Fabrication of novel high performance ductile poly(lactic acid) nanofiber scaffold coated with poly(vinyl alcohol) for tissue engineering applications

Materials Science and Engineering C 60 (2016) 143–150

Contents lists available at ScienceDirect

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

Fabrication of novel high performance ductile poly(lactic acid) nanofiber scaffold coated with poly(vinyl alcohol) for tissue engineering applications Abdalla Abdal-hay a,⁎, Kamal Hany Hussein b, Luca Casettari c, Khalil Abdelrazek Khalil d,e,⁎⁎, Abdel Salam Hamdy f a

Dept of Engineering Materials and Mechanical Design, Faculty of Engineering, South Valley of University, Qena 83523, Egypt Stem Cell Institute and College of Veterinary Medicine, Kangwon National University, Chuncheon, Gangwon 200-701, Republic of Korea c Department of Biomolecular Sciences, University of Urbino, Piazza Rinascimento, 6, Urbino, PU 61029, Italy d Dept. of Mechanical Engineering, College of Engineering, King Saud University, 800, Riyadh 11421, Saudi Arabia e Dept. of Mechanical Engineering, Faculty of Energy Engineering, Aswan University, Aswan, Egypt f Dept. of Manufacturing and Industrial Engineering, College of Engineering and Computer Science, University of Texas Rio Grande Valley, 1201 West University Dr., Edinburg, TX 78541-2999, USA b

a r t i c l e

i n f o

Article history: Received 3 May 2015 Received in revised form 3 October 2015 Accepted 8 November 2015 Available online 10 November 2015 Keywords: Poly lactic acid Poly vinyl alcohol Tissue regeneration Hydrothermal deposition Nanofiber scaffolds Cytocompatibility Biodegradable synthetic polymers

a b s t r a c t Poly(lactic acid) (PLA) nanofiber scaffold has received increasing interest as a promising material for potential application in the field of regenerative medicine. However, the low hydrophilicity and poor ductility restrict its practical application. Integration of hydrophilic elastic polymer onto the surface of the nanofiber scaffold may help to overcome the drawbacks of PLA material. Herein, we successfully optimized the parameters for in situ deposition of poly(vinyl alcohol), (PVA) onto post-electrospun PLA nanofibers using a simple hydrothermal approach. Our results showed that the average fiber diameter of coated nanofiber mat is about 1265 ± 222 nm, which is remarkably higher than its pristine counterpart (650 ± 180 nm). The hydrophilicity of PLA nanofiber scaffold coated with a PVA thin layer improved dramatically (36.11 ± 1.5°) compared to that of pristine PLA (119.7 ± 1.5°) scaffold. The mechanical testing showed that the PLA nanofiber scaffold could be converted from rigid to ductile with enhanced tensile strength, due to maximizing the hydrogen bond interaction during the heat treatment and in the presence of PVA. Cytocompatibility performance of the pristine and coated PLA fibers with PVA was observed through an in vitro experiment based on cell attachment and the MTT assay by EA.hy926 human endothelial cells. The cytocompatibility results showed that human cells induced more favorable attachment and proliferation behavior on hydrophilic PLA composite scaffold than that of pristine PLA. Hence, PVA coating resulted in an increase in initial human cell attachment and proliferation. We believe that the novel PVA-coated PLA nanofiber scaffold developed in this study, could be a promising high performance biomaterial in regeneration medicine. © 2015 Published by Elsevier B.V.

1. Introduction Biodegradable synthetic polymer fibers have attracted massive interest as effective substitute materials in regeneration medicine applications [1,2]. Specifically, poly(lactic acid) (PLA) is one of the most widely used synthetic polymers in this field due to the non-toxicity of lactic acid, which is naturally present in the human body, and FDAapproved [3–6]. However, there are some major limitations such as the hydrophobic nature and poor ductility of PLA which hinder its practical

⁎ Corresponding author. ⁎⁎ Correspondence to: K.A. Khali, Dept. of Mechanical Engineering, College of Engineering, King Saud University, P.O. Box 800, Riyadh 11421, Saudi Arabia. E-mail address: [email protected] (A. Abdal-hay).

http://dx.doi.org/10.1016/j.msec.2015.11.024 0928-4931/© 2015 Published by Elsevier B.V.

use as substitute materials in tissue regeneration [3]. It is known that the surface wettability reflects the adhesion, growth of cells, and protein absorption on the surface of the material [3,6]. Some researchers noticed that the porous scaffolds fabricated from PLA are floating in cell culture medium [7]. Thus, the hydrophobic nature of PLA is a serious problem in a predominantly hydrophilic bioenvironment where the cells fail to have initial attachment to the implanted scaffolds. Mechanical properties play a crucial role in determining the in vivo performance of the scaffolds in the tissue engineering field, such as vascular graft system [4], bone implants, and wound dressing [8]. The scaffold has to be composed of a durable biomaterial capable of withstanding physiological hemodynamic forces while maintaining structural integrity until mature tissue forms in vivo. Electrospun nanofibers represent an emerging class of biomimetic nanostructures that can act as proxies of the native tissue, while

144

A. Abdal-hay et al. / Materials Science and Engineering C 60 (2016) 143–150

providing topographical and biochemical cues that promote tissue healing. Development of advanced nanofiber scaffold in the second half of the 20th century resulted in a marked revolution in advanced structural materials [9] mainly because of the brittleness characteristics of PLA nanofibers as a potential scaffold material [4,10,11]. Many efforts have been invested to address the aforementioned limitations of PLA nanofibers. For instance, poly blend nanofibers prepared by simple pre-mixing solution or melt-blending of PLA with organic polymers [4,11–15] such as thermoplastic starch (TPS) were reported [14]. However, microscopic observations revealed non-uniform dispersed PLA inclusions within the TPS matrix. Pitarresi et al. [16], produced electrospun nanofiber scaffolds by employing PHEA-g-PLA copolymer as a starting material. However, these scaffolds didn't provide significant biocompatible properties. Pi et al. [15], noticed the occurrence of microphase separation in the coating layer through the film drying because of the dissimilarity of PLA and polyethylene oxide (PEO). This is because PEO is water soluble, while PLA is waterinsoluble. Therefore, the preparation of wettable and ductile PLA scaffold using a facile approach for biomedical applications remains a challenge. As a result, changing the mechanical properties of postelectrospun PLA nanofibers from brittle/rigid to ductile using a simple, cost-effective approach is one of the main challenges in the present study. Successful development of a hydrophilic and ductile PLA nanofiber scaffold will open a new era towards the designing of high performance advanced biomaterials in the field of tissue engineering. In this paper, we successfully developed a facile strategy that is based on a surface modification of the post-electrospun (PE) PLA nanofibers through deposition of a hydrophilic and elastic layer on each single PLA nanofiber using a simple hydrothermal route. The novelties of this paper are: a) optimizing of a simple and cost-effective hydrothermal process for fabrication of a novel PLA based-scaffold with unique morphology and mechanical and biological properties; b) overcoming the lacks of pristine PLA; and c) introducing the amine groups in the fibers for improving the cell adhesion and proliferation in tissue engineering. Recently, our research group has successfully exploited a simple and inexpensive hydrothermal strategy for in situ deposition of inorganic compounds onto PE nanofibers and creating advanced high performance 3D composite scaffolds for medical implants [17,18]. Interestingly, our process has no side effect on the properties of the polymer fibers. Furthermore, it provides sufficient interfacial bonding between the polymer fibers and deposited compounds, and subsequently improves the mechanical properties of the scaffold. This novel process has been successfully used by our group to apply an in situ deposition of a wettable and elastic polymer thin layer on PE PLA nanofiber. In this study, poly(vinyl alcohol) (PVA) was selected to conformal coat each single PLA nanofiber because it is a water-soluble synthetic polymer that possesses good biocompatibility, biodegradability, and excellent mechanical properties [4]. PVA, in general, has a high water content and tissue-like elasticity. The abundant hydroxyl groups on PVA can be readily modified to attach growth factors and adhesion proteins. We speculate that the hydrophilic properties of PVA molecules can create a good physical/chemical interaction with PLA nanofibers on the molecular level by forming strong hydrogen bonding throughout a hydrothermal treatment, and thereby enhancing the surface and mechanical properties of the PLA nanofibers. It is worthmentioning that exploiting our strategy can overcome the difficulty of mixing both PLA and PVA phases at the macromolecular level because PVA is more hydrophilic than PLA. Hence, the composite PVA/PLA nanofibrous scaffold that integrates the favorable wettability properties of PVA and elasticity as well as FDA approval of PLA is expected to significantly improve the material properties for tissue regeneration applications. Additionally, the cytocompatibility performance of the PLA nanofiber scaffold coated with a hydrophilic PVA thin layer using hydrothermal strategy was studied using the MTT test. The molecular interactions between PLA fiber and PVA molecules during the hydrothermal process were discussed in detail.

2. Experimental 2.1. Fabrication of nanofibers The electrospinning setting used in the current research for fabrication of nanofibers and PLA pellets is a type of Ingeo Biopolymer 2003D, (a Nature Works, LLC (USA) product supplied by Green Chemical Co., Ltd., Korea) as described in detail in our previous report [19]. Briefly, the solution for electrospinning was prepared by dissolving PLA in dichloromethane (Junsei Chemical Co., Japan) at a concentration of 10 wt.%. The injection rate and the applied voltage were 0.5 ml/h and 18 kV, respectively. The collected nanofiber mat was placed in a vacuum oven for 24 h at 40 °C to remove any potential residual solvents. A PVA-coated PLA electrospun mat was prepared following these procedures: (1) the PLA as-electrospun mat was cut into rectangular specimens (30 × 20 mm2). (2) These specimens were immersed into optimized freshly prepared 1 wt.% PVA aqueous solution (it was observed that increasing the PVA solution concentration N 1 wt.%, affects negatively the surface morphology of the nanofiber mats as shown in Fig. S1 of the supplementary materials). PVA solution viscosity at 1 wt.% was measured by a Brookfield, DV-III ultra programmable Rheometer at room temperature and the result was about 17 cP. (3) 40 ml of PVA solution containing the PLA electrospun nanofiber mat was placed in a Teflon-lined autoclave container and heat-treated at 150 °C for 30 min. Previous studies reported that PVA and PLA polymers have good thermal resistivity up to this temperature [20,21]. (4) After the reaction process is complete, the PVA-coated PLA mats were gently rinsed in distilled water to remove unattached PVA molecules from PLA nanofiber mats. (5) The resultant coated samples were left to dry at room temperature for four days until steady weight and then introduced into a vacuum oven (10 mbar) at 35 °C for 72 h. (6) The amount of PVA adhering to the surface was evaluated from the dry weight of the substrates. 2.2. Surface/material characterizations The surface micrographs of the pristine and composite mats were characterized by a field emission scanning electron microscope (FESEM; Hitachi S-7400, Japan). The micrographs of the coated samples were taken at an accelerating voltage of 5 kV and with magnifications of 5 and 25 K. To measure the fiber diameters, the FESEM images were processed and analyzed by means of ImageJ software (National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/). FTIR (MB100 spectrometer, Bomen, Canada) analysis in a transmission mode was used to identify the functional groups thereby reflecting phase structure of pristine and hydrothermally treated samples. Thermogravimetric analysis (TGA) was performed by a TGA-DSC, Q-20 Perkin-Elmer Inc., USA, at a heating rate of 20 °C/min with a constant purge of N2. Differential scanning calorimetry (DSC) data were obtained from a Perkin-Elmer Pyris Diamond DSC. Samples were scanned at a heating rate of 10 °C/min in N2 environment. The Tg values were measured as the change of the specific heat in the heat flow curves. 2.3. Surface wettability measurements Flat mats were used to evaluate the hydrophilicity of pristine PLA and PVA/PLA composite nanofiber (treated) mats, using the water contact angle (WCA) measurements. 3 μl of purified water (ultrapure grade) was pipetted out on top of the shiny side of 30 × 30 mm2 mats positioned on the stage of a bench-type contact angle goniometer (GBX; Digidrop, France), ensuring that the membrane mat was completely flat. The micrographs were taken after 1 s and WCA measurement was recorded. To confirm the coating homogeneity and coating distribution of a PVA layer on the PLA membrane mat, the WCA was measured at five different positions on each flat mat surface.

A. Abdal-hay et al. / Materials Science and Engineering C 60 (2016) 143–150

The PVA/PLA composite scaffold was compared to the pristine PLA scaffold for their wettability or absorbability of purified water. The purified water was dropped on top of flat mats and the time required for complete absorption of the water into the scaffold was estimated to see whether it is absorbable or un-absorbable. The flat mat was gently stretched and fixed on a standard microscope glass slide (dimensions approx. 76 × 26 mm for glass slide). Each prepared sample was measured three times, and the average value was recorded. 2.4. Mechanical properties Pristine and treated PLA mats were subjected to stress–strain analysis using an Instron universal tester model: LLOYD instruments, LR5K plus, UK. The samples were trimmed into a “dogbone” (see inset of Fig. 3) with offset ends via die cutting from the as-obtained mats to reduce grip effects according to ASTM D-638. Testing was conducted with the tissue grips moving at a rate of 10 mm/min. The applied load was conducted until the specimen experienced complete failure. The specimen thicknesses were measured using a digital micrometer with a precision of 1 μm (coating thickness gauge OMEGA instrument, OM179745). The tensile modulus was calculated as the slope of the initial linear portion of the stress–strain curve. The data acquisition rate was set to 20.0 Hz. Four membrane mat samples of each group were subjected to tensile testing at room temperature. The data presented were expressed as the mean ± standard deviation. Statistical analysis was performed using Student's t-test, and a p-value less than 0.05 was considered significant. 2.5. Cytocompatibility studies (cell attachment and MTT assay) To investigate the cell adhesion on the surfaces of as-prepared pristine PLA and treated PLA with PVA thin coated layer membrane samples, the pristine PLA and PLA/PVA hybrid scaffolds were placed and fixed in the bottom of each well of a 24-well plate and then sterilized with ethylene oxide (ETO) gas for 24 h at room temperature. Subsequently, EA.hy926 endothelial cells (passages 4–5) from American Type Culture Collection (ATCC) were then seeded onto the surfaces of the membrane samples at a density of 30 × 103 (300 μl of cell suspension) and cultured with Dulbecco's Modified Eagle's Medium (DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT, USA) and 1% penicillin/streptomycin (P/S; Gibco, Grand Island, NY, USA) in a humidified incubator at 37 °C and 5% CO2. Prior to cell culture on material surface, at 70% confluency, the cells were harvested by trypsinization and used for experiments. Samples containing cells were taken out after incubating the plates for 3 and 7 days, rinsed twice with phosphate-buffered solution (PBS) to remove the non-attached cells, and subsequently fixed with 2.5% glutaraldehyde for 2 h at 4 °C overnight. Then, the samples were rinsed in 0.1 M PBS and transferred to the critical-point dryer and dried with CO2. The dried samples were sputtered with a thin layer of gold for observation of cell morphology using low voltage Bio-SEM (Hitachi, S-3000N, Japan) at an acceleration voltage of 10 kV. To investigate cell proliferation, the attached cells viability (cell density) on the pristine and hybrid membrane scaffold samples was determined by the MTT (3-[4,-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide) assay. The MTT was prepared in phosphate buffered saline (PBS) at a final concentration of 5 mg/ml. For the assay, the scaffolds were fixed in a 24-well cell culture plate and were sterilized with ethylene oxide (ETO) steam for 24 h, using DMEM supplemented with 10% FBS and 1% P/S, then medium was aspirated, and replaced by 500 μl conditioned or control medium after adding 10% FBS. The EA.hy926 endothelial cell response was evaluated after incubating the plate for 3, 5 and 7 days. At the end of the incubation, 50 μl of MTT solution was added to each well followed by a 4-h incubation at 37 °C. The medium was aspirated, and 350 μl dimethyl sulfoxide (DMSO) was added to each well to dissolve the blue formazan crystal. Then the

145

absorbance was recorded at a test wavelength of 570 nm and a reference wavelength of 630 nm. A mean value was obtained from the measurement of four test runs. 2.6. Statistical analysis All the quantitative data were statistically analyzed to express as the mean (standard deviation (SD)). Statistical analysis was determined by single factor ANOVA. p-Values less than 0.05 were considered significant. 3. Results and discussion The degrees of hydrolysis affect the solubility of PVA in water, whereas PVA with high degree of hydrolysis has low solubility in water. Therefore, we used PVA (MW, 146,000–186,000) with 99% degree of hydrolysis. The selected PVA did not show solubility in water at room temperature, but showed solubility at N80 °C under stirring condition for 12 h. Once it dissolved, its solubility was maintained when the temperature goes down to room temperature. Fig. 1A–C (FESEM images) displays the fiber morphology of aselectrospun (pristine) PLA and PLA fibers coated with PVA. The pristine nanofibers exhibited a smooth surface and moderate uniform diameter along their lengths as well as a non-interconnected (linear) fiber structure (panel A). FESEM analysis of a pristine mat nanofiber verified the diameter of tissue scaffold to be 650 ± 180 nm. The PVA-coated PLA composite fiber mat showed cylindrical morphology with tiny defects along the fiber axis after the hydrothermal process (panels B and C). Moreover, PVA polymer is completely masking the individual PLA nanofiber along the fiber direction, and exhibits extremely extended nanobranches, of shooting the main nanofibers (Fig. 1B and C). No porous structure was observed on the fiber surface, indicating that PVA plays a significant role to protect PLA fibers from any breakdown or fragment formation during the deposition process (data are not shown). Ribeiro et al. [22] found that the nanofiber matrix of PLA changed into chunks consisting of short fiber fragments with porous structure on their surface during a simple immersion in aqueous solution. They attributed such behavior to the simple hydrolysis of the ester backbone of aliphatic polyester under aqueous conditions [23]. The imageJ software was used to determine the diameter of different fibers. Results indicated that the fiber diameter increased after PVA incorporation onto the fiber surface (panel C). The average fiber diameter of the composite nanofiber mat was about 1265 ± 222 nm, which is remarkably higher than its pristine counterpart (i.e., 650 ± 180 nm). PVA coating on the PLA fibers provides interconnected and pseudo coagglomeration of nanofibers in addition to sufficient amount of joint-welding of the fibers at their neighboring points (panels B) which is absent in the pristine one (panel A). High magnification (panel C) reveals that PVA completely covered each individual PLA fiber and the PVA layer grew on the PLA main nanofibers without any phase separation. The mat coated with PVA showed a significant increase in its weight by about 0.38 mg PVA/1.0 cm2 of the PLA mat. These data demonstrate an increase in diameter size of the composite scaffold due to the deposition of a PVA layer on PLA fibers. We believe that during hydrothermal reaction at elevated temperature (150 °C) the PVA solution viscosity decreases resulting in higher solution flow on the PLA single fibers causing the formation of a PVA layer with a highly branched bridge web-like structure (panels B and C). We noticed that the PVA web-like structure forms a kind of network between the coated fibers (panels B and C) while maintaining the initial 3D structure of the membrane mat. However, this behavior has not been detected at a temperature lower than 150 °C (data are not shown here). It seems that the web-like fibers act as bonding joints between the main fibers. It's likely that decreasing viscosity at elevated temperature might cause higher PVA molecule mobility which can explain the formation of branched PVA bridge fibers.

146

A. Abdal-hay et al. / Materials Science and Engineering C 60 (2016) 143–150

Fig. 1. SEM images of; (A) pristine PLA as-electrospun and PVA/PLA composite mat at low (B) and high (C) magnification. Dashed selected circular area in panel B is the respective position for panel C.

The reaction temperature of composite sample, 150 °C, was precisely controlled based on our previous study on thermal analysis of PLA fibers [3] whereas the melting temperature of PLA is about 156 °C (Fig. 2A). It was reported that the thermal bonding of semi-crystalline polymer fibers, such as PLA, can occur near the melting temperature [24,25]. Therefore, we hypothesize that some thermal interfiber-layer bonding was effectively achieved between the PLA fiber and the PVA coated layer during the hydrothermal treatment. To verify our hypothesis, FTIR was used to study the effect of hydrothermal treatment in the presence of PVA solution on the interfacial bonding of the composite polymers (Fig. 2B). From the IR spectra in Fig. 2B, the characteristic bands of pristine PVA occur at 838 cm−1 (rocking of CH), 919 cm−1 (bending

of CH2), 1088 (stretching of CO and bending of OH from amorphous sequence of PVA), 1143 cm−1 (stretching of CO from crystalline sequence of PVA), 1417 cm−1 (wagging of CH2 and bending of OH), 2906 cm−1 (symmetric stretching of CH2), 2935 cm−1 (asymmetric stretching of CH2) and 3278 cm−1 (stretching of OH). The comparison between the FTIR spectra of pristine and treated mats clearly shows that sufficient amount of PVA was deposited on the surface of PLA fibers as the intensity of different bands of PVA/PLA mats was significantly decreased and became narrower (Fig. 2B). It was reported that the bands of carbonyl group of PLA and the bands of hydroxyl group of PVA occur at 1758 and 3333 cm−1, respectively [26]. Accordingly, our results clearly identified both bands in the hydrothermally treated PVA-coated mats, while

Fig. 2. (A) DSC; (B) FTIR spectra and (C) TGA curves of the pristine and composite mats. Dashed arrows show the difference in absorption band of the pristine PLA nanofibers and PLA nanofibers coated with PVA molecules.

A. Abdal-hay et al. / Materials Science and Engineering C 60 (2016) 143–150

the 3333 cm−1 peak of pristine PLA disappeared (Fig. 2B). This provides another evidence for depression of this peak on PLA due to the deposition of PVA [26]. In the composite mat, the IR functional groups involved in strong intermolecular hydrogen bonding often exhibit obvious shifts in their vibration frequencies (Fig. 2B). The characteristic peaks of the carbonyl and hydroxyl groups of the composite mat were shifted towards higher frequencies due to the intermolecular hydrogen bonding interaction between the hydroxyl groups of PVA and carbonyl group of PLA. These results are in a good agreement with the FESEM morphology (Fig. 1B, C) where the interactions between the two polymers in the composite mat occurred without any phase-separation. Likewise, the crystallinerelated peaks at 1417 and 1326 cm−1 in pure PVA disappeared in the bands of composite fibers, which can be also attributed to intermolecular interaction between the polymers at the molecular level [26]. Furthermore, the band intensity of the composite mat was monotonically decreased after PVA deposition supporting the mechanism of formation of an outer PVA layer on the surface of PLA fibers. The hydrothermal treatment enhances the extent of shift of the carbonyl and hydroxyl groups indicating the formation of stronger hydrogen bonding and suggesting that such favorable interactions between the two polymers can lead to a miscible mat [26,27]. Additionally, it can also be identified that the intensity of the C–H stretching groups definitely varied (Fig. 2B) [28]. Taking the pristine mat as a base, the peak intensity of around 2944 and 2995 cm−1 significantly decreased after hydrothermal deposition of PVA. Indeed, the deposition of PVA using the hydrothermal approach can assure strong intermolecular hydrogen-bonding interactions [26,28]. Ribeiro et al. [22], reported that poly(L-lactide) electrospun fibers are amorphous but contain numerous crystal nuclei that could grow rapidly when the sample is heated up to 140 °C. Similarly, we expect that the degree of crystallinity of the fibers can be tailored and controlled by the hydrothermal treatment. The melting peaks of PLA showed no significant differences after PVA loading as revealed by DSC measurements (Fig. 2A). On the other hand, the Tm of pristine PVA could clearly be detected at 220 °C. After the hydrothermal process and formation of a PVA thin layer on PLA fibers, the PVA melting temperature slightly shifted to a lower value (218 °C). The typical TGA thermogram curves of pristine PLA, PVA and PLA coated with PVA samples are demonstrated in Fig. 2C. Notably, the TGA of pristine PVA polymer (prepared by sol–gel route at a concentration of 1 wt.%) exhibited three distinct weight loss stages at 30–210 °C (5 wt.% loss of weakly physisorbed water), 210–350 °C (decomposition of side chain of PVA) and 350–540 °C (decomposition of main chain of PVA), as shown in Fig. 2C. Subsequently, the pristine PLA nanofibers showed a single-step degradation at 305 °C, whereas the counterpart PVA/PLA composite nanofibers showed degradation in two steps; first step occurs at 329 °C due to the presence of PVA, and the second step occurs at 359 °C due to PLA (Fig. 2C). The onset temperatures of nanofibers were calculated to be 335 °C and 285 °C for pristine and composite nanofibers, respectively. This data provides further evidence for the successful incorporation of PVA onto the PLA nanofibers. For the thermal decomposition of PLA materials, the thermal decomposition temperature (Td) was reduced by 15% for PLA nanofibers modified with PVA, as shown in Fig. 2C. The decrease in Td during the hydrothermal treatment can be attributed to the thermal degradation of PLA at 150 °C. Also, it is likely that the PLA fibers are partially melted (Tm = 156 °C) during the treatment process to form an interfiber-layer bonding with PVA molecules and create a miscible system. Otherwise, it could be also due to the thermal instability of PLA [29]. Based on our experimental results, miscible polymer systems can produce new materials with designated properties superseding those of their constituents. Collectively, it is worthmentioning that the reduction in degradation temperature does not affect the properties of the scaffold when it is used in biological temperature (i.e., 37 °C) [30] and was therefore deemed to be insignificant in practical terms.

147

The glass transition temperature value of PLA slightly decreases after the treatment process and can be attributed to the presence of PVA phase. These observations provide another evidence that PLA and PVA, prepared by the hydrothermal process, can provide a good compatibility [26]. Tsuji and Muramatsu [31] illustrated that PLA and PVA fabricated at room temperature showed phase-separation in their blend films after solvent evaporation. Other studies confirmed that PLA is partially miscible with PVA [32]. Leiggener et al. [33] reported through an in vivo study that the higher degree of crystallinity results in a higher chemical strength and loading capacity which promises advantages for long-term implantation. Our results showed that the incorporation of PVA molecules at 150 °C induces the crystallization of composite samples (Fig. 2A) due to the enhanced chain mobility [12], where the measured heat of fusion (melting enthalpy) of the composite sample is 40.61 J/g compared to 31.12 J/g for the pristine sample. Increasing the enthalpy of fusion suggests that the crystallinity and perfection of the crystal structure are increased by PVA loading. The improvement in crystallinity after incorporation of a PVA thin layer onto the postelectrospun fibers indicates that there are interactions between PLA and PVA layer, i.e., formation of hydrogen bonding. In other words, the deposition of a PVA layer onto the PLA fibers readily induces a chain conformation without defects in the crystalline phase of PLA. In addition, it is known that the melting enthalpy reflects the crystallinity of polymer [40]. This can be explained by the decrease in the content of the amorphous domains due to crystalline growth [3]. This numerous growth (also known as nucleation) can be attributed to nonequilibrium chain conformations imposed by the electrospinning process that can be frozen upon the evaporation of the solvent as noted by Zong et al. [34]. The crystallization behavior of PLA has been discussed elsewhere [35] and results showed that the crystallinity can be obtained at more than 100 °C. The degree of crystallinity increases the treatment temperature below Tm. The temperature required for transition from a glassy state to a rubbery state will be higher. These results qualitatively disagree with the results presented previously [34] in which a relatively low crystallinity was observed for electrospun polymer fibers from the solution state. The chain conformation has also been noticed from FTIR results (Fig. 2B). The 921 cm−1 absorption band which is characteristic of the α-crystal in PLA [20] is clear in the pristine mat and becomes less intense in the composite mat (Fig. 2B). Fig. 2B shows that the increase in the crystallization degree is accompanied by significant narrows and change in the shape of the absorption band between 830 and 890 cm−1 for the composite mat [22]. The interfacial bonding and chain conformation can improve the mechanical properties of the composite mat. Collectively, hydrothermal treatment on PLA electrospun nanofibers in the presence of PVA molecules could create physical joint and chemical bonding between PLA fibers and PVA deposited layer. The substitute nanofiber materials used in tissue treatment should maintain adequate mechanical strength over critical phases of the tissue-healing process to withstand the physiological bioenvironment and tearing resistant during implantations. To investigate the mechanical properties of the pristine and composite mats, load displacement curves were obtained whereas the tensile strengths and percent of elongation (ductility) were calculated. The load displacement curves (Fig. 3) show that the tensile strength of the composite mat (12.9 MPa) was higher than that of the pristine mat (8.1 ± 1.5 MPa). Interestingly, composite fabric induced excellent ductility compared to pristine fabric (over than 2.5 times). This indicates that the PVA coating has a strong affinity on ductility improvement due to the increased elastic characteristic of the composite mat [21]. Ma and co-workers [36], found that the addition of PVA to the blend films improves the extensibility of the composite blend films as a potential tissue engineering matrix. Norton et al. [37] were able to tune the ductile properties of gellan through the incorporation of PVA as a secondary polymer network, for use in cartilage and skin as complex structures. The possible reasons could be due to that formation of a PVA layer during the hydrothermal treatment

148

A. Abdal-hay et al. / Materials Science and Engineering C 60 (2016) 143–150

Fig. 3. Selected stress–strain curves of pristine and composite nanofiber mats under tensile loading. Inset illustrates dog bone samples and their fixation at machine grip. Also, insets are their respective water contact angle images.

process resulting in an increase the stretching of the fiber composite mat. Our results are in good agreement with the previous study by Yang et al. on PVA/gelatin composite polymers [38]. It was also reported that the heat treatment further improves the mechanical properties of the membranes [39]. In contrast, Martin and Avérous [14], found that a simple blending of PLGA with PVA has a detrimental effect on the mechanical properties, with a loss of 31% in Young's modulus and more than 60% in tensile strength compared to the PLGA scaffold. The deposition of PVA on PLA fibers sharply improves the tensile strength and elongation at break for composite polymers, because of the good deformability and flexibility of PVA. We suggest that the improved tensile strength of the composite mat compared to pristine one can be attributed to increased tear resistance of the branched PVA fibers (web-like fibers, parallel PLA fibers bounded to each other by PVA fibers of different diameters) compared to linear (non-branched) membrane rather than the presence of physical/ chemical bonding between the polymers. The interconnected, network structure of the coated fibers resulted in higher tensile strength. It was already reported that the mechanical properties of nonwoven mats increase with increasing fiber junctions [40]. Thus, it is possible that the enhanced mechanical properties for the PVA-coated scaffold are also due to the fiber junction rather than the crystallinity improvement of the treated scaffold. These results describe the fact that the mechanical behavior of nonwoven mats depends mainly on the chemical composition, treatment condition, morphology, and bonding structure of fibers. Hence, the relative improvement in the mechanical properties of the composite mat may be explained in terms of the increased affinity between the two macromolecules. The miscibility resulted in difficult slippage of chains under loading because of more entanglements and strong physical/chemical interactions among the chains of composite polymers, such as hydrogen bond. Therefore, the addition of PVA in the presence of heat energy was quite helpful to improve the mechanical properties of PLA nanofiber scaffolds, overcome the brittleness nature of PLA and, to sharply convert it to ductile biomaterial. Contact angles, which depend on topographic pattern and chemical composition, reflect the hydrophilicity of scaffolds due to protein absorption and cell attachment [3,6,9]. The inset of Fig. 3 shows that the incorporation of a PVA thin layer sharply decreases the contact angle of PLA nanofibers and consequently, improves its hydrophilicity. We found out that the hydrophilicity of the coated PLA scaffold dramatically improved (36.11 ± 1.5°) compared to the pristine PLA (119.7 ± 1.5°) scaffold which is similar to the author's previous report [3] on the pristine PLA mat. The previous study also showed that the pristine PLA membrane surface possesses hydrophobic nature. The position-

dependent water contact angle changes for each PVA-coated PLA fiber mat surface were not significant, indicating the homogeneity of PVA coating on PLA fibers during the hydrothermal process. However, it was reported that the water contact angle measurements of nanofibers cannot exactly reflect the degree of wettability of the polymers and the results are purely qualitative, because the liquid drops on nanofibers cannot provide a full contact with the fiber surface [41]. Accordingly, the wettability was further confirmed in our study by the rate of water absorption to the prepared scaffolds. We noticed that the pristine PLA mat (Fig. S2A of supplementary materials) doesn't show wettability/absorbability until 80 min when water droplet has fallen on the surface, as expected, showing their hydrophobic characteristics. Conversely, the water absorption rate on the composite nanofiber mat was very fast within few seconds (Fig. S2B), which is qualitatively indicating that they are more hydrophilic. We confirmed these results using a video clip in the supporting information (Video clip S3 A and B). According to British Standard 4554:1970 (Method of Test for Wettability of 3D Fabrics), fabrics giving times greater than 200 s with water are considered to be unwettable. The fast absorption and wetting of the PVA/PLA composite scaffolds are highly desirable for tissue engineering applications because the cells can be seeded directly and cultured in this hydrophilic scaffold without any further modification. Therefore, deposition of a PVA thin layer on the surface of PLA fibers during the hydrothermal process could easily increase the hydrophilicity of as-spun fibers and increase its potential application in biological system. To prove this hypothesis together with investigating the effects of a PVA thin layer formation on PLA, we studied the influence of EA.hy926 endothelial cell attachments as well as proliferation on the pristine and treated scaffolds at different culture times. The cell attachment was confirmed using a bio-scanning electron microscope as shown in Fig. 4A–D. From the graphs, it can be seen that the endothelial cell has a good affinity to attach and grow on both pristine and composite scaffolds at different culture times. However, the cell adhering on the hydrophilic scaffold surface was evidently better than that of the pristine PLA scaffold. The cells have spread and the pseudopodia grew and extended along the composite scaffold (Fig. 4B and D) compared to that of the pristine PLA fibers (Fig. 4A and C), particularly after five days of culture, indicating confluence growth of EA.hy926 cell. Nuttelman et al. [42] reported that the inability of cells to attach to tissue scaffold material, in general, relates to their hydrophilicity, leading to minimal adsorption of cell adhesion proteins on the scaffold surface. In addition, the abundant hydroxyl groups on PVA (see FTIR data, Fig. 2B) during the hydrothermal process can be readily modified to attach growth factors, adhesion proteins, or other molecules of biological importance. Previous in vitro and in vivo studies documented that hydrophilicity is an important factor when considering permeation of nutrients across the membrane and cell compatibility [3,43]. These studies reported that hydrophobic materials are poorly wetted by cell culture medium, resulting in limited cell attachment to the scaffold and poor transfer of nutrients and waste products across the membrane. From the above analysis, the hydroxyl functional groups of PVA are present on the surface of the composite scaffolds and are absent on the PLA scaffold. The appearance of PVA functional groups increased the hydrophilicity of electrospun PLA scaffolds. Indeed, coating of PVA to PLA nanofibers to fabricate a PVA/PLA composite scaffold using hydrothermal strategy proved to be an effective approach for improving the hydrophilicity and attachment properties of the scaffold and thus, overcome the main constrain of using PLA nanofibers in tissue engineering applications. The cell proliferation (cell density or number of cells) was tested by MTT-assay and the results are shown in Fig. 5. In general, the cells on all scaffolds proliferated with increasing culture time points, indicating a good cytocompatibility of both pristine and composite scaffolds. Nevertheless, at each time point interval there is a remarkable difference in cell proliferation among the composite PVA/PLA scaffold (p less than 0.005). The composite scaffold contributed the best proliferation result

A. Abdal-hay et al. / Materials Science and Engineering C 60 (2016) 143–150

149

Fig. 4. SEM observations of human endothelial cell adhesion and growth on pristine PLA membrane scaffold (A and C) and composite membrane scaffold containing 1 wt.% PVA (B and D) after culture for 3 (A and B) and 5 (C and D) days.

after culturing for 5 and 7 days, while the pristine PLA scaffold yielded relatively low proliferation particularly after 7 days of culture, indicating that the PLA nanofiber scaffold coated with PVA might have accelerated the proliferation and differentiation of EA.hy926 human cells. It can be observed that EA.hy926 cells proliferated on the scaffold displays a time-dependent behavior, and the composite scaffold possesses higher cell proliferation than that of the pristine PLA scaffold. The better biochemical interaction between cells and the hydrophilic membrane surface of the composite scaffold might provide good interaction of EA.hy926 cells with PLA fibers coated with a PVA thin layer compared to the pristine PLA fibers. The durability of the repeated use of the composite fibrous mat was evaluated using different strategies which were mentioned above. Our

results showed that the as-prepared scaffolds can be used repeatedly without a significant decrease in its efficiency. Therefore, we are expecting that these 3D scaffold materials will have promising applications in tissue engineering. We performed some tests for at least three times to confirm our data. We followed very systematic procedures to achieve and optimize our results by using several evaluation techniques supported by several characterization tools to confirm our findings. Overall, based on our results, the PVA/PLA composite coating fabricated by the hydrothermal approach can be considered as a promising, simple, cost-effective and smart approach for fabrication of hydrophilized scaffolds.

4. Conclusion

Fig. 5. MTT cytotoxicity test on different mats after 3, 5, and 7 days of culture. The viability of control cells was set at 100%, and the viability relative to the control was expressed. The data is reported as the mean ± standard deviation (n = 4 and p b 0.05).

A novel polymer composite nanofiber scaffold was designed by exploiting a simple hydrothermal approach without using any surface modifier. The designed nanofibers demonstrated improved hydrophilicity, mechanical properties, and EA.hy926 cell attachment behavior. The new scaffold promises new advancement in tissue engineering to meet the current clinical challenges and needs in this field. Our results confirmed that a PVA thin layer was successfully in situ deposited and coated onto each single pre-electrospun PLA nanofiber. The deposition of PVA throughout the hydrothermal technique could improve the hydrogen bonding interaction and induce the crystalline conformation of PLA, resulting in sharp improvement in the PLA ductility with significant enhancement in the tensile strength as well. The designed PLA scaffold showed elongation over 2.5 times higher than that of neat PLA. The conventional blending of PLA-based materials with PVA causes severe degradation of the fibrous scaffold, which can compromise scaffold strength and affect the biodegradation rate before the healing process occurs completely. Using our novel coating approach, we were able to avoid degradation of the fibrous scaffold. However, further studies are needed to investigate the in vivo biocompatibility and biodegradability of the modified PLA scaffold under physiological conditions.

150

A. Abdal-hay et al. / Materials Science and Engineering C 60 (2016) 143–150

Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.msec.2015.11.024. Acknowledgments The authors would like to extend their sincere appreciation to the Deanship of Scientific Research at King Saud University for funding this Research Group NO. (RG 1435-001). References [1] T.H. Qazi, R. Rai, A.R. Boccaccini, Biomaterials 35 (2014) 9068–9086. [2] A. Rogina, Appl. Surf. Sci. 296 (2014) 221–230. [3] A. Abdal-hay, F.A. Sheikh, J.K. Lim, Colloids Surf. B: Biointerfaces 102 (2013) 635–643. [4] V. Grumezescu, G. Socol, A.M. Grumezescu, A.M. Holban, A. Ficai, R. Truşcǎ, C. Bleotu, P.C. Balaure, R. Cristescu, M.C. Chifiriuc, Appl. Surf. Sci. 302 (2014) 262–267. [5] J. Li, X.L. Lu, Y.F. Zheng, Appl. Surf. Sci. 255 (2008) 494–497. [6] S. Sarapirom, L.D. Yu, D. Boonyawan, C. Chaiwong, Appl. Surf. Sci. 310 (2014) 42–50. [7] A. Jordá-Vilaplana, V. Fombuena, D. García-García, M.D. Samper, L. Sánchez-Nácher, Eur. Polym. J. 58 (2014) 23–33. [8] S.I. Jeong, B.S. Kim, S.W. Kang, J.H. Kwon, Y.M. Lee, S.H. Kim, Y.H. Kim, Biomaterials 25 (2004) 5939–5946. [9] A. Abdal-hay, H.R. Pant, J.K. Lim, Eur. Polym. J. 49 (2013) 1314–1321. [10] D. Papkov, Y. Zou, M.N. Andalib, A. Goponenko, S.Z.D. Cheng, Y.A. Dzenis, ACS Nano 7 (2013) 3324–3331. [11] D. Cossement, R. Gouttebaron, V. Cornet, P. Viville, M. Hecq, R. Lazzaroni, Appl. Surf. Sci. 252 (2006) 6636–6639. [12] D. Cohn, A. Hotovely Salomon, Biomaterials 26 (2005) 2297–2305. [13] C. Liu, C. Gong, Y. Pan, Y. Zhang, J. Wang, M. Huang, Y. Wang, K. Wang, M. Gou, M. Tu, Colloids Surf. A Physicochem. Eng. Asp. 302 (2007) 430–438. [14] O. Martin, L. Avérous, Polymer 42 (2001) 6209–6219. [15] P. Pi, W. Mu, G. Fei, Y. Deng, Appl. Surf. Sci. 273 (2013) 184–191. [16] G. Pitarresi, C. Fiorica, F.S. Palumbo, F. Calascibetta, G. Giammona, J. Biomed. Mater. Res. Part A 100A (2012) 1565–1572.

[17] A. Abdal-hay, A. Salam Hamdy, Y. Morsi, K. Abdelrazek Khalil, J. Hyun Lim, Mater. Lett. 137 (2014) 378–381. [18] A. Abdal-hay, J. Lim, M. Shamshi Hassan, J.K. Lim, Chem. Eng. J. 228 (2013) 708–716. [19] A. Abdal-hay, N.A.M. Barakat, J.K. Lim, Colloids Surf. A Physicochem. Eng. Asp. 420 (2013) 37–45. [20] S. Kang, S.L. Hsu, H.D. Stidham, P.B. Smith, M.A. Leugers, X. Yang, Macromolecules 34 (2001) 4542–4548. [21] A. Abdal-hay, C.I. Kim, J.K. Lim, Ceram. Int. 40 (2014) 4995–5000. [22] C. Ribeiro, V. Sencadas, C.M. Costa, J.L.G. Ribelles, S. Lanceros-Méndez, Sci. Technol. Adv. Mater. 12 (2011) 015001. [23] E.A. Schmitt, D. Flanagan, R.J. Linhardt, Macromolecules 27 (1994) 743–748. [24] A.G. Mikos, N.A. Peppas, Polymer 30 (1989) 84–91. [25] Y. You, S.W. Lee, S.J. Lee, W.H. Park, Mater. Lett. 60 (2006) 1331–1333. [26] A. Asran, K. Razghandi, N. Aggarwal, G.H. Michler, T. Groth, Biomacromolecules 11 (2010) 3413–3421. [27] T. Kasuga, Y. Ota, M. Nogami, Y. Abe, Biomaterials 22 (2000) 19–23. [28] X. Shuai, Y. He, N. Asakawa, Y. Inoue, J. Appl. Polym. Sci. 81 (2001) 762–772. [29] B.-K. Chen, C.-H. Shen, S.-C. Chen, A.F. Chen, Polymer 51 (2010) 4667–4672. [30] G. Meneghello, D.J. Parker, B.J. Ainsworth, S.P. Perera, J.B. Chaudhuri, M.J. Ellis, P.A. De Bank, J. Membr. Sci. 344 (2009) 55–61. [31] H. Tsuji, H. Muramatsu, J. Appl. Polym. Sci. 81 (2001) 2151–2160. [32] J.M. Deitzel, J. Kleinmeyer, D. Harris, N.C. Beck Tan, Polymer 42 (2001) 261–272. [33] C.S. Leiggener, R. Curtis, A.A. Muller, D. Pfluger, S. Gogolewski, B.A. Rahn, Biomaterials 27 (2006) 202–207. [34] X. Zong, K. Kim, D. Fang, S. Ran, B.S. Hsiao, B. Chu, Polymer 43 (2002) 4403–4412. [35] X. Liu, M. Dever, N. Fair, R. Benson, J. Environ. Polym. Degrad. 5 (1997) 225–235. [36] G. Ma, D. Yang, D. Su, X. Mu, J.F. Kennedy, J. Nie, Polym. Adv. Technol. 21 (2010) 189–195. [37] A.B. Norton, R.D. Hancocks, L.M. Grover, Food Hydrocoll. 42 (Part 3) (2014) 373–377. [38] D. Yang, Y. Li, J. Nie, Carbohydr. Polym. 69 (2007) 538–543. [39] R.A. Franco, A. Sadiasa, B.-T. Lee, Polym. Adv. Technol. 25 (2014) 55–65. [40] Y. Jung, H. Kim, D. Lee, S. Park, M. Khil, Macromol. Res. 13 (2005) 385–390. [41] D.R. Nisbet, A.E. Rodda, D.I. Finkelstein, M.K. Horne, J.S. Forsythe, W. Shen, Colloids Surf. B: Biointerfaces 71 (2009) 1–12. [42] C.R. Nuttelman, S.M. Henry, K.S. Anseth, Biomaterials 23 (2002) 3617–3626. [43] S.H. Oh, S.G. Kang, E.S. Kim, S.H. Cho, J.H. Lee, Biomaterials 24 (2003) 4011–4021.