Hybrid nanobrous membranes of PLGA/chitosan fabricated via an electrospinning array

Bin Duan,1 Lili Wu,1 Xiaoyan Yuan,1 Zhen Hu,1 Xiulan Li,2 Yang Zhang,2 Kangde Yao,1 Min Wang3 1 School of Materials Science and Engineering, Tianjin University, Tianjin 300072, China 2 Institute of Orthopedics, Tianjin Hospital, Tianjin 300211, China 3 Department of Mechanical Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong
Received 27 August 2006; revised 2 March 2007; accepted 21 March 2007 Published online 13 June 2007 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jbm.a.31408 Abstract: Hybrid nanobrous membranes of poly(lacticco-glycolic acid) (PLGA) and chitosan with different chitosan amounts (32.3, 62.7, and 86.5%) were fabricated via a specially designed electrospinning setup consisting of two sets of separate syringe pumps and power supplies. After soaking in chloroform overnight to dissolve PLGA, the amount of chitosan in the hybrid membranes was determined. The structure, mechanical properties, water uptake, and cytocompatibilities of the nanobrous membranes were investigated by scanning electron microscopy, tensile testing, incubation in phosphate buffer solution, and human embryo skin broblasts culturing. Results showed that the chitosan amount in PLGA/chitosan membranes could be well controlled by adjusting the number of syringe for electrospinning of PLGA or chitosan, respectively. Because of the introduction of chitosan, which is a naturally hydrophilic polymer, the hybrid PLGA/chitosan membranes after chitosan crosslinking exhibited good mechanical and water absorption properties. The cytocompatibility of hybrid PLGA/chitosan membranes was better than that of the electrospun PLGA membrane. The electrospun hybrid nanobrous membranes of PLGA and chitosan appear to be promising for skin tissue engineering. The concept of using an electrospinning array to form multicomponent nanobrous membranes will lead to the creation of novel scaffolds for tissue engineering applications. 2007 Wiley Periodicals, Inc. J Biomed Mater Res 83A: 868878, 2007 Key words: electrospinning array; nanobrous membrane; PLGA; chitosan; cytocompatibility

INTRODUCTION As a new and multidisciplinary endeavor, tissue engineering has emerged in recent years to be a promising means to treat patients with tissue loss or organ failure. The aim of tissue engineering is the development of biological substitutes that restore, maintain, or improve tissue function.1,2 Growth of a tissue or organ requires suitable cells to be seeded on a scaffold for cell growth and differentiation. This may be initiated in vitro and followed by the implantation of the cell-scaffold construct in vivo. Here, the selection of appropriate, biocompatible, and biodegradable materials for the scaffold which should work in concert with a seeded cell is very important for the formation of the targeted tissue. From the materials science and engineering point of view, human tissues can be considered as cellular composites representing
Correspondence to: X. Yuan; e-mail: yuanxy@tju.edu.cn or xyuan28@yahoo.com Contract grant sponsor: Natural Science Foundation of China; contract grant numbers: 50273027, 50573055
' 2007 Wiley Periodicals, Inc.

multiphase systems, which consist of cells organized into functional units and the extracellular matrix (ECM).3 Biomedical composite materials provide the basis for the development of bioactive and bioresorbable scaffolds which can enhance tissue formation while providing customized physical, biochemical, and mechanical properties.4 No matter what the scaffold is made of, the cell-scaffold construct should adapt to the extracellular environment. Electrospinning, which was rst patented by Formhals in 1934,5 has been shown to be a simple yet versatile method for producing micro- or nanobrous membranes. However, it is only in recent years that electrospinning of nanobrous polymer scaffolds for tissue engineering applications has been attracting increasingly greater attention.6,7 Such interest in electrospun nanobers is mainly due to the fact that the size of most features existing in natural ECM, such as pore, ridge, and ber, are in the nanometer range.8 With high porosity and surface-to-volume ratio, electrospun nanobrous membranes could mimic the architecture of the ECM of native tissues, and thus have great potential in applications for tissue engineering.911

NANOFIBROUS PLGA/CHITOSAN

869

Electrospinning of multicomponent polymer solutions into nanobrous membranes has attracted much attention recently. Many types of polymers could not be electrospun in certain solvent systems while the addition of a second component could facilitate the electrospinning process.12,13 Also, electrospinning of polymer blends dissolved in the same solvent14 and electrospinning of the mixture of synthetic and natural polymers from different solution systems15 were studied in order to overcome the shortcomings of synthetic and natural polymers, which have resulted in new types of scaffolds with good biocompatibility and improved mechanical and physical/chemical properties. Ding et al.16 designed a multijet electrospinning device together with a rotatable grounded tubular collector so as to obtain a uniform thickness of blended nanobrous mats with good dispersions of the two components, poly(vinyl alcohol) and cellulose acetate, in the mats. This device can be used to fabricate polyblend nanobrous mats of multicomponent polymers which cannot be dissolved in the same solvent or kept in the same container. However, the electrospun parameters such as voltage, tip-to-collector distance, and ow rate cannot be varied and have to be the same for different polymer solutions forming the nanobrous mat. Min et al.17 built a setup for electrospinning of two polymer solutions simultaneously delivered by two syringe-pumps at different ow rates. With this setup, the tip-to-collector distances can be different but the voltage has to be maintained in the same for the two polymer solutions. Using this setup, a composite membrane of poly(lactic-co-glycolic acid) (PLGA) and chitin, within which chitin nanoparticles were uniformly distributed in the PLGA nanobrous structure, was prepared. However, the electrospinning setup could not be used for preparation of nanobrous membranes consisting of two or more polymers when different voltages were necessary. In our previous study, nanobrous PLGA-chitosan/PVA membranes were fabricated via a simultaneous mixing electrospinning process from PLGA and chitosan/PVA solutions.15 By this method, it was difcult to produce hybrid nanobrous PLGA/chitosan membranes with different amounts of each component because the parameters for electrospinning of these two polymer solutions could not be adjusted freely. To fabricate multicomponent nanobrous membranes of controlled compositions, in the present study, a novel electrospinning setup was specially designed, which consisted of two sets of syringe pumps and power supplies, and could be termed dual-source and dual-power electrospinning. It was demonstrated that using this novel setup, PLGA and chitosan nanobers could be electrospun separately and simultaneously, and deposited on a rotating drum to form a multicomponent nanobrous mem-

brane. Carefully selecting electrospinning parameters for the two polymers separately, the composition of the hybrid membrane could be controlled with relative ease. The fabricated nanobrous PLGA/chitosan membranes were subjected to physical, mechanical, and cytocompatibility evaluations for their potential applications in skin tissue engineering.

MATERIALS AND METHODS Materials

Chitosan (Mw 6 3 105, degree of deacetylation: 85%), provided by Qingdao Pharmaceutical Institute (Qingdao, China), was degraded to a lower molecular weight chitosan (Mw 2 3 105) using Co60-irradiation. PLGA (LA/GA 80/20) was supplied by Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, China. The molecular weight (Mw 2.52 3 105) and polydispersity (Mw /Mn 1.67) of PLGA were determined using gel permeation chromatography in tetrahydrofuran. Triuoroacetic acid (TFA) was obtained from KRS Fine Chemical, China. All solvents were used as-received without further purication. PLGA was dissolved in a mixed solvent of chloroform and N,N-dimethlformamide (DMF) (80/20, v/ v) at a concentration of 6% (w/v). The electrospinning of chitosan was conducted using a 10% (w/v) chitosan solution in TFA/dichloromethane (80/20 v/v).

Dual-source and dual-power electrospinning

A dual-source and dual-power electrospinning setup was designed in the present study. This setup comprised two high voltage power supplies (Tianjin Dong Wen High Voltage Power Supply Plant, China), two dual-syringe pumps (or one single-syringe pump and one dual-syringe pump) (Cole-Parmer Instrument Company, USA), and a grounded collecting drum, which is schematically shown in Figure 1. The grounded rotating drum (100 mm in diameter) covered by a piece of aluminum foil was placed in the middle of the two sets of syringe pumps. It rotated at 16 rpm and moved at a speed of 4 m/min along the track. Two power supplies were separately connected to the two sets of syringe needles on the two sets of pumps for controlling the electric eld separately. With the two sets of pumps, the ow rates of chitosan and PLGA solutions were also controlled separately. The number of syringes for chitosan and PLGA solutions, which were placed on the different pumps for electrospinning, was controlled to be 1:2, 2:2, and 2:1 to fabricate hybrid nanobrous PLGA/ chitosan membranes with three different compositions, which were designated as PLGA/chitosan-I, PLGA/chitosan-II, and PLGA/chitosan-III, respectively. Generally, a positive voltage of 15 kV, a ow rate of 0.2 mL/h, and a 12 cm distance between the needle tip and the collecting drum were used for the PLGA solution, and the chitosan solution was electrospun at 16 kV positive voltage, 12 cm working distance, and 0.2 mL/h ow rate. In this setup, if
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

870

DUAN ET AL.

Figure 1. Schematic diagram showing the setup for dual-source and dual-power electrospinning of hybrid nanobrous PLGA/chitosan membranes. [Color gure can be viewed in the online issue, which is available at www.interscience. wiley.com.]

the electrospinning of PLGA was not proceeding while the chitosan solution was electrospun, only mono-component nanobrous chitosan membrane was produced. Similarly, mono-component PLGA membranes could be produced. The electrospun PLGA/chitosan membranes as well as the nanobrous chitosan membranes were further crosslinked by glutaraldehyde vapor of a 25% glutaraldehyde aqueous solution at 378C for 8 h. After crosslinking, the electrospun nanobrous membranes were treated with a 0.1M glycine aqueous solution to block unreacted aldehyde groups and then soaked in phosphate buffered saline (PBS) for 24 h. They were nally dried overnight at room temperature.

Determination of compositions of nanobrous membranes

The amounts of chitosan and PLGA in hybrid nanobrous membranes fabricated were determined by immersing the dried membranes in chloroform to completely remove the PLGA component. Samples from all membranes were cut into a square shape with dimensions of 20 3 20 mm2 and the immersion time was 48 h. The wet nanobrous membranes were then dried at room temperature for 24 h. The amount of chitosan in the hybrid membranes was calculated by dividing the residual mass after immersion by the initial mass before immersion of the samples (n 3).

Philips XL-30). The average ber diameter of the electrospun bers in the membranes was measured from SEM micrographs in the original magnication of 10,000 using an Adobe Photoshop 7.0 software. To determine the water uptake of nanobrous membranes, the electrospun membranes of PLGA, PLGA/ chitosan-I, PLGA/chitosan-II, PLGA/chitosan-III, and chitosan were cut into a square shape with dimensions of 20 3 20 mm2. The samples were weighed using an electronic balance with a resolution of 0.01 mg before being placed in closed bottles containing 20 mL of PBS (pH 7.40) and incubated in vitro at (37.0 6 0.1)8C for 24 h. The wet mass of the samples after incubation was determined by weighing them immediately after removing from PBS and blotting them with lter paper to absorb water on the sample surface. The water uptake of electrospun membranes in PBS were then calculated using the following equation: Water uptake % m 1 m 0 3 100% m0

Physical and mechanical characterizations

For microstructural studies of the electrospun nanobrous membranes, samples were cut from the electrospun membranes, sputter-coated with a thin layer of gold and examined using a scanning electron microscope (SEM,
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

where, m1 and m0 are the mass of the membranes before and after immersion in PBS, respectively. Tensile properties of electrospun nanobrous membranes were determined using a universal testing machine (Testmetric M350-20KN, UK) equipped with a 100 N loadcell at a cross-head speed of 5 mm/min in the ambient environment. The gauge length of tensile samples was 40 mm. The samples were prepared in the rectangular shape with dimensions of 60 3 10 mm2 from dry electrospun nanobrous membranes. For obtaining mechanical properties of membranes in the wet condition, tensile samples were placed in closed bottles containing 10 mL of PBS (pH 7.40) and incubated at (37.0 6 0.1)8C for 24 h. The samples were then taken out of the bottles and tested under tension. The tensile modulus, tensile strength, and elonga-

NANOFIBROUS PLGA/CHITOSAN

871

tion at break presented in this article were averaged results of ve tests.

Cytocompatibility evaluation Cell culture and cell seeding

Human embryo skin broblasts (hESFs), supplied by the Cell Culture Center of the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences, China, were cultured in Dulbeccos Modied Eagle Medium (DMEM, Gibcol, USA) containing 10% fetal bovine serum, 50 U/mL penicillin, and 50 U/mL streptomycin. The medium was replaced every 3 days and cultures were maintained in a humidied incubator at 378C with 5% CO2. After reaching about 80% conuence, the cells were detached by 0.05% trypsin/0.05% EDTA (ethylenediaminetetraacetic acid) and counted with a hemocytometer. The electrospun PLGA, chitosan, and hybrid PLGA/chitosan membranes were cut into 10 3 10 mm2 pieces and placed in 24-well cell culture plates for sterilizing and prewetting by ethanol for 1 day and then washed by Hanks balanced salt solution. The resulting cells in a suspension were then seeded separately onto the electrospun nanobrous membranes as well as on the control tissue culture polystyrene (TCPS). The medium was changed every 3 days and the cell-scaffold constructs were maintained in the incubator for up to 5 days.

was transferred to a 96-well plate. Absorbance of the converted dye was measured at a wavelength of 570 nm using an ELISA plate reader. Samples with culture medium but without cells were set as control to obtain the background absorbance. Data of the proliferation measurement were collected from triplicate samples and expressed as the mean 6 standard deviation (SD). Statistical analysis was performed using an ANOVA with a Scheffe test. A value of p < 0.05 was considered to be statistically signicant.

Cell morphology
For cell morphology observations, the hESFs were seeded onto samples by placing the samples in a cell suspension of 1 3 105 cells/100 lL, and then the culture medium were lled into the culture wells to 1 mL after placing the culture wells in the incubator for 4 h. After 3 and 5 days of culture, the cell-scaffold constructs were harvested, rinsed twice with PBS, and subsequently xed with 2.5% glutaraldehyde at 48C for 4 h. They were subsequently dehydrated through a series of graded alcohol solutions and air-dried overnight. After drying, the samples were sputter-coated with gold and observed under SEM.

RESULTS AND DISCUSSION Fabrication of hybrid nanobrous PLGA/chitosan membranes The morphology of the nanobrous membranes of PLGA and chitosan fabricated by the electrospinning setup shown in Figure 1 is displayed in Figure 2. The ultrane PLGA bers [Fig. 2(a)] were electrospun from a 6% PLGA solution in a mixture solvent of chloroform and DMF (80/20, v/v), which had an average ber diameter of 232 6 76 nm [Fig. 2(a)], as measured from SEM micrographs. There were certain point-bonded structures present in the electrospun PLGA membrane. As for electrospinning of chitosan, TFA with dichloromethane was used as the solvent to

MTT assay
The hESFs were seeded onto samples by placing the samples in a cell suspension of 2 3 105 cells/100 lL, and then the culture medium were lled into the culture wells to 1 mL after placing the culture wells in the incubator for 4 h. After 1, 3, and 5 days of culture, the cell-scaffold constructs were rinsed with PBS to remove nonadhering cells, followed by incubation in 5 mg/mL MTT reagent in Hanks balanced salt solution (100 lL) for 4 h under the same conditions described. After removal of the medium, the converted dye was dissolved in 500 lL/well DMSO (dimethyl sulfoxide). Solution (100 lL) from each sample

Figure 2.

SEM micrographs of electrospun bers of (a) PLGA and (b) chitosan. (Original magnication 35000).
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

872

DUAN ET AL.

TABLE I Compositions of Hybrid Nanobrous PLGA/Chitosan Membranes After Crosslinking and Their Water Uptake
Sample Chitosan amount (% w/w) Average ber diameter (nm) Water uptake (%) PLGA 0 232 6 76 75.8 6 6.3 PLGA/Chitosan-I 32.3 6 11.8 286 6 72 289.4 6 6.1 PLGA/Chitosan-II 62.7 6 3.1 315 6 115 356.8 6 14.7 PLGA/Chitosan-III 86.5 6 0.7 326 6 122 437.3 6 18.3 Chitosan 100 347 6 104 499.8 6 17.1

facilitate the formation of the ultrane chitosan bers because the amino groups of chitosan were changed into salts with TFA and this salt formation destroyed the strong hydrogen bond interaction between chitosan molecules. The introduction of highly volatile dichloromethane in the chitosan-TFA solution could improve the homogeneity of the electrospun chitosan bers.18 A ber diameter distribution from 50 to 300 nm with an average ber diameter of 112 6 58 nm was observed for chitosan membranes [Fig. 2(b)], although some formation of beads as electrospinning defects was noticed in the membranes. Hybrid nanobrous chitosan/PLGA membranes were produced by simultaneous, dual-source and dual-power electrospinning of chitosan and PLGA solutions from two sets of separate electrospinning apparatus onto the same rotating drum (Fig. 1). Such an electrospinning setup was designed for two main reasons. First, the electrospinning parameters such as voltage distance between the needle tip and the collector, and ow rate of different polymer solutions could be controlled separately in order to be adjusted to their own appropriate electrospinning conditions to obtain ultrane bers. Second, by separately controlling different electrospinning sets, the composition of the multicomponent nanobrous membranes could be easily controlled, and hence membranes of specic compositions for certain tissue engineering applications could be produced through increasing or decreasing the number of syringes of the polymer solutions placed on each electrospinning set. Therefore, this setup for electrospinning of nanobrous membranes is a major advance from previous fabrication techniques including those employed by Ding et al.16 or Min et al.17 Furthermore, if required, more sets to form an electrospinning array can be added in order to produce multicomponent membranes with more than two components. Another possibility is the use of additional set for injecting or laying down biomolecules or drugs in the membrane being formed. The biomolecules incorporated in the membranes should enhance cell attachment and tissue growth in vitro and in vivo. Obviously, with more than two sets of electrospinning apparatus in the electrospinning array, careful consideration must be given to the conguration of the electrospinning setup. By controlling independently electrospinning parameters of individual polymer solutions, the concept of our design of the electrospinning setup as exemplied in
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

Figure 1 has opened up a large area for exploration in creating novel multicomponent nanobrous membranes for tissue engineering applications. With the current setup for dual-source and dualpower electrospinning (Fig. 1), chitosan and PLGA nanobers were alternatively and continuously deposited onto the collector as the drum rotated constantly and traversed at a constant speed. The amount of chitosan in the membranes was determined by completely removing the PLGA component from the hybrid membranes using chloroform. Table I lists the chitosan amount in the three types of hybrid membranes prepared by adjusting the number of the electrospun syringe. The amount of chitosan in the hybrid membranes was increased from (32.3 6 11.8)% in PLGA/chitosan-I (the syringe-number ratio of PLGA to chitosan was 2:1) to (86.5 6 0.7)% in PLGA/chitosan-III (the syringe-number ratio of PLGA to chitosan was 1:2) with increasing the number of syringes for the chitosan solution. The electrospun membranes containing chitosan were crosslinked by glutaraldehyde vapor from a 25% glutaraldehyde aqueous solution at 378C for 8 h to consolidate their structures and to prevent them from swelling and disintegrating in the culture medium in subsequent cytocompatibility studies. Figure 3 exhibits SEM micrographs showing the microstructures of crosslinked PLGA/chitosan membranes with different chitosan amounts. The corresponding histograms of diameters of electrospun bers are also included. It was observed that the crosslinked chitosan ultrane bers sustained the brous structure (Fig. 3) and that with crosslinking, chitosan nanobers had swollen from an average ber diameter of 112 6 58 nm [Fig. 2(b)] to 347 6 104 nm [Fig. 3(d)]. PLGA/chitosan-I [Fig. 3(a)] containing about 67.7% of PLGA had an average ber diameter of 286 6 72 nm and PLGA/chitosan-II [Fig. 3(b)] with 62.7% of chitosan showed an average ber diameter of 315 6 115 nm. PLGA/chitosan-III membranes [Fig. 3(c)] with an average ber diameter of 326 6 122 nm had a similar ber diameter distribution to that of chitosan bers in pure chitosan membranes that went through the corsslinking process. The average diameter of the hybrid nanobers was slightly increased with increasing chitosan amount in the hybrid membranes and meanwhile the ber diameter distributions of nanobrous membranes became wider, from 100 to 700 nm, because of crosslinking of chitosan bers.

NANOFIBROUS PLGA/CHITOSAN

873

Figure 3. SEM micrographs of crosslinked PLGA/chitosan membranes with different compositions and the crosslinked electrospun chitosan membranes. (a) PLGA/chitosan-I; (b) PLGA/chitosan-II; (c) PLGA/chitosan-III; (d) chitosan. (Original magnication 35000).

A few research groups measured the porosity of electrospun brous scaffolds.7,19,20 The porosity of electrospun brous scaffolds is normally in 7090% and generally around 80%. The porosity of the electrospun membranes was usually calculated from the apparent density of the electrospun membranes and the density of the raw materials or measured by mercury porosimeter. For the difcult evaluation of the density of the raw materials in hybrid PLGA/ chitosan, we did not measure the porosity of the membranes. According to the references above, it could be assumed that the porosity of the electrospun PLGA/chitosan membranes produced in the current investigation was about 80%. Water uptake The water uptake of the electrospun PLGA, chitosan, and hybrid PLGA/chitosan membranes was investigated by immersing membranes in PBS, and the results are summarized in Table I. The water uptake of PLGA membranes was about 75.8% after 24-h incubation in PBS, which was due to the unique nanober morphology which has extremely high specic surface area. With the introduction of chitosan nanobers into membranes, even the hybrid membranes containing only 32.3% chitosan (PLGA/chito-

san-I) could absorb as much as 289.4% of water, which was nearly four times of water absorption of the PLGA membranes. With larger amounts of chitosan, the water uptake of PLGA/chitosan-II (62.7% chitosan) and PLGA/chitosan-III (86.5% chitosan) membranes reached 356.8 and 437.3%, respectively. Among the samples tested, the electrospun chitosan membranes exhibited the highest PBS absorption (499.8%), which can be attributed to the hydrophilic nature of chitosan. In the present study, a large increase of water uptake was seen with the introduction of chitosan into the hybrid membranes, with further signicant, albeit gradual, increases in water uptake being detected when the amount of chitosan in the hybrid membranes was increased. Therefore, according to different application situations, the overall hydrophility of hybrid PLGA/chitosan membranes and hence their capability for water uptake could be controlled through control of the membrane composition. Mechanical properties For each of nanobrous membranes, ve samples were used for tensile testing. The average thickness of the ve dry membrane samples was 33 lm for PLGA, 64 lm for PLGA/chitosan I, 46 lm for PLGA/chitoJournal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

874

DUAN ET AL.

san II, 55 lm for PLGA/chitosan III, and 54 lm for chitosan, respectively. In wet conditions, after incubation in PBS at 378C for 24 h, the PLGA membranes and the crosslinked hybrid membranes containing chitosan shrank in about 2 and 3%, respectively. Mechanical properties of electorspun nanobrous membranes depend on a number of factors including the ber structure, properties of the constituent polymers, and their interactions. Figure 4 shows mechanical properties, including tensile strength, Youngs modulus, and elongation at break of electrospun PLGA, chitosan, and hybrid PLGA/chitosan membranes in both dry and wet states. For dry membranes, relatively low strength, modulus, and elongation were observed for the electrospun chitosan membranes as well as the electrospun PLGA/chitosan-III membranes because the later contained as high as 86% chitosan. As revealed in Figure 4, the PLGA component, which was increased from 13.5% in PLGA/chitosan-III to 37.3% in PLGA/chitosan-II, has more signicant effect on the tensile strength and Youngs modulus than on the elongation of membranes. When the PLGA content increased to about 67.7% in PLGA/ chitosan-I, there was a large increase in elongation from 2.8 to 46.0%, which was accompanied by reductions in tensile strength and Youngs modulus from 3.2 MPa and 110 MPa to 3.0 MPa and 106 MPa, respectively, for membranes from PLGA/chitosan-II to PLGA/chitosan-I. Such a large increase in elongation can be attributed to the increase of PLGA component as well as physical interactions between chitosan and PLGA bers. For hybrid PLGA/chitosan-I membranes having about 67.7% of PLGA, point-bonded structures in the electrospun PLGA membrane mentioned earlier and the interaction of chitosan bers at points of ber intersection due to crosslinking appeared to have a positive effect on hindering the slip of bers by providing frictional entanglements between bers. Similar observations were reported by Lee et al.,21 who studied mechanical behaviors of electrospun ber mats of poly(vinyl chloride)/polyurethane. In the current investigation, hybrid PLGA/ chitosan-II membranes were expected to have better mechanical properties as compared to other type membranes. Although the ultimate tensile strength and Youngs modulus were the highest in the dry state (Fig. 4), the elongation of PLGA/chitosan-II membranes decreased dramatically to the level of electrospun chitosan membranes. This was probably due to the discontinuity of chitosan bers dispersed in hybrid membranes of the higher PLGA content during the dual-source and dual-power electrospinning process of chitosan and PLGA solutions on the face to face direction. These hybrid membranes thus exhibited high tensile strength and high Youngs modulus before failure occurred in the region where
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

Figure 4. Dry and wet mechanical properties of electrospun PLGA, crosslinked chitosan, and hybrid PLGA/chitosan membranes with different compositions. (a) strength; (b) Youngs modulus; (c) elongation.

NANOFIBROUS PLGA/CHITOSAN

875

it was rich in chitosan bers and hence could not transfer the load to PLGA bers in the membranes. As for membranes in the wet condition, the tensile strength, Youngs modulus, and elongation of all electrospun membranes decreased to certain extent as compared to their dry state, except for the elongation of membranes having high chitosan amount. After the membranes had been placed in PBS, the absorption of PBS of membranes resulted in the stretching of polymer network bonds. In addition, the polymer chains became somewhat uncoiled and separated in the swollen gel structure and hence the sliding friction between polymer chains was very low.22 Moreover, the water molecules that entered nanobers may have played an important role as a plasticizer. Water uptake appears to have contributed to the reductions of tensile strength and Youngs modulus of nanobrous membranes in the wet condition and slight increases in elongation of PLGA/chitosan-II, PLGA/chitosan-III, and chitosan membranes. Cell proliferation The hESF proliferation on electrospun PLGA, chitosan, and hybrid PLGA/chitosan membranes, as well as on TCPS, was studied at days 1, 3, and 5. As shown in Figure 5, an increase in absorbance from day 1 to day 5 was recorded for hESFs, indicating a trend of cell proliferation on all membrane substrates. On the rst day, MTT assay showed a signicant increase (p < 0.05) in cell proliferation on chitosan and hybrid PLGA/chitosan membranes, when compared with the electrospun PLGA membrane. By contrast, there was no signicant difference on cell proliferation between the electrospun PLGA membranes and TCPS. This was probably because the hydrophilic membrane surface, due to the introduction of chitosan component in the membranes, accelerated the adhesion of hESFs and thus promoted cell proliferation at the initial stage. After 3 days of culture, the number of cells on the TCPS showed a dramatic increase (p < 0.05), and the cell viability of broblasts cultured on electrospun chitosan and hybrid membranes and TCPS was obviously higher than that on PLGA membranes. For the membranes with chitosan component, no signicant difference on cell proliferation was observed between the electrospun chitosan membranes and hybrid PLGA/chitosan-II and PLGA/chitosan-III membranes. However, the cell number on PLGA/chitosan-I membranes was signicant less than those on the other two hybrid membranes (p < 0.05) probably because of its lower chitosan content. From day 3 to day 5, the numbers of hESFs on PLGA and PLGA/chitosan-I membranes were close and there was no statistically signicant difference of cell proliferation between these two types of membranes.

Figure 5. Cell proliferation of hESFs by MTT assay. Mean for n 3 6 SD, p < 0.05 when compared with electrospun PLGA membranes.

Meanwhile, the increase in absorbance for hESFs on PLGA/chitosan-II membranes was much more obvious, increasing by about two times from day 1 to day 5. After 5-day culture, the number of cells on electrospun chitosan and PLGA/chitosan-II membranes was higher than that on hybrid PLGA/chitosan-I and PLGA/chitosan-III membranes. And, PLGA/chitosan-II membranes had even higher cell viability than chitosan membranes. From Refs. 23 and 24, it was known that, before cell attachment, a layer of proteins was rst absorbed onto the substance when it was contacted with the cell culture medium. And the surface energy of the substrate had a great inuence on the protein adsorption process. Low surface energy and hydrophobicity generally correlate with facilitating protein adsorption, but the adsorbed proteins easily change their original coiled conformation and then their bioactivities.25,26 Therefore, hydrophilic/hydrophobic balance of the substance surface is important for the protein absorption and the further cell attachment activity. The PLGA/chitosan-II membrane containing about 62.7% chitosan appeared to have achieved suitable hydrophilic/hydrophobic balance of surface and thus promoted the proliferation of broblasts.

Cell morphology Cell morphology and interaction between cells and nanobers of various electrospun membranes after 3 and 5 days of in vitro culture were examined using SEM as shown in Figures 6 and 7. After 3 days of culture, it was observed that hESFs interacted favorably with all nanobrous membranes, showing normal morphology and phenotype. Furthermore, the cells were found to migrate into porous nanobrous strucJournal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

876

DUAN ET AL.

Figure 6. Morphology of hESFs cultured for 3 days on the electrospun membranes. (a) PLGA; (b) PLGA/chitosan-I; (c) PLGA/chitosan-II; (d) PLGA/chitosan-III; (e) chitosan. (Original magnication 31000)

tures of membranes and integrate with the surrounding bers as well as cells to form a three-dimensional cellular network. It appears that in the electrospun hybrid PLGA/chitosan membranes, PLGA, and chitosan nanobers may be able to serve as substitutes of collagen and GAGs, and thus physically mimic the structure of nature ECM which is composed of randomly oriented collagen brils whose diameters are at the nanometer scale. Good cell growth on nanobrous membranes can be facilitated because of good hydrophilicity of the membranes, and the biomimetic nanobers can induce biological signals to emit. In addition, a proper match of the mechanical properties and the loosely interlaced brous structure could provide favorable
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

conditions for cells to enter the scaffolds.27 Therefore, the hybridization of nanobers of synthetic and natural polymers could successfully combine their advantages for processability, good mechanical properties, and biocompatibility. It has been demonstrated through the current investigation that the composition of hybrid nanobrous PLGA/chitosan membranes could be controlled and thus membrane properties including hydrophilicity, mechanical properties, and cytocompatibility could be varied by controlling the chitosan amount. The results obtained in the present study have shown that hybrid nanobrous PLGA/chitosan membranes fabricated can be a good candidate for skin wound dressings and for the scaffold in skin tissue engineering.

NANOFIBROUS PLGA/CHITOSAN

877

Figure 7. Morphology of hESFs cultured for 5 days on the electrospun membranes. (a) PLGA; (b) PLGA/chitosan-I; (c) PLGA/chitosan-II; (d) PLGA/chitosan-III; (e) chitosan. (Original magnication 31000)

CONCLUSIONS Through an electrospinning array, that is, dualsource and dual-power electrospinning using different power supplies and different syringe pumps with different number of syringes, hybrid nanobrous membranes of PLGA and chitosan with different compositions were successfully produced. The structure, mechanical properties, water uptake, and cytocompatibilities of the hybrid nanobrous PLGA/ chitosan membranes were systematically studied. Because of the introduction of large amounts of

hydrophilic chitosan (from 32.3 to 86.5%) into the membranes, the hybrid PLGA/chitosan membranes showed high water absorption as well as stable mechanical properties. The results from cytocompatibility study suggested favorable interactions between hESFs and PLGA/chitosan hybrid membranes. The hybrid nanobrous PLGA/chitosan membrane with suitable chitosan amount is a potential scaffold for skin tissue engineering.
We thank Professor Xuesi Chen in Changchun Institute of Applied Chemistry for his kind help.
Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a

878

DUAN ET AL.

References
1. Langer R, Vacanti JP. Tissue engineering. Science 1993;260: 920926. 2. Vacanti JP, Vacanti CA.The history and scope of tissue engineering. In: Lanza RP, Langer R, Vacanti J, editors. Principles of Tissue Engineering. San Diego: Academic Press; 2000. p 37. 3. Wintermantel E, Mayer J, Blum J, Eckert KL, Liischer P, Mathey M. Tissue engineering scaffolds using superstructures. Biomaterials 1996;17:8391. 4. Wang M. Developing bioactive composite materials for tissue replacement. Biomaterials 2003;24:21332151. 5. Formhals A. Process and apparatus for preparing articial threads. US Patent No. 1,975,504; 1934. 6. Li WJ, Laurencin CT, Caterson EJ, Tuan RS, Ko FK. Electrospun nanobrous structure: A novel scaffold for tissue engineering. J Biomed Mater Res 2002;60:613621. 7. Zong XH, Bien H, Chung CY, Yin LH, Fang DF, Hsiao BS, Chu B, Entcheva E. Electrospun ne-textured scaffolds for heart tissue constructs. Biomaterials 2005;26:53305338. 8. Abrams GA, Goodman SL, Nealey PF, Franco M, Murphy CJ. Nanoscale topography of the basement membrane underlying the corneal epithelium of the Rhesus Masaque. Cell Tissue Res 2000;299:3946. 9. Mo XM, Xu CY, Kotaki M, Ramakrishna S. Electrospun P(LLA-CL) nanober: A biomimetic extracellular matrix for smooth muscle cell and endothelial cell proliferation. Biomaterials 2004;25:18831890. 10. Boland ED, Matthews JA, Pawlowski KJ, Simpson DG, Wnek GE, Bowlin GL. Electrospinning collagen and elastin: Preliminary vascular tissue engineering. Front Biosci 2004;9:14221432. 11. Li WJ, Tuli R, Huang XX, Laquerriere P, Tuan RS. Multilineage differentiation of human mesenchymal stem cells in a three-dimensional nanobrous scaffold. Biomaterials 2005;26: 51585166. 12. Huang L, Nagapudi K, Apkarian RP, Chaikof EL. Engineered collagen-PEO nanobers and fabrics. J Biomater Sci Polym Ed 2001;12:979993. 13. Duan B, Dong CH, Yuan XY, Yao KD. Electrospinning of chitosan solutions in acetic acid with poly(ethylene oxide). J Biomater Sci Polym Ed 2004;15:797811. 14. Zhang YZ, Ouyang HW, Lim CT, Ramakrishna S, Huang ZM. Electrospinning of gelatin bers and gelatin/PCL composite brous scaffolds. J Biomed Mater Res Part B: Appl Biomater 2005;72:156165.

15. Duan B, Yuan XY, Zhu Y, Zhang YY, Li XL, Zhang Y, Yao KD. A nanobrous composite membrane of PLGA-chitosan/ PVA prepared by electrospinning. Eur Polym J 2006;42:2013 2022. 16. Ding B, Kimura E, Sato T, Fujita S, Shiratori S. Fabrication of blend biodegradable nanobrous nonwoven mats via multijet electrospinning. Polymer 2004;45:18951902. 17. Min BM, You Y, Kim JM, Lee SJ, Park WH. Formation of nanostructured poly(lactic-co-glycolic acid)/chitin matrix and its cellular response to normal human keratinocytes and broblasts. Carbohydr Polym 2004;57:285292. 18. Ohkawa K, Cha DI, Kim H, Nishida A, Yamamoto H. Electrospinning of chitosan. Macromol Rapid Commun 2004;25: 16001605. 19. Ma ZW, Kotaki M, Yong T, He W, Ramakrishna S. Surface engineering of electrospun polyethylene terephthalate (PET) nanobers towards development of a new material for blood vessel engineering. Biomaterials 2005;26:25272536. 20. Rho KS, Jeong L, Lee G, Seo BM, Park YJ, Hong SD, Roh S, Cho JJ, Park WH, Min BM. Electrospinning of collagen nanobers: Effects on the behavior of normal human keratinocytes and early-stage wound healing. Biomaterials 2006;27:1452 1461. 21. Lee KH, Kim HY, Ryu YJ, Kim KW, Choi SW. Mechanical behavior of electrospun ber mats of poly(vinyl chloride)/ polyurethane polyblends. J Polym Sci Part B: Polym Phys 2003;41:12561262. 22. Johnson BD, Beebe DJ, Crone W. Effects of swelling on the mechanical properties of a pH-sensitive hydrogel for use in microuidic devices. Mater Sci Eng C 2004;24:575581. 23. Anselme K. Osteoblast adhesion on biomaterials. Biomaterials 2000;21:667681. 24. Boyan BD, Hummert TW, Dean DD, Schwartz Z. Role of material surfaces in regulating bone and cartilage cell response. Biomaterials 1996;17:137146. 25. Kim JH, Kim SH, Kim HK, Akaike T, Kim SC. Adhesion and growth of endothelial cell on amphiphilic PU/PS IPN surface: Effect of amphiphilic balance and immobilized collagen. J Biomed Mater Res 2002;62:613621. 26. Moulin AM, OShea SJ, Badley RA, Doyle P, Welland ME. Measuring surface-induced conformational changes in proteins. Langmuir 1999;15:87768779. 27. Zhang YZ, Venugopal J, Huang ZM, Lim CT, Ramakrishna S. Characterization of the surface biocompatibility of the electrospun PCL-collagen nanobers using broblasts. Biomacromolecules 2005;6:25832589.

Journal of Biomedical Materials Research Part A DOI 10.1002/jbm.a