained with the G-CS-PVP/BG extract and the 45S5 BG extract. SEM images of pre-osteoblasts MC3T3-E1 cells seeded scaffolds after 4 and 7 days are shown in Fig. 9. After 4 days' culture, cells on all scaffolds are fully attached on the surface and extend their cytoplasm in the form of thin and long fibrils from the leading edges. Cells cultured on the BG scaffold (Fig. 9(a)) and the G-CS-PVP/BG (Fig. 9(e)) scaffold present polygonal morphology and expand relatively better than those on the CS-PVP/BG scaffold (Fig. 9(c)), which may be attributed to the relatively high toxicity of the CS-PVP/BG scaffold compared to other two scaffolds as we found from the CCK-8 assay. After 7 days' incubation, cells cultured on the BG scaffold and the CS-PVP/BG scaffold are grouped and exhibit similar morphology. Well interconnection between cells is observed. SEM results confirm that biocompatibility of the G-CS-PVP/BG scaffold is improved by cross-linking. The resulting G-CS-PVP/BG scaffold exhibits superb properties for promoting the spreading and proliferation of MC3T3-E1 cells. 3.7. In vitro drug release For purpose of developing novel scaffolds for multifunctional applications, we further investigated in vitro drug release from the as-synthesized scaffolds. The release level of vancomycin from the as-prepared scaffold was determined based on the total amount of vancomycin incorporated in the corresponding scaffold. Cumulative release results in PBS solution are shown in Fig. 10. There are two release stages observed at the scaffolds without genipin. The release curve on the G-CS-PVP/BG scaffold, however, exhibits three stages. At the first stage, a rapid drug release is observed for all samples, which is due to the dissolution of the surface attached vancomycin. The G-CS-PVP/BG scaffold shows the lowest initial burst release. Secondarily, the release at the BG scaffold completes within 24 h, which is in consistent with our previous work [34]. After introduction of CS-PVP onto BG, the release is improved to 48 h until a plateau is reached. Further cross-linkage of genipin to the CS-PVP/BG provides additional advantages for controlled and stained release. The drug entrapped at the G-CS-PVP/BG scaffold is released in a controlled manner over a 7-day period, which may be attributed to the cross-linking of CS with genipin. 4. Discussion
Fig. 6. CCK-8 counts of cells cultured on: the BG, the CS-PVP/BG and the G-CS-PVP/BG scaffolds. Error bars represent means ± SD for n = 4 (*P b 0.05; **P b 0.01).
Due to the unique biocompatibility and poor mechanic property of the BG, numerous natural and synthetic polymers have been used to increase the stiffness of BG scaffolds [29,34–38]. Hydrophilic polymers usually suffer from rapid degradation while hydrophobic nature of hydrophobic polymers always hinders cells' adhesion and expansion. In this work, we used genipin as the cross-linking reagent to overcome degradation issues of the CS-PVP/BG scaffold. Introduction of genipin
478
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Fig. 7. Fluorescence staining of cells cultured on: (a) the BG, (b) the CS-PVP/BG and (c) the G-CS-PVP BG scaffolds for 4 days.
remains the highly porous interconnected structures which are attractive for cell seeding and proliferation. The CCK-8 results show that G-CS-PVP coating on the BG scaffold does not affect cell viability and proliferation, suggesting that presence of genipin causes no negative effect on mitochondrial functions but enhanced biocompatibility of the scaffold. These results are consistent with the cell morphology results which show that increased cell
morphology on the CS-PVP/45S5 BG scaffold after cross-linking. These results were further evident by cell skeleton observations. MC3T3-E1 cells cultured on the resulting scaffold exhibit a favorable F-actin distribution. Enhanced cell proliferation is obtained after genipin cross-linking with the CS-PVP/BG scaffold. This may be due to presence of hydroxyl groups at genipin which enhance the hydrophilicity of the scaffold and facilitate cells' adhesion, dispersion and proliferation.
Fig. 8. Fluorescence images of cell skeletons on: (a, b) the BG, (c, d) the CS-PVP/BG, and (e, f) the G-CS-PVP/BG scaffolds. (a, c, e) Cells were cultured for 1 day, and (b, d, f) cells were incubated for 3 days.
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
479
Fig. 9. SEM micrographs of cell morphology on: (a, b) the 45S5 BG scaffold, (c, d) the CS-PVP/45S5 BG scaffold and (e, f) the G-CS-PVP/45S5 BG scaffold.
In vitro drug release results suggest that vancomycin entrapped in the resulting G-CS-PVP/BG scaffold shows a better controlled stained release kinetics than the other scaffolds. This result could be attributed to two reasons. Firstly, both CS and PVP are hydrophilic polymers and the degradation rate is quick high in PBS solution at 37 °C. But genipin cross-linking decreases the degradation of polymers. Secondly, the
cross-linking degree of CS is low in the acidic solution (19.6%). It is thus harder for genipin to react with protonated amino groups in CS via nucleophilic attack [2]. 5. Conclusions Novel G-CS-PVP/BG scaffolds were prepared using genipin as the cross-linker of the chitosan coating. The physicochemical properties of the scaffolds were investigated. Introduction of G-CS-PVP onto the BG scaffolds has improved resistance to enzymatic degradation of the scaffolds while remained good biocompatibility and unique porous structures of the BG scaffolds for good adhesion and proliferation of MC3T3-E1 cells. Cells on the scaffolds show well-stretched F-actin bundles after 4 days' incubation. Furthermore, the antibiotic vancomycin was entrapped into the BG scaffolds via CS-PVP coating. Vancomycin in the G-CS-PVP/BG scaffolds shows significantly improved release period of 7 days from 24 h of the BG scaffold. Our preliminary results suggest that the as-prepared G-CS-PVP/BG scaffolds are promising for tissue engineering applications considering their excellent bioactivity and in-situ antibiotic releasing function. Acknowlegments
Fig. 10. In vitro vancomycin release behaviors from the BG scaffold, the CS-PVP/BG scaffold and the G-CS-PVP/BG scaffold.
Financial support for this work from the Chinese National Nature Science Foundation (21374081, 51433005), and Zhejiang National Nature Science Foundations (LQ15H180003, Y13H180013) are acknowledged.
480
Q. Yao et al. / Materials Science and Engineering C 56 (2015) 473–480
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.06.046. References [1] M. Yazdimamaghani, D. Vashaee, S. Assefa, K.J. Walker, S.V. Madihally, G.A. Köler, L. Tayebi, J. Biomed. Nanotechnol. 10 (2014) 911–931. [2] C. Vitale-Brovarone, E. Verne', L. Robiglio, P. Appendino, F. Bassi, G. Martinasso, G. Muzio, R. Canuto, Acta Biomater. 3 (2007) 199–208. [3] K.K. Mallick, J. Winnett, J. Am. Ceram. Soc. 95 (2012) 2680–2686. [4] S. Bose, M. Roy, A. Bandyopadhyay, Trends Biotechnol. 30 (2012) 546–554. [5] Y. Liu, K. Gilmore, J. Chen, V. Misoska, G.G. Wallace, Chem. Mater. 19 (2007) 2721–2723. [6] K.T. Shalumon, J.P. Chen, Curr. Pharm. Des. 21 (2015) 1979–1990. [7] M.N. Rahaman, D.E. Day, B.S. Bal, Q. Fu, S.B. Jung, L.F. Bonewald, A.P. Tomsia, Acta Biomater. 7 (2011) 2355–2373. [8] G.G. Niederauer, M.A. Slivka, N.C. Leatherbury, D.L. Korvick, H.H. Harroff, W.C. Ehler, C.J. Dunn, K. Kieswetter, Biomaterials 21 (2000) 2561–2574. [9] P. Ducheyne, Q. Qiu, Biomaterials 20 (1999) 2287–2303. [10] S. Abdollahi, A.C.C. Ma, M. Cerruti, Langmuir (2013) 1466–1474. [11] A. Mohammadkhah, L.M. Marquardt, S.E. Sakiyama-Elbert, D.E. Day, A.B. Harkins, Mater. Sci. Eng. C 49 (2015) 632–639. [12] C. Wu, Y. Ramaswamy, P. Boughton, H. Zreiqat, Acta Biomater. 4 (2008) 343–353. [13] W. Li, H. Wang, Y. Ding, E.C. Scheithauer, O.-M. Goudouri, A. Grünewald, R. Detsch, S. Agarwal, A.R. Boccaccini, J. Mater. Chem. B 3 (2015) 3367–3378. [14] A. Di Martino, M. Sittinger, M.V. Risbud, Biomaterials 26 (2005) 5983–5990. [15] N.M. Julkapli, H.M. Akil, Z. Ahmad, Compos. Interfaces 18 (2011) 449–507. [16] R.A.A. Muzzarelli, Carbohydr. Polym. 77 (2009) 1–9.
[17] D.Y. Zuo, Y. Wang, W.L. Xu, H.T. Liu, Adv. Polym. Technol. 31 (2012) 310–318. [18] D. Yu, X. Wang, X. Li, E. Chian, Y. Li, Y. Liao, Acta Biomater. 9 (2013) 5665–5672. [19] M.F. Butler, Y.F. Ng, P.D.A. Pudney, J. Polym. Sci., Polym. Chem. Ed. 41 (2003) 3941–3953. [20] M. Mekhail, K. Jahan, M. Tabrizian, Carbohydr. Polym. 108 (2014) 91–98. [21] F.L. Mi, H.W. Sung, S.S. Shyu, C.C. Su, C.K. Peng, Polymer 44 (2003) 6521–6530. [22] T.H. Tsai, J. Westly, T.F. Lee, C.F. Chen, J. Liq. Chromatogr. 17 (1994) 2199–2205. [23] L. Bi, Z. Cao, Y. Hu, Y. Song, L. Yu, B. Yang, J. Mu, Z. Huang, Y. Han, J. Mater. Sci. Mater. Med. 22 (2011) 51–62. [24] Q. Chen, I.D. Thompson, A.R. Boccaccini, Biomaterials 27 (2006) 2414–2425. [25] V. Karageorgiou, D. Kaplan, Biomaterials 26 (2005) 5474–5491. [26] F.L. Mi, Biomacromolecules 6 (2005) 975–987. [27] F.L. Mi, H.W. Sung, S.S. Shyu, J. Appl. Polym. Sci. 81 (2014) 1700–1711. [28] C.D. Ji, N. Annabi, A. Khademhosseini, F. Dehghani, Acta Biomater. 7 (2011) 1653–1664. [29] Q. Yao, P. Nooeaid, R. Detsch, J.A. Roether, Y.M. Dong, O.M. Goudouri, D.W. Schubert, A.R. Boccaccini, J. Biomed. Mater. Res. A 102 (2014) 4510–4518. [30] B.S. Liu, T.B. Huang, Macromol. Biosci. 8 (2008) 932–941. [31] T. Jiang, Z. Zhang, Y. Zhou, Y. Liu, Z. Wang, H. Tong, X. Shen, Y. Wang, Biomacromolecules 11 (2010) 1254–1260. [32] A. Rustom, R. Saffrich, I. Markovic, P. Walther, H.H. Gerdes, Science 303 (2004) 1007–1010. [33] H.H. Gerdes, R.N. Carvalho, Curr. Opin. Cell Biol. 20 (2008) 470–475. [34] Q. Yao, P. Nooeaid, J.A. Roether, Y. Dong, Q. Zhang, A.R. Boccaccini, Ceram. Int. 39 (2013) 7517–7522. [35] W. Li, M.-I. Pastrama, Y. Ding, K. Zheng, C. Hellmich, A.R. Boccaccini, J. Mech. Behav. Biomed. Mater. 40 (2014) 85–94. [37] M. Montazeri, S. Karbasi, M.R. Foroughi, A. Monshi, R. Ebrahimi-Kahrizsangi, J. Mater. Sci. Mater. Med. 26 (2015) 62. [38] G. Yang, X. Yang, L. Zhang, M. Lin, X. Sun, X. Chen, Z. Gou, Mater. Lett. 75 (2012) 80–83.