<%BANNER%>

Contribution of modulus to the contact guidance of endothelial cells on microtextured siloxane elastomers


PAGE 1

CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE OF ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS By WADE RICHARD WILKERSON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2001

PAGE 2

Copyright 2001 by Wade Richard Wilkerson

PAGE 3

This thesis is dedicated to my family and friends, and in loving memory of James R. Wilkerson and James S. Blair.

PAGE 4

ACKNOWLEDGMENTS I would like to express my deepest gratitude to my advisor and committee chairman, Dr. Anthony Brennan, for his understanding, advice, and support during my time at the University of Florida. I would also like to thank the other members of my supervisory committee, Dr. Christopher Batich, and Dr. Ronald Baney, for their counsel and accommodation during this process. I am especially grateful to Dr. Laurie Gower, for agreeing to be a substitute member of my committee at such late notice. The members of Dr. Brennans research group have been a great source of advice, experience and camaraderie over the past two years. Specifically I would like to thank Jeanne Macdonald for taking responsibility of the lab and serving as group mentor and general well of knowledge. Licheng Zhao and Jeremy Mehlem both were valuable sources of experience and information. I would like to thank Adam Feinberg for his expertise with the AFM, SEM, and computer issues, as well as for his friendship. Chuck Seegert provided the wafers and the nuclear form method for this study, including a wealth of information on statistics, cell methods, and staining protocols. Amy Gibson and Leslie Wilson provided the modulus data and the formulations for the modified elastomer samples. Their hard work and flexibility are greatly appreciated. Other members of the Brennan research group, Clay Bohn, Brian Hatcher, and Nikhil Kothurkar, have been an invaluable source of friendship and support. Paul Martin was always accommodating with regard to cell culture techniques and advice that was crucial to the completion of this work. I would also like to thank many other iv

PAGE 5

graduate students in the Materials Science department and the Biomedical Engineering program, specifically Brad Willenberg, Brian Cuevas, Bob Hadba, Josh Stopek, Dan Urbaniak, and Jamie Rhodes. I am indebted to Dr. C. Keith Ozaki and the members of his research group, specifically Zaher Abouhamze. Zaher was always helpful and went above and beyond the call of duty in assisting me at various times during this project. Dr. Edward Block graciously supplied the endothelial cells used in this study, and Bert Herrara was very generous and patient in supplying cells on a weekly basis. Nina Klingmann and Dr. Tina Lam were also a great source of information and I am grateful to Nina for allowing use of her incubators and cell culture equipment. I would be remiss if I did not thank my family and friends who have supported me throughout this process and long before. My parents have provided me with every opportunity to succeed, and without their love and support, I could never achieve the goals that I have set. My wife, Laura, has been the greatest blessing and greatest friend throughout our relationship and especially during the first year of our marriage. Her understanding and support has no bounds, and her love has kept me going whenever I began to doubt. Most of all I would like to thank God for without Him, none of this would be possible. v

PAGE 6

TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES...........................................................................................................ix ABSTRACT.......................................................................................................................xi CHAPTERS 1 INTRODUCTION............................................................................................................1 2 BACKGROUND..............................................................................................................4 Contact Guidance............................................................................................................4 Surface Topography....................................................................................................4 Surface Chemistry.....................................................................................................11 Endothelial Cells...........................................................................................................14 Vascular Grafts.........................................................................................................16 Cell Adhesion............................................................................................................19 Silicones........................................................................................................................22 Use as a Biomaterial.................................................................................................23 Surface Energy..........................................................................................................25 Mechanical Properties...............................................................................................27 3 CHARACTERIZATION OF SILICONE ELASTOMER SUBSTRATES....................31 Introduction...................................................................................................................31 Materials.......................................................................................................................31 Methods.........................................................................................................................33 Elastomer Preparation...............................................................................................33 Modulus Determination............................................................................................34 Preparation of Textured Surfaces..............................................................................35 Surface Treatment.....................................................................................................36 Radiofrequency glow discharge treatment............................................................37 Fibronectin adsorption..........................................................................................37 Surface Energy..........................................................................................................38 Dynamic Contact Angles..........................................................................................38 vi

PAGE 7

Optical Profilometry.................................................................................................39 Results and Discussion.................................................................................................40 Modulus Results........................................................................................................40 Surface Energy of Silicone Elastomers.....................................................................41 Contact Angles of Treated Surfaces.........................................................................44 Plasma Treatment Issues...........................................................................................45 Dynamic Contact Angle............................................................................................49 Optical Profilometry.................................................................................................52 4 CONTACT GUIDANCE OF ENDOTHELIAL CELLS...............................................57 Introduction...................................................................................................................57 Materials.......................................................................................................................57 Elastomer Substrates.................................................................................................57 Cell Culture and Seeding..........................................................................................58 Methods.........................................................................................................................59 Elastomer Sample Preparation..................................................................................59 Surface Treatment by Fibronectin............................................................................59 Surface Treatment by RFGD Plasma........................................................................59 Cell Culture Techniques...........................................................................................60 Cell passage procedure.........................................................................................60 Determination of cell suspension concentration...................................................61 Cell Seeding on Samples..........................................................................................61 Cell Staining and Image Capture..............................................................................62 Image Analysis..........................................................................................................63 Results and Discussion.................................................................................................65 Contact Guidance on Textured Surfaces...................................................................65 Contact Guidance on Textured Surfaces of Varying Modulus.................................72 5 CONCLUSIONS AND FUTURE WORK.....................................................................80 Conclusions...................................................................................................................80 Microtextured Surfaces.............................................................................................80 Surface Energy and Treatment..................................................................................81 Contact Guidance on Textured Elastomers...............................................................82 Future Work..................................................................................................................83 Surface Treatment.....................................................................................................83 Topographical Design...............................................................................................83 Cell Studies...............................................................................................................84 LIST OF REFERENCES...................................................................................................85 BIOGRAPHICAL SKETCH.............................................................................................97 vii

PAGE 8

LIST OF TABLES Table Page 2.1 Enrichment of proteins adsorbed on polyethylene exposed to blood plasma...........14 3.1 Typical properties of Silastic T-2 Silicone Moldmaking Rubber (from Dow Corning product information sheet)..........................................................................32 3.2 PDMS additives to silicone elastomer system..........................................................33 3.3 Curing conditions for silicone elastomer samples....................................................34 3.4 Surface tension of liquids used for surface energy determination by contact angle analysis......................................................................................................................42 3.5 Contact angles and surface free energy of various substrates...................................43 3.6 Captive bubble contact angles on treated surfaces...................................................44 3.7 Water contact angles on plasma treated silicone samples after exposure to air over time...........................................................................................................................48 3.8 DCA data on silicone elastomer modified with non-functionalized PDMS oligomers...................................................................................................................51 3.9 Optical profilometer data of ridge widths and groove depths of the 1.5 and 5 m deep wafers and elastomer copies. The wafers should have a constant ridge width and the elastomer copies have ridge widths of 5 m, 10 m, and 20 m................55 4.1 Properties of elastomeric substrates for contact guidance experiments....................58 viii

PAGE 9

LIST OF FIGURES Figure Page 2.1 Light micrographs and measurement protocols of 6 mm diameter punches of corneal epithelial tissue explants (f) over (A & C) smooth PS surface and (B & D) a surface with 1 m deep microgrooves separated by 1 m.........................................9 2.2 Cutaway view of an artery showing the three main layers.......................................15 2.3 Light microscopy image of confluent porcine vascular endothelial cells (PVECs) grown on tissue culture polystyrene (Image taken by L. Zhao and W. Wilkerson).16 2.4 Polydimethylsiloxane, trimethylsiloxy terminated...................................................22 3.1 Representation of etched patterns on silicon wafers.................................................35 3.2 Sample layout for textured substrates examined using DCA...................................39 3.3 Elastic modulus of modified silicones as measured by tensile testing.....................41 3.4 Zisman plot of unmodified silicone for calculation of surface free energy..............44 3.5 Profilometer image of surface damage due to plasma of 15% vinyl tris textured elastomer sample (smooth area)................................................................................46 3.6 Profilometer image of surface damage due to plasma of 15% vinyl tris textured elastomer sample.......................................................................................................47 3.7 AFM Image of crack on plasma treated sample.......................................................47 3.8 Profilometry data for non-treated (A) and plasma treated (B) 1.5 m deep elastomer textured surfaces at the 5 m width spacing.............................................................48 3.9 Force-distance curve from DCA of unmodified silicone elastomer in water...........50 3.10 Force distance curve from DCA on a textured silicone elastomer substrate..........52 3.11 3-D images of 5 m deep elastomer samples copied off of epoxy taken with optical profilometer (50X). (A-C) are examples of the ridges at the different spacing. (D) demonstrates the smooth/textured interface.......................................53 ix

PAGE 10

3.12 3-D images of 5 m deep etched silicon wafers at each width spacing, taken with optical profilometer.................................................................................................54 3.13 3-D images of 5% LMW elastomer copies (A, B) of a 1.5 m wafer (C, D).........55 4.1 Representation of the values measured in the nuclear form factor...........................63 4.2 Steps in image processing technique of hematoxylin stained nuclei. The original picture (A) is contrasted (B), rotated to align the textures (C), and then measured for length (D) and width (not shown).............................................................................64 4.3 Main Effects Plot of the data means of fibronectin coated unmodified elastomers at various ridge widths (Feature) and groove depths (Depth)......................................67 4.4 Interaction plot representing the change in contact guidance with depth and feature width for fibronectin coated silicone elastomer........................................................68 4.5 Main Effects Plot of the data means for plasma treated unmodified elastomers......69 4.6 Interaction plot representing the change in contact guidance with depth and feature width for plasma treated silicone elastomer..............................................................70 4.7 Comparison of surface treatments on textured unmodified silicone elastomer by ridge width and groove depth (error bars represent standard error of mean)...........71 4.8 PVECs grown on untextured fibronectin coated LMW sample stained with hematoxylin...............................................................................................................72 4.9 PVECs grown on 5 m spacing 1.5 m deep untreated LMW stained with hematoxylin...............................................................................................................72 4.10 Main effects plot comparing log (L/W) to feature width, feature depth, and material used on fibronectin-coated elastomers.....................................................73 4.11 Interaction plot representing the change in factors of PVECs grown on fibronectin coated materials.....................................................................................................73 4.12 Comparison of materials on textured unmodified silicone elastomer by ridge width on 5 m deep grooves (error bars represent s.e.m.)...............................................75 4.13 Comparison of materials on textured unmodified silicone elastomer by ridge width on 5 m deep grooves (error bars represent s.e.m.)...............................................75 4.14 Comparison of ridge width and groove depth to the high modulus material (LMW) and the low modulus material (Tris) (error bars represent s.e.m.)......................76 x

PAGE 11

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE OF ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS By Wade Richard Wilkerson December, 2001 Chairman: Dr. Anthony B. Brennan Major Department: Biomedical Engineering Program Contact guidance is a term used to describe a materials ability to direct the alignment and growth patterns of biological cells or tissue. It has long been understood that the surface a cell grows on impacts its size, shape, and metabolism. Typically, when cells are exposed to mechanical features such as ridges and grooves, the cells align and travel along the length. Surface chemistry plays a significant role in the attachment of cells to a substrate and in their movement on that surface. The objective of this study was to study the effect of modulus as well as surface texture dimensions on vascular endothelial cells (ECs). To examine the effects of contact guidance on silicone elastomers, microtextured substrates were produced with reproducible and well-defined surfaces. Ridges of 10,000 m length were fabricated at 3 different widths: 5 m, 10 m, and 20 m, separated by 5 m wide grooves to determine the effect of separation of features on the alignment of porcine vascular endothelial cells (PVECs). Two depths were examined: 5 m and 1.5 m. xi

PAGE 12

Elastomer samples were examined with contact angles to determine their relative wettability and surface free energy. Formulations of elastomer with both functionalized and non-functionalized PDMS oligomer additives to alter the modulus were examined by contact angle, with no significant difference in surface energy. Surfaces were treated with fibronectin and radiofrequency glow discharge plasma in argon for 5 minutes at 50 W. Both treatments significantly increased the hydrophilicity after treatment, as measured by captive bubble contact angles. Dynamic contact angle analysis of textured surfaces showed a difference in smooth and textured areas as well. Contact guidance of PVECs on textured silicone elastomers was measured by the nuclear form factor, in which the log of the ratio of nuclear length to width was presented. Results demonstrated that as the ridge width decreased from 20 m to 5 m contact guidance increased, as well as when the depth of the grooves increased from 1.5 m to 5 m. Data analysis showed that the groove depth was the most important factor in nuclear alignment. Contact guidance on fibronectin-coated elastomers was examined to determine the effect of modulus. It was expected that higher modulus materials would increase the effect of contact guidance. Elastic modulus on 4 elastomers was measured by tensile tests and resulted in a range of values from 0.3 MPa to 2.3 MPa. There was no significant difference in the contact guidance on the deep 5 m grooves with varying modulus. The 1.5 m deep grooves showed a significant increase in the alignment of cells to the groove in the highest modulus material compared to the lowest modulus material for the 5 m and 10 m wide ridges. The conclusion to be taken from these data is that modulus does seem to play a role in the determination of contact guidance, but other factors such as groove width and especially depth are more significant. xii

PAGE 13

CHAPTER 1 INTRODUCTION The ability to predict and control a biological response to a biomedical device would be a skill of dramatic technological and economic importance. Since man first attempted to replace natures mechanisms and structures with artificial substitutes, he has met mainly with frustration marked with varying degrees of success. For implant materials, the factors that determine success are many, but the interaction between the surface and the surrounding tissue is one of the most important. The characteristics of that surface shape that interaction, and their secrets are slowly becoming known. Contact guidance is a term used to describe a materials ability to direct the alignment and growth patterns of biological cells or tissue. It has long been understood that the surface a cell grows on impacts its size, shape, and metabolism. By determining the aspects of the material and the surface that influence contact guidance, there is a better opportunity to design a hierarchical system to elicit the desired response. Contact guidance can be controlled by topography and surface chemistry. Typically, when cells are exposed to mechanical features such as ridges and grooves, the cells align and travel along the length. The addition of roughness at certain levels can improve a biomaterials ability to promote cell adhesion, while disrupting adhesion at different levels.[1-3] Patterning surface chemistry on samples to change the wettability and surface energy has been very successful in controlling cell growth. Alternating strips or islands of adhesive proteins and materials with different surface energies have been examined. The incorporation of texture with surface chemistry allows for mutual interactions to 1

PAGE 14

2 enhance the desired response.[2] As a better understanding of the principles involved is obtained, more subtle influences on the control of contact guidance can be examined. The objective of this study was to study the effect of modulus as well as surface texture dimensions on vascular endothelial cells (ECs). ECs are important regulators of homeostasis in the human body, and a crucial component of the cardiovascular system. Diseases of this system currently contribute to more deaths than any other disease or cause. By controlling the alignment and growth patterns of these cells and their tissues, improvement of medical devices such as vascular grafts is possible. Modulus has been shown to play an important role in adhesion of biofilms on substrates in marine environments,[4, 5] and coupling the effects of modulus with topographical features is another important step in designing the behavior of materials. The specific aims of this project involved the study of endothelial cells grown on microtextured silicone elastomers. Specifically, the effect of modulus was hypothesized to increase the effect of contact guidance as the modulus increased. This hypothesis was based on the observations of Kendall and others that modulus played an important role in the adhesion of biofilms on elastomer surfaces. These theories will be discussed in detail in the following chapters. To truly understand the system, another specific aim was to determine the importance of feature dimensions on this system. The hypothesis to be tested was that the depth of grooves in a surface played a more important role than the spacing between the grooves, and that a deeper groove increased the contact guidance of an EC on silicone. When the groove depth remained constant, the hypothesis being tested was that the grooves spaced closer together would improve the cells ability to direct cell growth. The importance of groove depth and spacing has been proven before

PAGE 15

3 on different systems to confirm these hypotheses, but the effect of modulus on similar materials has not truly been examined. To accomplish this objective, micropatterned silicone elastomers were fabricated with additives to change the elastic modulus while keeping the surface energy relatively constant. Surface topography was examined by various methods such as optical profilometry. Due to the low energy nature of the silicone, samples were treated with an argon plasma or coated with an adsorbed layer of fibronectin to improve cell adhesion. Porcine vascular endothelial cells (PVECs) were examined using the nuclear form factor, which is a measure of the alignment of a cell to defined topography by measuring the dimensions of its nucleus. The microtextured features were designed to be able to compare the degree of contact guidance by varying the feature width and depth. The modification of the samples with functionalized siloxane oligomers allowed for the variation of the modulus while studying the growth of cells on materials that are otherwise similar. These novel systems allow for the quantification of cell alignment, as well as a measure of a surfaces capability for contact guidance

PAGE 16

CHAPTER 2 BACKGROUND Contact Guidance The reaction of cells in vitro to the substrate they come into contact with has been traditionally separated into two main features: topography and surface chemistry. Cells adhere to surfaces via specific adhesion molecules that interact with proteins adsorbed onto the surface of the substrate. Thus, if the surface chemistry is favorable to adhesive protein adsorption, then the material should be favorable to cell adhesion. However, the topography of the surface is also important not only in the adhesion of the cells to the surface, but also in the behavior of the cells metabolism and growth patterns after the initial contact. The purpose of understanding and controlling cell and tissue growth on artificial materials is to be able to design and implant medical devices that improve biocompatibility and functionality. Surface Topography The response of the cell to topography has been referred to as contact guidance.[6, 7] The first known reference to the effect of substrate topography on growth characteristics comes from the growth of embryonic cells on plasma clots and spider webs.[8] Rovensky et al. used V-shaped grooves formed from copies of music records to show that chick embryo fibroblasts migrated from the bottom of the groove to the top over a period of hours and aligned along the texture.[9] Dunn and Heath examined chick heart fibroblasts growing on glass fibers to examine the effect of radius of curvature on 4

PAGE 17

5 the cells.[10] They discovered an important aspect of contact guidance in that the shape of the substratum causes mechanical impediments to the formation of cytoskeletal bundles important for cell locomotion. By correlating this discovery with linear slopes and discontinuities, they determined that an angle of deflection greater than 8 between two planes of a prism affected their cells and made them less likely to cross over the ridge. Another important aspect of this paper introduces the use of the nucleus to quantify contact guidance, by factoring the elongation of the nucleus with respect to the orientation of a fiber. This technique will be used in the results portion of this thesis, and will be explained in more detail later. Cells have also been shown to move along topographical features, and drastically change their morphology in response to this texture.[11-15] Clark et al. examined several cell lines in response to specific topography. They first examined baby hamster kidney (BHK) cells and embryonic neural cells in response to a single 5 m step. Surprisingly, this is one of the only studies to examine such a simple topography. They found that the cells were inhibited from both climbing up as well as down the step, but tended to align along the ridge.[16] As a follow-up, they examined the same cells on grooved substrata of varying widths and depths (4-24 m repeat width, 0.2-1.9 m depth) and found that groove depth increased cell alignment and proved to be more important.[17] Similar results were found using rat bone marrow cells on poly-L-lactic acid (PLA) grooves, with an increase of extracellular material being deposited along the grooves.[18] Schmidt and von Recum show distinct morphology changes on pitted surfaces for macrophages and increased spreading on smooth surfaces.[19] Macrophages were also demonstrated to have the unique ability to align to extremely shallow grooves, from 30-70 nm.[20, 21]

PAGE 18

6 The goal of substrate topography is to direct the cells to grow in a certain pattern and direction. Cells tend to align along a groove and move along this surface. Studies have shown that moderately porous materials improve cellular adhesion, which is possibly due to mechanical stability along with increased surface area for adhesion.[22-24] Endothelial cells (ECs) also have been shown to align along the direction of fluid flow.[25-27] In addition flow has been demonstrated to play a role in altering the mechanical properties of ECs. When subjected to a shear stress of 2 Pa over 24 hours, the endothelial cells gradually increased their stiffness as measured by the atomic force microscope.[26] The methods of producing the precise surface morphologies vary depending on the size of the pattern and the material on which the pattern is being replicated. The smallest patterns are produced with direct write laser lithography[28] and AFM lithography while those on the 2-10 um scale are produced with UV photolithography, followed by reactive ion etching to control the slope of the walls.[2, 29-32] The most common method used in these experiments for producing features is to first lithographically produce the pattern on a silicon wafer, then replicate that pattern by embossing or spin casting onto the substrate.[33] The grooves and ridges formed on these substrates have shown significant control over growth directions of cells. Current discussion focuses on the mechanisms behind the alignment of these cells to the surface topography. Von Recum and van Kooten question whether or not the actual geometries of the features are the defining factor, or the fact that there is a change in surface free energy due to edges and disruptions in the planar surface.[34] den Braber et al. concluded that parameters such as surface free energy and wettability influence fibroblast growth and proliferation on microtextured surfaces,

PAGE 19

7 but not the shape or orientation of cells in comparison to the texture.[11] By SEM, they conclude qualitatively that there is more alignment on the 2 m wide features, than on the 5 m and 10 m substrates. Other studies in this review seem to disagree with the their conclusion that fibroblasts do not align to wider features, but the fact that their groove depth was less than 1 m seems to be a limiting factor. Walboomers et al. examined fibroblasts[35] and rat bone marrow (RBM) cells[18, 36] on polystyrene (PS) and PLA textured radio-frequency glow discharge (RFGD) plasma treated surfaces with ridges and grooves with dimensions varying from 1 m to 10 m wide and depths of 0.5-1.5 m. These studies, along with another study they published demonstrated the importance of the ridge depth in that at deeper depths (up to 5.4 m), the cells were more aligned, but not as many cells grew on the surface, even with the increase in surface area.[35] They also showed similar results in examining alignment of intracellular and extracellular proteins, but found that the addition of the ridges and grooves did not alter proliferation of cells on the surface at all.[12, 37-39] Their group also took their textured samples from in vitro use to in vivo by implanting RFGD treated disks of textured and untextured silicones subcutaneously in rabbits and guinea pigs, and PS disks in goats. Their results were mainly inconclusive, but they noticed with the silicone substrates an increase vascularization of the capsules surrounding the textured surfaces compared to the untextured surfaces.[40-42] Typically, the more wettable the surface is, the more cell proliferation occurs. A study by Walboomers and Jansen et al. using rat dermal fibroblasts (RDF) on PS, PLA, silicone, and titanium coated PS substrates also show that the microtextures influence cell guidance, while surface chemistry influences morphology.[15] This study is of particular

PAGE 20

8 interest, in that they examined the effect of different substrates with the same features. They compared the wettability of RFGD plasma treated samples and the elastic modulus to the substrates ability to influence contact guidance. The features were the same dimensions as the previous studies, with the depth only 0.5 m. The moduli varied from 894 MPa for PS to 0.39 MPa for silicone. Elastic modulus of 5 x 15 x 0.2 mm bars was measured using an Instron mechanical testing machine with a crosshead speed of 0.5 mm/s. A proliferation study showed no statistical difference in the number of cells attached to each surface, although there was a significant increase once each surface was RFGD treated. Contact angles using only water gave their measure of wettability, but their relatively high value for silicone (33) after plasma treatment implies that the surface had rearranged or the plasma treatment was incomplete. This phenomenon will be discussed in more detail later. Their overall data were inconclusive, with their conclusions focusing more on the production of the patterns and the fact that different materials still induced contact guidance. A more recent study by this group addresses the depth of groove issue, by examining epithelial tissue and cell migration across and along PS microgrooves.[43] Briefly, 6 mm punches of bovine eye endothelium were place on the microgrooved surfaces and cultured for 6-9 days. They studied ridges and grooves at widths of 1, 2, 5, and 10 m and depths of 1 and 5 m. This interesting study demonstrated the importance of groove depth, in that they concluded the width variations not to be as important as depth, but more importantly, they concluded that the microgrooves have the capability to direct tissue growth. By placing intact epithelial tissue on a patterned substrate, they found that on 5 m deep ridges, the tissue was constrained to grow mainly

PAGE 21

9 in the direction of the ridges and grooves, and did not cross over perpendicular to the features very significantly. Examples of explant growth and the distances measured are shown in Figure 2.1. f Figure 2.1 Light micrographs and measurement protocols of 6 mm diameter punches of corneal epithelial tissue explants (f) over (A & C) smooth PS surface and (B & D) a surface with 1 m deep microgrooves separated by 1 m[43] There was more growth perpendicular to the grooves with the 1 m deep features compared to the 5 m deep grooves, but the growth was still directed mainly along the feature direction. While the direction of the growth was more polarized with the 5 m deep features, there was more tissue migration area for the shallower grooves. Similar results were found for cultured separated epithelial cells plated from a suspension. The study of defined patterns of topography stemmed from many observations that the random roughness of a biomaterial surface influenced cell and tissue response. The phenomena of rugophilia and rugophobia, defined as cells loving or hating rough surfaces was initially pointed out by Rich and Harris.[44] The luminal surface of conventional Dacron vascular grafts can be considered textured in a random roughness

PAGE 22

10 pattern. Studies in sheep have shown a distinct difference in the amount of cellular deposition on non-textured and textured polyurethane vascular surfaces after in vitro and ex vivo study.[45, 46] Using excimer laser micromachining, textured surfaces consisting of fibers of 25, 50, and 100 m in length sticking up from a smooth base plane were examined. When implanted in ovine carotid arteries thrombus formed more quickly on the textured surfaces compared to the non-textured surfaces, leading the authors to conclude that the textured surface acted as a promoter of a stabilized thrombus base. While this may accelerate the formation of a stable pseudointima, the thickness and composition of the thrombus was not controlled. Osteoblasts have also been shown to react more favorably to roughened titanium surfaces compared to smooth surfaces,[47] while surfaces with regularly spaced nanometrically sized pillars reduce tendon sheath (epitenon) cell adhesion.[21] This phenomenon is not seen for larger pillar sizes, in that astroglial cells showed preferential adhesion to pillars and wells on the scale of 0.5-2.0 m in width and 1.0 m in height.[48] In fact, 2 m pillars and holes 4 m deep showed changes not only in cell adhesion, but cell motility.[49] Neutrophils migrated much faster on holes than on smooth surfaces, while pillars slowed the process down. A group out of Harvard headed by Vacanti has recently moved into using the effects of contact guidance and directed cell growth for future clinical applications. They used micromachining technology to form many branching networks resembling capillary beds. Hepatocytes and endothelial cells were patterned and lifted as 2D sheets for the purpose of forming 3D tissue constructs.[50] While this technique is far from perfected, it

PAGE 23

11 is an interesting leap from the studying of contact phenomenon towards a more clinical application. The field of contact guidance using topographical cues is quickly becoming more noticeable in clinical fields. Several excellent reviews have been authored and are a good starting point for an overview on the topic.[1-3, 51] However, topographical features are not the only tools for directing cell growth, as the chemistry of the substrates is becoming more popular in terms of directed cell growth and adhesion. As more studies are performed and a better understanding of the issues involved is achieved, clinicians and researchers are discovering the importance of the texture of medical implants as well as the chemical moieties on the surfaces. Surface Chemistry Surface chemistry plays a large role in the field of contact guidance for controlling the results of cell proliferation. Carter originally demonstrated that cells exhibit a preference to hydrophilic areas of patterned cellulose acetate and palladium metal surfaces.[52, 53] The cells migrated towards the metal and, in a separate study, along a gradient of metal concentration densities towards the more dense and thereby more hydrophilic areas. These studies were later duplicated and confirmed by Harris.[54] Currently, surface treatments are typically deposited in regular patterns using the same photolithographic techniques as with the topographical substrates. Self-assembled monolayers[55] (SAMs) and areas of enhanced adsorption of proteins due to surface energy seem to be the most popular. A study by Britland, et al. deposited a pattern of alternating hydrophobic dimethyltrichlorosilane groups with aminosilane groups on glass slides. No residual topography resulted and BHK cells showed a definite preference for the aminosilane sections, as the cells were crowded and aligned along the border rather

PAGE 24

12 than cross to the hydrophobic groups.[56] A similar study was performed by Healys group using human bone-derived cells (HBDC) and showed the same results.[57] Another study by Britland et al., examined the effect of topography in combination with surface patterning. This paper examined quite a few of the relevant topics with a few simple experiments, although their choice of cells (BHK) does not give as much useful information for the future. They superimposed tracks of aminosilanes orthogonal to the ridges and grooves and showed that for shallow grooves (0.1 m to 0.5 m), the cells aligned almost exclusively with the chemical patterning, but as the grooves go deeper up to 6 m and closer together (5 m), alignment to both the texture and the chemistry was seen.[58] By comparing these results with the studies on topographical patterning and chemical patterning, it is apparent that there are critical levels for each system, in which topography and chemistry contribute differently to contact guidance. The use of silicones in the micropatterning of surfaces has also become quite popular.[59-63] Essentially, textured polydimethylsiloxane (PDMS) substrates are formed using the microfabrication techniques mentioned before. These substrates are then used as stamps or stencils to either directly apply surface treatments such as proteins to another substrate in an organized fashion, or to act as a mask allowing for microfluidics within channels in the silicone to pattern the surface.[59] Many times the silicone itself is the substrate that is used for cell studies. Whitesides group has demonstrated that by selectively adsorbing adhesive molecules in the form of fibronectin to the bottom of pits or wells in the surface, and by keeping a non-adhesive protein like albumin on the surface above the wells, endothelial cells will adhere only to the bottom of the wells where the fibronectin is adsorbed.[64] They have also shown that through the use of microcontact

PAGE 25

13 printing with PDMS stamps, they can pattern SAMs of alkanethiolates on gold to manufacture substrates with controlled islands of extracellular matrix (ECM). By restricting the size of ECM endothelial cells had to attach to, they controlled the size of the cell, which also resulted in altered metabolism.[65, 66] As the size of the area of attachment decreased, the cells changed from growth to apoptosis, or cell death. This technique is useful in its ability to isolate single cells on the same substrate for microarray examination of cell types. Whether the application of a material is for a vascular graft, dialysis machine, blood oxygenator, bioreactor vessel, dental material, or the surface of a ships hull, the first step in any biological response to a surface is the adsorption of proteins.[67-69] One cause of this adsorption is due to the highly varied structure of a protein in solution, both due to its conformation and primary structure of amino acids. Another cause is due to surface rearrangement of hydrophobic and hydrophilic areas of the polymer chain. In this fashion the surface properties and specifically surface energy play a very important role in protein adsorption.[70] An adsorbed layer of serum proteins after exposure to blood is not in the same concentration as the bulk liquid. Rather each protein has a different response to the material in contact, as illustrated by Table 2.1. The composition of the adsorbed protein layer changes over time, as early adsorbing proteins are displaced by others, exhibiting the Vroman effect.[71-73] Cells that adhere to a surface are able to deposit their own proteins, but only if the proteins are able to displace those already adhered. Hydrophobic materials, in the presence of high serum concentrations make this displacement very difficult.[34]

PAGE 26

14 Table 2.1 Enrichment of proteins adsorbed on polyethylene exposed to blood plasma[68] Protein Enrichmenta Fibrinogen 1.3 -globulin 0.53 Albumin 0.88 Hemoglobin 79 aEnrichment was calculated as the ratio of the surface fraction of the protein compared to the bulk fraction. Endothelial Cells Endothelial cells make up one of the most important tissues of the body, known as endothelium, which is the interior lining of all blood vessels. The three major types of blood vessels are arteries, veins, and capillaries. Arteries and veins have a complex structure, made up of three main layers as seen in Figure 2.2. The outermost layer is known as the tunica adventitia, and is composed of loosely woven collagen fibers that protect and anchor the blood vessel. The middle layer, or tunica media, is mostly smooth muscle cells and elastin. This portion of the vessel is elastic and plays a major role in regulating blood flow by relaxing and constricting. The innermost layer is known as the tunica intima, and contains the endothelium on a subendothelial layer of loose connective tissue to act as a basement membrane.[74] Cardiovascular endothelial cells (ECs) are diverse in their size and shape, since they are needed not only to act as a blood barrier in the aorta but also for nutrient exchange through tiny capillaries. For many years, endothelium was classified as an inert tissue, strictly a barrier to keep blood in its vessels. It is now known that the endothelium is an active tissue in homeostasis, producing factors such as endothelin, PDGF (prostaglandin-derived growth factor) and nitric oxide.[74] Endothelial cells are simple

PAGE 27

15 squamous epithelial cells that form a smooth, confluent monolayer characterized by a cobblestone pattern, as seen in Figure 2.3. Figure 2.2 Cutaway view of an artery showing the three main layers. From Seeley et al.[75] As they achieve confluence, the cells are contact inhibited and alter their metabolic behavior, going into a period of stable growth and turnover, known as quiescence. In the adult, the average endothelial cell only divides approximately twice in a lifetime.[76] However, the endothelial cell can rapidly proliferate upon damage. Disorders of the endothelium result in many pathologies, including atherosclerosis and cancer. It is generally accepted that a vascular graft or other blood contacting implant material will have less of an immune response if it is covered by an intact layer of endothelial cells. Endothelial cells effectively provide local delivery of endogenous endothelial secretory products to maintain prosthetic integrity after surgical

PAGE 28

16 implantation.[77] When exposed to a topography that disrupts their ability to achieve confluence, the cells typically respond by moving on the surface and extending philipods out in multiple directions, searching for other cells and more favorable conditions. Figure 2.3 Light microscopy image of confluent porcine vascular endothelial cells (PVECs) grown on tissue culture polystyrene (Image taken by L. Zhao and W. Wilkerson) Vascular Grafts In 1999, 529,544 people died from ischemic heart disease[78] and many undergo a procedure known as coronary artery bypass. This is necessitated by a blockage of the coronary arteries caused by fatty plaque accumulation or thickening of the artery wall. Coronary artery bypass graft (CABG) surgery is a procedure that supplies blood flow to the other side of the blockage by attaching a small diameter graft to the blood vessel. These grafts can take the form of native materials such as arteries and veins, or artificial polymers. The success and failure of these materials is crucial to the survival of the patient, and as of now no adequate small diameter vascular graft is available. The typical treatment in a CABG procedure is to use the saphenous vein from the leg and the internal mammary artery as the graft materials. These are superior to non

PAGE 29

17 natural materials because they do not have an immune response in the patient and are more similar to the native tissue. These grafts have patency rates of over 70% after 5 years, while the survival rate without the surgery is nearly zero.[79] In cases where the autograph material is not available or unusable, then the only option is to implant a polymeric vascular graft. Large diameter vascular grafts used in aortic repairs are made from a woven polyethylene terephthalate (PET) mesh or expanded polytetrafluoroethylene (e-PTFE) and are very successful. PET is a thrombogenic material, in that it causes a cascade response that clots the blood at the surface of the graft. This response is actually favorable and encouraged, in that it seals the porous graft and since the aorta is such a large diameter blood vessel, the flow is not significantly diminished. To better control this process, the grafts are typically coated with albumin or collagen, essentially pre-clotting the graft.[80] Small diameter artificial vascular grafts (< 6 mm) cause more of a problem in that the patency rate is less than 50% after 3 years.[79] Typically, the graft fails either due to thrombosis initially, or a buildup in the intimal layer of the blood vessel, essentially occluding the vessel with tissue. The thrombosis is due to the activation of the clotting cascade when the graft is exposed to the blood, activating platelets and the absorption of fibrinogen. Later in the life of the graft, the main threat to its success is occlusion by anastomotic intimal hyperplasia. The actual mechanisms of this are debated and are being examined, but a number of factors seem to contribute. First, there is a compliance mismatch between the native tissue and the graft. The artery has a very specific compliance to pressure waves during pulsatile flow that the graft interrupts. A mismatch causes stresses at the suture points of the anastomosis, which is typically where the

PAGE 30

18 intimal hyperplasia occurs. This is the main cause of failure in autologous saphenous vein (ASV) grafts, due to the fact that arteries have a much more muscular vascular wall compared to a vein.[81] Probably the most significant cause for the hyperplasia as well as thrombosis is the lack of a stable lining of endothelium, as in a normal vasculature. It typically only grows a short distance from the anastamoses, leaving exposed areas of pseudointima, consisting of fibrous materials and cells, including fibroblasts and smooth muscle cells. While current PET and e-PTFE grafts are inadequate in this regard, new materials are being investigated to improve cellular response to the tissues.[82-86] An ideal graft material would exhibit thromboresistant qualities, have similar mechanical properties as the native tissue, have ease of use by the surgical team, and be able to form a stable endothelial layer. Currently the materials used are thrombogenic, have varying yet acceptable mechanical properties, are easy to use by surgeons, but do not form a stable endothelial layer. Studies are in progress that have seen patency rates over two year periods double for artificial small diameter grafts due to endothelial seeding.[84] Currently, methods to improve endothelial cell seeding involve the adsorption of proteins on the surface or the use of fibrin glue.[22] Past results have shown that the substrate must tightly adhere the cells or they can be removed when subjected to shear flow. For surfaces without covalent linkages between the proteins and the surface or strong mechanical interlocking, it is difficult to form a stable intima that can withstand biological stresses. Many of these modified surfaces are attractive to proliferation of a fibrous pseudointima of fibroblasts and smooth muscle cells. With the proper surface chemistry and topography, endothelial

PAGE 31

19 cell attachment and spreading to an intact, confluent layer that can withstand biological stresses should be possible. Cell Adhesion Focal adhesions are typically the main area of adhesion of a cell to a substrate. They are the link between ECM proteins such as fibronectin, vitronectin, and collagen adsorbed to the surface, and the cell receptors, known as integrins, which bind to the ECM through the well-documented Arg-Gly-Asp (RGD) tripeptide sequence. These receptors link the focal adhesions with the cells cytoskeleton, thereby altering the cells shape and locomotion. Focal adhesions are typically elongated, and oriented in the direction of the stress fibers and the main axis of elongation.[87] A recent study by van Kooten and von Recum have shown that fibroblasts and human umbilical vein endothelial cells (HUVECs) formed focal adhesions within the first 24 hr of adhesion on textured silicone surfaces.[14] Recently work has been produced to examine binding polypeptide sequences to the surfaces of polymers.[88-93] The chemistry behind the grafting of these molecules on the surface can be accomplished by using plasma-induced graftcopolymerization. By exposing the surface to an argon plasma and then air, hydroperoxide groups are formed on the surface that can initiate radical polymerization.[92] Much work has been examined using the RGD sequence as the active adhesion area in proteins such as fibronectin. Hubbell examined the specific adhesion of endothelial cells to the Arg-Glu-Asp-Val (REDV) tetrapeptide through the 4-1 integrin and showed that while endothelial cells attach and spread on this ligand, fibroblasts, vascular smooth muscle cells, and platelets did not.[94, 95]

PAGE 32

20 Recent work has refocused on the traction forces that cells impart on the substrate on which they adhere. By using low moduli micropatterned substrates,[87] and unpatterned wrinkling substrates,[96-100] studies have begun to quantify these forces for specific systems. Harris et al. first introduced the study of cell locomotion and adhesion with wrinkling substrates by growing a wide variety of cells on a thin, heat-crosslinked film of silicone floating on a silicone fluid. As the cells grow, they pull the film underneath in circumferential folds that are smoothed out when the cells are trypsinized from the surface.[96] Pelham and Wang were able to grow fibroblasts on polyacrylamide substrates with very low moduli but varying 12-fold. They found that fibroblasts spread less and had increased motility and lamellipodial activity on more flexible substrates, while more rigid substrates promoted stable elongated focal adhesions.[98] Surprisingly, this is one of the few studies examining the effect of different moduli of similar materials on cell growth, and none to date have been reported that compare the effect of modulus to the contact guidance phenomenon. Balaban et al. used fibroblasts stained to expose the focal adhesions of fibroblasts grown on low modulus (E~15 kPa) silicone elastomers with either fluorescent patterns embedded into the surface, or features similar to pits and pillars as part of the surface. The textures on their surfaces were 0.3 m deep, because they wanted to minimize the contact guidance phenomenon. With features deeper than that, polarization and directed growth occurred along the features. In relating the displacement of the features with the locations of the focal adhesions, they were able to extract force measurements exerted by the focal points.[87]

PAGE 33

21 It has been well documented that cell behavior, in the form of growth, movement, and metabolism, is closely linked to the shape of the cell.[101, 102] The effects of adding forces and changing the tension on a cell have shown significant changes in the biochemistry of the cell.[103] Several cell types also align along lines of principle strain with external loading.[13, 104] Kato et al. showed that endothelial cells, when patterned on thin strips of adhesive regions that caused endothelial cells to become elongated, exhibited a decrease in mRNA expression for vascular adhesion molecule-1 (VCAM-1) and a higher mRNA expression for intracellular adhesion molecule-1 (ICAM-1).[102] Topography has been shown to alter cell shape, fibronectin mRNA level and stability, and the secretion of ECM by human fibroblasts.[101] In many cases, the shape of the cell is widely spread out, especially when lamellipods extend in multiple directions. This makes quantification of cell response to texture or chemistry difficult by simply looking at the outline of a cell. One approach has been to examine the shape of the nucleus of a cell to elucidate the prevailing cytoskeleton arrangement within the cell.[10] Ingbers group as well as others have demonstrated that the nucleus shape is indeed directly hard-wired such that changes in surface adhesion can affect the shape and orientation of the nucleus.[105] They examined bovine endothelial cells after attachment to a substrate, and pulled on them with pipets coated with adhesion molecules and found that as the cytoplasm stretches, the shape and orientation of the ECs change from round to elongated along the stress. As medical technology and surgical procedures improves, the need for effective biomaterials becomes greater. The ability of a non-native material to mimic the properties and functions of the tissue it is replacing is crucial to the success of the

PAGE 34

22 prosthesis or device. Similarly, a biomaterial used in an in vivo or ex vivo application should minimize unwanted biological interactions. Silicones Silicones are widely popular materials and have many commercial uses today. They are unique in many of the polymers used in biomedical devices in that they possess a silicone-oxygen backbone instead of a carbon backbone. Their chemical and physical properties allow for their use in a variety of applications. The simplest silicones are polydimethylsiloxanes, a linear organosilicon compound whose structure can be seen in Figure 2.4. SiCH3H3CCH3OSiOCH3CH3SiCH3CH3CH3n Figure 2.4 Polydimethylsiloxane, trimethylsiloxy terminated PDMS oils, when not cross-linked, are used in fields such as cosmetics, food-processing, and pharmaceutical preparations. Their lubricity and low surface tension make them excellent additives for anti-foaming.[106] The methyl groups in the backbone and end caps of the PDMS molecule can be replaced by both functional and non-functional molecules, including hydrogen, phenyl, and vinyl groups. The significance of these substitutions is in their changes to the chemical and physical property of the polymer, as well as the curing and processing capabilities. Cross-linked PDMS forms a silicone elastomer that has excellent elongation properties (20-700%) and moderate breaking strengths (~ 1000 psi). They have been

PAGE 35

23 cured in high-temperature vulcanizing (HTV) systems of methyl or vinyl groups using peroxides. Room temperature vulcanizing (RTV) systems are cured by condensing silanols with a moisture-sensitive silane cross-linking agent, or are condensed with a metal salt catalyst. Vinyl addition systems using platinum catalysts were initially used in low-temperature vulcanizing (LTV) systems, but has been extended to RTVs and HTVs.[106] The silicone used in this study is Dow-Cornings Silastic T-2, a filled RTV elastomer with vinyl terminated end caps that is addition polymerized with a platinum catalyst. Use as a Biomaterial Silicone elastomers are valuable polymers in the biomedical field. The use of silicone materials in vivo has become a heavily debated topic in the past decade, with the proliferation of procedures and studies on devices such as breast implants. These devices, made of a PDMS gel, allowed for the leaching or gel bleed of low molecular weight oils from the device into the surrounding environment and then through the body. The silicone biomaterials were widely criticized for causing a large range of complications and diseases, and as a result, were removed from the market as a medical device. A recent risk assessment on the effects of silicone gel-filled breast implants concluded that the adverse effect of exposure to these prostheses was minimal, and current stringent regulation should be discontinued.[107] One of the main problems with silicone implants, as with all implant materials, is the formation of fibrous capsules.[69] The capsules can cause discomfort as well as contraction on the device that can ultimately lead to failure. Silicones permeability to oxygen allows for their use in contact lenses and membrane oxygenators, and their flexibility and stability have seen uses in a wide variety

PAGE 36

24 of implant materials.[92, 106, 108, 109] Silicone orthopaedic devices have included finger joints and temporomandibular joints (TMJ), which were not sucessful.[110] They have also been used in the vascular field as components to heart assist devices and incorporated into vascular grafts.[106, 111] While the mechanical properties of silicones are ideal for vascular grafts, in that their compliance and modulus more closely match the native tissue than PTFE or PET, their lack of toughness prevents them from being used exclusively. Also, a major concern with PDMS is its affinity for lipids, which causes it to become more brittle in vivo over time.[111] The important issue to keep in mind for silicone systems as well as other biomaterial applications, is that the bioactivity of the system will typically dictate what use the material has in vivo. Silicone surfaces are mainly non-thrombogenic and biologically inert, which makes them an interesting material to prevent unwanted interactions, but this lack of bioactivity also precludes it from becoming incorporated in surrounding tissues without surface or bulk modifications. Another field of particular interest to our group is the use of silicones to prevent biofouling from marine organisms. Biofouling is an example of a problem concerning cellular materials accumulating on surfaces such as the hulls of ships and water treatment facilities. The marine spore Enteromorpha is the most common macroalga that fouls ships and submarines. Reproduction is mainly through motile spores that swim until a suitable surface on which to settle and adhere is located.[112] Adhesion involves secretion of a glycoprotein adhesive that anchors the spore to the surface.[113] Cues for settlement include phototaxis, chemotaxis and thigmotaxis. Previous anti-fouling coatings included biocides that did significant damage to marine life in harbors. Current research focuses on

PAGE 37

25 preventing adhesion forces able to withstand the shear forces during motion.[114, 115] By using low surface energy silicones, coatings are foul releasing rather than antifouling. Singer and others in this field have realized the importance of the mechanical properties in determining the ability of a cell to adhere and remain adhered to a surface.[5] He simulated barnacle pull-off tests by epoxying a stud onto a silicone surface, and determined the critical force to pull it off with respect to the materials thickness and elastic modulus. The result was for lower modulus materials (E* = 3 MPa) and thicker coatings (up to 4 mm), the force needed to detach was less than for higher modulus (23 MPa) and thinner (0.08 mm) coatings. Gatenholms group examined the use of microtextured surfaces in the marine biofouling environment by imparting 50-100 m deep and wide features through use of a wire mesh as a mold. They found that barnacle adhesion on the macro scale decreased on textured surfaces compared to smooth. Many of these same principles of concern in biofouling can be used in biomedical applications to improve the biocompatibility of polymeric surfaces in the body. Surface Energy Due to its hydrophobic nature, silicone experiences rather high amounts of protein adsorption and poor spreading of cells.[92] Thus, to improve cell adhesion to a silicone surface, the chemistry of that surface is usually modified. Unmodified silicone is hydrophobic with advancing contact angles around 110-120. The difference between the advancing and receding contact angle is known as hysteresis, and gives some understanding of a surfaces ability to remodel itself as well as the surface roughness. As the silicone is exposed to water, hydrophilic areas of the siloxane backbone migrate to the surface, masking the hydrophobic methyl groups.[70] This provides a more hydrophilic surface and a lower contact angle while receding. This rearrangement of the silicone

PAGE 38

26 backbone is an interesting phenomenon, and is especially important in PDMS systems due to its low Tg (-123C). Most of the surface rearrangement in plasma treated high surface energy PDMS samples is due to the migration of low surface energy, low MW PDMS oligomers from the bulk to the surface.[116, 117] Ostuni and Whitesides used the low surface energy of silicones to selectively pattern the surfaces of textured elastomer substrates with protein solutions. Since the contact angles of a fibronectin/physiological buffered saline (PBS) solution and a bovine serum albumin (BSA)/PBS solution on their silicones were over 100, they could trap air inside of wells by carefully placing drops of the solutions on the surfaces. This allowed for the flat surfaces above the well to have one of the proteins adsorbed, and by rapidly changing the solution and adding a vacuum to pull out the air bubbles, the bottom of the wells were patterned with another.[64] Schmidt and von Recum characterized Dow Cornings Medical Grade Silastic (MDX4-4210) silicone after texturing the surface with various pillars and wells. They determined the surface energy of their materials by using the stationary drop method[118] using diagnostic liquids that were not identified. They calculated Zisman plots that graph the cosine of the contact angle versus the surface tension of the diagnostic liquid. All of their samples, both textured and untextured had critical surface tensions of 20.5-21.5 dynes/cm.[119] They showed that the surface energy per unit of surface area decreased for higher densities of features, and concluded that the surface energy either was not greatly affected by their topography, or that contact angle methods are not sensitive enough to detect the difference.

PAGE 39

27 Mechanical Properties The importance of mechanical properties has been discussed in the uses for silicone materials and in the need for proper compliance for vascular graft applications. One area of interest is designing the mechanical properties to control the response of biofilms and tissues. Currently, the focus of our interest is in modifying the modulus of the elastomer without greatly affecting the surface chemistry or energy of the sample. Youngs modulus, E, may be written as E where and represent the tensile stress and strain respectively.[120] Essentially, Youngs modulus is a measure of the stiffness and compliance that is referenced in much of the biology-based research on these materials. The term compliance can be defined as the elongational compliance, J, which in regions far from thermal transitions is the inverse of Youngs modulus. E J1 This should not be confused with the compliance of vascular graft materials, which is a measure of the dynamic circumferential elastic properties of a vessel and is calculated using systolic and diastolic blood pressure and the vessel diameter by diastolicsystolicdiastoloicdiastolicsystolicPPDDDC where D is the vessel diameter, P is the pressure, and C is the compliance.[121]

PAGE 40

28 Typical glassy polymers such as polystyrene at room temperature have E values around 3 GPa, while soft rubbers are closer to 2 MPa.[120] PDMS based elastomers have a somewhat wide range of moduli, from 0.1 MPa to 10 MPa.[111] The change is due to the curing characteristics and the size of the oligomers, which affect the cross-link density, molecular weight between cross-links, and physical entanglements. Currently, we are able to vary the modulus of our base Silastic T-2 elastomer over a range of moduli from ~0.1 MPa to ~3 MPa with the addition of functionalized and unfunctionalized PDMS oils. These oils can take the form of linear PDMS or larger bulky oils with side chains. By varying the molecular weight and functionalization of these additives, the variables mentioned above can be altered, without significantly changing the surface chemistry of the samples. The majority of the development and characterization of this system was performed by other members of the group, and so will not be discussed in great detail in this thesis. However, since the materials with differing moduli are of critical importance to the main purpose of this work, a description of the specific oils and testing methods will be discussed in the characterization section. Research in our group has focused on the effect of mechanical properties on biofilm formation, whether the film is in the form of algal spores or endothelial cells. Kendall modeled adhesion behavior on elastomers by deriving formulae for the removal of rigid solids off of elastomers with varying thickness.[122] The critical pull-off force, Pc, required to remove a rigid cylinder with a radius of a from a film of thickness t is given by 2122tKwaPac

PAGE 41

29 where wa is Dupres work of adhesion and K is the bulk modulus, which is related to the elastic modulus, E, by Poissons ratio, by the following: 213EK At greater thickness where t >> a, the relationship between the pull-off force and the elastic modulus is given by 212218aEwaPaC The implication of this equation is that stiffer materials improve strength of adhesion.[5] One of the assumptions to this equation is that the attached surface is rigid, while for a cell, this is not the case. As discussed earlier with the low modulus wrinkling elastomers, a cell has the ability to create its own forces and change them depending on the substrate on which it is adhering. If a focal adhesion is modeled as a rigid body, then the equation can hold validity. If we imagine a cell applying traction forces on the substrate it is adhering to, then it is possible that a cell in equilibrium might exert forces just under the critical force in equal directions. Chicurel et al. concluded in their review of focal adhesion literature that a cell will continually contract on a substrate until the forces come into balance, much like a bow and a bowstring.[103] By that reasoning, for higher modulus materials the critical force is greater and thereby the traction forces applied. Since the ridges and grooves select the direction for a cell to align by directing its cytoskeleton, these increased forces would travel along the ridge and result in increasing the elongation of the cell, and the effect of contact guidance.

PAGE 42

30 By examining textured surfaces with varying moduli while keeping surface energy variations to a minimum, this project attempts to minimize some of the unknown factors and study the effects of modulus on contact guidance. Before this can be completed, a thorough investigation of the surface topography and chemistry must be performed, and the results are discussed in the next chapter.

PAGE 43

CHAPTER 3 CHARACTERIZATION OF SILICONE ELASTOMER SUBSTRATES Introduction The first step in this project was to fully characterize the silicone elastomers and textured substrates. Four separate formulations of silicone elastomer with functionalized additives and four with non-functional additives were used to vary the elastic modulus. The surface energy of each formulation was measured by contact angles of solvents with varying surface tensions. Dynamic contact angles in water were taken of the unmodified silicone elastomer with and without texture. Samples were treated with both radiofrequency glow discharge (RFGD) plasma and adsorbed fibronectin. Textured surfaces were provided at two depths of features, and the fidelity of the surface pattern was characterized by non-contact optical profilometry. Materials The base elastomer system, referred to in this thesis as unmodified, is the Silastic T-2 Silicone Moldmaking Rubber produced by Dow Corning. The resin is a dimethylvinyl-terminated polydimethylsiloxane (PDMS) composite mixture composed of PDMS and trimethylated silica for mechanical stability. It is a translucent resin that cures via addition polymerization with a platinum catalyst when added in a 100:10 ratio with the Silastic T-2 curing agent. The typical properties of the base resin and elastomer system are given in Table 3.1. Resin was provided in 45 lb. containers and curing agent 31

PAGE 44

32 provided in 4.5 lb. containers. To simplify the production, portions of the resin and curing agent were transferred into opaque HDPE bottles. Table 3.1 Typical properties of Silastic T-2 Silicone Moldmaking Rubber (from Dow Corning product information sheet) Test Unit Result As Supplied Base Color Translucent Viscosity Centipoise or mPas 50,000 Curing Agent Color Transparent Viscosity Centipoise or mPas 550 As Mixed 100 Parts Base to 10 Parts Curing Agent by Weight Viscosity Centipoise or mPas 55,000 Specific Gravity 1.12 As Cured 24 Hours at 25C Durometer Hardness, Shore A Points 42 Tensile Strength Psi 800 Elongation Percent 300 Tear Strength, Die B ppi 120 Linear Shrink Percent < 0.1 Two vinyl terminated linear PDMS oils and one bulky vinyl terminated oil were added to the resin to change the mechanical properties of the elastomer. A variety of non-functional PDMS oils of varying molecular weights were also examined for surface energy analysis. The molecular formula, viscosity, and molecular weight of each additive are given in Table 3.2. All were obtained from Gelest, Inc. and stored at room temperature. Five solvents were used for contact angle analysis: 1-Propanol (Aldrich, 99+% spectrophotometric grade), acetonitrile (Aldrich, 99.93+% HPLC grade), N,N-dimethylformamide (Aldrich, 99.9+% HPLC grade), diiodomethane (Aldrich, 99%), and

PAGE 45

33 ultra-high purity DI water. Bovine plasma fibronectin was received from Sigma (cat # F-4759, 2 mg) as a lyophilized powder. Table 3.2 PDMS additives to silicone elastomer system Name Molecular Structure Viscosity (cSt) MW (g/mol) Catalog Number Vinyl terminated PDMS H2CCHSiOCH3CH3SiOCH3CH3SiCHCH3CH3CH2n 2-3 1000 550 28,000 DMS-V03 DMS-V31 Vinyltris(trimethyl-siloxy)silane N/A 322.70 SIV-9300 Trimethylsiloxy terminated PDMS SiCH3H3CCH3OSiOCH3CH3SiCH3CH3CH3n 50 5000 3,780 49,350 DMS-T15 DMS-T35 Tris(trimethylsiloxy) silane N/A 296.66 SIT8721.0 Methods Elastomer Preparation Unmodified elastomer substrates were produced according to manufacturer instructions. In a tri-cornered polypropylene beaker, resin and curing agent were added in a 100:10 resin to curing agent ratio by weight. The two components were mixed with a metal spatula until well incorporated. During the mixing process, many air bubbles were trapped in the resin, and so the mixture was degassed under vacuum for 10-15

PAGE 46

34 minutes, periodically breaking the vacuum to rupture the bubbles formed. The uncured elastomer remains workable for approximately 1 hour. For modified samples, oils were added to the resin before the addition of curing agent at the appropriate concentrations and mixed together. Curing agent was then added and the procedure was followed as above. For non-functionalized oils, the resin to curing agent ratio remained 10:1. For vinyl-terminated oils, the curing agent percentage was increased to account for the increase in vinyl groups. Unmodified samples were cured at room temperature, while vinyl-terminated samples were cured at 80C for the appropriate amount of time, as seen in Table 3.3. Table 3.3 Curing conditions for silicone elastomer samples Sample Curing Agent Needed Cure Time and Temperature 5% 2-3 cSt vinyl term (5% LMW) 0.16 x mass resin 4.5 hr at 80C 5% 1000 cSt vinyl (5% HMW) 0.1011 x mass resin 2 hr at 80C 15% vinyl tris 0.26 x mass resin 12 hr at 80C Unmodified and nonfunctional oils 0.10 x mass resin 24 hr at RT Modulus Determination Modulus values were obtained by Leslie Wilson and Amy Gibson and the values are used for material selection purposes only. A brief explanation of the method used is included here. Tensile specimens were made using an ASTM D1822-68 type L dogbone die, resulting in a 1-inch gauge length. Each sample was individually measured and the thickness of the samples was ~1 mm. Tensile measurements were made according to ASTM D412-97 on an Instron model 1122 equipped with the TestWorks 3.07 software for analysis. The dogbones were tested via crosshead displacement at a rate of 2

PAGE 47

35 inch/min. Measurements of the modulus were calculated from the linear portion between the stress of 0.2 and 0.5 pounds and recorded. Preparation of Textured Surfaces All textured surfaces were taken from silica wafers with textures etched using microfabrication technology. Wafers were provided by Chuck Seegert at two etch depths, 5 m and 1.5 m. Briefly, silicon wafers were coated with a thin layer of a photosensitive polymer and then exposed to UV light through a photomask imparting various 5 m patterns onto the photoresist as seen in Figure 3.1. ABCD Figure 3.1 Representation of etched patterns on silicon wafers. Each square side is 10,000 m long and is made up of 5 m wide ridges, separated by varying groove depth. Each square is separated into thirds with the groove depth in each third 5 m, 10 m, or 20 m wide. Square A has ridges 60 m long with 40 m in between. Square B has 5 m square pillars. Square C has 10,000 m long continuous ridges and square D has 800 m long ridges with 200 m smooth space in between. After development, the wafer was etched using reactive ion etching to the desired depth. The 5 m deep wafer was patterned here at the University of Florida, and sent to Unaxis, USA for etching via the Bosch process. The 1.5 m wafers were etched in

PAGE 48

36 house. Each wafer was then glued to a metal backing with epoxy to improve mechanical stability. Before casting films for the first time, the wafers were treated with hexamethyldisilazane (HMDS) to minimize adhesion to the wafer. Textured substrates were cast either directly on the wafers or on epoxy master copies. For the direct wafer copies, wafers were placed on a thin sheet of polyethyleneterephthalate (PET) taped to a glass plate and uncured elastomer was poured onto the wafer. Any trapped air bubbles were pierced with a needle. A second glass plate with a PET sheet and spacers was placed on top to form a constant thickness film of 3 mm. After curing, the plates were separated and the film removed from the wafer. For the production of the epoxy master, PDMS copies of the wafers were made with an accelerated cure. The desired texture squares were then cut out and placed texture side down on a PET covered glass plate. Both 1.5 m and 5 m deep patterns were placed on the same plate. After positioning the textured surfaces, uncured elastomer was poured over the backside of the textured PDMS. The result was a PDMS film of constant thickness with individual squares of texture. This film was then used as a mold to cast epoxy (Epon 828 epoxy resin and Jeffamine D-230 hardener) over. The epoxy was cured at 80C overnight and the PDMS film removed. The resulting epoxy master is a direct replica of the wafer texture. Unmodified textured surfaces were cast off of both epoxy and wafers, while modified silicones were cast mainly off of the wafers directly due to problems with adhesion to the epoxy at elevated cure temperatures. Surface Treatment Samples for the cell adhesion portion of this project were either left untreated, treated with a RFGD plasma, or coated with an adsorbed layer of fibronectin.

PAGE 49

37 Radiofrequency glow discharge treatment To improve the wettability of the substrate, surfaces were exposed to an argon RFGD plasma treatment. The plasma system used was a RF Plasma Inc. HFS 401S instrument, set at 50 W. Four samples in a polystyrene petri dish were treated at a time, approximately 5.5 cm from the bottom of the plasma RF coils, and after the samples were in place, the plasma chamber was evacuated for 10 minutes. Pressure in the plasma chamber got as low as 10 mTorr before treatment and typically around 25 mTorr immediately after the plasma was switched on. The argon gas was introduced at a flow rate of 200 sccm and the RFGD treatment was activated for 5 minutes. After treatment, the pressure was equalized in the chamber with ambient air. Fibronectin adsorption Fibronectin adsorption on microtextured elastomer samples was performed via the method of Ostuni and Whitesides.[64] Lyophilized bovine plasma fibronectin was received from Sigma and the contents were dissolved in 2 mL of 0.22 m filtered water at 37C for 45 minutes. The solution was diluted to 50 g/mL in Hanks balanced salt solution (BSS). Elastomer samples cut into 15 mm disks were sterilized with 70% EtOH and rinsed 3X with BSS, then placed in 24-well culture plates and covered with 0.5 mL of the fibronectin solution. To ensure that fibronectin displaced air trapped in textured surfaces, the culture plate was exposed to house vacuum for 1 minute. During this time, small bubbles formed on the surface in the same pattern as the texture. After these bubbles detached, the vacuum was released and the fibronectin solution left on for 1 hour at ambient conditions. Samples were then rinsed three times with BSS.

PAGE 50

38 Surface Energy Untreated elastomers were examined for surface energy differences due to the additives. Sessile drop contact angles were measured for five solvents on each substrate. Five drops each of ultrapure water, methylene iodide (MeI), 1-propanol (1-prop), N,N-dimethylformamide (DMF), and acetonitrile (ACN) were measured for each substrate, and repeated on a second sample. Each drop was 2 L as dispensed from a 25 L pipet. An optical goniometer was used and the left and right contact angles were measured immediately after each drop was placed. Thus, 20 readings per liquid per sample were obtained. A Zisman plot was utilized to determine the surface energy of each substrate. Captive bubble contact angles were taken on untreated, plasma treated, and fibronectin adsorbed unmodified elastomer. The substrates were placed treatment side down in a PMMA jar containing BSS. Bubbles of 2-4 L of air were introduced into the BSS under the sample so they attached to the treated surface. Five bubbles for replicate samples were measured in the same manner as for the sessile drop method. Dynamic Contact Angles Dynamic contact angles were taken on a Cahn dynamic contact analyzer using the Wilhelmy plate technique. Briefly, Wilhelmy plate contact angles are taken by advancing and withdrawing a film into a liquid. The force on the film is measured and correlated to the film displacement. Textured and untextured unmodified elastomer and samples with non-functionalized PDMS oligomers were examined in ultra pure water in clean glass beakers. Samples examined were unmodified, 5% and 20% 50 cSt oil additive, and 5% and 20% 5000 cSt oil additive. Smooth films of measured thickness were cast in between glass plates with spacers. Textured surfaces were cast off polystyrene copies of the 5 m wafer.

PAGE 51

39 Films were cut into rectangles with dimensions approximately 10 mm wide and 30 mm long. The thickness was approximately 3 mm for the untextured films, but varied for the textured surfaces since they were cast off a PS copy with no back plate. Lower viscosity textured samples spread more and had decreased thickness when compared to the unmodified samples. The perimeter of the advancing cross-section was measured for each sample, and three samples were examined for each setting. First, clean mica strips were dipped in the water at a stage speed of 100 m/s. Surface tension of the water was calculated assuming perfect wetting of the mica. Textured surfaces were arranged so that 10 mm of untextured area is inserted into the water first, until 10 mm of textured lines running the width of the sample reached the water interface as demonstrated in Figure 3.2. Figure 3.2 Sample layout for textured substrates examined using DCA Sample depth was set at 25 mm to include all texture and single dips were used for all samples. Advancing and receding contact angles were calculated for the smooth and textured areas using the Cahn DCA4A software package. Optical Profilometry Topographical characterization of both the textured elastomer samples and the wafers directly was performed using optical profilometry. The Wyko NT1000 from Veeko instrumentation uses non-contact interferometry with optical light to map the

PAGE 52

40 surface in 3-D without affecting the surface properties or deforming the substrate. Magnification of the substrate can vary from 2.5X to 100X using the objective and field of view lenses provided. Samples were prepared as described above and needed no additional treatment. Both 1.5 m and 5 m deep samples and wafers were examined at 50X and 5X. The samples were leveled on the stage and focused before the scan was run. Profilometry data and 3-D rendering of the surface was accomplished using WYKO Vision32 version 2.210 software package. Results and Discussion Modulus Results Since the silicone formulations and modulus testing was performed by others in the group, an extensive discussion on the theory and practice will not be included here. The samples were provided as materials that should have similar surface energies but with varying moduli. Figure 3.3 is a graph of the modulus of each material. Note that there are two values for the unmodified elastomer, one which is cured at room temperature, and one cured at 80C. It is important to note that the accelerated cure produced an almost 25% reduction in modulus. For this reason, all samples were cured at the set conditions for that formulation. The striped columns in Figure 3.3 represent the samples with functionalized additives. The 5% LMW and 5% HMW represents a 5% addition of the 2-3 cSt (~550 g/mol MW) and the 1000 cSt (~28,000 g/mol MW) vinyl terminated PDMS oils respectively. The 15% vinyl Tris is the lowest modulus material, and represents a 15% addition of the vinyl tris(trimethylsiloxy)silane molecule. The non-functionalized oils all decreased the modulus and should not have crosslinked into the network due to its methyl endcaps. The 20% 5000 cSt oil additive actually had a

PAGE 53

41 greasy surface, where a film of oil that migrated to the surface was visible to the naked eye. Elastic Modulus of Modified Silicones00.511.522.53Elastic Modulus (MPa) Unmodified 5% 5000 cSt 5% 50 cSt Unmodified 80C 20% 5000 cSt 20% 50 cSt 20% Tris 5% LMW 5% HMW 15% Vinyl Tris Figure 3.3 Elastic modulus of modified silicones as measured by tensile testing. Striped bars represent elastomer with functionalized PDMS additives. (Data obtained by Leslie Wilson and Amy Gibson) Surface Energy of Silicone Elastomers As mentioned before, the surface energy of the elastomer formulations was determined using contact angles. Zisman and his group first introduced an empirical method of treating contact angle data to estimate s, the surface free energy of the solid.[118, 123] The plot of cos vs. l, the surface tension of the liquid, form approximately a straight line with the formula clb 1cos

PAGE 54

42 where c is the critical surface tension, below which is zero and the surface is perfectly wetted. This is considered to be a measure of the surface free energy of the solid. The rationale behind this stems from Youngs equation, represented as coslsls where represents the surface tension (or free energy) and the subscripts s, sl, and l refer to the solid-vapor, solid-liquid, and liquid-vapor interfaces respectively.[123] Essentially this is a force balance between interfaces of a drop on a surface, with the solid-vapor and solid-liquid opposing each other in the plane of the solid. The basis of this theory is that with decreasing l towards s, then the solid-liquid surface tension is minimized and the solid-vapor surface tension will equal the liquid-vapor surface tension or surface free energy. Contact angles were taken using the sessile drop method with 5 different liquids of known surface tension on the surface. Ten readings were taken for each sample and liquid by measuring the angle on both sides of the drop. The liquids used and their corresponding surface tensions can be found in Table 3.4. Table 3.4 Surface tension of liquids used for surface energy determination by contact angle analysis[124] Liquid Surface Tension (mN/m) Water 73.05 Methylene Iodide (MeI) 50.76 N,N-Dimethylformamide (DMF) 37.1 Acetonitrile (ACN) 29.30 1-Propanol 23.78

PAGE 55

43 Contact angles were measured for these liquids on unmodified elastomer, 5% LMW, 5% HMW, 15% vinyl tris, 20% tris, 5% 50 cSt, 20% 50 cSt, 5% 5000 cSt, and 20% 5000 cSt coated slides. For each sample, cos was plotted for each liquid versus the surface tension, and linear regression was performed to determine the surface free energy where 1cos The surface energy and contact angles of each substrate can be found in Table 3.5. A representative example of a Zisman plot from these data is shown in Figure 3.4. Table 3.5 Contact angles and surface free energy of various substrates Average Contact Angles (degrees) c Samples Water MeI DMF ACN 1-Prop (mN/m) Unmodified 109 4 67 4 55 2 47 4 32 2 19 5% 50 cSt 110 3 66 1 56 2 44 3 31 1 20 20% 50 cSt 109 1 64 2 54 2 46 3 25 1 21 5% 5000 cSt 108 2 65 2 55 1 48 2 26 2 20 20% 5000 cSt 103 2 64 2 52 2 48 2 26 2 19 20% Tris 106 2 64 2 56 2 46 2 29 1 19 5% LMW 107 5 65 3 55 2 47 2 29 3 19 5% HMW 112 2 64 2 56 2 46 3 25 3 21 15% vinyl Tris 107 2 66 3 58 3 50 5 33 3 17 As seen by Table 3.5, there is very little difference between all the samples in terms of surface free energy. This is an expected outcome since essentially the samples are all made of silicone elastomer with different PDMS oligomer additives. Since the material is the same, the surface energy should be similar. In addition, to minimize the energy at the interface, low surface tension oligomers have been shown to migrate to the surface.[116, 117] The fact that this does not affect a change in surface energy is supported by the results for the 20% 5000 cSt samples of no appreciable difference in surface energy even with a visible layer of oil on the surface.

PAGE 56

44 Zisman Plot of Unmodified Silicone5 Liquid Systeml = -42.301 cos + 61.235R2 = 0.968201020304050607080-0.4-0.20.00.20.40.60.81.0cos Surface Tension (mN/m) Figure 3.4 Zisman plot of unmodified silicone for calculation of surface free energy Contact Angles of Treated Surfaces The two main surface treatments employed in this study are RFGD plasma treatment and adsorbed fibronectin. To determine the relative wettability of the surfaces, captive bubble contact angles were measured for plasma treated, fibronectin treated, and untreated unmodified samples. Results are reported in Table 3.6. As seen by the low angles on fibronectin and plasma treated surfaces, the hydrophilicity of the substrate has been greatly increased compared to the hydrophobic unmodified PDMS. Table 3.6 Captive bubble contact angles on treated surfaces Contact Angle (degrees) Unmodified 87 4 Fibronectin treatment 15 4 Plasma treatment < 10

PAGE 57

45 Plasma Treatment Issues The fibronectin-coated surfaces change their hydrophilicity by adsorbing a layer of protein on the surface, but the plasma treated samples modify the surface chemistry of the base polymer. The literature implies that two factors seem to be at play here. One is the modification of the surface by creating free radicals that when exposed to air form peroxides on the surface.[92] This is the method used to add adhesion molecules and other chemistry to the surface through graft copolymerization. However, as the plasma treatment increases, an oxidized silica-like layer forms on the surface, especially if the plasma contains oxygen. This is typically a problem when using silicone coatings as high-voltage outdoor insulators. Studies have shown this silica-like layer to be up to 150 nm thick on the surface. This layer eventually cracks and then allows the low MW oligomers to migrate to the surface.[116] This was not considered to be a problem with the plasma treatment of the samples, but in analysis of the surfaces it was noticed that many of the plasma treated elastomers had cracks. These were not seen after removing the samples from the plasma chamber. As the samples were peeled off the dish to be placed in the wells, cracks were formed that were not there before or immediately after treatment. The main concerns with this phenomenon is the influence the cracks would have on the cells compared to the intended topography and the effect that a harder silica-like surface layer would have on the modulus. To examine this affect, a textured sample of 15% vinyl tris silicone was plasma treated under the same conditions as the samples and examined using optical profilometry and atomic force microscopy (AFM), as seen in Figure 3.5, Figure 3.6, and Figure 3.7. The sample was bent to simulate removal and

PAGE 58

46 placement in a culture dish. Cracks on the order of 0.5-1.0 m were observed, and appeared mainly as positive textures pushing up from the surface. Since the transformation of the material from PDMS to a silica-like oxidized layer results in a large decrease in specific volume, tensile stresses are formed in the oxidized layers. These stresses are relaxed by the cracking and result in an elevated texture, as seen and explained by Hillborg, Sandelin, and Gedde.[116] Of concern also is the affect that the plasma treatment has on the shape and size of the features. Profilometry data presented in Figure 3.8 shows a very dramatic change in the height of the ridges, changing from 1.43 m to 0.81 m. It is apparent that the plasma treatment at this level is ablating the surface and seriously altering the features. Figure 3.5 Profilometer image of surface damage due to plasma of 15% vinyl tris textured elastomer sample (smooth area)

PAGE 59

47 Figure 3.6 Profilometer image of surface damage due to plasma of 15% vinyl tris textured elastomer sample Figure 3.7 AFM Image of crack on plasma treated sample

PAGE 60

48 Figure 3.8 Profilometry data for non-treated (A) and plasma treated (B) 1.5 m deep elastomer textured surfaces at the 5 m width spacing. Height of ridge: A = 1.433 m, B = 0.809 m. Another feature of using a silicone surface is the effect of rearrangement of the surface properties due to mobility of the chains and migration of low MW oligomers. To examine this effect, unmodified elastomer samples were treated at two plasma power levels, 50W and 100W using the same procedure as in the methods section of this chapter. Contact angles of water on plasma treated samples were examined using the sessile drop method immediately after treatment, 15 minutes after treatment, and one week after treatment and the contact angles reported in Table 3.7. Table 3.7 Water contact angles on plasma treated silicone samples after exposure to air over time Average Contact Angles (degrees) Treatment conditions 50 W, 5 min 100 W, 5 min Before Treatment 103 8 114 3 Immediately after treatment < 10 <10 15 minutes after treatment 15 4 <10 1 week after treatment 77 7 45 28

PAGE 61

49 The data indicate that with the more intense plasma treatment, the hydrophilicity of the surface is more stable over a period of a week, however, the high standard deviation after 1 week exposure to air for 100 W sample implies that the surface has areas of homogeneity that are rather extreme. Some areas had contact angles as low as 25 while others had angles ~80. The stability of the surface is possibly due to a silica-like crust as mentioned above, and the areas of hydrophobicity are due to low MW oligomers migrating through cracks. Dynamic Contact Angle Dynamic contact angles are taken by advancing or removing a liquid interface on a surface. This can be accomplished with a variation of the sessile drop method by either adding or subtracting fluid, or by tilting the plate and measuring the angle. Another technique is the Wilhelmy plate method, where a film of material is dipped in a liquid and the force on the plate is measured. The force is related to the surface tension of the liquid by the equation: cosPFlv where P is the perimeter of the plate. By measuring the force and perimeter of a sample in a known liquid, the contact angle can be determined for both advancing (inserting) and receding (withdrawing) contact angles. Dynamic contact angle data was taken on unmodified silicone and the linear non-functionalized additives. Functionalized oligomers were not available at the time of examination and the method of production had yet to be determined. In addition, samples with both smooth and textured areas were examined. Figure 3.9 is a representative force-distance curve taken from DCA data. The lower linear portion of the curve represents the

PAGE 62

50 advancing or inserting portion of the dip. The force values are negative due to the fact that the low-energy surface is resisting being wetted by the water, and in essence a reverse meniscus is pushing up against the sample. The upper linear portion is the receding area of the curve. The separation of the two curves is a measure of the hysteresis, or the difference between the advancing and receding contact angles, represented as in Table 3.8. The slope of the linear portions is due to a buoyancy effect and is factored out by extrapolating the line to the point of sample contact with the water, or zero depth of immersion (ZDOI). DCA of Unmodified Silicone Elastomer in Water-500-400-300-200-1000100200300024681012141Position (mm)Force (mg) 6 Advancing AreaReceding AreaZDOI Figure 3.9 Force-distance curve from DCA of unmodified silicone elastomer in water Hysteresis is caused by multiple factors, especially roughness and surface heterogeneities. For most smooth surfaces, roughness under 0.5 m contributes minimally to the hysteresis of a sample.[125] Hysteresis due to rearrangement of the polymer backbone so that more hydrophilic moieties are exposed to the surface play a large role in the decrease of receding contact angles compared to advancing.[123, 125]

PAGE 63

51 Essentially, as the surface is in the receding phase, the meniscus is traveling over a previously wetted surface with a different surface energy than before. Due to this phenomenon, the advancing contact angle can be considered a measure of the low-energy portion of a heterogeneous surface and receding angles are more characteristic of high-energy parts.[125] In examining Table 3.8, one aspect that stands out is the small hysteresis of the 20% 5000 cSt samples. This is most likely attributed to the fact that these samples had a visible coating of oil on the surface, which may mask the rearrangement process, or at least minimize the effects. Table 3.8 DCA data on silicone elastomer modified with non-functionalized PDMS oligomers Viscosity Wt. % adv rec Unmodified 0 115.1 3.8 68.7 2.2 46.4 1.7 50 cSt 5% 113.9 1.8 77.5 1.8 36.4 0.3 50 cSt 20% 100.5 1.3 65.1 2.1 35.4 1.6 5000 cSt 5% 106.1 0.7 71.6 2.2 34.5 2.1 5000 cSt 20% 101.0 0.8 89.4 4.4 11.6 4.4 Figure 3.10 is an example of a force-distance curve taken on a textured substrate. The sample advances into a smooth area, and then into an area of texture. At the end of the advancing dip, the sample is retracted through the textured area and then on to the smooth area. Several factors preclude the presentation of numerical results for this study. For many samples, the lengths of smooth and textured areas were not long enough to achieve a linear region of stable contact angles. In addition, the varying thickness and the fact that only one side of samples were textured make numerical contact angles unreliable. However, for the majority of the samples, the trend was similar. The advancing textured areas had higher contact angles than the smooth areas, while the

PAGE 64

52 receding textured contact angles had wide variations. Due to the high aspect ratios and significant roughness of the textured areas, air is trapped in the grooves when the advancing front of the liquid passes over. As can be seen by Figure 3.10, the advancing force decreases as the liquid front reaches the texture, which in effect is increasing the resistance of the water to the film penetration. As it encounters areas of low surface energy air trapped in between the grooves, the water adheres to the ridges until the surface tension is overcome and it can bridge to the next ridge. Textured Silicone Elastomer20% wt/wt. 5000 cSt oil-350-300-250-200-150-100-50050100150051015202530Position (mm)Force (mg) Advancing Smooth AreaReceding Textured AreaReceding Smooth AreaAdvancing Textured AreaSmooth/Textured interface Figure 3.10 Force distance curve from DCA on a textured silicone elastomer substrate Optical Profilometry Optical profilometry is a useful tool for characterizing the shape and 3-D features of textured substrates. It uses white light interferometry to determine the topography of a substrate without contacting the surface. It also has the great advantage that no surface

PAGE 65

53 modification or contact is necessary to take measurements. Imaging procedures like SEM require a surface coating and high-energy treatment of the surface, which may change the properties as well as rendering the sample itself unusable in the future. Figure 3.11 shows representative 3-D renderings of an unmodified silicone copy off the 5 m deep textures. Figure 3.12 includes profilometry images taken directly off the wafer for comparison. The textures in Figure 3.11 were replicated from epoxy, demonstrating the ability of the epoxy system to faithfully recreate the wafers features. Note that these copies are in effect the negative of the wafers, in that the ridges have varying widths and the grooves are a constant width and depth of 5 m. Figure 3.11 3-D images of 5 m deep elastomer samples copied off of epoxy taken with optical profilometer (50X). (A-C) are examples of the ridges at the different spacing. (D) demonstrates the smooth/textured interface

PAGE 66

54 Figure 3.12 3-D images of 5 m deep etched silicon wafers at each width spacing, taken with optical profilometer. (50X A-C, 100X D) Figure 3.13 is an example of the features from a 1.5 m deep wafer. These were cast off the wafer with 5% LMW elastomer and cured at 80C. The left side of the image is elastomer copies of the wafer, which is represented on the right side. Replication is good, but the shape of the 1.5 m ridges are more rounded and pointed at the top after replication (see Figure 3.8A for 2D profilometry of 5 m spacing). Table 3.9 gives the quantitative values for ridge depth and width taken with the optical profilometer. All of the ridge widths for the wafers are 5 m since the groove spacing is varied. This translates to variations in ridge spacing for the elastomer copies. What is apparent is that

PAGE 67

55 the etching process does not give perfectly spaced features, especially with the 1.5 m wafer. The 5 m deep wafer has better retention of expected ridge values, most likely due to a more controlled reactive ion etching process. Table 3.9 Optical profilometer data of ridge widths and groove depths of the 1.5 and 5 m deep wafers and elastomer copies. The wafers should have a constant ridge width and the elastomer copies have ridge widths of 5 m, 10 m, and 20 m. Sample Ridge Width (m) Groove Depth (m) 5 m wafer 4.8 0.1 4.9 0.2 4.9 0.2 5.3 0.2 5 m elastomer 4.7 0.1 9.2 0.4 19.1 .1 4.9 0.1 1.5 m wafer 3.8 0.1 3.4 0.2 3.7 0.1 1.4 0.1 1.5 m elastomer 3.7 0.3 8.2 0.2 17.6 0.5 1.4 0.10 Figure 3.13 3-D images of 5% LMW elastomer copies (A, B) of a 1.5 m wafer (C, D).

PAGE 68

56 Some of the limitations of the optical profilometer system are a result of the scale and arrangement of the features. As can be seen by comparison of Figure 3.11 and Figure 3.12, both have similar shapes of ridges, with the sidewalls sloping downward so that the walls appear to not be vertical. Since the profilometer is non-contact and requires the reflectance of light, as the features get deeper and closer together, some data can be lost if the magnification is not high enough or if the material does not reflect light out of small features. For spacing greater than 5 m, and magnifications of 50X or more, this problem is not as great. This instrument is a good addition to the analytical capabilities of a researcher since it is easy and quick to use with little to no sample modification, and gives a quality replication of the image surface in three dimensions. With the analysis reported in this chapter, coupled with the background in contact guidance and cell growth, a better understanding of the factors and variables involved in the growth of cells on textured surfaces is possible.

PAGE 69

CHAPTER 4 CONTACT GUIDANCE OF ENDOTHELIAL CELLS Introduction In the previous chapters of this thesis, the principles of contact guidance and the materials used for this study have been detailed. The final step of this project was to examine the contact guidance phenomenon on a novel system to determine the effects of elastic modulus on a surfaces ability to direct cell growth. Contact guidance was quantified on a group of silicone elastomers with elastic modulus values varying almost 800% but with similar surface energies. Textured surfaces made from four elastomer formulations were seeded with porcine vascular endothelial cells (PVECs). Cell nuclei were imaged and the nuclear shape was compared for surfaces of varying ridge dimension, groove depth, and material modulus. Materials Elastomer Substrates As described in the previous chapter, the unmodified base elastomer is the Silastic T-2 Silicone Moldmaking Rubber produced by Dow Corning. Three vinyl terminated PDMS based oligomers were included to give the elastic modulus a range from 0.3 MPa to 2.3 MPa. Pertinent information about the elastomer systems, including modulus and surface free energy, s, is listed in Table 4.1. The vinyl-terminated oligomers were selected over non-functionalized oils due to their incorporation into the network structure and to minimize additive release. 57

PAGE 70

58 Table 4.1 Properties of elastomeric substrates for contact guidance experiments Sample Reference Additive Modulus (MPa) s (mN/m) Unmodified None 1.4 0.1 18.9 5% LMW 5% Vinyl terminated PDMS 550 g/mol 2.3 0.5 19.1 5% HMW 5% Vinyl terminated PDMS 28,000 g/mol 1.0 0.1 21.0 15% vinyl tris 15% Vinyltris(trimethylsiloxy)silane 0.3 0.1 17.1 Cell Culture and Seeding Bovine plasma fibronectin was received as a lyophilized powder from Sigma (cat# F-4759, 2 mg). PVECs were obtained from Dr. Edward Blocks lab at the Malcom Randall VAMC in Gainesville. Endothelial cells were obtained from the main pulmonary artery of 6 to 7 month old pigs and were propagated in monolayer cultures and characterized as described by Patel et al.[126] Third to sixth-passage cells in postconfluent monolayers maintained in RPMI 1640 medium (Life Technologies, Grand Island, NY) containing antibiotics (100 U/ml of penicillin, 100 g/ml of streptomycin, 20 g/ml of gentamicin, and 2 g/ml of Fungizone) were used in all studies. Fetal bovine serum (FBS, was added at a 10% concentration for plating and growth. Falcon tissue culture flasks and 24 well plates were used for cell passaging and seeding. Trypsin/EDTA 1X solution was stored in frozen 10 mL aliquots. Hanks balanced salt solution (BSS) was used for washing flasks and samples. PVEC nuclei were stained with Hematoxylin 2 from Richard Allan Scientific. Cytoplasm and various features were stained with a 1% aqueous crystal violet solution.

PAGE 71

59 Methods Elastomer Sample Preparation Silicone elastomer samples were prepared as described in the previous chapter. The four elastomer formulations reported in Table 4.1 were used for this portion of the study. Cure times and curing agent amounts are listed in the previous chapter, Table 3.3. For all samples, constant thickness was maintained at 3 mm with the use of appropriate spacers between glass plates. Textured sample were cast directly off silicon wafers with the patterns etched into the surface. Untextured films were cast between two PET sheets attached to glass plates. Samples were cut out using a punch with an interior diameter of 14.37 mm and outer diameter of 15.53 mm. All samples were sterilized in the same fashion by rinsing with 70% EtOH and drying overnight in a sterile hood. Surface Treatment by Fibronectin Some surfaces were coated with fibronectin (FN) as described in the previous chapter. Sterilized samples were placed in a 16 mm diameter well of a 24 well placte, and a 50 g/mL solution of FN was added in 0.5 mL aliquots to each well with a sample. After exposure to vacuum to remove trapped air, samples were left to incubate for 1 hour at room temperature. The FN solution was aspirated out and then the samples were washed 3X with Hanks BSS. Surface Treatment by RFGD Plasma Textured surfaces were exposed to an argon RFGD plasma at 50 W for 5 minutes as described in the previous chapter after sterilization by EtOH. The argon regulator was set at 20 psi and the flowrate was 200 sccm. Samples were treated 4 at a time, 5.5 cm below the RF coils, with one sample from each material per treatment to ensure similar surface modification between batches. After treatment, the samples were moved to a

PAGE 72

60 sterile hood and left exposed to air for 10 minutes for each sample, then transferred to a 24 well plate. Cell Culture Techniques PVECs were supplied by Bert Herrara from Dr. Edward Blocks lab weekly as suspensions in 12 mL of media. All cells received from Dr. Blocks lab were between passage 2 and 5. Cells were expanded by diluting the suspension to 20 mL with fresh media and subsequently transferred as 10 mL aliquots to a 75 cm2 angled neck, vented tissue culture flask. Flasks were incubated at 37C and 5% CO2 for 48 hours, and then existing media was exchanged for fresh. Media was changed every 72 hours after that. Typically, the PVECs formed a confluent monolayer on the culture flask within 72 hours from initial plating. Cell passage procedure To passage the cells, confluent flasks were washed 3X with Hanks BSS. A few mL of trypsin/EDTA 1X solution was poured in the flask and swirled to counteract any remaining serum proteins and the remaining liquid was poured off. A small amount (~1-2 mL) of fresh trypsin/EDTA solution was added again, with just enough to coat the bottom of the flask. The flask was then placed in the incubator at 37C for 5 minutes, and then checked with the inverted microscope to determine the appearance of the cells. Once the cells became rounded, the sides of the flasks were struck on each side to dislodge the remaining adherent cells. Serum containing media was added to counteract the trypsin, and the suspension was mixed by aspiration with a 10 mL pipet. The suspension was split into 3 flasks from each original flask and left to incubate as before.

PAGE 73

61 Determination of cell suspension concentration To determine the average amount of cells per mL in a cell suspension, a hemacytometer was used. A hemacytometer has a chamber with 10 separate grids each measuring 1 mm square. A coverslip was placed over the counting area, which resulted in a well 0.1 mm deep. Cell suspension was transferred by pipet to the edge of the coverslip, where the suspension was drawn into the chamber. Counts of 10 different squares were made, and the average count was multiplied by 104 to determine the average cells per mL in the suspension. If necessary, suspensions were diluted to achieve a cell count between 1 x 105 cells/mL and 2 x 105 cells/mL before seeding on samples. Cell Seeding on Samples For all samples, 1 mL of cell suspension was seeded into a well. Each sample has approximately 2 cm2 of surface area, and at a seeding density of 2 x 105 cells/mL, then approximately 1 x 105 cells/cm2 was added to the sample. For fibronectin-coated surfaces, the cells were suspended in serum free media since the adhesion protein was already adsorbed on the surface. Cells were seeded in normal 10% FBS media on plasma treated and untreated surfaces. After seeding, the samples were incubated at 37C and 5% CO2. After 48 hours, the media was changed to remove non-adherent cells and replace with fresh media. Plasma treated surfaces exhibited an initial period where endothelial cells clumped into groups (after 24 hours), and then eventually spread out to a monolayer. Cells on these surfaces were left to grow for 5 days. Fibronectin coated surfaces showed improved cell spreading and attachment, and cells were imaged after 48-72 hours.

PAGE 74

62 Cell Staining and Image Capture Many methods were examined to effectively stain the cells. To image the entire cell body and morphology, cells were stained in an aqueous 1% crystal violet solution. First, after removing the media, cells were washed 2X with BSS, then fixed with cold 10% n-buffered formalin for 20 minutes in the culture well. After removal of the formalin, fixed samples were washed with BSS and crystal violet was added and cells stained for 20 minutes. Samples were removed from the wells using a needle to minimize flexing the substrates. The samples were washed in saline and placed on a slide, then covered with a coverslip to prevent drying of the surface. To image only the nucleus for the nuclear form factor, a hematoxylin stain was used. Hematoxylin stains nuclear materials, specifically basophilic structures such as DNA and RNA and was therefore chosen so that nuclear elongation could be used to quantify contact guidance. The method followed the Sigma method for Gills hematoxylin staining. Briefly, cells were fixed in 95% EtOH for 10 minutes, rinsed 2X with tap water, and stained for 2 minutes in hematoxylin. Longer staining times increased the intensity of the nuclear stain, but also increased staining of the cytoplasm. After the hematoxylin, the samples were rinsed 2X in water and 2X in 95% EtOH. The substrates were placed on a slide and covered with a coverslip. Cells were imaged on the surface at 200X using a Nikon Optiphot microscope and Matrox image capturing system. Multiple images were taken at each feature width that included at least 5 nuclei. Images were saved as a jpeg format and imported to Adobe Photoshop for analysis.

PAGE 75

63 Image Analysis Images were analyzed to measure the ratio of the length to the width of each nucleus. A schematic of the measurements taken can be found in Figure 4.1. Figure 4.1 Representation of the values measured in the nuclear form factor. The left side of the image is a top view of a nucleus on a microtextured substrate, while the right side is a cross-sectional view. The length measured is represented by A, while the width is represented by B. The nuclear form factor is Log (A/B) (Drawing by Chuck Seegert) The imported image files were modified to improve the contrast between the nuclei and the background using Adobe Photoshop 6.0. A pictorial explanation of the steps can be found in Figure 4.2. All images were adjusted using the auto-contrast macro in the software package (Figure 4.2B). Then, batches of images were opened that came from the same sample since they all had the same angle of orientation. By drawing a measured line along the length of a groove, and then using the arbitrary rotate command, the image is automatically rotated by the same angle as the measured line. The end effect is to line up the ridges vertically on the computer screen (Figure 4.2C). By recording the commands and initiating a batch process over the range of the opened images, all of the features for each sample were aligned.

PAGE 76

64 50 m 50 m 50 m 15 m Figure 4.2 Steps in image processing technique of hematoxylin stained nuclei. The original picture (A) is contrasted (B), rotated to align the textures (C), and then measured for length (D) and width (not shown). The reported value is log(Length/Width) The benefit of the vertical alignment is to improve the efficiency of measuring the nuclei. The procedure introduced by Dunn and Heath[10] requires the measurement of the length and width of nucleus at its widest point, where the length measured is parallel to

PAGE 77

65 the ridge, and the width is perpendicular. With the images aligned vertically, the measurement lines can be constrained by the software to perfectly vertical or horizontal, and so by using that method the maximum length and maximum width of each nucleus was measured and entered into an Excel spreadsheet. A 5 x 5 grid was superimposed on the images, and 5 nuclei were chosen per image, each from a separate square of the grid. Using this method, at least 20 nuclei per sample setting were quantified. Results and Discussion Contact Guidance on Textured Surfaces As discussed in Chapter 2, surface ridges and grooves typically act to elongate cells along the ridge. This phenomenon is mainly reported in qualitative fashion, and fibroblasts are predominant as the cell system, since they are relatively easy to grow and have shown good results in terms of contact guidance. Quantitative studies of contact guidance on textured surfaces have used SEM and phase contrast microscopy to map the entire cell body and try to determine its alignment. A serious limitation to this method is the fact that unless the cells are isolated from others, the cellular dimensions and alignment can be difficult to determine. From personal experience, cells that are not in contact with others are typically more elongated and tend to show greater contact guidance than confluent cells. Other methods to quantify contact guidance stain actin filaments and other cytoskeleton components to detect the alignment of the interior stress fibers with fluorescent or confocal microscopy. For cells in this study, groups of cells were examined, rather than individual cells apart from the rest of the culture. The main purpose of this project was to examine the effects of modulus as a factor involved in the ability of a substrate to direct cell growth.

PAGE 78

66 While modifying the underlying substrates mechanical properties might alter a cells ability to adhere and change its morphology, it will not direct cells to grow in a certain pattern. In order to affect this change, the surface was patterned with microtexture in the form of ridges, as characterized in the previous chapter. Surface texture was examined using the unmodified elastomer as the reference material. For this portion of the experiment, two main factors were examined, ridge width and groove depth. As stated in the previous chapter, plasma treatment of the surface resulted in small random cracks and an unreliable texture dimension. Due to the concerns as to the effects of the plasma treatment to the fidelity of textures, fibronectin adsorbed materials were examined most closely. Figure 4.3 is a main effects plot of fibronectin coated unmodified elastomer. The average of all the data points is represented by the dotted line, and the individual means are compared to this overall mean. The term Feature in the graphs and subsequent analysis refers to the width in microns of the ridges in the section examined. All of the grooves were 5 m in width. The term depth refers to the depth of the grooves in microns, either 1.5 m or 5 m. Smooth images were taken from the same samples, at least 50 m from the closest texture. Log (L/W) is defined as the nuclear form factor and is the variable that measures the strength of nuclei alignment along the ridges. The more positive this variable is, the more the nuclei exhibit contact guidance. Values greater than 0.15-0.20 are typically very well aligned. The closer log (L/W) is to zero, the less the cell shows deference to the topography at all. A negative number implies that the cells are guided orthogonal to the textures.

PAGE 79

67 Feature Width (m)Groove Depth (m)51020smooth1.55.0 0.040.090.140.190.24 log (L/W) Main Effects Plot Data Means for log (L/W)Fibronectin Coated Figure 4.3 Main Effects Plot of the data means of fibronectin coated unmodified elastomers at various ridge widths (Feature) and groove depths (Depth) The main effects plot helps to elucidate which factors and levels play a role in the system. The more the data changes between levels, the greater the effect. Figure 4.4 is an interaction plot generated by Minitab that demonstrates the changes at the different factor levels for fibronectin coated unmodified elastomer. The important point that is illustrated here is that at the deeper groove depths, the greater the alignment of the nuclei to the ridges. One-way analysis of variance (ANOVA) and a multiple comparison test (Tukey, 95% CI) on each depth comparing the different feature widths demonstrate a significant difference between each level at the 5 m depth, but only a statistical difference at 1.5 m between 5 m wide ridges and smooth textures. One of the difficulties in the analysis of these samples is in the fact that for the smooth samples, the nuclei are not round, but elongated in random directions. The effect of this is to give a

PAGE 80

68 mean value for log (L/W) for the smooth areas around zero, but with a large standard deviation. 10 20 5 smooth 5.01.5 0.30.20.10.0 Depth Feature Mean Interaction Plot Data Means for log (L/W)Fibronectin Coated Figure 4.4 Interaction plot representing the change in contact guidance with depth and feature width for fibronectin coated silicone elastomer. Figure 4.5 and Figure 4.6 are main effects and interaction plots for plasma treated unmodified elastomers generated by Minitab. These are the same plots as Figure 4.3 and Figure 4.4, which are treated with FN. As can be seen in Figure 4.5, the difference between the feature widths is not as obvious, and the effect of the depth on the alignment does not play as large a role from 1.5 m to 5 m, as confirmed by the small slope. A two-way ANOVA comparison on plasma treated unmodified elastomer comparing the effects of feature size and depth show that when including the smooth areas in the analysis, the feature size is a very significant contributor to variance (p < 0.0001) while depth is not a significant influence. By removing the smooth terms and

PAGE 81

69 5.01.5smooth20105 0.160.120.080.040.00 log (L/W) Main Effects Plot Data Means for log (L/W)Plasma TreatedFeature Width (m)Groove Depth (m) Figure 4.5 Main Effects Plot of the data means for plasma treated unmodified elastomers. The left side of the plot compares effects due to the feature width, while the right side compares the groove depth. repeating the analysis such that the comparison is strictly on the textured surfaces, both the width and the depth are significant sources of variance (p < 0.05). One way ANOVA comparisons followed by a multi-comparison test (Tukey, 95% CI) between the ridge widths at constant depth on plasma treated unmodified elastomer show that for each width there is a significant difference when compared to smooth samples, but the difference is not significant when compared to the other widths. To put it simply, for plasma treated samples at both depths, each groove width shows significantly more alignment when compared to a smooth surface, but there is no statistical difference when comparing the different permutations of the three ridge widths.

PAGE 82

70 10 20 5 smooth 5.01.5 0.20.10.0 Depth (m) Feature width (m) Mean log (L/W) Interaction Plot Data Means for log (L/W)Plasma Treated Figure 4.6 Interaction plot representing the change in contact guidance with depth and feature width for plasma treated silicone elastomer. In this graph the contact guidance data for PVECs grown on the neat sample with no surface modification at 1.5 m depth is included. Due to substrate production limitations at the 5 m depth, untreated samples were not available for analysis. In comparing these results to the fibronectin coated surfaces, one point that stands out is that for the 1.5 m deep FN coated substrates only the 5 m wide ridges were significantly different from the untextured surfaces, while at 5 m there was a statistically significant (alpha = 0.05) higher degree of alignment. The 1.5 m plasma treated surfaces did have significant alignment compared to smooth surfaces, and a higher log (L/W) value than 1.5 m fibronectin samples. However, due to the surface irregularities of the plasma treated surface explained in the previous chapter and the fact that a silica-like layer would expose the cell to a seemingly harder substrate; direct

PAGE 83

71 -0.10.00.10.20.30.451020SmoothRidge Width (m)Log (L/W) 1.5 um deep Plasma 1.5 um deep FN 5 um deep Plasma 5 um deep FN 1.5 um deep Untreated Figure 4.7 Comparison of surface treatments on textured unmodified silicone elastomer by ridge width and groove depth (error bars represent standard error of mean) comparison of the two methods may not be appropriate due to material differences. The untreated samples show a high degree of alignment at all texture settings. Cell attachment and proliferation on these surfaces was significantly lower than on the treated surfaces, but those cells that were attached were aligned at all levels of texture. The morphology of these cells was highly elongated (compare Figure 4.8 and Figure 4.9) as the cells attempted spread and stabilize. The fact that almost all of the cells showed a high degree of alignment on the untreated surface suggests that when they come into contact with a surface with less than ideal adhesion capability, the presence of surface texture plays an increasingly important role.

PAGE 84

72 Figure 4.8 PVECs grown on untextured fibronectin coated LMW sample stained with hematoxylin Figure 4.9 PVECs grown on 5 m spacing 1.5 m deep untreated LMW stained with hematoxylin 50 m 50 m Contact Guidance on Textured Surfaces of Varying Modulus The main unknown factor in the design of this project is the modulus of the material and its effect on contact guidance. As discussed in Chapter 2, many different materials, from metals to elastomers, have been examined in the study of direct cell growth. In published research on contact guidance, the modulus of the sample substrate is very rarely reported. Most materials used with different mechanical properties have also significantly different surface energetics, which directly affects protein adsorption and adhesion, making a comparison due to modulus improbable. These effects would more than likely wash out any change due to the effects of mechanical properties. The purpose of this section is to examine whether or not the modulus, as a measure of the compliance, can significantly affect the ability of a material to direct cell growth in the range covered by these materials.

PAGE 85

73 FeatureDepthMaterial10205smooth1.55.015% vinyl tris5% HMW5% LMWUnmodified 0.000.050.100.150.20 log (L/W) Main Effects Plot Data Means for log (L/W) Figure 4.10 Main effects plot comparing log (L/W) to feature width, feature depth, and material used on fibronectin-coated elastomers. Unmodified5% LMW5% HMW15% vinyl tris5.01.50.300.150.000.300.150.00 Feature Width (m)Depth (m)Material 5.01.5smooth52010Interaction Plot Data Means for log (L/W)Log (L/W) Figure 4.11 Interaction plot representing the change in factors of PVECs grown on fibronectin coated materials.

PAGE 86

74 Figure 4.10 is a main effects plot from Minitab similar to Figure 4.3 and Figure 4.5 on fibronectin-coated surfaces. As mentioned previously, due to the uncertainties with the reproducibility of plasma treated surfaces, fibronectin materials will mainly be examined. As seen before, the effects of depth and width of ridges seem to play the largest role in the response of log (L/W). The relative flatness of the material interaction implies that there is little contribution due to a change in modulus. In qualitatively interpreting Figure 4.10 and Figure 4.11, it appears that contact guidance increases with decreasing feature width and increasing depth on each of the materials, but a direct correlation between the material choice and the degree of alignment is not obvious. One way ANOVA comparisons followed by a Tukey multi-comparison test (CI = 0.95) for the 5 m depth materials show no significant difference between any of the materials at the 5 m groove width. It is apparent from Figure 4.12 that the groove depth at that point is much more significant and few trends can be found in relating the modulus. In effect, the groove depth is overpowering any effect that modulus might have on the contact guidance. At the 1.5 m depth, there are more significant differences with respect to materials. At the 5 m and 10 m ridge width, there is a significant difference between the 5% LMW material and the 15% vinyl tris material with respect to log (L/W). The 1.5 m data is represented in Figure 4.13. Note that as reported in Table 4.1, these represent the high modulus (2.34 MPa) and low modulus (0.3 MPa) materials respectively. At 20 m ridge widths, there is no significant difference between the two. Figure 4.14 compares the effect of ridge width and groove depth strictly on the LMW and Tris surfaces. The trend of the materials is to

PAGE 87

75 -0.10.00.10.20.30.451020SmoothRidge Width (m)Log (L/W ) 5 FN LMW 5 FN Unmod 5 FN HMW 5 FN Tris Figure 4.12 Comparison of materials on textured unmodified silicone elastomer by ridge width on 5 m deep grooves (error bars represent s.e.m.) -0.10.00.10.251020SmoothRidge Width (m)Log (L/W) 1.5 FN LMW 1.5 FN Unmod 1.5 FN HMW 1.5 FN Tris Figure 4.13 Comparison of materials on textured unmodified silicone elastomer by ridge width on 5 m deep grooves (error bars represent s.e.m.)

PAGE 88

76 show increased contact guidance on the high modulus material. As stated before, this effect is not significant in the deeper grooves, and at 5 m the low modulus tris sample even shows higher contact guidance, albeit statistically insignificant. -0.10.00.10.20.30.451020SmoothRidge Width (m)Log (L/W ) 5 FN LMW 1.5 FN LMW 5 FN Tris 1.5 FN Tris Figure 4.14 Comparison of ridge width and groove depth to the high modulus material (LMW) and the low modulus material (Tris) (error bars represent s.e.m.) Since the 5 m deep grooves have shown the strongest factors for contact guidance, the reason there is not a significant difference between materials at the greater depth might be that the importance of the groove depth overpowers any change due to the modulus. At the lower depth where the effect of the depth is not as great, it appears that the modulus of the surface alters its ability to direct the growth of cells. According to the data and trends, the higher modulus material seems to enhance the contact guidance phenomenon.

PAGE 89

77 As discussed in Chapter 2, cells attach to surfaces at focal adhesions, and they impart a stress upon the substrate which is balanced by the material and acts as a counterweight so to speak improving adhesion. Researchers have examined the forces of cells such as fibroblasts on elastomeric substrates, using low modulus (~ 15 kPa) silicone films to quantify the forces of adhesion.[87] These cells continually pull and contract against these surfaces until they reach a balance, the equivalent of pulling the slack out of a rope until it is taut. Fibroblastic cells on higher modulus materials were shown to spread better and were of a more constant shape, while cells on lower modulus materials were more active and elongated.[98] Possibly the cells elongate along the grooves because they can pull along the length of the groove as opposed to the width, which is more compliant due to its reduced thickness. Since they have more resistance along the length, they have a more stable opposing force to pull and align to. At higher ridge widths, this effect is lessened by the increased continuous surface area for attachment as well as possibly the increased thickness of the ridge. In this study, vascular endothelial cells were shown to preferentially align along the length of a groove and increased that alignment as the groove depth increased. The fact that changing the modulus had little effect on the alignment for the deeper grooves is not completely unexpected, since it is well documented that groove depth plays an important role in contact guidance, while the contribution due to the mechanical properties is not well defined or studied. However, if mechanical stability is a main factor in directing the cell growth, then one should see a greater effect on the deeper features, since the ridges are even less mechanically stable at the higher aspect ratio. Another issue is that at deeper groove widths, the cells are more

PAGE 90

78 likely to span a groove without touching the bottom of the groove. This in effect lines up the possible areas of focal adhesions by leaving the only area for adhesion on the top of the ridge. At lower groove widths and shallower grooves, the modulus of the material seems to play a more important role, but one only seen by comparing the highest and lowest modulus silicone materials available. Substrates like titanium and polystyrene could be used as well for comparison as a much higher modulus material, but their behavior in regards to surface energy will play a significant role that cannot be overlooked. The nuclear form factor is a useful tool for elucidating the alignment of a cell, and one that may in the future become more widely used. The power of the model stems from the ease in imaging and data processing as compared to examining strictly actin filaments or trying to map an entire cell.[2, 37] The flaws in its use come from its sensitivity to less elongated cells and in comparing cells that are aligned at 45 angle to those with nuclei that are more rounded. Both types of nuclei would give similar results of zero using the nuclear form factor. While neither are aligned to the features, a group of cells aligned at 45 would result in an interpretation that the cells were not aligned or randomly aligned, when in fact they could be aligned all in one direction at 45 due to some other unknown factor. In reference to sensitivity, a simple model comparison can show one of the drawbacks of using the nuclear form factor. A highly elongated nucleus oriented 30 to the direction of the features can have the same dimensions in length and width as a nucleus that is perfectly aligned to the features, but not as elongated. With a large sample size, these problems can be minimized, but they should be issues to consider for potential researchers. By incorporating a corroborating measurement such as the

PAGE 91

79 angle of deflection of the nucleis long axis from the direction of the ridges, or combining cystoskeletal measurements, more confidence in the method would be assured for future studies and solve both of these issues.

PAGE 92

CHAPTER 5 CONCLUSIONS AND FUTURE WORK Conclusions Microtextured Surfaces To examine the effects of contact guidance on silicone elastomers, microtextured substrates were produced with reproducible and well-defined surfaces. Ridges of 10,000 m length were fabricated at 3 different widths: 5 m, 10 m, and 20 m, separated by 5 m wide grooves to determine the effect of separation of features on the alignment of porcine vascular endothelial cells. Two depths were examined: 5 m and 1.5 m. The silicone elastomer samples were produced by casting a film on a textured mold and allowing the samples to cure. Molds used were either silicon wafers or epoxy replicates of the wafers. The surfaces were created with micromachining technology, specifically photolithographic patterning followed by reactive ion etching. After examination of the surfaces by optical profilometry, it was determined that the silicone copies faithfully reproduced the textured surface and the textures were of the expected design. Closer examination demonstrated that the 1.5 m deep wafer and samples had a more rounded appearance and had ridge widths ~1 m less than expected. The depths measured corresponded with designed values of 1.5 m and 5 m within experimental error. All elastomer types used in this study faithfully reproduced the applied texture and gave a stable substrate for comparison. 80

PAGE 93

81 Surface Energy and Treatment Elastomer samples were examined with contact angles to determine their relative wettability and surface free energy. Formulations of elastomer with both functionalized and non-functionalized PDMS oligomer additives were examined, and there was no significant effect to the surface energy as determined by Zisman plots using 5 separate liquids. Sessile drop contact angles measured with a goniometer were measured, and the surface energy of the unmodified elastomer samples was found to be 18.9 mN/m. Water contact angles were typically ~ 110 for unmodified samples. Surfaces were treated with fibronectin and radiofrequency glow discharge plasma in argon for 5 minutes at 50 W. Both treatments significantly increased the hydrophilicity after treatment, as measured by captive bubble contact angles, from 86.7 4.3 for the unmodified sample to 14.5 3.5 for fibronectin adsorbed surfaces and < 10 for RFGD plasma treated surfaces. Dynamic contact angle analysis was performed on the unmodified elastomer and materials with non-functional PDMS oligomers added. Advancing contact angles varied between 100.5 and 115.1 and the hysteresis between the advancing and receding angles was between 35 45 except for high molecular weight, high weight percent additive which formed a visible layer of oil on the surface. The hysteresis on this surface was quite low (11.6 4.4). Dynamic contact angle analysis of textured surfaces showed a difference in smooth and textured areas, although quantitative data was not available due to experimental uncertainty. The trend of the graphs demonstrated an increase in observed advancing water contact angles, which is possibly due to composite surfaces with air trapped in the textures. Analysis of the RFGD treated PDMS elastomer surfaces revealed defects that were caused by the plasma treatment. Groove depths decreased by 40% in some cases as

PAGE 94

82 the material was ablated by the plasma. Cracks on the order of 0.5-1.5 m were seen to form after mechanical manipulation of treated surfaces. This phenomenon seems to be a result of the cracking of a hard silica-like layer on the surface.[116] Due to the uncertainties with the feature dimensions and surface mechanical properties, fibronectin adsorbed surfaces were used for the main analysis of the effects of contact guidance. Contact Guidance on Textured Elastomers Contact guidance of PVECs on textured silicone elastomers was measured by the nuclear form factor, in which the log of the ratio of nuclear length to width was presented. Results demonstrated that as the ridge width decreased from 20 m to 5 m contact guidance increased, as well as when the depth of the grooves increased from 1.5 m to 5 m. Data analysis showed that the groove depth was the most important factor in nuclear alignment. Average values of the nuclear form factor for 5 m deep, 5 m wide samples exceeded 0.3, which implies the length was more than twice the width on average. Shallower grooves increased the length of the nuclei by approximately 25% in comparison to the width. Contact guidance on fibronectin-coated elastomers was examined to determine the effect of modulus. It was expected that higher modulus materials would increase the effect of contact guidance. Elastic modulus on 4 elastomers was measured by tensile tests and resulted in a range of values from 0.3 MPa to 2.34 MPa. There was no significant difference in the contact guidance on the deep 5 m grooves with varying modulus. The 1.5 m deep grooves showed a significant increase in the alignment of cells to the groove in the highest modulus material compared to the lowest modulus material for the 5 m and 10 m wide ridges. The conclusion to be taken from this data is that modulus does seem to play a role in the determination of contact guidance, but

PAGE 95

83 other factors such as groove width and especially depth are more significant. With this knowledge, and with future work, ideal values to fine tune materials may be possible to direct and control cell growth. Future Work The importance of controlling cell growth and behavior cannot be underestimated and the phenomenon of contact guidance is only recently approaching maturity. Several areas for future study and improvement are possible and are listed below. Surface Treatment Optimization of plasma treatment needs to be examined, as well as the reasons for the unwanted effects. Argon RFGD at 50 W for 5 minutes is a standard treatment in the literature, and the reasons for the deviations should be investigated. Tether adhesion molecules to silicone surfaces to examine modulus affects with a permanently bound treatment. Topographical Design Smaller ridge widths and an intermediate groove depth should be examined since the data shows that smaller ridge widths improve contact guidance, but the deeper groove depths mask the effect of modulus. Ridge widths on the order of 0.5 m, 1 m, and 2.5 m would give useful data as to the limits of the ridge effect. Opposite or negative designs of the current elastomers would be useful in seeing the effect of a constant 5 m ridge separated by varying smooth areas.

PAGE 96

84 Cell Studies Examination of tissue growth from a central area to uncovered textured substrates should be relatively easy to set up and characterize. Growth patterns of endothelial cells along a textured interior surface of a cylinder would more closely model a seeded vascular graft response.

PAGE 97

LIST OF REFERENCES 1. Curtis, A.S.G. and P. Clark. The effects of topographic and mechanical properties of materials on cell behavior. Critical Reviews in Biocompatibility, 1990. 5(4): p. 343-362. 2. Curtis, A.S.G. and C.D.W. Wilkinson. Review: Topographical control of cells. Biomaterials, 1997. 18: p. 1573-1583. 3. Bhatia, S.N. and C.S. Chen. Tissue Engineering at the Micro-Scale. Biomedical Microdevices, 1999. 2(2): p. 131-144. 4. Brady, R.F. and I.L. Singer. Mechanical Factors Favoring Release from Fouling Release Coatings. Biofouling, 2000. 15(1-3): p. 73-81. 5. Singer, I.L., J.G. Kohl, and M. Patterson. Mechanical Aspects of Silicone Coatings for Hard Foulant Control. Biofouling, 2000. 16(2-4): p. 301-309. 6. Weiss, P. In vitro experiments on the factors determining the course of the outgrowing nerve fiber. The Journal of Experimental Zoology, 1934. 69: p. 393. 7. Weiss, P. Experiments on cell and axon orientation in vitro: the role of colloidal exudates in tissue organization. The Journal of Experimental Zoology, 1945. 100: p. 353. 8. Harrison, R.G. On the stereotropism of embryonic cells. Science, 1911. 34: p. 279. 9. Rovensky, Y.A., I.L. Slavnaja, and J.M. Vasiliev. Behaviour of fibroblast-like cells on grooved surfaces. Experimental Cell Research, 1971. 65: p. 193-201. 10. Dunn, G.A. and J.P. Heath. A new hypothesis of contact guidance in tissue cells. Experimental Cell Research, 1976. 101: p. 1-14. 11. den Braber, E.T., J.E. de Ruijter, H.T.J. Smits, L.A. Ginsel, A.F. von Recum, and J.A. Jansen. Effect of parallel surface microgrooves and surface energy on cell growth. Journal of Biomedical Materials Research, 1995. 29(4): p. 511-518. 85

PAGE 98

86 12. den Braber, E.T., J.E. de Ruijter, H.T.J. Smits, L.A. Ginsel, A.F. von Recum, and J.A. Jansen. Quantitative analysis of fibroblast morphology on microgrooved surfaces with various groove and ridge dimensions. Biomaterials, 1996. 17(21): p. 2037-2044. 13. Mudera, V.C., R. Pleass, M. Eastwood, R. Tarnuzzer, G. Schultz, P. Khaw, D.A. McGrouther, and R.A. Brown. Molecular Responses of Human Dermal Fibroblasts to Dual Cues: Contact Guidance and Mechanical Load. Cell Motility and the Cytoskeleton, 2000. 45: p. 1-9. 14. van Kooten, T.G. and A.F. von Recum. Cell Adhesion to Textured Silicone Surfaces: The Influence of Time of Adhesion and Texture on Focal Contact and Fibronectin Fibril Formation. Tissue Engineering, 1999. 5(3): p. 223-240. 15. Walboomers, X.F., H.J.E. Croes, L.A. Ginsel, and J.A. Jansen. Contact guidance of rat fibroblasts on various implant materials. Journal of Biomedical Materials Research, 1999. 47: p. 204-212. 16. Clark, P., P. Connolly, A.S.G. Curtis, J.A.T. Dow, and C.D.W. Wilkinson. Topographical control of cell behaviour: I. Simple step cues. Development, 1987. 99: p. 439-448. 17. Clark, P., P. Connolly, A.S.G. Curtis, J.A.T. Dow, and C.D.W. Wilkinson. Topographical control of cell behaviour: II. multiple grooved substrata. Development, 1990. 108: p. 635-644. 18. Matsuzaka, K., X.F. Walboomers, J.E. de Ruijter, and J.A. Jansen. The effect of poly-L-lactic acid with parallel surface micro groove on osteoblast-like cells in vitro. Biomaterials, 1999. 20: p. 1293-1301. 19. Schmidt, J.A. and A.F. von Recum. Macrophage response to microtextured silicone. Biomaterials, 1992. 13(15): p. 1059-1069. 20. Wojciak-Stothard, B., A.S.G. Curtis, W. Monaghan, K. Macdonald, and C.D. Wilkinson. Guidance and activation of murine macrophages by nanometric scale topography. Experimental Cell Research, 1996. 223(2): p. 426-435. 21. Wilkinson, C.D.W., A.S.G. Curtis, and J. Crossan. Nanofabrication in cellular engineering. Journal of Vacuum Science & Technology B, 1998. 16(6): p. 3132-3136. 22. Poole-Warren, L.A., K. Schindhelm, A.R. Graham, P.R. Slowiaczek, and K.R. Noble. Performance of small diameter synthetic vascular prostheses with confluent autologous endothelial cell linings. Journal of Biomedical Materials Research, 1996. 30(2): p. 221-229.

PAGE 99

87 23. Dalton, B.A., G.A. McFarland, and J.G. Steele. Stimulation of epithelial tissue migration by certain porous topographies is independent of fluid flux. Journal of Biomedical Materials Research, 2001. 56: p. 83-92. 24. Steele, J.G., G. Johnson, K.M. McLean, G.J. Beumer, and H.J. Griesser. Effect of porosity and surface hydrophilicity on migration of epithelial tissue over synthetic polymer. Journal of Biomedical Materials Research, 2000. 50: p. 475-482. 25. Kirkpatrick, C.J., M. Otto, T.V. Kooten, V. Krump, J. Kriegsmann, and F. Bittinger. Endothelial cell cultures as a tool in biomaterial research. Journal of Materials Science-Materials in Medicine, 1999. 10(10-11): p. 589-594. 26. Sato, M., K. Nagayama, N. Kataoka, M. Sasaki, and K. Hane. Local mechanical properties measured by atomic force microscopy for cultured bovine endothelial cells exposed to shear stress. Journal of Biomechanics, 2000. 33: p. 127-135. 27. Tsukurov, O.I., C.J. Kwolek, G.J. L'Italien, A. Benbrahim, B.B. Milinazzo, N.E. Conroy, J.P. Gertler, R.W. Orkin, and W.M. Abbott. The Response of Adult Human Saphenous Vein Endothelial Cells to Combined Pressurized Pulsatile Flow and Cyclic Strain, In Vitro. Annals of Vascular Surgery, 2000. 14: p. 260-267. 28. Vaidya, R., L.M. Tender, G. Bradley, M.J. O'Brien II, M. Cone, and G.P. Lopez. Computer-Controlled Laser Ablation: A Convenient and Versatile Tool for Micropatterning Biofunctional Synthetic Surfaces for Applications in Biosensing and Tissue Engineering. Biotechnology Progress, 1998. 14: p. 371-377. 29. Kappelt, M. and D. Bimberg. Wet chemical etching of high quality V-grooves with {111} a sidewalls on (001) InP. Journal of the Electrochemical Society, 1996. 143(10): p. 3271-3273. 30. Thomas, C.H., C.D. McFarland, M.L. Jenkins, A. Rezania, J.G. Steele, and K.E. Healy. The role of vitronectin in the attachment and spatial distribution of bone-derived cells on materials with patterned surface chemistry. Journal of Biomedical Materials Research, 1997. 37(1): p. 81-93. 31. Tien, J., A. Terfort, and G.M. Whitesides. Microfabrication through Electrostatic Self-Assembly. Langmuir, 1997. 13: p. 5349-5355. 32. Alaerts, J.A., V.M. De Cupere, S. Moser, P.V.B. de Aguilar, and P.G. Rouxhet. Surface characterization of poly(methyl methacrylate) microgrooved for contact guidance of mammalian cells. Biomaterials, 2001. 22(12): p. 1635-1642. 33. Schmidt, J.A. and A.F. von Recum. Texturing of polymer surfaces at the cellular level. Biomaterials, 1991. 12(4): p. 385-389.

PAGE 100

88 34. von Recum, A.F. and T.G. van Kooten. The influence of micro-topography on cellular response and the implications for silicone implants. Journal of Biomaterials Science Polymer Edition, 1995. 7(2): p. 181-198. 35. Walboomers, X.F., W. Monaghan, A.S.G. Curtis, and J.A. Jansen. Attachment of fibroblasts on smooth and microgrooved polystyrene. Journal of Biomedical Materials Research, 1999. 46: p. 212-220. 36. Matsuzaka, K., X.F. Walboomers, A. de Ruijter, and J.A. Jansen. Effect of microgrooved poly-l-lactic (PLA) surfaces on proliferation, cytoskeletal organization, and mineralized matrix formation of rat bone marrow cells. Clinical Oral Implants Research, 2000. 11: p. 325-333. 37. den Braber, E.T., J.E. de Ruijter, L.A. Ginsel, A.F. von Recum, and J.A. Jansen. Orientation of ECM protein deposition, fibroblast cytoskeleton, and attachment complex components on silicone microgrooved surfaces. Journal of Biomedical Materials Research, 1998. 40: p. 291-300. 38. den Braber, E.T., J.E. de Ruijter, H.T.J. Smits, L.A. Ginsel, A.F. von Recum, and J.A. Jansen. Quantitative analysis of cell proliferation and orientation on substrata with uniform parallel surface micro-grooves. Biomaterials, 1996. 17(11): p. 1093-1099. 39. Walboomers, X.F., J.J.E. Croes, L.A. Ginsel, and J.A. Jansen. Growth behavior of fibroblasts on microgrooved polystyrene. Biomaterials, 1998. 19: p. 1861-1868. 40. den Braber, E.T., J.E. de Ruijter, and J.A. Jansen. The effect of a subcutaneous silicone rubber implant with shallow surface microgrooves on the surrounding tissues in rabbits. Journal of Biomedical Materials Research, 1997. 37: p. 539-547. 41. Walboomers, X.F. and J.A. Jansen. Microgrooved silicone subcutaneous implants in guinea pigs. Biomaterials, 2000. 21: p. 629-636. 42. Walboomers, X.F., H.J.E. Croes, L.A. Ginsel, and J.A. Jansen. Microgrooved subcutaneous implants in the goat. Journal of Biomedical Materials Research, 1998. 42: p. 634-641. 43. Dalton, B.A., X.F. Walboomers, M. Dziegielewski, M.D.M. Evans, S. Taylor, J.A. Jansen, and J.G. Steele. Modulation of epithelial tissue and cell migration by microgrooves. Journal of Biomedical Materials Research, 2001. 56(2): p. 195-207. 44. Rich, A. and A.K. Harris. Anomalous preferences of cultured macrophages for hydrophobic and roughened substrata. Journal of Cell Science, 1981. 50(Aug): p. 1-7.

PAGE 101

89 45. Fujisawa, N., L.A. Poole-Warren, J.C. Woodard, C.D. Bertram, and K. Schindhelm. A novel textured surface for blood-contact. Biomaterials, 1999. 20: p. 955-962. 46. Fujisawa, N., R.A. Odell, L.A. Poole-Warren, C.D. Bertram, J.C. Woodard, and K. Schindhelm. Acute cellular interaction with textured surfaces in blood contact. Journal of Biomedical Materials Research, 2000. 52(3): p. 517-527. 47. Lincks, J., B.D. Boyan, C.R. Blanchard, C.H. Lohmann, Y. Liu, D.L. Cochran, D.D. Dean, and Z. Schwartz. Response of MG63 osteoblast-like cells to titanium and titanium alloy is dependent on surface roughness and composition. Biomaterials, 1998. 19: p. 2219-2232. 48. Turner, A.M.P., N. Dowell, S.W.P. Turner, L. Kam, M. Isaacson, J.N. Turner, H.G. Craighead, and W. Shain. Attachment of astroglial cells to microfabricated pillar arrays of different geometries. Journal of Biomedical Materials Research, 2000. 51: p. 430-441. 49. Tan, J., H. Shen, K.L. Carter, and W.M. Saltzman. Controlling human polymorphonuclear leukocytes motility using microfabrication technology. Journal of Biomedical Materials Research, 2000. 51: p. 694-702. 50. Kaihara, S., J. Borenstein, R. Koka, S. Lalan, E.R. Ochoa, M. Ravens, H. Pien, B. Cunningham, and J.P. Vacanti. Silicon Micromachining to Tissue Engineer Branched Vascular Channels for Liver Fabrication. Tissue Engineering, 2000. 6(2): p. 105-117. 51. Flemming, R.G., C.J. Murphy, G.A. Abrams, S.L. Goodman, and P.F. Nealey. Effects of synthetic microand nano-structured surfaces on cell behavior. Biomaterials, 1999. 20: p. 573-588. 52. Carter, S.B. Haptotactic Islands A Method of confining single cells to study individual cell reactions and clone formation. Experimental Cell Research, 1967. 48(1): p. 189-. 53. Carter, S.B. Haptotaxis and mechanism of cell motility. Nature, 1967. 213(5073): p. 256-. 54. Harris, A. Behavior of cultured cells on substrata of variable adhesiveness. Experimental Cell Research, 1973. 77: p. 285-297. 55. Lopez, G.P., M.W. Albers, S.L. Schreiber, R. Carroll, E. Peralta, and G.M. Whitesides. Convenient Methods for Patterning the Adhesion of Mammalian Cells to Surfaces Using Self-Assembled Monolayers of Alkanethiolates on Gold. Journal of the American Chemical Society, 1993. 115: p. 5877-5878.

PAGE 102

90 56. Britland, S., P. Clark, P. Connolly, and G. Moores. Micropatterned Substratum Adhesiveness: A Model for Morphogenetic Cues Controlling Cell Behavior. Experimental Cell Research, 1992. 198: p. 124-129. 57. McFarland, C.D., C.H. Thomas, C. DeFilippis, J.G. Steele, and K.E. Healy. Protein Adsorption and Cell Attachment to Patterned Surfaces. Journal of Biomedical Materials Research, 2000. 49(2): p. 200-210. 58. Britland, S., H. Morgan, B. Wojiak-Stondart, M. Riehle, A. Curtis, and C. Wilkinson. Synergistic and Hierarchical Adhesive and Topographic Guidance of BHK cells. Experimental Cell Research, 1996. 228: p. 313-325. 59. Folch, A. and M. Toner. Cellular Micropatterns on Biocompatible Materials. Biotechnology Progress, 1998. 14: p. 388-392. 60. Kane, R.S., S. Takayama, E. Ostuni, D.E. Ingber, and G.M. Whitesides. Patterning proteins and cells using soft lithography. Biomaterials, 1999. 20: p. 2363-2376. 61. Folch, A., B.-H. Jo, O. Hurtado, D.J. Beebe, and M. Toner. Microfabricated elastomeric stencils for micropatterning cell cultures. Journal of Biomedical Materials Research, 2000. 52: p. 346-353. 62. Schmid, H. and B. Michel. Siloxane Polymers for High-Resolution, High-Accuracy Soft Lithography. Macromolecules, 2000. 33: p. 3042-3049. 63. Kam, L. and S.G. Boxer. Cell adhesion to protein-micropatterned-supported lipid bilayer membranes. Journal of Biomedical Materials Research, 2001. 55: p. 487-495. 64. Ostuni, E., C.S. Chen, D.E. Ingber, and G.M. Whitesides. Selective deposition of proteins and cells in arrays of microwells. Langmuir, 2001. 17(9): p. 2828-2834. 65. Singhvi, R., A. Kumar, G.P. Lopez, G.N. Stephanopoulos, D.I.C. Wang, G.M. Whitesides, and D.E. Ingber. Engineering Cell Shape and Function. Science, 1994. 264(5159): p. 696-698. 66. Chen, C.S., M. Mrksich, S. Huang, G.M. Whitesides, and D.E. Ingber. Micropatterned Surfaces for Control of Cell Shape, Position, and Function. Biotechnology Progress, 1998. 14(3): p. 356-363. 67. Graham, L.M., D. Whittlesey, and B. Bevacqua. Cardiovascular Implantation, in Biomaterials Science, B.D. Ratner, A.S. Hoffman, F.J. Schoen, and J.E. Lemons, Editors. 1996, San Diego: Academic Press. p. 420-426.

PAGE 103

91 68. Horbett, T.A. Proteins: Structure, Properties, and Adsorption to Surfaces, in Biomaterials Science: An Introduction to Materials in Medicine, B.D. Ratner, A.S. Hoffman, F.J. Schoen, and J.E. Lemons, Editors. 1996, San Diego, CA: Academic Press. p. 133-141. 69. Anderson, J.M. Inflammation, Wound Healing, and the Foreign Body Response, in Biomaterials Science: An Introduction to Materials in Medicine, B.D. Ratner, A.S. Hoffman, F.J. Schoen, and J.E. Lemons, Editors. 1996, San Diego, CA: Academic Press. p. 165-173. 70. Bartzoka, V., M.A. Brook, and M.R. McDermott. Protein-Silicone Interactions: How Compatible are the Two Species? Langmuir, 1998. 14: p. 1887-1891. 71. Vroman, L., A.L. Adams, and M. Klings. Interactions among human blood proteins at interfaces. Federation Proceedings, 1971. 30(5): p. 1494-. 72. Vroman, L., A.L. Adams, G.C. Fischer, and P.C. Munoz. Interaction of high molecular-weight kininogen, factor-XII, and fibrinogen in plasma at interfaces. Blood, 1980. 55(1): p. 156-159. 73. Slack, S.M. and T.A. Horbett. The Vroman effect -A critical review. ACS Symposium Series-Proteins at Interfaces II, 1995. 602: p. 112-128. 74. Marieb, E. Human Anatomy & Physiology. 4 ed. 1998, Menlo Park: Benjamin/Cummings Science Publishing. 75. Seeley, R.R., T.D. Stephens, and P. Tate. Anatomy & Physiology. 4 ed. 1998, Boston: WCB/McGraw-Hill. 76. Bicknell, R. Introduction to the endothelial cell, in Endothelial Cell Culture, R. Bicknell, Editor. 1996, Cambridge: Cambridge University Press. p. 1-6. 77. Parikh, S.A. and E.R. Edelman. Endothelial cell delivery for cardiovascular therapy. Advanced Drug Delivery Reviews, 2000. 42(1-2): p. 139-61. 78. Kochanek, K.D., B.L. Smith, and R.N. Anderson. Deaths:Preliminary Data for 1999 National vital statistics report. 2001, Hyattsville, MD: National Center for Health Statistics, p. 49. 79. Veith, F.J., S.K. Gupta, E. Ascer, S. White-Flores, R.H. Samson, L.A. Scher, J.B. Towne, V.M. Bernhard, P. Bonier, W.R. Flinn, P. Astelford, J.S.T. Yao, and J.J. Bergan. Six year prospective multicenter randomized comparison of autologous saphenous vein and expanded polytetrafluoroethylene grafts in infrainguinal arterial reconstructions. Journal of Vascular Surgery, 1986. 3: p. 104-114.

PAGE 104

92 80. Dedisheim, P. and J.T. Watson. Cardiovascular Applications, in Biomaterials Science: An Introduction to Materials in Medicine, B.D. Ratner, A.S. Hoffman, F.J. Schoen, and J.E. Lemons, Editors. 1996, San Diego, CA: Academic Press. p. 283-297. 81. Graham, L.M., D. Whittlesey, and B. Bevacqua. Cardiovascular Implantation, in Biomaterials Science: An Introduction to Materials in Medicine, B.D. Ratner, A.S. Hoffman, F.J. Schoen, and J.E. Lemons, Editors. 1996, San Diego, CA: Academic Press. p. 420-426. 82. Bernex, F., J.P. Mazzucotelli, J.L. Roudiere, N. Benhaiem-Sigaux, J. Leandri, and D. Loisance. In vitro endothelialization of carbon-coated Dacron vascular grafts. The International Journal of Artificial Organs, 1992. 15(3): p. 172-180. 83. Bhat, V.D., B. Klitzman, K. Koger, G.A. Truskey, and W.M. Reichert. Improving endothelial cell adhesion to vascular graft surfaces: Clinical need and strategies. Journal of Biomaterials Science -Polymer Edition, 1998. 9(11): p. 1117-1135. 84. Laube, H.R., J. Duwe, W. Rutsch, and W. Konertz. Clinical experience with autologous endothelial cell-seeded polytetrafluoroethylene coronary artery bypass grafts. Journal of Thoracic and Cardiovascular Surgery, 2000. 120(1): p. 134-141. 85. Miwa, H., T. Matsuda, K. Kondo, N. Tani, Y. Fukaya, M. Morimoto, and F. Iida. Improved Patency of an Elastomeric Vascular Graft by Hybridization. ASAIO Journal, 1992. 38(3): p. M512-M515. 86. Shinoka, T., D. Shum-Tim, P.X. Ma, R.E. Tanel, N. Isogai, R. Langer, J.P. Vacanti, and J. Mayer, J.E. Creation of Viable Pulmonary Artery Autografts Through Tissue Engineering. Journal of Thoracic and Cardiovascular Surgery, 1998. 115(3): p. 536-545. 87. Balaban, N.Q., U.S. Schwarz, D. Riveline, P. Goichberg, G. Tzur, I. Sabanay, D. Mahalu, S. Safran, A. Bershadsky, L. Addadi, and B. Geiger. Force and focal adhesion assembly: a close relationship studied using elastic micropatterned substrates. Nature Cell Biology, 2001. 3: p. 466-472. 88. Chandy, T., G.S. Das, R.F. Wilson, and G.H.R. Rao. Use of plasma glow for surface-engineering biomolecules to enhance bloodcompatibility of Dacron and PTFE vascular prosthesis. Biomaterials, 2000. 21: p. 699-712. 89. Walpoth, B.H., R. Rogulenko, E. Tihvinskaia, S. Gogolewski, T. Schaffner, O.M. Hess, and U. Althaus. Improvement of patency rate in heparin-coated small synthetic vascular grafts. Circulation, 1998. 98(19): p. Suppl: II319-23; discussion II324.

PAGE 105

93 90. Patterson, R.B., A. Messier, and R.F. Valentini. Effects of Radiofrequency Glow Discharge and Oligopeptides on the Attachment of Human Endothelial Cells to Polyurethane. ASAIO Journal, 1995. 41(3): p. M625-M629. 91. Ferrari, S. and B.D. Ratner. ToF-SIMS quantification of albumin adsorbed on plasma-deposited fluoropolymers by partial least-squares regression. Surface and Interface Analysis, 2000. 29(12): p. 837-844. 92. Klee, D. and H. Hocker. Polymers for biomedical applications: Improvement of the interface compatibility, in Advances in Polymer Science. 1999. p. 1-57. 93. Qian, W.P., B. Xu, D.F. Yao, Y.H. Lin, L. Wu, C.X. Wang, F. Yu, Z.H. Lu, and Y. Wei. Site-directed immobilization of immunoglobulin G on 3-aminopropyltriethoxysilane modified silicon wafer surfaces. Materials Science & Engineering C -Biomimetic and Supramolecular Systems, 1999. 8-9: p. 475-480. 94. Hubbell, J.A., S.P. Massia, N.P. Desai, and P.D. Drumheller. Endothelial cell-selective materials for tissue engineering in the vascular graft via a new receptor. Biotechnology, 1991. 9: p. 568-572. 95. Massia, S.P. and J.A. Hubbell. Vascular endothelial cell adhesion and spreading promoted by the peptide REDV of teh IIICS region of plasma fibronectin is mediated by integrin alpha 4 beta 1. The Journal of Biological Chemistry, 1992. 267(20): p. 14019-14026. 96. Harris, A.K., P. Wild, and D. Stopak. Silicone Rubber Substrata: A New Wrinkle in the Study of Cell Locomotion. Science, 1980. 208(4440): p. 177-179. 97. Lee, J., M. Leonard, T. Oliver, A. Ishihara, and K. Jacobson. Traction forces generated by locomoting keratocytes. The Journal of Cell Biology, 1994. 127: p. 1957-1964. 98. Pelham, R.J. and Y.L. Wang. Cell Locomotion and Focal Adhesions are Regulated by Substrate Flexibility. Proceedings of the National Academy of Sciences of the United States of America, 1997. 94(25): p. 13661-13665. 99. Dembo, M. and Y.L. Wang. Stresses at the cell-to-substrate interface during locomotion of fibroblasts. Biophysical Journal, 1999. 76: p. 2307-2316. 100. Galbraith, C.G. and M.P. Sheetz. A micromachined device provides a new bend on fibroblast traction forces. Proceedings of the National Academy of Sciences of the United States of America, 1997. 94: p. 9114-9118.

PAGE 106

94 101. Chou, L., J.D. Firth, V.J. Uitto, and D.M. Brunette. Substratum surface topography alters cell shape and regulates fibronectin mRNA level, mRNA stability, secretion and assembly in human fibroblasts. Journal of Cell Science, 1995. 108: p. 1563-1573. 102. Kato, S., J. Ando, and T. Matsuda. mRNA expression on shape-engineered endothelial cells: Adhesion molecules ICAM-1 and VCAM-1. Journal of Biomedical Materials Research, 2001. 54: p. 366-372. 103. Chicurel, M.E., C.S. Chen, and D.E. Ingber. Cellular control lies in the balance of forces. Current Opinion in Cell Biology, 1998. 10: p. 232-239. 104. Wang, J.H., E.S. Grood, J. Florer, and R. Wenstrup. Alignment and proliferation of MC3T3-E1 osteoblasts in microgrooved silicone substrata subjected to cyclic stretching. Journal of Biomechanics, 2000. 33: p. 729-735. 105. Maniotis, A.J., C.S. Chen, and D.E. Ingber. Demonstration of mechanical connections between integrins, cytoskeletal filaments, and nucleoplasm that stabilize nuclear structure. Proceedings of the National Academy of Sciences of the United States of America, 1997. 94: p. 849-854. 106. Arkles, B. Look what you can make out of silicones. Chemtech, 1983. 13: p. 542-555. 107. Gerszten, P.C. A formal risk assessment of silicone breast implants. Biomaterials, 1999. 20(11): p. 1063-1069. 108. Rosato, D.V. Polymers, processes and properties of medical plastics: including markets and applications, in Biocompatible Polymers, Metals, and Composites, M. Szycher, Editor. 1983, Lancaster, PA: Technomic Publ. p. 1019-1067. 109. Visser, S.A., R.W. Hergenrother, and S.L. Cooper. Polymers, in Biomaterials Science: An Introduction to Materials in Medicine, B.D. Ratner, A.S. Hoffman, F.J. Schoen, and J.E. Lemons, Editors. 1996, San Diego, CA: Academic Press. p. 50-60. 110. Quinn, K.J. and J.M. Courtney. Silicones as Biomaterials. British Polymer Journal, 1988. 20: p. 25-32. 111. Widenhouse, C.W. Surface Modification of Vascular Prostheses and Intracorneal Lens Polymers [dissertation]. Gainesville: University of Florida; 1996. 112. Callow, M.E., J.A. Callow, J.D. Pickett-Heaps, and R. Wetherbee. Primary adhesion of Enteromorpha (Chlorophyta, Ulvales) propagules: Quantitative settlement studies and video microscopy. Journal of Phycology, 1997. 33(6): p. 938-947.

PAGE 107

95 113. Stanley, M.S., M.E. Callow, and J.A. Callow. Monoclonal antibodies to adhesive cell coat glycoproteins secreted by zoospores of the green alga Enteromorpha. Planta, 1999. 210(1): p. 61-71. 114. Schultz, M.P., C.J. Kavanagh, and G.W. Swain. Hydrodynamic forces on barnacles: Implications on detachment from fouling-release surfaces. Biofouling, 1999. 13(4): p. 323-335. 115. Swain, G.W., W.G. Nelson, and S. Preedeekanit. The influence of biofouling adhesion and biotic disturbance on the development of fouling communitites on non-toxic surfaces. Biofouling, 1998. 12(1-3): p. 257-269. 116. Hillborg, H., M. Sandelin, and U.W. Gedde. Hydrophobic recovery of polydimethylsiloxane after exposure to partial discharges as a function of crosslink density. Polymer, 2001. 42: p. 7349-7362. 117. Toth, A., I. Bertoti, M. Blazso, G. Banhegyi, A. Bognar, and P. Szaplonczay. Oxidative damage and recovery of silicone rubber surfaces. 1. X-ray photoelectron spectroscopic study. Journal of Applied Polymer Science, 1994. 52(9): p. 1293-1307. 118. Zisman, W.A. Contact Angle: Wettability and Adhesion. 1964. Washington, D.C.: American Chemical Society. 119. Schmidt, J.A. and A.F. von Recum. Surface characterization of microtextured silicone. Biomaterials, 1992. 13(10): p. 675-81. 120. Sperling, L.H. Introduction to Physical Polymer Science. 2 ed. 1992, New York: Wiley-Interscience. 594. 121. Mergerman, J., J.E. Hasson, D.F. Warnock, G.J. L'Italien, and W.M. Abbott. Noninvasive measurements of nonlinear arterial elasticity. American Journal of Physiology, 1986. 250: p. H181-188. 122. Kendall, K. The adhesion and surface energy of elastic solids. Journal of Physics D -Applied Physics, 1971. 4(8): p. 1186-1195. 123. Good, R.J. Contact angle, wetting and adhesion: a critical review. Journal of Adhesion Science and Technology, 1992. 6(12): p. 1269-1302. 124. CRC handbook of chemistry and physics. 78 ed. 1998, Cleveland, Ohio: CRC Press. 125. Johnson, R.E., Jr. and R.H. Dettre. Wettability and Contact Angles. Surface and Colloid Science, 1969. 2: p. 85-153.

PAGE 108

96 126. Patel, J.M., D.A. Edwards, E.R. Block, and M.K. Raizada. Effect of nitrogen dioxide on surface membrane fluidity and insulin receptor binding of pulmonary endothelial cells. Biochem. Pharmacol., 1988. 37(8): p. 1497-1507.

PAGE 109

BIOGRAPHICAL SKETCH Wade Richard Wilkerson was born on July 7, 1976, in Jacksonville, FL, to Dick and Nancy Wilkerson. He attended The Bolles School in Jacksonville where he received his high school diploma in 1994. In the Fall of 1994, Wade headed north to the University of Virginia, where he studied chemical engineering and received his Bachelor of Science degree in the Spring of 1998. His senior thesis involved the design of a testing apparatus for magnetic bearings in a vascular assist device. After graduating, Wade moved to Greenville, NC, and began work at Metrics, Inc. There he gained valuable experience in analytical chemistry and laboratory techniques under GLP and c-GMP guidelines. After one year in Greenville, Wade moved to Gainesville to begin his graduate work in Dr. Anthony Brennans group as part of the Biomedical Engineering Program. During this time, he married Laura Holcomb on June 24, 2000. After graduation, Wade will begin a new phase of his education while studying to become a medical doctor at Wake Forest University Medical School in Winston-Salem, NC. 97


xml version 1.0 encoding UTF-8
METS:mets LABEL CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS OBJID UFE0000364 TYPE monograph xmlns:METS http:www.loc.govMETS xmlns:daitss http:www.fcla.edudlsmddaitss xmlns:dc http:purl.orgdcelements1.1 xmlns:mods http:www.loc.govmodsv3 xmlns:palmm http:www.fcla.edudlsmdpalmm xmlns:rightsmd http:www.fcla.edudlsmdrightsmd xmlns:techmd http:www.fcla.edudlsmdtechmd xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation ..E20051012_AAAABMlinks_20051012234823www.loc.govstandardsmetsmets_LOC.xsd ..E20051012_AAAABMlinks_20051012234823dublincore.orgschemasxmlssimpledc20021212_LOC.xsd ..E20051013_AAAAAKlinks_20051013064438www.loc.govstandardsmodsv3mods-3-0_LOC.xsd ..E20051012_AAAABMlinks_20051012234823www.fcla.edudlsmdtechmd_LOC.xsd ..E20060621_AAAELKlinks_20060621194313www.fcla.edudlsmdpalmm_LOC.xsd ..E20051012_AAAABMlinks_20051012234823www.fcla.edudlsmdrightsmd_LOC.xsd ..E20060502_AAACYYlinks_20060502001940www.fcla.edudlsmddaitssdaitss_LOC.xsd
METS:metsHdr CREATEDATE 2002-04-05T00:00:00Z ID LASTMODDATE 2006-09-14T16:33:20Z RECORDSTATUS NEW
METS:agent OTHERROLE MXF CREATOR ROLE OTHER ORGANIZATION
METS:name FCLA
METS:note directory=L:\Common 1\Data\UFE_2001_fall\UFE0000364\
makerules=etd
server=TD
formats=application/pdf
projects=ETD
OTHERTYPE SOFTWARE
MXFClient
INDIVIDUAL
emh
METS:dmdSec DMD1
METS:mdWrap MDTYPE MODS MIMETYPE textxml
METS:xmlData
mods:mods
mods:titleInfo
mods:title CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE OF ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS
mods:name
mods:namePart WILKERSON, WADE R
mods:role
mods:roleTerm type text creator
Anthony B. Brennan
contributor
mods:originInfo
mods:publisher University of Florida
mods:dateIssued 2001
20011215
mods:language
mods:languageTerm English
mods:abstract Contact guidance is a term used to describe a material's ability to direct the alignment and growth patterns of biological cells or tissue. It has long been understood that the surface a cell grows on impacts its size, shape, and metabolism. Typically, when cells are exposed to mechanical features such as ridges and grooves, the cells align and travel along the length. Surface chemistry plays a significant role in the attachment of cells to a substrate and in their movement on that surface.
The objective of this study was to study the effect of modulus as well as surface texture dimensions on vascular endothelial cells (ECs). To examine the effects of contact guidance on silicone elastomers, microtextured substrates were produced with reproducible and well-defined surfaces. Ridges of 10,000 �m length were fabricated at 3 different widths: 5 �m, 10 �m, and 20 �m, separated by 5 �m wide grooves to determine the effect of separation of features on the alignment of porcine vascular endothelial cells (PVECs). Two depths were examined: 5 �m and 1.5 �m.
Elastomer samples were examined with contact angles to determine their relative wettability and surface free energy. Formulations of elastomer with both functionalized and non-functionalized PDMS oligomer additives to alter the modulus were examined by contact angle, with no significant difference in surface energy. Surfaces were treated with fibronectin and radiofrequency glow discharge plasma in argon for 5 minutes at 50 W. Both treatments significantly increased the hydrophilicity after treatment, as measured by captive bubble contact angles. Dynamic contact angle analysis of textured surfaces showed a difference in smooth and textured areas as well.
Contact guidance of PVECs on textured silicone elastomers was measured by the nuclear form factor, in which the log of the ratio of nuclear length to width was presented. Results demonstrated that as the ridge width decreased from 20 �m to 5 �m contact guidance increased, as well as when the depth of the grooves increased from 1.5 �m to 5 �m. Data analysis showed that the groove depth was the most important factor in nuclear alignment. Contact guidance on fibronectin-coated elastomers was examined to determine the effect of modulus. It was expected that higher modulus materials would increase the effect of contact guidance. Elastic modulus on 4 elastomers was measured by tensile tests and resulted in a range of values from 0.3 MPa to 2.3 MPa. There was no significant difference in the contact guidance on the deep 5 �m grooves with varying modulus. The 1.5 �m deep grooves showed a significant increase in the alignment of cells to the groove in the highest modulus material compared to the lowest modulus material for the 5 �m and 10 �m wide ridges. The conclusion to be taken from these data is that modulus does seem to play a role in the determination of contact guidance, but other factors such as groove width and especially depth are more significant.
mods:subject
mods:topic contact guidance, microtexture, endothelial cell, PDMS, silicone, biomaterial
mods:accessCondition useAndReproduction Public
METS:amdSec
METS:rightsMD RMD1
OTHERMDTYPE RIGHTSMD
rightsmd:versionStatement Electronic version created 2002, State University Sytem of Florida.
METS:sourceMD SMD1
PALMM
palmm:entityDesc SOURCE UF
METS:digiprovMD DPMD1
DAITSS
daitss:daitss
daitss:AGREEMENT_INFO ACCOUNT PROJECT ETD
METS:fileSec
METS:fileGrp
METS:file CHECKSUM 385b65485133bb22589be77b57e3643f CHECKSUMTYPE MD5 CREATED 2002-03-06T13:38:33Z GROUPID GID1 FID1 applicationpdf SIZE 2377657
METS:FLocat LOCTYPE OTHERLOCTYPE SYSTEM xlink:href MASTER.pdf
METS:structMap
METS:div ADMID DMDID
main
file
METS:fptr FILEID


xml version 1.0 encoding ISO-8859-1
METS:mets LABEL CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS OBJID UFE0000364 TYPE monograph xmlns:METS http:www.loc.govMETS xmlns:daitss http:www.fcla.edudlsmddaitss xmlns:dc http:purl.orgdcelements1.1 xmlns:mods http:www.loc.govmodsv3 xmlns:palmm http:www.fcla.edudlsmdpalmm xmlns:rightsmd http:www.fcla.edudlsmdrightsmd xmlns:techmd http:www.fcla.edudlsmdtechmd xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmetsversion14mets.xsd http:dublincore.orgschemasxmlssimpledc20021212.xsd http:www.loc.govstandardsmodsv3mods-3-0.xsd http:www.fcla.edudlsmdtechmd.xsd http:www.fcla.edudlsmdpalmm.xsd http:www.fcla.edudlsmdrightsmd.xsd http:www.fcla.edudlsmddaitssdaitss.xsd
METS:metsHdr CREATEDATE 2002-04-05T00:00:00Z ID LASTMODDATE 2006-09-14T16:33:20Z RECORDSTATUS NEW
METS:agent OTHERROLE MXF CREATOR ROLE OTHER ORGANIZATION
METS:name FCLA
METS:note directory=L:\Common 1\Data\UFE_2001_fall\UFE0000364\
makerules=etd
server=TD
formats=application/pdf
projects=ETD
OTHERTYPE SOFTWARE
MXFClient
INDIVIDUAL
emh
METS:dmdSec DMD1
METS:mdWrap MDTYPE MODS MIMETYPE textxml
METS:xmlData
mods:mods
mods:titleInfo
mods:title CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE OF ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS
mods:name
mods:namePart WILKERSON, WADE R
mods:role
mods:roleTerm type text creator
Anthony B. Brennan
contributor
mods:originInfo
mods:publisher University of Florida
mods:dateIssued 2001
20011215
mods:language
mods:languageTerm English
mods:abstract Contact guidance is a term used to describe a material's ability to direct the alignment and growth patterns of biological cells or tissue. It has long been understood that the surface a cell grows on impacts its size, shape, and metabolism. Typically, when cells are exposed to mechanical features such as ridges and grooves, the cells align and travel along the length. Surface chemistry plays a significant role in the attachment of cells to a substrate and in their movement on that surface.
The objective of this study was to study the effect of modulus as well as surface texture dimensions on vascular endothelial cells (ECs). To examine the effects of contact guidance on silicone elastomers, microtextured substrates were produced with reproducible and well-defined surfaces. Ridges of 10,000 m length were fabricated at 3 different widths: 5 m, 10 m, and 20 m, separated by 5 m wide grooves to determine the effect of separation of features on the alignment of porcine vascular endothelial cells (PVECs). Two depths were examined: 5 m and 1.5 m.
Elastomer samples were examined with contact angles to determine their relative wettability and surface free energy. Formulations of elastomer with both functionalized and non-functionalized PDMS oligomer additives to alter the modulus were examined by contact angle, with no significant difference in surface energy. Surfaces were treated with fibronectin and radiofrequency glow discharge plasma in argon for 5 minutes at 50 W. Both treatments significantly increased the hydrophilicity after treatment, as measured by captive bubble contact angles. Dynamic contact angle analysis of textured surfaces showed a difference in smooth and textured areas as well.
Contact guidance of PVECs on textured silicone elastomers was measured by the nuclear form factor, in which the log of the ratio of nuclear length to width was presented. Results demonstrated that as the ridge width decreased from 20 m to 5 m contact guidance increased, as well as when the depth of the grooves increased from 1.5 m to 5 m. Data analysis showed that the groove depth was the most important factor in nuclear alignment. Contact guidance on fibronectin-coated elastomers was examined to determine the effect of modulus. It was expected that higher modulus materials would increase the effect of contact guidance. Elastic modulus on 4 elastomers was measured by tensile tests and resulted in a range of values from 0.3 MPa to 2.3 MPa. There was no significant difference in the contact guidance on the deep 5 m grooves with varying modulus. The 1.5 m deep grooves showed a significant increase in the alignment of cells to the groove in the highest modulus material compared to the lowest modulus material for the 5 m and 10 m wide ridges. The conclusion to be taken from these data is that modulus does seem to play a role in the determination of contact guidance, but other factors such as groove width and especially depth are more significant.
mods:subject
mods:topic contact guidance, microtexture, endothelial cell, PDMS, silicone, biomaterial
mods:accessCondition useAndReproduction Public
METS:amdSec
METS:rightsMD RMD1
OTHERMDTYPE RIGHTSMD
rightsmd:versionStatement Electronic version created 2002, State University Sytem of Florida.
METS:sourceMD SMD1
PALMM
palmm:entityDesc SOURCE UF
METS:digiprovMD DPMD1
DAITSS
daitss:daitss
daitss:AGREEMENT_INFO ACCOUNT PROJECT ETD
METS:fileSec
METS:fileGrp
METS:file CHECKSUM 385b65485133bb22589be77b57e3643f CHECKSUMTYPE MD5 CREATED 2002-03-06T13:38:33Z GROUPID GID1 FID1 applicationpdf SIZE 2377657
METS:FLocat LOCTYPE OTHERLOCTYPE SYSTEM xlink:href MASTER.pdf
METS:structMap
METS:div ADMID DMDID
main
file
METS:fptr FILEID


Permanent Link: http://ufdc.ufl.edu/UFE0000364/00001

Material Information

Title: Contribution of modulus to the contact guidance of endothelial cells on microtextured siloxane elastomers
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000364:00001

Permanent Link: http://ufdc.ufl.edu/UFE0000364/00001

Material Information

Title: Contribution of modulus to the contact guidance of endothelial cells on microtextured siloxane elastomers
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0000364:00001


This item has the following downloads:


Full Text











CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE OF
ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS


















By

WADE RICHARD WILKERSON


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2001




























Copyright 2001

by

Wade Richard Wilkerson



























This thesis is dedicated to my family and friends, and in loving memory of James R.
Wilkerson and James S. Blair.















ACKNOWLEDGMENTS

I would like to express my deepest gratitude to my advisor and committee

chairman, Dr. Anthony Brennan, for his understanding, advice, and support during my

time at the University of Florida. I would also like to thank the other members of my

supervisory committee, Dr. Christopher Batich, and Dr. Ronald Baney, for their counsel

and accommodation during this process. I am especially grateful to Dr. Laurie Gower,

for agreeing to be a substitute member of my committee at such late notice.

The members of Dr. Brennan's research group have been a great source of advice,

experience and camaraderie over the past two years. Specifically I would like to thank

Jeanne Macdonald for taking responsibility of the lab and serving as group mentor and

general well of knowledge. Licheng Zhao and Jeremy Mehlem both were valuable

sources of experience and information. I would like to thank Adam Feinberg for his

expertise with the AFM, SEM, and computer issues, as well as for his friendship. Chuck

Seegert provided the wafers and the nuclear form method for this study, including a

wealth of information on statistics, cell methods, and staining protocols.

Amy Gibson and Leslie Wilson provided the modulus data and the formulations

for the modified elastomer samples. Their hard work and flexibility are greatly

appreciated. Other members of the Brennan research group, Clay Bohn, Brian Hatcher,

and Nikhil Kothurkar, have been an invaluable source of friendship and support. Paul

Martin was always accommodating with regard to cell culture techniques and advice that

was crucial to the completion of this work. I would also like to thank many other









graduate students in the Materials Science department and the Biomedical Engineering

program, specifically Brad Willenberg, Brian Cuevas, Bob Hadba, Josh Stopek, Dan

Urbaniak, and Jamie Rhodes.

I am indebted to Dr. C. Keith Ozaki and the members of his research group,

specifically Zaher Abouhamze. Zaher was always helpful and went above and beyond

the call of duty in assisting me at various times during this project. Dr. Edward Block

graciously supplied the endothelial cells used in this study, and Bert Herrara was very

generous and patient in supplying cells on a weekly basis. Nina Klingmann and Dr. Tina

Lam were also a great source of information and I am grateful to Nina for allowing use of

her incubators and cell culture equipment.

I would be remiss if I did not thank my family and friends who have supported me

throughout this process and long before. My parents have provided me with every

opportunity to succeed, and without their love and support, I could never achieve the

goals that I have set. My wife, Laura, has been the greatest blessing and greatest friend

throughout our relationship and especially during the first year of our marriage. Her

understanding and support has no bounds, and her love has kept me going whenever I

began to doubt. Most of all I would like to thank God for without Him, none of this

would be possible.
















TABLE OF CONTENTS

Page

A C K N O W L E D G M E N T S ................................................................................................. iv

L IS T O F T A B L E S ........ ............................................................... .......... .. ............. ...... v iii

LIST OF FIGURES ......... ......................... ...... ........ ............ ix

ABSTRACT .............. ................. .......... .............. xi

CHAPTERS

1 IN TR OD U CTION ....................................... ...... .. ........ .... .............. .

2 B A C K G R O U N D ............... ......... .............................. ........ ........ ................ .4

C contact G u id an ce ....................................................................... 4
Surface Topography. ..................................... ............. 4
Su rface C h em istry .................................................... .. ... .......... .. ................ 11
Endothelial Cells.................................. .............. 14
Vascular Grafts ......................................... 16
C ell A dhesion............................ .................... 19
Silicones...................... ....................... .............. 22
Use as a Biomaterial ........................................ 23
Su rface E n ergy ................................................................. 2 5
M echanical Properties............................. .............. 27

3 CHARACTERIZATION OF SILICONE ELASTOMER SUBSTRATES....................31

In tro d u ctio n ....................................................................... 3 1
M materials ..................................... .................... 31
M e th o d s................................................................................................... .............. ...... 3 3
Elastomer Preparation ..... ........... ........ .......... ........33
M odulus D eterm nation ........................................................ ..................... 34
Preparation of Textured Surfaces............................................... .................... 35
Surface Treatm ent .................................................... 36
Radiofrequency glow discharge treatment.............................. .................... 37
Fibronectin adsorption .............. ......... ................ 37
Surface Energy .................................. .............. 38
D ynam ic Contact A ngles ...................... ..................... .. ........................... 38









O p tical P ro fi lo m etry ................................................................................................. 3 9
Results and Discussion ........................................... ................ 40
M odulu s R esults................................. .. ......... ... ....... .... ............ 40
Surface Energy of Silicone Elastom ers............................................ ....... ...... 41
Contact Angles of Treated Surfaces ........................................ ....... .............. 44
Plasm a Treatm ent Issues............................ ................. .... ........ .............. 45
D ynam ic Contact A ngle..................... ....................................... .......................... 49
O optical P rofi lom etry ................................................................................................. 52

4 CONTACT GUIDANCE OF ENDOTHELIAL CELLS ............................................57

In tro d u ctio n ......................................................... ............... 5 7
M materials ...................................... ........ .............. 57
Elastom er Substrates ........... ...... ......... .. .. .............................. 57
C ell C u ltu re an d S eedin g .......................................................................................... 5 8
M e th o d s........................................................................................................... .. ........ 5 9
E lastom er Sam ple P reparation ............................................................ .............. 59
Surface Treatm ent by Fibronectin ........................................ .......... .............. 59
Surface Treatm ent by RFG D Plasm a................................................. .. .................. 59
Cell Culture Techniques .. ..... .......................................... ...... .............. 60
C ell passage procedure ......................................................... ................... 60
Determination of cell suspension concentration............................................... 61
C ell S eedin g on S am p les .......................................................................................... 6 1
C ell Staining and Im age C apture .................................................................. ...... 62
Image Analysis......................................... ............ 63
Results and Discussion ............................................................... ............ 65
Contact Guidance on Textured Surfaces.......................... ..... .............. 65
Contact Guidance on Textured Surfaces of Varying Modulus............................. 72

5 CONCLUSIONS AND FUTURE WORK ........................................80

C o n c lu sio n s ......................................................... ............... 8 0
M icrotextured Surfaces ................................................... ......... .............. 80
Surface Energy and Treatment......................... ................... .. .............. 81
Contact Guidance on Textured Elastom ers.......................................... ... ................. 82
F u tu re W o rk ......................................... .. .............................................. 8 3
Surface Treatment ................ ............. ....................... ..... ........ .. 83
Topographical D design ............. ..... ......... ..... ............... ................. 83
C e ll S tu d ie s ...................... .. ............. .. .................................................. 8 4

LIST OF REFEREN CES .................................................................... ............... 85

B IO G R A PH IC A L SK E TCH ..................................................................... ..................97
















LIST OF TABLES


Table Page

2.1 Enrichment of proteins adsorbed on polyethylene exposed to blood plasma...........14

3.1 Typical properties of Silastic T-2 Silicone Moldmaking Rubber (from Dow
Corning product inform ation sheet).................................. ..................................... 32

3.2 PDMS additives to silicone elastomer system ......................................................33

3.3 Curing conditions for silicone elastomer samples ......................................... 34

3.4 Surface tension of liquids used for surface energy determination by contact angle
a n a ly sis ........................................................ ................ 4 2

3.5 Contact angles and surface free energy of various substrates...............................43

3.6 Captive bubble contact angles on treated surfaces ............................................. 44

3.7 Water contact angles on plasma treated silicone samples after exposure to air over
tim e .............................................................................. 4 8

3.8 DCA data on silicone elastomer modified with non-functionalized PDMS
o lig o m e rs...................................................... ................ 5 1

3.9 Optical profilometer data of ridge widths and groove depths of the 1.5 and 5 [tm
deep wafers and elastomer copies. The wafers should have a constant ridge width
and the elastomer copies have ridge widths of 5 [tm, 10 [tm, and 20 m ................55

4.1 Properties of elastomeric substrates for contact guidance experiments ..................58















LIST OF FIGURES


Figure Page

2.1 Light micrographs and measurement protocols of 6 mm diameter punches of
corneal epithelial tissue explants (f) over (A & C) smooth PS surface and (B & D) a
surface with 1 [m deep microgrooves separated by 1 [m ......................................9

2.2 Cutaway view of an artery showing the three main layers.................. .............15

2.3 Light microscopy image of confluent porcine vascular endothelial cells (PVECs)
grown on tissue culture polystyrene (Image taken by L. Zhao and W. Wilkerson) .16

2.4 Polydimethylsiloxane, trimethylsiloxy terminated.................................................22

3.1 -Representation of etched patterns on silicon wafers..............................................35

3.2 Sample layout for textured substrates examined using DCA .................................39

3.3 Elastic modulus of modified silicones as measured by tensile testing ...................41

3.4 Zisman plot of unmodified silicone for calculation of surface free energy.............44

3.5 Profilometer image of surface damage due to plasma of 15% vinyl tris textured
elastomer sample (smooth area) ................. .. ................. .. ............... 46

3.6 Profilometer image of surface damage due to plasma of 15% vinyl tris textured
elastom er sam ple ................................................ ................. 47

3.7 AFM Image of crack on plasma treated sample ....................................... .......... 47

3.8 Profilometry data for non-treated (A) and plasma treated (B) 1.5 [im deep elastomer
textured surfaces at the 5 ism width spacing .......................................................48

3.9 Force-distance curve from DCA of unmodified silicone elastomer in water...........50

3.10 Force distance curve from DCA on a textured silicone elastomer substrate ..........52

3.11 3-D images of 5 [m deep elastomer samples copied off of epoxy taken with
optical profilometer (50X). (A-C) are examples of the ridges at the different
spacing. (D) demonstrates the smooth/textured interface ....................................53









3.12 3-D images of 5 .im deep etched silicon wafers at each width spacing, taken with
optical profilom eter ........ .... ...................... .... .............. .. ........ .... 54

3.13 3-D images of 5% LMW elastomer copies (A, B) of a 1.5 tmm wafer (C, D).........55

4.1 Representation of the values measured in the nuclear form factor.........................63

4.2 Steps in image processing technique of hematoxylin stained nuclei. The original
picture (A) is contrasted (B), rotated to align the textures (C), and then measured for
length (D) and width (not shown) ............................ .. .... .. .................64

4.3 Main Effects Plot of the data means of fibronectin coated unmodified elastomers at
various ridge widths (Feature) and groove depths (Depth) ....................................67

4.4 Interaction plot representing the change in contact guidance with depth and feature
width for fibronectin coated silicone elastomer.................................................. 68

4.5 Main Effects Plot of the data means for plasma treated unmodified elastomers......69

4.6 Interaction plot representing the change in contact guidance with depth and feature
width for plasma treated silicone elastomer.............. ...............................................70

4.7 Comparison of surface treatments on textured unmodified silicone elastomer by
ridge width and groove depth (error bars represent standard error of mean) ...........71

4.8 PVECs grown on untextured fibronectin coated LMW sample stained with
h em ato x y lin .................................................... ................ 7 2

4.9 PVECs grown on 5 [tm spacing 1.5 [tm deep untreated LMW stained with
h em ato x y lin .................................................... ................ 7 2

4.10 Main effects plot comparing log (L/W) to feature width, feature depth, and
material used on fibronectin-coated elastomers..............................................73

4.11 Interaction plot representing the change in factors of PVECs grown on fibronectin
coated m materials. .....................................................................73

4.12 Comparison of materials on textured unmodified silicone elastomer by ridge width
on 5 tm deep grooves (error bars represent s.e.m.).............................. ........ 75

4.13 Comparison of materials on textured unmodified silicone elastomer by ridge width
on 5 tm deep grooves (error bars represent s.e.m.).............................. ........ 75

4.14 Comparison of ridge width and groove depth to the high modulus material (LMW)
and the low modulus material (Tris) (error bars represent s.e.m.)....................76















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

CONTRIBUTION OF MODULUS TO THE CONTACT GUIDANCE OF
ENDOTHELIAL CELLS ON MICROTEXTURED SILOXANE ELASTOMERS

By

Wade Richard Wilkerson

December, 2001

Chairman: Dr. Anthony B. Brennan
Major Department: Biomedical Engineering Program

Contact guidance is a term used to describe a material's ability to direct the

alignment and growth patterns of biological cells or tissue. It has long been understood

that the surface a cell grows on impacts its size, shape, and metabolism. Typically, when

cells are exposed to mechanical features such as ridges and grooves, the cells align and

travel along the length. Surface chemistry plays a significant role in the attachment of

cells to a substrate and in their movement on that surface.

The objective of this study was to study the effect of modulus as well as surface

texture dimensions on vascular endothelial cells (ECs). To examine the effects of contact

guidance on silicone elastomers, microtextured substrates were produced with

reproducible and well-defined surfaces. Ridges of 10,000 [im length were fabricated at 3

different widths: 5 itm, 10 itm, and 20 itm, separated by 5 tm wide grooves to determine

the effect of separation of features on the alignment of porcine vascular endothelial cells

(PVECs). Two depths were examined: 5 [m and 1.5 im.









Elastomer samples were examined with contact angles to determine their relative

wettability and surface free energy. Formulations of elastomer with both functionalized

and non-functionalized PDMS oligomer additives to alter the modulus were examined by

contact angle, with no significant difference in surface energy. Surfaces were treated

with fibronectin and radiofrequency glow discharge plasma in argon for 5 minutes at 50

W. Both treatments significantly increased the hydrophilicity after treatment, as

measured by captive bubble contact angles. Dynamic contact angle analysis of textured

surfaces showed a difference in smooth and textured areas as well.

Contact guidance of PVECs on textured silicone elastomers was measured by the

nuclear form factor, in which the log of the ratio of nuclear length to width was

presented. Results demonstrated that as the ridge width decreased from 20 .im to 5 .im

contact guidance increased, as well as when the depth of the grooves increased from 1.5

lm to 5 rm. Data analysis showed that the groove depth was the most important factor

in nuclear alignment. Contact guidance on fibronectin-coated elastomers was examined

to determine the effect of modulus. It was expected that higher modulus materials would

increase the effect of contact guidance. Elastic modulus on 4 elastomers was measured

by tensile tests and resulted in a range of values from 0.3 MPa to 2.3 MPa. There was no

significant difference in the contact guidance on the deep 5 [m grooves with varying

modulus. The 1.5 rm deep grooves showed a significant increase in the alignment of

cells to the groove in the highest modulus material compared to the lowest modulus

material for the 5 rm and 10 rm wide ridges. The conclusion to be taken from these data

is that modulus does seem to play a role in the determination of contact guidance, but

other factors such as groove width and especially depth are more significant.














CHAPTER 1
INTRODUCTION

The ability to predict and control a biological response to a biomedical device

would be a skill of dramatic technological and economic importance. Since man first

attempted to replace nature's mechanisms and structures with artificial substitutes, he has

met mainly with frustration marked with varying degrees of success. For implant

materials, the factors that determine success are many, but the interaction between the

surface and the surrounding tissue is one of the most important. The characteristics of

that surface shape that interaction, and their secrets are slowly becoming known.

Contact guidance is a term used to describe a material's ability to direct the

alignment and growth patterns of biological cells or tissue. It has long been understood

that the surface a cell grows on impacts its size, shape, and metabolism. By determining

the aspects of the material and the surface that influence contact guidance, there is a

better opportunity to design a hierarchical system to elicit the desired response. Contact

guidance can be controlled by topography and surface chemistry. Typically, when cells

are exposed to mechanical features such as ridges and grooves, the cells align and travel

along the length. The addition of roughness at certain levels can improve a biomaterial's

ability to promote cell adhesion, while disrupting adhesion at different levels.[1-3]

Patterning surface chemistry on samples to change the wettability and surface

energy has been very successful in controlling cell growth. Alternating strips or islands

of adhesive proteins and materials with different surface energies have been examined.

The incorporation of texture with surface chemistry allows for mutual interactions to









enhance the desired response.[2] As a better understanding of the principles involved is

obtained, more subtle influences on the control of contact guidance can be examined.

The objective of this study was to study the effect of modulus as well as surface

texture dimensions on vascular endothelial cells (ECs). ECs are important regulators of

homeostasis in the human body, and a crucial component of the cardiovascular system.

Diseases of this system currently contribute to more deaths than any other disease or

cause. By controlling the alignment and growth patterns of these cells and their tissues,

improvement of medical devices such as vascular grafts is possible. Modulus has been

shown to play an important role in adhesion of biofilms on substrates in marine

environments,[4, 5] and coupling the effects of modulus with topographical features is

another important step in designing the behavior of materials.

The specific aims of this project involved the study of endothelial cells grown on

microtextured silicone elastomers. Specifically, the effect of modulus was hypothesized

to increase the effect of contact guidance as the modulus increased. This hypothesis was

based on the observations of Kendall and others that modulus played an important role in

the adhesion ofbiofilms on elastomer surfaces. These theories will be discussed in detail

in the following chapters. To truly understand the system, another specific aim was to

determine the importance of feature dimensions on this system. The hypothesis to be

tested was that the depth of grooves in a surface played a more important role than the

spacing between the grooves, and that a deeper groove increased the contact guidance of

an EC on silicone. When the groove depth remained constant, the hypothesis being

tested was that the grooves spaced closer together would improve the cells ability to

direct cell growth. The importance of groove depth and spacing has been proven before









on different systems to confirm these hypotheses, but the effect of modulus on similar

materials has not truly been examined.

To accomplish this objective, micropatterned silicone elastomers were fabricated

with additives to change the elastic modulus while keeping the surface energy relatively

constant. Surface topography was examined by various methods such as optical

profilometry. Due to the low energy nature of the silicone, samples were treated with an

argon plasma or coated with an adsorbed layer of fibronectin to improve cell adhesion.

Porcine vascular endothelial cells (PVECs) were examined using the nuclear form factor,

which is a measure of the alignment of a cell to defined topography by measuring the

dimensions of its nucleus. The microtextured features were designed to be able to

compare the degree of contact guidance by varying the feature width and depth. The

modification of the samples with functionalized siloxane oligomers allowed for the

variation of the modulus while studying the growth of cells on materials that are

otherwise similar. These novel systems allow for the quantification of cell alignment, as

well as a measure of a surface's capability for contact guidance














CHAPTER 2
BACKGROUND


Contact Guidance

The reaction of cells in vitro to the substrate they come into contact with has been

traditionally separated into two main features: topography and surface chemistry. Cells

adhere to surfaces via specific adhesion molecules that interact with proteins adsorbed

onto the surface of the substrate. Thus, if the surface chemistry is favorable to adhesive

protein adsorption, then the material should be favorable to cell adhesion. However, the

topography of the surface is also important not only in the adhesion of the cells to the

surface, but also in the behavior of the cells' metabolism and growth patterns after the

initial contact. The purpose of understanding and controlling cell and tissue growth on

artificial materials is to be able to design and implant medical devices that improve

biocompatibility and functionality.

Surface Topography

The response of the cell to topography has been referred to as "contact

guidance."[6, 7] The first known reference to the effect of substrate topography on growth

characteristics comes from the growth of embryonic cells on plasma clots and spider

webs.81 Rovensky et al. used V-shaped grooves formed from copies of music records to

show that chick embryo fibroblasts migrated from the bottom of the groove to the top

over a period of hours and aligned along the texture.[9] Dunn and Heath examined chick

heart fibroblasts growing on glass fibers to examine the effect of radius of curvature on









the cells.[10] They discovered an important aspect of contact guidance in that the shape of

the substratum causes mechanical impediments to the formation of cytoskeletal bundles

important for cell locomotion. By correlating this discovery with linear slopes and

discontinuities, they determined that an angle of deflection greater than 80 between two

planes of a prism affected their cells and made them less likely to cross over the ridge.

Another important aspect of this paper introduces the use of the nucleus to quantify

contact guidance, by factoring the elongation of the nucleus with respect to the

orientation of a fiber. This technique will be used in the results portion of this thesis, and

will be explained in more detail later.

Cells have also been shown to move along topographical features, and drastically

change their morphology in response to this texture.[1-15] Clark et al. examined several

cell lines in response to specific topography. They first examined baby hamster kidney

(BHK) cells and embryonic neural cells in response to a single 5 rm step. Surprisingly,

this is one of the only studies to examine such a simple topography. They found that the

cells were inhibited from both climbing up as well as down the step, but tended to align

along the ridge.[16] As a follow-up, they examined the same cells on grooved substrata of

varying widths and depths (4-24 rm repeat width, 0.2-1.9 rm depth) and found that

groove depth increased cell alignment and proved to be more important.[17] Similar

results were found using rat bone marrow cells on poly-L-lactic acid (PLA) grooves, with

an increase of extracellular material being deposited along the grooves.[18] Schmidt and

von Recum show distinct morphology changes on pitted surfaces for macrophages and

increased spreading on smooth surfaces.[19] Macrophages were also demonstrated to have

the unique ability to align to extremely shallow grooves, from 30-70 nm.[20, 21]









The goal of substrate topography is to direct the cells to grow in a certain pattern

and direction. Cells tend to align along a groove and move along this surface. Studies

have shown that moderately porous materials improve cellular adhesion, which is

possibly due to mechanical stability along with increased surface area for adhesion.[22-24]

Endothelial cells (ECs) also have been shown to align along the direction of fluid flow.[25-

27] In addition flow has been demonstrated to play a role in altering the mechanical

properties of ECs. When subjected to a shear stress of 2 Pa over 24 hours, the endothelial

cells gradually increased their stiffness as measured by the atomic force microscope.[26]

The methods of producing the precise surface morphologies vary depending on

the size of the pattern and the material on which the pattern is being replicated. The

smallest patterns are produced with direct write laser lithography[28] and AFM

lithography while those on the 2-10 um scale are produced with UV photolithography,

followed by reactive ion etching to control the slope of the walls.12' 29-32] The most

common method used in these experiments for producing features is to first

lithographically produce the pattern on a silicon wafer, then replicate that pattern by

embossing or spin casting onto the substrate.[33]

The grooves and ridges formed on these substrates have shown significant control

over growth directions of cells. Current discussion focuses on the mechanisms behind

the alignment of these cells to the surface topography. Von Recum and van Kooten

question whether or not the actual geometries of the features are the defining factor, or

the fact that there is a change in surface free energy due to edges and disruptions in the

planar surface.[34] den Braber et al. concluded that parameters such as surface free energy

and wettability influence fibroblast growth and proliferation on microtextured surfaces,









but not the shape or orientation of cells in comparison to the texture.[1ll By SEM, they

conclude qualitatively that there is more alignment on the 2 tim wide features, than on the

5 tm and 10 tm substrates. Other studies in this review seem to disagree with the their

conclusion that fibroblasts do not align to wider features, but the fact that their groove

depth was less than 1 .im seems to be a limiting factor. Walboomers et al. examined

fibroblasts[35] and rat bone marrow (RBM) cells[18, 36] on polystyrene (PS) and PLA

textured radio-frequency glow discharge (RFGD) plasma treated surfaces with ridges and

grooves with dimensions varying from 1 [tm to 10 [tm wide and depths of 0.5-1.5 [im.

These studies, along with another study they published demonstrated the importance of

the ridge depth in that at deeper depths (up to 5.4 itm), the cells were more aligned, but

not as many cells grew on the surface, even with the increase in surface area.1351 They

also showed similar results in examining alignment of intracellular and extracellular

proteins, but found that the addition of the ridges and grooves did not alter proliferation

of cells on the surface at all.112' 37-39] Their group also took their textured samples from in

vitro use to in vivo by implanting RFGD treated disks of textured and untextured

silicones subcutaneously in rabbits and guinea pigs, and PS disks in goats. Their results

were mainly inconclusive, but they noticed with the silicone substrates an increase

vascularization of the capsules surrounding the textured surfaces compared to the

untextured surfaces.[40-42]

Typically, the more wettable the surface is, the more cell proliferation occurs. A

study by Walboomers and Jansen et al. using rat dermal fibroblasts (RDF) on PS, PLA,

silicone, and titanium coated PS substrates also show that the microtextures influence cell

guidance, while surface chemistry influences morphology.[151 This study is of particular









interest, in that they examined the effect of different substrates with the same features.

They compared the wettability of RFGD plasma treated samples and the elastic modulus

to the substrate's ability to influence contact guidance. The features were the same

dimensions as the previous studies, with the depth only 0.5 im. The moduli varied from

894 MPa for PS to 0.39 MPa for silicone. Elastic modulus of 5 x 15 x 0.2 mm bars was

measured using an Instron mechanical testing machine with a crosshead speed of 0.5

mm/s. A proliferation study showed no statistical difference in the number of cells

attached to each surface, although there was a significant increase once each surface was

RFGD treated. Contact angles using only water gave their measure of wettability, but

their relatively high value for silicone (330) after plasma treatment implies that the

surface had rearranged or the plasma treatment was incomplete. This phenomenon will

be discussed in more detail later. Their overall data were inconclusive, with their

conclusions focusing more on the production of the patterns and the fact that different

materials still induced contact guidance.

A more recent study by this group addresses the depth of groove issue, by

examining epithelial tissue and cell migration across and along PS microgrooves.143]

Briefly, 6 mm punches of bovine eye endothelium were place on the microgrooved

surfaces and cultured for 6-9 days. They studied ridges and grooves at widths of 1, 2, 5,

and 10 rm and depths of 1 and 5 im. This interesting study demonstrated the

importance of groove depth, in that they concluded the width variations not to be as

important as depth, but more importantly, they concluded that the microgrooves have the

capability to direct tissue growth. By placing intact epithelial tissue on a patterned

substrate, they found that on 5 um deep ridges, the tissue was constrained to grow mainly








in the direction of the ridges and grooves, and did not cross over perpendicular to the

features very significantly. Examples of explant growth and the distances measured are

shown in Figure 2.1.


A




C D

f a e



Flat-cast surface Microgrooved surface

Figure 2.1 Light micrographs and measurement protocols of 6 mm diameter punches of
corneal epithelial tissue explants (f) over (A & C) smooth PS surface and (B & D) a
surface with 1 [im deep microgrooves separated by 1 [im[43]


There was more growth perpendicular to the grooves with the 1 im deep features

compared to the 5 [m deep grooves, but the growth was still directed mainly along the

feature direction. While the direction of the growth was more polarized with the 5 rm

deep features, there was more tissue migration area for the shallower grooves. Similar

results were found for cultured separated epithelial cells plated from a suspension.

The study of defined patterns of topography stemmed from many observations

that the random roughness of a biomaterial surface influenced cell and tissue response.

The phenomena of rugophilia and rugophobia, defined as cells loving or hating rough

surfaces was initially pointed out by Rich and Harris.[44] The luminal surface of

conventional Dacron vascular grafts can be considered textured in a random roughness









pattern. Studies in sheep have shown a distinct difference in the amount of cellular

deposition on non-textured and textured polyurethane vascular surfaces after in vitro and

ex vivo study.145' 46] Using excimer laser micromachining, textured surfaces consisting of

fibers of 25, 50, and 100 rm in length sticking up from a smooth base plane were

examined. When implanted in ovine carotid arteries thrombus formed more quickly on

the textured surfaces compared to the non-textured surfaces, leading the authors to

conclude that the textured surface acted as a promoter of a stabilized thrombus base.

While this may accelerate the formation of a stable pseudointima, the thickness and

composition of the thrombus was not controlled.

Osteoblasts have also been shown to react more favorably to roughened titanium

surfaces compared to smooth surfaces,[47] while surfaces with regularly spaced

nanometrically sized pillars reduce tendon sheath (epitenon) cell adhesion.[21] This

phenomenon is not seen for larger pillar sizes, in that astroglial cells showed preferential

adhesion to pillars and wells on the scale of 0.5-2.0 rm in width and 1.0 rm in height.[48]

In fact, 2 rm pillars and holes 4 rm deep showed changes not only in cell adhesion, but

cell motility.[491 Neutrophils migrated much faster on holes than on smooth surfaces,

while pillars slowed the process down.

A group out of Harvard headed by Vacanti has recently moved into using the

effects of contact guidance and directed cell growth for future clinical applications. They

used micromachining technology to form many branching networks resembling capillary

beds. Hepatocytes and endothelial cells were patterned and lifted as 2D sheets for the

purpose of forming 3D tissue constructs.[50] While this technique is far from perfected, it









is an interesting leap from the studying of contact phenomenon towards a more clinical

application.

The field of contact guidance using topographical cues is quickly becoming more

noticeable in clinical fields. Several excellent reviews have been authored and are a good

starting point for an overview on the topic.[1 3 51] However, topographical features are

not the only tools for directing cell growth, as the chemistry of the substrates is becoming

more popular in terms of directed cell growth and adhesion. As more studies are

performed and a better understanding of the issues involved is achieved, clinicians and

researchers are discovering the importance of the texture of medical implants as well as

the chemical moieties on the surfaces.

Surface Chemistry

Surface chemistry plays a large role in the field of contact guidance for

controlling the results of cell proliferation. Carter originally demonstrated that cells

exhibit a preference to hydrophilic areas of patterned cellulose acetate and palladium

metal surfaces.152' 53] The cells migrated towards the metal and, in a separate study, along

a gradient of metal concentration densities towards the more dense and thereby more

hydrophilic areas. These studies were later duplicated and confirmed by Harris.[54]

Currently, surface treatments are typically deposited in regular patterns using the same

photolithographic techniques as with the topographical substrates. Self-assembled

monolayers[551 (SAMs) and areas of enhanced adsorption of proteins due to surface

energy seem to be the most popular. A study by Britland, et al. deposited a pattern of

alternating hydrophobic dimethyltrichlorosilane groups with aminosilane groups on glass

slides. No residual topography resulted and BHK cells showed a definite preference for

the aminosilane sections, as the cells were crowded and aligned along the border rather









than cross to the hydrophobic groups.[56] A similar study was performed by Healy's

group using human bone-derived cells (HBDC) and showed the same results.[57] Another

study by Britland et al., examined the effect of topography in combination with surface

patterning. This paper examined quite a few of the relevant topics with a few simple

experiments, although their choice of cells (BHK) does not give as much useful

information for the future. They superimposed tracks of aminosilanes orthogonal to the

ridges and grooves and showed that for shallow grooves (0.1 im to 0.5 pm), the cells

aligned almost exclusively with the chemical patterning, but as the grooves go deeper up

to 6 rm and closer together (5 .im), alignment to both the texture and the chemistry was

seen.1581 By comparing these results with the studies on topographical patterning and

chemical patterning, it is apparent that there are critical levels for each system, in which

topography and chemistry contribute differently to contact guidance.

The use of silicones in the micropatteming of surfaces has also become quite

popular.[59-63] Essentially, textured polydimethylsiloxane (PDMS) substrates are formed

using the microfabrication techniques mentioned before. These substrates are then used

as stamps or stencils to either directly apply surface treatments such as proteins to another

substrate in an organized fashion, or to act as a mask allowing for microfluidics within

channels in the silicone to pattern the surface.[59] Many times the silicone itself is the

substrate that is used for cell studies. Whitesides' group has demonstrated that by

selectively adsorbing adhesive molecules in the form of fibronectin to the bottom of pits

or wells in the surface, and by keeping a non-adhesive protein like albumin on the surface

above the wells, endothelial cells will adhere only to the bottom of the wells where the

fibronectin is adsorbed.[64] They have also shown that through the use of microcontact









printing with PDMS stamps, they can pattern SAMs of alkanethiolates on gold to

manufacture substrates with controlled islands of extracellular matrix (ECM). By

restricting the size of ECM endothelial cells had to attach to, they controlled the size of

the cell, which also resulted in altered metabolism.[65, 66] As the size of the area of

attachment decreased, the cells changed from growth to apoptosis, or cell death. This

technique is useful in its ability to isolate single cells on the same substrate for

microarray examination of cell types.

Whether the application of a material is for a vascular graft, dialysis machine,

blood oxygenator, bioreactor vessel, dental material, or the surface of a ship's hull, the

first step in any biological response to a surface is the adsorption of proteins.[67-69] One

cause of this adsorption is due to the highly varied structure of a protein in solution, both

due to its conformation and primary structure of amino acids. Another cause is due to

surface rearrangement of hydrophobic and hydrophilic areas of the polymer chain. In this

fashion the surface properties and specifically surface energy play a very important role

in protein adsorption.[70]

An adsorbed layer of serum proteins after exposure to blood is not in the same

concentration as the bulk liquid. Rather each protein has a different response to the

material in contact, as illustrated by Table 2.1. The composition of the adsorbed protein

layer changes over time, as early adsorbing proteins are displaced by others, exhibiting

the "Vroman effect."[71-73] Cells that adhere to a surface are able to deposit their own

proteins, but only if the proteins are able to displace those already adhered. Hydrophobic

materials, in the presence of high serum concentrations make this displacement very

difficult. [34]









Table 2.1 Enrichment of proteins adsorbed on polyethylene exposed to blood plasma[68]
Protein Enrichment"
Fibrinogen 1.3
y-globulin 0.53
Albumin 0.88
Hemoglobin 79
aEnrichment was calculated as the ratio of the surface fraction of the protein compared to
the bulk fraction.


Endothelial Cells

Endothelial cells make up one of the most important tissues of the body, known as

endothelium, which is the interior lining of all blood vessels. The three major types of

blood vessels are arteries, veins, and capillaries. Arteries and veins have a complex

structure, made up of three main layers as seen in Figure 2.2. The outermost layer is

known as the tunica adventitia, and is composed of loosely woven collagen fibers that

protect and anchor the blood vessel. The middle layer, or tunica media, is mostly smooth

muscle cells and elastin. This portion of the vessel is elastic and plays a major role in

regulating blood flow by relaxing and constricting. The innermost layer is known as the

tunica intima, and contains the endothelium on a subendothelial layer of loose connective

tissue to act as a basement membrane.[74]

Cardiovascular endothelial cells (ECs) are diverse in their size and shape, since

they are needed not only to act as a blood barrier in the aorta but also for nutrient

exchange through tiny capillaries. For many years, endothelium was classified as an inert

tissue, strictly a barrier to keep blood in its vessels. It is now known that the endothelium

is an active tissue in homeostasis, producing factors such as endothelin, PDGF

(prostaglandin-derived growth factor) and nitric oxide.[74] Endothelial cells are simple











squamous epithelial cells that form a smooth, confluent monolayer characterized by a


cobblestone pattern, as seen in Figure 2.3.


Figure 2.2 Cutaway view of an artery showing the three main layers. From Seeley et
al. [ ]




As they achieve confluence, the cells are contact inhibited and alter their


metabolic behavior, going into a period of stable growth and turnover, known as


quiescence. In the adult, the average endothelial cell only divides approximately twice in


a lifetime.[76] However, the endothelial cell can rapidly proliferate upon damage.


Disorders of the endothelium result in many pathologies, including atherosclerosis


and cancer. It is generally accepted that a vascular graft or other blood contacting


implant material will have less of an immune response if it is covered by an intact layer


of endothelial cells. Endothelial cells effectively provide local delivery of endogenous


endothelial secretary products to maintain prosthetic integrity after surgical


Vasa vascrum

Nerve

Tunica
S.- adventitia
External
elastic membrane Tunica
.. 4.x.media
4 Smooth muscle
._. Internal
elastic membrane
Tunica
SL amina propria n
0r --,smooth muscle and intima
connect ive tissue)
Basement
S.... membrane
,Enclothelium









implantation.[77] When exposed to a topography that disrupts their ability to achieve

confluence, the cells typically respond by moving on the surface and extending philipods

out in multiple directions, searching for other cells and more favorable conditions.















Figure 2.3 Light microscopy image of confluent porcine vascular endothelial cells
(PVECs) grown on tissue culture polystyrene (Image taken by L. Zhao and W.
Wilkerson)



Vascular Grafts

In 1999, 529,544 people died from ischemic heart disease[78] and many undergo a

procedure known as coronary artery bypass. This is necessitated by a blockage of the

coronary arteries caused by fatty plaque accumulation or thickening of the artery wall.

Coronary artery bypass graft (CABG) surgery is a procedure that supplies blood flow to

the other side of the blockage by attaching a small diameter graft to the blood vessel.

These grafts can take the form of native materials such as arteries and veins, or artificial

polymers. The success and failure of these materials is crucial to the survival of the

patient, and as of now no adequate small diameter vascular graft is available.

The typical treatment in a CABG procedure is to use the saphenous vein from the

leg and the internal mammary artery as the graft materials. These are superior to non-









natural materials because they do not have an immune response in the patient and are

more similar to the native tissue. These grafts have patency rates of over 70% after 5

years, while the survival rate without the surgery is nearly zero.[791 In cases where the

autograph material is not available or unusable, then the only option is to implant a

polymeric vascular graft. Large diameter vascular grafts used in aortic repairs are made

from a woven polyethylene terephthalate (PET) mesh or expanded

polytetrafluoroethylene (e-PTFE) and are very successful. PET is a thrombogenic

material, in that it causes a cascade response that clots the blood at the surface of the

graft. This response is actually favorable and encouraged, in that it seals the porous graft

and since the aorta is such a large diameter blood vessel, the flow is not significantly

diminished. To better control this process, the grafts are typically coated with albumin or

collagen, essentially pre-clotting the graft.[80]

Small diameter artificial vascular grafts (< 6 mm) cause more of a problem in that

the patency rate is less than 50% after 3 years.[79] Typically, the graft fails either due to

thrombosis initially, or a buildup in the intimal layer of the blood vessel, essentially

occluding the vessel with tissue. The thrombosis is due to the activation of the clotting

cascade when the graft is exposed to the blood, activating platelets and the absorption of

fibrinogen. Later in the life of the graft, the main threat to its success is occlusion by

anastomotic intimal hyperplasia. The actual mechanisms of this are debated and are

being examined, but a number of factors seem to contribute. First, there is a compliance

mismatch between the native tissue and the graft. The artery has a very specific

compliance to pressure waves during pulsatile flow that the graft interrupts. A mismatch

causes stresses at the suture points of the anastomosis, which is typically where the









intimal hyperplasia occurs. This is the main cause of failure in autologous saphenous vein

(ASV) grafts, due to the fact that arteries have a much more muscular vascular wall

compared to a vein.[81]

Probably the most significant cause for the hyperplasia as well as thrombosis is

the lack of a stable lining of endothelium, as in a normal vasculature. It typically only

grows a short distance from the anastamoses, leaving exposed areas of pseudointima,

consisting of fibrous materials and cells, including fibroblasts and smooth muscle cells.

While current PET and e-PTFE grafts are inadequate in this regard, new materials are

being investigated to improve cellular response to the tissues.[82-86]

An ideal graft material would exhibit thromboresistant qualities, have similar

mechanical properties as the native tissue, have ease of use by the surgical team, and be

able to form a stable endothelial layer. Currently the materials used are thrombogenic,

have varying yet acceptable mechanical properties, are easy to use by surgeons, but do

not form a stable endothelial layer.

Studies are in progress that have seen patency rates over two year periods double

for artificial small diameter grafts due to endothelial seeding.[84] Currently, methods to

improve endothelial cell seeding involve the adsorption of proteins on the surface or the

use of fibrin glue.122] Past results have shown that the substrate must tightly adhere the

cells or they can be removed when subjected to shear flow. For surfaces without covalent

linkages between the proteins and the surface or strong mechanical interlocking, it is

difficult to form a stable intima that can withstand biological stresses. Many of these

modified surfaces are attractive to proliferation of a fibrous pseudointima of fibroblasts

and smooth muscle cells. With the proper surface chemistry and topography, endothelial









cell attachment and spreading to an intact, confluent layer that can withstand biological

stresses should be possible.

Cell Adhesion

Focal adhesions are typically the main area of adhesion of a cell to a substrate.

They are the link between ECM proteins such as fibronectin, vitronectin, and collagen

adsorbed to the surface, and the cell receptors, known as integrins, which bind to the

ECM through the well-documented Arg-Gly-Asp (RGD) tripeptide sequence. These

receptors link the focal adhesions with the cell's cytoskeleton, thereby altering the cell's

shape and locomotion. Focal adhesions are typically elongated, and oriented in the

direction of the stress fibers and the main axis of elongation. 871 A recent study by van

Kooten and von Recum have shown that fibroblasts and human umbilical vein

endothelial cells (HUVECs) formed focal adhesions within the first 24 hr of adhesion on

textured silicone surfaces.[14]

Recently work has been produced to examine binding polypeptide sequences to

the surfaces of polymers.[88-93] The chemistry behind the grafting of these molecules on

the surface can be accomplished by using plasma-induced graftcopolymerization. By

exposing the surface to an argon plasma and then air, hydroperoxide groups are formed

on the surface that can initiate radical polymerization.[92] Much work has been examined

using the RGD sequence as the active adhesion area in proteins such as fibronectin.

Hubbell examined the specific adhesion of endothelial cells to the Arg-Glu-Asp-Val

(REDV) tetrapeptide through the ac4- 31 integrin and showed that while endothelial cells

attach and spread on this ligand, fibroblasts, vascular smooth muscle cells, and platelets

did not.[94, 95]









Recent work has refocused on the traction forces that cells impart on the substrate

on which they adhere. By using low moduli micropatterned substrates, [87 and

unpatterned "wrinkling" substrates,196-100] studies have begun to quantify these forces for

specific systems. Harris et al. first introduced the study of cell locomotion and adhesion

with wrinkling substrates by growing a wide variety of cells on a thin, heat-crosslinked

film of silicone floating on a silicone fluid. As the cells grow, they pull the film

underneath in circumferential folds that are smoothed out when the cells are trypsinized

from the surface.[96] Pelham and Wang were able to grow fibroblasts on polyacrylamide

substrates with very low moduli but varying 12-fold. They found that fibroblasts spread

less and had increased motility and lamellipodial activity on more flexible substrates,

while more rigid substrates promoted stable elongated focal adhesions.[98] Surprisingly,

this is one of the few studies examining the effect of different moduli of similar materials

on cell growth, and none to date have been reported that compare the effect of modulus to

the contact guidance phenomenon.

Balaban et al. used fibroblasts stained to expose the focal adhesions of fibroblasts

grown on low modulus (E' 15 kPa) silicone elastomers with either fluorescent patterns

embedded into the surface, or features similar to pits and pillars as part of the surface.

The textures on their surfaces were 0.3 [im deep, because they wanted to minimize the

contact guidance phenomenon. With features deeper than that, polarization and directed

growth occurred along the features. In relating the displacement of the features with the

locations of the focal adhesions, they were able to extract force measurements exerted by

the focal points.[87]









It has been well documented that cell behavior, in the form of growth, movement,

and metabolism, is closely linked to the shape of the cell.1101' 102] The effects of adding

forces and changing the tension on a cell have shown significant changes in the

biochemistry of the cell.1103] Several cell types also align along lines of principle strain

with external loading. 13, 104] Kato et al. showed that endothelial cells, when patterned on

thin strips of adhesive regions that caused endothelial cells to become elongated,

exhibited a decrease in mRNA expression for vascular adhesion molecule-1 (VCAM-1)

and a higher mRNA expression for intracellular adhesion molecule-1 (ICAM-1).[102]

Topography has been shown to alter cell shape, fibronectin mRNA level and stability,

and the secretion of ECM by human fibroblasts.1101l

In many cases, the shape of the cell is widely spread out, especially when

lamellipods extend in multiple directions. This makes quantification of cell response to

texture or chemistry difficult by simply looking at the outline of a cell. One approach has

been to examine the shape of the nucleus of a cell to elucidate the prevailing cytoskeleton

arrangement within the cell.1101 Ingber's group as well as others have demonstrated that

the nucleus' shape is indeed directly "hard-wired" such that changes in surface adhesion

can affect the shape and orientation of the nucleus.[105] They examined bovine

endothelial cells after attachment to a substrate, and pulled on them with pipets coated

with adhesion molecules and found that as the cytoplasm stretches, the shape and

orientation of the ECs change from round to elongated along the stress.

As medical technology and surgical procedures improves, the need for effective

biomaterials becomes greater. The ability of a non-native material to mimic the

properties and functions of the tissue it is replacing is crucial to the success of the









prosthesis or device. Similarly, a biomaterial used in an in vivo or ex vivo application

should minimize unwanted biological interactions.


Silicones

Silicones are widely popular materials and have many commercial uses today.

They are unique in many of the polymers used in biomedical devices in that they possess

a silicone-oxygen backbone instead of a carbon backbone. Their chemical and physical

properties allow for their use in a variety of applications. The simplest silicones are

polydimethylsiloxanes, a linear organosilicon compound whose structure can be seen in

Figure 2.4.

CH3 CH3 CH3

H3C- Si- 0 Si- O Si- CH3

CH3 CH3 CH3

Figure 2.4 Polydimethylsiloxane, trimethylsiloxy terminated



PDMS oils, when not cross-linked, are used in fields such as cosmetics, food-

processing, and pharmaceutical preparations. Their lubricity and low surface tension

make them excellent additives for anti-foaming.[106]

The methyl groups in the backbone and end caps of the PDMS molecule can be

replaced by both functional and non-functional molecules, including hydrogen, phenyl,

and vinyl groups. The significance of these substitutions is in their changes to the

chemical and physical property of the polymer, as well as the curing and processing

capabilities. Cross-linked PDMS forms a silicone elastomer that has excellent elongation

properties (20-700%) and moderate breaking strengths (- 1000 psi). They have been









cured in high-temperature vulcanizing (HTV) systems of methyl or vinyl groups using

peroxides. Room temperature vulcanizing (RTV) systems are cured by condensing

silanols with a moisture-sensitive silane cross-linking agent, or are condensed with a

metal salt catalyst. Vinyl addition systems using platinum catalysts were initially used in

low-temperature vulcanizing (LTV) systems, but has been extended to RTVs and

HTVs.[1061 The silicone used in this study is Dow-Corning's Silastic T-2, a filled RTV

elastomer with vinyl terminated end caps that is addition polymerized with a platinum

catalyst.

Use as a Biomaterial

Silicone elastomers are valuable polymers in the biomedical field. The use of

silicone materials in vivo has become a heavily debated topic in the past decade, with the

proliferation of procedures and studies on devices such as breast implants. These

devices, made of a PDMS gel, allowed for the leaching or 'gel bleed' of low molecular

weight oils from the device into the surrounding environment and then through the body.

The silicone biomaterials were widely criticized for causing a large range of

complications and diseases, and as a result, were removed from the market as a medical

device. A recent risk assessment on the effects of silicone gel-filled breast implants

concluded that the adverse effect of exposure to these prostheses was minimal, and

current stringent regulation should be discontinued.[1071

One of the main problems with silicone implants, as with all implant materials, is

the formation of fibrous capsules.[69] The capsules can cause discomfort as well as

contraction on the device that can ultimately lead to failure.

Silicone's permeability to oxygen allows for their use in contact lenses and

membrane oxygenators, and their flexibility and stability have seen uses in a wide variety









of implant materials.192' 106, 108, 109] Silicone orthopaedic devices have included finger

joints and temporomandibular joints (TMJ), which were not sucessful.11101 They have

also been used in the vascular field as components to heart assist devices and

incorporated into vascular grafts.[106' 111] While the mechanical properties of silicones are

ideal for vascular grafts, in that their compliance and modulus more closely match the

native tissue than PTFE or PET, their lack of toughness prevents them from being used

exclusively. Also, a major concern with PDMS is its affinity for lipids, which causes it to

become more brittle in vivo over time.[111' The important issue to keep in mind for

silicone systems as well as other biomaterial applications, is that the bioactivity of the

system will typically dictate what use the material has in vivo. Silicone surfaces are

mainly non-thrombogenic and biologically inert, which makes them an interesting

material to prevent unwanted interactions, but this lack of bioactivity also precludes it

from becoming incorporated in surrounding tissues without surface or bulk

modifications.

Another field of particular interest to our group is the use of silicones to prevent

biofouling from marine organisms. Biofouling is an example of a problem concerning

cellular materials accumulating on surfaces such as the hulls of ships and water treatment

facilities. The marine spore Enteromorpha is the most common macroalga that fouls

ships and submarines. Reproduction is mainly through motile spores that swim until a

suitable surface on which to settle and adhere is located.[112] Adhesion involves secretion

of a glycoprotein adhesive that anchors the spore to the surface.[113] Cues for settlement

include phototaxis, chemotaxis and thigmotaxis. Previous anti-fouling coatings included

biocides that did significant damage to marine life in harbors. Current research focuses on









preventing adhesion forces able to withstand the shear forces during motion.[114' 115] By

using low surface energy silicones, coatings are foul releasing rather than antifouling.

Singer and others in this field have realized the importance of the mechanical properties

in determining the ability of a cell to adhere and remain adhered to a surface.[5] He

simulated barnacle pull-off tests by epoxying a stud onto a silicone surface, and

determined the critical force to pull it off with respect to the material's thickness and

elastic modulus. The result was for lower modulus materials (E* = 3 MPa) and thicker

coatings (up to 4 mm), the force needed to detach was less than for higher modulus (23

MPa) and thinner (0.08 mm) coatings. Gatenholm's group examined the use of

microtextured surfaces in the marine biofouling environment by imparting 50-100 [im

deep and wide features through use of a wire mesh as a mold. They found that barnacle

adhesion on the macro scale decreased on textured surfaces compared to smooth. Many

of these same principles of concern in biofouling can be used in biomedical applications

to improve the biocompatibility of polymeric surfaces in the body.

Surface Energy

Due to its hydrophobic nature, silicone experiences rather high amounts of protein

adsorption and poor spreading of cells.[92] Thus, to improve cell adhesion to a silicone

surface, the chemistry of that surface is usually modified. Unmodified silicone is

hydrophobic with advancing contact angles around 110-120. The difference between

the advancing and receding contact angle is known as hysteresis, and gives some

understanding of a surface's ability to remodel itself as well as the surface roughness. As

the silicone is exposed to water, hydrophilic areas of the siloxane backbone migrate to the

surface, masking the hydrophobic methyl groups.[701 This provides a more hydrophilic

surface and a lower contact angle while receding. This rearrangement of the silicone









backbone is an interesting phenomenon, and is especially important in PDMS systems

due to its low Tg (-123C). Most of the surface rearrangement in plasma treated high

surface energy PDMS samples is due to the migration of low surface energy, low MW

PDMS oligomers from the bulk to the surface.1116' 117]

Ostuni and Whitesides used the low surface energy of silicones to selectively

pattern the surfaces of textured elastomer substrates with protein solutions. Since the

contact angles of a fibronectin/physiological buffered saline (PBS) solution and a bovine

serum albumin (BSA)/PBS solution on their silicones were over 1000, they could trap air

inside of wells by carefully placing drops of the solutions on the surfaces. This allowed

for the flat surfaces above the well to have one of the proteins adsorbed, and by rapidly

changing the solution and adding a vacuum to pull out the air bubbles, the bottom of the

wells were patterned with another.[64]

Schmidt and von Recum characterized Dow Corning's Medical Grade Silastic

(MDX4-4210) silicone after texturing the surface with various pillars and wells. They

determined the surface energy of their materials by using the stationary drop method I18]

using diagnostic liquids that were not identified. They calculated Zisman plots that graph

the cosine of the contact angle versus the surface tension of the diagnostic liquid. All of

their samples, both textured and untextured had critical surface tensions of 20.5-21.5

dynes/cm.[119] They showed that the surface energy per unit of surface area decreased for

higher densities of features, and concluded that the surface energy either was not greatly

affected by their topography, or that contact angle methods are not sensitive enough to

detect the difference.









Mechanical Properties

The importance of mechanical properties has been discussed in the uses for

silicone materials and in the need for proper compliance for vascular graft applications.

One area of interest is designing the mechanical properties to control the response of

biofilms and tissues. Currently, the focus of our interest is in modifying the modulus of

the elastomer without greatly affecting the surface chemistry or energy of the sample.

Young's modulus, E, may be written as

07
E=-


where cr and E represent the tensile stress and strain respectively.[120] Essentially,

Young's modulus is a measure of the stiffness and compliance that is referenced in much

of the biology-based research on these materials. The term compliance can be defined as

the elongational compliance, J, which in regions far from thermal transitions is the

inverse of Young's modulus.

1
E

This should not be confused with the compliance of vascular graft materials, which is a

measure of the dynamic circumferential elastic properties of a vessel and is calculated

using systolic and diastolic blood pressure and the vessel diameter by


c (Dsystolic diastolic)
C=
Sdiastoloic /the press diastolic a

where D is the vessel diameter, P is the pressure, and C is the compliance.[121]









Typical glassy polymers such as polystyrene at room temperature have E values

around 3 GPa, while soft rubbers are closer to 2 MPa.1120] PDMS based elastomers have

a somewhat wide range of moduli, from 0.1 MPa to 10 MPa.111ll The change is due to

the curing characteristics and the size of the oligomers, which affect the cross-link

density, molecular weight between cross-links, and physical entanglements.

Currently, we are able to vary the modulus of our base Silastic T-2 elastomer over

a range of moduli from -0.1 MPa to -3 MPa with the addition of functionalized and

unfunctionalized PDMS oils. These oils can take the form of linear PDMS or larger

"bulky" oils with side chains. By varying the molecular weight and functionalization of

these additives, the variables mentioned above can be altered, without significantly

changing the surface chemistry of the samples. The majority of the development and

characterization of this system was performed by other members of the group, and so will

not be discussed in great detail in this thesis. However, since the materials with differing

moduli are of critical importance to the main purpose of this work, a description of the

specific oils and testing methods will be discussed in the characterization section.

Research in our group has focused on the effect of mechanical properties on

biofilm formation, whether the film is in the form of algal spores or endothelial cells.

Kendall modeled adhesion behavior on elastomers by deriving formulae for the removal

of rigid solids off of elastomers with varying thickness.[122] The critical pull-off force, Pc,

required to remove a rigid cylinder with a radius of a from a film of thickness t is given

by


2waK
p.=^[t









where wa is Dupre's work of adhesion and K is the bulk modulus, which is related to the

elastic modulus, E, by Poisson's ratio, v, by the following:

E
K=
3(1-2v)

At greater thickness where t >> a, the relationship between the pull-off force and

the elastic modulus is given by


P 8Ew 2
PC =)\ -


The implication of this equation is that stiffer materials improve strength of adhesion.[5]

One of the assumptions to this equation is that the attached surface is rigid, while for a

cell, this is not the case. As discussed earlier with the low modulus wrinkling elastomers,

a cell has the ability to create its own forces and change them depending on the substrate

on which it is adhering. If a focal adhesion is modeled as a rigid body, then the equation

can hold validity. If we imagine a cell applying traction forces on the substrate it is

adhering to, then it is possible that a cell in equilibrium might exert forces just under the

critical force in equal directions. Chicurel et al. concluded in their review of focal

adhesion literature that a cell will continually contract on a substrate until the forces come

into balance, much like a bow and a bowstring.[103] By that reasoning, for higher

modulus materials the critical force is greater and thereby the traction forces applied.

Since the ridges and grooves select the direction for a cell to align by directing its

cytoskeleton, these increased forces would travel along the ridge and result in increasing

the elongation of the cell, and the effect of contact guidance.






30


By examining textured surfaces with varying moduli while keeping surface

energy variations to a minimum, this project attempts to minimize some of the unknown

factors and study the effects of modulus on contact guidance. Before this can be

completed, a thorough investigation of the surface topography and chemistry must be

performed, and the results are discussed in the next chapter.














CHAPTER 3
CHARACTERIZATION OF SILICONE ELASTOMER SUBSTRATES


Introduction

The first step in this project was to fully characterize the silicone elastomers and

textured substrates. Four separate formulations of silicone elastomer with functionalized

additives and four with non-functional additives were used to vary the elastic modulus.

The surface energy of each formulation was measured by contact angles of solvents with

varying surface tensions. Dynamic contact angles in water were taken of the unmodified

silicone elastomer with and without texture. Samples were treated with both

radiofrequency glow discharge (RFGD) plasma and adsorbed fibronectin. Textured

surfaces were provided at two depths of features, and the fidelity of the surface pattern

was characterized by non-contact optical profilometry.


Materials

The base elastomer system, referred to in this thesis as "unmodified", is the

Silastic T-2 Silicone Moldmaking Rubber produced by Dow Coming. The resin is a

dimethylvinyl-terminated polydimethylsiloxane (PDMS) composite mixture composed of

PDMS and trimethylated silica for mechanical stability. It is a translucent resin that cures

via addition polymerization with a platinum catalyst when added in a 100:10 ratio with

the Silastic T-2 curing agent. The typical properties of the base resin and elastomer

system are given in Table 3.1. Resin was provided in 45 lb. containers and curing agent









provided in 4.5 lb. containers. To simplify the production, portions of the resin and

curing agent were transferred into opaque HDPE bottles.


Table 3.1 Typical properties of Silastic T-2 Silicone Moldmaking Rubber (from Dow
Corning product information sheet)
Test Unit Result
As Supplied
Base Color Translucent
Viscosity Centipoise or mPa-s 50,000
Curing Agent Color Transparent
Viscosity Centipoise or mPa-s 550

As Mixed 100 Parts Base to 10 Parts Curing Agent by Weight
Viscosity Centipoise or mPa-s 55,000
Specific Gravity 1.12

As Cured 24 Hours at 25C
Durometer Hardness, Shore A Points 42
Tensile Strength Psi 800
Elongation Percent 300
Tear Strength, Die B ppi 120
Linear Shrink Percent < 0.1


Two vinyl terminated linear PDMS oils and one bulky vinyl terminated oil were

added to the resin to change the mechanical properties of the elastomer. A variety of

non-functional PDMS oils of varying molecular weights were also examined for surface

energy analysis. The molecular formula, viscosity, and molecular weight of each

additive are given in Table 3.2. All were obtained from Gelest, Inc. and stored at room

temperature. Five solvents were used for contact angle analysis: 1-Propanol (Aldrich,

99+% spectrophotometric grade), acetonitrile (Aldrich, 99.93+% HPLC grade), N,N-

dimethylformamide (Aldrich, 99.9+% HPLC grade), diiodomethane (Aldrich, 99%), and









ultra-high purity DI water. Bovine plasma fibronectin was received from Sigma (cat # F-

4759, 2 mg) as a lyophilized powder.


Table 3.2 PDMS additives to silicone elastomer system
Name Molecular Structure Viscosity MW Catalog
(cSt) (g/mol) Number
Vinyl terminated CH3 CH3 CH3 2-3 550 DMS-V03
PDMS H2C=C-Si-d-Si- -Si-c=cH2 1000 28,000 DMS-V31
H I n I H
CH3 CH3 CH3

Vinyltris(trimethyl- / N/A 322.70 SIV-9300
siloxy)silane s---o s'

Si- -



Trimethylsiloxy CH3 CH3 CH3 50 3,780 DMS-T15
terminated PDMS H3C-- i-O -- I--CH3 5000 49,350 DMS-T35
I L I J n I
CH3 CH3 CH3
Tris(trimethylsiloxy) N/A 296.66 SIT8721.0
silane -o --


SI--
s-Sl--/


Methods

Elastomer Preparation

Unmodified elastomer substrates were produced according to manufacturer

instructions. In a tri-cornered polypropylene beaker, resin and curing agent were added

in a 100:10 resin to curing agent ratio by weight. The two components were mixed with

a metal spatula until well incorporated. During the mixing process, many air bubbles

were trapped in the resin, and so the mixture was degassed under vacuum for 10-15









minutes, periodically breaking the vacuum to rupture the bubbles formed. The uncured

elastomer remains workable for approximately 1 hour.

For modified samples, oils were added to the resin before the addition of curing

agent at the appropriate concentrations and mixed together. Curing agent was then added

and the procedure was followed as above. For non-functionalized oils, the resin to curing

agent ratio remained 10:1. For vinyl-terminated oils, the curing agent percentage was

increased to account for the increase in vinyl groups. Unmodified samples were cured at

room temperature, while vinyl-terminated samples were cured at 800C for the appropriate

amount of time, as seen in Table 3.3.




Table 3.3 Curing conditions for silicone elastomer samples
Sample Curing Agent Needed Cure Time and
Temperature
5% 2-3 cSt vinyl term (5% LMW) 0.16 x mass resin 4.5 hr at 80C
5% 1000 cSt vinyl (5% HMW) 0.1011 x mass resin 2 hr at 80C
15% vinyl tris 0.26 x mass resin 12 hr at 80C
Unmodified and nonfunctional oils 0.10 x mass resin 24 hr at RT


Modulus Determination

Modulus values were obtained by Leslie Wilson and Amy Gibson and the values

are used for material selection purposes only. A brief explanation of the method used is

included here. Tensile specimens were made using an ASTM D1822-68 type L dogbone

die, resulting in a 1-inch gauge length. Each sample was individually measured and the

thickness of the samples was -1 mm. Tensile measurements were made according to

ASTM D412-97 on an Instron model 1122 equipped with the TestWorks 3.07 software

for analysis. The dogbones were tested via crosshead displacement at a rate of 2









inch/min. Measurements of the modulus were calculated from the linear portion between

the stress of 0.2 and 0.5 pounds and recorded.

Preparation of Textured Surfaces

All textured surfaces were taken from silica wafers with textures etched using

microfabrication technology. Wafers were provided by Chuck Seegert at two etch

depths, 5 |tm and 1.5 |tm. Briefly, silicon wafers were coated with a thin layer of a

photosensitive polymer and then exposed to UV light through a photomask imparting

various 5 |tm patterns onto the photoresist as seen in Figure 3.1.


A






5,000 p --25,000 jm


10,000 mn .....






Figure 3.1 Representation of etched patterns on silicon wafers. Each square side is
10,000 rm long and is made up of 5 [m wide ridges, separated by varying groove depth.
Each square is separated into thirds with the groove depth in each third 5 rim, 10 rim, or
20 rm wide. Square A has ridges 60 rm long with 40 rm in between. Square B has 5
lm square pillars. Square C has 10,000 [m long continuous ridges and square D has 800
rm long ridges with 200 [m smooth space in between.



After development, the wafer was etched using reactive ion etching to the desired

depth. The 5 |tm deep wafer was patterned here at the University of Florida, and sent to

Unaxis, USA for etching via the Bosch process. The 1.5 |tm wafers were etched in









house. Each wafer was then glued to a metal backing with epoxy to improve mechanical

stability. Before casting films for the first time, the wafers were treated with

hexamethyldisilazane (HMDS) to minimize adhesion to the wafer.

Textured substrates were cast either directly on the wafers or on epoxy master

copies. For the direct wafer copies, wafers were placed on a thin sheet of

polyethyleneterephthalate (PET) taped to a glass plate and uncured elastomer was poured

onto the wafer. Any trapped air bubbles were pierced with a needle. A second glass

plate with a PET sheet and spacers was placed on top to form a constant thickness film of

3 mm. After curing, the plates were separated and the film removed from the wafer.

For the production of the epoxy master, PDMS copies of the wafers were made

with an accelerated cure. The desired texture squares were then cut out and placed

texture side down on a PET covered glass plate. Both 1.5 |tm and 5 |tm deep patterns

were placed on the same plate. After positioning the textured surfaces, uncured

elastomer was poured over the backside of the textured PDMS. The result was a PDMS

film of constant thickness with individual squares of texture. This film was then used as

a mold to cast epoxy (Epon 828 epoxy resin and Jeffamine D-230 hardener) over. The

epoxy was cured at 800C overnight and the PDMS film removed. The resulting epoxy

master is a direct replica of the wafer texture. Unmodified textured surfaces were cast off

of both epoxy and wafers, while modified silicones were cast mainly off of the wafers

directly due to problems with adhesion to the epoxy at elevated cure temperatures.

Surface Treatment

Samples for the cell adhesion portion of this project were either left untreated,

treated with a RFGD plasma, or coated with an adsorbed layer of fibronectin.









Radiofrequency glow discharge treatment

To improve the wettability of the substrate, surfaces were exposed to an argon

RFGD plasma treatment. The plasma system used was a RF Plasma Inc. HFS 401S

instrument, set at 50 W. Four samples in a polystyrene petri dish were treated at a time,

approximately 5.5 cm from the bottom of the plasma RF coils, and after the samples were

in place, the plasma chamber was evacuated for 10 minutes. Pressure in the plasma

chamber got as low as 10 mTorr before treatment and typically around 25 mTorr

immediately after the plasma was switched on. The argon gas was introduced at a flow

rate of 200 sccm and the RFGD treatment was activated for 5 minutes. After treatment,

the pressure was equalized in the chamber with ambient air.

Fibronectin adsorption

Fibronectin adsorption on microtextured elastomer samples was performed via the

method of Ostuni and Whitesides.[64] Lyophilized bovine plasma fibronectin was

received from Sigma and the contents were dissolved in 2 mL of 0.22 [m filtered water

at 37C for 45 minutes. The solution was diluted to 50 [g/mL in Hank's balanced salt

solution (BSS). Elastomer samples cut into 15 mm disks were sterilized with 70% EtOH

and rinsed 3X with BSS, then placed in 24-well culture plates and covered with 0.5 mL

of the fibronectin solution. To ensure that fibronectin displaced air trapped in textured

surfaces, the culture plate was exposed to house vacuum for 1 minute. During this time,

small bubbles formed on the surface in the same pattern as the texture. After these

bubbles detached, the vacuum was released and the fibronectin solution left on for 1 hour

at ambient conditions. Samples were then rinsed three times with BSS.









Surface Energy

Untreated elastomers were examined for surface energy differences due to the

additives. Sessile drop contact angles were measured for five solvents on each substrate.

Five drops each of ultrapure water, methylene iodide (Mel), 1-propanol (1-prop), N,N-

dimethylformamide (DMF), and acetonitrile (ACN) were measured for each substrate,

and repeated on a second sample. Each drop was 2 gL as dispensed from a 25 gL pipet.

An optical goniometer was used and the left and right contact angles were measured

immediately after each drop was placed. Thus, 20 readings per liquid per sample were

obtained. A Zisman plot was utilized to determine the surface energy of each substrate.

Captive bubble contact angles were taken on untreated, plasma treated, and

fibronectin adsorbed unmodified elastomer. The substrates were placed treatment side

down in a PMMA jar containing BSS. Bubbles of 2-4 [L of air were introduced into the

BSS under the sample so they attached to the treated surface. Five bubbles for replicate

samples were measured in the same manner as for the sessile drop method.

Dynamic Contact Angles

Dynamic contact angles were taken on a Cahn dynamic contact analyzer using

the Wilhelmy plate technique. Briefly, Wilhelmy plate contact angles are taken by

advancing and withdrawing a film into a liquid. The force on the film is measured and

correlated to the film displacement. Textured and untextured unmodified elastomer and

samples with non-functionalized PDMS oligomers were examined in ultra pure water in

clean glass beakers. Samples examined were unmodified, 5% and 20% 50 cSt oil

additive, and 5% and 20% 5000 cSt oil additive. Smooth films of measured thickness

were cast in between glass plates with spacers. Textured surfaces were cast off

polystyrene copies of the 5 gm wafer.









Films were cut into rectangles with dimensions approximately 10 mm wide and

30 mm long. The thickness was approximately 3 mm for the untextured films, but varied

for the textured surfaces since they were cast off a PS copy with no back plate. Lower

viscosity textured samples spread more and had decreased thickness when compared to

the unmodified samples. The perimeter of the advancing cross-section was measured for

each sample, and three samples were examined for each setting.

First, clean mica strips were dipped in the water at a stage speed of 100 gm/s.

Surface tension of the water was calculated assuming perfect wetting of the mica.

Textured surfaces were arranged so that 10 mm of untextured area is inserted into the

water first, until 10 mm of textured lines running the width of the sample reached the

water interface as demonstrated in Figure 3.2.



Direction of
Sample
Textured Immersion
SUntextured



Figure 3.2 Sample layout for textured substrates examined using DCA



Sample depth was set at 25 mm to include all texture and single dips were used

for all samples. Advancing and receding contact angles were calculated for the smooth

and textured areas using the Cahn DCA4A software package.

Optical Profilometry

Topographical characterization of both the textured elastomer samples and the

wafers directly was performed using optical profilometry. The Wyko NT1000 from

Veeko instrumentation uses non-contact interferometry with optical light to map the









surface in 3-D without affecting the surface properties or deforming the substrate.

Magnification of the substrate can vary from 2.5X to 100X using the objective and field

of view lenses provided. Samples were prepared as described above and needed no

additional treatment. Both 1.5 [im and 5 [im deep samples and wafers were examined at

50X and 5X. The samples were leveled on the stage and focused before the scan was

run. Profilometry data and 3-D rendering of the surface was accomplished using WYKO

Vision32 version 2.210 software package.


Results and Discussion

Modulus Results

Since the silicone formulations and modulus testing was performed by others in

the group, an extensive discussion on the theory and practice will not be included here.

The samples were provided as materials that should have similar surface energies but

with varying moduli. Figure 3.3 is a graph of the modulus of each material. Note that

there are two values for the unmodified elastomer, one which is cured at room

temperature, and one cured at 800C. It is important to note that the accelerated cure

produced an almost 25% reduction in modulus. For this reason, all samples were cured at

the set conditions for that formulation. The striped columns in Figure 3.3 represent the

samples with functionalized additives. The 5% LMW and 5% HMW represents a 5%

addition of the 2-3 cSt (-550 g/mol MW) and the 1000 cSt (-28,000 g/mol MW) vinyl

terminated PDMS oils respectively. The 15% vinyl Tris is the lowest modulus material,

and represents a 15% addition of the vinyl tris(trimethylsiloxy)silane molecule. The non-

functionalized oils all decreased the modulus and should not have crosslinked into the

network due to its methyl endcaps. The 20% 5000 cSt oil additive actually had a










"greasy" surface, where a film of oil that migrated to the surface was visible to the naked

eye.


Figure 3.3 Elastic modulus of modified silicones as measured by tensile testing.
Striped bars represent elastomer with functionalized PDMS additives. (Data obtained by
Leslie Wilson and Amy Gibson)



Surface Energy of Silicone Elastomers

As mentioned before, the surface energy of the elastomer formulations was

determined using contact angles. Zisman and his group first introduced an empirical

method of treating contact angle data to estimate Ys, the surface free energy of the

solid.1118, 123] The plot of cos 0vs. yi, the surface tension of the liquid, form

approximately a straight line with the formula


coso = -b(y1 -yV)


Elastic Modulus of Modified Silicones

3


25
M5% LMW
SUnmodified
2 05% 5000 cSt
55% 50 cSt
O Unmodified 80C
1 5
S15% HMW
T 20% 5000 cSt
S20% 50 cSt
S120% Tris
S15% Vinyl Tris
05









where y/ is the critical surface tension, below which 0is zero and the surface is perfectly

wetted. This is considered to be a measure of the surface free energy of the solid. The

rationale behind this stems from Young's equation, represented as


Ys = s/+ 1 Cos 0

where represents the surface tension (or free energy) and the subscripts s, sl, and 1 refer

to the solid-vapor, solid-liquid, and liquid-vapor interfaces respectively.[123] Essentially

this is a force balance between interfaces of a drop on a surface, with the solid-vapor and

solid-liquid opposing each other in the plane of the solid. The basis of this theory is that

with decreasing y/ towards ,, then the solid-liquid surface tension is minimized and the

solid-vapor surface tension will equal the liquid-vapor surface tension or surface free

energy.

Contact angles were taken using the sessile drop method with 5 different liquids

of known surface tension on the surface. Ten readings were taken for each sample and

liquid by measuring the angle on both sides of the drop. The liquids used and their

corresponding surface tensions can be found in Table 3.4.




Table 3.4 Surface tension of liquids used for surface energy determination by contact
angle analysis[124]
Liquid Surface Tension (mN/m)
Water 73.05
Methylene Iodide (Mel) 50.76
N,N-Dimethylformamide (DMF) 37.1
Acetonitrile (ACN) 29.30
1-Propanol 23.78









Contact angles were measured for these liquids on unmodified elastomer, 5% LMW, 5%

HMW, 15% vinyl tris, 20% tris, 5% 50 cSt, 20% 50 cSt, 5% 5000 cSt, and 20% 5000 cSt

coated slides. For each sample, cos 0was plotted for each liquid versus the surface

tension, and linear regression was performed to determine the surface free energy where

cos = 1. The surface energy and contact angles of each substrate can be found in Table

3.5. A representative example of a Zisman plot from these data is shown in Figure 3.4.




Table 3.5 Contact angles and surface free energy of various substrates
Average Contact Angles (degrees) Yc
Samples Water Mel DMF ACN 1-Prop (mN/m)
Unmodified 109 +4 67 + 4 55 + 2 47 + 4 32 + 2 19
5% 50 cSt 110+3 66+ 1 56+2 44+3 31 1 20
20% 50 cSt 109 1 64 2 54 +2 46 +3 25 +1 21
5% 5000 cSt 108 +2 65 + 2 55 + 1 48 + 2 26 + 2 20
20% 5000 cSt 103 +2 64 + 2 52 + 2 48 + 2 26 + 2 19
20% Tris 106 +2 64 +2 56 +2 46 +2 29 + 1 19
5% LMW 107 5 65 3 55 +2 47 +2 29 +3 19
5% HMW 112 2 64 2 56 +2 46 +3 25 +3 21
15% vinyl 107 +2 66 3 58 3 50 5 33 + 3 17
Tris

As seen by Table 3.5, there is very little difference between all the samples in

terms of surface free energy. This is an expected outcome since essentially the samples

are all made of silicone elastomer with different PDMS oligomer additives. Since the

material is the same, the surface energy should be similar. In addition, to minimize the

energy at the interface, low surface tension oligomers have been shown to migrate to the

surface.[116' 117] The fact that this does not affect a change in surface energy is supported

by the results for the 20% 5000 cSt samples of no appreciable difference in surface

energy even with a visible layer of oil on the surface.











Zisman Plot of Unmodified Silicone
5 Liquid System



8 700 7,y = -42.301 cos 0 + 61.235


S40
30
8 20
I-0

C 0
u, 4-

-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
cos 0


Figure 3.4 Zisman plot of unmodified silicone for calculation of surface free energy



Contact Angles of Treated Surfaces

The two main surface treatments employed in this study are RFGD plasma

treatment and adsorbed fibronectin. To determine the relative wettability of the surfaces,

captive bubble contact angles were measured for plasma treated, fibronectin treated, and

untreated unmodified samples. Results are reported in Table 3.6. As seen by the low

angles on fibronectin and plasma treated surfaces, the hydrophilicity of the substrate has

been greatly increased compared to the hydrophobic unmodified PDMS.


Table 3.6 Captive bubble contact angles on treated surfaces
Contact Angle (degrees)
Unmodified 87 4
Fibronectin treatment 15 4
Plasma treatment < 10










Plasma Treatment Issues

The fibronectin-coated surfaces change their hydrophilicity by adsorbing a layer

of protein on the surface, but the plasma treated samples modify the surface chemistry of

the base polymer. The literature implies that two factors seem to be at play here. One is

the modification of the surface by creating free radicals that when exposed to air form

peroxides on the surface.[92] This is the method used to add adhesion molecules and other

chemistry to the surface through graft copolymerization. However, as the plasma

treatment increases, an oxidized silica-like layer forms on the surface, especially if the

plasma contains oxygen. This is typically a problem when using silicone coatings as

high-voltage outdoor insulators. Studies have shown this silica-like layer to be up to 150

nm thick on the surface. This layer eventually cracks and then allows the low MW

oligomers to migrate to the surface.[116] This was not considered to be a problem with the

plasma treatment of the samples, but in analysis of the surfaces it was noticed that many

of the plasma treated elastomers had cracks. These were not seen after removing the

samples from the plasma chamber. As the samples were peeled off the dish to be placed

in the wells, cracks were formed that were not there before or immediately after

treatment.

The main concerns with this phenomenon is the influence the cracks would have

on the cells compared to the intended topography and the effect that a harder silica-like

surface layer would have on the modulus. To examine this affect, a textured sample of

15% vinyl tris silicone was plasma treated under the same conditions as the samples and

examined using optical profilometry and atomic force microscopy (AFM), as seen in

Figure 3.5, Figure 3.6, and Figure 3.7. The sample was bent to simulate removal and










placement in a culture dish. Cracks on the order of 0.5-1.0 ipm were observed, and

appeared mainly as positive textures pushing up from the surface. Since the

transformation of the material from PDMS to a silica-like oxidized layer results in a large

decrease in specific volume, tensile stresses are formed in the oxidized layers. These

stresses are relaxed by the cracking and result in an elevated texture, as seen and

explained by Hillborg, Sandelin, and Gedde.[116]

Of concern also is the affect that the plasma treatment has on the shape and size of

the features. Profilometry data presented in Figure 3.8 shows a very dramatic change in

the height of the ridges, changing from 1.43 pm to 0.81 pm. It is apparent that the

plasma treatment at this level is ablating the surface and seriously altering the features.


1.06

Surface Stats:
Ra: 72.08 nrn 70
Rq: 137.05 mn o50
RtI: 2.08 um


Measurement Info: .
Ma.niflcati.ll 5150 .
Measurente Mode: VSI J -0 30
Sampling: 163.10 nm
Array Size: 736 X 480


.1 01
Title: 1.5 um 15% tris
Note: Plasma treated smooth


Figure 3.5 Profilometer image of surface damage due to plasma of 15% vinyl tris
textured elastomer sample (smooth area)








47




urn
SO.BG

o 70
Surface Stats:

Ra: 292.16 mrn 0.50

Rq: 348.73 m o ao







XM.anifcainI 51.50 91 4 -so

Measurmen= Mode: 151
Sampling: 163.10 im 5

Array Siz: 736 X480 -a 70




0 05
Title: 1.5 um 15% tries

Note: Plasma treated 20



Figure 3.6 Profilometer image of surface damage due to plasma of 15% vinyl tris
textured elastomer sample


Figure 3.7 AFM Image of crack on plasma treated sample


nm
551


Surface Stats: 400
Ra 132 07 rinm
P 300
Rq 186 91 nm
.I-m200
Rt 1 13 um
10

Measurement Info:
Magnification 1 00 -100
Measurement Mode VSI -20
Sampling 35 32 nra30
Array Size 512 X 512
-400

500
584
Title:

Note:















SI I

I I II I I II I. II


X 44 umr




I I 1
i ,, ', ,, I *


SB 10 eC 6 0 100 120


Figure 3.8 Profilometry data for non-treated (A) and plasma treated (B) 1.5 [im deep
elastomer textured surfaces at the 5 [im width spacing. Height of ridge: A = 1.433 rim, B
= 0.809 rm.



Another feature of using a silicone surface is the effect of rearrangement of the

surface properties due to mobility of the chains and migration of low MW oligomers. To

examine this effect, unmodified elastomer samples were treated at two plasma power

levels, 50W and 100W using the same procedure as in the methods section of this

chapter. Contact angles of water on plasma treated samples were examined using the

sessile drop method immediately after treatment, 15 minutes after treatment, and one

week after treatment and the contact angles reported in Table 3.7.


Table 3.7
over time


- Water contact angles on plasma treated silicone samples after exposure to air

Average Contact Angles (degrees)
Treatment conditions 50 W, 5 min 100 W, 5 min
Before Treatment 103 8 114 3
Immediately after treatment < 10 <10
15 minutes after treatment 15 4 <10
1 week after treatment 77 + 7 45 28


X47um
_14 -


_^-i I
- I I I
" ", ',17.
_F7 II









The data indicate that with the more intense plasma treatment, the hydrophilicity

of the surface is more stable over a period of a week, however, the high standard

deviation after 1 week exposure to air for 100 W sample implies that the surface has areas

of homogeneity that are rather extreme. Some areas had contact angles as low as 250

while others had angles -800. The stability of the surface is possibly due to a silica-like

crust as mentioned above, and the areas of hydrophobicity are due to low MW oligomers

migrating through cracks.

Dynamic Contact Angle

Dynamic contact angles are taken by advancing or removing a liquid interface on

a surface. This can be accomplished with a variation of the sessile drop method by either

adding or subtracting fluid, or by tilting the plate and measuring the angle. Another

technique is the Wilhelmy plate method, where a film of material is dipped in a liquid

and the force on the plate is measured. The force is related to the surface tension of the

liquid by the equation:

F = y Pcos 0

where P is the perimeter of the plate. By measuring the force and perimeter of a sample

in a known liquid, the contact angle can be determined for both advancing (inserting) and

receding (withdrawing) contact angles.

Dynamic contact angle data was taken on unmodified silicone and the linear non-

functionalized additives. Functionalized oligomers were not available at the time of

examination and the method of production had yet to be determined. In addition, samples

with both smooth and textured areas were examined. Figure 3.9 is a representative force-

distance curve taken from DCA data. The lower linear portion of the curve represents the










advancing or inserting portion of the dip. The force values are negative due to the fact

that the low-energy surface is resisting being wetted by the water, and in essence a

reverse meniscus is pushing up against the sample. The upper linear portion is the

receding area of the curve. The separation of the two curves is a measure of the

hysteresis, or the difference between the advancing and receding contact angles,

represented as AOin Table 3.8. The slope of the linear portions is due to a buoyancy

effect and is factored out by extrapolating the line to the point of sample contact with the

water, or zero depth of immersion (ZDOI).


DCA of Unmodified Silicone Elastomer in Water

300
200
100
Receding Area
0 0
-100 2 4 6 10 12 14 15
S-200 -ZDOI -
-300 Advancing Area
-400
-500
Position (mm)


Figure 3.9 Force-distance curve from DCA of unmodified silicone elastomer in water



Hysteresis is caused by multiple factors, especially roughness and surface

heterogeneities. For most smooth surfaces, roughness under 0.5 .m contributes

minimally to the hysteresis of a sample.[125] Hysteresis due to rearrangement of the

polymer backbone so that more hydrophilic moieties are exposed to the surface play a

large role in the decrease of receding contact angles compared to advancing.[123, 125]









Essentially, as the surface is in the receding phase, the meniscus is traveling over a

previously wetted surface with a different surface energy than before. Due to this

phenomenon, the advancing contact angle can be considered a measure of the low-energy

portion of a heterogeneous surface and receding angles are more characteristic of high-

energy parts.[125] In examining Table 3.8, one aspect that stands out is the small hysteresis

of the 20% 5000 cSt samples. This is most likely attributed to the fact that these samples

had a visible coating of oil on the surface, which may mask the rearrangement process, or

at least minimize the effects.




Table 3.8 DCA data on silicone elastomer modified with non-functionalized PDMS
oligomers
Viscosity Wt. % 0adv 0rec AO
Unmodified 0 115.1 3.8 68.7 2.2 46.4 1.7
50 cSt 5% 113.9 1.8 77.5 1.8 36.4 0.3
50 cSt 20% 100.5 + 1.3 65.1 + 2.1 35.4 + 1.6
5000 cSt 5% 106.1 + 0.7 71.6 2.2 34.5 2.1
5000 cSt 20% 101.0 +0.8 89.4 4.4 11.6 4.4


Figure 3.10 is an example of a force-distance curve taken on a textured substrate.

The sample advances into a smooth area, and then into an area of texture. At the end of

the advancing dip, the sample is retracted through the textured area and then on to the

smooth area. Several factors preclude the presentation of numerical results for this study.

For many samples, the lengths of smooth and textured areas were not long enough to

achieve a linear region of stable contact angles. In addition, the varying thickness and the

fact that only one side of samples were textured make numerical contact angles

unreliable. However, for the majority of the samples, the trend was similar. The

advancing textured areas had higher contact angles than the smooth areas, while the










receding textured contact angles had wide variations. Due to the high aspect ratios and

significant roughness of the textured areas, air is trapped in the grooves when the

advancing front of the liquid passes over. As can be seen by Figure 3.10, the advancing

force decreases as the liquid front reaches the texture, which in effect is increasing the

resistance of the water to the film penetration. As it encounters areas of low surface

energy air trapped in between the grooves, the water adheres to the ridges until the

surface tension is overcome and it can bridge to the next ridge.



Textured Silicone Elastomer
20% wt/wt. 5000 cSt oil


150
100
50
0
-50
-100
-150
-200
-250
-300
-350


Position (mm)


Figure 3.10


Force distance curve from DCA on a textured silicone elastomer substrate


Optical Profilometry

Optical profilometry is a useful tool for characterizing the shape and 3-D features

of textured substrates. It uses white light interferometry to determine the topography of a

substrate without contacting the surface. It also has the great advantage that no surface








modification or contact is necessary to take measurements. Imaging procedures like

SEM require a surface coating and high-energy treatment of the surface, which may

change the properties as well as rendering the sample itself unusable in the future.

Figure 3.11 shows representative 3-D renderings of an unmodified silicone copy

off the 5 gm deep textures. Figure 3.12 includes profilometry images taken directly off

the wafer for comparison. The textures in Figure 3.11 were replicated from epoxy,

demonstrating the ability of the epoxy system to faithfully recreate the wafer's features.

Note that these copies are in effect the "negative" of the wafers, in that the ridges have

varying widths and the grooves are a constant width and depth of 5 gm.


A 5 npm wide


B 10 pm wide


C 20 pm wide D 10 pm wide

Figure 3.11 3-D images of 5 gm deep elastomer samples copied off of epoxy taken with
optical profilometer (50X). (A-C) are examples of the ridges at the different spacing. (D)
demonstrates the smooth/textured interface


l UJi MUInBf






















A 5 pnm wide B 10 pm wide











C 20 pm wide D 20 pm wide

Figure 3.12 3-D images of 5 .im deep etched silicon wafers at each width spacing,
taken with optical profilometer. (50X A-C, 100X D)



Figure 3.13 is an example of the features from a 1.5 .im deep wafer. These were

cast off the wafer with 5% LMW elastomer and cured at 800C. The left side of the image

is elastomer copies of the wafer, which is represented on the right side. Replication is

good, but the shape of the 1.5 [tm ridges are more rounded and pointed at the top after

replication (see Figure 3.8A for 2D profilometry of 5 [im spacing). Table 3.9 gives the

quantitative values for ridge depth and width taken with the optical profilometer. All of

the ridge widths for the wafers are 5 [im since the groove spacing is varied. This

translates to variations in ridge spacing for the elastomer copies. What is apparent is that









the etching process does not give perfectly spaced features, especially with the 1.5 [tm

wafer. The 5 tim deep wafer has better retention of expected ridge values, most likely

due to a more controlled reactive ion etching process.


Table 3.9 Optical profilometer data of ridge widths and groove depths of the 1.5 and 5
lm deep wafers and elastomer copies. The wafers should have a constant ridge width
and the elastomer copies have ridge widths of 5 rim, 10 rim, and 20 rim.
Sample Ridge Width (tm) Groove Depth (rtm)
5 um wafer 4.8 0.1 4.9 0.2 4.9 0.2 5.3 0.2
5 nm elastomer 4.7 0.1 9.2 0.4 19.1 0.1 4.9 0.1

1.5 m wafer 3.8 0.1 3.4 0.2 3.7 0.1 1.4 0.1
1.5 [m elastomer 3.7 0.3 8.2 0.2 17.6 0.5 1.4 0.10


S5 d P


A 5 1il wide PDM)lS


B 20 jpm wide PDMS


ru uTV


C 5 lun xvide wafer


D 20 pim w-ide wafer


Figure 3.13 3-D images of 5% LMW elastomer copies (A, B) of a 1.5 tim wafer (C, D).









Some of the limitations of the optical profilometer system are a result of the scale

and arrangement of the features. As can be seen by comparison of Figure 3.11 and

Figure 3.12, both have similar shapes of ridges, with the sidewalls sloping downward so

that the walls appear to not be vertical. Since the profilometer is non-contact and requires

the reflectance of light, as the features get deeper and closer together, some data can be

lost if the magnification is not high enough or if the material does not reflect light out of

small features. For spacing greater than 5 rim, and magnifications of 50X or more, this

problem is not as great. This instrument is a good addition to the analytical capabilities

of a researcher since it is easy and quick to use with little to no sample modification, and

gives a quality replication of the image surface in three dimensions. With the analysis

reported in this chapter, coupled with the background in contact guidance and cell

growth, a better understanding of the factors and variables involved in the growth of cells

on textured surfaces is possible.














CHAPTER 4
CONTACT GUIDANCE OF ENDOTHELIAL CELLS


Introduction

In the previous chapters of this thesis, the principles of contact guidance and the

materials used for this study have been detailed. The final step of this project was to

examine the contact guidance phenomenon on a novel system to determine the effects of

elastic modulus on a surface's ability to direct cell growth. Contact guidance was

quantified on a group of silicone elastomers with elastic modulus values varying almost

800% but with similar surface energies. Textured surfaces made from four elastomer

formulations were seeded with porcine vascular endothelial cells (PVECs). Cell nuclei

were imaged and the nuclear shape was compared for surfaces of varying ridge

dimension, groove depth, and material modulus.


Materials

Elastomer Substrates

As described in the previous chapter, the unmodified base elastomer is the

Silastic T-2 Silicone Moldmaking Rubber produced by Dow Corning. Three vinyl

terminated PDMS based oligomers were included to give the elastic modulus a range

from 0.3 MPa to 2.3 MPa. Pertinent information about the elastomer systems, including

modulus and surface free energy, Ys, is listed in Table 4.1. The vinyl-terminated

oligomers were selected over non-functionalized oils due to their incorporation into the

network structure and to minimize additive release.









Table 4.1 Properties of elastomeric substrates for contact guidance experiments
Sample Reference Additive Modulus (MPa) 4 (mN/m)
Unmodified None 1.4 + 0.1 18.9

5% LMW 5% Vinyl terminated PDMS 2.3 0.5 19.1
550 g/mol
5% HMW 5% Vinyl terminated PDMS 1.0 + 0.1 21.0
28,000 g/mol
15% vinyl tris 15% Vinyltris- 0.3 0.1 17.1
(trimethylsiloxy)silane


Cell Culture and Seeding

Bovine plasma fibronectin was received as a lyophilized powder from Sigma

(cat# F-4759, 2 mg). PVECs were obtained from Dr. Edward Block's lab at the Malcom

Randall VAMC in Gainesville. Endothelial cells were obtained from the main

pulmonary artery of 6 to 7 month old pigs and were propagated in monolayer cultures and

characterized as described by Patel et al.11261 Third to sixth-passage cells in postconfluent

monolayers maintained in RPMI 1640 medium (Life Technologies, Grand Island, NY)

containing antibiotics (100 U/ml of penicillin, 100 [g/ml of streptomycin, 20 [g/ml of

gentamicin, and 2 [g/ml of Fungizone) were used in all studies. Fetal bovine serum

(FBS, was added at a 10% concentration for plating and growth. Falcon tissue culture

flasks and 24 well plates were used for cell passaging and seeding. Trypsin/EDTA 1X

solution was stored in frozen 10 mL aliquots. Hank's balanced salt solution (BSS) was

used for washing flasks and samples. PVEC nuclei were stained with Hematoxylin 2

from Richard Allan Scientific. Cytoplasm and various features were stained with a 1%

aqueous crystal violet solution.









Methods

Elastomer Sample Preparation

Silicone elastomer samples were prepared as described in the previous chapter.

The four elastomer formulations reported in Table 4.1 were used for this portion of the

study. Cure times and curing agent amounts are listed in the previous chapter, Table 3.3.

For all samples, constant thickness was maintained at 3 mm with the use of appropriate

spacers between glass plates. Textured sample were cast directly off silicon wafers with

the patterns etched into the surface. Untextured films were cast between two PET sheets

attached to glass plates. Samples were cut out using a punch with an interior diameter of

14.37 mm and outer diameter of 15.53 mm. All samples were sterilized in the same

fashion by rinsing with 70% EtOH and drying overnight in a sterile hood.

Surface Treatment by Fibronectin

Some surfaces were coated with fibronectin (FN) as described in the previous

chapter. Sterilized samples were placed in a 16 mm diameter well of a 24 well place,

and a 50 [tg/mL solution of FN was added in 0.5 mL aliquots to each well with a sample.

After exposure to vacuum to remove trapped air, samples were left to incubate for 1 hour

at room temperature. The FN solution was aspirated out and then the samples were

washed 3X with Hank's BSS.

Surface Treatment by RFGD Plasma

Textured surfaces were exposed to an argon RFGD plasma at 50 W for 5 minutes

as described in the previous chapter after sterilization by EtOH. The argon regulator was

set at 20 psi and the flowrate was 200 sccm. Samples were treated 4 at a time, 5.5 cm

below the RF coils, with one sample from each material per treatment to ensure similar

surface modification between batches. After treatment, the samples were moved to a









sterile hood and left exposed to air for 10 minutes for each sample, then transferred to a

24 well plate.

Cell Culture Techniques

PVECs were supplied by Bert Herrara from Dr. Edward Block's lab weekly as

suspensions in 12 mL of media. All cells received from Dr. Block's lab were between

passage 2 and 5. Cells were expanded by diluting the suspension to 20 mL with fresh

media and subsequently transferred as 10 mL aliquots to a 75 cm2 angled neck, vented

tissue culture flask. Flasks were incubated at 370C and 5% CO2 for 48 hours, and then

existing media was exchanged for fresh. Media was changed every 72 hours after that.

Typically, the PVECs formed a confluent monolayer on the culture flask within 72 hours

from initial plating.

Cell passage procedure

To passage the cells, confluent flasks were washed 3X with Hank's BSS. A few

mL of trypsin/EDTA 1X solution was poured in the flask and swirled to counteract any

remaining serum proteins and the remaining liquid was poured off. A small amount (-1-

2 mL) of fresh trypsin/EDTA solution was added again, with just enough to coat the

bottom of the flask. The flask was then placed in the incubator at 370C for 5 minutes,

and then checked with the inverted microscope to determine the appearance of the cells.

Once the cells became rounded, the sides of the flasks were struck on each side to

dislodge the remaining adherent cells. Serum containing media was added to counteract

the trypsin, and the suspension was mixed by aspiration with a 10 mL pipet. The

suspension was split into 3 flasks from each original flask and left to incubate as before.









Determination of cell suspension concentration

To determine the average amount of cells per mL in a cell suspension, a

hemacytometer was used. A hemacytometer has a chamber with 10 separate grids each

measuring 1 mm square. A coverslip was placed over the counting area, which resulted

in a well 0.1 mm deep. Cell suspension was transferred by pipet to the edge of the

coverslip, where the suspension was drawn into the chamber. Counts of 10 different

squares were made, and the average count was multiplied by 104 to determine the average

cells per mL in the suspension. If necessary, suspensions were diluted to achieve a cell

count between 1 x 105 cells/mL and 2 x 105 cells/mL before seeding on samples.

Cell Seeding on Samples

For all samples, 1 mL of cell suspension was seeded into a well. Each sample has

approximately 2 cm2 of surface area, and at a seeding density of 2 x 105 cells/mL, then

approximately 1 x 105 cells/cm2 was added to the sample. For fibronectin-coated

surfaces, the cells were suspended in serum free media since the adhesion protein was

already adsorbed on the surface. Cells were seeded in normal 10% FBS media on plasma

treated and untreated surfaces.

After seeding, the samples were incubated at 370C and 5% CO2. After 48 hours,

the media was changed to remove non-adherent cells and replace with fresh media.

Plasma treated surfaces exhibited an initial period where endothelial cells clumped into

groups (after 24 hours), and then eventually spread out to a monolayer. Cells on these

surfaces were left to grow for 5 days. Fibronectin coated surfaces showed improved cell

spreading and attachment, and cells were imaged after 48-72 hours.









Cell Staining and Image Capture

Many methods were examined to effectively stain the cells. To image the entire

cell body and morphology, cells were stained in an aqueous 1% crystal violet solution.

First, after removing the media, cells were washed 2X with BSS, then fixed with cold

10% n-buffered formalin for 20 minutes in the culture well. After removal of the

formalin, fixed samples were washed with BSS and crystal violet was added and cells

stained for 20 minutes. Samples were removed from the wells using a needle to

minimize flexing the substrates. The samples were washed in saline and placed on a

slide, then covered with a coverslip to prevent drying of the surface.

To image only the nucleus for the nuclear form factor, a hematoxylin stain was

used. Hematoxylin stains nuclear materials, specifically basophilic structures such as

DNA and RNA and was therefore chosen so that nuclear elongation could be used to

quantify contact guidance. The method followed the Sigma method for Gill's

hematoxylin staining. Briefly, cells were fixed in 95% EtOH for 10 minutes, rinsed 2X

with tap water, and stained for 2 minutes in hematoxylin. Longer staining times

increased the intensity of the nuclear stain, but also increased staining of the cytoplasm.

After the hematoxylin, the samples were rinsed 2X in water and 2X in 95% EtOH. The

substrates were placed on a slide and covered with a coverslip.

Cells were imaged on the surface at 200X using a Nikon Optiphot microscope and

Matrox image capturing system. Multiple images were taken at each feature width that

included at least 5 nuclei. Images were saved as ajpeg format and imported to Adobe

Photoshop for analysis.










Image Analysis

Images were analyzed to measure the ratio of the length to the width of each

nucleus. A schematic of the measurements taken can be found in Figure 4.1.


0 = 90* Line b
Linea w x



Nucleus
Line a -






Figure 4.1 Representation of the values measured in the nuclear form factor. The left
side of the image is a top view of a nucleus on a microtextured substrate, while the right
side is a cross-sectional view. The length measured is represented by A, while the width
is represented by B. The nuclear form factor is Log (A/B) (Drawing by Chuck Seegert)



The imported image files were modified to improve the contrast between the nuclei and

the background using Adobe Photoshop 6.0. A pictorial explanation of the steps can be

found in Figure 4.2. All images were adjusted using the auto-contrast macro in the

software package (Figure 4.2B). Then, batches of images were opened that came from

the same sample since they all had the same angle of orientation. By drawing a measured

line along the length of a groove, and then using the arbitrary rotate command, the image

is automatically rotated by the same angle as the measured line. The end effect is to line

up the ridges vertically on the computer screen (Figure 4.2C). By recording the

commands and initiating a batch process over the range of the opened images, all of the

features for each sample were aligned.











Image Processing Technique






00-1
[. 1 1 .



r";=~ipl~U


Figure 4.2 Steps in image processing technique of hematoxylin stained nuclei. The
original picture (A) is contrasted (B), rotated to align the textures (C), and then measured
for length (D) and width (not shown). The reported value is log(Length/Width)




The benefit of the vertical alignment is to improve the efficiency of measuring the

nuclei. The procedure introduced by Dunn and Heathl10 requires the measurement of the

length and width of nucleus at its widest point, where the length measured is parallel to









the ridge, and the width is perpendicular. With the images aligned vertically, the

measurement lines can be constrained by the software to perfectly vertical or horizontal,

and so by using that method the maximum length and maximum width of each nucleus

was measured and entered into an Excel spreadsheet. A 5 x 5 grid was superimposed on

the images, and 5 nuclei were chosen per image, each from a separate square of the grid.

Using this method, at least 20 nuclei per sample setting were quantified.


Results and Discussion

Contact Guidance on Textured Surfaces

As discussed in Chapter 2, surface ridges and grooves typically act to elongate

cells along the ridge. This phenomenon is mainly reported in qualitative fashion, and

fibroblasts are predominant as the cell system, since they are relatively easy to grow and

have shown good results in terms of contact guidance. Quantitative studies of contact

guidance on textured surfaces have used SEM and phase contrast microscopy to map the

entire cell body and try to determine its alignment. A serious limitation to this method is

the fact that unless the cells are isolated from others, the cellular dimensions and

alignment can be difficult to determine. From personal experience, cells that are not in

contact with others are typically more elongated and tend to show greater contact

guidance than confluent cells. Other methods to quantify contact guidance stain actin

filaments and other cytoskeleton components to detect the alignment of the interior stress

fibers with fluorescent or confocal microscopy.

For cells in this study, groups of cells were examined, rather than individual cells

apart from the rest of the culture. The main purpose of this project was to examine the

effects of modulus as a factor involved in the ability of a substrate to direct cell growth.









While modifying the underlying substrate's mechanical properties might alter a cells

ability to adhere and change its morphology, it will not direct cells to grow in a certain

pattern. In order to affect this change, the surface was patterned with microtexture in the

form of ridges, as characterized in the previous chapter.

Surface texture was examined using the unmodified elastomer as the reference

material. For this portion of the experiment, two main factors were examined, ridge

width and groove depth. As stated in the previous chapter, plasma treatment of the

surface resulted in small random cracks and an unreliable texture dimension. Due to the

concerns as to the effects of the plasma treatment to the fidelity of textures, fibronectin

adsorbed materials were examined most closely. Figure 4.3 is a main effects plot of

fibronectin coated unmodified elastomer. The average of all the data points is

represented by the dotted line, and the individual means are compared to this overall

mean. The term "Feature" in the graphs and subsequent analysis refers to the width in

microns of the ridges in the section examined. All of the grooves were 5 [m in width.

The term "depth" refers to the depth of the grooves in microns, either 1.5 rm or 5 jm.

Smooth images were taken from the same samples, at least 50 [im from the closest

texture.

Log (L/W) is defined as the nuclear form factor and is the variable that measures

the strength of nuclei alignment along the ridges. The more positive this variable is, the

more the nuclei exhibit contact guidance. Values greater than 0.15-0.20 are typically

very well aligned. The closer log (L/W) is to zero, the less the cell shows deference to

the topography at all. A negative number implies that the cells are guided orthogonal to

the textures.













Main Effects Plot Data Means for log (L/W)
Fibronectin Coated
Feature Width (pm) Groove Depth (pm)
024

019

014

0 009

004


5 10 20 smooth 15 50






Figure 4.3 Main Effects Plot of the data means of fibronectin coated unmodified
elastomers at various ridge widths (Feature) and groove depths (Depth)



The main effects plot helps to elucidate which factors and levels play a role in the

system. The more the data changes between levels, the greater the effect. Figure 4.4 is

an interaction plot generated by Minitab that demonstrates the changes at the different

factor levels for fibronectin coated unmodified elastomer. The important point that is

illustrated here is that at the deeper groove depths, the greater the alignment of the nuclei

to the ridges. One-way analysis of variance (ANOVA) and a multiple comparison test

(Tukey, 95% CI) on each depth comparing the different feature widths demonstrate a

significant difference between each level at the 5 [im depth, but only a statistical

difference at 1.5 .im between 5 .im wide ridges and smooth textures. One of the

difficulties in the analysis of these samples is in the fact that for the smooth samples, the

nuclei are not round, but elongated in random directions. The effect of this is to give a







68


mean value for log (L/W) for the smooth areas around zero, but with a large standard

deviation.



Interaction Plot Data Means for log (L/W)
Fibronectin Coated Feature
10
20
0.3 5
-smooth

0.2






0.0
0.0 - - - - - - --
I I
1.5 5.0
Depth




Figure 4.4 Interaction plot representing the change in contact guidance with depth and
feature width for fibronectin coated silicone elastomer.



Figure 4.5 and Figure 4.6 are main effects and interaction plots for plasma treated

unmodified elastomers generated by Minitab. These are the same plots as Figure 4.3 and

Figure 4.4, which are treated with FN. As can be seen in Figure 4.5, the difference

between the feature widths is not as obvious, and the effect of the depth on the alignment

does not play as large a role from 1.5 .im to 5 rim, as confirmed by the small slope.

A two-way ANOVA comparison on plasma treated unmodified elastomer

comparing the effects of feature size and depth show that when including the smooth

areas in the analysis, the feature size is a very significant contributor to variance (p <

0.0001) while depth is not a significant influence. By removing the smooth terms and












Main Effects Plot Data Means for log (UW)

Plaswm Treated
Feature With (um) roove Depth (m)

0.16

0.12

0.08

0.04

0.00

5 10 20 snooth 1.5 5.0



Figure 4.5 Main Effects Plot of the data means for plasma treated unmodified
elastomers. The left side of the plot compares effects due to the feature width, while the
right side compares the groove depth.



repeating the analysis such that the comparison is strictly on the textured surfaces, both

the width and the depth are significant sources of variance (p < 0.05).

One way ANOVA comparisons followed by a multi-comparison test (Tukey, 95%

CI) between the ridge widths at constant depth on plasma treated unmodified elastomer

show that for each width there is a significant difference when compared to smooth

samples, but the difference is not significant when compared to the other widths. To put

it simply, for plasma treated samples at both depths, each groove width shows

significantly more alignment when compared to a smooth surface, but there is no

statistical difference when comparing the different permutations of the three ridge widths.







70




Interaction Plot Data Means for log (L/W)
Plasma Treated
Feature, Jili .1''Ji
0.2
-* 10
20
smooth

0.1





0.0


1.5 5.0
Depth (ptm)




Figure 4.6 Interaction plot representing the change in contact guidance with depth and
feature width for plasma treated silicone elastomer.



In this graph the contact guidance data for PVECs grown on the neat sample with no

surface modification at 1.5 [im depth is included. Due to substrate production limitations

at the 5 itm depth, untreated samples were not available for analysis.

In comparing these results to the fibronectin coated surfaces, one point that stands

out is that for the 1.5 [tm deep FN coated substrates only the 5 itm wide ridges were

significantly different from the untextured surfaces, while at 5 [im there was a

statistically significant (alpha = 0.05) higher degree of alignment. The 1.5 [tm plasma

treated surfaces did have significant alignment compared to smooth surfaces, and a

higher log (L/W) value than 1.5 [tm fibronectin samples. However, due to the surface

irregularities of the plasma treated surface explained in the previous chapter and the fact

that a silica-like layer would expose the cell to a seemingly harder substrate; direct














0.4
-L
0.3

0.2

0.1

0.0
5 10 20 Smooth
-0.1

Ridge Width (um)
-- 1.5 um deep Plasma -- 1.5 um deep FN 5 um deep Plasma
5 um deep FN -- 1.5 um deep Untreated


Figure 4.7 Comparison of surface treatments on textured unmodified silicone elastomer
by ridge width and groove depth (error bars represent standard error of mean)



comparison of the two methods may not be appropriate due to material differences. The

untreated samples show a high degree of alignment at all texture settings. Cell

attachment and proliferation on these surfaces was significantly lower than on the treated

surfaces, but those cells that were attached were aligned at all levels of texture. The

morphology of these cells was highly elongated (compare Figure 4.8 and Figure 4.9) as

the cells attempted spread and stabilize. The fact that almost all of the cells showed a

high degree of alignment on the untreated surface suggests that when they come into

contact with a surface with less than ideal adhesion capability, the presence of surface

texture plays an increasingly important role.























Figure 4.8 PVECs grown on untextured Figure 4.9 PVECs grown on 5 [im spacing
fibronectin coated LMW sample stained 1.5 [im deep untreated LMW stained with
with hematoxylin hematoxylin



Contact Guidance on Textured Surfaces of Varying Modulus

The main unknown factor in the design of this project is the modulus of the

material and its effect on contact guidance. As discussed in Chapter 2, many different

materials, from metals to elastomers, have been examined in the study of direct cell

growth. In published research on contact guidance, the modulus of the sample substrate

is very rarely reported. Most materials used with different mechanical properties have

also significantly different surface energetic, which directly affects protein adsorption

and adhesion, making a comparison due to modulus improbable. These effects would

more than likely wash out any change due to the effects of mechanical properties. The

purpose of this section is to examine whether or not the modulus, as a measure of the

compliance, can significantly affect the ability of a material to direct cell growth in the

range covered by these materials.







73



Main Effects Plot Data Means for log (L/W)


.d .


Figure 4.10 Main effects plot comparing log (L/W) to feature width, feature depth, and
material used on fibronectin-coated elastomers.


Interaction Plot Data Means for log (LAV)
\- <


~\ r~o~go


Feature Width (im) 0.30

S10 -o0.15
-20
smooth -0.00
-0.30
Depth (um)
*5.0 --- .15

S1.5
5 0.00

Material






Figure 4.11 Interaction plot representing the change in factors of PVECs grown on
fibronectin coated materials.









Figure 4.10 is a main effects plot from Minitab similar to Figure 4.3 and Figure

4.5 on fibronectin-coated surfaces. As mentioned previously, due to the uncertainties

with the reproducibility of plasma treated surfaces, fibronectin materials will mainly be

examined. As seen before, the effects of depth and width of ridges seem to play the

largest role in the response of log (L/W). The relative flatness of the material interaction

implies that there is little contribution due to a change in modulus. In qualitatively

interpreting Figure 4.10 and Figure 4.11, it appears that contact guidance increases with

decreasing feature width and increasing depth on each of the materials, but a direct

correlation between the material choice and the degree of alignment is not obvious.

One way ANOVA comparisons followed by a Tukey multi-comparison test (CI = 0.95)

for the 5 [im depth materials show no significant difference between any of the materials

at the 5 [tm groove width. It is apparent from Figure 4.12 that the groove depth at that

point is much more significant and few trends can be found in relating the modulus. In

effect, the groove depth is overpowering any effect that modulus might have on the

contact guidance.

At the 1.5 tm depth, there are more significant differences with respect to

materials. At the 5 itm and 10 [tm ridge width, there is a significant difference between

the 5% LMW material and the 15% vinyl tris material with respect to log (L/W). The 1.5

ltm data is represented in Figure 4.13.

Note that as reported in Table 4.1, these represent the high modulus (2.34 MPa)

and low modulus (0.3 MPa) materials respectively. At 20 [tm ridge widths, there is no

significant difference between the two. Figure 4.14 compares the effect of ridge width

and groove depth strictly on the LMW and Tris surfaces. The trend of the materials is to


























Ridge Width (pm)
-- 5 FN LMW -- 5 FN Unmod -- 5 FN HMW 5 FN Tris


Figure 4.12 Comparison of materials on textured unmodified silicone elastomer by
ridge width on 5 [im deep grooves (error bars represent s.e.m.)


Ridge Width (4m)

-- 1.5 FN LMW 1.5 FN Unmod 1.5 FN HMW --1.5 FN Tris


Figure 4.13 Comparison of materials on textured unmodified silicone elastomer by
ridge width on 5 mrn deep grooves (error bars represent s.e.m.)









show increased contact guidance on the high modulus material. As stated before, this

effect is not significant in the deeper grooves, and at 5 [im the low modulus tris sample

even shows higher contact guidance, albeit statistically insignificant.


0.4


0.3


S0.2 F


0.1


0.0
5 10 20 Smooth
-0.1
Ridge Width (pm)

--5 FN LMW -- 1.5 FN LMW 5 FN Tris -- 1.5 FN Tris

Figure 4.14 Comparison of ridge width and groove depth to the high modulus material
(LMW) and the low modulus material (Tris) (error bars represent s.e.m.)



Since the 5 [im deep grooves have shown the strongest factors for contact

guidance, the reason there is not a significant difference between materials at the greater

depth might be that the importance of the groove depth overpowers any change due to the

modulus. At the lower depth where the effect of the depth is not as great, it appears that

the modulus of the surface alters its ability to direct the growth of cells. According to the

data and trends, the higher modulus material seems to enhance the contact guidance

phenomenon.









As discussed in Chapter 2, cells attach to surfaces at focal adhesions, and they

impart a stress upon the substrate which is balanced by the material and acts as a

counterweight so to speak improving adhesion. Researchers have examined the forces of

cells such as fibroblasts on elastomeric substrates, using low modulus (- 15 kPa) silicone

films to quantify the forces of adhesion.[871 These cells continually pull and contract

against these surfaces until they reach a balance, the equivalent of pulling the slack out of

a rope until it is taut. Fibroblastic cells on higher modulus materials were shown to

spread better and were of a more constant shape, while cells on lower modulus materials

were more active and elongated.[98]

Possibly the cells elongate along the grooves because they can pull along the

length of the groove as opposed to the width, which is more compliant due to its reduced

thickness. Since they have more resistance along the length, they have a more stable

opposing force to pull and align to. At higher ridge widths, this effect is lessened by the

increased continuous surface area for attachment as well as possibly the increased

thickness of the ridge. In this study, vascular endothelial cells were shown to

preferentially align along the length of a groove and increased that alignment as the

groove depth increased. The fact that changing the modulus had little effect on the

alignment for the deeper grooves is not completely unexpected, since it is well

documented that groove depth plays an important role in contact guidance, while the

contribution due to the mechanical properties is not well defined or studied. However, if

mechanical stability is a main factor in directing the cell growth, then one should see a

greater effect on the deeper features, since the ridges are even less mechanically stable at

the higher aspect ratio. Another issue is that at deeper groove widths, the cells are more









likely to span a groove without touching the bottom of the groove. This in effect lines up

the possible areas of focal adhesions by leaving the only area for adhesion on the top of

the ridge.

At lower groove widths and shallower grooves, the modulus of the material seems

to play a more important role, but one only seen by comparing the highest and lowest

modulus silicone materials available. Substrates like titanium and polystyrene could be

used as well for comparison as a much higher modulus material, but their behavior in

regards to surface energy will play a significant role that cannot be overlooked.

The nuclear form factor is a useful tool for elucidating the alignment of a cell, and

one that may in the future become more widely used. The power of the model stems

from the ease in imaging and data processing as compared to examining strictly actin

filaments or trying to map an entire cell.12' 37] The flaws in its use come from its

sensitivity to less elongated cells and in comparing cells that are aligned at 450 angle to

those with nuclei that are more rounded. Both types of nuclei would give similar results

of zero using the nuclear form factor. While neither are aligned to the features, a group

of cells aligned at 450 would result in an interpretation that the cells were not aligned or

randomly aligned, when in fact they could be aligned all in one direction at 450 due to

some other unknown factor. In reference to sensitivity, a simple model comparison can

show one of the drawbacks of using the nuclear form factor. A highly elongated nucleus

oriented 300 to the direction of the features can have the same dimensions in length and

width as a nucleus that is perfectly aligned to the features, but not as elongated. With a

large sample size, these problems can be minimized, but they should be issues to consider

for potential researchers. By incorporating a corroborating measurement such as the






79


angle of deflection of the nuclei's long axis from the direction of the ridges, or combining

cystoskeletal measurements, more confidence in the method would be assured for future

studies and solve both of these issues.














CHAPTER 5
CONCLUSIONS AND FUTURE WORK


Conclusions

Microtextured Surfaces

To examine the effects of contact guidance on silicone elastomers, microtextured

substrates were produced with reproducible and well-defined surfaces. Ridges of 10,000

[lm length were fabricated at 3 different widths: 5 [m, 10 [m, and 20 [m, separated by 5

[lm wide grooves to determine the effect of separation of features on the alignment of

porcine vascular endothelial cells. Two depths were examined: 5 [im and 1.5 im. The

silicone elastomer samples were produced by casting a film on a textured mold and

allowing the samples to cure. Molds used were either silicon wafers or epoxy replicates

of the wafers. The surfaces were created with micromachining technology, specifically

photolithographic patterning followed by reactive ion etching. After examination of the

surfaces by optical profilometry, it was determined that the silicone copies faithfully

reproduced the textured surface and the textures were of the expected design. Closer

examination demonstrated that the 1.5 [m deep wafer and samples had a more rounded

appearance and had ridge widths -1 [m less than expected. The depths measured

corresponded with designed values of 1.5 [m and 5 [m within experimental error. All

elastomer types used in this study faithfully reproduced the applied texture and gave a

stable substrate for comparison.









Surface Energy and Treatment

Elastomer samples were examined with contact angles to determine their relative

wettability and surface free energy. Formulations of elastomer with both functionalized

and non-functionalized PDMS oligomer additives were examined, and there was no

significant effect to the surface energy as determined by Zisman plots using 5 separate

liquids. Sessile drop contact angles measured with a goniometer were measured, and the

surface energy of the unmodified elastomer samples was found to be 18.9 mN/m. Water

contact angles were typically 1100 for unmodified samples.

Surfaces were treated with fibronectin and radiofrequency glow discharge plasma

in argon for 5 minutes at 50 W. Both treatments significantly increased the

hydrophilicity after treatment, as measured by captive bubble contact angles, from 86.70

4.30 for the unmodified sample to 14.50 3.50 for fibronectin adsorbed surfaces and <

100 for RFGD plasma treated surfaces. Dynamic contact angle analysis was performed

on the unmodified elastomer and materials with non-functional PDMS oligomers added.

Advancing contact angles varied between 100.5 and 115.10 and the hysteresis between

the advancing and receding angles was between 350 450 except for high molecular

weight, high weight percent additive which formed a visible layer of oil on the surface.

The hysteresis on this surface was quite low (11.60 4.40). Dynamic contact angle

analysis of textured surfaces showed a difference in smooth and textured areas, although

quantitative data was not available due to experimental uncertainty. The trend of the

graphs demonstrated an increase in observed advancing water contact angles, which is

possibly due to composite surfaces with air trapped in the textures.

Analysis of the RFGD treated PDMS elastomer surfaces revealed defects that

were caused by the plasma treatment. Groove depths decreased by 40% in some cases as









the material was ablated by the plasma. Cracks on the order of 0.5-1.5 [im were seen to

form after mechanical manipulation of treated surfaces. This phenomenon seems to be a

result of the cracking of a hard silica-like layer on the surface.[116] Due to the

uncertainties with the feature dimensions and surface mechanical properties, fibronectin

adsorbed surfaces were used for the main analysis of the effects of contact guidance.

Contact Guidance on Textured Elastomers

Contact guidance of PVECs on textured silicone elastomers was measured by the

nuclear form factor, in which the log of the ratio of nuclear length to width was

presented. Results demonstrated that as the ridge width decreased from 20 .im to 5 .im

contact guidance increased, as well as when the depth of the grooves increased from 1.5

lm to 5 rm. Data analysis showed that the groove depth was the most important factor

in nuclear alignment. Average values of the nuclear form factor for 5 [im deep, 5 [im

wide samples exceeded 0.3, which implies the length was more than twice the width on

average. Shallower grooves increased the length of the nuclei by approximately 25% in

comparison to the width.

Contact guidance on fibronectin-coated elastomers was examined to determine

the effect of modulus. It was expected that higher modulus materials would increase the

effect of contact guidance. Elastic modulus on 4 elastomers was measured by tensile

tests and resulted in a range of values from 0.3 MPa to 2.34 MPa. There was no

significant difference in the contact guidance on the deep 5 [m grooves with varying

modulus. The 1.5 rm deep grooves showed a significant increase in the alignment of

cells to the groove in the highest modulus material compared to the lowest modulus

material for the 5 rm and 10 rm wide ridges. The conclusion to be taken from this data

is that modulus does seem to play a role in the determination of contact guidance, but









other factors such as groove width and especially depth are more significant. With this

knowledge, and with future work, ideal values to fine tune materials may be possible to

direct and control cell growth.


Future Work

The importance of controlling cell growth and behavior cannot be underestimated

and the phenomenon of contact guidance is only recently approaching maturity. Several

areas for future study and improvement are possible and are listed below.

Surface Treatment

Optimization of plasma treatment needs to be examined, as well as the reasons for

the unwanted effects. Argon RFGD at 50 W for 5 minutes is a standard treatment

in the literature, and the reasons for the deviations should be investigated.

Tether adhesion molecules to silicone surfaces to examine modulus affects with a

permanently bound treatment.

Topographical Design

Smaller ridge widths and an intermediate groove depth should be examined since

the data shows that smaller ridge widths improve contact guidance, but the deeper

groove depths mask the effect of modulus.

Ridge widths on the order of 0.5 rim, 1 rim, and 2.5 [m would give useful data as

to the limits of the ridge effect.

Opposite or "negative" designs of the current elastomers would be useful in

seeing the effect of a constant 5 mrn ridge separated by varying smooth areas.






84


Cell Studies

Examination of tissue growth from a central area to uncovered textured substrates

should be relatively easy to set up and characterize.

Growth patterns of endothelial cells along a textured interior surface of a cylinder

would more closely model a seeded vascular graft response.
















LIST OF REFERENCES


1. Curtis, A.S.G. and P. Clark. The effects of topographic and mechanical properties
of materials on cell behavior. Critical Reviews in Biocompatibility, 1990. 5(4): p.
343-362.

2. Curtis, A.S.G. and C.D.W. Wilkinson. Review: Topographical control of cells.
Biomaterials, 1997. 18: p. 1573-1583.

3. Bhatia, S.N. and C.S. Chen. Tissue Engineering at the Micro-Scale. Biomedical
Microdevices, 1999. 2(2): p. 131-144.

4. Brady, R.F. and I.L. Singer. Mechanical Factors Favoring Release from Fouling
Release Coatings. Biofouling, 2000. 15(1-3): p. 73-81.

5. Singer, I.L., J.G. Kohl, and M. Patterson. Mechanical Aspects of Silicone
Coatings for Hard Foulant Control. Biofouling, 2000. 16(2-4): p. 301-309.

6. Weiss, P. In vitro experiments on the factors determining the course of the
outgrowing nerve fiber. The Journal ofExperimental Zoology, 1934. 69: p. 393.

7. Weiss, P. Experiments on cell and axon orientation in vitro: the role of colloidal
exudates in tissue organization. The Journal ofExperimental Zoology, 1945. 100:
p. 353.

8. Harrison, R.G. On the stereotropism of embryonic cells. Science, 1911. 34: p.
279.

9. Rovensky, Y.A., I.L. Slavnaja, and J.M. Vasiliev. Behaviour of fibroblast-like
cells on grooved surfaces. Experimental Cell Research, 1971. 65: p. 193-201.

10. Dunn, G.A. and J.P. Heath. A new hypothesis of contact guidance in tissue cells.
Experimental CellResearch, 1976. 101: p. 1-14.

11. den Braber, E.T., J.E. de Ruijter, H.T.J. Smits, L.A. Ginsel, A.F. von Recum, and
J.A. Jansen. Effect of parallel surface microgrooves and surface energy on cell
growth. Journal ofBiomedical Materials Research, 1995. 29(4): p. 511-518.









12. den Braber, E.T., J.E. de Ruijter, H.T.J. Smits, L.A. Ginsel, A.F. von Recum, and
J.A. Jansen. Quantitative analysis of fibroblast morphology on microgrooved
surfaces with various groove and ridge dimensions. Biomaterials, 1996. 17(21): p.
2037-2044.

13. Mudera, V.C., R. Pleass, M. Eastwood, R. Tarnuzzer, G. Schultz, P. Khaw, D.A.
McGrouther, and R.A. Brown. Molecular Responses of Human Dermal
Fibroblasts to Dual Cues: Contact Guidance and Mechanical Load. CellMotility
and the Cytoskeleton, 2000. 45: p. 1-9.

14. van Kooten, T.G. and A.F. von Recum. Cell Adhesion to Textured Silicone
Surfaces: The Influence of Time of Adhesion and Texture on Focal Contact and
Fibronectin Fibril Formation. Tissue Engineering, 1999. 5(3): p. 223-240.

15. Walboomers, X.F., H.J.E. Croes, L.A. Ginsel, and J.A. Jansen. Contact guidance
of rat fibroblasts on various implant materials. Journal ofBiomedical Materials
Research, 1999. 47: p. 204-212.

16. Clark, P., P. Connolly, A.S.G. Curtis, J.A.T. Dow, and C.D.W. Wilkinson.
Topographical control of cell behaviour: I. Simple step cues. Development, 1987.
99: p. 439-448.

17. Clark, P., P. Connolly, A.S.G. Curtis, J.A.T. Dow, and C.D.W. Wilkinson.
Topographical control of cell behaviour: II. multiple grooved substrata.
Development, 1990. 108: p. 635-644.

18. Matsuzaka, K., X.F. Walboomers, J.E. de Ruijter, and J.A. Jansen. The effect of
poly-L-lactic acid with parallel surface micro groove on osteoblast-like cells in
vitro. Biomaterials, 1999. 20: p. 1293-1301.

19. Schmidt, J.A. and A.F. von Recum. Macrophage response to microtextured
silicone. Biomaterials, 1992. 13(15): p. 1059-1069.

20. Wojciak-Stothard, B., A.S.G. Curtis, W. Monaghan, K. Macdonald, and C.D.
Wilkinson. Guidance and activation of murine macrophages by nanometric scale
topography. Experimental Cell Research, 1996. 223(2): p. 426-435.

21. Wilkinson, C.D.W., A.S.G. Curtis, and J. Crossan. Nanofabrication in cellular
engineering. Journal of Vacuum Science & Technology B, 1998. 16(6): p. 3132-
3136.

22. Poole-Warren, L.A., K. Schindhelm, A.R. Graham, P.R. Slowiaczek, and K.R.
Noble. Performance of small diameter synthetic vascular prostheses with
confluent autologous endothelial cell linings. Journal ofBiomedical Materials
Research, 1996. 30(2): p. 221-229.









23. Dalton, B.A., G.A. McFarland, and J.G. Steele. Stimulation of epithelial tissue
migration by certain porous topographies is independent of fluid flux. Journal of
Biomedical Materials Research, 2001. 56: p. 83-92.

24. Steele, J.G., G. Johnson, K.M. McLean, G.J. Beumer, and H.J. Griesser. Effect of
porosity and surface hydrophilicity on migration of epithelial tissue over synthetic
polymer. Journal ofBiomedical Materials Research, 2000. 50: p. 475-482.

25. Kirkpatrick, C.J., M. Otto, T.V. Kooten, V. Krump, J. Kriegsmann, and F.
Bittinger. Endothelial cell cultures as a tool in biomaterial research. Journal of
Materials Science-Materials in Medicine, 1999. 10(10-11): p. 589-594.

26. Sato, M., K. Nagayama, N. Kataoka, M. Sasaki, and K. Hane. Local mechanical
properties measured by atomic force microscopy for cultured bovine endothelial
cells exposed to shear stress. Journal ofBiomechanics, 2000. 33: p. 127-135.

27. Tsukurov, O.I., C.J. Kwolek, G.J. L'Italien, A. Benbrahim, B.B. Milinazzo, N.E.
Conroy, J.P. Gertler, R.W. Orkin, and W.M. Abbott. The Response of Adult
Human Saphenous Vein Endothelial Cells to Combined Pressurized Pulsatile
Flow and Cyclic Strain, In Vitro. Annals of Vascular Surgery, 2000. 14: p. 260-
267.

28. Vaidya, R., L.M. Tender, G. Bradley, M.J. O'Brien II, M. Cone, and G.P. Lopez.
Computer-Controlled Laser Ablation: A Convenient and Versatile Tool for
Micropatterning Biofunctional Synthetic Surfaces for Applications in Biosensing
and Tissue Engineering. Biotechnology Progress, 1998. 14: p. 371-377.

29. Kappelt, M. and D. Bimberg. Wet chemical etching of high quality V-grooves
with { 111 } a sidewalls on (001) InP. Journal of the Electrochemical Society,
1996. 143(10): p. 3271-3273.

30. Thomas, C.H., C.D. McFarland, M.L. Jenkins, A. Rezania, J.G. Steele, and K.E.
Healy. The role of vitronectin in the attachment and spatial distribution of bone-
derived cells on materials with patterned surface chemistry. Journal of
Biomedical Materials Research, 1997. 37(1): p. 81-93.

31. Tien, J., A. Terfort, and G.M. Whitesides. Microfabrication through Electrostatic
Self-Assembly. Langmuir, 1997. 13: p. 5349-5355.

32. Alaerts, J.A., V.M. De Cupere, S. Moser, P.V.B. de Aguilar, and P.G. Rouxhet.
Surface characterization of poly(methyl methacrylate) microgrooved for contact
guidance of mammalian cells. Biomaterials, 2001. 22(12): p. 1635-1642.

33. Schmidt, J.A. and A.F. von Recum. Texturing of polymer surfaces at the cellular
level. Biomaterials, 1991. 12(4): p. 385-389.









34. von Recum, A.F. and T.G. van Kooten. The influence of micro-topography on
cellular response and the implications for silicone implants. Journal of
Biomaterials Science Polymer Edition, 1995. 7(2): p. 181-198.

35. Walboomers, X.F., W. Monaghan, A.S.G. Curtis, and J.A. Jansen. Attachment of
fibroblasts on smooth and microgrooved polystyrene. Journal ofBiomedical
Materials Research, 1999. 46: p. 212-220.

36. Matsuzaka, K., X.F. Walboomers, A. de Ruijter, and J.A. Jansen. Effect of
microgrooved poly-l-lactic (PLA) surfaces on proliferation, cytoskeletal
organization, and mineralized matrix formation of rat bone marrow cells. Clinical
Oral Implants Research, 2000. 11: p. 325-333.

37. den Braber, E.T., J.E. de Ruijter, L.A. Ginsel, A.F. von Recum, and J.A. Jansen.
Orientation of ECM protein deposition, fibroblast cytoskeleton, and attachment
complex components on silicone microgrooved surfaces. Journal ofBiomedical
Materials Research, 1998. 40: p. 291-300.

38. den Braber, E.T., J.E. de Ruijter, H.T.J. Smits, L.A. Ginsel, A.F. von Recum, and
J.A. Jansen. Quantitative analysis of cell proliferation and orientation on substrata
with uniform parallel surface micro-grooves. Biomaterials, 1996. 17(11): p. 1093-
1099.

39. Walboomers, X.F., J.J.E. Croes, L.A. Ginsel, and J.A. Jansen. Growth behavior of
fibroblasts on microgrooved polystyrene. Biomaterials, 1998. 19: p. 1861-1868.

40. den Braber, E.T., J.E. de Ruijter, and J.A. Jansen. The effect of a subcutaneous
silicone rubber implant with shallow surface microgrooves on the surrounding
tissues in rabbits. Journal ofBiomedical Materials Research, 1997. 37: p. 539-
547.

41. Walboomers, X.F. and J.A. Jansen. Microgrooved silicone subcutaneous implants
in guinea pigs. Biomaterials, 2000. 21: p. 629-636.

42. Walboomers, X.F., H.J.E. Croes, L.A. Ginsel, and J.A. Jansen. Microgrooved
subcutaneous implants in the goat. Journal of Biomedical Materials Research,
1998. 42: p. 634-641.

43. Dalton, B.A., X.F. Walboomers, M. Dziegielewski, M.D.M. Evans, S. Taylor,
J.A. Jansen, and J.G. Steele. Modulation of epithelial tissue and cell migration by
microgrooves. Journal ofBiomedical Materials Research, 2001. 56(2): p. 195-
207.

44. Rich, A. and A.K. Harris. Anomalous preferences of cultured macrophages for
hydrophobic and roughened substrata. Journal of Cell Science, 1981. 50(Aug): p.
1-7.