Endothelial cell scaffolds generated by 3D direct writing of biodegradable polymer microfibers

Biomaterials 32 (2011) 1872e1879

Contents lists available at ScienceDirect

Biomaterials journal homepage: www.elsevier.com/locate/biomaterials

Endothelial cell scaffolds generated by 3D direct writing of biodegradable polymer microfibers Scott M. Berry a,1, Sean P. Warren b, DeVonnah A. Hilgart b, Adam T. Schworer b, Santosh Pabba c, 2, Andrea S. Gobin b, Robert W. Cohn c, Robert S. Keynton a, b, c, * a b c

Department of Mechanical Engineering, University of Louisville, Louisville, KY, USA Department of Bioengineering, University of Louisville, Louisville, KY, USA The Electro Optics Research Institute and Nanotechnology Center, University of Louisville, Louisville, KY, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 8 September 2010 Accepted 14 November 2010 Available online 8 December 2010

The engineering of large (thickness > 100 mm) tissues requires a microvascular network to supply nutrients and remove waste. To produce microvasculature in vitro, a scaffold is required to mechanically support and stimulate endothelial cell (EC) adhesion and growth. Scaffolds for ECs are currently produced by patterning polymers or other biomaterials into configurations which often possess isotropic morphologies such as porous films and fibrous mats. We propose a new “direct-write” process for fabricating scaffolds composed of suspended polymer microfibers that are precisely oriented in 3D, providing directional architecture for selectively guiding cell growth along a desired pathway. The diameters of the fibers produced with this process were predictably and repeatably controlled through modulation of the system parameters, enabling production of fibers with microvascular-scale diameters (5e20 mm) from a variety of biodegradable polymers. These scaffolds were successfully seeded with ECs, which conformed to the geometry of the fibers and proliferated over the course of one week. Ó 2010 Elsevier Ltd. All rights reserved.

Keywords: Direct-write Endothelial cell scaffold Microfibers Biodegradable materials Tissue engineering

1. Introduction The capability to selectively promote generation of new blood vessels from existing vessels or precursor cells is a critical consideration in the emerging field of tissue engineering. Currently, the only commercially engineered tissues are skin and cartilage, both of which can be manufactured without the requirement for vascular or neural networks [1]. The development of complex microvascular networks is an essential prerequisite for the establishment of more advanced and larger tissues, which require adjacent capillary networks to provide delivery of nutrients and removal of waste products. However, such networks cannot be generated spontaneously; they require three-dimensional mechanical and chemical support, provided in vivo by the extracellular matrix (ECM). The

* Corresponding author. 419 Lutz Hall, Department of Bioengineering, University of Louisville, Louisville, KY 40292, USA. Tel.: þ1 502 852 6356; fax: þ1 502 852 6806. E-mail addresses: [email protected] (S.M. Berry), santosh.pabba@ intel.com (S. Pabba), [email protected] (R.S. Keynton). 1 Present address: Department of Biomedical Engineering, University of WisconsineMadison, 1111 Highland Ave., WIMR 6th Floor, Madison, WI 53726, USA. 2 Present address: D1C Litho-Immersion Scanner, Intel Corporation, Hillsboro, OR, USA. 0142-9612/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2010.11.023

development of synthetic cellular scaffolds provides a potential alternative by mimicking the relevant properties of the ECM required to initiate and sustain neovascularization. Additionally, the ability to generate isolated microvasculature in vitro would enable detailed evaluation of angiogenic phenomena, including studies of neovascularization during wound healing, cancerous growth, or rheumatoid arthritis inflammation [2]. A synthetic cellular scaffold must meet several design criteria. First, the scaffold must be capable of guiding and supporting endothelial cell (EC) growth and differentiation. Second, the scaffold must allow access to the cells, such that chemical (growth factors) and/or mechanical (shear stress) stimuli can be applied as required, thereby promoting development into functional microvasculature. Lastly, the scaffold must maintain its structural integrity during proliferation of the EC layer until such a time as the developing microvasculature possesses sufficient robustness to be mechanically self-supportive [3,4]. The scaffold morphology must be taken into consideration as scaffolds with micro- or nanoscale architecture have been shown to influence cellular growth through the modulation of cell shape, adhesion, and growth [5e7]. The utilization of fibrous polymer structures as synthetic cell scaffolds has been previously demonstrated by several groups [8e22]. These scaffolds have been produced in a myriad of 3D geometries including electrospun nanofibrous mats

S.M. Berry et al. / Biomaterials 32 (2011) 1872e1879

1873

Fig. 1. Schematic of the direct-write fiber fabrication process.

[8e13], fibrous monoliths [10], suspended microfibers [18e22], and combinations thereof [15e17]. Furthermore, many of these materials have been augmented with bioactive molecules in order to enhance cellular interaction with the scaffolds [14,17,20]. While electrospinning remains a popular methodology for generating biodegradable fibrous scaffolds, a variety of other fiber drawing techniques have been harnessed to produce cellular scaffolds. Vozzi et al. have employed a pressure-assisted microsyringe to “write” stacked arrays of lines of PLGA with widths of w20 mm [18]. Kullenberg et al. utilized this process to develop a cellular scaffold for neuroblastoma cells and found that cellular metabolism increased three fold on a multi-layer scaffold relative to a planar design [19]. Hadjizadeh et al. functionalized commercially-available 100 mm-diameter poly (ethylene terephtalate) fibers with cellular adhesion ligands to pattern cylinders of ECs [20] while Hwang et al. cultured fibroblasts on microscale (20e230 mmdiameter) wet spun fibers [21]. Nain et al. used arrays of aligned microfibers to influence cellular behavior of mouse C2C12 progenitor cells, finding that cellular proliferation, migration, and differentiation is manipulated by scaffold morphology [22]. As cellular behavior is directly linked to scaffold architecture, the development of new mechanisms to generate fibrous scaffolds with a high degree of control is of critical importance to the elucidation of next-generation scaffolds. Recently, our group and others have developed a technique through which suspended microscale and sub-microscale polymer fibers are precisely fabricated on a substrate [23,24]. In this process, a viscous (h z 5000 to 30,000 cP) polymer solution is loaded into a syringe controlled by a programmable micromilling machine (Dover Instruments Inc.). The syringe is pressurized to expel the solution from the needle into contact with a substrate to form the initial fiber attachment point (Fig. 1a). The syringe is then lifted, translated, and descended to establish a second attachment point (Fig. 1b,c). The result of this motion is a liquid filament of solution suspended between the two attachment points. This filament thins due to surface tensiondriven necking as it dries, resulting in a stable fiber with a diameter well below that of the syringe needle. This automated process can be repeated several times in rapid succession to produce an array of fibers in a “connect-the-dots” fashion. Because the final diameter of the fiber is determined by the degree of surface tension-driven thinning that occurs prior to solidification (Fig. 1b), fiber diameter can be predictably influenced through the modulation of the material properties of the solution, including viscosity, surface tension, and fiber solidification rate (e.g. solvent evaporation rate) [25]. Of these three material properties, viscosity can most easily be manipulated over several orders of magnitude, specifically through variation in polymer solution concentration or polymer molecular weight, thereby enabling statistically significant control over fiber diameter [23].

The ability to accurately control both fiber diameter and 3D fiber orientation makes the direct-write system ideal for producing microfibrous arrays when precise control of array architecture is desired. Also, the previously-demonstrated ability [23] to repeatably produce suspended fibers with microvascular-scale diameters indicates that the direct-write technique may be an attractive alternative to other fiber fabrication processes in the production of microvascular cell scaffolds. Therefore, the purpose of this investigation is to utilize the direct-write system to fabricate microvascular-scale scaffolds and evaluate the functionality of these structures in terms of cellular adhesion, proliferation, and scaffold degradation. 2. Materials and methods 2.1. Solution preparation and processability characterization Polymer solutions were prepared by dissolving a variety of biodegradable polymers (LACTEL Absorbable Polymers, Cupertino, CA) in organic solvents. These included crystalline (L) and amorphous (DL) forms of PLA, multiple compositions (50:50 and 75:25) of PLGA, and PCL as described by Table 1. Solvents (Sigma Aldrich, St. Louis, MO) were selected based on their ability to dissolve elevated quantities (>30% by wt.) of polymer at room temperature. Preliminary direct-write experiments were performed with multiple concentrations of each polymer/solvent pair to determine the concentration ranges that possess the requisite viscosity to be successfully directly-written into fibers. Solutions from the high and low ends of these ranges were then processed into test arrays of fibers with lengths of 16 mm, which were imaged via electron microscopy (Zeiss Supra 35VP, Thornwood, NY) to quantify direct-write process yield and determine fiber diameters. The process yield was determined by making 48 fiber drawing attempts and observing the fraction of attempts that resulted in an unbroken fiber; whereas, the diameters of 16 individual fibers were measured in three locations (at the midpoint and w200 mm from each end). 2.2. Biodegradation rates of fibrous scaffolds Test arrays containing 8 to 11 fibers of each polymer on polystyrene frames were constructed with the direct-write apparatus and weighed to determine scaffold mass. Each test array was placed in a Petri dish and covered with 3 mL of phosphate buffered saline (PBS), then incubated at 37  C for a period of 1e6 weeks. PBS was refreshed on a weekly basis to prevent accumulation of the acidic biodegradation products which would influence the hydrolysis reaction rate. At the end of each

Table 1 Polymer solutions utilized with the direct-write process.(Note: 1,2 Dichlorethane is abbreviated 1,2 DCE). Polymer

Abbreviation Intrinsic Viscosity Solvent (dL/g)

L-PLA Poly(L-lactide) DL-PLA Poly(D,L-lactide) 50:50 PLGA 50:50 Poly (DL-lactideecoeglycolide) 75:25 Poly 75:25 PLGA (DL-lactideecoeglycolide) Poly(caprolactone) PCL

0.90e1.20 0.55e0.75 0.76e0.94

Chloroform 1,2 DCE 1,2 DCE

0.55e0.75

Chlorobenzene

1.00e1.30

1,2 DCE

1874

S.M. Berry et al. / Biomaterials 32 (2011) 1872e1879

Fig. 2. Schematic illustrating the construction of branched fibers.

week, selected arrays of each material were rinsed 3 or more times with DI water, then dried under vacuum and weighed to quantify hydrolysis rate. 2.3. Fibrous scaffold fabrication for cellular studies Arrays of polymer fibers were drawn on square and circular polystyrene frames using the direct-write system equipped with a 22 gauge needle (Internal diameter ¼ 0.4 mm) at a draw-rate of 20 mm/s. Scaffolds were produced from all five biodegradable materials using concentrations which were shown to produce fibers with microvascular-scale diameters (5e15 mm) at maximum yield. After drying overnight, the fiber arrays were sterilized by UV irradiation (l ¼ 253.7 nm) and placed into Petri dishes or well plates which had been previously coated with bovine serum albumin to prevent unwanted cellular adhesion as described by Klebe et al. [26]. Each fiber array was then coated with 400 mL of an aqueous solution containing either 5 mg/mL fibronectin (FN) (SigmaeAldrich), 0.1% gelatin attachment factor (AF) (Cascade Biologics), or both adhesion promoters and incubated at 37  C for 1 h to promote adsorption. “Branched” fibers were fabricated as indicated by Fig. 2. First, a suspended fiber was drawn across a polystyrene (PS) frame (Fig. 2a). Then a solution filament was drawn over top of the fiber and briefly lowered to attach the two structures (Fig. 2b). Lastly, the syringe needle was lifted and moved to the final attachment point (Fig. 2c) to complete the construction (Fig. 2d). This process is done quickly (V ¼ 20 mm/s) such that the top filament still possesses some solvent when it contacts the bottom fiber, promoting adhesion between the two. 2.4. Cell culture Human umbilical vein endothelial cells (HUVECs; Cascade Biologics, Portland, OR) were cultured in M-200 media supplemented with antibiotics (penicillin and streptomycin) and Low Serum Growth Supplement containing 2% fetal bovine serum, basic fibroblast growth factor, heparin, hydrocortisone, and epidermal growth factor. All cells in this study were used between passages 3 and 8. 2.5. Cellular adhesion quantification Cellular adhesion to the various scaffold polymers was evaluated using films of each material in order to mitigate both the inaccuracy associated with variations in fiber scaffold surface area and the difficulty of counting cells in a three-dimensional configuration. Films of each polymer were deposited in 24-well plates via solvent casting from 150 mL of a 2.5% solution of each polymer in chloroform (total film area ¼ 2 cm2). Films were heated to 65  C for 30 min and then left overnight at room temperature to remove residual solvent. Each film was then coated with 400 mL of an aqueous solution containing FN, AF, both, or PBS control as previously described for

fiber arrays in Section 2.3. After washing twice with PBS, HUVECs were seeded at a density of w40,000 cells/cm2 and incubated for 24 h at 37  C. Cells were then fixed in 3.7% formaldehyde for 10 min at room temperature, permeabilized in 0.1% Tween 20 for 5 min, and stained with a 300 nM 40 ,6-diamidino-2-phenylindole (DAPI) nuclear stain (Molecular Probes, D1306) for 3 min. Stained specimens were imaged in at least 3 locations per film on at least 3 films with a fluorescent microscope (Nikon Eclipse TE-2000-U, Melville, NY) equipped with a charge-coupled device (CCD) camera (QImaging). Stained nuclei were counted by utilizing ImageJ software [27] to filter out any light noise from the image and then counting all objects with a size of at least 150 pixels. 2.6. Evaluation of fibrous scaffold Fiber arrays were then coated with one of the cellular adhesion promoters (FN, AF, or both) as previously described with polymer films. Next, fiber arrays were submerged in media, subsequently suspensions of HUVECs were deposited in the vicinities of the arrays at densities of w40,000 cells/cm2 (based on Petri dish area). Eight arrays were incubated at 37  C for 24 h before fixing, staining, and imaging with fluorescent or confocal (Nikon D-Eclipse C1 laser scanning confocal microscope) microscopy to confirm cellular adhesion. Briefly, cells were fixed in 3.7% formaldehyde for 10 min at room temperature, permeabilized in 0.1% Tween 20 for 5 min, and stained at room temperature with 300 nM 40 ,6-diamidino-2-phenylindole (DAPI) nuclear stain (Molecular Probes, D1306) for 3 min and 165 nM rhodamine phalloidin F-actin stain (Molecular Probes, R415, Eugene, OR) for 20 min at room temperature. Other fiber arrays were continuously cultured for 1 week in order to monitor cellular proliferation on the scaffolds via daily imaging of the same fiber locations. 2.7. Statistical analyses In the fiber fabrication experiments, the diameter measurements and process yields were compared statistically using a Student’s t-test. For the cell adhesion and proliferation experiments, all experiments were conducted in at least triplicate and the results were statistically analyzed using a Student’s t-test. Results from all of the experiments were considered significantly different if the p-value was less than or equal to 0.05.

3. Results and discussion 3.1. Fiber fabrication Polymer fibers were successfully drawn from solutions containing each of the five biodegradable materials listed in Table 1. It

Table 2 Characterization of the direct-write process for the five biodegradable materials. Yield data represents 48 drawing attempts while fiber diameter data represents 16 fibers each measured in three locations (fiber midpoint and w200 mm from each end) and  diameter values represent standard deviation of fiber diameter. Polymer

L-PLA DL-PLA

50:50 PLGA 75:25 PLGA PCL

Solvent

Chloroform 1,2 DCE 1,2 DCE Chlorobenzene 1,2 DCE

Workable Concentration Range

28e29% 30e32% 22.5e23% 29e30% 28e30%

Fiber Diameter (mm)

Process Yield At Low End of Range

At High End of Range

At Low End of Range

18% 83% 63% 33% 71%

29% 71% 92% 83% 83%

11.57 12.70 4.42 6.40 11.79

    

7.86 2.28 3.13 5.05 3.67

At High End of Range 14.49 12.01 25.40 16.87 26.75

    

7.67 6.39 6.81 1.66 3.11

S.M. Berry et al. / Biomaterials 32 (2011) 1872e1879

1875

Fig. 3. SEM images of representative fibers fabricated with the direct-write system. A) End of PCL fiber, B) Array of 50:50 PLGA fibers, C) Suspended, branched 50:50 PLGA structure with diameter measurements, D) Overlapping 75:25 PLGA fibers.

was discovered that the ability to generate fibers from a given solution was influenced by solution concentration, with the workable concentration range falling between 22% and 32% by weight, depending on the specific polymer/solvent pair. Preliminary experimentation illustrated that solutions with concentrations below this range typically experienced capillary breakup prior to solidification. Alternatively, solutions above the aforementioned limit dried significantly prior to ejection from the direct-write tip. This accumulation of dried material at the ejection site essentially modified the tip geometry in an uncontrolled manner, resulting in inconsistent ejection or complete blockage of the solution. However, within the workable range, unbroken fibers with microvascular-scale diameters (5e20 mm) were repeatably fabricated from each material (Table 2, Fig. 3). The majority of the test solutions were processed into fibers with yields exceeding 60%. The exceptions included the lower limit solution of 75:25 PLGA (29%), which frequently failed due to capillary breakup, and all solutions containing L-PLA, which failed due to extensive drying of the solution during the direct-write process since solvent evaporation was accelerated due to the extreme volatility of the chloroformbased solutions. Fibers drawn from two of the polymer solutions (L-PLA and DL-PLA) exhibited no change in diameter between the low and high concentrations while the other three polymer solutions (50:50 PLGA, 75:25 PLGA, and PCL) displayed significant increases in diameter as concentration increased.

polymers degraded most rapidly, with mass losses of 75% and 90%, respectively, over the course of 6 weeks. This is consistent with the result by Farnia et al. which suggests that the decreased hydrophobicity of glycolide (relative to lactide) facilitates faster hydrolysis [28]. The scaffolds composed of L-PLA and PCL degraded more gradually, only losing 63% and 56% of the polymer mass, respectively, over the course of the trial. Initially, the DL-PLA scaffold exhibited a large (>50%) mass loss after the first week, but this rapid decline abated thereafter and a loss of only 57% was measured after week 6. These findings correspond well with results reported in the literature [29]. Specifically, L-lactides are highly crystalline, which impedes the ability for water to interact with hydrolysis sites in the polymer. On the other hand, DL-lactides degrade more rapidly in water due to their amorphous structure. Thus, co-polymers with

3.2. Scaffold hydrolysis rate Test arrays of each material were exposed to a PBS environment (Initial pH ¼ 7.4) for 1e6 weeks, dried under vacuum, and then weighed to determine the extent of degradation (Fig. 4). It was observed that scaffolds composed of the 75:25 and 50:50 PLGA

Fig. 4. Biodegradation profile of micro fiber (Dia. ¼ 5 to 20 mm) test arrays.

1876

S.M. Berry et al. / Biomaterials 32 (2011) 1872e1879

Fig. 5. Quantification of EC adhesion on films of different biodegradable polymers with pre-treatments of fibronectin (FN), gelatin-based attachment factor (AF), a combination of both (AF þ FN), and no treatment (PBS control). Error bars represent the standard deviation of the data and three measurements were recorded on each of three independent samples. Values statistically higher than the negative controls are denoted with asterisks.

glycolides will vary the degradation rate through the modulation of the polymers’ crystallinity and the availability of methyl groups in the structure. The pH of all materials excluding 50:50 PLGA remained within the range of 7.0e7.5 throughout the entire period of the experiment. The pH of the 50:50 PLGA samples also remained stable at 7.0 to 7.5 through week 4, but acidified to 6.0e6.5 for the remainder of the trial. Understanding the rates of mass loss of each fiber material is critical, because it determines the length of time ECs must be cultured on the fibrous scaffold before the material degrades. 3.3. Cellular adhesion Endothelial cells adhered to films of each of the polymers, albeit at lower densities than the positive-control environment of tissue culture polystyrene (TCPS). This result is likely due to the modification of the TCPS via plasma treatment, which grafts highly energetic oxygen ions onto the PS surface, thus making the surface

Fig. 6. Various cell scaffolds drawn with the direct-write technique. A) Parallel fibers of PLGA fibers, and D) Multiple circle frames with fibers in 48-well plate.

more negatively charged and more conducive to protein adsorption leading to enhanced cellular attachment [30]. The combination of gelatin attachment factor and fibronectin (AF þ FN) as well as FN alone were shown to significantly increase cellular adhesion for all materials. These two treatments caused statistically similar increases, except in the case of 50:50 PLGA, where AF þ FN were significantly higher than all other treatments. Treatment with AF alone yielded no effect for films of PCL and L-PLA, but promoted an increase in adhesion for the other three biodegradable materials relative to the negative control (Fig. 5). Unmodified, the polymers used in the study were poorly suited for EC adhesion. However, adsorption of biomolecules such as FN or AF provided focal adhesions for cell attachment, thus increasing adhesion [26]. 3.4. Cellular scaffold fabrication Biodegradable fiber arrays were drawn on PS frames using the direct-write process. Both parallel fiber and branched fiber (as

DL-PLA,

B) Branched 75:25 PLGA fibers, C) Square PS frame with multiple branched 75:25

S.M. Berry et al. / Biomaterials 32 (2011) 1872e1879

1877

Fig. 7. HUVECs were cultured on directly-written fibrous scaffolds and stained to highlight the cytoskeletons (red) and nuclei (green/blue). Scaffolds were composed of A,F,H) 75:25 PLGA; B,C,G) 50:50 PLGA; D) L-PLA; and, E) DL-PLA. Panel E illustrates growth on two fibers positioned near the scaffold substrate (PS Petri dish), resulting in cellular adhesion on the surface between adjacent fibers. Panel F is a confocal image with cross-sectional representation (inset), which confirms cellular presence around entire fiber circumference.

Fig. 8. HUVEC growth on same area of L-PLA fiber over 6 days.

1878

S.M. Berry et al. / Biomaterials 32 (2011) 1872e1879

described in Fig. 2) arrays were constructed in order to investigate the growth patterns of EC on these differing geometries. Fibers were treated with either FN or FN þ AF to promote cellular adhesion. The fibers on a selection of the frames were imaged (Fig. 6) and it was determined that the mean fiber diameter was 17.36 mm  3.93 mm. 3.5. Cellular growth on fibrous scaffold Endothelial cells were successfully seeded on microfibrous scaffolds composed of each of the five tested biodegradable polymers. Fig. 7 shows several such scaffolds which were fixed 24 h after seeding and stained to highlight the nucleus and cytoskeleton. The cells appear to orient in a confluent monolayer on the fiber surface and preferentially spread along the axial direction on the fiber. Confocal imaging of a selection of these scaffolds confirms that the endothelial cells are tightly enveloping the entire circumference of the fiber (Fig. 7F, inset). This outcome may not be achieved if the scaffolds are resting on an external surface (as shown by Fig. 7E), highlighting the importance of having the capability to produce truly suspended scaffolds with this directwrite technique. Examination of the same scaffold area on consecutive days demonstrates an increase in cellular density as existing cells propagated to fill non-cellularized regions of the scaffold (Fig. 8). The cellular density of the scaffold peaked between 2 and 5 days after seeding then steadily declined. This behavior is likely due to contact inhibition, which is typically observed in monolayer cultures of ECs when no additional chemotactic or mechanical stimuli are present [31e34]. The formation of a confluent cylinder of endothelial cells on a biodegradable scaffold may be the first step in the in vitro fabrication of a functional vasculature. Since the scaffold is responsible for guiding cellular growth, the scaffold architecture will be a strong determinant of the specific nature of this vasculature. A scaffold consisting of an array of precisely-positioned, branching fibers with microvascular-scale diameters will stimulate the formation of endothelial cell cylinders of this size, making them ideal precursors of a functional microvascular network. The next challenge involves coercing the cells to differentiate from growth into maturation. This can likely be achieved through the external application of physiological shear stress and/or the application of signaling molecules. This obstacle provides an excellent opportunity to study endothelial differentiation in vitro on fibrous scaffolds and quantify the factors that influence their migration and maturation. 4. Conclusions A direct-write technique has been utilized to fabricate preciselypositioned arrays of biodegradable fibers. These fibers have microscale diameters which are controllable through the modulation of the system geometry and the intrinsic properties of the solutions from which they are drawn. This ability to predictably construct suspended, branched, arrays of microvascular-scale fibers facilitated the production of endothelial cell scaffolds with unique geometries. Through hydrolysis, these scaffold lost between 50% and 90% of their masses over the course of six weeks, depending on material. A high level of cellular adhesion (similar to TCPS) was successfully demonstrated on the biodegradable materials with the aid of fibronectin and gelatin-based adhesion promoters. Cells cultured on fibrous scaffolds were observed to proliferate both along the axis and around the circumference of the fibers, producing confluent cylinder-shaped monolayers of cells, often within <2 days of seeding.

Acknowledgements This investigation was funded by National Science Foundation NIRT Program (ECS-0506941), NSF PFI Program (EEC-0438604), and National Aeronautics and Space Administration cooperative agreement (NCC5-571). Appendix Figures with essential color discrimination. Figs. 1, 2 and 4 in this article are difficult to interpret in black and white. The full color images can be found in the online version, at doi:10.1016/j. biomaterials.2010.11.023. References [1] Hellman KB. Engineered tissues: the regulatory path from concept to market; 2007. pp. 363e376. [2] Jain RK. Molecular regulation of vessel maturation. Nat Med 2003;9:685e93. [3] Nomi M, Atala A, Coppi PD, Soker S. Principals of neovascularization for tissue engineering. Mol Aspects Med 2002;23:463e83. [4] Soker S, Machado M, Atala. A systems for therapeutic angiogenesis in tissue engineering. World J Urol 2000;18:10e8. [5] Blawas AS, Reichert WM. Protein patterning. Biomaterials 1998;19:595e609. [6] Dike LE, Chen CS, Mrksich M, Tien J, Whitesides GM, Ingber DE. Geometric control of switching between growth, apoptosis, and differentiation during angiogenesis using micropatterned substrates. Vitro Cell Dev Biol 1999;35:441e8. [7] Kane RS, Takayama S, Ostuni E, Ingber DE, Whitesides GM. Patterning proteins and cells using soft lithography. Biomaterials 1999;20:2363e76. [8] He W, Yong T, Ma ZW, Inai R, Teo WE, Ramakrishna S. Biodegradable polymer nanofiber mesh to maintain functions of endothelial cells. Tissue Eng 2006;12 (9):2457e66. [9] Kameoka J, Orth R, Yang YN, Czaplewski D, Mathers R, Coates GW, et al. A scanning tip electrospinning source for deposition of oriented nanofibres. Nanotechnology 2003;14:1124e9. [10] Xu CY, Inai R, Kotaki M, Ramakrishna S. Aligned biodegradable nanofibrous structure: a potential scaffold for blood vessel engineering. Biomaterials 2004;25:877e86. [11] Yang F, Murugan R, Wang S, Ramakrishna S. Electrospinning of nano/micro scale poly(l-lactic acid) aligned fibers and their potential in neural tissue engineering. Biomaterials 2005;26:2603e10. [12] You Y, Min B-M, Lee SJ, Lee TS, Park WH. In vitro degradation behavior of electrospun polyglycolide, polylactide, and poly(lactide-co-glycolide). J Appl Polym Sci 2005;95:193e200. [13] Dalton PD, Joergensen NT, Groll J, Moeller M. Patterned melt electrospun substrates for tissue engineering. Biomed Mater 2008;3(3):11. [14] Kwon IK, Matsuda T. Co-electrospun nanofiber fabrics of poly(l-lactide-coepsilon-caprolactone) with type I collagen or heparin. Biomacromolecules 2005;6(4):2096e105. [15] Jeong SI, Kim SY, Cho SK, Chong MS, Kim KS, Kim K, et al. Tissue-engineered vascular grafts composed of marine collagen and PLGA fibers using pulsatile perfusion bioreactors. Biomaterials 2007;28(6):1115e22. [16] Ekaputra AK, Prestwich GD, Cool SM, Hutmacher DW. Combining electrospun scaffolds with electrosprayed hydrogels leads to three-dimensional cellularization of hybrid constructs. Biomacromolecules 2008;9(8):2097e103. [17] Santos MI, Tuzlakoglu K, Gomes ME, Fuchs S, Unger RE, Piskin E, et al. Nanoand micro-fiber combined scaffolds: an innovative design for improving endothelial cell migration in bone tissue engineering approaches. Tissue Eng 2006;12(4):986e7. [18] Vozzi G, Flaim C, Ahluwalia A, Bhatia S. Fabrication of PLGA scaffolds using soft lithography and microsyringe deposition. Biomaterials 2003;24:2533e40. [19] Kullenberg J, Rosatini F, Vozzi G, Bianchi F, Ahluwalia A, Domenici C. Optimization of PAM scaffolds for neural tissue engineering: preliminary study on an SH-SY5Y cell Line. Tissue Eng Part A 2008;14(6):1017e23. [20] Hadjizadeh A, Doillon CJ, Vermette P. Bioactive polymer fibers to direct endothelial cell growth in a three-dimensional environment. Biomacromolecules 2007;8:864e73. [21] Hwang CM, Khademhosseini A, Park Y, Sun K, Lee S-H. Microfluidic chip-based fabrication of PLGA microfiber scaffolds for tissue engineering. Langmuir 2008;24:6845e51. [22] Nain AS, Phillippi JA, Sitti M, MacKrell J, Campbell PG, Amon C. Control of cell behavior by aligned micro/nanofibrous biomaterial scaffolds fabricated by spinneret-based tunable engineered parameters (STEP) technique. Small 2008;4(8):1153e9. [23] Berry SM, Harfenist SA, Cohn RW, Keynton RS. Characterization of micromanipulator-controlled dry spinning of micro- and sub-microscale polymer fibers. J Micromech Microeng 2006;16:1825e32. [24] Nain AS, Wong JC, Amon C, Sitti M. Drawing suspended polymer micro-/ nanofibers using glass micropipettes. Appl Phys Lett 2006;89:183105.

S.M. Berry et al. / Biomaterials 32 (2011) 1872e1879 [25] Tripathi A, Whittingstall P, McKinley GH. Using filament stretching rheometry to predict strand formation and “processability” in adhesives and other nonnewtonian fluids. Rheol Acta 2000;39:321e37. [26] Klebe RJ, Bentley KL, Schoen RC. Adhesive substrates for fibronectin. J Cell Physiol 1981;109(3):481e8. [27] Abramoff MD, Magelhaes PJ, Ram SJ. Image processing with ImageJ. Biophotonics Int 2004;11(7):36e42. [28] Farnia SMF, Mohammadi-Rovshandeh J, Sarboluki MN. Synthesis and characterization of novel biodegradable triblock copolymers from l-lactide, glycolide, and PPG. J Appl Polym Sci 1999;73(5):633e7. [29] Bronzino JD. The biomedical engineering handbook. Boca Raton: CRC Press; 2000.

1879

[30] Ratner BD. Biomaterials science: an introduction to materials in medicine. San Diego: Elsevier Academic Press; 2004. [31] Ryan U. Endothelial cells volume 2. Boca Raton: CRC Press; 1988. [32] Bicknell R. Endothelial cell culture. Cambridge: Cambridge University Press; 1996. [33] Kirkpatrick CJ, Otto M, Van Kooten T, Krump V, Kriegsmann J, Bittinger F. Endothelial cell cultures as a tool in biomaterial research. J Mater Sci e Mater Med 1999;10:589e94. [34] Traub O, Berk BC. Laminar shear stress: mechanisms by which endothelial cells transduce an atheroprotective force. Arterioscler Throb Vasc Biol 1998; 18(5):677e85.