Novel poly(L-lactide-co-caprolactone)/gelatin porous scaffolds for use in articular cartilage tissue engineering: comparison of electrospinning and wet spinning processing methods.
Nowadays, the most widely used cell-based surgical procedure for the repair of articular cartilage involves a technique known as autologous chondrocyte implantation (ACI) [1, 2]. In ACI, a patient's own cartilage cells (chondrocytes) are extracted from the cartilage matrix and then cultured in vitro for typically 4-6 weeks until there are enough cells to be implanted in the damaged area or defect site of the patient's cartilage. This implantation can either be in combination with a biomembrane (periosteal cover) or preseeded in a porous biodegradable scaffold. It is the latter of these two methods which is the subject of this research.
Although the use of porous scaffolds has been known for some time, there is still an urgent need for scaffolds with improved material and handling properties. In addition to being porous, biocompatible and biodegradable, pore size and internal volume need to be of suitable dimensions to allow for cell infiltration. The scaffold also needs to have the appropriate mechanical and handling properties to enable it to be molded by the surgeon to fit the size and shape of the defect. Consequently, scaffolds tend to be elastic, sponge-like materials but with a strength and toughness that can withstand the stress loads that are likely to be experienced in the human body.
Polymers for use in this type of application can be either natural or synthetic or blends of both. Examples of natural polymers include collagen, chitosan, keratin, gelatin, silk fibroin, and hyaluronan [3-10], while the synthetic polymers include polyglycolide, polylactide, polycaprolactone, polyhydroxybutyrate and their copolymers and blends [10-29]. Poly(L-lactide-co-caprolactone), P(LL-CL), has attracted particular attention for ACI as it exhibits rubber-like elasticity within a certain composition range which can be tailored to meet specific requirements. However, P(LL-CL) by itself is a relatively hydrophobic material which limits its cell interaction. Therefore, to promote cell attachment and proliferation, P(LL-CL) scaffolds tend to be subjected to some form of modification to make them more hydrophilic and cell interactive.
This work now describes the molecular design, preparation and fabrication of some novel porous scaffolds from solution blends of P(LL-CL) and gelatin in a common solvent (2,2,2-trifluoroethanol). Solution blending ensures that the two components are mixed together at the molecular level so that, after solvent evaporation or coprecipitation, the gelatin content in the solid scaffold is distributed throughout the bulk of the material. Two methods of scaffold fabrication are compared, namely electrospinning and wet spinning, resulting in nanofibrous and microfibrous nonwoven meshes, respectively, each with its own characteristic pore structure. As far as we are aware, a comparison of P(LL-CL)/gelatin scaffolds prepared by these two methods has not been previously reported. Although some wet spinning studies have been reported [30, 31], wet spinning as an alternative to the more widely used electrospinning, porogen leaching and phase separation methods has received scant attention in the literature for scaffold preparation. In this article, the electrospinning and wet spinning methods are compared and the potential of these scaffolds for use in articular cartilage tissue engineering is discussed.
L-Lactide monomer was synthesized from L-lactic acid (Grand Chemical Far East Ltd., 88%) by well-established procedures and purified by repeated recrystallization from ethyl acetate . After drying to constant weight in a vacuum oven at 55[degrees]C for 24 h, pure L-lactide was obtained as a white, needlelike crystalline solid with a chemical purity of [is greater than or equal to] 99.5% (from DSC purity analysis). [epsilon]-Caprolactone (Acros, 99%) was purified by vacuum distillation over calcium hydride (b.pt. 70[degrees]C/5 mm Hg). Tin(II) octoate (Sigma-Aldrich, 95%) was purified by heating with stirring under vacuum to remove the octanoic acid impurity. Gelatin powder (Merck), 2,2,2-trifluoroethanol (Sigma-Aldrich, [is greater than or equal to] 99%) and absolute ethanol (VWR, AnalaR Normapur[R], >99.9%) were used as supplied without further purification.
Copolymer composition was determined by proton nuclear magnetic resonance ([sup.1]H-NMR) spectroscopy using a Bruker Avance NMR Spectrometer operating at a field frequency of 400 MHz. Deuterated chloroform (CD[Cl.sub.3]) was used as the solvent with tetramethylsilane (TMS) as the internal standard.
Thermal analysis for determining copolymer temperature transitions by differential scanning calorimetry (DSC) was performed using a Perkin-Elmer DSC7 Differential Scanning Calorimeter (-20 [right arrow] 200[degrees]C, heating rate = 10[degrees]C/min) equipped with an Intracooler subambient accessory.
Copolymer molecular weight determination by gel permeation chromatography (GPC) was performed using a Waters Alliance e2695 Separations Module High-Performance Liquid Chromatograph (tetrahydrofuran solvent, 35[degrees]C, flow-rate 1 ml/ min) calibrated with narrow molecular weight distribution polystyrene standards.
For microscopic analysis of the scaffolds, specimens were mounted on aluminum stubs with a conductive carbon tape, gold-coated and then imaged using a Hitachi S-3400N Scanning Electron Microscope (SEM) operating at a 20 kV accelerating voltage and 10 mm working distance.
Poly(l-lactide-co-caprolactone), P(LL-CL), copolymers were synthesized via the bulk ring-opening copolymerization of L-lactide (LL) and caprolactone (CL) at 120[degrees]C for 72 h under a dry nitrogen atmosphere. Initial comonomer feeds of LL:CL = 50:50, 75:25 and 80:20 mol% were used together with 0.02 mol% tin(II) octoate, Sn[(Oct).sub.2], as the initiator. The copolymers were purified by cutting into small pieces before heating under vacuum at 60[degrees]C to constant weight to remove any residual monomers. Percent yields were around 95% which was consistent with the expected levels of residual monomers from theoretical equilibrium monomer concentration calculations. The comonomer and copolymer structures are shown in Fig. 1.
Electrospinning of the P(LL-CL) copolymers and the P(LL-CL)/gelatin blends was performed from solution in 2,2,2-trifluroethanol (TFE) as a common solvent at room temperature (25 30[degrees]C). Even though TFE has a boiling point (b.pt.) of 74[degrees]C, it still has a volatility suitable for use as a solvent in electrospinning. Blend solutions were prepared by mixing together 10% w/ v solutions of P(LL-CL) and gelatin in TFE in various ratios in order to give P(LL-CL):gelatin dry blend compositions ranging from 70:30 to 95:5 wt%. The processing conditions that were found to give the best results in terms of the uniformity of fiber diameter and pore size with little or no bead formation were:
Solution concentration = 10% w/v
Power supply = 15 kV
Syringe needle diameter = 0.8 mm
Needle tip to receiver plate distance = 15 cm
Under these conditions, nonwoven mats of nanofibers could be gradually built up to a thickness of 4 mm over a period of about 24 h. Cylindrical plugs of 7 mm diameter, similar in size and shape to those that might be used in articular cartilage repair, were then cut from these mats using a cork borer for SEM analysis and property testing.
For wet spinning, a 10% w/v solution of P(LL-CL) in TFE was mixed with either a 10 or 5% w/v gelatin solution in TFE to give P(LL-CL):gelatin weight % ratios over the same composition range of 70:30 to 95:5 as for electrospinning. For precipitation in the form of a nonwoven mesh, each solution mixture was slowly injected through a syringe needle of diameter 1.2 mm into a fivefold excess of chilled ethanol as nonsolvent. After leaving for a period of 2 h to ensure complete precipitation. the supernatant liquid was poured off and the nonwoven mat dried to constant weight in a vacuum oven at 60[degrees]C for 24 h. When dry, the mat was compressed in a cylindrical mold so that test specimens of similar dimensions to those from electrospinning could be cut for analysis and testing.
It was found that both the copolymer composition and the blend composition had marked effects on the mechanical properties of the wet-spun mesh. First, as the CL content in the copolymer increased, the blend became less brittle and more malleable. Second, as the gelatin content in the blend decreased, the mesh became denser and more coherent. Thus, the best mechanical and handling properties for cutting test specimens and molding them into different shapes were observed for the P(LL-CL)50:50/gelatin 95:5 blend.
Water Absorption Measurements
To determine their water absorbability, the cylindrical-shaped scaffolds (7 mm diameter X 4 mm thickness) were immersed in distilled water at 37[degrees]C, removed at various time intervals over a period of 2 h, patted dry with a lint-free cloth, and the increase in weight due to water absorption measured by weighing. From the increase in weight, the water absorption (WA) was then calculated as a percentage from the following equation:
Water Absorption = WA = [([W.sub.w] - [W.sub.d])/[W.sub.d]] X 100%
where [W.sub.w] = wet weight, and [W.sub.d] = dry weight. Determinations were performed in triplicate with fresh samples being used for each sampling time.
Cytotoxicity testing was performed by the direct contact method in which the scaffolds (7 mm diameter X 4 mm thickness discs), presterilized with ethylene oxide gas, were seeded with L929 mouse fibroblast cells (6 X [10.sup.4] cells/scaffold) in Dulbecco's Modified Eagle's Medium supplemented with 10% (v/v) fetal calf serum and 1% penicillin/streptomycin. The cell-seeded scaffolds in a polystyrene cell culture plate were then incubated at 37[degrees]C under a 5% C[O.sub.2] atmosphere for 48 h after which the cells were stained with 0.01% neutral red in phosphate buffer saline (PBS) for membrane integrity. Cell morphology and the toxic zone were evaluated by examination under an inverted phase contrast light microscope. The polystyrene culture plate and natural rubber containing carbon black were used as negative and positive controls, respectively. All cytotoxicity tests were performed in triplicate.
RESULTS AND DISCUSSION
Copolymer composition was determined by high-resolution [sup.1]H-NMR spectroscopy. The [sup.1]H-NMR spectrum of the P(LL-CL)50:50 copolymer is shown in Fig. 2 as an example. The copolymer composition was calculated from the peak area integrations of the LL methine proton peak (b) at [delta] 5.1-5.2 and the CL [epsilon]-methylene peak (c) at [delta] 4.1-4.2. As all of the copolymers were obtained in near-quantitative yields ([approximately equal to]95%), their compositions were very similar to their initial comonomer feeds, as shown in Table 1.
Regarding their chain microstructures, P(LL-CL) copolymers synthesized under these conditions are statistical copolymers in which the monomer sequencing is determined mainly by the respective monomer reactivity ratios. However, as the LL ring is considerably more reactive than the CL ring, the monomer sequencing tends to be tapered and hence partially blocky rather than purely statistical. Earlier work reported reactivity ratios (r) of [r.sub.LL] = 34.7 and [r.sub.CL] = 0.24 for LL-CL bulk copolymerization at 130[degrees]C using Sn[(Oct).sub.2] as the initiator , Conversely, this blocky character is offset to a certain extent by the transesterification reactions which occur in the melt during synthesis and which tend to randomize the monomer sequencing. Monomer sequencing in P(LL-CL) copolymers has been studied extensively by [sup.13]C-NMR [34-37],
Copolymer number-average molecular weights from GPC were in the range of [M.sub.n] = 5.84-7.60 X [10.sup.4] g/mol, as listed in Table 1, with normal unimodal molecular weight distributions (PDI = 1.80-2.04), as shown in Fig. 3 for the P(LL-CL)80:20 copolymer as an example. These molecular weights were considered to be suitable for fiber spinning which for aliphatic polyesters typically requires [M.sub.n] > 30,000 for the mechanical properties to be fully developed, although this also depends on the degree of chain orientation . However, unlike in melt spinning where chain orientation can be introduced by hot-drawing, the electrospinning and wet spinning processes introduce relatively little chain orientation and so the mechanical properties depend almost entirely on the molecular weight.
Thermal analysis by DSC enabled the copolymers' glass transition temperatures, [T.sub.g], cold crystallization temperatures, [T.sub.c], and melting temperatures, [T.sub.m], to be determined. From the DSC thermograms in Fig. 4 and the results in Table 1, the P(LL-CL)50:50 copolymer is seen to be amorphous with no observable [T.sub.c] or [T.sub.m] transitions. In addition, its [T.sub.g] transition is barely visible in its DSC thermogram and was more clearly observed at -10[degrees]C by dynamic mechanical analysis (DMA) instead. In contrast, the 75:25 and 80:20 copolymers were both semicrystallizable materials with clear [T.sub.g], [T.sub.c], and [T.sub.m] transitions in their DSC curves.
For electrospinning, only the P(LL-CL)75:25 and P(LL-CL)80:20 copolymers were used. This was because previous work had shown that the microsized pores of electrospun P(LL-CL)50:50 scaffolds tended to gradually merge together during storage at ambient temperature due to the copolymer's subzero [T.sub.g] , Consequently, a [T.sub.g] of at least 20[degrees]C was considered necessary for electrospun pore stability during storage before use. Examples of the electrospun P(LL-CL) and P(LL-CL)/gelatin scaffolds of various compositions, as processed from 10% w/v blend solutions in TFE as the solvent, are shown in the SEM images in Figs. 5 and 6.
From the SEM images in Fig. 5, both the P(LL-CL)75:25 and P(LL-CL)80:20 copolymers exhibit good fiber-forming properties although the 75:25 copolymer shows evidence of some bead formation. This could have been due to its slightly lower [T.sub.g] and/or higher molecular weight (and hence higher solution viscosity) than the 80:20 copolymer (Table 1). The SEM images in Fig. 5 also suggest that increasing the CL content of the copolymer leads to decreases in both the fiber and pore diameters (Table 2).
As the P(LL-CL)80:20 copolymer gave the better quality electrospun scaffolds, it was therefore chosen for use in blending with gelatin. The SEM images in Fig. 6 compare the electrospun blends with gelatin contents ranging from 0 to 30% by weight. A gelatin content of 30% was considered to be the upper limit above which the scaffold would start to become too hydrophilic and biodegrade (absorb) in the human body too quickly. Despite gelatin's incompatibility with P(LL-CL), the blends all gave good quality scaffolds with the addition of gelatin tending to give a greater proportion of smaller fiber and pore diameters (Fig. 6). However, the differences in the fiber and pore diameters in Table 2 are relatively small and so it can be concluded that gelatin, up to 30% by weight, does not greatly affect the electrospun scaffold-forming properties of the main P(LL-CL) component.
Although these electrospun scaffolds are attractive in terms of their uniformity and through-thickness interconnecting pore structure, they still need to have a pore size which is large enough for cells to be able to penetrate and migrate inside but small enough for the surface area to be adequate for efficient cell binding and cell growth , A suitable pore size is usually considered to be of the order of tens of microns (20-100 pm) depending on the particular tissue engineering application, although pore sizes in excess of 100 pm may be required for bone cell regeneration. With respect to articular cartilage cells, previous studies have suggested that the optimum pore size for cell ingrowth is in the range of 20-60 [micro]m [23, 24]. A large enough pore size also serves to facilitate the effective exchange of nutrients and waste products in the implant.
Consequently, these electrospun scaffolds in which the vast majority of their pore diameters are less than 10 pm would need to swell quite considerably in the culture medium before cell seeding. Indeed, this was the main reason for blending the P(LL-CL) copolymer with gelatin to increase its hydrophilicity. Apart from increasing its ability to absorb and retain water, increasing the hydrophilicity of the scaffold can also affect its surface tension which in turn can have an impact on cell adhesion [41, 42], Despite their attractiveness in terms of their uniformity of fiber and pore structure, the small pore sizes of electrospun scaffolds remains a disadvantage in applications where cell ingrowth rather than just surface attachment is an essential requirement.
As an alternative method to electrospinning which could give much larger-sized pores, wet spinning was used to produce a microporous nonwoven mesh of fibers. As the pores were much larger, pore merging in the case of the P(LL-CL)50:50 copolymer was less of a problem in wet spinning and so all three copolymer compositions (50:50, 75:25, and 80:20) gave stable pore structures. However, when comparing their mechanical properties, in particular their shape-forming ability, the P(LL-CL)50:50 copolymer was found to be the most malleable due to its flexibility arising from its subzero [T.sub.g] (Table 1). A shape-forming ability is an important property requirement for enabling the cell-impregnated scaffold to be hand-molded and compressed by the surgeon into the often irregular shape of the defect site of a patient's cartilage.
SEM images of the wet-spun P(LL-CL)50:50 copolymer are shown in Fig. 7. In marked contrast to the electrospun copolymer in Fig. 5, wet spinning produces an irregular fibrous mesh with pore sizes ranging from less than 10 to hundreds of microns in diameter. This wide distribution of pore sizes makes the wet-spun scaffolds more porous and sponge-like, both of which are useful properties for facilitating water absorption and cell infiltration. However, when the P(LL-CL)50:50 copolymer was blended with gelatin, it was found that the wet-spun scaffold rapidly lost its sponge-like properties and became increasingly brittle and nonuniform as the amount of gelatin increased over the 0-30% by weight range. Consequently, a low gelatin content of 5% by weight gave the best balance of properties between increasing the scaffold's hydrophilicity while still preserving its shape-forming ability. SEM images of the wet-spun P(LL-CL)50:50/gelatin 95:5 (wt%) scaffold are also shown in Fig. 7 alongside those of the copolymer alone. As the amount of gelatin was relatively small, its effect on the pore structure of the scaffold was also relatively small.
Water Absorption Measurements
The ability of a porous scaffold to absorb water in a cell culture medium in which it is immersed is an essential prerequisite for cell infiltration in the ACI technique and is dependent upon a range of factors:
* the hydrophilicity of the material of which the scaffold is made
* the capillarity of the scaffold which in turn depends on pore size
* the ability of the scaffold to expand (swell) as water enters the bulk
* the temperature of the immersion medium and its ionic strength
In order for the seeded cells to infiltrate the porous interior rather than just coat the surface, the scaffold must be sufficiently water-absorbent and able to swell without breaking apart (loss of mass integrity). Therefore, a certain degree of sponge-like elasticity together with a through-thickness connectivity are required properties for an effective scaffold.
In this work, it was found that, when immersed in water, the electrospun scaffolds both with and without gelatin fell short of these requirements based on the observations that (a) water absorption was limited to less than 5% by the small pore size; (b) elasticity was limited by the copolymers' high-LL contents; and (c) the scaffolds tended to break apart as water was absorbed due to their loosely connected layered structure. In contrast, the wet-spun scaffolds with their much larger pores and greater elasticity were able to absorb about 10% water without gelatin and more than 50% water with gelatin. They were also able to swell intact without any loss of mass integrity over the full 2-h period of immersion.
Typical examples of the water absorption--time profiles for the wet-spun scaffolds are shown in Fig. 8 and illustrate clearly the effect of gelatin in increasing the scaffolds' hydrophilicity. The profiles both show a rapid initial increase in water absorption which then slows down with time toward an equilibrium value at which the positive osmotic force drawing water in is balanced by the negative restraining force resisting further expansion. It was also significant to note that the P(LL-CL)50:50/gelatin 95:5 scaffold, even at its equilibrium water content, was still malleable enough to be hand-shaped to fit cavities of different shapes and sizes such as would be needed in actual surgery. The various properties of the electrospun and wet-spun scaffolds are compared in Table 3.
Even though P(LL-CL) copolymers and gelatin have already been widely used in biomedical applications and have been shown to be biocompatible, it is still a necessary part of any research project such as this to test for possible cytotoxicity. The main concern here was the possibility that there could be trace amounts of residual LL and CL monomers, Sn[(Oct).sub.2] initiator and TFE solvent present in the P(LL-CL) copolymers. Such contamination would be likely to be toxic.
After staining with 0.01% neutral red in phosphate buffer saline and imaging using a phase-contrast light microscope, it was found that well-stained cells with a normal rounded morphology, indicating that they were alive, were similarly observed when cultured both in the polystyrene culture plate alone (as negative control) and in contact with each of the scaffold samples tested both from electrospinning and wet spinning. In contrast, dead unstained cells were detected when they were incubated with the positive control due to loss of integrity of the cell surface and inability of the lysosomal membranes to accumulate and retain the intracellular neutral red dye. A typical example is shown in Fig. 9 for the P(LL-CL)50:50/gelatin 95:5 scaffold from wet spinning. These results were therefore able to demonstrate quite convincingly that the scaffolds were noncytotoxic.
Further confirmation of the noncytotoxicity of the scaffolds was provided by a cell viability assay (Alamar Blue assay) in which the P(LL-CL)50:50/gelatin 95:5 wet-spun scaffolds (n = 4) were seeded with primary human articular chondrocytes at a density of 5 X 105 cells/scaffold and incubated at 37[degrees]C in a humidified atmosphere of 5% C[O.sub.2] in air. Cell viability after 72 h exposure to the scaffolds was found to be approximately 77% which exceeds the cytotoxicity cutoff level (70%) established by ISO-10993-513. In the wider context of this work, these combined noncytotoxic test results are important insofar that they also confirm that the synthesis and purification procedures used for making the P(LL-CL) copolymers were appropriate for the production of biomedical-grade materials.
In addition to its biocompatibility, each polymeric device to be used in a biomedical application has its own unique set of property requirements. This is especially true in the case of an absorbable device where it is not only the polymer itself but also its degradation products which need to be biocompatible. These property requirements are determined by the end-users of the device--in this case the medical team--and take into account not only the device's in vivo performance but also the ease and convenience with which it can be used throughout the ACI procedure.
In this work, the main objective was to develop a porous polymeric scaffold which could be seeded with cells, was malleable enough to be hand-molded into the required shape to fit snugly into a cartilage defect site, and which would biodegrade at an appropriate rate. This rate should be such that the gradual breakdown and disappearance of the scaffold allows sufficient time (typically 6-12 months) for cell growth to fill the defect space. Based on the foregoing results, it is concluded that this objective has been met more closely by the wet-spun scaffolds than the electrospun scaffolds. The main reasons for this are (a) the much larger pore sizes and (b) the ability to use a P(LL-CL)50:50 copolymer which is within the copolymer's rubber-like composition range which makes it both flexible and pliable. To offset the copolymer's inherent hydrophobicity, its water affinity and hence absorbability could be significantly enhanced by the addition of as little as 5% gelatin without any adverse effects on the pore size and mechanical properties.
In conclusion, it is therefore considered that the wet-spun P(LL-CL)50:50/gelatin 95:5 scaffold shows sufficient potential to warrant further development as an absorbable scaffold for use in articular cartilage tissue engineering. It has been shown to be noncytotoxic and its properties in general appear to be well-suited to the requirements of the ACI procedure. Furthermore, P(LL-CL) copolymers are known to be absorbed slowly in the human body over an appropriate timescale of up to one year or more depending on composition. The higher the CL content, the slower the rate of absorption. As mentioned in the "Introduction," there is still a pressing need for scaffolds with improved properties. The ultimate objective of this research is to be able to meet this need by working in close collaboration with tissue engineers and orthopedic surgeons in hospitals here in Thailand.
The authors wish to thank the Department of Chemistry, Faculty of Science, Chiang Mai University for providing the facilities used in this research and the National Metal and Materials Technology Center (MTEC), Thailand, and the National Research University Project under Thailand's Office of the Higher Education Commission for financial support. The authors also thank Dr. Peraphan Pothacharoen of the Center of Excellence for Tissue Engineering and Stem Cells, and Dr. Dumnoensun Pruksakorn of the Musculoskeletal Research Laboratory, Department of Orthopedics, both of whom are affiliated to the Faculty of Medicine, Chiang Mai University, for their practical suggestions and medical advice.
[1.] J.D. Harris, R.A. Siston, X. Pan, and D.C. Flanigan, J. Bone Joint Surg. Am., 92, 2220 (2010).
[2.] D.G. Jones, and L. Peterson, J. Bone Joint Surg. Am., 88, 2501 (2006).
[3.] X. Wu, Y. Liu, X. Li, P. Wen, Y. Zhang, Y. Long, X. Wang, Y. Guo, F. Xing, and J. Gao, Acta Biomater., 6, 1167 (2010).
[4.] M. Kabiri, S.H. Emami, M. Rafinia, and M. Tahriri, Curr. Appl. Phys., 11, 457 (2011).
[5.] D. Daranarong. B. Thapsukhon, N.S. Wanandy, R. Molloy. W. Punyodom, and L.J.R. Foster, Polym. Int., 63, 1254 (2014).
[6.] C. Gartner and B.L. Lopez, Polym. Eng. Sci., 52. 2269 (2012).
[7.] W. Zhou, Y. Feng, J. Yang, J. Fan, J. Lv, L. Zhang, J. Guo, X. Ren, and W. Zhang, J. Mater. Sci. Mater. Med., 26, 56 (2015).
[8.] C.-L. Ko, Y.-C. Tien, J.-C. Wang, and W.-C. Chen, J. Mech. Behav. Biomed. Mater., 14, 227 (2012).
[9.] M. Salehi, M.N. Nosar, A. Amani, M. Azami, S. Tavakol, and H. Ghanbari, Int. J. Polym. Mater., 64, 675 (2015).
[10.] S. Torres-Giner, R. Perez-Masia, and J.M. Lagaron, Polym. Eng. Sci., 56, 500 (2016).
[11.] J.B. Jonnalagadda, I.V. Rivero, and J.S. Dertien, J. Biomater. Sci. Polym. Ed., 26, 401 (2015).
[12.] Z. Zhou, Y. Zhou, Y. Chen, H. Nie, Y. Wang, F. Li, and Y. Zheng, Appl. Surf. Sci., 258, 1670 (2011).
[13.] X. Wang, L. Sang, Z. Wei. L. Zhai, H. Wang, K. Song, and M. Qi, Polym. Eng. Sci., 54, 2902 (2014).
[14.] G.-Y. Liao, L. Chen, X.-Y. Zeng, X.-P. Zhou, X.-L. Xie, E.-J. Peng, Z.-Q. Ye, and Y.-W. Mai, J. Appl. Polym. Sci., 120, 2154 (2011).
[15.] Y. Jung, S.H. Kim. H.J. You, S.-H. Kim, Y.H. Kim, and B.G. Min, J. Biomater. Sci. Polym. Ed., 19, 1073 (2008).
[16.] P. Torricelli. M. Gioffre, A. Fiorani, S. Panzavolta. C. Gualandi, M. Fini, M.L. Focarete, and A. Bigi, Mater. Sci. Eng. C Mater. Biol. Appl., 36, 130 (2014).
[17.] Z. Chen, L. Cao, L. Wang, H. Zhu, and H. Jiang, J. Appl. Polym. Sci., 127, 4225 (2013).
[18.] Y. Wan, X. Cao, Q. Wu, S. Zheng, and S. Wang, Polym. Adv. Technol., 19, 114 (2008).
[19.] J.K. Perron, H.E. Naguib, J. Daka, and A. Chawla, Polym. Eng. Sci., 49, 2062 (2009).
[20.] Y. Wan, Y. Fang, H. Wu, and X. Cao, J. Biomed. Polym. Res. A, 80A, 776 (2007).
[21.] M.T. Khorasani, S.A. Mirmohammadi, and S. Irani, Int. J. Polym. Mater., 60, 562 (2011).
[22.] G. Liu, Q. Tang, Y. Yu, J. Li, J. Luo, and M. Li, Polym. Adv. Technol.. 25, 1596 (2014).
[23.] P. Sarazin, X. Roy, and B.D. Favis, Biomaterials, 25, 5965 (2004).
[24.] W. Ji, F. Yang, J. Ma, M.J. Bouma. O.C. Boerman, Z. Chen, J.J.J.P. van den Beucken, and J.A. Jansen, Biomaterials, 34, 735 (2013).
[25.] M.C. Bottino, V. Thomas, and G.M. Janowski, Acta Biomater., 7, 216 (2011).
[26.] D. Gupta, J. Venugopal, S. Mitra, V.R. Giri Dev. and S. Ramakrishna, Biomaterials, 30, 2085 (2009).
[27.] J. Lee, G. Tae, Y.H. Kim, b, I.S. Park, S-H. Kim, and S.H. Kim, Biomaterials, 29, 1872 (2008).
[28.] A.R. Padalhin, N.T.B. Linh, Y.K. Min, and B.-T. Lee, J. Biomater. Sci. Polym. Ed., 25, 487 (2014).
[29.] N.G. Rim, J.H. Lee, S.I. Jeong, B.K. Lee, C.H. Kim, and H. Shin, Macromol. Biosci., 9, 795 (2009).
[30.] D. Puppi. A.M. Piras, F. Chiellini, E. Chiellini, A. Martins, I.B. Leonor, N. Neves, and R. Reis, J. Tissue Eng. Regen. Med., 5, 253 (2011).
[31.] B. Xu, L. Du, J. Zhang, M. Zhu, S. Ji, Y. Zhang, D. Kong, X. Ma, Q. Yang, and L. Wang, RSC Adv., 5, 42705 (2015).
[32.] S.-H. Hyon, K. Jamshidi, and Y. Ikada, Biomaterials, 18, 1503 (1997).
[33.] D.W. Grijpma, and A.J. Pennings, Polym. Bull., 25, 335 (1991).
[34.] K. Nalampang, R. Molloy, and W. Punyodom, Polym. Adv. Technol., 18, 240 (2007).
[35.] H.R. Kricheldorf, and I. Kreiser, J. Macromol. Sci. A, 24, 1345 (1987).
[36.] J. Kasperczyk, and M. Bero, Makromol. Chem., 192, 1777 (1991).
[37.] J. Kasperczyk, and M. Bero, Makromol. Chem., 194, 913 (1993).
[38.] J. Siripitayananon, R. Molloy, S. Bunkird, A. Kleawkla, R. Panjakha, and P. Chooprayoon, Int. Polym. Proc., 23, 161 (2008).
[39.] B. Thapsukhon, N. Thadavirul, P. Supaphol, P. Meepowpan, R. Molloy, and W. Punyodom, J. Appl. Polym. Sci., 130, 4357 (2013).
[40.] L. Liu, Y. Wang, S. Guo, Z. Wang, and W. Wang, J. Biomed. Mater. Res. B, 100B, 956 (2012).
[41.] M. Lampin, R. Warocquier-Clerout, C. Legris, M. Degrange, and M.F. Sigot-Luizard, J. Biomed. Mater. Res. A, 36, 99 (1997).
[42.] G. Chang, D.R. Absolom, A.B. Strong, G.D. Stubley, and W. Zingg, J. Biomed. Mater. Res. A, 22, 13 (1988).
Wichaya Kalaithong, (1) Robert Molloy, (1,2) Tharinee Theerathanagorn, (3) Wanida Janvikul (3)
(1) Biomedical Polymers Technology Unit, Department of Chemistry, Faculty of Science, Chiang Mai University, Chiang Mai, Thailand 50200
(2) Materials Science Research Center, Faculty of Science, Chiang Mai University, Chiang Mai, Thailand 50200
(3) National Metal and Materials Technology Center, National Science and Technology Development Agency, Thailand Science Park, Pathum Thani, Thailand 12120
Correspondence to: R. Molloy; e-mail: Robert.firstname.lastname@example.org
Caption: FIG. 1. Ring-opening copolymerization of L-lactide and [epsilon]-caprolactone.
Caption: FIG. 2. 400 MHz [sup.1]H-NMR spectrum of the P(LL-CL)50:50 copolymer recorded in CD[Cl.sub.3] as solvent.
Caption: FIG. 3. GPC chromatogram of the P(LL-CL)80:20 copolymer.
Caption: FIG. 4. DSC thermograms of the (a) P(LL-CL)50:50, (b) P(LL-CL)75:25 and (c) P(LL-CL)80:20 copolymers. (Second scans; heating rate = 10[degrees]C/min).
Caption: FIG. 5. SEM images of the electrospun P(LL-CL)80:20 and 75:25 scaffolds at various magnifications.
Caption: FIG. 6. SEM images of the electrospun P(LL-CL)80:20/gelatin scaffolds at various weight % compositions.
Caption: FIG. 7. SEM images of the wet-spun P(LL-CL)50:50 and P(LL- CL)50:50/gelatin 95:5 scaffolds at various magnifications.
Caption: FIG. 8. Comparison of the water absorption--time profiles for the (a) P(LL-CL)50:50 and (b) P(LL-CL)50:5()/gelatin 95:5 wet-spun scaffolds showing the effect of gelatin.
Caption: FIG. 9. Optical micrographs of L929 mouse fibroblast cells after direct contact for 48 h with (top) P(LL-CL)50:50/gelatin 95:5 wet-spun scaffold, (below left) polystyrene culture plate (negative control) and (below right) natural rubber containing carbon black (positive control); magnification X160. [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 1. P(LL-CL) copolymer compositions, molecular weights and temperature transitions from [sup.1]H-NMR, GPC and DSC, respectively. [sup.1]H-NMR P(LL-CL) copolymer code (a) LL:CL (b) (mol%) P(LL-CL) 50:50 51:49 P(LL-CL) 75:25 74:26 P(LL-CL) 80:20 80:20 GPC P(LL-CL) copolymer code (a) [M.sub.n] X [10.sup.4] P(LL-CL) 50:50 7.15 P(LL-CL) 75:25 7.60 P(LL-CL) 80:20 5.84 GPC P(LL-CL) copolymer code (a) [M.sub.w] X [10.sup.-5] P(LL-CL) 50:50 1.46 P(LL-CL) 75:25 1.42 P(LL-CL) 80:20 1.05 GPC DSC (d) P(LL-CL) copolymer code (a) PDI (c) [T.sub.g] ([degrees]C) P(LL-CL) 50:50 2.04 -10 (e) P(LL-CL) 75:25 1.87 22 P(LL-CL) 80:20 1.80 28 DSC (d) P(LL-CL) copolymer code (a) [T.sub.c] ([degrees]C) P(LL-CL) 50:50 -- (f) P(LL-CL) 75:25 98 P(LL-CL) 80:20 102 DSC (d) P(LL-CL) copolymer code (a) [T.sub.m] ([degrees]C) P(LL-CL) 50:50 -- (f) P(LL-CL) 75:25 154 P(LL-CL) 80:20 161 (a) LL:CL ratios are the comonomer feeds (mol%) used in synthesis. (b) LL:CL ratios are the actual copolymer compositions (mol%) from [sup.1]H-NMR. (c) PDI = polydispersity index = [M.sub.w]/[M.sub.n] (where [M.sub.w] and [M.sub.n] have units of g/mol). (d) DSC data obtained from second heating scan. (e) [T.sub.g] determined by dynamic mechanical analysis (DMA): temperature scan from -80 to 100[degrees]C at 2[degrees]C/min, frequency 1 Hz, on thin film sample in tension mode. (f) 50:50 copolymer was completely amorphous with no observed [T.sub.c] or [T.sub.m] transitions. TABLE 2. Fiber and pore diameter ranges ([micro]m) and fiber surface areas (%) for the electrospun P(LL-CL) 80:20 and 75 Copolymer/blend sample code Fiber diameter range ([micro]m) (c) P(LL-CL) 80:20 (a) 0.8-1.2 P(LL-CL) 75:25 (a) 0.3-0.5 P(LL-CL)80:20/Gelatin 100:0 (b) 0.8-1.2 P(LL-CL)80:20/Gelatin 90:10 (b) 0.5-0.8 P(LL-CL)80:20/Gelatin 80:20 (b) 0.5-1.0 P(LL-CL)80:20/Gelatin 70:30 (b) 0.5-1.0 Copolymer/blend sample code Pore diameter range ([micro]m) (c) P(LL-CL) 80:20 (a) 6.0-11.0 P(LL-CL) 75:25 (a) 3.0-6.0 P(LL-CL)80:20/Gelatin 100:0 (b) 6.0-11.0 P(LL-CL)80:20/Gelatin 90:10 (b) 3.0-9.5 P(LL-CL)80:20/Gelatin 80:20 (b) 4.0-10.0 P(LL-CL)80:20/Gelatin 70:30 (b) 3.0-10.0 Copolymer/blend sample code Fiber surface area (%) (c,d) P(LL-CL) 80:20 (a) 21 P(LL-CL) 75:25 (a) 29 P(LL-CL)80:20/Gelatin 100:0 (b) 21 P(LL-CL)80:20/Gelatin 90:10 (b) 25 P(LL-CL)80:20/Gelatin 80:20 (b) 24 P(LL-CL)80:20/Gelatin 70:30 (b) 26 (a) LL-CL ratios are the comonomer feeds (mol%) used in synthesis. (b) P(LL-CL)80:20/gelatin ratios are the blend compositions (wt%) used in electrospinning. (c) Estimated using image analysis software (ImageJ, National Institutes of Health, USA). (d) Expressed as a percentage (%) of the total surface area. TABLE 3. Comparison of the various properties of the porous scaffolds prepared by electrospinning and wet spinning relevant to their use in the ACI procedure. Electrospinning Wet spinning Regular fiber and pore Irregular fiber and pore structures structures Narrow distribution of fiber Broad distribution of fiber and pore diameters and pore diameters Small pore sizes < 10 Large pore sizes > 100 [micro]m diameter [micro]m diameter Pore stability during storage Pore stability during storage requires [T.sub.g] > ambient not limited by [T.sub.g] Low capillarity and water Higher capillarity and water absorption absorption Densely packed with low Sponge-like and malleable for pliability for molding shape-forming Tended to disintegrate when Remained intact when immersed immersed in water in water Requires specialized Requires only simple apparatus apparatus
|Printer friendly Cite/link Email Feedback|
|Author:||Kalaithong, Wichaya; Molloy, Robert; Theerathanagorn, Tharinee; Janvikul, Wanida|
|Publication:||Polymer Engineering and Science|
|Date:||Aug 1, 2017|
|Previous Article:||Mechanism and kinetics of curing of Diglycidyl Ether of Bisphenol a (DGEBA) resin by chitosan.|
|Next Article:||Preparation of PLLA/PBAT-strengthened degradable expanded polystyrene.|