Original Article
Electrospun composite polycaprolactone scaffolds for optimized tissue regeneration
Proc IMechE Part N: J Nanoengineering and Nanosystems 226(3) 111–121 Ó IMechE 2012 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1740349912450828 pin.sagepub.com
M Tyler Nelson1, Joshua P Keith1, Bing-Bing Li2,3, David L Stocum1,3 and Jiliang Li1,3
Abstract There is evidence to suggest that the development of a stable microvasculature at the site of a critical-sized bone defect or fracture aids in repairing or regenerating bone. Identifying a tissue engineering scaffold that optimizes bone tissue and blood vessel development could improve regenerative capabilities. In this paper we study the proliferation and directed differentiation potentials of endothelial colony forming cells and mesenchymal stem cells cultured on electrospun polycaprolactone matrices and compare them with data obtained for composite polycaprolactone–hydroxyapatite, polycaprolactone–hydroxyapatite/b-tricalcium phosphate and polycaprolactone–small intestine submucosa electrospun matrices. Polycaprolactone–hydroxyapatite and polycaprolactone–hydroxyapatite/b-tricalcium phosphate fibers on average displayed a two-fold increase in fiber diameter and average pore-size area as compared with polycaprolactone or polycaprolactone–small intestine submucosa scaffolds. X-ray diffraction showed that significant additions of hydroxyapatite, hydroxyapatite/b-tricalcium phosphate and small intestine submucosa were present in the composite scaffolds. Incorporating hydroxyapatite or hydroxyapatite/b-tricalcium phosphate into the polycaprolactone fiber increased the modulus and ultimate tensile strength significantly. Both endothelial colony forming cell and mesenchymal stem cell proliferation was two-fold greater on polycaprolactone–small intestine submucosa scaffolds; whereas on polycaprolactone– hydroxyapatite and polycaprolactone–hydroxyapatite/b-tricalcium phosphate scaffolds only endothelial colony forming cell proliferation was observed to be significant. Alkaline phosphatase analysis for mesenchymal stem cell-seeded scaffolds indicated that only polycaprolactone–small intestine submucosa scaffolds displayed significant increases after 10 days of culture, suggesting an osteoblast phenotype. Electrospun polycaprolactone–small intestine submucosa scaffolds stimulated proliferation of both cell types and directed mesenchymal stem cell differentiation, providing a stable platform to investigate the potential of endothelial colony forming cell in directing bone tissue repair or regeneration.
Keywords Electrospinning, polycaprolactone, small intestinal submucosa, hydroxyapatite, tricalcium phosphate, endothelial colony forming cells, mesenchymal stem cells
Date received: 6 March 2012; accepted: 16 May 2012
Introduction The development of biocompatible and biodegradable materials to repair, restore or replace living tissue has become a major interest for tissue engineering and regenerative medicine. These materials must combine mechanical, chemical and biocompatible characteristics that complement native tissue.1,2 In recent years, electrospinning has become a viable technology to create microporous three-dimensional (3D) scaffolds with microstructures that closely resemble the native tissue’s extracellular matrix (ECM).1,3,4 The ECM provides the structural framework for cell adhesion, proliferation
and tissue development. Electrospinning uses a highvoltage static electrical field to extrude a polymer melt or solution resulting in ultrafine and continuous 1
Department of Biology, Indiana University–Purdue University Indianapolis, USA 2 Department of Chemistry, Central Michigan University, USA 3 Center for Regenerative Biology and Medicine, Indiana University– Purdue University Indianapolis, USA Corresponding author: Jiliang Li, Department of Biology, Indiana University–Purdue University Indianapolis, Indianapolis, IN 46202, USA. Email:
[email protected]
112 non-woven fibers.1,3–8 Synthetic and natural polymers have both been explored as suitable biomaterials to create electrospun fiber matrices for tissue engineering. Polycaprolactone (PCL), poly-lactic-glycolic acid and poly-glycolic acid are examples of synthetic polymers that have been used extensively to create electrospun fiber scaffolds.9–12 PCL electrospun fiber has been investigated as a possible tissue engineering scaffold material for bone, nerve and blood vessel tissue generation.6,13–16 Synthetic polymer scaffolds have characteristically exhibited improved mechanical properties and degradation stability compared with natural polymers.10,17–20 However, it has been suggested that synthetic polymers lack key cell and protein recognition sites resulting in instances of reduced cellular affinity and proliferation.11,21 Therefore, natural polymers such as, collagen, laminin, fibrin and gelatin, have been extensively examined as bioactive materials for electrospinning tissue engineering scaffolds.16,22–27 Tissue-engineered skin has largely seen success with electrospun collagen I matrices, showing improved kerotinocyte and fibroblast cellular attachment and proliferation.16,18,24 The development of composite natural and synthetic electrospun scaffolds can provide enhancements to tissue engineering scaffolds allowing for tailoring of mechanical and biological properties.28–31 Small intestinal submucosa (SIS), a biomaterial derived from porcine intestine, has been successfully used to repair soft tissue defects in animal models and in clinical use.32–35 The submucosa of the small intestine is located between its mucosal and muscular layers. SIS is harvested from the intestine and processed to remove cells and other components, such as lipids.36 The resulting biomaterial is an acellular, naturally collagenous ECM material. SIS contains key extracellular matrix proteins and chemokines, such as collagens I and III, beta-fibroblastic growth factor (b-FGF) and glycosaminoglycans, etc.37–40 All these components are known to play important roles in tissue remodeling. SIS has been used primarily as a scaffold for the repair of soft tissues, including vascular graft application, dural substitute and the repair of a bladder’s wall and diaphragm.32–36 Recently, SIS has been trialed in orthopedic applications such as the repair of tendons and menisci as well as a bone graft biomaterial in the repair of skeletal defects.41–44 In this study electrospun PCL composites were analyzed for endothelial colony forming cell (ECFC) and mesenchymal stem cell (MSC) viability, proliferation and differentiation. Blood vessel formation near the cartilage transition regions of bone has been shown to support endochondral ossification of cartilage into bone. ECFCs grown in a biodegradable scaffold and surgically implanted at the fracture site and bone defective area in a rat model significantly enhanced new vessel formation, subsequently leading to greater new bone formation in both defective areas.45 Hydroxyapatite (HA) and beta-tricalcium phosphate (TCP) are two
Proc IMechE Part N: J Nanoengineering and Nanosystems 226(3) bioceramics that display osteoconductive properties and are major components of native bone matrix.46–49 Many studies have shown efficacy of use for composite electrospun PCL–HA and PCL–TCP scaffolds in providing an environment that promotes MSC viability, proliferation and osteoblastic cell differentiation in vitro.46–48 ECFCs grown in a HA/TCP porous matrix produced far greater and extensive microvascular networks that supported significantly higher bone matrix formation when surgically implanted in a rat femur fracture site or critical size defect model.32 In order for tissue engineering scaffolds to succeed as suitable replacements for more traditional bone allograft procedures,32–36 they must be biocompatible, closely match the mechanical properties of bone, biodegrade at the rate of bone tissue development, and provide microstructural and chemical environments that support cell infiltration, growth and proliferation.32–39 Electrospun PCL–SIS scaffolds have been shown to significantly increase MSC proliferation and aid in directing differentiation towards an osteoblastic cell phenotype;50 while also supporting endothelial cell development of extensive, stable microvascular networks in vitro.40,42–46 In this paper we compare PCL–HA and PCL–HA/TCP composite electrospun scaffolds against pure PCL fibers, and against PCL–SIS composite scaffolds in vitro to determine optimal scaffold compositions that support ECFC and MSC viability, proliferation and differentiation. Our long-term objective is to provide a suitable platform for stable microvascular networks to aid in facilitating MSC development for bone tissue repair or regeneration.51–56
Materials and methods Materials PCL (Mn 80,000), HA (particle size \ 200 nm, . 99% purity), TCP (particle size \ 200 nm, . 99% purity) and a solvent of 1,1,1,3,3,3-hexafluoro-2-propanol (HFP, . 99% purity) were all obtained from SigmaAldrich (St. Louis, Missouri). Lyophilized and sterilized porcine SIS (\ 100 nm particle size, . 99% purity) was obtained from Cook Biotech Inc. (West Lafayette, Indiana). All materials were used without further purification in the preparation of electrospinning dopes.
Preparation of electrospinning dopes The 5 wt% PCL stock solutions were prepared by dissolution in HFP and stirring at room temperature until a transparent viscous liquid was obtained. Composite PCL/HA and PCL/SIS dopes were prepared from the 5 wt% PCL stock by adding HA or SIS in an 8:2 (PCL: HA, PCL: SIS) dry weight ratio. Composite PCL–HA/ TCP scaffolds were prepared in a similar manner from a 5 wt% PCL stock solution by adding HA and TCP in a 8:1:1 (PCL: HA: TCP) dry weight ratio. Doped
Nelson et al. solutions were stirred at room temperature until the particles were fully dispersed and dissolved.
Electrospinning After PCL solution and the full dissolution of the composite dopes the solutions were placed into 20 cc syringes each fitted with a 20 gage blunt-tipped needle. Each solution was then electrospun to produce a scaffold. The 20 cc syringes with polymer or composite dopes were placed into a syringe pump (Braintree Scientific Inc., Braintree, Massachusetts) set to a flowrate of 10 mL/h. The distance between the needle tip and grounded collector was set to a 20 cm working distance. An adjustable high-voltage DC power source (Gamma High Voltage, Ormond Beach, Florida) was used to produce a strong static electric field. A highvoltage electrode was attached at the needle tip, and the voltage was set to 25 kV. Each of the scaffolds were spun onto a grounded 2.5 3 2.5 in2 aluminum plate for 15 min, producing an average scaffold thickness of 150 6 12 mm. Prior to cell culture all electrospun scaffolds were placed in a vacuum chamber overnight to remove any residual HFP solvent, and then stored in a desiccator for a period no longer than 2 days until use.
Morphology analysis of electrospun fiber Fiber morphology and microstructure was observed using a Jeol JSM-5800LV scanning electron microscope (SEM) (Jeol, Tokyo, Japan). The fiber diameter and the between fibers pore size of 50 fibers and 25 pores from each of the different scaffold types was determined using the free software ImageJ. Prior to the SEM imaging the scaffolds were cut into 10 mm discs, mounted on aluminum mounts using double-sided carbon tape, and gold sputter coated (Denton Vacuum Desk II, Moorestown, New Jersey) for 60 s to create a gold thickness of 15 mm. Gold-coated samples were then observed using the secondary electron detector at an accelerating voltage of 12 kV.
X-ray Diffraction The addition of dopants was confirmed by analyzing the unique crystallography patterns for each of the material components obtained in X-ray diffraction (XRD) (Ultima III, Rigaku Inc., Japan) studies. A monochromatic Cu X-ray source (40 kV) was utilized to characterize the effects of the dopants on the crystalline nature of PCL over the range of 18–35° 2u. A 30 mm disc with an average mat thickness of 148 6 13 mm of each composition was punched out of fiber mats and then adhered to holders for XRD analysis. HA, TCP and SIS pure powders were also examined over the same 2u range as the electrospun samples for reference profiles.
113
Water contact-angle measurement The hydrophobicity of electrospun scaffolds was determined using a Kru¨ss Easy Drop water contact and surface tension instrument (Kru¨ss Inc., Hamburg, Germany). After removal from the desiccator the samples for examination were cut into 30 3 10 mm strips. These strips were placed on a level stage, with a zero background mirror in the field of the optical lens. A 300 mL drop of deionized water was placed on the fiber’s surface and the contact angle was determined using the analysis software provided with the measuring instrument. Five measurements were taken from each of the different scaffold types, and the measured contact angles are reported as an average angle 6 standard deviation. The contact angles reflect the wetting ability of the semi-flat fibrous scaffolds.
Tensile testing The mechanical properties of PCL and the composite scaffolds were determined using an Instron 3365 uniaxial tensile testing machine (Instron, Norwood, Massachusetts). Dog bone-shaped samples were cut from each of the different scaffolds with a gage length of 30 mm and a gage width of 4 mm. Scaffolds were fixed at the ends of the dog bone-shaped samples which were pulled at a cross-head rate of 10 mm/min using a 50 lbf load cell at 37 °C in phosphate buffered saline (PBS). Five samples were tested for each type of electrospun scaffold with average ultimate tensile strength (UTS), % elongation and modulus being reported with standard deviations.
Cell culture Cell isolation. ECFCs were isolated and purified using the method described in Alvarez et al.57 Briefly, vessel wall-derived ECFCs were isolated from the outer edges of the rat pulmonary microvasculature, cultured in Dulbecco’s modified eagle’s medium (DMEM) (Gibco, Grand Island, New York), 10 w/v % fetal bovine serum (FBS) (Gibco, Grand Island, New York), and incubated at 37 °C with 5% CO2–21% O2. MSCs were isolated from rat bone marrow of femurs by mechanically flushing out with DMEM, strained using a 40 mm filter, centrifuged and re-suspended in DMEM/FBS/ antibiotics (100 units/mL penicillin and 100 mg/mL streptocycin, Gibco, Grand Island, New York), and plated for 1 week to isolate mononucleated, adherent cells. Adherent cells underwent rapid proliferation, and were trypsinized, counted and cultured at approximately 5 3 104 cells/ 75 mm dish for further expansion and passage after reaching 80–90% confluence. All MSC cultures were incubated in 95% humidity/5% CO2 at 37 °C in DMEM with 10% FBS and 1% antibiotics. The medium was changed every 3 days. When the culture dishes reached 80–90% confluence, cells were detached using trypsin (Invitrogen, Carlsbad, California), and subcultured. Passage 3 cells were used in the seeding of the electrospun fiber scaffolds.
114 Cell seeding. Electrospun fiber samples were cut into 13 mm diameter discs, and placed in the bottom of a tissue culture polystyrene 24-well plate. Scaffolds were sterilized under UV light for a 24 h period, followed by a single ethanol rinse for 30 min, and two phosphate buffered saline (PBS) rinses each of 30 min duration. After the PBS rinse, the scaffolds were placed into a clean, sterile 24-well plate with 3 mL of DMEM medium overnight. ECFC and MSC were cultured on each of the different scaffold materials (n = 6) at a cell density of 25,000 cells/cm2 for 10 days. The culture medium was changed every day. Cell proliferation analysis. ECFC and MSC cell proliferation was observed using premixed WST-1 cell proliferation assay (Clontech, Mountain View, California). ECFC and MSC-seeded scaffolds (n = 6) for each of the different scaffold types were cultured for 1, 3, 5 and 7 day time intervals. At a given time point scaffolds were removed from the medium, rinsed with deionized water and placed into a clean 24-well plate. WST-1 assay was then added to the scaffolds in a 1:15 ratio to fresh culture medium, and incubated at 37 °C for 2 h to allow for full reactivity. Measurements were taken from the reacted WST-1: culture medium solution by removing a 200 mL aliquot of solution (n = 2 replicates for each sample) and placing the solution into a 96-well plate. These plates were then placed into a Nicolet 6700 multi-well plate reader (Thermo Scientific, Cambridge, Massachusetts) set to a wavelength of 460 nm and run in absorbance mode. Reported values are the average and standard deviation of the absorbance values at a given time point, after averaging within a sample’s replicates and across samples of similar scaffold types to achieve a high degree of accuracy and reduce error. MSC differentiation. MSC-seeded scaffolds (n = 6) for each of the different scaffold types were observed for osteoblast activity by use of a differentiation detecting alkaline phosphate (ALP) assay (Abcam, Cambridge, Massachusetts). After 10 days of culture scaffolds were removed and placed into new sterile 24-well plates with a 1:10 ratio of ALP assay solution to culture medium. Scaffolds were placed on a shaking table and incubated for 2 h at 37 °C to allow for full reactivity. Absorbance measurements were then taken on 200 mL aliquots of the ALP: culture medium solution (n = 2) using the Nicolet 6700 96-well plate reader at a wavelength of 430 nm. The reported values are the average and standard deviation per cent ALP reactivity, calculated by averaging the replicate measurements of each sample and averaging across samples. Cell morphology. ECFC and MSC morphology was observed using the SEM. All cultured nanofiber scaffolds at 3 and 7 day time points were gently rinsed with PBS buffer solution for 30 min, then fixed using formalin for 1 h at 37 °C. Samples were then dehydrated using two rinses with a graded series of ethanol (50%, 75%, 85%, 90%, 100%) and then followed by graded HDMS
Proc IMechE Part N: J Nanoengineering and Nanosystems 226(3) (50:50, 75:25, and 100% hexamethlydisilazane (HDMS): ethanol) drying. The samples were then fixed on aluminum mounts for the SEM and then sputter coated with a 15 nm gold layer before examination using the SEM under the same protocols described previously.
Statistical analysis Statistical analyses of the WST-1 and ALP results were conducted using a two-factor full interaction layout with the factors being the scaffold type and culture time. The statistical software Minitab 16 (Minitab, State College, Pennsylvania) was used to compare the differences in WST-1 absorbance and ALP per cent activity using Tukey’s multiple comparisons method at a 95% confidence level. All results are reported as average 6 standard deviation.
Results Morphology of the electrospun nanofibers The unique microstructure and morphology exhibited by electrospun fiber scaffolds plays a major role in tissue engineering and regeneration. Figure 1(a) shows an SEM image of electrospun PCL fibers. The electrospun PCL scaffold exhibits a continuous, uniform and porous microstructure and has a smooth cylindrical fiber morphology. Figure 1(d) shows PCL–SIS fibers exhibiting a similar fiber morphology to that of PCL, however, with less uniformity. Figure 1(c) shows that the PCL– HA fibers have a wide range of diameters, with some areas of fiber–fiber bonding. Figure 1(b) shows that PCL–HA/TCP fibers have a large diameter, resulting in a considerably different microstructure to that observed for the other samples. Figure 2 shows the average fiber diameter for each of the scaffold types. The PCL fiber has a mean fiber diameter of 286.25 6 136 nm, showing no significant statistical difference from the value for the PCL–SIS fiber. The PCL–HA fibers with a mean diameter of 649.9 6 111 nm and the PCL–HA/TCP fibers with a mean diameter of 868.7 6 180 nm have mean diameters that are significantly different from the values for the PCL and PCL-SIS scaffolds. Figure 3 shows that there is a strong positive correlation between fiber diameter and fiber–fiber pore-size areas.
Characterization studies XRD studies were performed to determine the presence of HA, TCP and SIS in the electrospun scaffolds. Figure 4 shows the XRD patterns for PCL, PCL–SIS, PCL–HA, PCL–HA/TCP, HA powder, SIS powder and TCP powder. As-spun PCL exhibits strong peaks at 21.25 and 23.75° in 2Y. PCL–SIS fibers exhibit a reduction in intensity of the peaks, suggesting a more amorphous morphology, due to the heterogeneity of the amorphous SIS phase. However, small peaks
Nelson et al.
115
Figure 1. SEM micrographs showing the morphology and microstructure of (a) the pure PCL scaffold, (b) the PCL + n-HA composite scaffold, (c) the PCL + n-HA + n-TCP composite scaffold and (d) the PCL + SIS composite scaffold. The ceramic composite scaffolds ((b) and (c)) show a crystalline-like structure on the interior and exterior edges of the fibers, whereas the PCL + SIS scaffold shows no significant difference in fiber morphology.
Figure 2. Average fiber diameter for each of the scaffolds, showing no significant difference between pure PCL and PCL–SIS scaffolds. However, PCL–HA and PCL–HA/TCP scaffolds exhibited significantly greater fiber diameters than PCL or PCL–SIS scaffolds. Significance denoted by (*), p \ 0.05 versus pure PCL.
Figure 3. A positive correlation between average pore-size area and fiber diameter suggests that as fiber diameter increases the pore-size area increases. Table 1. Tensile testing data.
PCL: polycaprolactone; SIS: small intestine submucosa; HA: hydroxyapatite; TCP: b-tricalcium phosphate.
seen at 21.25 and 23.75° in 2Y represent a PCL component of the composite suggesting a blended material. PCL–HA fibers exhibit a decrease in the crystalline peaks seen for PCL, however, there are additional peaks found at 32.1, 33.25 and 33.75° in 2Y that correlate closely with the XRD profile of powdered HA. PCL–HA/TCP displays a similar decrease in the overall intensity of the PCL peaks, but in addition to the PCL peaks, a blend of HA and TCP peaks are seen to be present at lower intensities than the HA or TCP
Pure PCL PCL–HA PCL–HA/TCP PCL–SIS
UTS (MPa)
%elongation
Modulus (MPa)
0.91 6 0.12 1.40 6 0.29a 1.31 6 0.15a 0.85 6 .08
139.07 6 13.0 64.41 6 11.0a 52.54 6 8.5a 165.66 6 25
1.50 6 0.24 2.81 6 0.28a 2.46 6 0.10a 1.32 6 0.12
a
p \ 0.05.UTS: ultimate tensile strength; PCL: polycaprolactone; HA: hydroxyapatite; TCP: b-tricalcium phosphate; SIS: small intestine submucosa.
powders alone. XRD analysis adds additional confirmation that electrospun composite scaffolds were produced.
116
Proc IMechE Part N: J Nanoengineering and Nanosystems 226(3) 5. The ECFCs exhibited a normal cellular morphology with rounded or spherical shaped cells. Long projections are seen connecting from cell to cell, and matrix to cell. After 8 days of culture it is very noticeable that in the case of the composite scaffolds, cells seem to have over-grown the scaffold, and in some places the ECM is very thick and fractured (Figures 5(b) and (c)). Fracturing of the ECM from the cells is most likely due to the fixation process or dehydration of the scaffolds for SEM examination.
ECFC and MSC proliferation and ALP assay results Figure 4. XRD patterns for the electrospun fiber and pure powder dopants. PCL shows high crystallinity with peaks at 21.75 and 23.5° in 2Y. The addition of the dopants results in a decreased crystallinity for PCL, and the new peaks observed in the patterns closely correlate with pure powder dopant profiles. SIS is completely amorphous and thus does not create a diffraction pattern. PCL: polycaprolactone; SIS: small intestine submucosa; HA: hydroxyapatite; TCP: b-tricalcium phosphate.
Tensile testing results help further confirm the results of the XRD analysis. Table 1 shows the UTS, %elongation and elastic modulus (E) for the electrospun scaffolds tested in PBS at 37 °C. PCL–HA and PCL–HA/ TCP scaffolds exhibit significantly higher average UTS and modulus values. However, the elongation is decreased significantly leading to a more brittle fracture at failure. PCL–SIS scaffolds show no significant difference in mechanical properties from those for PCL fiber, thus exhibiting a more ductile fracture at failure. To allow possible hydrophobicity effects to be investigated, the average values for the water contact angle for the composite scaffolds are listed in Table 2. PCL has an average wetting angle of 135.4 613°. This is extremely high, resulting in a highly hydrophobic material. The PCL–HA and PCL–HA/TCP scaffolds exhibit wetting angles significantly lower than that for pure PCL. PCL + SIS scaffolds have an average wetting angle of 10.8 6 15° (Table 2, p \ 0.001). This is a 12-fold decrease in the wetting angle making this material very hydrophilic. This improvement in hydrophilic behavior could translate to better cellular viability and adhesion.
Cellular morphology on fibers ECFCs grew very rapidly, with high proliferation after 3 days of culture as seen in the SEM images in Figure
Figures 6(a) and (b) display the WST-1 proliferation assay results collected on each of the scaffolds at 1, 3, 5 and 7 days for ECFC and MSC. Proliferation can be seen for all ECFC and MSC inoculated scaffolds. Figure 6(a) shows PCL–HA/TCP scaffolds exhibiting 1.5-fold greater proliferation at early, 1 and 3 day time points. ECFC-inoculated PCL–HA/TCP and PCL–SIS scaffolds exhibit the greatest proliferation after 7 days with three-fold greater values than pure PCL. No significant increase in proliferation is observed between days 3 and 5. Figure 6(b) shows consistent increases in MSC proliferation for all scaffolds. No significant differences are detected for days 1, 3 and 5; however, MSC-inoculated PCL–SIS scaffolds show a 1.5-fold increase in proliferation compared to all other scaffold types. ALP colorimetric assay detects the ALP activity of cells, which is often used as a marker for ostoegenic cells. Figure 7 shows the ALP activity after 14 days of inoculation on electrospun scaffolds. Composite scaffolds exhibit a significant 1.6-fold increase in ALP activity as compared to pure PCL fiber on average. However, no significant difference was detected between composite scaffolds.
Discussion Fiber morphology and microstructure Many researchers have published results showing the ability of nanofiber scaffolds to provide a 3D microenvironment in which cells can adhere, proliferate and differentiate much like the behavior of cells developing in native ECM.2,3,16 The porosity, mechanical properties, and surface topography and chemistry play a crucial role in sustaining cell viability, proliferation, and in some instances, direct or control differentiation.2–8,46 Figures 1 and 2 show that PCL–HA and PCL–HA/
Table 2. Water contact angle (hydrophobicity test).
Average angle (deg) a
Pure PCL
PCL–HA
135.4613.4
90.2625.2a
PCL–HA/TCP 81.8610.6a
p \ 0.05. p \ 0.01. PCL: polycaprolactone; HA: hydroxyapatite; TCP: b-tricalcium phosphate; SIS: small intestine submucosa. b
PCL–SIS 10.8615.8b
Nelson et al.
117
Figure 5. SEM images of NFS seeded with ECFC cells at 3 and 7 day time points showing significant cellular proliferation on the composite scaffolds (a) pure PCL, (b) PCL + HA, (c) PCL + HA/TCP and (d) PCL + SIS.
TCP scaffolds have a significantly increased fiber diameter, whereas PCL and PCL–SIS scaffolds show no significant difference. Correlation between fiber–fiber pore size and fiber diameter can provide some evidence as to the level of scaffold porosity.4,5 Figure 3 shows the correlation between pore size and fiber diameter; from this figure it is clear that the larger the fiber diameter the greater the pore size, suggesting an increased porosity.5 However, fiber density will also play a role based on the packing or orientation of the fibers in 3D space. PCL–HA/TCP fibers have the greatest pore size whereas PCL and PCL–SIS scaffolds show no significant difference. Nanofiber scaffolds exhibit high surface areas and porosities.3–5 The high surface area allows easy attachment of cells, proteins and other cellular components that might be able to bind to the surface of the nanofiber scaffold.5,17 The porosity allows cellular, chemical and mechanical communication among adjacent cells and their microenvironment.2–4,46 Figure 4 shows that composite electrospun scaffolds have an overall decrease in the extent of crystalline PCL, however, new peaks emerge and correlate at 2Y angles that closely match the XRD profiles of the pure-powder dopants. This confirms that significant amounts of the dopants (HA, TCP or SIS) are contained within the composite fibers. The mechanical properties when tested in an aqueous environment at 37 °C for the PCL–HA and PCL–HA/TCP scaffolds are significantly different from the PCL and PCL–SIS scaffolds as a result of the dopants. With the addition of more brittle and stronger inorganic additives such as HA and TCP, the UTS and elastic modulus increases two-fold, while the % elongation decreases two-fold compared with PCL and PCL–SIS scaffolds. No significant difference in mechanical properties was found between the PCL and PCL–SIS scaffolds. However, the hydrophobicity of the composite scaffolds as compared with that for pure PCL is significantly different, which could greatly affect a cell’s ability to adhere to the fibers.42–44 Hydrophobicity can also affect the absorption of
nutrient, protein and other culture media-based growth factors due to the scaffold surface’s aid of cell adhesion and viability.4–6 PCL–SIS scaffolds exhibit a significant decrease in water contact angle as seen in Table 2, and due to the solubility of type I collagen in water or aqueous environments, it can be hypothesized that significant amounts of the contained growth factors and cytokines are released into the media.20–22,40–43 The release of growth factors or cytokines such as b-FGF could help stimulate MSC and ECFC development.40–43 Creating a scaffold that optimizes bone regeneration or repair by providing suitable environments for both ECFCs to set-up a stable microvasculature and MSCs to differentiate into bone requires the correct mechanical properties. Evidence suggests that high modulus materials help support the differentiation of MSCs into osteoblasts.48 ECFC grown within a 3D Matrigel matrix has been shown to support microvascularization, and bone regeneration was maximized when HA and TCP were added as dopants to the supportive microvasculature.32 Chemical cues such as the ostoegenic properties of HA and TCP, the unique growth factor of SIS and protein attributes such as collagen type I, cytokines and bFGF, have been shown to effectively improve cellular adhesion and proliferation helping drive tissue development in vitro and in vivo.34– 44 The combination of the distinctive electrospun fiber microstructure, mechanical properties and improved chemical composition created by the composite scaffolds suggest that optimization of the scaffold could provide suitable microenvironments for ECFC and MSC development.
Cellular response The morphology and development of the ECFC and MSC were similar on each of the different scaffolds. The validity of using composite PCL–HA, PCL–HA/ TCP or PCL–SIS scaffolds for MSC culture has been widely tested and shown in each case to have efficacy
118
Proc IMechE Part N: J Nanoengineering and Nanosystems 226(3)
Figure 6. (a) WST-1 proliferation results showing significant increases in proliferation for ECFC- seeded composite scaffolds (* denotes statistical significance for the PCL + HA /TCP and PCL + SIS groups (p + 0.05)) and (b) WST-1 proliferation results for MSC-seeded scaffolds (where only PCL + SIS scaffolds showed statistical significance from the pure PCL group (p \ 0.05)). PCL: polycaprolactone; HA: hydroxyapatite; TCP: b-tricalcium phosphate; SIS: small intestine submucosa.
as a potential scaffold for the development of bone tissue.30–44 Looking at Figure 5, ECFC-inoculated scaffolds show a normal cellular morphology and spherical shape. Long extensions appear to connect the cells, and based on visual and quantitative analysis of Figure 6(a) overall ECFC proliferation and growth is greater on the composite scaffolds as compared with pure PCL. Early indication of proliferation is seen in the PCL– HA/TCP and PCL–HA scaffolds. This finding suggests that incorporation of the either HA or TCP into the PCL fiber composition improves early cell growth and proliferation. This could be a result of the adsorption of components such as fibronectin during the early
stages of cellular attachment, which act as the foundation for cellular adhesion.33,34 Both PCL–SIS and PCL–HA/TCP scaffolds seem to have a greater impact on ECFC and MSC proliferation at later time points. In all cases PCL–SIS and PCL–HA/TCP scaffolds exhibit far greater proliferation values than pure PCL. ECFC growth on all of the composite scaffolds shows complete confluence after only 8 days with significant ECM deposition and mechanical connections between cells. While PCL–SIS scaffolds display very little proliferative improvement over pure PCL scaffolds at early time points, they prove to be far better at promoting proliferation at later time points for both cell types.
Nelson et al.
119
Figure 7. ALP assay results for MSC-seeded electrospun scaffolds, displaying standardized percent area. * and ** denote statistical significance, p value \ 0.05 and \ 0.01, respectively. PCL: polycaprolactone; HA: hydroxyapatite; TCP: b-tricalcium phosphate; SIS: small intestine submucosa.
The mechanical properties, microstructure and fiber morphology are relatively unchanged for PCL–SIS scaffolds as compared with pure PCL, however, there is a significant increase in the overall proliferation and ALP activity. This observation suggests that PCL–SIS scaffolds provide a purely chemical improvement on the PCL composition whereas PCL–HA/TCP scaffolds provide both mechanical and chemical alterations to the PCL fiber. Increased levels of SIS or HA/TCP would be required in both cases to determine the effects of each on the biological activity. However, based on the presented results, PCL–SIS or PCL–HA/TCP scaffolds appear to provide the necessary microstructural, mechanical and chemical properties to optimize ECFC and MSC development. Further studies investigating co-cultures are necessary to confirm the efficacy of the microvasculature effect of ECFC on the development of tissue in vitro or in vivo for bone regeneration or repair.
proliferation and differentiation among the PCL/SIS fibers with different diameters. The properties of SIS have been extensively studied and reported in the literature. It should be noted that in the orthopedic realm, concerns have been stated about potential inflammation caused with the use of this material.43,59 However, SIS has been used in a variety of clinical applications: including hernia, pelvic floor wound care and neurosurgery.32–35 The method used to process biomaterial grade ECM (e.g. SIS) is probably the key factor that determines the success or failure of experiments and clinical use. As such, care should be taken in determining the processing route and source of the ECM. The results detailed in this paper are indicative of the SIS material used in this study and may not translate to alternative preparations.
Conclusions Limitations and future perspective Our PCL/SIS data are consistent with those reported by Yoon and Kim58 who demonstrated that electrospun PCL/SIS fibers have a greater hydrophilic nature and are therefore much more favorable for PC-12 cell growth, compared with a pure scaffold. However, the average diameter of PCL/SIS fibers in this study is three times greater than the previous study by Yoon and Kim.58 We expect that the hydrophilic nature of the PCL/SIS scaffold and cellular behavior of MSCs and ECFCs would markedly improve as the diameter of the PCL/SIS fiber decreases. In order to test this hypothesis in future studies we will compare cell
Optimization of the structural, mechanical and chemical properties of a scaffold is a critical step in tissue engineering. The proximity and connection to a stable microvasculature in many tissues is the key to sustaining cell viability, proliferation and differentiation. The microstructure of electrospun scaffolds provides a framework that closely mimics natural ECM, providing a microenvironment that helps promote cellular growth and proliferation. The addition of inorganic components, HA and TCP, into the PCL core results in improved mechanical properties, including a significant increase in the modulus. ECFC and MSC growth and proliferation on PCLHA/TCP and PCL–SIS scaffolds has been shown to be significantly greater than for pure
120 PCL. PCL–SIS scaffolds displayed the only unique ALP result, showing that MSC differentiation favored an osteoblastic cell line. The use of growth factors, cytokines and protein-rich SIS components provided chemical enhancements to the PCL composite for optimizing ECFC and MSC development. From the presented results it can be concluded that electrospun PCL–SIS composite scaffolds are an optimized tissue engineering platform that can be used to test the benefits of ECFCderived stable microvascular networks on MSC development for bone tissue repair or regeneration. Funding This study was supported by research funds provided by the Purdue School of Science and Multidisciplinary Undergraduate Research Institute (MURI), Indiana University Purdue University Indianapolis (JL).
Conflict of interest No competing financial interests exist. References 1. Huang ZM, Zhang YZ, Kotaki M, et al. A review on polymer nanofibers by electrospinning and their applications in nanocomposites. Compos Sci Technol 2003; 63: 2223–2253. 2. Yoshimoto H, Shin YM, Terai H, et al. A biodegradable nanofiber scaffold by electrospinning and its potential for bone tissue engineering. Biomaterials 2003; 24: 2077– 2082. 3. Kadler KE, Holmes DF, Trotter JA, et al. Collagen fibril formation. Biochem J 1996; 316: 1–11. 4. Ma ZW, Kotaki M, Inai R, et al. Potential of nanofiber matrix as tissue-engineering scaffolds. Tissue Engng 2005; 11: 101–109. 5. Lannutti J, Nam J, Huang Y, et al. Improved cellular infiltration in electrospun fiber via engineered porosity. Tissue Engng 2007; 13: 2249–2257. 6. Lannutti J, Reneker D, Ma T, et al. Electrospinning for tissue engineering scaffolds. Mat Sci Engng C-Biomimet Supermol Syst 2007; 27: 504–509. 7. Li WJ, Laurencin CT, Caterson EJ, et al. Electrospun nanofibrous structure: a novel scaffold for tissue engineering. J Biomed Mater Res 2002; 60: 613–621. 8. Smith LA and Ma PX. Nano-fibrous scaffolds for tissue engineering. Colloid Surf B, Biointerfaces 2004; 39: 125–131. 9. Dong CM, Guo YZ, Qiu KY, et al. In vitro degradation and controlled release behavior of D,L-PLGA50 and PCL-b-D,L-PLGA50 copolymer microspheres. J Controlled Release 2005; 107(1): 53–64. 10. Lannutti J, Johnson J, Niehaus A, et al. Electrospun PCL in vitro: a microstructural basis for mechanical property changes. J Biomater Sci-Polym Ed 2009; 20: 467–481. 11. Li M, Mondrinos MJ, Chen X, et al. Co-electrospun poly(lactide-co-glycolide), gelatin, and elastin blends for tissue engineering scaffolds. J Biomed Mater Res A 2006; 79A(4): 963–973.
Proc IMechE Part N: J Nanoengineering and Nanosystems 226(3) 12. Xu CY, Inai R, Kotaki M, et al. Aligned biodegradable nanotibrous structure: a potential scaffold for blood vessel engineering. Biomaterials 2004; 25: 877–886. 13. Wutticharoenmongkol P, Sanchavanakit N, Pavasant P, et al. Preparation and characterization of novel bone scaffolds based on electrospun polycaprolactone fibers filled with nanoparticles. Macromol Biosci 2006; 6: 70–77. 14. Bowlin G, Sell S, Barnes C, et al. Extracellular matrix regenerated: tissue engineering via electrospun biomimetic nanofibers. Polym Int 2007; 56: 1349–1360. 15. Mey J, Schnell E, Klinkhammer K, et al. Guidance of glial cell migration and axonal growth on electrospun nanofibers of poly-epsilon-caprolactone and a collagen/ poly-epsilon-caprolactone blend. Biomaterials 2007; 28: 3012–3025. 16. Powell HM and Boyce ST. Engineered human skin fabricated using electrospun collagen-PCL blends: morphogenesis and mechanical properties. Tissue Engng Part A 2009; 15: 2177–2187. 17. He AH, Li JX, Zheng JF, et al. Gelatin and gelatinhyaluronic acid nanofibrous membranes produced by electrospinning of their aqueous solutions. Biomacromolecules 2006; 7: 2243–2247. 18. Powell HM and Drexler JW. Dehydrothermal crosslinking of electrospun collagen. Tissue Engng Part C: Methods 2011; 17: 9–17. 19. Yu DS, Lee CF, Chen HI, et al. Bladder wall grafting in rats using salt-modified and collagen-coated polycaprolactone scaffolds: Preliminary report. Int J Urol 2007; 14(10): 939–944. 20. Zhang YZ, Venugopal J, Huang ZM, et al. Crosslinking of the electrospun gelatin nanofibers. Polymer 2006; 47: 2911–2917. 21. Cai Q, Yang JA, Bei JZ, et al. A novel porous cells scaffold made of polylactide-dextran blend by combining phase-separation and particle-leaching techniques. Biomaterials 2002; 23: 4483–4492. 22. Matthews JA, Boland ED, Wnek GE, et al. Electrospinning of collagen type II: a feasibility study. J Bioact Compat Polym 2003; 18: 125–134. 23. Matthews JA, Wnek GE, Simpson DG, et al. Electrospinning of collagen nanofibers. Biomacromolecules 2002; 3: 232–238. 24. Powell HM, Supp DM and Boyce ST. Influence of electrospun collagen on wound contraction of engineered skin substitutes. Biomaterials 2008; 29: 834–843. 25. Yung LYL, Zhong SP, Teo WE, et al. An aligned nanofibrous collagen scaffold by electrospinning and its effects on in vitro fibroblast culture. J Biomed Mater Res A 2006; 79: 456–463. 26. Zhang YZ. Electrospinning of gelatin fibers and gelatin/ PCL composite fibrous scaffolds. J Biomed Mater Res B, Appl Biomater 2005; 72: 156–165. 27. Zhang YZ, Venugopal J, Huang ZM, et al. Characterization of the surface biocompatibility of the electrospun PCL-collagen nanofibers using fibroblasts. Biomacromolecules 2005; 6: 2583–2589. 28. Coombes AGA, Verderio E, Shaw B, et al. Biocomposites of non-crosslinked natural and synthetic polymers. Biomaterials 2002; 23: 2113–2118. 29. Dai NT, Williamson MR, Khammo N, et al. Composite cell support membranes based on collagen and
Nelson et al.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
polycaprolactone for tissue engineering of skin. Biomaterials 2004; 25: 4263–4271. Venugopal J, Ma LL, Yong T, et al. In vitro study of smooth muscle cells on polycaprolactone and collagen nanofibrous matrices. Cell Biol Int 2005; 29: 861–867. Venugopal J, Zhang YZ and Ramakrishna S. Fabrication of modified and functionalized polycaprolactone nanofibre scaffolds for vascular tissue engineering. Nanotechnology 2005; 16: 2138–2142. Bejjani GK and Zabramski J. Safety and efficacy of the porcine small intestinal submucosa dural substitute: results of a prospective multicenter study and literature review. J Neurosurg 2007; 106: 1028–1033. Franklin ME, Trevino JM, Portillo G, et al. The use of porcine small intestinal submucosa as a prosthetic material for laparoscopic hernia repair in infected and potentially contaminated fields: long-term follow-up. Surg Endosc 2008; 22: 1941–1946. Mostow EN, Haraway GD, Dalsing M, et al. Effectiveness of an extracellular matrix graft (OASIS Wound Matrix) in the treatment of chronic leg ulcers: a randomized clinical trial. J Vasc Surg 2005; 41: 837–843. Wiedemann A and Otto M. Small intestinal submucosa for pubourethral sling suspension for the treatment of stress incontinence: first histopathological results in humans. J Urology 2004; 172: 215–218. Kropp BP. Small-intestinal submucosa for bladder augmentation: a review of preclinical studies. World J Urol 1998; 16: 262–267. Badylak SF, Record R, Lindberg K, et al. Small intestinal submucosa: a substrate for in vitro cell growth. J Biomater Sci-Polym Ed 1998; 9: 863–878. Suckow MA, Voytik-Harbin SL, Terril LA, et al. Enhanced bone regeneration using porcine small intestinal submucosa. J Invest Surg 1999; 12: 277–287. Voytik-Harbin SL, Brightman AO, Waisner BZ, et al. Small intestinal submucosa: a tissue-derived extracellular matrix that promotes tissue-specific growth and differentiation of cells in vitro. Tissue Engng 1998; 4: 157–174. Voytik-Harbin SL, Rajwa B and Robinson JP. Threedimensional imaging of extracellular matrix and extracellular matrix-cell interactions. Methods Cell Biol 2001; 63: 583–597. Badylak S, Arnoczky S, Plouhar P, et al. Naturally occurring extracellular matrix as a scaffold for musculoskeletal repair. Clin Orthop Relat Res 1999; 367: S333–S343. Dejardin LM, Arnoczky SP, Ewers BJ, et al. Tissue-engineered rotator cuff tendon using porcine small intestine submucosa - histologic and mechanical evaluation in dogs. Am J Sports Med 2001; 29: 175–184. Derwin K, Androjna C, Spencer E, et al. Porcine small intestine submucosa as a flexor tendon graft. Clin Orthop Relat Res 2004; 423: 245–252. Moore DC, Pedrozo HA, Crisco JJ, et al. Preformed grafts of porcine small intestine submucosa (SIS) for bridging segmental bone defects. J Biomed Mater Res A 2004; 69: 259–266. Chandrasekhar KS, Zhou H, Zeng P, et al. Blood vessel wall-derived endothelial colony-forming cells enhance fracture repair and bone regeneration. Calcif Tissue Int 2011; 89: 347–357.
121 46. Araujo JV, Martins A, Leonor IB, et al. Surface controlled biomimetic coating of polycaprolactone nanofiber meshes to be used as bone extracellular matrix analogues. J Biomater Sci-Polym Ed 2008; 19: 1261–1278. 47. Fujihara K, Kotaki M and Ramakrishna S. Guided bone regeneration membrane made of polycaprolactone/calcium carbonate composite nano-fibers. Biomaterials 2005; 26: 4139–4147. 48. Nandakumar A, Yang L, Habibovic P, et al. Calcium phosphate coated electrospun fiber matrices as scaffolds for bone tissue engineering. Langmuir 2010; 26: 7380– 7387. 49. Yang F, Both SK, Yang XC, et al. Development of an electrospun nano-apatite/PCL composite membrane for GTR/GBR application. Acta Biomater 2009; 5: 3295– 3304. 50. Yoon H, Ahn S and Kim G. Three-dimensional polycaprolactone hierarchical scaffolds supplemented with natural biomaterials to enhance mesenchymal stem cell proliferation. Macromol Rapid Commun 2009; 30: 1632– 1637. 51. Badylak S, Liang A, Record R, et al. Endothelial cell adherence to small intestinal submucosa: an acellular bioscaffold. Biomaterials 1999; 20: 2257–2263. 52. Buschmann J, Welti M, Hemmi S, et al. Three-dimensional co-cultures of osteoblasts and endothelial cells in DegraPol foam: histological and high-field magnetic resonance imaging analyses of pre-engineered capillary networks in bone grafts. Tissue Engng Part A 2011; 17: 291–299. 53. Rutledge GC, Lowery JL and Datta N. Effect of fiber diameter, pore size and seeding method on growth of human dermal fibroblasts in electrospun poly (epsilon-caprolactone) fibrous mats. Biomaterials 2010; 31: 491–504. 54. Sturgis JE, Robinson JP and Voytik-Harbin SL. Threedimensional (3D) culture of human vascular cells in a complex extracellular matrix (ECM). Mol Biol Cell 1998; 9: 168A–169A. 55. Voytik-Harbin SL, Critser PJ, Kreger ST, et al. Collagen matrix physical properties modulate endothelial colony forming cell-derived vessels in vivo. Microvasc Res 2010; 80: 23–30. 56. Voytik-Harbin SL, Pizzo AM, Kokini K, et al. Extracellular matrix (ECM) microstructural composition regulates local cell-ECM biomechanics and fundamental fibroblast behavior: a multidimensional perspective. J Appl Physiol 2005; 98: 1909–1921. 57. Alvarez DF, Huang L, King JA, et al. Lung microvascular endothelium is enriched with progenitor cells that exhibit vasculogenic capacity. Am J Physiol Lung Cell Mol Physiol 2008; 294:419–430. 58. Yoon H and Kim G. Micro/nanofibrous scaffolds electrospun from PCL and small intestinal submucosa. J Biomater Sci-Polym Ed 2010; 21: 553–562. 59. Zheng MH, Chen J, Kirilak Y, et al. Porcine small intestine submucosa (SIS) is not an acellular collagenous matrix and contains porcine DNA: possible implications in human implantation. J Biomed Mater Res B, Appl Biomater 2005; 73: 61–67.
