Materials Science and Engineering C 54 (2015) 50–60

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

The effect of coating type on mechanical properties and controlled drug release of PCL/zein coated 45S5 bioactive glass scaffolds for bone tissue engineering Zeinab Fereshteh a,b,c,⁎, Patcharakamon Nooeaid a, Mohammadhossein Fathi c,d, Akbar Bagri e, Aldo R. Boccaccini a,⁎⁎ a

Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, Cauerstrasse 6, 91058 Erlangen, Germany Institute of Science, High Technology and Environmental Sciences, Graduate University of Advanced Technology, 76315117 Kerman, Iran c Biomaterials Research Group, Department of Materials Engineering, Isfahan University of Technology, Isfahan 8415683111, Iran d Dental Materials Research Center, Isfahan University of Medical Sciences, Isfahan, Iran e Department of Materials Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA b

a r t i c l e

i n f o

Article history: Received 22 July 2014 Received in revised form 7 March 2015 Accepted 2 May 2015 Available online 5 May 2015

1. Introduction A highly investigated branch of tissue engineering involves the implantation of a highly porous biomaterial scaffold populated with appropriate cells to provide a three-dimensional (3D) environment in which new tissue can grow [1,2]. After tissue repair, the ideal biomaterial will be resorbed and removed by the body itself or integrated within the new tissue without the need for further surgery [1–3]. Scaffolds are engineered materials with high porosity which should perform as 3D templates for cell adhesion, proliferation, migration and ultimately the formation of new tissue [3,4]. One of the scientific challenges faced by tissue engineers is associated with the complex combination of properties required in an ideal scaffold [2,3]. 45S5 bioactive glass (BG) is a promising material for application in bone tissue engineering as it possesses osteoconductive and bioactive properties as well as angiogenic effects [5–9]. In 2006, the foam replica technique was introduced to produce 45S5 BG scaffolds [10]. Highly porous foam-like BG-based scaffolds have been developed which exhibit macroporous structure very similar to that of spongy bone; with ~90% completely open trabecular porosity [10]. Also irregular shapes can be created to match the morphology and size of the bone imperfection. Moreover, the foam replication method

⁎ Correspondence to: Z. Fereshteh, Institute of Biomaterials, Department of Materials Science and Engineering, University of Erlangen-Nuremberg, Cauerstrasse 6, 91058 Erlangen, Germany. ⁎⁎ Corresponding author. E-mail addresses: [email protected] (Z. Fereshteh), [email protected] (A.R. Boccaccini).

http://dx.doi.org/10.1016/j.msec.2015.05.011 0928-4931/© 2015 Elsevier B.V. All rights reserved.

does not include the use of toxic chemicals being more rapid and cost effective compared to other standard processing techniques [3,10]. As a drawback, the mechanical properties of foam-like BG-based scaffolds are in general poor in terms of strength and toughness for bone tissue engineering applications. Polymer coatings have been proposed to toughen brittle scaffolds, which can be achieved by the formation of a surface coating and by impregnation of microcracks and surface pores with a suitable polymer [3,11,12]. One of the most studied biodegradable polymers for this purpose is poly(ε-caprolactone) (PCL). This polymer is biocompatible and possesses an appropriate rate of biodegradation, degrading hydrolytically without leaving toxic residues [13, 14]. It has been also reported that polymer coatings can act as carrier for drugs and other biomolecules [15]. Consequently, the functionality and bioactivity of scaffolds can be notably enhanced by the coating approach [3,11,15,16]. Zein, a corn protein, has received attention in recent years regarding its pharmaceutical application, where it has been successfully used in tableting, films, active food packaging materials, encapsulation of essential oils, aromas and flavors [17], controlled release of active additives or drug [18]. This vegetable protein has been extensively used in pharmaceuticals as an encapsulation material and as a controlled-release delivery material for drugs and vaccines [19,20] and in tissue engineering [18]. It was first identified based on its solubility in aqueous alcohol solutions. The high fraction of non-polar amino acid residues is responsible for its solubility characteristic. The molecular structure of zein is a helical wheel with nine homologous repeating units arranged in an anti-parallel form stabilized by hydrogen bonds and disulfide bonds [21]. The average hydrophobicity of zein is reported to be 50 times larger than albumin or fibrinogen, among others. Also its regular geometry allows self-assembling into chains, layers or films [22]. The aim of the present work was to fabricate highly porous (90% porosity) bioactive 45S5 BG-based scaffolds and introduce a dual PCL/zein coating on the scaffold surface in order to achieve improved mechanical properties and impart a local drug release function using tetracycline as a model drug. The possibility to increase the mechanical properties of the BG scaffolds by coating with PCL was investigated. In addition, we hypothesized that the drug release capability can be improved by

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

coating the struts with a thin PCL/zein layer. The effects of the PCL and PCL/zein concentration on the mechanical strength and porosity of the scaffolds were analyzed. Moreover, the ability of PCL, zein and PCL/ zein coatings to release tetracycline hydrochloride (TCH) in a sustained manner was verified by immersion studies in phosphate buffered saline (PBS).

zein

2.1. Fabrication of bioglass-based scaffolds Fully reticulated polyurethane (PU) foams (45 ppi, Eurofoam, Germany) and 45S5 BG powder (composition in wt.%: 45% SiO2, 24.5% Na2O, 24.4% CaO and 6% P2O5) of particle size ∼ 5 μm have been used to fabricate the scaffolds. The scaffolds were fabricated by the foam replication method, following a similar process as described elsewhere [10]. Briefly, polyvinyl alcohol (PVA) (MW ∼ 30,000, Merck, Germany) was dissolved in deionized water at 80 °C to produce a polymer weight to solvent volume ratio of 6% (w/v), and then 45S5 BG powder was added to the PVA solution up to a concentration of 50 wt.%. The mixture was then stirred 2 h to obtain a homogeneous solution. PU foams were immersed in the slurry and rotated to ensure homogeneous slurry infiltration. Then, the scaffolds were extracted from the slurry, and the extra slurry was completely squeezed out. The samples were dried overnight at room temperature, and then the procedure described above was repeated again to obtain a homogeneous BG cover over PU foams. The samples were finally heated at 400 °C (2 °C/min) for 1 h in air to decompose the PU foam, and then at 1100 °C (2 °C/min) for 2 h to densify the bioactive glass network. 2.2. Polymer coating procedure and drug loading Poly (ε-caprolactone) solutions with 1, 2, 3 and 5% (w/v) (PCL, Mw = 80,000, Sigma-Aldrich) were prepared by dissolving PCL in chloroform (Merck, Germany). The as sintered scaffolds of dimensions 10 mm × 5 mm × 5 mm were then completely immersed in the PCL solution (20 ml) for 2.5 min. For homogeneous coatings, the container was manually shaken during the coating procedure. After that, the scaffolds were taken out and dried in air at room temperature, and then the process of polymer coating was repeated for each BG scaffold two times to obtain a homogeneous polymer coating. In order to obtain a dual PCL/zein coating, PCL/zein (Sigma-Aldrich, USA) (50:50% w/w) solution was prepared by dissolving PCL in chloroform and zein in ethanol (Merck, Germany). Then, scaffolds were coated by this solution for two times. For the drug loading together with the PCL/zein polymer, TCH (0.5% w/v) was dissolved in ethanol solution and the coating procedure was performed similarly as for the unloaded coatings. 2.3. Characterization Porosity is one of the parameters that should be measured in characterization of the scaffolds. The porosity of the samples was measured according to our previous work. The porosity of scaffolds (Pi) before and after preparing the coating was calculated using Eqs. (1) and (2) or (3), respectively:
1−

W1 ρBG V 1

  100

ð1Þ

2

% P PCL or zein

3 W 1 ðW 2 −W 1 Þ þ 6 ρBG ρPCL or zein 7 7  100 ¼6 41− 5 V2

zein

ρPCL ¼

2. Experimental procedure

% P BG ¼

2

% P PCL

ð2Þ

3 W 1 ðW 2 −W 1 Þ þ 6 7 ρBG ρPCL 6 7 zein ¼ 61− 7  100 4 5 V2

51

ðρPCL þ ρzein Þ 2

ð3Þ

ð4Þ

where W and V are the weight and volume of the scaffold, and the subscripts 1 and 2 refer to the values before and after coating, respectively. Also, ρBG (2.7 g/cm3) was considered to be the theoretical density of 45S5 bioactive glass [23] (although this value could differ slightly from the value of the crystallized matrix), ρPCL = 1.145 g/cm3 and ρzein = 1.22 g/cm3, respectively. The macroscopic cross-section of the scaffolds, which were cut using a razor blade, was observed under a light microscope (LEICA M50) with attached camera (LEICA IC80 HD). Samples were then sputter coated and observed at an accelerating voltage of 10 kV using a LEO 435 VP scanning electron microscope (SEM). The pore size distribution of scaffolds was quantitatively measured using SEM images by ImageJ software. The Fourier transform infrared (FTIR) spectra of different scaffolds were recorded on a Nicolet 6700 FTIR spectrometer in transmittance mode in the mid infrared region (4000–400 cm−1). Selected scaffolds were also characterized using XRD (Philips diffractometer, 40 kV, Cu Kα) analysis. Data were collected over the 2θ range from 5° to 80° using a step size of 0.05°. Energy dispersive X-ray spectroscopy (EDS) associated with the scanning electron microscopy (SERON AIS2300C) was utilized to obtain elemental analyses. The compressive strength of uncoated and coated scaffolds was measured using a Zwick/RoellZ050 mechanical tester equipped with a 50 N loading cell at across head speed of 0.5 mm/min. The cylindrical samples were 10 mm in diameter and 10 mm in height. During compressive strength test, the load was applied until the strain reached 80%. The compressive strength was determined from the maximum load of the obtained stress–strain curve. Five samples were tested for each condition and results were expressed as the mean values and standard deviations. The bioactivity of scaffolds was investigated by immersion in Kokubo's SBF [24] for up to 28 days. The scaffolds were immersed in 50 ml of SBF and maintained at 37 °C, 90 rpm in a shaking incubator (KS 4000 i control, IKA, Germany). The SBF solution was replaced twice a week. Samples were collected after 1, 3, 7, 14, and 28 days of incubation (5 samples for each date), respectively. At each time point, the samples were removed, cleaned with deionized water, and dried at room temperature in a desiccator for further examination. According to ASTMF 1635-95, in vitro degradation assay was used to study the weight loss of the PCL/zein composite films within 4 weeks of soaking in phosphate buffered saline (PBS) (pH 7.4, Sigma-Aldrich, USA). The PBS solution was changed every 3 days and after each specific time point (1, 7, 14 and 28 days), pH of the PBS solution was determined by using a pH meter (Metrohm, Germany). The films were rinsed in PBS, freeze-dried overnight and the degradation percentage of each sample was calculated by dividing the weight loss to the initial dry weight. The results were expressed as average value and standard deviations (5 samples for each date). The weight loss was calculated according to: Weight loss ð%Þ ¼

W 0− W t  100 W0

ð5Þ

where W0 is the initial dry weight and Wt is the dry sample weight after removal. Static contact angle measurements were carried out on film samples using a DSA30 contact angle measuring instrument (Kruess, Germany). The films were prepared by casting the above PCL-zein solution into glass Petri dish. Water (3 μl) was added by a motor-driven syringe at

52

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

room temperature. Reported data were obtained by averaging the results of five measurements. In order to determine the released TCH concentration in PBS, the uncoated and PCL, PCL/zein, and zein coated scaffolds were immersed in glass vials containing 10 ml of the phosphate buffered saline solution (pH 7.4, Sigma-Aldrich, USA) and placed in a shaking incubator (90 rpm, 37 °C). At pre-determined time intervals aliquots of 5 ml were withdrawn and replenished by an equal volume of fresh PBS solution to maintain a constant volume. The amount of released TCH was measured using the UV–vis spectrophotometer (Specord 40, Analytik Jena, Germany) at a wavelength of 358 nm, which is λmax of TCH. The tetracycline concentration calibration curve was established using known tetracycline concentrations in the PBS solution. Amounts of the released drugs at specified time periods were plotted as percentage of drug released versus time. The drug release evaluation was performed using 5 samples for each time up to 14 days and obtained results were expressed as the mean values and standard deviations. 3. Results and discussion 3.1. Development and characterization of coated scaffolds By coating the BG scaffolds with PCL or PCL/zein blend the mechanical properties of the scaffolds were substantially improved, i.e. the compressive strength increased from 0.004 ± 0.001 MPa (uncoated BG scaffolds) to 0.15 ± 0.02 MPa (PCL/zein coated BG scaffolds). The compressive strength values of the coated scaffolds are close to the lower bound strength value for spongy bone. The compressive strength of superficial of osteochondral tissue is 0.079 MPa [25-27]. Fig. 1 shows typical compressive stress–strain curves for uncoated and coated scaffolds with different concentrations of PCL and PCL/zein indicating the increase of both the compressive strength and the area under the stress displacement curve, which is related to the work of fracture (toughness of the material). By increasing the concentration of polymer, the compressive strength of coated scaffolds was increased. During compressive strength test, it was observed that uncoated scaffolds and zein only

coated scaffolds completely crumbled into powder while the PCL and PCL/zein coated scaffolds partly retained the shape and did not collapse. It is also evident that the curve for the scaffold coated by a high polymer concentration is much less jagged, and the shape resembles the ideal curve for highly porous foams [25] (compare compressive stress–strain curves of scaffolds coated by 3% w/v PCL and 5% w/v PCL shown in Fig. 1b and c). By comparing Fig. 1b and c, it is apparent that the jagging of compressive stress–strain curves for scaffolds coated by 2.5% w/v% PCL/2.5% w/v% zein is lower than that of the other samples. As discussed in previous studies [3,12,16,28], it is likely that the polymer layer covers the strut and fill microcracks on the strut surface. After polymer coating, the mechanical stability of the flaw sensitive and brittle material is improved and the original weak and fragile struts become stronger. Also, the jagging in the stress–strain curve for the uncoated and the zein coated samples is due to microcracks in the struts. Because the stress–strain curves for the coated samples are smoother (without the jagging), it can be stated that the coating process enabled PCL to fill the cracks, and crack bridging by the polymer could be the active mechanism improving the mechanical stability of the scaffold. On the other hand, the coating process did not improve the compressive strength in the case of pure zein coated scaffolds because the pure zein film might be too brittle and rigid [29,30]. This issue is also confirmed by SEM images of PCL, PCL/ zein and zein-coated scaffolds (discussed below), which showed a smooth surface of the struts and no large microcracks present on scaffolds coated by PCL. In a recent study on similar polymer coated BG scaffolds, Hum et al. [30] concluded that PCL can infiltrate the cracks present in the struts increasing the stiffness of the overall structure. It was shown that PCL significantly increased the stiffness of the scaffolds without sacrificing the intrinsic bioactive properties of BG [30]. It should be noted that the compressive strength achieved with the present scaffolds was sufficient for safe handling. Digital photographs of the top surface and cross section of scaffolds are shown in Fig. 2. It should be noted that all coated scaffolds exhibit homogeneous surfaces. There were no distinct domains or particulate matter on the surface of samples, which could have resulted from poor mixing of PCL and zein during coating. It was also confirmed by

Fig. 1. Effect of different concentrations of a) PCL, b) PCL/zein composite coating with 3 wt.% and c) with 5 wt.%, on the stress–strain behavior of BG-based scaffolds in compression. d) Stress–strain curve of PCL/zein coated BG-based scaffolds with 5 wt.% of PCL/zein under compressive loads before and after immersion in SBF for 28 days.

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

53

Fig. 2. Digital photographs of coated scaffolds: (a) top surface and (b) cross section of the scaffold coated by PCL (5% w/v), (c) top surface and (d) cross section of the scaffold coated by PCL/ zein (2.5%:2.5% w/v) and (e) top surface and (f) cross section of the scaffold coated by zein (5% w/v).

visual inspection that no significant blocking of pores occurred by the coating process. Fig. 3 shows the porosity of scaffolds coated by various polymer concentrations. All coated scaffolds have high porosity, N 91%, and up to 97%. The porosity of the coated scaffolds is seen to (almost) linearly increase by decreasing the polymer–zein concentration. With increasing polymer concentration, the actual volume fraction occupied by the material itself increased. According to SEM images shown in Fig. 4, it can be estimated that the thickness of the polymer coating is in the range of

2–5 μm. As a consequence of the coating, the porosity decreased to ~91%, which is still within the desired range for bone tissue engineering applications [5–7]. Based on this result, the scaffold coated using a solution of 2.5% PCL:2.5% zein w/v (1:1) was chosen for further studies. The porosity and the compressive strength of the present coated scaffolds were ∼94% and 0.15 MPa, respectively. Table 1 shows the water contact angle values of uncoated, coated scaffolds and films. The contact angle increased in the presence of

Fig. 3. Porosity of composite scaffolds coated by different concentrations of PCL, zein and PCL/zein.

54

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

Fig. 4. SEM micrographs of the surface morphology of: (a, b) the uncoated BG scaffold, the BG scaffolds coated with (c, d) PCL 5% w/v, (e, f) PCL/zein (2.5%:2.5% w/v) and (g, h) zein 5% w/v at different magnifications.

polymer coating. However, the values were still lower than that of the pure PCL film. The inclusion of zein increased the sample wettability which is in agreement with the more hydrophilic nature of zein

Table 1 Mean value and standard deviation of contact angles measured on uncoated BG scaffold, coated BG scaffold, and pure films. Sample

BG scaffold (°)

Pure film (°)

BG-pure Zein coated PCL/zein coated PCL coated

15 ± 2 23 ± 4 42 ± 2 58 ± 2

– 36 ± 2 56 ± 2 89 ± 2

compared to PCL [18]. Surface hydrophilicity significantly affects the biological performance of materials, having an impact on protein adsorption, cell attachment, migration and spreading [19,28,31]. It is anticipated that increasing hydrophilicity of the surface of the scaffolds will be beneficial to effective cell–material interaction, cell adhesion, growth, and ultimately for superior performance of the scaffold in bone tissue engineering applications [32]. The microstructures of the uncoated and coated scaffolds are illustrated in Fig. 4. The replica method allows obtaining scaffolds with an open and interconnected porous structure [10]. The polymer concentration in the coating solution plays an important role in determining the thickness of the coatings. It is observed that the struts of the scaffolds are to some extent homogeneously coated with PCL and PCL/zein. It

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

was confirmed that by increasing the polymer concentration from zero (coated with pure zein) to 5% w/v (coated with pure PCL), the thickness of the coatings is increased. The PCL film on the scaffold is seen to lead to some pore closure due to the high concentration of PCL (Fig. 4a and b). In this case, the zein layer on the outer surface of the bioactive glass scaffold indicates that there was no good adhesion at the interface. From Fig. 4g and h, it can be seen that there is a thin gap between the surface of the BG scaffold and the zein layer. After coating with PCL/zein, the open pore structure was maintained, as confirmed by SEM images (Fig. 4e and f). In the case of PCL/zein coating, only a few pores were blocked as porosity of the coated scaffolds was ∼94%, and the average weight percentage increase of PCL/zein coated scaffolds was 15%. The PCL/zein coating was very uniform on the outer surface of the scaffolds (as can be seen in Fig. 2) which shows that there was a suitable interaction and adhesion between the PCL/zein coating and the BG surface without any additional surface treatment. As assessed by SEM images the pore size was in the range of 200–450 μm, which is appropriate for bone tissue engineering applications [11]. Fig. 4e and f shows that the BG scaffold is partially coated with polymer. This is not considered a disadvantage because if the surface would be totally covered by the polymer, the nucleation of bone-like apatite on the scaffold surfaces, which is the indication of bioactive behavior (see next section), would not be possible or it would be retarded [13]. 3.2. Assessment of bioactivity in simulated body fluid (SBF) The surface reactivity of the PCL/zein coated BG scaffold was examined in SBF as a qualitative indication of their in vitro bioactivity [24]. The goal was to study the formation of bone-like apatite or amorphous calcium phosphate on the surface of coated BG scaffolds during immersion in the SBF, which is a typical feature in BG derived scaffolds [10,16, 28] to ascertain if the bioactivity was retarded by the PCL/zein coating. Fig. 1d shows the mechanical behavior of the PCL/zein coated scaffolds before and after immersion in SBF for 28 days. The compressive strength of the scaffolds decreased markedly during 28 days of immersion in SBF (0.094 ± 0.004 MPa). The presence of bone-like apatite on the uncoated and coated scaffolds was confirmed by XRD analysis. Fig. 5a presents the XRD patterns obtained for uncoated and coated scaffolds before and after immersion in SBF for 2 weeks. As can be seen, the two patterns are analogous, probably due to the very thin polymeric covering layer deposited. The main crystalline phase is identified as Na2Ca2Si3O9, and the present BG-based scaffolds are glass–ceramic materials, which agree with previous results [10,16]. Characteristic hydroxyapatite (HA) peaks can be seen on both uncoated and coated scaffolds after immersion in SBF. Hence, the bioactivity of the scaffolds was sustained in the PCl/zein coated scaffolds. A search-match data base analysis confirmed that the formed apatite on the surface of the uncoated and coated scaffolds after immersion in

55

SBF has the phase composition corresponding to the JCPDS card no. 03-0747. The XRD patterns of the PCL, zein and PCL/ zein films are shown in Fig. 5b. The neat PCL scaffold pattern shows the PCL characteristic peaks (2Teta = 21° and 24°) [14,34]. Zein is naturally amorphous, and the diffraction pattern of zein exhibits a diffused background pattern with two diffraction halos (~10° and 20° 2θ) [35]. The XRD pattern of the composite film exhibits the peaks corresponding to both materials. In Fig. 5a (PCL/zein coated BG before immersion in SBF), the PCL characteristic peaks are shown by red star around 2Teta = 21° and 24°, and can also be found in PCL/zein coated BG after 14 days immersion in SBF. Fig. 6 represents SEM images of the uncoated scaffolds and the PCL/ zein coated scaffolds after immersion in SBF for 14 days. On these images, the crystalline apatite on the surface of the uncoated and coated scaffolds was clearly observed by SEM after immersion in SBF for 14 days. “Bone-like” apatite crystals can be recognized by their globular, cauliflower shape (see insets in Fig. 6e). This morphology is typical of carbonated hydroxyapatite (HCA), which has been reported previously on bioactive glass–polymer composite scaffolds after incubating in SBF [36]. The HA layer on both uncoated and PCL/zein coated scaffolds is seen to have grown fairly homogeneously throughout the struts. Li et al. [36] have investigated bioactivity of BG based scaffolds similar to those studied here, before and after applying a polymer coating on the scaffold struts. They reported the presence of areas of coated scaffolds which are exposed to SBF providing paths for SBF to penetrate into the interface gap between the BG surface and the coating, thus establishing direct contact with the bioactive glass surface. On the other hand, after 3 days of immersion, bone-like apatite precipitates were seen to unevenly cover the strut surfaces. Zhou et al. [37] modified the surface of mesoporous bioglass (MBG) scaffolds by hydroxypropyltrimethyl ammonium chloride chitosan (HACC) and zein which was shown to facilitate the adhesion of hydroxyapatite and the loading of drugs on the surface of MBG. The presence of HA crystals was observed with more clarity on the surface of modified scaffolds in comparison to non-modified scaffolds. The energy dispersive X-ray spectroscopy (EDX) technique was used to determine the elemental composition of the sample surfaces. Fig. 7 shows the EDX spectra of the uncoated and coated scaffolds before and after immersion in SBF during 14 days. The EDX spectra confirmed the characteristic elements of the bioactive glass composition, such as silicon, calcium, phosphorus, sodium and oxygen. Moreover, due to the polymer coating, the EDX analysis of the PCL/zein coated scaffold shows carbon. By comparing scaffolds before and after immersion in SBF, results from EDX analysis show a significant alteration of the surface composition. As it can be observed, in the uncoated BG scaffolds, the Ca/P ratio is higher than in coated scaffolds. After immersion of the scaffolds in SBF for 14 days, the peaks of Ca and P raised more than the Si peaks. This result is due to the crystals of HA formed on

Fig. 5. XRD patterns for a) uncoated and PCL/zein coated BG scaffolds before and after immersion in SBF for 2 weeks, showing HA (JCPDS file No: 03-0747) and Na2Ca2Si3O9 phase peaks (indexed) from sintered 45S5 BG and b) zein, PCL and PCL/zein films.

56

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

Fig. 6. SEM micrographs of (a, b) uncoated BG scaffolds and (c, d) PCL/zein coated BG scaffolds after immersion in SBF for 14 days. The insets in (b) and (d) indicate the globular and cauliflower shape of hydroxyapatite crystals.

the surface of the scaffolds. The carbon peak can be seen in the spectrum of the immersed uncoated BG scaffolds (Fig. 7b) which is suggested to be from the formation of HCA.

In order to confirm these observations, FTIR analyses of the SBF immersed scaffolds were carried out. The FTIR spectra of both uncoated and PCL/zein coated scaffolds before and after immersion in SBF for 1,

Fig. 7. The EDX spectra on BG scaffolds: a) uncoated and c) PCL/zein coated before immersion in SBF; b) uncoated and d) PCL/zein coated after immersion in SBF for 14 days.

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

7 and 14 days are presented in Fig. 8. Characteristic bands of PCL were observed at 2918 cm−1 (asymmetric CH2 stretching), 2845 cm−1 (symmetric CH2 stretching), 1722 cm−1 (carbonyl stretching), 1290 cm−1 (C–O and C–C stretching), 1238 cm−1 (asymmetric C–O–C stretching) and 1159 and 1173 cm−1 (symmetric C–O–C stretching) [38]. As can be seen in Fig. 8b, the characteristic absorptions bands of zein in asfabricated scaffolds consist of amide A from 3600 to 3100 cm− 1, amide I from 1750 to 1600 cm−1, and amide II from 1500 to 1400 cm−1. The peak at ∼ 3200 to 3500 cm− 1, according to N–H stretching, also indicates the presence of amine group [39]. The FTIR spectra of the PCL/zein coated scaffolds showed similar peaks corresponding to C–O and P–O bonds of hydroxyapatite. All FTIR spectra of the uncoated scaffolds after 7 and 14 days of immersion in SBF presented the characteristic peaks of biological hydroxyapatite, namely the peak at 1051 cm−1 that can be attributed to the stretching vibration of P–O bonds and the small peaks at 1427 and 868 cm− 1 that correspond to the stretching and bending vibrational modes of C–O, respectively, suggesting that the formed HA was carbonated hydroxyapatite (HCA) [40,41]. The peak at 550–620 cm−1 indicates the formation of the characteristic double peak of hydroxyapatite, which corresponds to the bending vibration of P–O bonds and is formed at the spectral area of 550–615 cm−1 [40]. The FTIR characterization thus confirmed that both uncoated and coated scaffolds exhibited formation of HCA. The increasing amount of HCA crystals formed on coated scaffolds was clearly detected by FTIR (after immersion for 1 day and longer). Therefore, zein is confirmed to have slightly enhanced the formation of HCA on the surface of the coated scaffolds. As can be seen, coating is one of the important factors for improving the mechanical adherence and bioactivity of the scaffolds [3]. Salerno et al. [18] demonstrated that the presence of zein enhances the hydrophilicity and accelerates the degradation of PCL. They reported that PCL/zein–HA composites are interesting biomaterials for bone tissue engineering [42]. These results suggest that the PCL/zein coating will positively affect the bioactive character of 45S5 BG scaffolds. Indeed a cell biology assessment of the present scaffolds, in comparison to similar polymer coated BG scaffolds reported recently [13,16,36] remains to be carried out. 3.3. Biodegradability assessment of PCL/zein coating Effectiveness of formulation on the weight evolution during degradation in PBS at 37 °C is reported in Fig. 9. In vitro degradation study was carried out to investigate the effect of zein content on the degradability of the PCL/zein coating. Fig. 9 shows the weight loss of the films after immersion in PBS for different times. An insignificant weight loss was observed for the PCL coating, which lost about 1.3% of its initial weight after immersion in PBS for 28 days. However, the PCL/zein coating presented noticeable degradation during the 28-day immersion in

57

PBS. Furthermore, the weight loss of the PCL/zein coating increased by increasing the amount of zein. The weight loss for the PCL/zein (50:50) and zein films was 16% and 39%, respectively. For the case of pure zein film, it is estimated that a majority of the zein component degraded after immersion in PBS for 28 days. Fig. 9 also illustrates the pH variations of PBS during the different times of immersion. For the case of pure PCL film, no significant pH variation was observed during the whole immersion period. However, the pH variation for the pure zein film decreased after 28 days of immersion. Therefore, the only slight decrease of pH value for the PCL/zein coating might be attributed to the degradation of zein. The weight loss results suggest that a majority of the zein component degraded after 28 days of immersion, while the hydrolysis of PCL was not recognizable, as reported elsewhere. Jaiswal et al. showed that because of hydrophobicity of poly-L-lactide (PLLA), the mass loss of the pure PLLA scaffold occurred slowly [43]. They prepared PLLA/Gelatin and PLLA/Gelatin/HA mineralized composites scaffold by utilizing the advantages of alternate soaking method. The addition of gelatin accelerated the degradation of PLLA/Gel scaffold due to the increase of hydrophilicity of the scaffold. Salerno et al. demonstrated that the presence of zein enhances hydrophilicity and accelerates degradation rate of PCL. The weight loss of PCL/zein (60/40) scaffold increased up to 12.1% at 14 days, and remained almost unchanged during the following 6 weeks [18]. They found that the composition of the systems plays an important role in regulating the cell behavior. When molten PCL was blended with zein and HA, PCL formed multiphase systems wettability and degradation enhanced. The PCL/ zein–HA composite proved the most interesting biomaterial for bone tissue engineering application purpose [44]. According to the present results, tailoring of zein concentration can control in vitro degradation behavior of polymer composite coatings. 3.4. Drug release study In order to establish the applicability of the present scaffolds as controlled drug release devices, their drug release behavior must be determined. The uncoated and the PCL, PCL/zein and zein coated scaffolds were loaded with TCH. The TCH was selected as a model drug to study the release kinetics. However, tetracycline is an attractive choice for bone tissue engineering as a marker of bone growth for biopsies in humans. In addition it has a broad-spectrum antimicrobial effect [45]. TCH has a UV absorbance peak around 358 nm. Hence, the amount of TCH released from the scaffolds was determined by UV–visible spectroscopy. The cumulative percentage of drug release was normalized to the total amount of TCH. The cumulative percentage of drug released from the uncoated and the PCL, PCL/zein and zein coated scaffolds in PBS for 14 days is shown in Fig. 10. A continuous release pattern along the study period is observed for all scaffolds. However, there was a

Fig. 8. FTIR spectra of (a) uncoated BG scaffolds and (b) PCL/zein coated BG scaffolds before and after immersion in SBF for 1, 7 and 14 days. The characteristic peaks are discussed in the text.

58

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

Fig. 9. Weight loss of the different composite films and pH trends of PBS as function of soaking time. Pure PCL, PCL/zein (50:50) and pure zein films.

remarkable difference among the uncoated and the different coated scaffolds depending on the composition of the coating. The rapid release of drugs directly adsorbed on the uncoated scaffold was observed, and the drug was completely released in just 48 h. The absorbed drug within the coated scaffolds was released in a sustained manner over 10 days. Three trends were observed for the drug release profiles. In the first type, the scaffolds showed rapid drug release during immersion in PBS. This was the case for the uncoated scaffold (94% release during 48 h). The second trend was related to a very slow release rate. In this case, the release rate decreased and the zein coated BG scaffolds demonstrated the longest release period compared to the other scaffolds. It can be seen in Fig. 10 that ~10% of TCH remained in the zein coated BG scaffolds after 14 days. In the third curve, the rate of the drug release was intermediate in comparison to the previous cases. It was confirmed that the drug release rate from the PCL/zein coated BG scaffolds was lower than that of the PCL coated scaffolds. Consequently, the PCL/zein coated BG scaffolds presented a released rate in a more controlled manner compared to the other scaffolds, which could provide a drug release to suitable induce initial antibacterial effects and a further sustained release to retain an anti-interface formation during long term healing. The PCL/zein coated scaffolds showed a much lower initial burst release of b1%. The data confirmed that the release of TCH occurred in a controlled manner over a period of 14 days from the PCL/zein coated scaffolds (99.5%). The PCL and the zein coated scaffolds exhibited 98% and 89.5% release, respectively. To shed some light on the release rate of TCH from PCL/zein coated scaffolds a logarithmic function is fit to the experimental data and the coefficient of determination (R-squared) is

Fig. 10. Released TCH from BG scaffolds uncoated and coated with PCL, PCL/zein and zein TCH-loaded. The inserted curve indicates more details of the release curve in the first 50 h.

calculated. The R-squared parameter is always between 0 and 1, and in general, the higher the R-squared, the better the model fits our data. The function that we fit is D = 17.9 Ln (t) − 4.4785; where D is the cumulative drug release percentage and “t” is the release time in hours. The R-squared value of our fitting model for the PCL/zein coated scaffolds is 0.9998. This clearly indicates that the equation fit the data very well. The analysis of the data demonstrated that the total amount of TCH in PCL/zein coated scaffolds can be released after less than 15 days, as observed in previous studies [15,16,36]. On the other hand, the formulations containing zein presented a lower cumulative release in comparison to the other scaffolds. The coated BG scaffolds containing zein have reduced release level compared to the uncoated and coated scaffolds without protein. This result suggests that the entrapment of the drug by zein protein played an important role in the release behavior. The formation of protein–drug interactions is likely to contribute to the kinetics of the process, which should be related to the establishment of hydrogen bonds, van der Waals forces, and the release of solvent molecules [19–21,43,46]. The prediction of the actual mechanism is challenging. Some of the reasons that make such prediction difficult are related to the fact that the interactions also depend on the degradation rate of both components. In addition, there are difficulties for the evaluation of the enthalpy and entropic contribution of the solvent [43]. ATR-FTIR spectra of the TCH-loaded scaffolds are shown in the Supplementary data. Because of a large amount of nonpolar amino acids, zein might be able to form stable protein–drug complexes. Consequently, it can control the release of drugs. Therefore, zein is used in various types of carrier systems such as film, nano fiber, micro/nano particle, gel, micelle and composite. When zein is used as a film, there are a number of advantages for these carrier systems. It has large surface area and this is respectable for tissue adhesion and as active therapeutic cargo. It can be combined with polymers; otherwise, it has poor mechanical strength as observed in previous investigations [47]. Jiang et al. used ketoprofen (KET) as the model drug, poly vinylpyrrolidone (PVP) and zein as the sheath polymer and core matrix, respectively [48]. Core–sheath PVP/zein nanofibers were electrospun for drug delivery. KET molecules interacted with polymer and zein molecules which is thought to be mainly through the KET–PVP and KET– zein hydrogen bonding. Their studies showed that the PVP/zein nanofibers could deliver an immediate release of 42.3% of KET after 10 h. This initial burst effect was out of control and should be avoided when the sustained release profile is desired [47]. In other studies, drug release from poly lactic-co-glycolic acid (PLGA) and zein matrix has been found to be governed by a diffusion mechanism. It demonstrated that TCH release from PLGA/zein matrix was conditioned by the interaction between the drug and zein. It was

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

also observed that the formulations containing higher amounts of zein showed the highest amount of loaded drug and resulted in a slower release of TCH [19]. Domingues et al. [49] found that cyclodextrin enhanced the time of tetracycline release which was used as a modulator for tetracycline molecules. They showed that there was interaction between glass anionic phosphate groups and the tetracycline acid moiety in uncoated bioactive glass. However, the same result was not observed in cyclodextrin coated BG because the cyclodextrin protected the tetracycline acid upon inclusion [49]. In our study the PCL/zein coated BG scaffolds exhibited the best performance in providing sustained release profile in comparison with the other scaffolds investigated. Future experiments will be conducted to investigate the effect of drug release on biocompatibility of the PCL/ zein coated BG scaffolds. 4. Conclusions Bioactive glass-based scaffolds prepared using the replica method have suitable pore macrostructures for bone tissue engineering applications (average pore size 300 μm and porosity ~ 96%). By optimizing a method for dip coating of BG scaffolds by PCL and zein, the compressive strength and mechanical stability were significantly increased. By increasing the concentration of the polymer, the compressive strength of the coated scaffold was increased. It was also evident that the stress–strain curve for the scaffold coated with the high concentration of polymer solution was much less jagged, and the shape more resembles the ideal curve for relatively tough highly porous foams. Both uncoated and PCL/zein coated scaffolds possessed appropriate bioactivity to form bone-like apatite layers after 2 weeks of immersion in SBF. The PCL/zein composite coating was also developed to enhance the biological activity of BG scaffolds, e.g. inducing faster HA formation. On the other hand, because of hydrophilic nature of zein, the water contact angle and the weight loss of coating were augmented by increasing zein concentration. It was found that the presence of zein accelerates degradation rate of the coating in the time period investigated. The present results show a way for controlling the in vitro degradation behavior of the coating by engineering the concentration of zein. The study described here offers an example of the organized design and preparation of novel type of drug controlled-release composite scaffolds and TCH for providing the drug release profiles. In order to impart control drug delivery function the scaffold was coated by a blend of PCL and zein. Sustained release of tetracycline was confirmed, and the proportion of zein in the coating had a great impact on the drug release behavior. The PCL/zein coated BG scaffolds presented a controlled drug release rate indicating that these scaffolds are promising candidates for in-situ drug release in bone tissue engineering applications. Acknowledgments The authors are grateful to the Isfahan University of Technology for providing financial support to Z. Fereshteh. Dr. Menti Goudori, Mr Bapi Sarker, Ms. Alina Grünewald and Dr Judith A. Roether (University of Erlangen-Nuremberg, Germany) are acknowledged for experimental support. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.05.011. References [1] L.L. Hench, J.M. Polak, Third generation biomaterials, Science 295 (2002) 1014–1017.

59

[2] D.W. Hutmacher, J.T. Schantz, C.X. Lam, K.C. Tan, T.C. Lim, State of the art and future directions of scaffold-based bone engineering from a biomaterials perspective, J. Tissue Eng. Regen. Med. 1 (2007) 245–260. [3] D.M. Yunos, O. Bretcanu, A.R. Boccaccini, Polymer–bioceramic composites for tissue engineering scaffolds, J. Mater. Sci. 10 (2008) 4433–4442. [4] Y. Ikada, Challenges in tissue engineering, J. R. Soc. Interface 3 (10) (2006) 589–601. [5] M.N. Rahaman, D.E. Day, B. Sonny Bal, Q. Fu, S.B. Jung, L.F. Bonewald, A.P. Tomsia, Bioactive glass in tissue engineering, Acta Biomater. 7 (2011) 2355–2373. [6] F. Baino, C. Vitale-Brovarone, Three-dimensional glass-derived scaffolds for bone tissue engineering: current trends and forecasts for the future, J. Biomed. Mater. Res. A 97 (2011) 514–535. [7] A.R. Boccaccini, L.C. Gerhardt, Carbon nanotube composite scaffolds and coatings for tissue engineering applications, Key Eng. Mater. 441 (2010) 31–52. [8] M.H. Fathi, A. Doostmohammadi, Bioactive glass nanopowder and bioglass coating for biocompatibility improvement of metallic implant, J. Mater. Process. Technol. 9 (2008) 1385–1391. [9] A.A. Gorustovich, J.A. Roether, A.R. Boccaccini, Effect of bioactive glasses on angiogenesis: a review of in vitro and in vivo evidences, Tissue Eng. B. Rev 16 (2010) 199–207. [10] Q.Z. Chen, I.D. Thompson, A.R. Boccaccini, 45S5 Bioglass®-derived glass–ceramic scaffolds for bone tissue engineering, Biomaterials 27 (2006) 2414–2425. [11] Q.Z. Chen, A.R. Boccaccini, Poly (D, L-lactic acid) coated 45S5 Bioglass®-based scaffolds: processing and characterization, J. Biomed. Mater. Res. A 77 (2006) 445–457. [12] M. Peroglio, L. Gremillard, J. Chevalier, L. Chazeau, C. Gauthier, T. Hamaide, Toughening of bio-ceramics scaffolds by polymer coating, J. Eur. Ceram. Soc. 27 (2007) 2679–2685. [13] N. Johari, M.H. Fathi, M.A. Golozar, Fabrication, characterization and evaluation of the mechanical properties of poly (ε-caprolactone)/nano-fluoridated hydroxyapatite scaffold for bone tissue engineering, Compos. Part B 43 (2012) 1671–1675. [14] M. Diba, M.H. Fathi, M. Kharaziha, Novel forsterite/polycaprolactone nanocomposite scaffold for tissue engineering applications, Mater. Lett. 65 (2011) 1931–1934. [15] V. Mourino, A.R. Boccaccini, Bone tissue engineering therapeutics: controlled drug delivery in three-dimensional scaffolds, J. R. Soc. Interface 7 (2010) 209–227. [16] Q. Yao, P. Nooeaid, J.A. Roether, Y. Dong, Q. Zhang, A.R. Boccaccini, Bioglass®-based scaffolds incorporating polycaprolactone and chitosan coatings for controlled vancomycin delivery, Ceram. Int. 39 (2013) 7517–7522. [17] D. Alkan, L.Y. Aydemir, I. Arcan, H. Yavuzdurmaz, H.I. Atabay, C. Ceylan, A. Yemenicioğlu, Development of flexible antimicrobial packaging materials against Campylobacter jejuni by incorporation of gallic acid into zein based films, J. Agric. Food Chem. 59 (2011) 11003–11010. [18] A. Salerno, S. Zeppetelli, M. Oliviero, E. Battista, E. Di Maio, S. Iannace, P.A. Netti, Microstructure, degradation and in vitro MG63 cells interactions of a new poly(εcaprolactone), zein, and hydroxyapatite composite for bone tissue engineering, J. Bioact. Compat. Polym. 27 (2012) 210–226. [19] F.O. de Sousa, J. Blanco-Méndez, A. Pérez-Estévez, R. Seoane-Prado, A. LuzardoÁlvarez, Effect of zein on biodegradable inserts for the delivery of tetracycline within periodontal pockets, J. Biomater. Appl. 27 (2012) 187–200. [20] Y. Luo, B. Zhang, M. Whent, L.L. Yu, Q. Wang, Preparation and characterization of zein/chitosan complex for encapsulation of α-tocopherol, and its in vitro controlled release study, Colloids Surf. B 85 (2011) 145–152. [21] Q. Wang, L.L. Yin, G.W. Padua, Effect of hydrophilic and lipophilic compounds on zein microstructures, Food Biophys. 3 (2008) 174–181. [22] H.D. Belitz, R. Kieffer, W. Seilmeier, H. Wieser, Structure and function of gluten proteins, Cereal Chem. 63 (1986) 336–341. [23] L.L. Hench, Bioceramics, J. Am. Ceram. Soc. 81 (1998) 1705–1728. [24] T. Kokubo, H. Takadama, How useful is SBF in predicting in vivo bone bioactivity? Biomaterials 27 (2006) 2907–2915. [25] L.J. Gibson, M.F. Ashby, Cellular Solids: Structure and Properties, 2nd ednPergamon, Oxford 1999, pp. 429–452. [26] P. Nooeaid, V. Salih, J.P. Beier, A.R. Boccaccini, Osteochondral tissue engineering: scaffolds, stem cells and applications scaffolds for osteochondral, Tissue Eng. 16 (2012) 2247–2270. [27] O. Bretcanu, S. Misra, I. Roy, C. Renghini, F. Fiori, A.R. Boccaccini, V. Salih, In vitro biocompatibility of 45S5 Bioglass®-derived glass–ceramic scaffolds coated with poly(3-hydroxybutyrate), J. Tissue Eng. Regen. Med. 3 (2009) 139–148. [28] I. Arcan, A. Yemenicioglu, Incorporating phenolic compounds opens a new perspective to use zein membranes as flexible bioactive packaging materials, Food Res. Int. 44 (2011) 550–556. [29] Y. Ma, X. Li, P. Jia, Y. Ma, N. Liu, H. Zhang, Preparation of zein-based membranes and their pervaporation for ethanol aqueous solution, Desalination 299 (2012) 70–78. [30] J. Hum, K.W. Luczynski, P. Nooeaid, P. Newby, O. Lahayne, C. Hellmich, A.R. Boccaccini, Stiffness improvement of 45S5 Bioglass®-based scaffolds through natural and synthetic biopolymer coatings: an ultrasonic study, Strain 49 (2013) 431–439. [31] B.D. Boyan, T.W. Hummert, D.D. Dean, Z. Schwartz, Role of material surfaces in regulating bone and cartilage cell response, Biomaterials 17 (1996) 137–146. [32] C. Zhang, W. Zhang, H. Yao, H. Zhu, L. Mao, S. Yu, Bioinspired crystallization of continuous calcium phosphate films on a langmuir monolayer of zein protein: their mechanical, Cryst. Growth Des. 13 (2013) 3505–3513. [33] B. Olalde, N. Garmendia, V. Sáez-Martínez, N. Argarate, P. Nooeaid, F. Morin, A.R. Boccaccini, Multifunctional bioactive glass scaffolds coated with layers of poly(D, Llactide-co-glycolide) and poly(n-isopropylacrylamide-co-acrylic acid) microgels loaded with vancomycin, Mater. Sci. Eng. C Mater. Biol. Appl. 33 (2013) 3760–3767. [34] J.M. Yang, L. Zha, D.G. Yu, J. Liu, Coaxial electrospinning with acetic acid for preparing ferulic acid/zein composite fibers with improved drug release profiles, Colloids Surf. B Biointerfaces 102 (2013) 737–743.

60

Z. Fereshteh et al. / Materials Science and Engineering C 54 (2015) 50–60

[35] D. Mohn, C. Bruhin, N.A. Luechinger, W.J. Stark, T. Imfeld, M. Zehnder, Composites made of flame-sprayed bioactive glass 45S5 and polymers: bioactivity and immediate sealing properties, Int. Endod. J. 43 (2010) 1037–1046. [36] W. Li, P. Nooeaid, J.A. Roether, D.W. Schubert, A.R. Boccaccini, Preparation and characterization of vancomycin releasing PHBV coated 45S5 Bioglass®-based glass–ceramic scaffolds for bone tissue engineering, J. Eur. Ceram. Soc. 34 (2014) 505–514. [37] P. Zhou, Y. Xia, J. Wang, C. Liang, L. Yu, W. Tang, S. Gu, S. Xu, Antibacterial properties and bioactivity of HACC- and HACC–Zein-modified mesoporous bioactive glass scaffolds, J. Mater. Chem. B 1 (2013) 685–692. [38] S. Gautam, A.K. Dinda, N.C. Mishra, Fabrication and characterization of PCL/gelatin composite nanofibrous scaffold for tissue engineering applications by electrospinning method, Mater. Sci. Eng. C. Mater. Biol. Appl. 33 (2013) 1228–1235. [39] L.A. Forato, T.D.E.C. Bicudo, L.A. Colnago, Conformation of α-zeins in solid state by Fourier transform IR, Biopolymers 72 (2003) 421–426. [40] O.P. Filho, G.P. La Torre, L.L. Hench, Effect of crystallization on apatite-layer formation of bioactive glass 45S5, J. Biomed. Mater. Res. 30 (1996) 509–514. [41] Y. Zhu, S. Kaskel, Comparison of the in vitro bioactivity and drug release property of mesoporous bioactive glasses (MBGs) and bioactive glasses (BGs) scaffolds, Microporous Mesoporous Mater. 118 (2009) 176–182. [42] A. Salerno, M. Oliviero, E. Di Maio, P.A. Netti, C. Rofani, A. Colosimo, V. Guida, B. Dallapiccola, P. Palma, E. Procaccini, A.C. Berardi, F. Velardi, A. Teti, S. Iannace, Design

[43]

[44]

[45]

[46]

[47] [48] [49]

of novel three-phase PCL/TZ-HA biomaterials for use in bone regeneration applications, J. Mater. Sci. Mater. Med. 21 (2010) 2569–2581. A.K. Jaiswal, S.S. Kadam, V.P. Soni, J.R. Bellare, Improved functionalization of electrospun PLLA/gelatin scaffold by alternate soaking method for bone tissue engineering, Appl. Surf. Sci. 268 (2013) 477–488. A. Salerno, M. Oliviero, E. Di Maio, P.A. Netti, C. Rofani, A. Colosimo, et al., Design of novel three-phase PCL/TZ-HA biomaterials for use in bone regeneration applications, J. Mater. Sci. Mater. Med. 21 (2010) 2569–2581. D.J. Simmons, S.L. Chang, J.E. Russell, B. Grazman, D. Webster, C. Oloff, The effect of protracted tetracycline treatment on bone growth and maturation, Clin. Orthop. Relat. Res. 180 (1983) 253–259. F.F.O. Sousa, A. Luzardo-Álvarez, J. Blanco-Méndez, M. Martín-Pastor, NMR techniques in drug delivery: application to zein protein complexes, Int. J. Pharm. 439 (2012) 41–48. R. Paliwal, S. Palakurthi, Zein in controlled drug delivery and tissue engineering, J. Control. Release 189 (2014) 108–122. Y.N. Jiang, H.Y. Mo, D.G. Yu, Electrospun drug-loaded core–sheath PVP/zein nanofibers for biphasic drug release, Int. J. Pharm. 438 (2012) 232–239. Z.R. Domingues, M.E. Cortés, T.A. Gomes, H.F. Diniz, C.S. Freitas, J.B. Gomes, A.M. Faria, R.D. Sinisterra, Bioactive glass as a drug delivery system of tetracycline and tetracycline associated with β-cyclodextrin, Biomaterials 25 (2004) 3 27–333.