|
|
||||||||
RESEARCH COMMUNICATION |
a Laboratory of Biophysics and Surface Analysis, School of Pharmaceutical Sciences, The University of Nottingham, Nottingham NG7 2RD, United Kingdom
b Department of Chemical Engineering, Massachusetts Institute of Technology, E25342, Cambridge Massachusetts 02139, USA
| ABSTRACT |
|---|
|
|
|---|
Key Words: tissue engineering neurite extension integrins patterning
| INTRODUCTION |
|---|
|
|
|---|
An important challenge in tissue engineering is to apply the principles of molecular biology to control the spatial organization of cells. This is vital in the bioengineering of tissues that require precisely defined cellular architectures. For example, the functioning of tissues such as nerves and blood vessels is dependent on the controlled orientation of cells. For many tissue types, the spatial organization of cells is required to ensure that cell-to-cell interactions occur (18, 27, 28). These interactions are essential in cell phenotype preservation (29).
Achieving spatially controlled cell engineering requires parallel development in molecular biology and biomimetic materials technology. In the field of molecular biology, there has been a continuing refinement in the our understanding of ligand-to-cell adhesion receptor interactions. This has generated a large number of cell adhesion motifs that can control cell adhesion behavior via specific receptor interactions (30, 31). In the design of biomimetic materials, the aim is to use these motifs to promote cell interactions with synthetic materials. Most biomimetic materials for cell engineering have been based on biodegradable polymers, which are removed from the engineered tissues by hydrolysis and dissolution of breakdown products (32). The major technological challenge in the fabrication of these biomimetic materials is the development of techniques of engineering biodegradable polymer surfaces (1, 33). This surface engineering has been used to immobilize motifs homogeneously over surfaces. However, to fabricate biomimetic materials for spatially controlled cell engineering, it is necessary to restrict motif immobilization to predefined micron-scale patterns.
Recent advances in patterning technology have generated a range of techniques with which biomolecules can be immobilized on surfaces with micron-scale precision. These techniques include lithographic methods (34) that use patterned masks to restrict the location of interactions between a beam of light, ions or electrons, and a surface (12) and micro-contact printing techniques (35). Most patterning techniques have been developed on a small set of materials with chemical modifications that require highly specialized surface reactions. This can restrict the types of ligands and surfaces that can be patterned. The objective of the current work was to develop a generic technology that can form patterns of any ligand onto a biodegradable polymer surface.
As with many aspects of tissue engineering research, this work requires a multidisciplinary strategy that combines molecular biology with biodegradable polymer synthesis, surface engineering and analysis, and patterning technology. Our method generates micron-scale patterns of any biotinylated ligand on the surface of a biodegradable block copolymer, polylactide-poly(ethylene glycol) (PLA-PEG).2 Nanometer control over the location of biomolecule has been achieved. This control has been confirmed by molecular resolution of protein molecules on the patterned surfaces using atomic force microscopy (AFM). Spatial control over cell development has been observed when using these templates to culture bovine aortic endothelial cells and PC12 nerve cells. In addition, directional control over neurite extension from PC12 cells has been achieved on patterns composed of peptides containing the IKVAV sequence (15, 16) through precise control of the location of receptor-mediated interactions.
| MATERIALS AND METHODS |
|---|
|
|
|---|
-Amine
-hydroxy PEG (Shearwater Polymers, Inc.; average mol wt, 3.8k) was stirred overnight with NHS-biotin (Fluka, Milan, Italy) and triethylamine in methylene chloride and acetonitrile, at room temperature, under argon. The biotinylated PEG was then isolated by vacuum filtration and dried azeotropically from toluene. Then recrystallized (l-)lactide was polymerized from the
-hydroxy PEG-biotin end by refluxing in toluene, using stannous 2-ethylhexanoate as a catalyst. PLA-PEG-biotin was precipitated from a methylene chloride solution by the addition of cold ether.
1H-NMR (nuclear magnetic resonance) at each stage confirmed the attachment of biotin to the PEG chain. Specifically, attachment of biotin-NHS to the end group amine of
-amine
-hydroxy PEG to form an amide bond was confirmed by shift of the free amine protons to an amide proton at 7.8 ppm and the appearance of a triplet (methylene from biotin arm alpha to the amide) at 2.05 ppm. The proton signals from the bicyclic biotin structure owing to the (2)methine protons (4.3 and 4.2 ppm) and urea protons (6.45 and 6.35 ppm) can be seen throughout the synthesis of PLA-PEG-biotin: the biotin structure remains intact and is not damaged from the lactide polymerization onto HO-PEG-biotin.
The molecular weight of the PLA-PEG-biotin was determined by 1H-NMR, using the PEG signal as a reference. Gel permeation chromatography revealed one peak indicative of pure material. PLA molecular weight for this study was 9.2k.
Mold fabrication
A patterned poly(dimethyl siloxane) (PDMS) mold was formed by curing its prepolymer (Sylgard 184, Dow Corning) on a patterned master prepared photolithographically by exposing and developing a photoresist pattern on gallium/arsenide wafers (36). The PDMS mold bearing the negative pattern of the master was peeled off and washed repeatedly with ethanol, hexane, and de-ionized water. After drying under argon, the mold was placed onto an epitaxially grown gold surface (see
Fig. 2) (37). The capillaries were rendered hydrophilic by plasma treatment with O2 (load coil power Å 200 W) for 1 s (Bio-Rad RF Plasma Barrel Etcher)
|
Avidin pattern formation
Within 1 min of plasma treatment, the mold was placed onto a film of the PLA-PEG-biotin. The polymer film was formed from a 1 mg/mL solution of the polymer in trifluoroethanol, drop cast onto a polystyrene substrate, and dried under vacuum. A 500 µg/ml solution of rhodamine-labeled avidin (av-R) (Sigma, Dorset, U.K.) was prepared using distilled water. Approximately 1 mL of this av-R solution was dropped onto the PLA-PEG-biotin surface. The drop was positioned so that the av-R solution wetted an end of the mold where the capillaries were open. After 1 h of contact, the av-R solution was removed by blotting and replaced with approximately 20 ml of distilled water; after 5 min, the water was removed. This washing procedure was repeated five times. Then the sample was immersed in water and the mold removed by peeling the PDMS and PLA-PEG-biotin surfaces apart. The sample was washed with an additional 100 mL of water.
Atomic force microscopy
Images were obtained using a Digital Instruments (Santa Barbara, Calif.) multimode scanning probe microscope with a Digital Instruments Nanoscope IIIa controller. Images were acquired in tapping mode using sharpened silicon nitride tips with cantilever resonance frequencies in the range of 307375 kHz. Simultaneous topographic and phase images were obtained at a scan rate of 1 Hz.
Cell culture
All reagents were obtained from Sigma unless otherwise noted. Bovine aortic endothelial cells (BAECs) were maintained in low glucose Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum, 0.5% penicillin, 0.5% streptomycin, and 1% L-glutamine in a humidified incubator at 37°C and 5% CO2. Cells were passaged by trypsinization before reaching confluence, usually every fifth day. Fresh media were added every other day. PC12 cells (ATCC CRL-1721) were grown in suspension culture in F-12K media supplemented with 15% horse serum, 2.5% fetal bovine serum, 0.5% penicillin, 0.5% streptomycin, and 1% L-glutamine in a humidified incubator at 37°C and 5% CO2. Cells were passaged 1:5 every third day
Cell spreading experiment
The biotinylated polypeptides were synthesized and their sequences confirmed at the MIT Biopolymer Processing Laboratory. The samples were sterilized under UV light for 10 min. A 1 mg/mL solution of peptide (biotin-G11GRGDS for endothelial cell experiments, biotin-G5CSRARKQAASIKVAVSADR for PC-12 cell experiments) was added to the tissue culture plate and allowed to incubate at 37°C on a shaker plate for 30 min. The samples were washed three times with sterile phosphate-buffered saline (PBS) prior to seeding with cells.
BAECs between passages 7 and 9 were removed from tissue culture flasks by trypsinization, pelleted by centrifugation, resuspended, washed three times, and diluted to the appropriate concentration in serum-free DMEM. PC12 cells were primed with culture medium containing 50 ng/mL 7S nerve growth factor (NGF) 48 h prior to the experiment. The cells were pelleted by centrifugation, washed three times with serum-free F-12K medium, passed through a 22-gauge needle to obtain a single cell suspension, and diluted to the appropriate concentration with serum-free F-12K medium supplemented with 50 ng/mL 7S NGF. For both cell types, approximately 10,000 cells/cm2 were added to each sample and the plates returned to the incubator for the duration of the experiment (48 h). At the end point of the experiment, the samples were washed gently with PBS to remove unattached cells and visualized with a Nikon Diaphot TMD inverted microscope equipped with a Hitachi HV-C20 high-resolution CCD video camera using phase contrast objectives. Images were digitized using NIH Image (v1.61) image analysis software. In addition, some BAEC samples were fixed in 10% neutral buffered formalin for 10 min, washed with water, and stained with hematoxylin for visualization.
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
As shown schematically in
Fig. 1,
when films of PLA-PEG-biotin are exposed to aqueous solutions of avidin, the polymer surface immobilizes avidin molecules. Each avidin molecule can then act as a bridge between the biotinylated polymer and a biotinylated ligand. PLA-PEG was chosen as the foundation for the surface engineering because this polymer has become established as a material that possesses excellent bulk and surface properties for biomaterial applications (38). In addition, the PEG chains act as flexible, hydrophilic spacers for the biotin moieties. The preservation of the biotin structure during the synthesis of PLA-PEG-biotin was verified by 1H-NMR, and the avidin binding ability of the tethered biotin was confirmed by surface plasmon resonance analysis (39).
|
A microfluidic network system (40, 41) was used to form a pattern of avidin molecules on PLA-PEG-biotin film surfaces. The pattern was composed of avidin functionalized lines, with widths of 70, 50, 30, or 12 µm, separated by 250 µm gaps that were free of the protein. To achieve pattern formation, an aqueous solution of avidin was allowed to flow through capillaries, created by placing a PDMS mold (see
Fig. 2)
on PLA-PEG-biotin surfaces. The mold must protect the polymer surface from the avidin solution over the gap regions while allowing the avidin solution to flow over the polymer in the capillaries. In its native state, the hydrophobicity of the PDMS ensured that regions of the polymer surface in contact with the mold base were protected from the avidin solution. However, this hydrophobicity also inhibited solution flow through the capillaries, so an oxygen plasma technique was developed to render the capillary walls hydrophilic while maintaining the hydrophobicity of the mold base (
Fig. 2A). Base regions of the mold were shielded from the activity of the plasma by conformal contact with an epitaxially grown gold surface. The mold was then transferred to the PLA-PEG-biotin surface. Contact angle analysis confirmed the differential hydrophobicity/hydrophilicity of shielded and unshielded regions of the mold. Shielded regions displayed contact angles of 105° with water, whereas unshielded regions were saturated with water drops. Further evidence of the successful treatment of the capillary walls was provided by the rapid flow of water through the capillaries.
The flow of the avidin solution through the hydrophilic capillaries was monitored by fluorescence microscopy using avidin labeled with rhodamine (av-R). After 1 h of exposure of PLA-PEG-biotin to the av-R solution, the sample was washed thoroughly and the mold was removed by peeling. Resulting patterns of av-R on the PLA-PEG-biotin surface are shown in
Fig. 3A,B.
The presence of a sharp boundary between avidin-immobilized and protein-free regions has been confirmed by phase detection AFM (
Fig. 3C), in which differences in viscoelastic and adhesive properties of avidin and PLA-PEG-biotin generate image contrast (42).
|
The thickness of the avidin layer deposited on the PLA-PEG-biotin was measured by tapping mode AFM. The AFM image in
Fig. 3D shows the topography of a PLA-PEG-biotin surface at a protein boundary. The edge of the line was resolved, and cross-sectional analysis recorded a step height of less than 5 nm. Given that the dimensions of the avidin molecule have been estimated as 5.6 nm x 5.0 nm x 4.0 nm by X-ray crystallography (43), a step height of 5 nm is indicative of a monolayer coverage. The inset image shows a 100 nm x 100 nm scan of the channel on which molecular-resolution of the protein has been achieved. On some areas of the avidin channel, protein aggregates are evident. These aggregates were resistant to washing with water. The AFM image in
Fig. 3D also demonstrates the exceptional control of avidin distribution along the channel edge. Lateral deviations of this edge from a straight line are small. The largest lateral deviation on this image is 30 nm, equivalent to approximately 6 avidin molecules. Most deviations were found to be less than 20 nm in length.
The micron-scale patterns of avidin on the PLA-PEG-biotin surfaces were then used to immobilize biotinylated peptides containing either the fibronectin fragment RGD or the laminin fragment IKVAV. Initial cell culture studies were performed on the RGD-presenting patterns, using BAECs to prove that the patterns could spatially organize the adhesion of an anchorage-dependent cell type. As shown in
Fig. 4,
the BAECs adhered and spread on the RGD-functionalized lines but did not adhere to unfunctionalized areas between the lines. Complete cell coverage of the 70 and 50 µm width lines was achieved, but little cell adhesion occurred to the 30 or 12 µm-wide lines.
|
Having proved that patterns on the biodegradable surface could be used in spatially controlled cell engineering, we investigated the control of a cell developmental process, namely, directed neurite extension. Directed neurite extension occurs during peripheral nerve regeneration, and it is known that one component of this extension is integrin-to-laminin interactions (15, 4447). PC12 cells were used in this study because of their known laminin-induced neurite extension. As shown in
Fig. 5,
PC12 cells showed selective adherence to the IKVAV functionalized lines, with only a small degree of cell adhesion between the lines (
Fig. 5). In addition, no cell adhesion was observed on negative control samples consisting of PLA-PEG-biotin-avidin patterns in the absence of the biotinylated IKVAV sequence. Directionally controlled neurite outgrowth was stimulated by the IKVAV micropattern, with neurites extending between groups of cells often hundreds of microns apart (
Fig. 5B, C). The extent of control over neurite growth was demonstrated by the morphology of many neurites that approached the boundary between the functionalized and unfunctionalized surfaces, but were always effectively restricted from crossing the interface (
Fig. 5D).
|
| CONCLUSIONS |
|---|
|
|
|---|
The patterning technique described in this paper provides a generic technology by which any biotinylated ligand can be patterned onto the surface of a biodegradable polymer. The polymer structure can be tailored to meet the requirements of specific bioengineering applications. The ability to pattern such a wide range of ligands opens new possibilities in spatially controlled cell engineering, because the ability to control cell organization on the templates becomes limited by our understanding of the molecular biology of ligand-to-receptor binding and its influence on cellular development, not by surface engineering constraints.
Advances in cell patterning have recently been reported, primarily those on gold surfaces patterned using molecular self-assembly (11, 13). These studies utilize the ability to precisely control the patterning procedures with nanometer precision of molecular immobilization. If the field of tissue engineering is to exploit these fundamental studies, it is necessary to achieve similar patterning precision on biodegradable polymer templates. The detailed surface analysis of our patterned surfaces has revealed that our technique retains the precision of ligand immobilization.
Finally, the ability to direct neurite extension on the patterned templates demonstrates that natural biological organizational principles can be mimicked successfully. Research on the mechanisms of peripheral and central nerve regeneration continues to reveal mechanisms by which medical interventions can augment natural regenerative processes (4751). Patterned templates that determine the nature and location of neurite-to-matrix interactions offer the potential to refine our ability to guide regeneration.
|
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Abbreviations: AFM, atomic force microscopy; av-R, rhodamine-labeled avidin; BAECs, bovine aortic endothelial cells; DMEM, Dulbecco's modified Eagle's medium; NGF, nerve growth factor; NMR, nuclear magnetic resonance; PBS, phosphate-buffered saline; PDMS, poly(dimethyl siloxane); PLA-PEG, polylactide-poly(ethylene glycol). ![]()
Received for publication March 27, 1998.
Revision received June 17, 1998.
| REFERENCES |
|---|
|
|
|---|
-V-ß-3-mediated fibroblast spreading and 140 nm for focal contact and stress fiber formation. J. Cell Biol. 114, 10891100
This article has been cited by other articles:
![]() |
S.-H. Mirmalek-Sani, R. S. Tare, S. M. Morgan, H. I. Roach, D. I. Wilson, N. A. Hanley, and R. O.C. Oreffo Characterization and Multipotentiality of Human Fetal Femur-Derived Cells: Implications for Skeletal Tissue Regeneration Stem Cells, April 1, 2006; 24(4): 1042 - 1053. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. L. Hench and J. M. Polak Third-Generation Biomedical Materials Science, February 8, 2002; 295(5557): 1014 - 1017. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |