July 2009
Published by the American Chemical Society
Volume 10, Number 7
Copyright 2009 by the American Chemical Society
Articles Evaluation of Cross-Linking Methods for Electrospun Gelatin on Cell Growth and Viability Kristin Sisson,† Chu Zhang,‡,§ Mary C. Farach-Carson,†,‡,§ D. Bruce Chase,† and John F. Rabolt*,† Departments of Materials Science and Engineering and Biological Sciences, and Center for Translational Cancer Research, University of Delaware, Newark, Delaware 19716 Received January 8, 2009; Revised Manuscript Received May 4, 2009
The creation of a tissue engineering scaffold via electrospinning that has minimal toxicity and uses a solvent system composed of solvents with low toxicity and different cross-linking agents was investigated. First, a solvent system of acetic acid/ethyl acetate/water (50:30:20) with gelatin as a solute was evaluated. The optimum system for electrospinning a scaffold with the desired properties resulted from a gelatin concentration of 10 wt %. Several different methods were used to cross-link the electrospun gelatin fibers, including vapor-phase glutaraldehyde, aqueous phase genipin, and glyceraldehyde, as well as reactive oxygen species from a plasma cleaner. Because glutaraldehyde at high concentrations has been shown to be toxic, we explored other cross-linking methods. Using reactive oxygen species from a plasma cleaner is an easy alternative; however, the degradation reaction dominated the cross-linking reaction and the scaffolds degraded after only a few hours in aqueous medium at 37 °C. Glyceraldehyde and genipin were established as good options for cross-linking agents because of the low toxicity of these cross-linkers and the resistance to dissolution of the cross-linked fibers in cell culture medium at 37 °C. MG63 osteoblastic cells were grown on each of the cross-linked scaffolds. A proliferation assay showed that the cells proliferated as well or better on the cross-linked scaffolds than on traditional two-dimensional polystyrene culture plates.
Introduction Electrospinning is a technique that first was described in patents in the early 1900s and can be used to create micro- and nanosized fibers. These nonwoven membranes have a variety of potential applications including filtration,1–3 sensors,3,4 photovoltaic cells,1–5 biomedical materials,1,2 tissue engineering constructs,1,3 and chemical and biologically resistant clothing.2,6 In recent years there is increasing interest in creating biomaterials for purposes such as drug delivery,7 wound dressings,1,2 and tissue scaffolds.1,3 Frequently, collagen type I has been used to create tissue engineered electrospun scaffolds,8–10 however, gelatin, which is simply denatured collagen, is less costly. Gelatin has many other advantages over synthetic polymers such * To whom correspondence should be addressed. Fax: (302)831-4545. E-mail:
[email protected]. † Department of Materials Science and Engineering. ‡ Department of Biological Sciences. § Center for Translational Cancer Research.
as biological origin, biodegradability, biocompatibility, commercial availability, and nonimmunogenicity.11 This study evaluated the feasibility of creating an electrospun tissue engineered scaffold composed of gelatin with minimum toxicity so that the scaffolds can be readily used in future tissue engineering applications. Typical solvent systems used to electrospin gelatin include 1,1,1,3,3,3-hexaflouro-2-propanol (HFIP)12 and 2,2,2-triflourothanol (TFE),11,13 both of which are highly toxic.14 To avoid this, Song et al. developed a waterbased cosolvent approach to electrospin gelatin using acetic acid and ethyl acetate,14 both of which are relatively nontoxic. Cross-linking is another major consideration when using collagen and gelatin to create tissue engineered scaffolds. Without cross-linking, collagen and gelatin scaffolds dissolve in aqueous media, making them unsuitable for long-term applications. It is generally accepted that glutaraldehyde is a good choice for cross-linking electrospun fibers because of the simplicity of using this vapor-phase method.11 The cross-linking
10.1021/bm900036s CCC: $40.75 2009 American Chemical Society Published on Web 05/20/2009
1676
Biomacromolecules, Vol. 10, No. 7, 2009
reaction occurs between the carboxyl groups on the glutaraldehyde and the amine groups of the gelatin.11 However, glutaraldehyde in certain concentrations has been shown to be cytotoxic15,16 and can disrupt the electrospun fiber morphology.11 Hence, other methods for cross-linking electrospun scaffolds are needed. New methods were selected for this study based on the premise of reduced toxicity relative to glutaraldehyde. The first method involved the use of a plasma cleaner that creates reactive oxygen species to cross-link the gelatin fibers. Other methods used liquid phase cross-linking methods employing the molecules D,L-glyceraldehyde17–19 and genipin,20–22 both of which have been investigated previously using gelatin films, gels, and microspheres.17–22 Cross-linking electrospun gelatin using the following techniques has not previously been demonstrated on electrospun fibers: reactive oxygen species via a plasma cleaner and aqueous phase cross-linking methods using D,L-glyceraldehyde and genipin. Further, these methods of crosslinking require relatively simple procedures when compared to methods such as carbodiimide activation.
Materials and Methods Fabrication of Gelatin Scaffolds. Gelatin (courtesy of Eastman Kodak Corporation, Rochester, NY) was dissolved in a combination of 50:30:20 acetic acid (ACS reagent, g99.7%, Sigma Aldrich, Milwaukee, WI)/ethyl acetate (Fisher Scientific, Pittsburgh, PA)/distilled water, at concentrations of 6-12% (w/w). The gelatin solutions were stirred overnight. Electrospun scaffolds were created and collected using an electrospinning setup that consists of a syringe pump (KD Scientific, Holliston, MA), a high-voltage power supply (CZE1000R, Spellman, Hauppauge, NY), and a rotating mandrel collector set at a low rotation speed to allow the formation of a scaffold of uniform thickness. The syringe pump generates a constant flow from the needle and was set to a 0.06 mL/min flow rate. The high voltage power supply was set at +12 kV. The rotating mandrel is approximately 3 in. in width and 3 in. in diameter and was placed approximately 8 cm from the tip of the needle (0.51 mm inner diameter, 21 gauge). Cross-Linking Methods. Several cross-linking methods were examined in this study: glutaraldehyde vapor, reactive oxygen species generated using a plasma cleaner, D,L-glyceraldehyde (g90% (GC), Sigma Aldrich, Saint Louis, MO), and genipin (Wako Chemicals, Richmond, VA) in 70% (v/v) ethanol/water solution. Glutaraldehyde (Electron Microscopy Sciences, Hatfield, PA) was used in the vapor phase to cross-link electrospun scaffolds at a concentration of 0.5% (w/w) for 19 h. Reactive oxygen species generated using a plasma cleaner/sterilizer (PDC-32G, Harrick Plasma, Ithaca, NY) were used to cross-link the gelatin scaffolds by placing the fibers in a metal perforated box and the plasma cleaner energized until a purple glow was achieved by adjusting the air inlet valve for 2 min. Genipin and D,L-glyceraldehyde both were dissolved in 70% (v/v) ethanol/water. The electrospun scaffolds were cut into 8 mm discs and placed into 24-well plates and cross-linked by D,L-glyceraldehyde and genipin for 19 h. The amounts of D,L-glyceraldehyde and genipin were varied from 0.1 to 0.5% (w/w) and 0.1 to 2.0% (w/w), respectively, to determine how much cross-linker was necessary for use in tissue engineering applications. For the dissolvability study, 1 mL of the cross-linking solution was used on 12 mm diameter discs of the electrospun gelatin scaffold. In the cellular viability and proliferation assay, the scaffolds were placed on a dot blotter, which has 2 mm diameter wells. The effective size of the exposed scaffolds was 2 mm diameter and 200 µL of cross-linking solution was added to each of the scaffolds. Evaluation of Electrospun Scaffolds. A scanning electron microscope (FE-SEM, JSM 7400, JEOL, Tokyo, Japan) operating at 2 kV and 10 µA was used to analyze the structure and size of the electrospun fibers and to compare the cross-linked and precross-linked fibers. All samples were mounted using carbon tape on aluminum SEM stubs.
Sisson et al. An Attenuated Total Reflectance Fourier Transform Infrared Spectrometer (Nexus 670 FT-IR, Thermo Nicolet, Waltham, MA; ATRFTIR) was used to evaluate any chemical changes in the gelatin by examining gelatin both pre- and postelectrospinning. Due to the low concentration of cross-linking agents, no discernible difference in the ATR-FTIR spectra were observed for the fibers pre- and postcrosslinking. Dissolvability Study. The glutaraldehyde and reactive oxygen crosslinked discs were placed directly into the 24-well plates postcrosslinking. Scaffolds cross-linked with glyceraldehyde and genipin were cross-linked directly in the 24-well plates. These 24-well plates then were sterilized for 45 min under UV light. After sterilization, 1 mL of DMEM (Dulbecco’s modified Eagle medium, 11965, GIBCO, Carlsbad, CA) was added to each well, and the plates were held at 37 °C for a period of time to test the aqueous solubility of the cross-linked scaffolds. Cell Seeding and Attachment. MG63 cells (CRL-1427, American type Culture Collection (ATCC), Manassas, VA) are a human osteosarcoma cell line that has osteoblastic characteristics and can serve as a model for testing methods for bone tissue engineering. These cells were cultured in medium that consists of DMEM, 10% (v/v) fetal bovine serum (FBS), and 1% (v/v) penicillin/streptomycin (p/s; GIBCO; Grand Island, NY). MG63 cells were cultured in a 75 cm2 flask with approximately 1 week between passages. To passage the cells, all medium was removed from the flask. MG63 cells were washed using 5 mL of DPBS (Dulbecco’s phosphate buffered solution, 14190, GIBCO), and the PBS was removed. Next, 2 mL of 0.25% (w/v) trypsin-EDTA (25200, GIBCO) was added to the flask and allowed to sit for 5 min to detach cells from the flask. A total of 5 mL of medium was added and used to wash the cells from the flask bottom, and the cells and medium were removed and placed into a conical tube and centrifuged at 2000 rpm for 2.5 min. The medium was removed and 10 mL of medium was added to distribute the cells. Around 0.5 mL of the cell suspension was added into a fresh 75 cm2 flask with 20 mL of medium. Medium was changed every 2-3 days. To prevent contamination, each scaffold was rinsed in 70% (v/v) ethanol for 5 min. The scaffolds were rinsed with PBS three times and UV irradiated for 30 min before cell seeding. The cells seeded first were counted using a hemocytometer. The cells were added to the medium to be seeded onto the scaffolds at 2 × 105 cells/well density. The cell suspension was added to the scaffolds slowly in the middle of the scaffolds to allow cells to attach to the scaffold instead of sliding off to the side of the well. The scaffolds then were placed in the incubator at 37 °C and 5% CO2 and cells were allowed to grow. Evaluation of Cellular Viability and Proliferation. A WST-1 assay (Roche, Indianapolis, IN) was used as an index of cellular viability and proliferation. Cells were cultured on a 96-well plate for the twodimensional control. A dot blotter apparatus (BioRad, Hercules, CA) was used to test the scaffold samples. First, a piece of parafilm was placed over the holes in the bottom section of the dot blotter. Strips of the electrospun scaffolds were placed over the parafilm in the desired locations on the dot blotter. Finally, the blotter was sealed to create a mechanism to hold the electrospun scaffolds in place for cell culture. To prevent contamination, each scaffold was rinsed in 70% ethanol for 5 min. The scaffolds were rinsed with PBS three times and UV irradiated for 30 min before cell seeding. The cells were seeded at a density of 5 × 103 cells/well and allowed to grow for a total of 9 days. At 2 day intervals during the culture period, the WST assay was performed. The medium was removed from each well and 100 µL of medium was then added to each of these wells along with 10 µL of WST-1 reagent. A background control of 100 µL of medium plus 10 µL of WST-1 reagent was placed in a sterile 96-well plate. The plates were incubated at 37 °C for 1 h to allow the WST reaction to occur. The medium with the WST-1 reagent was removed from each well and placed in the 96-well plate along with the control medium. The plate was placed into a spectrophotometric plate reader (ELISA reader) set to read an absorbance at 450 nm with a reference filter at 600 nm to
Electrospun Gelatin on Cell Growth and Viability
Biomacromolecules, Vol. 10, No. 7, 2009
1677
Figure 1. FE-SEM images of electrospun gelatin from 50:30:20 acetic acid/ethyl acetate/water solutions: (A) 6, (B) 7, (C) 8, (D) 9, (E) 10, (F) 11% (w/w). Beads were present in A and B. The white scale bar represents 1 µm. Table 1. ATR-FTIR of Gelatin Peaks and Band Assignments31–33 peak position (cm-1)
band assignment
3300 1636-1640 1542-1544 1240
amide A (N-H stretching vibration) amide I (CdO stretch) amide II (N-H bend and C-H stretch) amide III (C-N stretch plus N-H in-phase bending)
Figure 2. Fiber diameter vs gelatin weight percent varied from 6 to 11% (w/w). Diameters were measured using SEM images and 25 randomly selected places were averaged.
determine the amount of Formazan formed, which directly correlates to the number of metabolically active cells in the culture. The data were analyzed using student’s two-sample t test (Microsoft Excel). All p-values are two-tailed. Figure 3. ATR-FTIR data of raw gelatin and electrospun gelatin.
Results Gelatin Concentration Effect on Fiber Size. Gelatin was electrospun from a solution of 50:30:20 acetic acid: ethyl acetate: water at 6-11% (w/w), and the fibers were collected on a flat collector plate covered in aluminum foil. To determine the fiber diameters and morphologies, the scaffolds were examined using FE-SEM. The fibers for 8-10% (w/w) were uniform and contained no beads. The fibers for 6 and 7% (w/w) contained beads. The fiber diameters also were measured in 25 randomly selected places. These results are shown in Figures 1 and 2. The highest fiber diameter (∼300 nm) was obtained from 10% (w/w) gelatin solutions. Although the 11% (w/w) concentration did have high fiber diameters, the lack of uniformity left a wider
distribution of fiber sizes leading to a lower average fiber diameter than the 10% (w/w). Because of the high fiber diameter and uniformity of the fibers produced, 10% (w/w) gelatin concentration was used in the remainder of the study. Characterization of Electrospun Scaffolds by FTIR Studies of Chemical Changes. To assess whether electrospinning gelatin affected the primary structure, ATR-FTIR measurements were made on raw gelatin before electrospinning and a membrane of the electrospun gelatin (Table 1). The results can be seen in Figure 3. Both the pre- and postelectrospun gelatin had very similar IR spectra with only slight differences in relative intensities, which could result from differences in the broad background underneath each spectrum. Thus, no change
1678
Biomacromolecules, Vol. 10, No. 7, 2009
Sisson et al.
Figure 4. FE-SEM images of electrospun gelatin at 10% (w/w) from 50:30:20 acetic acid/ethyl acetate/water solution and cross-linked using different methods: (A) uncross-linked, (B) glutaraldehyde cross-linked (0.5% (w/w)), (C) reactive oxygen species cross-linked, (D) D,L-glyceraldehyde cross-linked (0.5% (w/w)), and (E) genipin cross-linked (1.0% (w/w)). The white scale bar represents 1 µm.
in the chemical structure of the gelatin occurred during the electrospinning process. Membrane Morphology of the Cross-Linked Electrospun Scaffolds. Gelatin was electrospun from a solution of 50: 30:20 acetic acid/ethyl acetate/water at 10% (w/w), and the fibers were collected on a rotating mandrel at a slow speed so as not to draw the fibers, but rather to create a uniform electrospun scaffold. These fibers were cross-linked using the various methods described in Materials and Methods. The uncross-linked fibers are shown in Figure 4A. The fibers cross-linked with glutaraldehyde vapor are shown in Figure 4B. The morphology of the glutaraldehyde vapor cross-linked fibers was different than that of the uncross-linked fibers. The fibers appeared to be more rubbery and many fibers at fiber junctions were fused. The water vapor present in the air may have caused this change in morphology during cross-linking. In Figure 4C, the reactive oxygen species cross-linked fibers are shown, and there was no change in the morphology of the fibers when cross-linked with reactive oxygen species. The fibers cross-linked with glyceraldehyde and genipin are depicted in Figure 4D and E, respectively. The glyceraldehyde cross-linked fibers better maintained their morphology than did genipin cross-linked fibers. This could have been due to the slower reaction rate of the genipin as compared to the glyceraldehyde, allowing for the gelatin to dissolve before the genipin could completely crosslink the fibers. Note that although the fibers cross-linked with glyceraldehyde and genipin did not maintain their morphology, many pores still remained. Dissolvability. A dissolvability study of each of the crosslinked scaffolds was completed to test the solubility of the scaffolds in DMEM at 37 °C for at least 2 weeks. The reactive oxygen species cross-linked scaffold only lasted 12 h before it dissolved. Many efforts were made to optimize the cross-linking time to obtain a sample that would not dissolve after 12 h; however, those attempts were futile. The 0.5% (w/w) glutaraldehyde cross-linked fibers remained for over two weeks. Scaffolds cross-linked with glyceraldehyde at concentrations of
0.1, 0.2, 0.3, 0.4, and 0.5% (w/w) all remained for the entire two-week test; however, all but the 0.5% (w/w) sample were visibly weaker mechanically and slightly translucent. On the contrary, the 0.5% (w/w) was light yellow in color and opaque and appeared to be mechanically stronger. The scaffolds cross-linked with genipin were cross-linked at 0.1, 0.2, 0.5, 1.0, and 2.0% (w/w) and, at all concentrations, except 0.1% (w/w), remained over two weeks. The 0.1% (w/w) dissolved before the test was completed, likely contributable to incomplete crosslinking. The genipin scaffolds cross-linked at 0.2 and 0.5% (w/ w) appeared to be visibly weaker than the scaffolds cross-linked with a higher concentration. The scaffolds cross-linked with 1.0 and 2.0% (w/w) genipin were blue in color and opaque and appeared mechanically stronger than the scaffolds cross-linked with lesser concentrations of genipin. Although the genipin scaffolds cross-linked with 1.0 and 2.0% (w/w) genipin performed identically in the dissolvability study, the lesser concentration of 1.0% (w/w) genipin was preferred because the additional cross-linking agent was unnecessary and excessive. A concentration of 1.0% (w/w) genipin and 0.5% (w/w) glyceraldehyde was selected for upcoming work to be sure there would be enough cross-linking agents to form stable scaffolds in the future. From this study, it was determined that the use of reactive oxygen species is not a viable method for cross-linking durable gelatin scaffolds. Cellular Viability and Proliferation. A cell proliferation and viability assay was used to compare the cytotoxicity of the crosslinked scaffolds in cell culture conditions. The WST assay measures the metabolic activity of cultured cells, which correlates to the degree of proliferation. The cells were seeded onto each of the different scaffolds and remained in culture for 9 days. In Figure 5, it can be seen that cells grown on each of the different cross-linked scaffolds had equal or better cell proliferation and viability than did cells grown on the two-dimensional control (two-dimensional tissue culture polystyrene (PS)) on days 3 and 9. This indicates that the cells on the cross-linked scaffolds could migrate into the scaffolds thus allowing for more
Electrospun Gelatin on Cell Growth and Viability
Figure 5. Cell proliferation and viability study using a WST assay. The symbol * denotes a significant difference from two-dimensional PS, # denotes a significant difference from glutaraldehyde, and @ denotes a significant difference from genipin (p < 0.05).
proliferation because the cells have more area to grow in a 3-dimensional environment. This three-dimensional environment resembles that of the natural extracellular matrix. On day 7, the glyceraldehyde cross-linked scaffold shows higher proliferation and viability than the glutaraldehyde and genipin crosslinked scaffolds and the two-dimensional tissue culture polystyrene.
Discussion Because gelatin itself is nontoxic, it is important that the crosslinking agents also are nontoxic to the cells that will attach to the scaffold. However, many chemical cross-linking agents are cytotoxic as a result of the residual unreacted cross-linking agent or the degradation products that result from hydrolytic or enzymatic degradation.23,24 Glutaraldehyde is the most commonly used agent for crosslinking collagen and gelatin. It is inexpensive, soluble in aqueous solution and has a rapid reaction rate due to the many available amine groups in proteins.23 Glutaraldehyde reacts to form crosslinks in two different ways. First, the aldehyde group can react with the amine group of lysine, hydroxlysine, or arginine in the collagen polypeptide and form a Schiff base.23 Second, two adjacent aldehydes can undergo aldol condensation.23 Glutaraldehyde has, however, been shown to be cytotoxic at high concentrations.15,16 As described earlier, at low concentrations glutaraldehyde vapor disrupts the electrospun fiber morphology with fibers at fiber junctions fused together presumably due to the presence of water vapor during vapor-phase cross-linking.11 The residual unreacted glutaraldehyde, cross-linked collagen and gelatin, and the products from the hydrolytic degradation of the cross-linked collagen are cytotoxic.11,25 Cytotoxicity can be reduced by using low concentrations of glutaraldehyde for crosslinking8 and thoroughly washing the scaffolds before use in tissue engineering applications.11 D,L-Glyceraldehyde has been used to cross-link gelatin and is nontoxic. Furthermore, glyceraldehyde is a natural product of a metabolic process.17–19 Studies suggest that gelatin crosslinked with glyceraldehyde is well tolerated in vivo.17 However, the mechanism of the cross-linking reaction between glyceraldehyde and amine groups remains widely debated and the subject of continued discussion.26–28
Biomacromolecules, Vol. 10, No. 7, 2009
1679
Genipin is derived from geniposide, which is extracted from the fruit of Gardenia jasminoides Ellis. It has been used in herbal medicine as an antiphlogistics, an agent against inflammation and fever, and cholagogue, an agent that increases production of bile.29 Its anti-inflammatory properties could be useful for tissue engineering or would healing. Genipin forms dark blue pigments when reacted with amino acids or proteins and is used in the fabrication of food dyes.30 Although the mechanism of genipin cross-linking is not well understood,23 it is known that genipin reacts with free amine groups on proteins such as lysine, hydroxylysine, and arginine.23 One study of cross-linked gelatin hydrogels indicated that genipin cross-linked gelatin is about 10000 times less cytotoxic than glutaraldehyde cross-linked gelatin.23 Other studies have also shown a higher amount of cytotoxicity in glutaraldehyde cross-linking compared to genipin using MTT assays to compare cellular proliferation in vitro.21,22 The electrospinning process does not alter the chemical architecture of the gelatin. Amide A (or N-H stretching vibration) is found at 3300 cm-1 for both samples and is typical for gelatin.31,32 The amide I (CdO stretch) is located at 1640 cm-1 and 1636 cm-1 for the electrospun gelatin and the raw gelatin, respectively. This is close to other reported values of 1633 cm-1 and 1640 cm-1 for gelatin.31,33 The amide II (N-H bend and C-H stretch) vibrations for the raw gelatin and the electrospun gelatin are found at 1542-1544 cm-1. The reported values are between 1535 and 1550 cm-1, depending on the amount of R-helix, β-sheet, and disordered conformation present.31–33 Hence, the broadness of amide II may reflect the populations of these three conformations that are present. Finally, the amide III (C-N stretch plus N-H in-phase bending) band is observed at 1240 cm-1 in both the raw and the electrospun gelatin and is identical to that reported in the literature.31–33 In contrast, exposure to glutaraldehyde at low concentrations changes the fiber morphology due to the presence of water vapor during cross-linking.11 In addition, during cross-linking in the aqueous phase with either glyceraldehyde or genipin, the glyceraldehyde cross-linked fibers better maintain their morphology than do genipin cross-linked fibers. This could be attributed to the slower reaction rate of the genipin as compared to the glyceraldehyde allowing the gelatin to dissolve before the genipin can completely cross-link the fibers. The fibers crosslinked with reactive oxygen species maintained their morphology during cross-linking; however, during the dissolvability study, the reactive oxygen species cross-linked scaffolds were unstable and typically dissolved after 12 h. The reactive oxygen species not only initiate cross-linking, but also initiate a competing degradation reaction. The exact reaction mechanism is not well understood. We suggest that the degradation process predominates because these reactive oxygen species cross-linked scaffolds dissolve. On the other hand, the scaffolds cross-linked with glutaraldehyde, glyceraldehyde, and genipin remained stable in DMEM culture medium at 37 °C for over two weeks. To test the cytotoxicity of the cross-linked scaffolds, a cellular viability and proliferation assay was performed. All cross-linked scaffolds were compared to traditional polystyrene culture. The results indicated that each of the cross-linked scaffolds with glutaraldehyde, glyceraldehyde, and genipin performed as well or better than did the traditional two-dimensional polystyrene culture. Three-dimensional scaffolds give cells more room to spread and proliferate and also provide a transferrable support that can be transplanted into living organisms. Because there appears to be no difference between the glutaraldehyde crosslinked scaffold and the two-dimensional control, there remains
1680
Biomacromolecules, Vol. 10, No. 7, 2009
the possibility that, if the scaffolds are washed well with PBS to remove any residual glutaraldehyde, they may not prove cytotoxic.
Conclusions To create a scaffold that has the lowest toxicity possible, a system that uses less toxic solvents and different cross-linking agents was investigated. First, a solvent system of 50:30:20 acetic acid/ethyl acetate/water with gelatin at 10% (w/w) was determined to be the optimum system for electrospinning a scaffold with the desired properties. Next, several different methods were examined to cross-link the gelatin electrospun fibers. Because glutaraldehyde is known to be cytotoxic at high concentrations, it was necessary to explore other methods of cross-linking. Using reactive oxygen species from a plasma cleaner was an easy alternative; however, the degradation reaction dominated the cross-linking reaction and, thus, the scaffolds degraded after only a few hours in cell culture medium at 37 °C. Each of the cross-linked scaffolds supported cellular proliferation. As discussed previously, glyceraldehyde and genipin are both good options as cross-linking agents for their low toxicity and reliability. Follow up studies will further examine cell migration and differentiation in these biocompatible scaffolds. Acknowledgment. The author would like to acknowledge the funding for this research: NSF IGERT Proteins at Surfaces, NSF DMR-0704970, and NIH/NCI P01 CA098912 (to M.C.F.C.). The project described was partially supported by Grant No. 2 P20 RR016472-08 under the INBRE Program of the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH).
References and Notes (1) Gopal, R.; Kaur, S.; Ma, Z. W.; Chan, C.; Ramakrishna, S.; Matsuura, T. J. Membr. Sci. 2006, 281, 581–586. (2) Doshi, J.; Reneker, D. H. J. Electrostat. 1995, 35, 151–160. (3) Teo, W. E.; Ramakrishna, S. Nanotechnology 2006, 17, R89-R106. (4) Wang, X. Y.; Drew, C.; Lee, S. H.; Senecal, K. J.; Kumar, J.; Samuelson, L. A. J. Macromol. Sci., Part A: Pure Appl. Chem. 2002, A39, 1251–1258. (5) Drew, C.; Wang, X. Y.; Senecal, K.; Schreuder-Gibson, H.; He, J. N.; Kumar, J.; Samuelson, L. A. J. Macromol. Sci., Part A: Pure Appl. Chem. 2002, A39, 1085–1094. (6) Gibson, P.; Schreuder-Gibson, H.; Rivin, D. Colloids Surf., A 2001, 187, 469–481.
Sisson et al. (7) Ma, Z. W.; Kotaki, M.; Inai, R.; Ramakrishna, S. Tissue Eng. 2005, 11, 101–109. (8) Casper, C. L.; Yang, W. D.; Farach-Carson, M. C.; Rabolt, J. F. Biomacromolecules 2007, 8, 1116–1123. (9) Matthews, J. A.; Wnek, G. E.; Simpson, D. G.; Bowlin, G. L. Biomacromolecules 2002, 3, 232–238. (10) Li, W. J.; Laurencin, C. T.; Caterson, E. J.; Tuan, R. S.; Ko, F. K. J. Biomed. Mater. Res., Part A 2002, 60, 613–621. (11) Zhang, Y. Z.; Venugopal, J.; Huang, Z. M.; Lim, C. T.; Ramakrishna, S. Polymer 2006, 47, 2911–2917. (12) Li, M. Y.; Mondrinos, M. J.; Gandhi, M. R.; Ko, F. K.; Weiss, A. S.; Lelkes, P. I. Biomaterials 2005, 26, 5999–6008. (13) Huang, Z. M.; Zhang, Y. Z.; Ramakrishna, S.; Lim, C. T. Polymer 2004, 45, 5361–5368. (14) Song, J. H.; Kim, H. E.; Kim, H. W. J. Mater. Sci.: Mater. Med. 2008, 19, 95–102. (15) Tomihata, K.; Burczak, K.; Shiraki, K.; Ikada, Y. In ACS Symp. Ser. 1994, 540, 275–286. (16) Han, B.; Jaurequi, J.; Tang, B. W.; Nimni, M. E. J. Biomed. Mater. Res., Part A, 2003, 65A, 118–124. (17) Vandelli, M. A.; Rivasi, F.; Guerra, P.; Forni, F.; Arletti, R. Int. J. Pharm. 2001, 215, 175–184. (18) Wollensak, G.; Spoerl, E. J. Cataract Refract. Surg. 2004, 30, 689– 695. (19) Kosmala, J. D.; Henthorn, D. B.; Brannon-Peppas, L. Biomaterials 2000, 21, 2019–2023. (20) Nickerson, M. T.; Patel, J.; Heyd, D. V.; Rousseau, D.; Paulson, A. T. Int. J. Biol. Macromol. 2006, 39, 298–302. (21) Lien, S. M.; Li, W. T.; Huang, T. J. Mater. Sci. Eng., C 2008, 28, 36–43. (22) Bigi, A.; Cojazzi, G.; Panzavolta, S.; Roveri, N.; Rubini, K. Biomaterials 2002, 23, 4827–4832. (23) Sung, H. W.; Huang, D. M.; Chang, W. H.; Huang, R. N.; Hsu, J. C. J. Biomed. Mater. Res., Part A 1999, 46, 520–530. (24) Ulubayram, K.; Aksu, E.; Gurhan, S. I. D.; Serbetci, K.; Hasirci, N. J. Biomater. Sci., Polym. Ed. 2002, 13, 1203–1219. (25) Vanluyn, M. J. A.; Vanwachem, P. B.; Damink, L. O.; Dijkstra, P. J.; Feijen, J.; Nieuwenhuis, P. J. Biomed. Mater. Res. 1992, 26, 1091– 1110. (26) Tessier, F. J.; Monnier, V. M.; Sayre, L. M.; Kornfield, J. A. Biochem. J. 2003, 369, 705–719. (27) Gerrard, J. A.; Brown, P. K.; Fayle, S. E. Food Chem. 2002, 79, 343– 349. (28) Wollensak, G.; Iomdina, E. Acta Ophthalmol. 2008, 86, 887–893. (29) Akao, T.; Kobashi, K.; Aburada, M. Biol. Pharm. Bull. 1994, 17, 1573– 1576. (30) Touyama, R.; Takeda, Y.; Inoue, K.; Kawamura, I.; Yatsuzuka, M.; Ikumoto, T.; Shingu, T.; Yokoi, T.; Inouye, H. Chem. Pharm. Bull. 1994, 42, 668–673. (31) Yannas, I. V. J. Macromol. Sci. 1972, C7, 49–104. (32) Cao, H.; Xu, S. Y. Food Chem. 2008, 108, 439–445. (33) Tengroth, C.; Gasslander, U.; Andersson, F. O.; Jacobsson, S. P. Pharm. DeV. Technol. 2005, 10, 405–412.
BM900036S
