Bioresorbable elastomeric vascular tissue engineering scaffolds via melt spinning and electrospinning

Acta Biomaterialia 6 (2010) 1958–1967

Contents lists available at ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

Bioresorbable elastomeric vascular tissue engineering scaffolds via melt spinning and electrospinning Sangwon Chung a,b, Nilesh P. Ingle a,b, Gerardo A. Montero a, Soo Hyun Kim c, Martin W. King a,d,* a

Fiber and Polymer Science, North Carolina State University, Raleigh, NC 27695-8301, USA Joint Department of Biomedical Engineering, North Carolina State University and University of North Carolina at Chapel Hill, Raleigh, NC 27695-7115, USA c Biomaterials Research Center, Korea Institute of Science and Technology, P.O. Box 131, Cheongryang, Seoul 130-650, Republic of Korea d College of Textiles, Donghua University, Songjiang, Shanghai 201620, People’s Republic of China b

a r t i c l e

i n f o

Article history: Received 4 August 2009 Received in revised form 7 October 2009 Accepted 3 December 2009 Available online 11 December 2009 Keywords: Tissue engineering scaffolds Elastomeric PLCL Melt spinning Electrospinning

a b s t r a c t Current surgical therapy for diseased vessels less than 6 mm in diameter involves bypass grafting with autologous arteries or veins. Although this surgical practice is common, it has significant limitations and complications, such as occlusion, intimal hyperplasia and compliance mismatch. As a result, cardiovascular biomaterials research has been motivated to develop tissue-engineered blood vessel substitutes. In this study, vascular tissue engineering scaffolds were fabricated using two different approaches, namely melt spinning and electrospinning. Small diameter tubes were fabricated from an elastomeric bioresorbable 50:50 poly(L-lactide-co-e-caprolactone) copolymer having dimensions of 5 mm in diameter and porosity of over 75%. Scaffolds electrospun from two different solvents, acetone and 1,1,1,3,3,3-hexafluoro-2-propanol were compared in terms of their morphology, mechanical properties and cell viability. Overall, the mechanical properties of the prototype tubes exceeded the transverse tensile values of natural arteries of similar caliber. In addition to spinning the polymer separately into melt-spun and electrospun constructs, the approach in this study has successfully demonstrated that these two techniques can be combined to produce double-layered tubular scaffolds containing both melt-spun macrofibers (<200 lm in diameter) and electrospun submicron fibers (>400 nm in diameter). Since the vascular wall has a complex multilayered architecture and unique mechanical properties, there remain several significant challenges before a successful tissue-engineered artery is achieved. Ó 2010 Published by Elsevier Ltd. on behalf of Acta Materialia Inc.

1. Introduction One of the more severe forms of heart disease is associated with atherosclerosis, which affects more than 8 million Americans, leading to a significant process that causes narrowing of the arteries [1,2]. Recent statistics show that peripheral arterial disease increases morbidity and mortality [3]. Surgical replacement of vessel segments or bypass surgery is the most common intervention for coronary and peripheral atherosclerotic disease, with at least 550,000 bypass cases performed per year [4]. Synthetic vascular prostheses made of poly(ethylene terephthalate) or expanded polytetrafluoroethylene have been effective for large diameter grafts. However, in spite of many years of research using a wide variety of biomaterials, clinical success for small diameter (<6 mm) vessels has yet to be demonstrated due to complications such as occlusion, thrombosis and intimal hyperplasia [1,2,4–7].

* Corresponding author. Address: Fiber and Polymer Science, North Carolina State University, Raleigh, NC 27695-8301, USA. Tel.: +1 919 515 1011. E-mail address: [email protected] (M.W. King).

Several attempts have been made to construct a blood vessel replacement with biological functionality. The use of protein coatings and the seeding of endothelial cells on the lumenal surface have shown some promise to increase biocompatibility, but permanent synthetic materials are unlikely to be fully accepted, given the chronic inflammatory response they provoke and the increased risk of infection. On the other hand, implantation of native vessels is limited by the mismatch of dimensional and mechanical properties [2,6]. Given the limitations of these current techniques, the desirability of a biocompatible engineered blood vessel is pushing research into the development of tissue-engineered small-diameter blood vessels. For tissue engineering, the design of the scaffold plays a significant role since the matrix provides the cells with a tissue-specific environment and architecture. In particular, designing a scaffold structure that enables cells to proliferate and grow into a threedimensional (3-D) tissue while allowing the necessary amount of oxygen and nutrient diffusion is important. The key factors in creating a 3-D scaffold include tailoring the degradation rate to meet the requirements of new tissue growth, providing interconnected pores, generating high porosity so as to promote cell–cell and cell–matrix communication, and having sufficient mechanical

1742-7061/$ - see front matter Ó 2010 Published by Elsevier Ltd. on behalf of Acta Materialia Inc. doi:10.1016/j.actbio.2009.12.007

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

stability [8–12]. Developing scaffolds that can maintain their mechanical integrity while exposing cells to long-term cyclic mechanical strains is especially necessary in cardiovascular applications when engineering smooth muscle cellular constructs [13– 16]. To achieve this, scaffolds should be elastic enough to withstand cyclic mechanical strains without any significant permanent deformation or creep. So, in the selection of a scaffold material for the fabrication of engineered tissues, a candidate polymer should possess appropriate mechanical properties with a low elastic modulus or high compliance, which are suitable for the target applications. In addition, its degradation products during implantation should be nontoxic. Over the past years, hydrolyzable and biocompatible copolymers of e-caprolactone and L-lactide have been of great interest for medical applications [13,17]. Polylactide (PLA) is a crystallizable hard and brittle material, whereas poly(e-caprolactone) is a semi-crystalline material with rubbery properties. Copolymers of L-lactide and e-caprolactone (poly(L-lactide-co-ecaprolactone), PLCL) exhibit a range of mechanical properties from rigid solids to elastomers, depending on their composition [18,19]. There have been previous studies using PLCL copolymers to fabricate scaffolds by various methods, including extrusion, particulate leaching [13,14] and electrospinning [19–21]. In particular, a 50:50 ratio of L-lactide and e-caprolactone monomers has proven to have high elastomeric properties with breaking strains in excess of 100% [13,14,18,19]. Compared to other bioresorbable polymers such as polyglycolide and PLA, this PLCL copolymer has a slow rate of degradation, with in vivo studies reporting 81% mass retained after 15 weeks of implantation [22]. The objective of this study was to design a porous biodegradable multilayered tubular construct that mimics the structural configuration and mechanical properties of a native vessel. To achieve this goal, the highly elastomeric 50:50 PLCL copolymer was selected to produce a biconstituent tubular scaffold with distinct layers with different geometries. Both melt spinning and electrospinning techniques were chosen so as to create separate layers with different fiber diameters (macrofibers vs. submicron fibers) and different fiber alignments within the scaffold’s architecture. Recently, fibrous structures have gained considerable interest as tissue engineering scaffolds due to their high surface-to-volume ratios, interconnected pores and high total porosity, while giving versatile material selection and better control over pore size distribution, as well as consistent and flexible processing. Melt spinning and winding were undertaken to achieve larger-diameter fibers and to control the angle between them so as to modulate the pore size. Scaffolds were electrospun from two different solvents, acetone and 1,1,1,3,3,3hexafluoro-2-propanol (HFIP), and were evaluated in terms of their mechanical performance, morphology, ease of processing, cytotoxicity and cell viability. The scaffolds were designed to have values for total porosity in excess of 75%, an interconnected network of pores, and values for mechanical strength and elongation exceeding those of natural arteries. The transverse tensile strength and strain as well as the initial tensile modulus were compared between the different types of scaffolds. Finally, based on the characterization of the individual layers of melt-spun and electrospun structures, a biconstituent structural concept was attempted by combining concentric layers of melt-spun macrofibers and electrospun submicron fibers within the same tubular scaffold without the need for an adhesive.

1959

procedure described previously [23]. Briefly, L-lactide (100 mmol), e-caprolactone (100 mmol) and 1,6-hexanediol (0.5 mmol) were polymerized at 150 °C for 24 h, using stannous octoate (1 mmol) as the catalyst. After reaction, the product was dissolved in chloroform and precipitated in methanol, filtered and dried under vacuum. The molar ratio of the two monomers (L-lactide and e-caprolactone) in the PLCL copolymer was 50:50. The number and weight average molecular weights of the copolymer were Mn = 240,000 and Mw = 350,000, respectively. The molecular weights were determined using a gel permeation chromatograph (Waters 410, Milford, MA) equipped with microstyragel columns calibrated with polystyrene standards. The copolymer was stored at room temperature under dry conditions in sealed plastic bags in a vacuum desiccator. 2.2. Melt spinning Monofilament fiber samples were produced by melt spinning using a co-rotating twin screw extruder Thermo Haake Z 4.1 Mini Lab (Thermo Electron Corp., Hamburg, Germany) at a fixed temperature of 155 °C. The circular single spinneret orifice for extruding the fibers was 0.25 mm in diameter. A 6 g quantity of polymer was fed into the chamber and the extruder was run for 5 min before melt spinning. The speed of the screws was set to 150 rpm based on preliminary melt viscosity measurements. A customized wind-up unit provided an automated traverse motion for collecting the melt-spun monofilament fibers at a predetermined winding angle. The speed of the motor for winding was controlled at 100 rpm, and the molten fibers were wound up for approximately 5 min on a TeflonÒ FEP tube (ID = 3 mm, OD = 5 mm) mounted on a rotating stainless steel mandrel. 2.3. Electrospinning The custom-designed electrospinning apparatus consisted of a high-voltage power supply (Gamma High Voltage Research, Inc.), an infusion pump (New Era Pump System, Inc.), a plastic syringe, a stainless steel blunt-ended needle (20 gauge) and a metal mandrel collector, as described previously [24]. The syringe was mounted horizontally on the infusion pump and the sample solution was fed at a constant rate through the syringe to the needle tip. The distance between the needle tip and the collector was maintained at 15 cm. The applied voltage to the needle tip was 10–14 kV and the flow rate of the solution was 0.5–1.0 ml h1, depending on the solvent system. The two solvents used for dissolving the polymer were acetone (Sigma–Aldrich, USA) and HFIP (Sigma–Aldrich, USA). Based on preliminary trials [24], the optimal polymer concentration for electrospinning using acetone as the solvent was determined to be 15% (w/v), whereas, using HFIP as the solvent, the optimal spinnable concentration was 9% (w/v). 2.4. Scanning electron microscopy (SEM)

2. Materials and methods

In order to determine the morphology and diameter of the filaments, the scaffolds were viewed under a scanning electron microscope. Images were acquired from a JEOL JSM 5900-LV scanning electron microscope using an accelerating voltage of 15 kV. Specimens were mounted on aluminum stubs using conductive carbon tape. They were then coated with gold/palladium using a Hummer™ 6.2 Sputter Coating System (Anatech, CA, USA) to obtain a conductive coating about 100 Å thick.

2.1. Materials

2.5. Pore size and fiber diameter measurement by image analysis

PLCL copolymer was received as a solid bulk polymer from the Biomaterials Research Center at the Korea Institute of Science and Technology. The polymer had been synthesized following the

The SEM images of melt-spun filaments and electrospun webs were analyzed by measuring the fiber diameters and pore size distributions. The measurements were made using the java-based

1960

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

image processing program Image J (NIH). The filaments were analyzed manually, by measuring the diameters of >100 randomly selected fibers. To measure the pore size distribution >100 randomly selected areas between fibers were measured manually from the captured images. Before undertaking these measurements, the contrast and threshold of each image was optimized so as to ensure that the fibers in the top layers were clearly in focus. Filament diameter and pore size values are reported as means ± standard deviations. 2.6. Porosity Since the overall porosity of the web could not be measured directly, it was calculated by an indirect approach using the density of the tubular scaffold and the density of the polymer. The porosity was reported in percent as means ± standard deviation values:

porosity ðPÞ ¼ f1  ðds =dp Þg  100     ms ¼ 1 dp  100 vs * ( !, )+ ms  100 dp ¼ 1 s  ðR2  r 2 Þ  ‘

ð1Þ

where ds is the density of the scaffold (g cm3), dp is the density of the polymer (g cm3), ms is the mass of the scaffold (g), vs is the volume of the scaffold (cm3), R is the outer diameter of the tubular scaffold (cm), r is the inner diameter of the tubular scaffold (cm) and ‘ is the length of the scaffold (cm). The density of the polymer (dp) was reported as 1.21 g cm3 for the 50:50 PLCL polymer [19]. The mass of the specimens was measured after they had been cut into 5 mm lengths. 2.7. Mechanical testing Transverse tensile mechanical testing of the melt-spun and electrospun tubular structures was performed on an MTS ReNew Model 1122 system (Eden Prairie, MN, USA). For the electrospun tubes, 10 and 250 N load cells were used with a crosshead speed of 0.5 mm s1, whereas for the melt-spun tubes, a 250 N load cell was used with a crosshead speed of 3.5 mm s1. All specimens were tested to failure. A custom made mounting frame was designed and used to mount the tubular structures on the tensile tester (Fig. 1). The values for peak load (gf) at failure were converted into peak stress (MPa) using the following equation:

peak stress ðMPaÞ ¼ force=areaðN mm2 Þ ¼

0:0098  loadðgÞ thicknessðmmÞ  lengthðmmÞ  2

The porosity of the specimens was also considered in this calculation since the void areas of the pores are not load bearing. The values of peak elongation at failure (mm) were converted into peak strain (%) using the following equation:

Compliance ðml mm Hg1 Þ ¼ DV=DP ¼ change in volumeðmlÞ=change in pressureðmm HgÞ

All the experimental mechanical measurements are reported in Section 3 as mean ± standard error values. 2.8. Cytotoxicity and cell viability Electrospun samples for biological testing were first cleaned three times with deionized water followed by isopropanol rinsing, then stored under vacuum overnight. Samples were then sterilized by ethylene oxide. Specimens measuring 0.5 cm  0.5 cm were placed into empty wells of a 96-well culture plate. NIH 3T3 fibroblast cells (ATCC) were then seeded into the wells at a density of 10,000 cells per well. The plate was incubated at 37 °C for up to 2 weeks. After 1, 3, 7 and 14 days of incubation, the cytotoxicity and cell viability were determined using the lactate dehydrogenase (LDH) and water-soluble tetrazolium salt (WST) assay (Dojindo, Japan), respectively. The LDH cytotoxicity assay quantifies cytotoxicity based on the measurement of activity of LDH released from damaged cells. The WST cell proliferation assay is a colorimetric assay that is based on the cleavage of a tetrazolium salt by mitochondrial dehydrogenases to form formazan in viable cells [25].

3. Results and discussion 3.1. Melt-spun scaffolds 3.1.1. Morphological properties The melt-spun monofilaments were wound up directly after extrusion on a rotating Teflon-coated mandrel and were removed as a tube after collection once they had cooled to room temperature (Fig. 2). The inner diameter of the tubular scaffolds was 5 mm. The tubes were then cut up in 5 mm lengths for SEM and mechanical testing. When removed from the Teflon-coated mandrel, the tubular construct maintained its shape and integrity because the translucent monofilaments had bonded well to each other. This is the first report of the conversion of this particular copolymer into continuous filaments, which were wound to form a tubular construct in a single process. As shown in the SEM photomicrograph in Fig. 2, the average diameter of the melt-spun monofilaments was 253 ± 36 lm and the winding angle was ±10°. As the filaments cooled after extrusion, cold drawing and increased orientation occurred, as evidenced by visual necking of the filaments prior to wind-up. The filaments were still in a softened state when they reached the wind-up unit as they bonded well to each other on contact. The average porosity of the melt-spun tubes was 76.17 ± 0.76%. Since the pore geometry can be modulated by the wind-up angle, the pore size distribution can be controlled depending on the end use. Modulating the micro- to macroscale morphology is important when designing a scaffold, since individual cells can recognize comparable dimensions in their environment [26] and the bulk mechanical properties can be influenced by the scale of organization.

strain ð%Þ ¼ change in lengthðmmÞ  100=initial gauge lengthðmmÞ The initial transverse tensile modulus of each specimen was determined from the slope of the initial straight line portion of the stress/strain curve:

initial modulus ðMPaÞ ¼ stressðMPaÞ=strain The average values for compliance of the tubes were theoretically estimated by the following equation:

3.1.2. Mechanical properties Overall, the peak stress and strain values for the melt-spun tubes exceeded those of natural arteries. The peak tensile stress and strain values for human arterial tissues (brachial and popliteal arteries) are in the range of 0.78–1.37 MPa and 65–83%, respectively [27]. As shown in Table 2 and Fig. 7, the transverse breaking stress for the melt-spun tubes was 26.1 ± 1.2 MPa, the initial modulus was 23.5 ± 0.9 MPa and the breaking extension was high, at 578.2 ± 17.1%. This value should be compared to that reported

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

1961

Fig. 1. Custom-made clamping frame (A) and schematic showing area calculation (B).

for PLCL scaffolds fabricated by an extrusion-salt particulate leaching method, which have an elongation of more than 200% [22]. The estimated compliance value was 0.0159 ml mm Hg1, which was of the same order of magnitude as natural arteries [28]. Since the melt-spun tubes were prepared by winding the monofilaments continuously and directly from the spinneret without a secondary drawing process, the very high breaking extensions were achieved in part because the filaments experienced some degree of fiber drawing and polymer chain alignment prior to reaching their ultimate failure point. Fiber stiffness and initial tensile modulus depend on the levels of crystallinity and orientation of the polymer chains within the fiber’s microstructure. Such microstructural features are influenced by the fiber-spinning conditions and, in particular, by the level of drawing experienced by the threadline as it leaves the spinneret hole.

3.2. Electrospun scaffolds 3.2.1. Morphological properties Two types of tubular constructs with an inner diameter of 5 mm were produced using two different solvent systems. Due to the rapid evaporation of acetone, the polymer tended to clog the needle and form unstable multiple jets. This required manual unblocking, leading to a low spinning efficiency. On the other hand, when HFIP was used as the solvent, fibers could be collected continuously without manual intervention. Hence, it was possible to improve the ease of processing and the spinning efficiency with a more stable jet (Fig. 3). Representative photomicrographs of tubes spun from HFIP are presented in Fig. 4. After the tubular scaffolds had been spun, they were removed from the mandrel and cut into small pieces for characterization. The appearance of the electrospun tubes was opaque, white and film-like. To assist with the

1962

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

Fig. 2. Macroscopic image and SEM photomicrograph of the melt-spun PLCL tubes.

Table 1 Porosity of electrospun PLCL tubes.

a

Electrospun tubes (length: 5 mm)

Mass (g)

Thickness (mm)

Volume (mm3)

Density (g/cm3)

Calculated porositya (%)

Acetone HFIP

0.00112 0.00442

0.033 0.118

5.94 23.31

0.188 0.189

84.46 84.38

Calculated porosity assuming density of PLCL = 1.21 g/cm3 [19].

removal process, the samples were immersed in a 1% aqueous Triton X detergent solution for 1 h. On removal, the samples maintained both their tubular shape and mechanical integrity. Representative SEM photomicrographs comparing the morphology of the electrospun scaffolds with two different solvents are shown in Fig. 5. The fibers appear to be interconnected with various sizes and shapes of pores. The acetone spun scaffolds consisted of fibers with diameters ranging from 200 to 800 nm, with a normal unimodal distribution and an average diameter of 540 ± 160 nm. However, when HFIP was used as the solvent, the average diameter was larger, at 840 ± 210 nm, and the diameters ranged from 400 to 1200 nm in a skewed distribution (Fig. 6). Generally, the pores in both types of construct were roughly triangular or oval in shape, forming interconnected networks. The pore size distribution for

Fig. 3. Jet formation during electrospinning. (A) Acetone solvent; (B) HFIP solvent.

the acetone spun scaffolds showed that most of the pores were in the range of <1–10 lm2 and the mean pore area was 2.79 ± 2.20 lm2, whereas the HFIP spun scaffolds showed a larger range of pores, in the range of 1–16 lm2, with a mean pore area of 5.31 ± 4.50 lm2. The large values for the standard deviation not only indicate a wide range of pore sizes, they also remind us of the limitation associated with measuring only a two-dimensional pore size by image analysis since the actual resolution is limited to the surface images provided by SEM. Previously reported pore size dimensions for electrospun PLCL webs are in a similar range, namely from 0.2 to 30 lm [19]. Electrospinning has recently received much attention with regard to healthcare applications, especially biomedical and tissue engineering end uses [19,20,29–36], providing an alternative approach for the fabrication of unique matrices and scaffolds. The nanoscale diameter of the fibers and the structure of the nonwoven web resemble certain supramolecular features of extracellular matrix (ECM) [33]. Since the fibers, pores, ridges and grooves in the basement membrane of ECM all have dimensions on the nanoscale, this characteristic is especially important for the design of blood vessel tissue engineering scaffolds because the monolayer of endothelial cells in native blood vessels grows directly on the basement membrane [37]. The small fiber diameter provides a high surface area to volume ratio, which, together with a high length to diameter ratio, provides favorable parameters for cell attachment, growth and proliferation. It is hypothesized that the large surface area of nanofibers with specific surface chemistries facilitates the attachment of cells and controls their cellular functions [38]. The results for the total porosity of the electrospun tubes are reported in Table 1. The overall porosities for tubes spun from both solvents are similar, at approximately 80%. These results exceed the previously reported values for electrospun PLCL webs, which were in the range of 56–63% [19].

3.2.2. Mechanical properties Again, the peak stress and strain values for the two types of electrospun tubes exceeded those of natural arteries. As shown in

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

1963

Fig. 4. Electrospun tubes from PLCL copolymer spun by HFIP. (A) Macroscopic view; (B, C, and D) SEM photomicrographs of the cross-sectional view.

Fig. 5. SEM photomicrographs showing the morphology of PLCL submicron fibers. (A) Acetone solvent; (B) HFIP solvent.

Table 2 and Fig. 7, the transverse breaking stress for the electrospun tubes using acetone as the solvent was 17.8 ± 2.0 MPa and using HFIP was 44.8 ± 3.5 MPa. This maximum stress value of over 40 MPa from the HFIP electrospun tube is impressive compared to the values that have been reported for PLCL scaffolds made by particulate extrusion, which are only 0.80 MPa [13]. Peak transverse tensile strains for both electrospun tubes exceeded 140%, thus confirming that they have similar compliance to that of natural vessels [28]. The stress and strain values were different for electrospun tubes using different solvent systems, confirming that not only the morphology, but also the mechanical properties are dependent

on the choice of solvent and electrospinning conditions. In contrast, the estimated compliance values were similar for both solvent systems, as shown in Table 2. The tube electrospun from acetone had an initial modulus of 24.6 ± 1.9 MPa which was similar to the melt-spun tube of 23.5 ± 0.9 MPa, suggesting that the initial deformation is caused by an inherent material property of the polymer regardless of fiber diameter. However, the initial modulus of the electrospun tube using HFIP as the solvent was relatively low at 9.34 ± 0.59 MPa, indicating that this tube was not as stiff as the other two. Clearly the drawing or attenuation of the fibers in these three types of

1964

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

Fig. 6. Fiber diameter (A) and pore size distribution (B) of PLCL fibers electrospun from HFIP and acetone.

Fig. 7. Breaking stress (A), breaking strain (B) and initial modulus (C) of PLCL tubes.

Table 2 Comparison of transverse mechanical properties of PLCL tubes. Transverse tensile breaking properties

Natural arteries Melt-spun tubes Electrospun using acetone Electrospun using HFIP

Stress (MPa)

Strain (%)

Initial modulus (MPa)

Estimated compliance (ml mm Hg1)

0.78–1.37 [22] 26.1 ± 1.2 17.8 ± 2.0 44.8 ± 3.5

65–83 [22] 578.2 ± 17.1 142.3 ± 11.7 362.2 ± 9.7

– 23.5 ± 0.9 24.6 ± 1.9 9.34 ± 0.59

0.0080 [28] 0.0159 0.052 0.053

spun tubes were different, which is reflected in a difference of microstructure and polymer orientation, even though the starting 50:50 PLCL copolymer was the same raw material. The initial tensile modulus of compression-molded 50:50 PLCL copolymer film reported in the literature is only 0.6 MPa, which is substantially lower than the values obtained in this study [39]. In addition, Kwon et al. [19] reported the Young’s modulus of electrospun PLCL scaffolds with a thickness of approximately 140 lm to be in the range of 0.8–2.2 MPa, which is also significantly lower than the experimental results reported here. In summary, Fig. 7 shows that the electrospun tubes using HFIP as the solvent exhibited the highest breaking stress and lowest stiffness.

3.2.3. Cytotoxicity and cell viability To investigate the presence of residual solvent and its effect to cell behavior on the tubular constructs after electrospinning, cell adhesion and proliferation studies were performed using WST and LDH assays, respectively, followed by SEM observation (Figs. 8 and 9). As reported in Fig. 9, both solvent systems resulted in cell death decreasing after 3 days of culture and cell viability increasing after 7 days. This suggests that, regardless of the type of solvent used during electrospinning, cell proliferation is initially slow, but increases with time. These results indicate that electrospun structures can

promote cell proliferation and that the effect of any residual solvent on cell behavior is not significantly detrimental.

3.3. Combining melt spinning and electrospinning for vascular tissue engineering Melt-spun tubes were prepared for use as the base mandrel for the collection of electrospun PLCL nanofibers. Photomicrographs of the resulting structures are seen in Fig. 10. Despite the concern that the nanofiber web would not adhere to the melt-spun filaments, the combination of electrospinning and melt spinning of 50:50 PLCL copolymer to fabricate a multilayered tubular construct was successfully demonstrated. The two layers were well bonded together and were not separable with manual manipulation. Micro- and nanocombined structures can be advantageous, since the nanolayer can mimic the ECM, whereas the microlayer will provide larger pores which facilitate superior cell infiltration [40]. There have been several approaches combining electrospinning techniques to obtain a multilayered structure [33,40,41]. Tuzlakoglu et al. [42] previously reported electrospinning starch based materials on top of a flat fibrous mesh for bone tissue engineering applications, and Martins et al. [43] combined rapid prototyping and electrospinning techniques to achieve multilayered structures. This multiscaled

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

Fig. 8. SEM photomicrographs of cell attachment on PLCL scaffolds after 7 days of culture.

Fig. 9. WST and LDH assays of electrospun PLCL tubes.

Fig. 10. Bi-layered PLCL tube showing both electrospun and melt-spun fibers. (A) 70; (B) 200.

1965

1966

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

technique to fabricating a 3-D matrix structure is an attractive approach to mimicking living systems since nature often derives properties from multi-scalar and hierarchical structures [26]. In this work we have combined two 3-D concentric tubular layers using a rotating mandrel for both techniques, enabling the nanolayer to adhere completely to the microlayer. This combined structure is particularly attractive for vascular tissue engineering applications, since natural blood vessels consist of three layers (tunica intima, media and adventitia), each with its own respective cells (endothelial cells, smooth muscle cells and fibroblasts). By designing a multilayered tube, it is now possible to seed or co-culture different cell lines in layers with controlled orientation. Fig. 10 shows that the layer of horizontally aligned electrospun fibers was bonded to the vertically oriented melt-spun macrofibers. These two types of fibers are easily distinguished, since the horizontal electrospun submicron fibers have diameters in the 400 nm to 2 lm range, whereas the vertical melt-spun macrofibers had diameters in the range of 120– 250 lm. This phenomenon of aligned electrospun fibers is interesting because the rotation speed of the mandrel was much lower than 1000 rpm, which was previously claimed to be the minimum speed necessary in order to observe some fiber alignment [44]. It now appears that the aligned and wound macrofibers were acting as electrodes when the electrospun fibers were collected, and that this method has already been used as one of the ways to align nanofibers while electrospinning [45,46]. This degree of alignment may provide more options for controlling the cell orientation and tissue growth on the scaffolds [47,48]. It has been reported previously that cells cultured on electrospun fiber scaffolds tend to proliferate in the direction of fiber alignment [49]. For example, in native blood vessels the shear stresses caused by blood flow on the luminal surface align the endothelial cells in the direction of the blood flow, whereas within the wall of the blood vessels the smooth muscle cells in the tunica media tend to be aligned concentrically [37]. Therefore, the alignment of fibers in a circumferential direction around a tube is the preferred architecture for tissue engineering blood vessels. Successfully electrospinning submicron fibers on top of the meltspun fibers provides opportunities to fabricate scaffolds with sufficient mechanical strength and surface architecture on the nanoscale. Controlling the electrospinning time and the fiber alignment are two challenges that the further development of this novel approach must face. By modifying the sequence of fabrication steps, it will be possible to obtain different hierarchical layers, each layer providing unique and distinct characteristics to the scaffold matrix. This approach to controlling multiple levels of the structural design of scaffolds will enable more precise mimicking of the natural living system. 4. Conclusions A 50:50 PLCL copolymer was successfully melt-spun and electrospun to form individual and combined porous tubular scaffolds. This is the first time that this particular elastomeric biodegradable copolymer has been melt-spun. Using two alternative solvent systems – acetone and HFIP – for electrospinning provided different results in terms of fiber dimensions, mechanical properties and cytotoxicity, but the HFIP solvent was preferred as it gave a more stable threadline. The mechanical properties of both types of tubes demonstrated greater strength and compliance than natural arteries of equivalent caliber. For future vascular applications, the two-layered tubular scaffold containing both fine electrospun and thicker melt-spun fibers will facilitate the formation of distinctive endothelial and smooth muscle tissue layers, hence mimicking the structure of native vessels. Acknowledgements The authors acknowledge the financial support from the Boston Scientific Corporation and the College of Textiles at North Carolina

State University. We also thank Dr. Bhupender Gupta, Dr. Ajit Moghe and Hai Bui for their insights and technical assistance. Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figures 1–4, are difficult to interpret in black and white. The full colour images can be found in the on-line version, at doi: 10.1016/j.actbio.2009.12.007. References [1] Fuchs JR, Nasseri BA, Vacanti JP. Tissue engineering: a 21st century solution to surgical reconstruction. Ann Thorac Surg 2001;72:577–91. [2] Nerem RM, Seliktar D. Vascular tissue engineering. Annu Rev Biomed Eng 2001;3:225–43. [3] Allison MA, Ho E, Denenberg JO, Langer RD, Newman AB, Fabsitz RR, et al. Ethnic-specific prevalence of peripheral arterial disease in the United States. Am J Prev Med 2007;32:328–33. [4] Rabkin E, Schoen FJ. Cardiovascular tissue engineering. Cardiovasc Pathol 2002;11:305–17. [5] Brewster LP, Bufallino D, Ucuzian A, Greisler HP. Growing a living blood vessel: insights for the second hundred years. Biomaterials 2007;28:5028–32. [6] Hoerstrup SP, Zund G, Sodian R, Schnell AM, Grunenfelder J, Turina MI. Tissue engineering of small caliber vascular grafts. Eur J Cardiothorac Surg 2001;20:164–9. [7] Niklason LE, Langer RS. Advances in tissue engineering of blood vessels and other tissue. Transpl Immunol 1997;5:303–6. [8] Hutmacher DW. Scaffold design and fabrication technologies for engineering tissues-state of the art and future perspectives. J Biomater Sci Polym Ed 2001;12:107–24. [9] Griffith LG. Emerging design principles in biomaterials and scaffolds for tissue engineering. Ann NY Acad Sci 2002;961:83–95. [10] Yang S, Leong K-F, Du Z, Chua C-K. The design of scaffolds for use in tissue engineering. Part 1. Traditional factors. Tissue Eng 2001;7:679–89. [11] Matsuda T. Recent progress of vascular graft engineering in Japan. Artif Organs 2004;28:64–71. [12] Salgado AJ, Coutinho OP, Reis RL. Bone tissue engineering: state of the art and future trends. Macromol Biosci 2004;4:743–65. [13] Jeong SI, Kim SH, Kim YH, Jung Y, Kwon JH, Kim B-S, et al. Manufacture of elastic biodegradable PLCL scaffolds for mechano-active vascular tissue engineering. J Biomater Sci Polym Ed 2004;15:645–60. [14] Jeong SI et al. Mechano-active tissue engineering of vascular smooth muscle using pulsatile perfusion bioreactors and elastic PLCL scaffolds. Biomaterials 2005;26:1405–11. [15] Kim B-S, Mooney DJ. Scaffold for engineering smooth muscle under cyclic mechanical strain conditions. J Biomech Eng 2000;122:210–5. [16] Kim B-S, Nikolovski J, Bonadio J, Mooney DJ. Cyclic mechanical strain regulates the development of engineered smooth muscle tissue. Nat Biotechnol 1999;17:979–83. [17] JHd Groot, Zijlstra FM, Kuipers HW, Pennings AJ, Klompmaker J, Veth RPH, et al. Meniscal tissue regeneration in porous 50/50 copoly(L-lactide/ e-caprolactone) implants. Biomaterials 1996;18:613–22. [18] Hiljanen-Vainio M, Karjalainen T, Seppala J. Biodegradable lactone copolymers. 1. Characterization and mechanical behavior of e-caprolactone and lactide copolymers. J Appl Polym Sci 1996;59:1281–8. [19] Kwon IK, Kidoaki S, Matusda T. Electrospun nano- to microfiber fabrics made of biodegradable copolyesters: structural characteristics, mechanical properties and cell adhesion potential. Biomaterials 2005;26:3929–39. [20] Mo XM, Xu CY, Kotaki M, Ramakrishna S. Electrospun P(LLA-CL) nanofiber: a biomimetic extracellular matrix for smooth muscle cell and endothelial cell proliferation. Biomaterials 2004;25:1883–90. [21] Inoguchi H, Kwon IK, Inoue E, Takamizawa K, Maehara Y, Matsuda T. Mechanical responses of a compliant electrospun poly(L-lactide-co-epsiloncaprolactone) small-diameter vascular graft. Biomaterials 2006;27:1470–8. [22] Jeong SI, Kim B-S, Lee YM, Ihn KJ, Kim SH, Kim YH. Morphology of elastic poly(L-lactide-co-E-caprolactone) copolymers and in vitro and in vivo degradation behavior of their scaffolds. Biomacromolecules 2004;5:1303–9. [23] Jung Y, Kim SH, You HJ, Kim S-H, Ha Kim Y, Min BG. Application of an elastic biodegradable poly(L-lactide-co-e-caprolactone) scaffold for cartilage tissue regeneration. J Biomater Sci Polym Ed 2008;19:1073–85. [24] Chung S, Moghe AK, Montero GA, Kim S-H, King MW. Nanofibrous scaffolds electrospun from elastomeric biodegradable poly(L-lactide-co-e-caprolactone) copolymer. Biomed Mater 2009;4:015019. [25] Baek MO, Kim SZ, So JW, Roh HW, Lee NR. Cytotoxicity measurements of tissue-engineered scaffold. Tissue Eng Regen Med 2007;4:571–6. [26] Lee J, Cuddihy MJ, Kotov NA. Three-dimensional cell culture matrices: state of art. Tissue Eng B 2008;14:61–86. [27] Yamada H. Strength of biological materials. Baltimore, MD: Williams & Wilkins; 1970. [28] Romney JS, Lewanczuk RZ. Vascular compliance is reduced in the early stages of type 1 diabetes. Diabetes Care 2001;24:2102–6. [29] Boland ED, Wnek GE, Simpson DG, Pawlowski KJ, Bowlin GL. Tailoring tissue engineering scaffolds using electrostatic processing techniques: a study of

S. Chung et al. / Acta Biomaterialia 6 (2010) 1958–1967

[30]

[31] [32] [33]

[34]

[35] [36]

[37] [38] [39]

poly(glycolic acid) electrospinning. J Macromol Sci Pure Appl Chem 2001;A38: 1231–43. Deitzel JM, Kleinmeyer J, Harris D, Tan NCB. The effect of processing variables on the morphology of electrospun nanofibers and textiles. Polymer 2001;42: 261–72. Doshi J, Reneker DH. Electrospinning process and applications of electrospun fibers. J Electrostat 1995;35:151–60. Fong H, Chun I, Reneker DH. Beaded nanofibers formed during electrospinning. Polymer 1999;40:4585–92. Kidoaki S, Kwon IK, Matsuda T. Mesoscopic spatial designs of nano- and microfiber meshes for tissue engineering matrix and scaffold based on newly devised multilayering and mixing electrospinning techniques. Biomaterials 2005;26:37–46. Yoshimoto H, Shin YM, Terai H, Vacanti JP. A biodegradable nanofiber scaffold by electrospinning and its potential for bone tissue engineering. Biomaterials 2003;24:2077–82. Ramakrishna S, Fujihara K, Teo W-E, Yong T, Ma Z, Ramaseshan R. Electrospun nanofibers: solving global issues. Mater Today 2006;9:40–50. Li W-J, Laurencin CT, Caterson EJ, Tuan RS, Ko FK. Electrospun nanofibrous structure: a novel scaffold for tissue engineering. J Biomed Mater Res 2002;60: 613–21. Ma Z, Kotaki M, Inai R, Ramakrishna S. Potential of nanofiber matrix as tissue engineering scaffolds. Tissue Eng 2005;11:101–9. Venugopal J, Ramakrishna S. Applications of polymer nanofibers in biomedicine and biotechnology. Appl Biochem Biotechnol 2005;125:147–58. Jeon O, Lee S-H, Kim SH, Lee YM, Kim YH. Synthesis and characterization of poly(L-lactide)–poly(e-caprolactone) multiblock copolymers. Macromolecules 2003;36:5585–92.

1967

[40] Pham QP, Sharma U, Mikos AG. Electrospinning of polymeric nanofibers for tissue engineering applications: a review. Tissue Eng 2006;12:1197–211. [41] Vaz CM, Tuijl SV, Bouten CV, Baaijens FP. Design of scaffolds for blood vessel tissue engineering using a multi-layering electrospinning technique. Acta Biomater 2005;1:575–82. [42] Tuzlakoglu K, Bolgen N, Salgado AJ, Gomes ME, Piskin E, Reis RL. Nano-and micro-fiber combined scaffolds: a new architecture for bone tissue engineering. J Mater Sci Mater Med 2005;16:1099–104. [43] Martins A, Chung S, Pedro AJ, Sousa RA, Marques AP, Reis RL, et al. Hierarchical starch-based fibrous scaffold for bone tissue engineering applications. J Tissue Eng Regen Med 2009;3:37–42. [44] Teo WE, Kotaki M, Mo XM, Ramakrishna S. Porous tubular structures with controlled fibre orientation using a modified electrospinning method. Nanotechnology 2005;16:918–24. [45] Bhattarai N, Edmondson D, Veiseh O, Matsen FA, Zhang M. Electrospun chitosan-based nanofibers and their cellular compatibility. Biomaterials 2005;26:6176–84. [46] Park SA, Park K, Yoon H, Son JG, Min T, Kim GH. Apparatus for preparing electrospun nanofibers: designing an electrospinning process for nanofiber fabrication. Polym Int 2007;56:1361–6. [47] Murugan R, Ramakrishna S. Design strategies of tissue engineering scaffolds with controlled fiber orientation. Tissue Eng 2007;13:1845–66. [48] Xu CY, Inai R, Kotaki M, Ramakrishna S. Electrospun nanofiber fabrication as synthetic extracellular matrix and its potential for vascular tissue engineering. Tissue Eng 2004;10:1160–8. [49] Xu CY, Inai R, Kotaki M, Ramakrishna S. Aligned biodegradable nanofibrous structure: a potential scaffold for blood vessel engineering. Biomaterials 2004;25:877–86.