1 PHYSICO CHEMICAL PROPERTIES OF HYDROPHILIC AND AMPHIPHILIC CROSSLINKED SYSTEMS THAT INFLUENCE BIOLOGICAL RESPONSES By ANGEL EJIASI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012
2 2012 ANGEL EJIASI
3 To my family and friends
4 ACKNOWLEDGEMENTS First and foremost, I would like to thank my family and friends. I would also like to thank my adviser, Dr. Anthony Brennan for his support and academic guidance; as well as my former co workers Adwoa Baah Dwomah, Joe Decker, Laura Villada, Dr. Canan Kizilkia Jiun Jeng Chen, Scott Cooper, Dave Jackson, Julian Sheets and Chelsea Magin I would like to thank my committee members for their input on my dissertation research: Dr. Christopher Batich, Dr. Josephine Allen, Dr. Thomas Angelini, and Dr. Jennifer Andrew. I am also grateful for the financial support provided by the UF Alumni Associat ion, Office of Naval Research ( Grant N00014 10 1 0579 ) and International Cent e r for Materials Research at University of California, Santa Barbara
5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ ............................... 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 12 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 17 Physicochemical Factors that Influence Biological Responses ............................... 17 Scope of Research ................................ ................................ ................................ 18 Specific Aims ................................ ................................ ................................ .......... 19 Specific Aim 1 ................................ ................................ ................................ ... 19 Specific Aim 2 ................................ ................................ ................................ .. 20 2 BACKGROUND ................................ ................................ ................................ ...... 21 Significance of Biofouling in the Marine Environment ................................ ............. 21 Signifi cance of Vascular Grafts for Blood Vessel Replacement .............................. 23 Development and Composition of Blood Vessels ................................ ............. 26 Phenotype Modulation of Smooth Muscle Cells ................................ ............... 27 Influence of Physical and Chemical Factors on Mammalian Cell Behavior ............. 28 Physicochemical Factors that Influenc e Marine Organism Adhesion ..................... 32 Physical and Chemical Factors that Influence Bioadhesion ................................ .... 35 Influence of Surface Energy on Fouling ................................ ........................... 35 Influence of Mechanical Properties on Fouling ................................ ................. 37 Crossli nked Network Systems ................................ ................................ ................ 38 3 PHYSICAL AND CHEMICAL PROPERTIES OF POLY(2 HYDROXYETHYL METHACRYLATE) HYDROGELS ................................ ................................ .......... 42 Introduction ................................ ................................ ................................ ............. 42 Experimental Section ................................ ................................ .............................. 46 Materials ................................ ................................ ................................ ........... 46 HEMA Hydrogel Preparation ................................ ................................ ............ 46 Volume Fraction of Polymer ................................ ................................ ............. 47 Compression Testing ................................ ................................ ....................... 48 Contact Angle Measurements ................................ ................................ .......... 48 Fourier Transform Infrared Spectroscopy Attenuated Total Reflectance (FTIR ATR) ................................ ................................ ................................ ... 49
6 Amido Black Assay ................................ ................................ ........................... 49 Statistical Analysis ................................ ................................ ............................ 50 Results ................................ ................................ ................................ .................... 50 Conclusion ................................ ................................ ................................ .............. 57 4 SURFACE PROPERTIES OF AMPHIPHILIC CROSSLINKED NETWORKS WITH TUNABLE MECHANICAL PROPERTIES ................................ ..................... 59 Introduction ................................ ................................ ................................ ............. 59 Experimental Section ................................ ................................ .............................. 62 Materials ................................ ................................ ................................ ........... 62 Sample Preparation ................................ ................................ .......................... 63 E quilibrium Water Content ................................ ................................ ................ 64 Volume Fraction of Polymer ................................ ................................ ............. 64 Compression Testing ................................ ................................ ....................... 65 Tensile Testing ................................ ................................ ................................ 65 Contact Angle Measurements ................................ ................................ .......... 65 Fourier Transform Infrared Attenuated Total Reflectance (FTIR ATR) ............ 67 Atomic Force Microscopy ................................ ................................ ................. 67 X ray Photoelectron Spectroscopy ................................ ................................ ... 67 Amido Black Assay ................................ ................................ ........................... 68 Statistical Analysis ................................ ................................ ............................ 69 Results ................................ ................................ ................................ .................... 69 Conclusion ................................ ................................ ................................ .............. 84 5 ANTIFOULING PROPERTIES OF AMPHIPHILIC CROSSLINKED NETWORKS .. 86 Introduction ................................ ................................ ................................ ............. 86 Experimental Section ................................ ................................ .............................. 90 Materials ................................ ................................ ................................ ........... 90 Preparation of Amphiphilic Hydrogels ................................ .............................. 90 Synthesis of PDMS samples ................................ ................................ ............ 91 Ulva Spore Assay ................................ ................................ ............................. 92 Ulva Sporeling Assay ................................ ................................ ....................... 92 Statistical Analysis ................................ ................................ ............................ 93 Results ................................ ................................ ................................ .................... 93 Conclusions ................................ ................................ ................................ .......... 101 6 PEPTIDE CONJUGATED AMPHIPHILIC CROSSLINKED NETWORKS WITH TUNABLE MECHANICAL PROPERTIES ................................ ............................. 103 Introduction ................................ ................................ ................................ ........... 103 Experimental Section ................................ ................................ ............................ 106 Materials ................................ ................................ ................................ ......... 106 Synthesis of Amphiphilic Crosslinked Networks ................................ ............. 107 RGDS Fu nctionalization of Crosslinked Networks ................................ .......... 108
7 Fourier Transform Infrared Spectroscopy Attenuated Total Reflectance (FTIR ATR) ................................ ................................ ................................ 109 Fluorescence Microscopy ................................ ................................ ............... 109 Equilibrium Water Content ................................ ................................ .............. 110 Compression Testing ................................ ................................ ..................... 110 Cell Culture ................................ ................................ ................................ ..... 110 Cell Proliferation ................................ ................................ ............................. 110 Statistical Analysis ................................ ................................ .......................... 111 Results ................................ ................................ ................................ .................. 111 Conclusions ................................ ................................ ................................ .......... 122 7 CONCLUSIONS AND FUTURE WORK ................................ ............................... 124 Conclusions ................................ ................................ ................................ .......... 124 Chemical and Mechanical Properties of Poly(HEMA co DEGDMA) Hydrogels ................................ ................................ ................................ .... 124 Chemical and Mechanical Properties of Amphiphilic Hydrogels ..................... 125 Fouling Properties of Amphiphilic Hydrogels ................................ .................. 125 RGDS Functionalized Amphiphilic Hydrogels ................................ ............... 127 Future Work ................................ ................................ ................................ .......... 128 RGDS Functionalized Amphiphilic Hydrogels ................................ ............... 128 AF and FR P roperties of Amphiphilic Hydrogels ................................ ............ 129 APPENDIX: FABRICATION OF TOPOGRAHPICALLY MODIFIED PHEMA BASED HYDROGELS ................................ ................................ .......................... 130 LIST OF REFERENCES ................................ ................................ ............................. 133 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 144
8 LIST OF TABLES Table page 3 1 Composition of poly(HEMA) based hydrogels ................................ ................... 47 3 2 Swelling and mechanical properties of PHEMA h ydrogels ................................ 51 3 3 Surface properties of poly(HEMA co DEGDMA) hydrogels. ............................... 55 4 1 Chemical composition of amphiphilic crosslinked networks with varying crosslink ratio. ................................ ................................ ................................ ..... 64 4 2 Mechanical properties of amphiphilic networks from tensile tests. ..................... 72 4 3 Surface properties of amphiphilic gels with varying MPTSDS/(TRIS+HEMA) molar ratio using the sessile contact angle technique ................................ ........ 74 4 4 Surface properties of amphiphilic crosslinked networks using the dynamic contact angle technique. ................................ ................................ ..................... 76 4 5 Captive bubble measurements for amphiphilic crosslinked networks. ................ 77 4 6 Theoretical elemental composition of amphiphilic networks. .............................. 79 4 7 Elemental compositi on of amphiphilic crosslinked networks using XPS. ............ 79 4 8 Roughness R q of amphiphilic crosslinked networks in a dehydrated state. ........ 81 4 9 Network parameters of amphiphilic crosslinked networks. ................................ 84 5 1 Feed composition of amphiphilic crosslinked networks. ................................ ..... 91 5 2 Contact angle measurements of amp hiphilic crosslinked networks equilibrated in sea water for 48h ................................ ................................ ....... 101 6 1 Chemical composition of amphiphilic crosslinked net works ............................. 108
9 LIST OF FIGURES Figure page 2 1 Relationship between relative bioadhesion and critical surface tension of a material. Adapted from Baier et al. 64 ................................ ................................ 36 3 1 Standa rd curve for amido black protein adsorption assay. ................................ 50 3 2 Compression modulus and M c of PHEMA hydrogels with varying DEGDMA/HEMA molar ratio (mmol/mmol) ................................ ......................... 52 3 3 Compression stress deflection curve of PHEMA hydrogels with varying DEGDMA/HEMA molar ratio (mmol/mmol). ................................ ........................ 53 3 4 FTIR ATR spectra of PHEMA hydrogels of varying DEGDMA/HEMA molar ratio (mmol/mmol). ................................ ................................ .............................. 53 3 5 Protein adsorption of BSA onto PHEMA of varying D EGDMA/HEMA molar ratio (mmol/mmol), TCP, and NC ................................ ................................ ....... 56 3 6 The relationship between protein adsorption and either modulus (blue diamond) or surface energy (red square) of amphiphilic crosslinked networks. 57 4 1 Monomers included in the polymerization of amphiphilic crossli nked networks. ................................ ................................ ................................ ............ 63 4 2 Standard curve for amido black protein adsorption assay. ................................ 69 4 3 EWC(%) of amphiphilic hyd rogels with varying MPTSDS/(TRIS+HEMA) molar ratio. Asterisk (*) denotes statistically significant EWC (%) from gel with molar ratio of 50 (mmol/mmol). Double asterisk (**) denotes statistically significant EWC (%) from gel with molar ratio of 25 (mmol/mmo l) (p < 0.05). .... 70 4 4 Compression modulus (red square) and M c of amphiphilic hydrogels of varying MPTSDS/(TRIS+HEMA) molar ratio. All moduli were statistically significant ( p < 0.05) between all gels. ................................ ................................ 71 4 5 Stress deflection curve from compression testing of amphiphilic crosslinked networks with varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol). ......... 71 4 6 Stress strain curve from tensile testing of amphiphilic crosslinked networks with varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol). ........................ 73 4 7 Sessile water drop and captive air droplet on crosslinked networks with an MPTSDS/(TRIS+HEMA) molar ratio of 2.5 mmol/mmol. ................................ .... 74
10 4 8 FTIR ATR spectra of amphiph ilic crosslinked networks with varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol). ................................ ............ 78 4 9 Morphology of crosslinked networks with an MPTSDS/(TRIS+HEMA) molar ratio ................................ ................................ ................................ .................... 80 4 10 Protein adsorption on NC, TCP, and amphiphilic hyd rogels of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) ................................ ............ 81 4 1 1 The relationship betwe en protein adsorption and M c of amphiphilic hydrogels. 83 5 1 Settlement density of spores on amphiphilic crosslinked networks with varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) and the PDMSe control ................................ ................................ ................................ ................. 94 5 2 Biomass of sporeling biomass after 7 days of growth on amphiphilic crosslink ed networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) and PDMSe ................................ ................................ .................. 95 5 3 Images of sporeling biomass on amphiphilic crosslinked networks with varying MPTSDS/(TRIS+HEMA) molar ratio and PDMSe after 7 days of growth ................................ ................................ ................................ ................ 96 5 4 Percent removal of sporeling biomass on amphiphilic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) and PDMSe ........... 98 5 5 Relationship between percent removal of sporeling biomass and M c (top).Correlation between percent removal of sporelin g biomass and substrate modulus (bottom). ................................ ................................ ............... 99 5 6 Relationship between percent removal of sporeling biomasss and M c (top) of crosslinked networks. Relationship between sporeling bioma ss and substrate modulus (bottom) of crosslinked networks. ................................ ...................... 100 6 1 Synthesis of amphiphilic crosslinked networks ................................ ................. 108 6 2 Synthesis of RGDS conjugated amphiphilic crosslinked networks ................... 109 6 3 FTIR ATR spectra of amphiphilic crosslinked networks with an MPTSDS/(TRIS+HEMA) molar ratio of 2.5, 10 and 75 mmol/mmol. ................ 112 6 4 FTIR ATR Spectra of RGDS conjugated amphiphilic crosslinked networks with an MPTSDS/(TRIS+HEMA) molar ratio of (top) 2.5 and (bottom) 75 mmol/mmol. ................................ ................................ ................................ ...... 113 6 5 Fluorescence of FITC RGDS conjugated amphiphilic crosslinked networks. Fluorescence signal normalized to gel with highest fluorescence signal amongst all hydrogels (top) and fluores cence ................................ .................. 114
11 6 6 Swelling behavior of RGDS conjugated amphiphilic crosslinked networks of vary ing MPTSDS/(TRIS+HEMA) molar ratio ................................ ................... 116 6 7 Compression modulus as a function of initial concentration of RGDS o n amphiphilic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio ................................ ................................ ................................ .................. 116 6 8 SMC proliferation on TCPS and RGDS conjugated amphiphilic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) ........... 118 6 9 Proliferation of SMCs on conjugated amphiphilic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) .............................. 120 6 10 SMC attachment on smooth or topographically modified (Sharklet AF TM +1SK2x2, wafer no. 5816) amphiphilic hydrogels with MPTSDS/ (TRIS+HEMA) molar ratio of 2.5, 10, and 75 mmol/mmol after 24h ................. 122 A 1 Scheme of mold used for hydrogel polymerization by UV or heat. ................... 131 A 2 Image of poly(HEMA co DEGDMA) hydrogel with the +2.8 SK 2x2 pattern using the SEM at 2000x magnification(left) and Zeiss optical microscope (right) ................................ ................................ ................................ ................ 132 A 3 Image of the amphiphilic gels with the +2.8 SK 2x2 pattern using the SEM at 2000x magnification(left) and Zeiss optical microscope (right). ........................ 132
12 LIST OF ABBREVIATION S AF Antifouling AFM Atomic Force Microscopy APS Ammonium persulfate BSA Bovine Serum Albumin DEGDMA Di(ethylene glycol) dimethacrylate E Modulus ECM Extracellular matrix EWC Equilibrium water content FITC Fluorescein isothiocyanate FP Fluoropolymer FR Fouling Release FTIR ATR Fourier Transform Infrared Attenuated Transform Reflectance HEMA 2 hydroxyethyl methacrylate MPTSDS 1,3 bis(3 methacryloxypropy(trimethylsiloxy)disiloxane PDMS E Poly(dim ethylsiloxane) elastomer PEG Polyethylene glycol RGDS Arg Gly Asp Ser RT PCR Reverse transcriptase polymerase chain reaction SMBS Sodium metabisulfite SMC Smooth Muscle Cell TEVG Tissue Engineered Vascular Graft TRIS Tris(trimethylsiloxy) 3 methacryloxypr opylsilane UV Ultraviolet XPS X ray Photoelectron Spectroscopy
13 C Critical surface energy
14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree o f Doctor of Philosophy PHYSICO CHEMICAL PROPERTIES OF HYDROPHILIC AND AMPHIPHILIC CROSSLINKED SYSTEMS THAT INFLUENCE BIOLOGICAL RESPONSES By Angel Ejiasi August 2012 Chair: Anthony Brennan Major: Materials Science and Engineering The effect of physica l, chemical, and biological cues on the behavior of smooth muscle cells (S MCs) and attachment of marine organisms was investigated. Both h ydrop hilic and amphiphilic crosslinked polymer networks with varying chemical and mechanical properties were used to d irect biologic al responses. Poly(2 hydroxy ethyl methacrylate) (PHEMA) hydrogels were fabricated with tunable mechanical properties by varying the di functional monomer concentration in the feed composition. Amphiphilic hydrogels composed of 2 hydroxy ethyl methacrylate (HEMA), 1,3 bis(3 methacryloxy propyl)tetrakis(trimethylsiloxy)disiloxane (MPTSDS), and tris(trimethylsiloxy) 3 methacryloxypropylsilane (TRIS) were copolymerized using ultraviolet (UV) light and a photo initiator. Hydrogels prepared with vary ing concentration of di functional monomer, MPTSDS, exhibited an order of mag nitude difference in elastic moduli. Not only were the bulk material properti es influenced by the crosslinking agent concentration in the feed composition, bu t the surface propert ies ( i.e., contact angle and hysteresis) were influenced as well. Modulus (E) has been reported to be positively correlated wit h the settlement of marine organisms However, this was not the case for the amphiphilic gels tested
15 against biomolecules and ma rine organisms. Stiffer gels inhibited fouling of proteins and marine organism, Ulva linza to a greater extent than the softer gels. Furthermore, the network structure, in regards to the molecular weight between crosslinks M c was found to have a greater influence on fouling A strong correlation was observed between protein adsorption and M c of the a mphiphilic crosslinked networks compared to just the modulus and surface energy ( ) alone. A higher correlation was also obtained between M c and Ulva sporeling biomass than between sporeling biomass and elastic modulus E exhibiting R 2 value of 0.98 and 0.38 respectively The percent removal of sporeling biomas s growth wa s shown to be posi tively correlated with the ( E ) 1/2 which is a contrast to what has previously been reported. Again, there was a higher correlation between M c and percent removal of sporeling biomas s than between ( E ) 1/2 and percent removal of sporelings (R 2 value of 0.83 and 0.57, respectively). The differences in biofouling ability is most likely due to differences in mesh size between hydrogel compositions. Biomolecule accumulation and absorption was made easier by the larger mesh size in hydrogels with lower cro sslinking concentration in the feed composition. The influence of chemical and physical properties on mammalian cells was also investigated. Amphiph i lic crosslinked networks were fabricated with tunable mechanical properties and their ability to modulat e smooth muscle cell ( SMC ) phenotype was studied by assessing cell proliferation Bioactive molecules, Arg Gly Asp Ser (RGDS), were incorporated into the crosslinked matrix to promote adhesion and facilitate cell growth. The elastic modulus of the substrat e a nd the concentration of RGDS were shown to positively correlate with the attachment and proliferation of SMCs ; indicating
16 that the physic chemical network properties play a large role in behavior of unicellular organisms.
17 CHAPTER 1 INTRODUCTION Physicochemical Factors that Influence Biological Responses When developing anti microbial coatings and biomaterials, the interaction between the surface of a material and biomolecules or cells must be considered. Studies have shown that biological interac tion influences the performance of man made materials in various environments. Although the mechanisms for adhesion and criteria for settlement may differ between species, the physicochemical factors that control attachment must be energeti cally favorable for survival; so understanding what conditions promote adsorption of biomolecules and attachment and growth of cells is necessary. For this rea son, hydrogels were fabricated to investigat e the effect of physico chemical properties on the biological respon ses of two model systems: Ulva zoo spores and smooth muscle cells (SMCs). Smooth muscle cells are found in the native blood vessel and tend to play a large role in cardiovascular diseases. These cells are often used in tissue engineered constructs for repla cement of damaged blood vessels; therefore, controlling the behavior of these cells is done to create physiologically functioning tissue engineered constructs 1 Ulva linza was examined because it is the most common macroalga fouler on marine vessels and structures in the ocean 2 Amphiphilic networks were made with tunable mechanical properties to assess the effe ct of network properties on fouling release properties towards Ulva zoospores. The same networks were functionalized with the peptide, Arg Asp Ser Gly (RGDS), and the effect of elasti c modulus and chemistry on biological responses of smooth muscle cells was investigated
18 Scope of Research In the marine environment, f ouling of biomolecules and unicellular organisms on the hulls of vessels is problematic for the Navy. The increased roug hness of the biofilms and presence of macrofoulers leads to increased frictional drag and affect s the hydrodynamics of ships.; acco unting for an estimated total cost of 56 million dollars per year for all DDG 51 class ships due to heavy slime hull fouling and increases to 119 million dollars for small calcareous fouling 3 Surfaces that are environmentally benign and foul resistant are being investigated to curtail these issues In the following work, materials were developed w ith tunable mechanical properties to investigat e the effect of network properties on the adhesion o f marine organisms and protein s These same materials were used to investigate the biological responses of mammalian cells. Controlling the behavior of mamma lian cells is necessary to obtain physiologically functioning tissue. Constructs made out of synthetic and biological materials have b een used to grow cells in easily tailored scaffold s These types of constructs have been used previously as vascular graft s, which are used to replace damaged blood vessels that impair blood circulation due to atherosclero sis. The biological responses of SMCs to chemical and mechan ical cues will be investigated, due to their importance in controlling vas o constriction and ECM production in the native blood vessel.
19 Specific Aims Specific Aim 1 : Develop an amphiphilic hydrogel with tunable mechanical properties that inhibits settlement of Ulva zoospores and sporelings greater than the poly(dimethylsiloxane) elastomer (PDMSe) con trol. Determine the relationship between substrate modulus and fouling of spores and sporelings Hydr ogels have been shown to inhibit fouling of bacteria, and marine organisms 4 5 6 However, these materials tend to exhibit poor mechanical integrity 5 7 An approach used to endow hydrogels with improved e lastic modulus was done by incorporating hydrophobic di functional and mono functional monomers into the crosslinked networks 8 This type of network may also exhibit unique antifouling and fouling release properties due to the presence of hydrophobic and hydrophilic domains at the surface These types of am phip hilic coatings are postulated to create an uninhabitable surface for settlement of many types of marine organisms that adhere strongly to either hydrophilic or hydrophobic surfaces, i.e.: Ulva spores and diatoms, respec tively 9 Novel amphiphilic hydrogels were made with tunable mechanical properties with the intent of studying the material properties that influence bioadhesion. The elastic moduli of a series of these crossil nked systems were characterize d. The surface energy of these materials was determined by contact angle measurements Both parameters were influenced by the concentration of di functional monomer in the feed composition. The ability of these hydrogels to deter settlement of Ulva spores and attachment of proteins was assessed It was hypothesized that a better correlation would be found between percent removal and ( E ) 1/2 than between percent removal of Ulva spores and modulus E or surface energy alone. The amphiphilic hydr ogels should exhibit a statistically significant lower Ulva spore settlement. A greater Ulva spore percent
20 removal will be observed on the hydrogel with the lowest crosslink molar ratio compared to the smooth PDMSe control. Specific Aim 2 : Dete rmine w heth er the modulus and concentration of peptide correlate with higher levels of attachment and proliferation of smooth muscle cells Peptide type and concentration have been shown to influence SMC attachment, proliferation and migration. In order to promote t he mature phenotype of SMCs, certain chemical and mechanical cues must be sent to the cells via integrins. Chemical composition, in the form of peptide ligands can induce certain responses from cells. For example, a negative correlation between peptide (RG DS) concentration and number of proliferating cells has been reported and s tiff su bstrates appear to induce proliferation in SMCs 10 11 Expression of contractile markers (calponin and caldesmon) appears to decrease with higher modulus 12 There appears to be a critical ligand density and substrate modulus that promotes heightened expression of the contractile phenotype markers and proliferation that is indicative of the contractile phenotype These chemical and mechanical cues were studied for their influence on smooth muscle cells behavior with amphiphilic hydrogels. Amphiphilic hydrogels with tunable mechanical properties were fabrica ted and functionalized with varying RGDS concentration : 0.05 mM, 0.5 mM, 5 mM Modulation of SMC ph enotype was studied on amphiphilic gels with differing elastic modulus and peptide concentration by examin ing SMC proliferation It was hypothesized that the substrate with the highest concentrat ion of peptide RGDS (5 mM ) will exhibit the highest prolifera tion rate in cells. In addition, the hydrogels exhibiti ng the highest modulus will also result in the highest proliferation rate in cells compared to the substrate with the lowest modulus.
21 CHAPTER 2 BACKGROUND Significance of Biofouling in the Marine Env ironment The importance of studying biofouling of biomolecules and unicellular organisms stems from the fact that it occurs quite frequently in the marine environment. This issue becomes especially problematic for marine coatings that are used for Navy ves sels The enhanced roughness from the biofilms and presence of macrofoulers leads to increased frictional drag and affects the hydrodynamics of ships. For example, the change in total resistance between a hydraulically smooth Arleigh Burke class destroyer (DDG 51, a ship that accounts for 30% of all US Navy ships) moving at a speed of 7.7 ms 1 and an Arleigh Burke class destroyer with small, medium, or heavy calcareous fouling is 29%, 44%, and 69%, respectively. The fuel costs associated with the hull con dition of DDG 51 class ships increases to 1.15 million dollars per ship per year when the hull is coated with heavy slime as opposed to hydraulically smooth ships, due to an increase in fuel consumption by 10.3%. The estimated total cost for all DDG 51 cla ss ships due to heavy slime hull fouling is 56 million dollars per year and increases to 119 million dollars for small calcareous fouling or weed 3 Factors that contribute to fouling of biomolecules and marine organisms onto man made materials are being investigated to curtail these issues. Settl ement of marine organisms (i.e., spores of macroalgae and larvae of barnacles) onto surfaces has been reported to be preceded by a conditioning film composed of biomolecules and bacte ria 13 In some instances, successional colonization has not been observed; thus, the elimination of preliminary colonization of
22 biofilm will not always lead to the inh ibition of macrofoulers 2 This behavior has been observed with the crustacean, Balanus amphitrite ( B. amphitrite ). B. amphitrite has been reported to exhibit low settlement on surfaces in the presence of biofilms 13 Another approach to controlling biofouling of marine organisms involves disrupting their life cycle. In the case of algae, the zygote that forms from the fusion of the sperm and egg from released spores must develop and compete for settlement and recruit other sporelings or larvae to the corresponding area. It is the last stage that is pursued to control bio fouling 13 Currently, coatings are being developed with antifouling (AF) and fou ling release (FR) p roperties. Antifouling coati ngs are those that inhibit initial settlement of marine organism s. Fouling release coatings exhibit minimal adhesion between the ma terial and fouling organisms, which results in easy release when subjected to hydrodynamic forces 2 F or example, 0.2 to 1 MPa is needed to remove barnacles fro m the PDMS coatings 14 The coatings that are currently used to deter settlement of ma rine organisms by emission of chemicals into the marine environment are known as AF biocides. The AF biocides that are c urrent ly used include copper, copper/zinc pyrithione, triphenylborane pyridine (TPBP), chlorothlonil, cichlofluanid, cuprous thiocyanate, diuron, SeaNine 211 (DCOIT), Iragrol 1051, naphthenic acid copper salts, thiram, etc. Many of these coatings have been reported to be deleterious to the marine environment. Copper based biocides are considered to be toxic to mussel, oyster, sea urchin, and cyanobacteria. Organic biocides, such as Irgarol 1051 and diuron, inhibit photosynthesis of marine
23 organisms. Irgarol 1051 is also toxic to diatom Navicula pelliculosa macrophytes Chara vulgaris Apium nodiflorum, and some marine animals 15 Natural fouling agents have been shown to deter marine organism settlement as by using mechanical, chemical, and physical strategies 16 For example, t he analogues of natural organic compounds (3 acetyl 2,5 dimethyl furan, ethyl 3 furoate, and 2 furyl n pentyl ketone) that are released from octocorals inhibit diatom Nitzschia settlement equivalent to inorganic copper (CuSO 4 5H 2 0) 16 Microencapsulated octocoral extract ( Renilla reniformis ) fouled substantially slower than the bare fiberglass rod s, but fouled faster than the Navy standard copper paint and was deemed non fouling (as determined by percent coverage of marine organisms in the ocean) by day 93 17 Remnants of sponges Decitopsis ceylonia Haliclona cribricutis Sprastella vagabunda, and Aplysina fistularis as well as an unidentified gorgonian were shown to be antifouli ng towards the tubeworm Protolaespira eximia in the Los Angeles harbor 18 Although, there has been some success with natural compounds and their analogs, their limited availability is what makes these compounds unpractical as AF coatings. Furthermore, the goal is to provide surfaces that are non toxic, non eluting, and wi ll not induce significant changes to the marine ecology. These natural coatings do not meet those criteria Coatings that do not contain biocides have bee n studied for their AF and FR properties. The goal is to create a surface that either prevents attachm ent o f marine organisms or reduce attachment strength so that shear force can be used for easy removal 2 Significance of Vascular Grafts for Blood Vessel Replacement Directing mammalian cell behavior using chemical and mechanical c ues at the biomaterial interface can facilitate integration and regeneration of new tissue which
24 may increase the likelihood of a physiologically functional tissue engineered construct. Recently, vascular grafts have been designed using this approach. Va s cular grafts have been utilized to replace damaged blood vessels that impair blood circulation due to atherosclerosis, which is a disease of the vessel where the thickening and stiffening of the arterial wall adversely affects the blood circulation. The at herosclerotic plaque consists of a central core of lipid and cholesterol crystals, which is separated from the lumen by fibrous tissue 19 The mature plaque also contains cells (macrophages, smooth muscle cells, etc. ), proteins and necrotic debris. All of these components contribute to the occlusion of the vessel if left untreated 20 Current Treatment for Arthrosclerosis Many patients treated for coronary atherosclerosis received perc utaneous coronary intervention or cardiac revascularization, such as a coronary artery bypass graft 19 Autologous (i.e. saphenous vein) or synthetic (i.e. Dacron or ePTFE) blood vessels can be used to replace damaged blood v essels. Autologous vessels are the primary choice for da maged or diseased blood vessels, because they rarely provoke an immune response and they contain all of the components found in the native blood vessel. However, t hese blood vessels are not used if ac cessibility is limited or if the blood vessel is in a disease ridden state. If these blood vessels are unavailable or inadequate, surgeons turn to synthetic grafts for treatment 21 Synthetic grafts, such as expanded polytetrafluoroethylene (ePTFE) or poly(ethylene tere phthalate) (Dacron), are used to replace damaged vessels. Expanded PTFE is a fluorocarbon polymer that was patented by Gore in 1969. This non biodegradable material is made by an extruding and sintering process, which endows
25 the material with a tensile mod ulus of 0.5 GPa 22 23 The graft is produced by stretching the melt extruded solid polymer tube, forming a porous tube. The structure can be described as a node fibril structure, where solid nodes are connected through fibrils. The average distance between nodes is 30 m. The Dacron fiber material was patented by DuPont in 1950 and used as a material for a vascular graft in 1957 22 Dacron grafts are available in knitted or woven forms, which have a tensile modulus of 14 GPa 23 The woven grafts have low porosity and minimal creep, due to the over an d under pattern of the threads. The knitted grafts have higher porosity and high ability to stretch or dilate, which is due to the way the threads are looped. To prevent leakage of blood to the surrounding milieu, the knitted Dacron is usually coated with gelatin, albumin, and collagen 22 The pate ncy of synthetic grafts to replace large diameter vessels located in a high blood flow area is relatively high, reaching a 5 to 10 year patency rate of 90% for aortofemoral bypass. This is not the case with synthetic small diameter grafts. Poor performance of synthetic vascular grafts is prevalent when their diameter is less than 6 mm, exhibiting a 5 year patency rate of less than 50% 20 Complications stem from adhesion of platelets and the presence of this fibri n on the inner wall, creating a layer called the pseudointima. Reduction of the cross sectional area of the graft adversely affects the blood circulation due to the higher resistance to blood flow 19 Polyurethane is another ma terial that has been studied for its use as a vascular graft. However, polyurethane vascular grafts and PTFE grafts have been reported to have similar patency rates 24
26 Biological inspired materials have been used more recently to improve biocompatibility. The tissue engineering approach to regenerate functional tissue has increased the popularity of biodegradable po lymers such as polylactic acid (PLA), polycaprolactone (PCL), and polyglycolic acid (PGA). Composites, constructs made out of synthetic material and biological material, have become popular as of late; where the goal is to provide mammalian cells with a hi ghly reliable and easily tailored scaffold to promote growth and restore function. The tissue engineering approach has been used for controlling organogenesis of vascular graft constructs. Development and Composition of Blood Vessels In order to obtai n tissue engineered vascular constructs (TECV), man y developmental processes (e.g., angiogenesis) have been mimicked. Angiogenesis describes the formation of blood vessels from pre existing vasculature. The endothelial cells (ECs) migrate along a certain d irection and proliferate to reach a certain length. The layers of the blood vessel are formed by the recruitment and differentiation of smooth muscle cells (SMCs), fibroblasts, and pericytes by growth factors secreted by the ECs 25 26 Vasculogenesis involves the formation of blood vessels in the embryo from ECs or in an avascular area; whereby stem cells, angioblasts, and hemangioblasts (precursor ECs) form blood vessel s from environmental cues (i.e., vascular growth factors, ex tracellular matrix) 27 The walls of these newly formed blood vessels will consist of three layers: the adventitia, media, and intima. The adventitia, or outside layer, is composed of connective tissue with elastic fibers which imparts the vessel with the ability to stretch and recoil. The intima, or inner layer, consists of a confluent layer of endothelial cells and the basement membrane and is pivotal in preventing blood clots. The media, or
27 middle layer, provides mechanical strength and vessel s ize control and consists of layers of SMCs, collagen, and elastic fibers 28 Collagen, elastin, proteoglycans, glycosami noglycans, and glycoproteins are secreted by SMCs to make the extracellular matrix (ECM) in the media layer 29 The manner in which the SMCs and ECM c omponents are organized a ffect s the structural integrity as well as the vasoactivity, which is the constriction and dilation of the blood vessel 26 The ECM is an important component of tissue because it provid es bioactive molecules and a 3 dimensional structure for cell adhesion. It is these cells and this layer that will be the focus of this study. Phenotype Modulation of Smooth Muscle Cells SMCs must be considered when developing tissue engineered vascular grafts. Modulation of the SMC s to their mature phenotype is necessary to provide physiological ly function al tissue engineered vascular construct. SMCs play a key role in contraction and ECM production by synthesis and degradation of the matrix components when expressing their mature p henotype, the contractile phenotype. The media layer can provide mechanical support for the blood vessel and exhibit a bipolar spindle shape once the contractile phenotype is expressed 30 31 This can be tricky due to the phenotype that is displayed during the developmental stages of tissue regeneration ( in vivo and in vitro ). During initial stages of blood vessel formation, phenotype which is considered to be in a migratory, highly proliferative state, and exhibit s a fibroblast like morphology. The synthetic phenotype is also observed in diseased blood vessels, such as those afflicted by athero sclerosis. These diseased 32 In
28 order to promo te phenotype modulation from the synthetic phenotype to the contractile phenotype, the SMCs will need to be direc ted by environmental cues (i.e., chemistry, topo graphy, mechanical stimulus ). Influence of Physical and Chemical Factors on Mammalian Cell Beh avior Biomolecules have been utilized in the past to direct cells towards tissue regeneration. Proteins, which consist of 20 or more amino acids, are polymerized from the condensation reaction of an amino group of one amino acid and a carboxyl group of ano ther amino acid 19 This linear sequence of the polypeptide chain is called the primary structure. This sequence of amino acids will dictate folding and coiling of the polypeptide chain through hydrogen bonding between an amin e and a carbonyl group in the polypeptide backbone, creating the secondary structure. These functional groups in the backbone chain can participate in intramolecular interaction or engage in interaction with the surrounding milieu (such as water molecules) 33 Further intramolecular interaction (hydrogen bonding, ionic bonding, disulfide bonding, and hydrophobic interactions) of side chains results in a tertiary structure, which governs the 3 dimensional structure of the protein. Intermolecular interactions between multiple polypeptide subunits create the quaternary structure 19 This conformation of the protein can determine its abili ty to adsorb strongly to a substrate. Proteins, along with growth factors and synthetic compounds can act as chemical cues. These chemical cues are sent by means of transmembrane molecules, such as integrins. Integrins are glycoproteins consisting of two 34 35 These transmembrane molecules can bind to amino acid sequences to act as anchors to the cytoskeleton 36 nique ligand binding specificity 34
29 Integrins have been stated to exist in 3 conformational states: (1) bent conformer, (2) intermediate conformer, (3) and the extended conformer. The bent conformer and the extended conformer are indicative of an integrin exhibiting low and high a ffinity, respectively. The intermediate conformer is a transitional state. To induce cell adhesion, integrin activation is needed. The bent conformer is energetically favorable. Peng et al describes the reaction rates of activation k a+ and inactivation k a as a reversible stochastic process, where E a k B and T temperature, respectively. The energy in the form of chemical factors E c or mechani cal factors E m are needed for the transition to the extended conformer 37 The chemical composition and mechanical behavior of the 3 dimensional network of bioactive molecules known as the extracellular matrix (ECM) has been shown to influence adhesion and proliferation of cells 38 Naito et al. showed a positive correlation between fibronectin and type I collagen concentration and cell adhesion 39 Laminin has been shown to resist cell proliferation in the presence of platelet derived growth factor (PDGF) which suggests modulation of phenotype to the contractile state 40 32 The linkage between this protein and the 7 1 integrin is postulated to induce enhanced contractile phenotype 41 i ntegrin + ligand bound ligand
30 Peptides presence, type, and concentration have been shown to influence SMC proliferation. SMCs were found to adhere at a higher rate to glass slides coated with the peptide, KQAGDV, compa red to slides coated with RGDS. Cell migration for KQAGDV and RGDS conjugated glass slides decreased as the protein concentration increased from 0.2 nmol/cm 2 2.0 nmol/cm 2 10 In regards to SMC modulation, fibronectin does not seem to promote maturation to the contractile phenotype. Laminin has been shown to promote the contractile phenoty pe 40 32 Beamish et al. observed higher levels of expression of contractile actin, calponin, and SM slips compared to those on RGD incorporated hydrogels. However, contraction was not observed in response to exposure to carbachol, which was used to examine ligand induced contraction 42 Heparin is another biomolecule that has been found to promote phenotypic modulation of SMCs towards the contractile state by perhaps the inhibition of signaling of exogenous or autocrine bFGF and inducing expression of smooth musc le actin 42 It is well known that mammalian cells can also sense a nd respond to physical forces. Mechanical signals can be sent across the cell membrane via integrins. The amount of stress exerted by the cell at the adhesion sites and the deformation of the substratum in response to the stress is a way in which cells ret rieve information about their mechanical environment. The amount of deformation at the adhesion site can control cell behavior, which is known as mechanosensitivity 36 The modulus or st iffness of a substrate has been shown to effect the phenotypic expression of differentiated tissue cells and stem cells 43 44 45 The matrix elasticity of co llagen coated
31 polyacrylamide (PA) gels appeared to affect the differentiation lineage of human mesenchymal stem cells (MSCs). The collagen coated PA gels exhibiting an elastic modulus of 0.1 1 kPA, 8 17 kPa, and 25 40 kPa induced differentiation of MSCs to neurogenic, muscle, and bone like cells respectively. They confirmed that once the MSCs can feel and respond to the matrix, differentiation into their expected lineage follows 43 The effect of substrate properties on SMC spreading has a lso been investigated, to see if the morphology in the native tissue can be mimicked on biologically inspired materials. The spreading of SMCs was similar on materials with a range of moduli between 0.001 0.008 MPa for polyacrylamide (PA) gels and for po ly(L lysine)/hyaluronic acid (PLL/HA) gels exhibiting a modulus of 0.085 MPa (uncrosslinked) and 1.30 MPa (crosslinked) 46 Modulus did not influence SMC spreading on poly(ethylene glycol) PEG based hydrogels exhibiting moduli in the range of 0.014 0.423 MPa. However, modulus was found to have a more profound effect on cell spreading than ligand density. RGDS concentration was found to have a statistically significant effect on cell spreading, which increased with increasing peptide coverage (ranging from 33 200 nmol/cm 2 ). Although there was a positive correlation Interestingly enough, Peyton et al. found that there was a negative correlation between expression of contractil e markers (calponin and caldesmon) and modulus 12 It seems that at a critical ligand density and mod ulus of the substrate (in the range of 0.014 0.423 MPa) has a substantial effect on SMC morphology, but it does not guarantee a
32 statistically significant difference in cell shape or SMC marker expression of the contractile phenotype. Physicochemical Facto rs that Influence Marine Organism Adhesion Hydrophilic materia ls, such as hydrogels, have been used because of their antifouling abilities towards proteins, bacteria (i.e.: Escherichia coli and Pseudomonas aeruginosa ), and problematic marine organisms (i.e .: Ulva linza Balanus amphitrite ) 4 5 6 It has previously been shown that poly(ethylene glycol) dimethacrylate (PEGDMA) based hydrogels outperformed poly(dimethylsiloxane) elastomer (PDMSe) substrates in inhibition of Ulva spore settlement. The antifouling ability of the PEG based hydrogel s was postulated to stem from their low modulus (approximately 0.01 MPa) and chemical composition 47 The hydrophilicity, or wettability, of these PEG ba sed hydrogels stems from the presence of ethylene oxide (EO) in the side chains or backbone chain. These EO chains bind water strongly via hydrogen bonding. These strongly bound water 48 49 Superhydrophilic materials, such as zwitterionic materials, have shown great antifouling potential towards proteins and marine organisms. These chemistries, i.e.: poly(sulfobetaine) and poly(carboxybetaine), possess negative and positive charges that are believed to create a water layer via e lectrostatic attraction 2 Hydrophobic surfaces tend to exhibit FR properties. A positive correlation was found between percent removal of sporelin gs and degree of hydrophobicity of polyolefin coatings: Polypropylene (PP), high density polyethylene (HDPE), polypropylene polyethylene (PPPE) copolymer, and ethylene vinyl acetate (EVA 12). A correlation of 0.93 was found between the contact angles (rang
33 surfaces and Ulva spore percent removal. The percent removal in this assay was 22.4, 14.2, 23.8, 31.3, 35.3 for PDMSe, EVA 12, PPPE, HDPE, and PP coatings, respectively 50 PDMSe samples with topographically modified channels were found to exhibit a positive correlation between hydrophobic ity (i.e.: contact angle measurements) and Ulva spore density. However, the most hydrophobic surface, the PDMSe with the 51 Superhydrophobic surfaces usually These surfaces are appealing as foul resistant surfaces due to their inherent self cleaning capabilities. These surfaces tend to carry away dirt as the droplets roll away off the surface. This phenomenon is obser ved in nature with t he lotus leaf. This leaf has hierarchical topography consi sting of microtopography (3 10 m protrusions), as well as nanotopography (70 100 nm features of epicuticular wax crystalloids). The lotus leaf has been reported to have a contac Surfaces with micro scale topography and absence of hierarchy, such as the rame leaf, have also been reported to exhibit superhydrophobic characteristics ( exhibiting a due to presence of 1 Wet chemistry has been used to create superhydrophobic surfaces from materials made out of metals and polymers. Exposure of copper to n tetradecanoic acid or exposure of nickel to monoalkylphosphonic acid has been proven to create superhydrophobic surfaces Furthermore, Polycrystalline metals exposed to basic or acidic solutions create superhydrophobic surfaces due mainly to topography 52 Not all organisms found in the marine environment are inhibited from settlement by hydrophobic surfaces th ough Diatoms adhere strongly to hydrophobic surfaces
34 compared to hydrophilic surfaces 2 Differences in the substrata settlement preference even exists between species. For instance, B. amphitrite tend to exhibit higher settlement on hydrophilic surfaces, while Balanus improvises ( B. improvises ) prefer to settle on hydrophobic surfaces 13 This is why amphip hilic coatings have gained popularity. An amphiphilic coating contains both hydrophobic (i.e.: fluoropolymer) and hydrophilic (i.e.: poly(ethylene oxide)) moieties. Incorporating both of these components into the coatings is postulated to create an uninhab itable surface for settlement of many types of marine organisms that adhere strongly to either hydrophili c or hydrophobic surfaces, i.e., Ulva spores and diatoms, respectively 9 A majority of the amphiphilic coatings tested against Ulva zo ospores contain a combination of hydrophilic PEG and hydrophobic fluoropolymers (FP). Low Ulva spore settlement tends to be found on PEG FP brushes with the highest PEG content; however, this type of surface also requires high water jet pressure for spore removal 53 54 Other marine organisms, e.g., diatoms and barnacles, seem to be inhibited by amphiphilic coatings as well, performing well in laboratory barnacle settlement assays and in field tests by panel immersion in the m arine environment 9 55 56 PEG FP based crosslinked networks tested against Ulva spores and diato m Nitzschia exhibited lower density compared to the glass standards or PDMSe control 57 58 It has been reported that incorporating a high concentration of low molecular weight, hydrophilic monomers like hydroxyethyl acrylate (HEA) in PDMS co polyisocyanate crosslinke d networks exhibited high percent removal of diatom Navicula and Ulva spores 59
35 Physical and Chemical Factors that Influence Bioadhesion Influence of Surface Energy on Fouling Bio adhesi on has been reported to be influenced by surface wettability Wettability is a term used to de scribe the tendency of a liquid (i.e., water) to spread on a surface 60 This parameter can be assessed by performing contact angle measurements; whereby, droplets are placed onto a surface and the cont act angle between the solid vapor interface and the liquid vapor interface is measured. The contact angle of the droplet on the s urface is shaped by the balance of forces between the liquid, solid, and vapor. Contact angle measurements are commonly used to determine the surface energy of a 61 62 where and represents the interfacial free energies of the solid vapor, solid liquid, and liquid vapor interfaces, respectively. The Owens and Wendt two liquid geometric mean approach can b e used to calculate the dispersive and the polar force components from contact angle measurements performed using liquid pairs, such as methylene iodide water (MeI 2 water) and methylene iodide ethylene glycol (MeI 2 EG). The and represent the surface free energies of the solid and liquid, respectively. The parameters and represen t the dispersion force components and polar force components of the solid and liquid, respectively 63
36 Figure 2 1. Relationship between relative bioadhesion and critical surface tension of a material. Adapted from Baier et al. 64 Baier et al. determined that the critical surface tension needed to act as a fouling release coating should be betwe en 20 and 30 mNm 1 as seen in Figure 2 1 The rate of fouling was reported to be low for surfaces within that surface tension range; in addition, cleaning of the surface was deemed to be easy as well. PDMS was shown to be the best (compared to fluorinate d polymers, polyurethanes, epoxides, copper based polymer, and antibiotic containing polymer) at providing low toxicity to marine organisms and exhibiting low fouling properties 65 The foul resistant nature of these materials stem from poor adhesion between the substratum and the fouling organism s This phenomenon is supported by the work of adhesion equation which describes the amount of work needed to separate a solid from a liquid. The level of adhesive strength is described in the following equation,
37 where and represents the surface energy of the solid and liquid, respectively, and the represents the interfacial tension between the solid and liquid. Based on the work of adhesion equation, it would be advantageous to design a material with very low surface energy in order to obtain weak adhesion 66 Influence of Mechanical Properties on Fouling However, this does not explain the fouling release properties seen for coatings with varying degree of hydrophobicity. RTV 3140 silicone (Dow Corning), a commercial si licon, FEP Teflon (Dupont), and epoxy Mil Spec F150 required a shear water jet removal pressure of 0.11 0.02, 0.12 0.4, 0.87 0.33, and 2.23 0.55 Pa, respectively. The FEP Teflon coating with the lowest surface (17 mNm 1 ) energy did not possess w eaker adhesion to barnacles 14 To receive a better correlation between adhesion and surface properties, it has been reported that the elastic modulus must be considered. The weaker adhesion between the barnacles and the silicone, compared to Teflon is most lik ely due to the low modulus of the silicone Berglin et al. have shown that the detachment stress of pseudobarnacles is proportional to (E ) 1/2 67 The correlation between the relative adhesion and (E ) 1/2 has been found to be greater (R 2 = 0.89) compared to the relative adhesion versus the surface energy or elastic modulus alone (R 2 = 0.75 and 0.82, respectively) 68 as been used to describe this phenomenon. The critical pull off force P c required for formation of an elliptical cavity of a substrate,
38 where E represents the ela stic modulus of the substrate, represents the contact radius, and G c represents the critical fracture energy. The G c can be calculated from the following equation, where A represents the area at which the pull off force is acting on and e represents the pull off force required for a ri gid disk from a thin elastomer film (t << a) attached to a rigid material, where t, a, and K, represent the film thickness, contact radius, bulk modulus of the film, respectively. In the instance when the elastomeric a dhesive film is thick (t >> a), the equation modified further 68 The influence of substrate on other marine organisms has also been investigated. Ulva spore settlement has been shown to be influenced by the mechanical properties of crosslinked networks. Chaudhury et al. have sho wn that the percent removal of spores was negatively correlated with the elastic modulus of the PDMS crosslinked networks 69 Crosslinked Network Systems Hydrogels have gained immense popularity as a biomaterial due to their low antigenicity and hydrophilicity. Hydrogels are insoluble, crosslinked networks that are synthesized often times from monomers with acrylate or methacrylate groups, such as 2 hydroxy ethyl methacrylate (HEMA). H ydrogels can be polymerized by hea t, ultraviolet
39 (UV), or gamma radiation in the presence of redox initiators (i.e.: ammonium persulfate sodium metabisulfite), photo initiators (i.e.: Irgacure 2949), or no initiators, respectively. This type of crosslinked network is highly permeable and t end s to imbibe large amounts of water which is why these materials have been studied for their use in drug delivery, wound healing, and ophthalmic applications 70 71 The amount of water that can be imbibed is dependent upon steric effects of the polymer chai ns and polarity effects of the chain. The polarity effects are determined by the chemical composition of the polymer chain in the network system. A disadvantage of using hydrogels lies in their poor mechanical integrity, which stem from their high swellin g capabilities. The water in the hydrogels acts as a plasticizer, decreasing the stiffness of the material In order to improve the robustness of hydrogels, hydrophobic monomers have been incorporated into hydrogels to reduce the plasticizing effect of wat er. Siloxane based mono functional and di functional monomers have been reported to impart hydrogels with improved mechanical properties, i.e. elastic modulus and tear strength 8 Flexibility of the network will still be intact through the presence of Si O Si l inkages. The flexibility of the Si O bond is attributed its length of 0.164 nm, compared to the C C bond length of 0.154 nm. In addition, t he Si O C C 72 The molecular weight or the concent ration of the mono functional and di functional monomer s in a crosslinked network can affect the elastic modulus of the network as well There is a direct correlation between the network structure and the mechanical
40 properties of the network system accordi ng to the following equation that describes shear modulus G where T and M c represents the density of the polymer, gas constant, temperature, and molecular weight between crosslinks respectively. The shear modulus is related t o where Therefore, if stiffer gels are desired, i ncreasing the concentration of the siloxane crosslinker should crea te a network that is tighter (i.e., M c ) and mor e dense. Amphiphilic hydrogels composed of 2 hydroxyethyl methacrylate (HEMA), 1,3 bis(3 methacryloxypropy(trimethyl siloxy)disiloxane (MPTSDS), and tris(trimethylsiloxy) 3 methacryloxypropylsilane (TRIS) were copolymerized using ultraviolet (UV) light an d Irgacure 2959 as the photo initiator The siloxane monomers, MPTSDS and TRIS, will impart the PHEMA based hydrogels with hydrophobic moieties. TRIS is included in the network to maintain the concentration of the siloxane in the material, while not cont ributing to crosslinking. This allows for changes in mechanical properties without there being noticeable changes in the chemical composition of hydrogel. The influen ce of the hydrogel network structure on protein adsorption and Ulva linza settlement was studied. The surface properties (i.e., surface energy, hysteresis, mesh size) were postulated to have a higher impact on protein adsorption compared to the bulk properties (i.e., elastic modulus). On the other hand, the bulk mechanical properties (i.e., el astic modulus) were postulated to generate a more profound effect on
41 fouling of Ulva linza Similarities in trends found between fouling of protein and fouling of the marine organism may mean that the mechanism for fouling for th e unicellular marine organi sm is controlled by the presence of adhesive pad that is produced by the Ulva organism that contains high molecular weight glycoproteins 73 Proteins and peptides h ave previously been grafted (b y plasma, IR/gamma radiation, or chemical modification) or physi adsorbed to hydrogels with improved results in regards to biocompatibility 74 75 76 The advantages of using peptide sequences as opposed to protein molecules include lower likelihood of conformational changes, higher control of ligand density a nd orientation, lower prevalence of imm une response s and stability during sterilization 77 For example, su rface coverage was higher for fibronectin adhesion peptide (F AP) tethered poly(HEMA co MMA) hydrogels 2 ) compared to laminin or fibronectin, tethered poly(HEMA co MMA) 2 ), which may have been due to the lower molecular weight of the peptide compared to the protein molecule 78 In order to fabricate a material that is suitable for smooth muscle cell growth and maturation the mechanical properties were also considered in order to induce physiologically functional smooth muscle cells (SMCs). Amphiphilic crosslinked networks were fabricated to have tunable elas tic moduli in the range that is relevant for native small diameter blood vessel tissue 79 Peptide conjugation to the hydrogels was performed by reaction of the epoxy group in the glycidyl methacrylate to the pep tide, RGDS. The influence of mechanical and chemical cues on SMC attachment and proliferation was studied.
42 CHAPTER 3 PHYS ICAL AND CHEMICAL PROPERTIES OF POLY(2 HYDROXYETHYL METHACRYLATE) HYDROGELS Introduction Poly(2 hydroxyethyl methacrylate) (PHEMA) based hydrogels have numerous appl ications in medicine, due to their similarity to the native tissue High water content, low modulus, high wettability are just some of the reasons this type of crosslinked network have been studied for drug delivery, opht halmological applications, and artificial skin 80 This water imbibing material has been shown to display resista nce to protein, which makes them desir able for applications that require implantation into the body Aside fr om the biomedical applications, hydrogels have been studied as marine antifouling applications due to their resistance to biomolecules and bacteri a PHEMA based hydrogels have been shown to e xhibit resistance to attachment of bacteria, Escherichia coli and Pseudomonas Aeruginosa 4 These hydrogels have been shown to inhibit settlement of marine organisms: cyprids of barnacles ( Balanus amphitrite ), zoospores ( Enteromorpha intestinalis ), and diatoms ( Me losira nummuloides ) 81 6 The antifouling abilities of these brushes and crosslinked networ ks stem from the hydro xyl and carbonyl moieties which engage in hydrogen bonding with water molecules in aqueous environment s Biomolecules, such as proteins, cannot easily displace the tightly bound layer of water that is formed 48 49 This resistance to protein adsorption may be why these hydrophilic crosslinked networks tend to exhibit antifouling properties towards marine organisms 49 The low modulus ( approximately 0.01 MPa) of PEGDMA based hydrogels has also been postulated to play a role in antifouling towards marine organisms 47 Elastic modulus was shown to influence percent removal of m arine organisms on poly(dimethyl
43 siloxane) (PDMS). Chaudhury et al. found that the percent removal of Ulva spores and sporelings tends to be negatively correlated with substrate elastic modulus (ranging from 0.2 9.4 MPa). However, the percent removal lev eled off at a certain critical modulus, 0.8 MPa and 2.4 MPa, for spores and sporelings, respectively 69 To date, the correlation bet ween modulus of hydrogels t owards resistance of marine organisms has not been inves tigated. T he mechanical properties of hydrogels stem from their network structure. Physical or chemical linkages are needed to form the 3 dimensional network of a hydrogel Physical crosslinks can be formed through van der Waals bonding, hydrogen bonding or even physical entanglements. Chemical crosslinking of hydrogels is quite popular since it provides improved mechanical integrity over physical crosslinks in various environments. Chemical crosslinks are formed by covalent bonding with a monomer that h as a functionality greater than 2. PHEMA hydrogels can be polymerized by heat, ultraviolet (UV) light or gamma radiation in the presence of redox initiators, photo initiator s, or no initiators, respectively. Heat and the redox initiators, ammonium persulf ate and sodium metabisulfite, were used to polymerize PHEMA hydrogels. The mol ecular weight between crosslink junction s and density of cro sslink junctions has been shown to influence the mechanical properties of PHEMA hydrogels 82 83 Since hydrogels are crosslinked elastomeric materials the shear modulus G and the molecular weight between crosslinks M c is related in the following manner where R and T represents the density of the polymer network, gas constant, and absolute temperature, respectively. The molecular weight between crosslinks M c is
44 related to the number of active network chain segments per unit volume n in the following manner 84 Either increasing the crosslink density or decreasing the M c results in a hy drogel with increased modulus. PHEMA hydrogels were fabricated with tunable mechanical properties by varying the di functional monomer, di(ethylene glycol) dimethacrylate (DEGDMA), concentration in the feed composition to vary the M c Many of the parameter s that influence the mechanical properties also affect the swelling behavior. The swelling behavior of hydrogels has been shown to be influenced by the prepolymerization parameters: chemical composition of monomers, crosslinking concentration and molecular weight, curing method, and diluent concentration and type 70 85 86 After polymerization, the hydrogel network may be affected by solvent composition, pH, and temperature. T he equilibrium swelling behavior of crosslinked networks can be described by the Flory and Rehner equation The swelling behavior was theorized to be dictated by the (1) e ntropy changes from the polymer and swelling agent interaction, (2) entropy changes from the reduction in possible chain conformations upon swelling, and (3) the heat of mixing of the polymer and swelling agent. The equation is often times used to calculate number of active network chain segments per unit volume n by performing some swelling tests; whereby the volume fraction of the polym er the molar volume of the solvent V 1 and the Flory Huggins polymer solvent interaction parameter 1 is either obtained or known. The interaction
45 parameter 1 for the fabricated PHEMA hydrogels which is influenced by the molecular interactions between the solvent and the network chains was calculated from the following equation 87 The molecular interactions between the solvent and network chains influence the wettability of the PHEMA hydrogels These hydrogels should exhibit hydrophilic characteristics in a swollen state, as seen by captive air bubble measurem ents. The hydrophilicity stems from the hydroxyl and carbonyl groups, which participate in hydrogen bonding. The presence of these hydrophilic moieties was detected using Fourier transform infrared (FTIR) spectroscopy. FTIR ATR uses measurements made about the re flected signal from the sample crystal interface. The attenuated total reflectance (ATR) mode was used to assess the presence of hydrophilic moieties within approximately 2 d p for this material was calculated from the following equation, where and represent the wavelength, incident angle, and index of refraction of the crystal (diamond, n 1 2 88 Hydrogels were fabricated with varying di functional monomer, DEGDMA, concentration with the intention of obtaining a range in modulus. The network properties of HEMA hydrogels were investigated, to dete rmine if this is a suitable composition to study the effect of mechanical properties on biofouling. The effect of DEGDMA concentration on the surface properties, network structure and mechanical properties
46 was investigated using contact angle measurements swell testing, and compression testing Experimental Section Materials Diethylene glycol dimethacrylate (DEGDMA) and 2 hydroxyethyl methacrylate (HEMA) were purchased from Sigma Aldrich and used as received Sodium metabisulfite (SMBS) and ammonium per sulfate (APS) were purchased from Fisher Scientific and Sigma Aldrich, respectively. N octane was obtained from Fisher Scientific. cm) was obtained from the Thermo Scientific Branstead EASYpure II system and was used for all polymerization mixtures and swelling experiments. HEMA Hydrogel P reparation HEMA (9 grams) and DEGDMA were added to glass bottles. Initiators, APS SMBS (1:1 ratio, 0.3 mol% relative to monomer), and DI water were added to the glass bottles (Table 3 1) The prepolymerization mixtures contained a DEGDMA molar ratio (DEG DMA /HEMA molar rati o = mol DEGDMA/mol HEMA) of 1, 5, 10, and 25 mmol/mmol The prepolymerization mixtures were purged with nitrogen for approximately 2 minutes and then placed in a centrifuge (2000 RPM) to degas approximately 2 minutes. The prepolymerizati on mixture was transferred by a pipette into a mold consisting of a polydimethylsiloxane elastomer gasket sandwiched between two hexamethyldisilazane treated glass plates, which were held together with paper clamps. The mold was then placed in the oven in a vertical position for 3 hours at 50 C. After curing, the hydrogel films were soaked in deionized water for 24 hours to remove residual monomer and initiator.
47 Table 3 1. Composition of poly(HEMA) based hydrogels Volume Fraction of Polymer Discs (6 mm) were punched out of the hydrogel films. The weight after polymerization was ob tained in water W a,r and hexane W h,r The samples were soaked in water until an equilibrium water content was reached which occurred when the weight of the hydrogel remained constant over time The weight of the hydrogels in a swollen state was measured i n water W a,s and hexane W h,s The samples were then placed in a vacuum oven and left under vacuum (28 in Hg) until an equilibrium dry weight W a,d was reached which occurred when the weight of the hydrogel remained constant over time The volume fraction o f the sample after the reaction was calculated from the following equation, p s represents the density of the non swelling solvent. The water equilibrated weight mea surements and the volume of the polymer V p were used to calculate the volume fraction of the polymer in the swollen state 87 DEGDMA/HEMA Molar Ratio (mmol /mmol ) HEM A (g) Water (g) APS (g) SMBS (g) DEGDMA (mL) 1 9 6 0.02 0.02 0.02 5 9 6 0.02 0.02 0.08 10 9 6 0.02 0.02 0.15 25 9 6 0.02 0.02 0.39
48 Compression Testing The moduli of the crosslinked networks were determined by compression testing using the TA.XT.plus Texture Analyzer. Discs (8mm) were punched out from swollen gels. Mechanical testing was performed in compression at a 1 mm/min strai n rate in DI water at room temperature. The compression modulus of the crosslinked networks was calculated from the stress 89 Contact Angle Measurements Contact angle measurements were performed using the captive air bubble technique with the Ram Hart Goniometer; whereby swollen samples were placed upside down on top of rubber stoppers in a tank filled with DI water. Air or n octane bubbles (3 l) were placed in five different spots for each composition. The two liquid geometric mean approach was used to calculate the surface energy of the solid whe re and represent the dispersion surface tension components and polar surface tension components of the solid and water, respectively 90 The and represent the surface free energy of octane, the water octane interfacial tension, and contact angle for octane and air as the probe liquid, respectively. The surface free energy polar and dispersion force components for water are 50.8
49 mNm 1 and 21.8 mNm 1 respectively. The surface energy of n octane and interfacial tension for water octane is 21.8 mNm 1 and 50.8 mNm 1 re spectively 90 91 Fourier Transform Infrared Spectroscopy Attenuated Total Reflectance (FTIR ATR) Dried samples were evaluated with the Thermo Scientific Nicolet 6700 FTIR ATR equipped with a diamond crystal flat plate. T he scan number and scan resolution was set at 36 and 4 cm 1 respectively. The OMNIC software was used for spectrum analysis. Amido Black Assay The amido black assay was performed to evaluate the protein adsorption onto the PHEMA hydrogels, as well as the assay standards (TCP = tissue culture polystyrene and NC = nitrocellulose). The following assay w as adapted from a previous report 92 All rinsed with DI water to remove unadsorbed protein. The samples were soaked in the amido b lack stain solution (methanol, DI water, Naphthol Blue Black) for 3 minutes. After rinsing wit h DI water, the sample s were rins ed with the wash solution (90 v/v% methanol, 8 v/v% water, and 2 v/v % acetic acid) twice to remove unbound dye. The samples wer e rinsed with DI water once again. This was followed by soaking the hydrogels in the eluent solution (50 v/v% ethanol, 50 v/v% 50 mM NaOH with 0.1 mM EDTA) for 30 minutes on a plate shaker (at 300 RPM). The BioTek microplate reader equipped with a Gen5 Sof tware was used to measure the UV absorbance at a aliquot of the dye labeled protein that was previously adsorbed to the samples. Five replicates of each
50 composition were used f or the protein adsorption assay. A standard curve was made from serial dilutions of 6 mg/mL of naphthol blue dye conjugated BSA in PBS. Figure 3 1. Standard curve for amido black protein adsorption assay. Statistical Analysis The data was reported as mean standa rd deviation. Analysis of variance (ANOVA) was used to make inferences about the sample means W procedure was used to make multiple comparisons between sample means ( = 0.05) Results The poly(HEMA co DEGDMA) hydrogels with varying crosslink molar ratio were found to exhibit differing crosslink structure, as indicated by the difference in volume fraction of polymer in a swollen state as well as differing mole cular weight between crosslinks M c (Table 3 2) T he increased with increasing concentration of crosslinker in the feed composition. C hanges in the volume fraction of the polymer were not substantial enough to make any conclusions about the effect o f the crosslink molar ratio and the network structure. As expected, there was a negative correlation between the y = 0.3719x + 0.0133 R = 0.9963 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 Absorbance (a.u.) Protein Concentration (mg/mL)
51 crosslink molar ratio and the M c The increased concentration of the di functional monomer, DEGDMA, resulted y due to the higher presence of crosslink junctions. lower EWC as seen in Figure 3 2 ; a lt hough gnificantly between networks with different chemical composition The molecular weight of the crosslinker is shown to influence the network structure As expected, the M c increased with crosslinking agents exhibiting a higher molecular wei gh t. A M c of 1000 g/mol was not obtained due to the high concentration of low molecular weight monomer, HEMA. It seems as though the directly affected the EWC of the gels, which was expected. The increase in the volume fraction of polymer results in lower water adsorption due in part to the denser network Table 3 2. Swelling and mechanical properties of PHEMA hydrogels DEGDMA/H EMA Molar Ratio (mmol/mmol) M c (g/mol) EWC (%) DEGDMA_ 1 0.69 0.04 0.85 149 29 1 DEGDMA 5 0.67 0.02 0.83 180 25 5 DEGDMA 10 0.73 0.04 0.88 110 26 3 DEGDMA 25 0.75 0.04 0.90 89 22 2 The higher presence of crosslink junctio ns, as the crosslink molar ratio is increased, resulted in hydrogels with higher modulus (as seen in Figure 3 2 and Figure 3 3 ). The higher number of chains contributing to the resistance of the compressive stress explains the higher modulus obtained for t he hydrogels with increased crosslink molar ratio. This behavior is co nsistent with what has been reporte d in the literature concerning PHEMA based hydrogels 82 Although there was a statistically significant difference in modulus between hydrogels prepared, the modulus range (0.4 1.1 MPa)
52 does not seem to vary as much as anticipated. A hydrogel with a modulus lower than 0.4 MPa upon handling could not be obtained with the monomers that a re uti lized in the prepolymerization mixture. A molar ratio of more than 50 mmol/mmol was found to lead to defects (i.e., cracks) that compromised the mechanical integrity of the gels. The initial water content, crosslinking agent concentration, and molec ular weight of monomer in the feed composition have been used to obtain hydrogels v arying modulus 70 93 An increase in crosslinking agent co n cen tration has been shown to increase tensile strength of poly(HEMA co EGDMA) hydrogels. The largest increase in tensile strength (0.73 4.8 MPa) was seen at higher crosslinking concentrations (0.4 5.7 mol/ cm 3 ) 83 Th e same behavior was observed wit h the fabricated poly(HEMA co DEGDMA) hydrogels Figure 3 2 Compression m odulus and M c of PHEMA hy drogels with varying DEGDMA/HEMA molar ratio (mmol/mmol) Modulus is stati stically significant ( p < 0.05) between all gels 0 2 4 6 8 10 12 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 5 10 15 20 25 30 1/M c (mol/kg) Modulus (MPa) DEGDMA/HEMA Molar Ratio (mmol/mmol)
53 Figure 3 3 Compression s tress deflection curve of PHEMA hydrogels with varying DEGDMA/HEMA molar ratio (mmol/mmol) Figure 3 4 FTIR A TR spectra of PHEMA hydrogels of varying DEGDMA/HEMA molar ratio (mmol/mmol) -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0 Stress (MPa) Deflection (mm/mm) 1 5 10 25 25 10 5 1 Composition
54 The chemical composition of the crosslinked networks was evaluated using FTIR ATR spectroscopy (Figure 3 4 ) As expected, the spectra of all of the compositions looked similar. T he hydroxyl group in the HEMA was detected at the characteristic peak, 3400 cm 1 The peak around 1700 cm 1 originates from the carbonyl group and the signal at approximately 1100 cm 1 comes from ethylene oxide group. The CH stretching and CH 2 bending peak was present at 2949 and 1452cm 1 respectively. These hydrophilic moieties are what give the hydrogels high wettability, which is defined as the ability o f water to spread on a surface. As seen in Table 3 3, the hydrogels exhibit low contact angles using the captive air bubble t echnique. This indicates that hydrophilic moieties are present at the surface in an aqueous environment Furthermore, these surfaces do not appear to be oleophilic due to the low contact angle obtained when n octane is used as a pro be liquid. The surface energy calculated from the contact angle measurements with air and n octane as probe liquids confirm that the surfaces of these hyd rogels are hydrophilic in w ater, exhibiting a surface energy 61 mN m 1 Th e surface energies determined for these PHEMA hydrogels is consistent with what was found in the literature with similar compositions 94 The increased hydrophilicity obser ved for the hydrogel with the higher DEGDMA /HEMA molar ratio most likely stems from the higher concentration of ethylene oxide groups in the crosslinked network, due to the higher concentration of the DEGDMA and HEMA The only hydrogel that is deemed stati stically different (p < 0.05) from the others is the hydrogel with a DEGDMA /HEMA molar ratio of 25 mmol/mmol.
55 Table 3 3. Surface properties of poly(HEMA co DEGDMA) hydrogels DEGDMA /HEMA Molar Ratio (mmol /mmol ) Captive Bubble Contact Angle Surface Ener gy (mNm 1 ) Air n Octane 1 33 4 34 10 62 2 5 35 2 42 9 61 2 10 32 3 41 7 63 2 25 26 2 34 6 66 1 The hydrophilicity ( > 60 mNm 1 ) and low modulus (E < 1.1 MPa) may endow these hydrogels with antifouling pro perties to wards biomolecules and marine organisms The hydrophilicity and low modulus most likely stem from the three states of water that can exist in a polymer network in most hydrogels: (1) bound water, (2) interfacial water, and (3) bulk water. The bulk water is not associated with the network system and the interfacial water is loosely associated with the network while the bound water is strongly ass ociated with the chain segments 95 96 Bulk water and interfacial water tend to decrease with decreasing molecular weight between cross links and higher crosslink density 83 However, the bound water has been postulated to remain constant and has been detected in PHEMA based hydrogels with low w ater content (i.e.: < 18 % EWC ) 95
56 Figure 3 5 Protein adsorption of BSA onto PHEMA of varying DEGDMA/HEMA molar ratio (mmol/mmol) TCP, and NC Ast eris k (*) denotes protein adsorption that is statistically significant (p < 0.05) from TCP. The amido black assay was used to assess protein adsorption of the different hydrogel compositions and assay controls, NC and TCP. The protein coverage between all of the hydrogels was deemed not statistically different (Figure 3 5 ) The protein coverage of all of the hydrogel samples were not statistically different to the protein coverage found on the TCP samples. The protein coverage found on NC surfaces was statistically different from protein coverage found on all other surfaces ( = 0.05) This sort of preliminary bio adhesion study shows that the gels may have similar fouling abilities, which may stem from the similarity between gels in regards to wettability (captive air bubble measurements ranging from 62 to 66 mNm 1 ) Intere stingly enough, the adsorption between the hydrophilic surfaces was similar to the hydrophobic surface, TCP, which is composed of polystyrene. The albumin has been shown to prefer more 0 10 20 30 40 50 60 70 80 1 5 10 25 TCP NC Protein Coverage ( g/cm 2 ) Composition
57 hydrophobic surfaces, which may be a result of the presence of hydropho bic domains on the protein itself 97 98 The modulus was not expected to have an effect on the protein adsorption due to the narrow range in bulk modulus of the samples and also due to the small size of the protein. The R 2 value for the surface energy or modulus E versus the protein conce ntration is 0.16 and 0.43 respectively ( Figure 3 6 ) Figure 3 6 The relationship between p rotein adsorption and either modulus (blue diamond) or surface energy (red square) of amphiphilic crosslinked network s Conclusion No noticeable changes in the volume f ra ction of polymer were observed by varying crosslinker concentration in the feed composition The water content between hydrogels also did not differ substantially between compositions and were found to c orrespond well to what is reported in the literature 86 70 The compression modulus for these hydrogels did not vary as much as anticipated, which was also the case for the surface en ergy. The protein adsorption was not statistically different between all hydrogels. R = 0.4315 R = 0.1568 0 10 20 30 40 50 60 70 80 0 0.5 1 1.5 2 0 2 4 6 8 10 Surface Energy (mN/m) Modulus (MPa) Protein Coverage (g/cm 2 )
58 Surprisingly enough, t he amount of protein attaching onto a sample correlated the strongest (R 2 = 0.43) with t he modulus of the hydrogel The protein adhesion may be dictat ed by a surface parameter, such as hysteresis, that is influenced by the network structure of the hydrogel A hydrogel with a wider range in elastic modulus will be created and tested with unicellular organisms in order to invest igate the validity of the c laim that ( E ) 1/2 exhibits a stronger correlation to relative adhesion than modulus and surface energy of the substrate alone.
59 CHAPTER 4 SURFACE PROPERTIES OF AMPHIPHILIC CROSSLINKED NETWORKS WITH TUNABLE MECHANICAL PROPERTIES Introduction Biofoul ing, which is defined as the attachment of macromolecules and living organisms onto a surf ace, occurs quite frequently in the marine environment. Fouling on the hull of navy vessels is problematic, because it results in high operation and maintenance costs The antifouling coatings currently used to curtail this problem are deemed harmful to the marine environment due to the release of toxic biocides; therefore, non toxic, foul resistant materials are being studied as an alternative 2 Hyd rophilic coatings, such as hydrogels, have shown great potential as an antifouling material towards proteins, bacteria (i.e.: Escherichia coli and Pseudomonas aeruginosa ), Ulva spores, and barnacles 4 5 6 47 The antifouling ability of the hydrophilic, ethylene oxid e (EO) based coatings stem from th e steric effects of the long EO chains and the strong interaction of water with the hydrophilic chains 48 This strong interaction with the water molecules, known as hydrogen bonding, may be inhibiting adsorption of biomolecules and consequently marine organism settlement on to EO chains 48 4 9 H ydrogels possess favorable antifouling properties in short labora tory assays, but have been shown to be unsuitable for long term assays in a marine environment, due surface instability and degradation 5 7 Hydrogels that have been fabricated as interpenetrating polymer networks (IPN) tend to exhibi t improved mechanical robustness compared to the hydrogels composed of only one of the two monomers. An interpenetrating polymer network of poly(2 acrylamide 2 methyl 1 propanesulfonate)/ polyacrylamide ( PAMPS /PAAm DN ) hydrogels have be en shown to be resistant to
60 degradation in the marine environment for up to 3 months and exhibit acceptable antifouling properties towards barnacles ( resulting in barnacle surface area coverage < 50 %) 7 Another approach to creating robust hydrogels invo lves incorporating hydrophobic moieties into the hydrophilic crosslinked network to lower the uptake of water and reduce the prevalence of hydrolysis The presence of water has a plasticizing effect on the network structure and can compromises the mechanic al integrity of the material, which limits their use for many applications. The increased hydrophobicity of the crosslinked structure should in fact lead to improved mechanical integrity of hydrogels in an aqueous environment 8 In this work, amphiphilic crossl inked networks were fabricated as potential antifouling materials. HEMA was copolymerized with hydrophobic monomers, 1,3 bis(3 methacryloxy propyl)tetrakis(trimethylsiloxy)disiloxane (MPTSDS) and tris(trimethyl siloxy) 3 methacryloxy propylsilane (TRIS). T he amphiphilicity of the gels stems from the inclusion of hydrophilic monomer, HEMA, and the hydrophobic monomers, TRIS and MPTSDS in the crosslinked networks. The surface properties of these crosslinked networks were studied as a function of the di functi onal monomer (MP TSDS) concentration These properties are important to characterize due to their influence on biofouling 99 100 101 The MPTSDS concentration was defined as the MPTSDS /(TRIS+HEMA) molar ratio (equivalent to mol MPTSDS/(mol HEMA + mol TRIS) ) Since these materials have hydrophobic and hydrophilic moieties the surface propert ies may vary with differing feed composition, but not by much. changes in the
61 chemical composition of the hydrogel, due to the inclusion of TRIS in the network to keep the siloxane concentr ation in the material constant. The surface properties of amphiphilic coatings are highly dependent upon the surrounding milieu. For example, when the surface is exposed to a polar solvent (such as water) the surface will reconfigure to lower the total free energy of the system by expos ing the polar moieties to the polar solvent. For HEMA based hydrogels in particular, the hydroxyl and carbonyl groups will interact with water molecules via hydrogen bonding. When the material is exposed to air, or a nonpolar solvent, the nonpolar moieties (such as the ethylene backbone methyl and siloxane pendant groups ) of the hydrogel will be exposed to the surface. This reconfiguration is possible due to the flexibility of the polymer chains 90 102 103 When the di functional monomer ( MPTSDS/(TRIS+HEMA) ) molar ratio is varied, t he chain mobility, chemical composition, and topogra phy of a surface may vary as well All of these surface characteristics influence the hysteresis of a substrate 103 104 Karlsson et al. reported that hydrogels possessing a higher crosslinker concentration exhibited low er measured hysteresis due to the lower mobility of polymer chains and decreased ability of the probe liquid to penetrate the substrate 103 Similar behavior is expected to be seen with the following amphiphilic hydrogels. The hysteresis in the siloxane crosslinked hydrogels should be influenced by th e very mobile pendant groups and the crosslinker concentration. The motivation for investigating this surface parameter stems from the belief that it may also influence the foul resistant properties of a material. Previous studies suggest that the magnitu
62 fouling release properties of marine foulers. The dynamic contact angle derived hysteresis was reported to be positively correlated with the percent removal of Ulva spores on polyolefin co atings: high density polyethylene, polypropylene, polypropylene polyethylene, and ethylene vinyl acetate 50 I t has also been reported that the hysteresis of perfluoroalkyl coatings was positively correlated with fouling of marine organisms in the Chesapeake Bay 105 In or der to evaluate the surface properties of the siloxane crosslinked amphiphilic crosslinked networks, static and dynamic contact angle measurements were performed. Swelling tests were performed to determine the equilibrium water content and mechanical testi ng was done to examine the effect of crosslinker concentration on the compression modulus of the amphiphilic crosslinked networks. The amido black assay was used to evaluate the fouling properties of these crosslinked networks toward protein Experimental Section Materials Hexanol and 2 hydroxyethyl methacrylate (HEMA) were used as purchased from Fisher Scientific and Sigma Aldrich, respectively. T ris(trimethylsiloxy) 3 methacryloxypropyl silane (TRIS) and 1,3 bis(3 methacryloxypropyl)tetrakis (trimethyl si loxy)disiloxane (MPTSDS) were purchased from Sigma Aldrich and Geles t Inc. respectively. Irgacure 2959 was purchased from Ciba Specialty Chemicals. Deionized (DI) water (>17.8 cm ) obtained from Thermo Scientific Branstead EASYpure II system was used fo r contact angle measurements, swelling tests, compression tests, and the protein adsorption assay.
63 Sample Preparation The monomers (as shown in Figure 4 1 and Table 4 1), diluent, and photo initiator were transferred to glass bottles and mixed until the photo initiator dissolved completely The prepolymerization mixtures were purged with nitrogen for approximately 2 minutes. The prepolymerization mixtures were then transferred by a pipette into the molds. These molds consisted of a poly(dimethylsiloxane) elastomer (PDMSe) gasket sandwiched between two hexamethyldisilazane treated quartz crystal plates, held together with paper clamps. The prepolymerization mixtures were then exposed to ultraviolet (UV) light ( BLAK RAY Lamp, Model UVL 56, 365 nm wavelength, 6.75 mW/cm 2 ) for 70 minutes at room temperature. To remove residual monomer and hexanol from the hydrogels, the hydrogels were soaked in swelling solvents. The amphiphilic hydrogels were submerged in 50 v/v % ethanol in water solution for 24 hours, a 25 v/v % ethanol in water solution for 24 additional hours, followed by DI water until characterization or protein ad s orption assays were performed. Figure 4 1. Monomers included in the polymerization of amphiphilic crosslinked networks
64 Table 4 1. Chemical composition of amphiphilic crossl i nked networks with varying crosslink ratio MPTSDS/(TRIS+HEMA) Molar Ratio ( mmol /mmol ) HEMA MPTSDS TRIS (mol %) (mol %) (mol %) 2.5 91.7 0.3 8.0 5 .0 91.7 0.5 7.8 10 .0 91.7 1.0 7.3 25 .0 91.7 2.4 5.9 50 0 91.7 4.8 3.5 75 .0 91.7 7.0 1.3 Equilibrium Water Content Discs (8mm) were punched out of the water equilibrated films. The crosslinked networks were weighed w d after an equilibrium dry weight was reached in a vacuum oven (28 mmHg) at room temperature The crosslinked networks were then soaked in DI water until an equilibrium swollen weight w s was reached. The weight measurements were used to calculate the equilibrium water content EWC Volume Fraction of Polymer Discs (8 mm) were punched out of the hydrogel films. The samples were soaked in water until an equilibrium water content was reached which occurred when the weight of the hydrogel remained consta nt over time The weight of the hydrogels in a swollen state was measured in water W a,s and toluene W t ,s The samples were then placed in a vacuum oven and left under vacuum (28 in Hg) until an equilibrium dry weight W a,d was reached, which occurred when t he weight of the hydrogel remained constant over time. The water equilibrated weight measurements and the volume of the polymer V p were used to calculate the volume fraction of the polymer in the swollen state
65 p t represents the density of the non swelling solvent (toluene) 87 C ompression Testing The elastic modulus of the crosslinked networks was determined by compression testing using the TA.XT.plus Texture Analyzer. Discs (8mm) were punched out from the water equilibrated gels. Mechanical testing was performed in compression a t a 1 mm/min strain rate in DI water at room temperature. The compression modulus of the crosslinked networks was c alculated from the stress deflection curve corresponding to the linear elastic region of the curve Tensile Testing The elastic modulus of t he water equilibrated crosslinked networks was determined by tensile testing using the method described in ASTM D638 and the Instron 1122 frame with MTS ReNew package and extensometer. The Inst ron frame was equipped with a 1000 lb load cell and the samples were subjected to a tensile force at a rate of 2 mm/min at room temperature The elastic modulus o f each composition was calculated from the linear elastic region from the stress Contact Angle Measurements Static c ontact angle analysis was performed using the Ram Hart Goniometer equipped with DropAdvance Software. Before any measurements were recorded, water film/drops were blown off the samples with nitrogen gas The initial drop size was set at
66 ach sample and three samples of each composition were evaluated. The Owens and Wendt two liquid geometric mean approach was used to calculate the surface energy for contact angle measurements performed using the methylene iodide water (MeI 2 water) and meth ylene iodide ethylene glycol (MeI 2 EG) liquid pairs 63 The and represent the surface free energies of the solid and liquid respectively. The following variables and represent the dispersion force components and polar force components of the solid and liquid respectively 63 The surface free energy polar and dispersion force components for water are 50.8 mNm 1 and 21.8 mNm 1 respectively. The surface energy dispersion force component and polar force component for methylene iodide (MeI 2 ) are 48.5 mNm 1 and 2.3 mNm 1 respectively. The dispersion force component and polar force component of the surface energy for ethylene glycol (EG) are 29.3 mNm 1 and 18.9 mNm 1 respectively 90 57 Contact a ngle measurements using the captive air bubble technique were obtained wi th the Ram Hart Goniometer ; whereby swollen samples were placed upside down on top of rubber stoppers in a tank filled with DI water. Air bubbles (3 l) were placed in five different spots for each composition Three samples of the same composition were evaluated The method used to determine the surface energy of the hydrogels has previously been reported by Nguyen 90 Dynamic contact angle measurements were performed using the Krss DSA 100 drop shape analysis instrument equipped with DSA3 software. The a dvancing contact
67 Fourier Transfo rm Infrared Attenuated Total Reflectance (FTIR ATR) Dried samples were evaluated with the Thermo Scientific Nicolet 6700 FTIR ATR equipped with OMNIC software. A diam ond crystal flat plate was utilized, and the scan number and scan resolution were set at 3 6 and 4 cm 1 respectively. Atomic Force Microscopy The surface morphology and the root me an square average of the vertical deviations from the data plane (R q ) of dry films were examined using the Veeco Dimen sion 3100 AFM equipped with a silicon tip opera ting under tapping mode. The scan rate was set at 1 Hz. The R q is described in the following equation was obtained using the Nanoscope Software 7.0 where is the height of the surface at point and N represents the number of values 106 The samples were dried in the vacuum oven prior to analysis in air at room temperatu re X ray Photoelectron Spectroscopy The element al composition of dry films was examined using the Pe rkin Elmer PH1 5100 ESCA System ray source at a take off angle of 8 Torr. AugerScan 3.2 was used to perform data analysis for each composition.
68 Amido Black Assay The amido black assay was performed to evaluat e the protein adsorption of the amphiphilic hydrogels, as well as the assay standards (TCP and NC ). The following assay was adapted from a previous report 92 All of the samples were co ated with 100 BSA in PBS solution. The samples were left in an incubator for 60 in solution was rinsed with DI water to remove unadsorbed protein. The samples were soaked in the amido black stain solution (methanol, DI water, Naphthol Blue Black) for 3 minutes. After rinsing with DI water, the sample s were washed with the wash soluti on (90 v/v% methanol, 8 v/v% water, and 2 v/v % acetic acid) twice to remove unbound dye. The samples were rinsed with DI water once again. This was followed by soaking the hydrogels in the eluent solution (50 v/v% ethanol, 50 v/v% 50 mM NaOH with 0.1 mM E DTA) for 30 minutes on a plate shaker (at 300 RPM). The BioTek m icroplate reader equipped with Gen5 Software was used to measure the UV absorbance at a wavelength of 620 nm (with a 450 nm reference filter) aliquot of the dye labeled protein tha t was adsorbed to the samples. Five replicates of each composition were used for the protein adsorption assay. A standard curve was made from serial dilutions of 6 mg/mL of naphthol blue dye conjugated BSA in PBS.
69 Figure 4 2. Standard curve for amido black protein adsorption assay. Statistical Analysis The data was reported as mean standard deviation. Analysis of variance (ANOVA ) was used to make inferences about sample means and m ultiple comparisons W procedure. The p value of 0.05 was used for determining statistical significance. Results As expected, the crosslink mo lar ratio influenced the modulus and the swelling behavior. The correlation between the crosslinker molar ratio and the equilibrium water content EWC and modulu s is shown in Figure 4 3 and Figure 4 4 T here hydrogel s were not statistically different in rega rds to EWC, which did not surpass 10% for any compositions. The low EWC stems from the presence of hydrophobic moieties in the bulk material inhibiting the upt ake of large quantities of water. Furthermore, the y = 0.3602x + 0.0211 R = 0.9969 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 Absorbance Protein Concentration ( g/mL)
70 diminished swelling capabilities with increasing crosslinker molar ratio most likely a M c Figure 4 3 EWC(%) of amphiph ilic hydrogels with varying MPTSDS/(TRIS+HEMA) molar ratio. Asteris k (*) denotes statistically significant EWC (%) from gel with molar ratio o f 50 (mmol/mmol). Double asterisk (**) denotes statistically significant EWC (%) from gel with molar ratio of 25 ( mmol/mmol) (p < 0.05) 0 2 4 6 8 10 12 14 16 0 10 20 30 40 50 60 70 80 EWC (%) MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol) **
71 Figure 4 4 Compression modulus (red square) and M c of amphiphilic hydrogels of varying MPTSDS/(TRIS+HEMA) mol ar ratio All moduli were statistically significant ( p < 0.05) between all gels. Figure 4 5 Stress deflection cu rve from compression testing of amphiphilic crosslinked n etworks with varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol). 0 2 4 6 8 10 12 14 16 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 10 20 30 40 50 60 70 80 Modulus (MPa) 1/M c (mol/g) MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol) -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 -0.5 -0.4 -0.3 -0.2 -0.1 0 Stress (MPa) Deflection (mm/mm) 2.5 5 10 25 50 75 Composition
72 An increase in crosslinker concentration resulted in network structures with higher modulus. This sort of behavior has been reported p reviously with siloxane based hydrogels 8 increase in crosslink junctions in the network, which results in a higher concentration of chain segments contributing to the retractive stress or deformation. In contrast to the EWC, e ach fabricated amphiphilic hydrogel had a statistically different compression modulus. This increase in modulus with increasing MPTSDS/(TRIS+HEMA) molar ratio is mirrored in the data obtained through tensile testi ng, as seen in Table 4 2 and Figure4 6 The gel with 75 mmol/mmol molar ratio is statistically different ( p < 0.05) from all gels, except for the gel with an MPTSDS/(TRIS+HEMA) molar ratio of 50 mmol/mmol. The gel with 50 mmol/mmol molar ratio of crosslink ing agent is statistically different ( p < 0.05) from gels with an MPTSDS/(TRIS+HEMA) molar ratio of 2.5, 5, and 10 mmol/mmol. Lastly, the gel with 25 mmol/mmol molar ratio is statistically different (p < 0.05) from the gel that has a molar ratio of 2.5 mmo l/mmol. As expected, the increase in stiffness of a hydrogel resulted in a lower measured elongation to break. High crosslinking agent in the feed composition lead to higher measured tensile and yield stress. Table 4 2. Mechanical properties of amphiphi lic networks from tensile tests. MPTSDS /(TRIS+HEMA) Molar Ratio (mmol/mmol) Elastic Modulus (MPa) Tensile Stress (MPa) Yield Stress (MPa) Elongation at Break (%) 2.5 0.17 0.02 0.23 0.01 0.06 0.01 153 6 5.0 0.61 0.05 0.32 0.01 0.11 0.01 107 10 10.0 1.21 0.23 0.54 0.03 0.15 0.02 85 6 25.0 3.72 1.10 0.80 0.04 0.27 0.04 52 12 50.0 5.48 0.92 0.66 0.05 0.34 0.04 20 2 75.0 8.55 3.98 0.65 0.02 0.35 0.14 8 2
73 Figure 4 6 Stress strain curve from tensile testing of amphiphilic crosslinked n etworks with varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) Contact angle measurements performed on the amphiphilic crosslinked networks indicate that a ll of these surfaces display a hydrophobic surface (contact an in air When exposed to a nonpolar environment, the hydrophobic moieties presented themselves to achieve a low interfacial surface energy. As expected, the surface properties were depe ndent upon the feed composition. High MPTSDS/(TRIS+HEMA) mola r ratio in the feed composition resulted in lower measured sessile water drop contact angles (Table 4 3 ). Furthermore, the cross l ink molar ratio and surface energy were positively correlated; indicating that the hyd rophobic moieties were present to a high extent on the surface with highly crosslinked networks 0 0.2 0.4 0.6 0.8 1 1.2 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Stress (MPa) Strain (mm/mm) 75 2.5 5 10 25 50 Composition
74 Table 4 3 Surface properties of a mphiphilic gels with varying MPTSDS/(TRIS+HEMA) molar ratio using the sessile contact angle technique MPTSDS/(TRIS+HEMA) Molar Ratio ( mmol/mmol) Sessile Contact A ngle Surface Energy (mNm 1 ) Water EG MeI 2 Water MeI 2 EG MeI 2 2.5 110 4 93 3 55 5 30 3 33 3 5 104 3 90 2 65 4 25 2 27 3 10 101 2 85 3 63 4 27 2 29 3 25 95 2 75 5 53 5 33 3 35 3 50 93 2 65 4 41 4 39 2 42 3 75 91 2 65 2 33 3 42 1 46 2 Figure 4 7 Sessile water drop and captive air droplet on crossl inked networks with an MPTSDS/(TRIS+HEMA) molar ratio of 2.5 mmol/mmol The contact angle measurements were most likely influenced by th e mobility of the crosslinked network, which is influenced by the MPTSDS/(TRIS+HEMA) molar ratio. The gels with a lower crosslinker concentration most likely consist of a network structure with higher M c st energetically favorable moieties to the surrounding milieu in a quicker fashion than those gels with a higher crosslinker concentration This behavior was observed in the contact angle measurements using the dynamic contac t angle technique as well.
75 The advancing contact angles adv networks (Table 4 4). The adv were higher than those contact angles reported for PHEMA hydrogels, which tend to exhibit adv in the range of 58 107 The high contact angles may be attributed to the surface ro ughness and c hemical heterogeneity The se surfaces exhibit behavior as those described in the Wenzel regime ; where the co ntact angle measured is i nfluenced by the chemical composition and enhanced by the surface roughness. This behavior is described in the following eq uation, where the roughness is correlated to the contact angle in the following manner. The and represent the contact angle of the smooth and rough surfaces respectively In the Wenzel state, the probe liqui d layer follows the topography of the surface. Here the enhanced hydrophobicity would be explained by the higher surface area involved in creating the solid liquid interface 108 However, problems arise due to assumptions made about the surface in this regime This regi me exists on homogeneously rough surfaces, which may not be the case for these random copolymers. The receding contact angles rec are less than 60 which is indicative of hydrophilic surfaces. The rec is low due to the wetting that occurs on the surface. The presence of hydrophilic domains is caused by the hydration and penetration of the water droplet into the surface. Pinnin g of the water droplet to this hydrated region leads to low contact angles 109 As seen in Table 4 4 the rec decreases with increasing crosslinker molar ratio. It seems as though the hydrophilic domains are constrained to the surface
76 to a higher extent with increasing crosslinker in the network, as detected by the lower adv and rec Table 4 4 Surface pro perties of amphiphilic crosslinked networks using the dynamic contact angle technique MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol ) Dynamic Contact Angle Hysteresis Advancing Receding a r | 2.5 134 6 52 7 82 8 5 124 5 44 6 80 9 10 115 6 41 4 74 10 25 109 4 33 4 77 4 50 102 4 30 4 71 6 75 96 4 32 5 64 5 The hysteresis, calculated from the difference in measurements between the advancing and receding contact angles, was influenced by the cros slinker concentrat ion as well (a s shown in Table 4 4 ). The hysteresis was inversely proportional to the MPTSDS/(TRIS+HEMA) molar ratio. This has been previously reported with rubber materials, mainly natural rubber and butadiene rubber; increased crosslink concentration res ulted in increased measured hysteresis 110 The gels presented here show no substantial changes in hysteresis between compositions The lower hysteresis of the highly crosslinked networks may be due to higher crosslink junctions created with increased MPTSDS concentration; which would then re duce the ability of chains to reorient upon hydration Deformation at the contact line due to contact with the probe liquid can cause this ins tability 110 The captive air bubble measurements show a surface that is more hydrophilic. The contact angles in an aqueous environment increase with increasing MPTSDS/(TRIS+HE MA) molar ratio (Table 4 5) This is in contrast with contact angle measurements in air. This behavior is again the result of chain mobility. The restricted
77 groups), resu lting in a surface that is more hydrophobic t han others. The surface energy, ranging from 56 66 mNm 1 obtained for all compositions is similar to what is found in the literature for PHEMA hydrogels 94 The gels with an MPTSDS/(TRIS+HEMA) molar ratio of 75 mmol/mmol were statistically significant ( p < 0.05) from all of the other gels. Furthermore, the gel with an MPTSDS/(TRIS+HEMA) molar ratio of 50 mmol/mmol is statistically significan t ( p < 0.05) from all gels, with an exception for the gel with a molar ratio of 25 mmol/mmol. Table 4 5 Captive bubble measurements for amphiphilic crosslinked networks. MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol ) Captive Bubble Contact Angle Surface Energy (mNm 1 ) Air n Octane 2.5 25 2 21 2 66 1 5 .0 25 2 26 2 66 1 10 .0 28 2 27 3 65 1 25 .0 30 2 31 3 63 1 50 .0 34 3 36 4 61 2 75 .0 41 5 44 4 56 4 Chemical composition a nalysis of the amphiphil ic networks was performed using FTIR ATR and XPS to determine which functional groups are present at the surface when the gels are in a dehydrated state The hydroxyl groups indicating the presence of HEMA, are detected at 3400 cm 1 in the ATR FTIR spectr um (Figure 4 8 ) CH stretching, CO stretching, and CH 2 bending are found in all monomers and are present at 2956 cm 1 1721cm 1 and 1402 cm 1 respectively. The characteristic peaks for Si CH 3 and Si O Si are seen at 1251 and 1053 cm 1 respectively. The chemical composition of the hydrogels was not expected to deviate substantially between compositions due to the constant siloxane concentration in all gels. As the MPTSDS
78 concentration increased, the TRIS concentration decreased and vice versa. The TRIS a cting as mono the gel by provid i ng additional crosslink junctions. Figure 4 8 FTIR ATR spectra of amphiphilic crosslinked networks with varying MPTSDS/(TRIS+HEMA) molar rati o (mmol/mmol) Table 4 6 shows that the detected elemental composition of amphiphilic cross l inked networks using XPS. The concentration of TRIS and/or MPTSDS at the surface is higher than what was predicted theoretically (as seen in Table 4 6 and Table 4 7 ). In regards to elemental composition, there seems to be no stark difference between crosslinked networks and no trends were found.
79 Table 4 6 Theoretical elemental composition of amphiphilic networks MTSDS /(TRIS+HEMA) Molar Ratio (mmol/mmol) Si (%) O (%) C (%) C/O C/Si 2.5 1.3 32.2 66.4 2.1 51.1 5 .0 1.3 32.2 66.4 2.1 51.1 10 .0 1.3 32.2 66.4 2.1 51.1 25 .0 1.3 32.3 66.4 2.1 51.1 50 .0 1.3 32.3 66.4 2.1 51.1 75 .0 1.2 32.3 66.4 2.1 55.3 PHEMA 0.0 33.3 66.7 2.0 Table 4 7 Elemental composition o f amphiphilic crosslinked networks using XPS MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol) Si (%) O (%) C (%) C/O C/Si 2.5 10.0 25.3 64.7 2.6 6.5 5 .0 9.7 25.9 64.4 2.5 6.6 1 0. 0 9.2 28.0 62.8 2.2 6.8 25 .0 11.1 27.1 61.8 2.3 5.6 50 .0 9.0 27.9 63.1 2.3 7 .0 75 .0 8.4 28.2 63.4 2.2 7.5 The morphology of the crosslinked networks was examined using the AFM. Nanotopograpy was observed for all surf aces, as seen in Figure 4 9 Furthermore, the molar ratio of MPTSDS/(TRIS+HEMA) influenced the surface morphology of the crosslinked networks The root mean square roughness (R q ) generally de creased with increasing MPTSDS, as seen in Table 4 8 was the hydrogel with the lowest crosslinker concentration. The difference in m orphology is most likely influenced by the network structure, which is heavily influenced by the crosslink molar ratio. The lower crosslinking concentration in the network structure resulted in a network with phase segregation, comprising of hydrophilic an d hydrophobic domains.
80 This may explain the varying hysteresis obtained between compositions. More likely than not, the roughness is perhaps too small (less than 100 nm) to influence the relatively high sessile contact angle and advancing contact angle in air in polymers with lower MPTSDS concentration. There was no apparent correlation between hysteresis and varying levels of nanotopography (less than 100 nm) on polyvinylpyrrolidone (PVP) and polyvinyl acetate (PVAc) grafted silicon wafers and fluoropolym er coated silicon wafer 111 112 Figure 4 9 Morphology of crosslinked networks with an MPTSDS/(TRIS+HEMA) molar ratio of (a) 2.5 (b ) 5 (c) 10 (d) 25 (e) 50 and (f) 75 mmol/mmol was examined using the Veeco Dimension 3100 AFM equipped with a Si 3 N 4 cantilever
81 Table 4 8 Roughness R q of amphiphilic crosslinked networks in a dehydrated state MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mm ol ) R q (nm) 2.5 14.6 5 .0 39.1 10 .0 26.6 25 .0 20.9 50 .0 22.9 75 .0 10.8 The results for the amido black assay show that the adsorption differed between h ydrogels, as shown in Figure 4 10 Lower protein adsorption was observed on hydrogels with hi gh crosslinking concentration; how ever, the protein adsorption was not statistically different between all compositions except that obtained on NC. The R 2 value between protein adsorption versus surface energy (captive bu bble technique) and modulus E was found to be 0.66 and 0.67 respectively. Figure 4 10 Protein adsorption on NC, TCP, and amphiphilic hydrogels of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) Asterisk ( *) denotes statistically significant difference with the TCP control (p < 0.05) 0 10 20 30 40 50 60 70 80 90 2.5 5 10 25 50 75 TCP NC Protein Coverage ( g/cm 2 ) Composition
82 The chain mobility and roughness of the surface each have been shown to play a role in protein adsorption which tends to be a function of the hysteresis 113 However, it has been reported that the roughness needs to be > 100 n m in order to affect the hysteresis of a surface. Since this is not the case with the amphiphilic crosslinked networks, the hysteresis most likely stems from the chain mobility and chemical heterogeneities. Otsuka et al. observed this phenomenon on polyethylene glycol block polylactide (PEG b PL) copolymer brushes. BSA adsorption onto PEG b PL based brushes was negatively correlated to surface hysteresis 109 I n this assay, the positive correlation between hysteresis and BSA adsorption was seen; t he R 2 value for the protein adsorption versus hysteresis was 0.72. In regards to the amido black assay, the macroporous structure may be entrapping protein molecules, resulting in higher protein adsorption or absorption. This has previously been observed with BSA by Mequanint et a l; whereby coatings with larger molecular weight between crosslinks M c (6,253 g/mol) were shown to exhibit higher BSA ads orption than coatings with smaller M c (177g/mol) 114 The relationship between the shear modulus G and the molecular weight between crosslinks M c can be seen in the following equation, where R and T represents the density of the polymer network, gas constant, and absolute temperature, respectively. According to this equation, the M c de creased almost 50 fold as the crosslinker molar ratio in creased from 2.5 mmol/mm ol to 75 mmol/mmol. The M c of these amphiphilic gels ranged from 999 47,846 g/mol. I t is possible that the least crosslinked amphiphilic gel s would result in BSA penetration as well, if protein
83 absorption was observed in hydrogels exhibiting a M c of 6,25 3 g/mol This parameter was found to exhibit a high correlation with protein adsorption (Fi gure 4 11 ) with an R 2 value of 0.93 Figure 4 1 1 The r elationship between protein adsorption and M c of amphiphilic hydrogels The mesh size of the crosslinked networks was obtained by first calculating the unperturbed end to end distance of the PHEMA chains, where and represent the molecular weight of the repeating unit of HEMA, the Flory characteristic ratio, number of chains between two crosslinks, and the carbon carbon bond length (0.154 nm). The characteristic ratio of 6.9 was used because of the R = 0.9318 0 5 10 15 20 25 0 10 20 30 40 50 60 Protein Coverage ( g/cm 2 ) M c (kg/mol)
84 similarities between PHEMA and poly(meth yl methacrylate) 87 The mesh size was calculated from the following equation (Table 4 9) Table 4 9 Network parameters of amphiphilic crosslinked networks. MPTSDS/(TR IS+HEMA) Molar Ratio (mmol/mmol) (nm) (nm) 2.5 0.75 0.04 11.0 12.1 5 0.77 0.01 5.9 6.5 10 0.74 0.01 4.3 4.7 25 0.81 0.01 2.4 2.6 50 0.86 0.01 2.0 2.1 75 0.87 0.01 1.9 2.0 The radius of BSA has been re ported to have a hydrodynamic radius of 3.4 nm 115 It seems this protein may be more easily absorbed by gels exhibiting a n MPTSDS/(TRIS+HEMA) molar ratio of 10 mmol/mmol or less. Conclusion Amphiphilic hydrogels composed of 2 hydroxyethyl methacrylate (HEMA), 1,3 bis(3 methacryloxypropyl)tetrakis(trimethylsiloxy)disiloxane (MPTSDS ), and tris(trimethylsiloxy) 3 methacryloxypropylsilane (TRIS) were copolymerized using ultraviolet (UV) light and Irgacure 2959 as the photo initiator Swelling tests reveal that all of these hydrogels exhibit low water content (< 10%) that is dependent u pon the crosslinker (MPTSDS) concentration. Hydrogels prepared with varying concentration o f di functional monomer (MPTSDS) exhibit ed tunable mechanical properties, with a compression modulus ranging from 0.4 to 4.7 MPa. Not only were the bulk properties influenced by the crosslinker concentration in the feed composition, but the surface
85 properties were affected as well. AFM images show nanotopography that is dependent upon the crosslinker c oncentration in the dehydrated state. Static contact angle measure ments indicate that the networks displayed a hydrophobic surface in air. The hydrophobic moieties, such as the siloxane groups or alkyl backbone chain, were exposed to air to a greater extent in the networks with low crosslink molar ratio than the hydrogel s with higher crosslinking concentration due to the lower restriction in mobility of the surface. Furthermore, reconstruction at the surface was evident from dynamic contact angle measurements. The advancing and receding c ontact angle measurements indicate that the hydrophilic moieties that are initially buried in the bulk phase in the gas environment present themselves at the surface when de wetting occurs. The hysteresis was large for all gels and is inversely proportional to the crosslinking concentratio n. The protein adsorption assay indicates that the highly crosslinked networks exhibit better anti fouling properties than the networks with a lower crosslink molar ratio H owever, differences in protein adsorption between gels were not statistically sign ificant. Network hysteresis and M c seems to play a large role in adsorption of BSA. The network pro perties of the gels with lower crosslinking agent concentration may facilitat e absorption of proteins due to the mesh size that is larger in size than the pr otein.
86 CHAPTER 5 ANTIFOULING PROPERTIES OF AMPHIPHILIC CROSSLINKED NETWORKS Introduction Fouling of macromolecules and living organisms occurs frequently in the marine environment, which can be problematic for navy vessels. In order to combat issues such as this, coatings that serve as antifouling (AF) and foul release (FR) materials have been studied extensively. Green macroalga, Ulva linza a re common foulers of natural and man made surfaces. These organisms that are present in the upper intertidal zon e of the ocean, are readily found on the hull of ships and for this reason are used as a model system to study the foul resistant properties of marine coatings 100 Settleme nt and recruitment of spores or larvae is a vit al part of the algae life cycle; without success at this stage survival is minimal 13 It is this stage that was targeted to control biofouling of Ulva As stated previously, low modulus and high wettability of hydrogels provides these cross linked networks with AF properties towards mari ne organisms 47 116 Hydrogel s contain hydrophilic mo ieties that participate in hydrogen bonding with water molecules. The water molecules closely associated with the polymer chain plays a large role in protein resistance 99 The lack of presence of a conditioning layer composed of biomolecules may be the reason for inhibition of permanent settlement of marine organisms. However, there are some marine organisms that adhere rather strongly to hydroph ilic surfaces, i.e.: Balanus Amphitrite 66 To inhibit settlement of these types of organisms, the surface s should exhibit FR properties as well. An approach that has been used to produc e coating s with FR properties involves creating a chemically ambiguous surface. This can be done by cre ating an amphiphilic
87 coating, which contains both hydrophilic (> 50 mNm 1 ) and hydrophobic (< 25 mNm 1 ) domains This is believed to create uninhabitable surfaces for settlement of different types of marine organisms that adhere strongly to either hydrop hilic or hydrophobic surfaces, i.e.: Ulva spores and diatoms, respectively 9 A majority of the amphiphilic coatings tested against Ulva zoospores contain a combination of hydrophilic poly(ethylene glycol) PEG and hydrophobic fluoropolymers (FP). It has been reported that settlement and percent removal of Ulva spores are dependent upon chain length and concentration of PEG in these amphiphilic brushes. Low spore settlement tends to be found on high molecular weight brushes with the high PEG content. Unfortunately, these surfaces may require high critical pressure due to the strong intermolecular interactions between the settlin g organism and the surface. Park et al. found t his to be the case with amphi philic brushes containing PEG chains exhi biting a molecular weight of 20 kg/mol In contrast, t he brushes with a lowest molecular weight (10 kg/mol) of PEG required lower critical pressure (< 21 kPa) to remove spores compared to the PDMSe control but this was only the case for the coating with t he highest PEG content 53 An amphiphilic brush coating consisting of polyoxyethylene polytetrafluoroethylene (PEO PTFE) side groups exhibited similar Ulva spore settlement compared to the poly(styrene b (ethylene co butylene) b polystyrene) (SEBS) control. The se coatings were found to have FR properties; p ercent removal of spores was higher for PEO PTFE coatings (50 80 %) compared to the SEBS coating (30%) and PDMSe control (30%) 9 Coatings composed of dendritic diisocyanate tethered with the am phiphilic functional moiety, PEO and fluoroalkyl chain s exhibited
88 lowest density of spores and highest percent removal compared to PDMSe and glass samples 54 Other marine organisms, i.e. diatoms and barnacles, seem to be inhibited by amp hiphilic brushes as well 55 Poly( poly(ethylene g lycol)methyl ether methacrylate) block poly(2,3, 4,5,6 pentafluorosty rene) (P(PEGMA) b PPFS) brushes were reported to inhibit barnacle settlement in laboratory settlement assays and field test s by panel immersion in the marine environment 56 T he percent removal of diatom Navicula exposed to the PEO PTF E brushes was lower (ranging from 60 to 85 %) than the SEBS control (which was ~90%). Diatoms have been reported to adhere more strongly to hydrophobic surfaces. This wa s perhaps why the they were hard to remove from th e amphiphilic surfaces, which was mor e hydrophobic than the SEBS control 9 C rosslinked networks have also been successful at inhibiting settlement of marine organisms. A PEG and perfluoropolyether crosslinked network had a 29% reduction of diatom Nitzschia density compared to the PDMSe control 58 Moreover, s pore settlement was lowest on polyethylene glycol hyperbranched fluoropolymers crosslinked networks (HBFP PEG) surfaces with the highest PEG content (45 and 55 wt %), exhibiting an 83% reduction of Ulva spore settlement compared to the glass standard. The HBFP PEG45 hydrogel with 45 wt% PEG had the highest percent removal of spores, reaching 40 %. Percent removal for 8 day old Ulva was 90% for HFPE PEG45 surface, which was twice as much as that for the PDMSe surface 57 Percent removal of diatom Navicula and Ulva spores was higher with acrylic polyol grafted PDMS polyisocyanate crosslinked coatings containing a high concentration of
89 hydroxyethyl acrylate (HEA) (and to a lesser extent butyl acrylate (BA)). The highest percent removal of Ulva spores observed was 100% 59 The m odulus of a substrate has also been shown to play a role in attachment strength of marine organisms. This phenomenon was observed by Berglin et al. hydrophobic materials were studied fo r their fouling rele ase abilities towards barnacles. They found that the adhesion was weaker for the barnacles attached to silicone (possessing a = 22 mNm 1 ), compared to the barnacles attached to Teflon According to the work of adhesion W sl equation, ( where and represents the surface energy of the solid the surface energy of the liquid, and the interfacial tensi on between the solid and liquid respectively ) the surface with the lowest surface energy (Teflon = 17 mNm 1 ) w ill generate a weaker adhesion 14 However this was not observed. The low modulus of the silicone coa tings was reported to be the reason for the lower adhesive strength between the silicone and the barnacles Berglin et al. determined that the detachment stress of b arnacles is proportional to (E ) 1/2 67 The correlation betwee n the relative adhesion and (E ) 1/2 was found to be greater (R 2 = 0.89) than the correlation between to the relative adhesion versus the surface energy or elastic modulus alone, R 2 = 0.75 and 0.82, respectively 68 A mphiph ilic crosslinked gels were fabricate d to investigate its AF and FR capab ilities towards marine organism, Ulva linza These amphiphilic hydrogels were tested against Ulva to evaluate the effect of modulus, which is influenced by the network structure parame ter M c on settlement density and strength of adhesion. The hydrogels were composed of 2 hydroxyethyl methacrylate (HEMA), 1,3 bis(3 methacryloxy
90 propyl)tetrakis(trimethylsiloxy)disiloxane (MPTSDS), and tris(trimethylsiloxy) 3 methacryloxy propylsilane (TR IS). These three monomers were copolymerized using molar ratios of MPTSDS to the sum of TRIS plus HEMA ranging from 2.5 75 mmol/mmol The HEMA concentration was kept constant at 91.7 mol % These gels exhibit tunable mechanical properties due to the diff erences in the feed composition which influence s the M c Experimental Section Materials Allyltrimethoxysilane 2 hydroxyethyl methacrylate (HEMA), and tris(trimethylsiloxy) 3 methacryloxypropylsilane (TRIS) were purchased from Sigma Aldrich and used as received The crosslinker, 1,3 bis(3 methacryloxypropyl)tetrakis (trimethyl siloxy)disiloxane (MPTSDS), was purchased from Gelest, Inc. Sodium metabisulfite (SMBS) and ammonium persulfate (APS) were purchased from Fisher Scientific and Sigma Aldrich, respe ctively. Irgacure 2959 was purchased from Ciba Specialty Chemicals. N octane and 1 hexanol were obtained from Fisher Scientific. cm) was obtained from the Thermo Scientific Branstead EASYpure II system Silastic T2 resin and curi ng agent was purchased from Dow Corning. Preparation of Amphiphilic Hydrogels The monomers, diluent, and photo initiator (Table 5 1) were transferred to glass bottles and mixed until the photo initiator dissolved. The prepolymerization mixtures were purged with nitrogen for approximately 2 minutes. Methacryloxypropyltriethoxy silane treated glass slides were prepared and included in a mold consisting of a PDMSe gasket sandwiched between two hexamethyldisilazane treated quartz crystal plat es
91 (held together w ith clamps). The prepolymerization mixtures were then transferred by a pipette into the molds. The prepolymerization mixtures were then exposed to ultraviolet (UV) light ( BLAK RAY Lamp, Model UVL 56, 365 nm 6.75 mW/cm 2 ) with the mold in a vertical po sition for approximately 70 minutes at room temperature. To remove residual monomer and hexanol from the hydrogels, the hydrogels were soaked in swelling solvents. The amphiphilic hydrogels were submerged in 50 v/v % ethanol in water solution for 24 hours, a 25 v/v % ethanol in water solution for 24 additional hours, followed by DI water until assays were performed. Table 5 1. Feed composition of amphiphilic crosslinked networks MPTSDS/(TRIS+HEMA) Molar Ratio ( mmol/mmol ) HEMA MPTSDS TRIS (mol %) (mol % ) (mol %) 2.5 91.68 0.25 8.07 5 91.68 0.50 7.82 10 91.68 0.99 7.33 25 91.68 2.44 5.88 50 91.68 4.77 3.55 75 91.68 6.98 1.34 Synthesis of PDMS samples Silastic T2 resin and curing agent were weigh ed out at a 10:1 weight ratio and mixed for 5 minutes The mixture was then placed in desiccator under vacuum for 20 minutes to degas. Glass slides were treated with a llyltrimethoxysilane (APS) to enable adhesion of the siloxane mixture to the glass slide The PDMS was poured over the treated glass slides an d pressure was applied to the mold consisting of two glass slides and 2 mm spacers. The siloxane mixture was allowed to cure for 24h at room temperature.
92 Ulva Spore Assay Three replicates of each composition were soaked in DI water prior to being equilibr ated in artificial seawater (ASW) for 2 hours. Three replicates of each composition were tested in the Ulva spore assay. Ulva linza was collected from mature plants in the Llantwit Major. Preparation of Ulva linza is described elsewhere 69 The samples were soaked in a 1 x 10 6 spores/mL suspension for 45 minutes in the dark at 20 C Unattached spores were washed away by passing 10 times through a beaker filled with seawater Three replicates from each compos ition wer e fixed with 2.5% gluta raldehyde in seawater solution. Thirty counts were taken from three of the replicates (n = 90). A Zeiss Kontron 3000 imaging system equipped with a Zeiss epi fluorescence microscope with a 10 x objective was used to count the attache d spores on the samples. Ulva Sporeling Assay The zoospores of Ulva linza plants were the source of the sporelings and were prepared as described previously 69 These zoospores were suspended in seawater at a concen tration of 0.1 (OD 66nm) Six replicates of each composition we re soaked in the zoospore suspension for 45 minutes followed by a gentle washing to remove unat tached zoospores. The samples were transferred t o a dish with enriched seawater and incubated at 18 C with 16:8 light : dark cycle, while media was replaced every 2 days Cell density was measured at day 8 via chlorophyll a extraction. Before chlorophyll a extraction and spectrophotometric evaluation the s lides were subject ed to a wall shear stress o f 5 2 Pa for 5 minutes A water jet apparatus operating at a water pressure of 100 kPa was used to assess t he sporeling biomass removal on amphiphilic coatings (3 replicates) and the PDMSe control (3 replicates).
93 Statistical Analysis The data was reported in mean standard deviation. Analysis of variance (ANOVA as make multiple comparisons between sample means. A p value of 0.05 was used to determine statistical significance Result s Ulva spore attachment and sporeling biomass assays were performed to assess the antifouling and fouling release properties of amphiphilic crosslinked gels with varying crosslinking agent in the feed composition. The assay to evaluate spore density on the various test surfaces show that all of the gels perform better than the PDMS control as seen in Figure 5 1 The lowest settlement observed was with the composition that has a n MPTSDS/(TRIS+HEMA) molar ratio of 50 mmol/mmol exhibiting an 80 % reduction of spore settlement compared to the PDMSe control This gel has excellent antifouling abilities due most likely to the high surface energy of 61 mNm 1 However, the surface ener gy between gels with an MPTSDS/(TRIS+HEMA) molar ratio of 50 and 25 mmol/mmol we re not sta tistically different ( p > 0.05) and both received noticeably different density of spore settlement. The surface energy between gels was shown to be statistically different between some compositions, but the di fferences were not expected to influe nce a biological response. Another substrate parameter(s) must account for the difference in settlement density of spores between these gels.
94 F igure 5 1. Settlement density of spores on amphiphilic crosslinked networks with varying MPTSDS/(TRIS+HEM A) molar ratio (mmol/mmol) and the PDMSe control Asterisk (*) denotes statistical s ignificance from PDMSe control (p < 0.05) A general trend was observed; the spore density decrease d with increasing MPTSDS/(TRIS+HEMA) molar ratio However, the least cros slinked gel did not follow this trend. The influence of substrate modulus on marine settlement has been documented previously. Howev er, these results contradict what has been described in the literature. For example, the settlement density of bryozoan, Bug ula neritina ( B. neritina ), has previously been shown to be positively correlated with substrate modulus of dimethyl silicones (possessing an elastic modulus ranging from 0.01 0.1 MPa). Only the surface with the lowest moduli was statistically different. T he same behavior was seen with Zoothamnium sp., Astylozoon sp., Folliculina sp. in field experiments at the Rhode River estuary 117 A similar tr end was observe d with sporeling biomass after 7 days of se ttlement onto test samples (Figure 5 2 and Figure 5 3 ) further substantiating the evidence found 0 500 1000 1500 PDMSe 2.5 5 10 25 50 75 Spore Density (no./mm 2 ) Composition *
95 with the Ulva spore settlement onto the test surfaces These young plants were found to preferentially adhere to PD MSe and the hydrogel with the lowest crosslinking agent in the f eed composition (2.5 mmol/mmol). In this experiment, the coating with an MPTSDS/(TRIS+HEMA) molar ratio of 75 mmol/mmol exhibited the lowest biomass of sporelings ; with a percent reduction com pared to the PDMSe control of 70. Again, the surface energy of the substrate is not the only parameter influencing the settlement behavior on these subs trates. The compositions with a n MPTSDS/(TRIS+ HEMA) molar ratio of 2.5, 5, 10, and 25 mmol/mmol did not exhibit statistically different surface energy (p > 0.05) F igure 5 2. Biomass of spore ling biomas s after 7 days of growth on amphi philic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) and PDMSe. The RFU represents relat ive fluorescence units. Asterisk (*) denotes statistical sig nificance from the gel with an MPTSDS/(TRIS+HEMA) molar ratio of 75 mmol/mmol (p < 0.05) 0.0E+00 1.0E+04 2.0E+04 3.0E+04 PDMSe 2.5 5 10 25 50 75 Biomass (RFU) Composition
96 F igure 5 3. Images of sporeling biomas s on amphiphilic crosslinked networks with varying MPTSDS/(TRIS+H EMA) molar ratio and PDMSe after 7 days of growth. Left to Right: PDMSe, 2.5, 5, 10, 25, 50, and 75 mmol/mmol. Ulva spore adhesion has been shown to be influenced by the mechanical properties of crosslinked networks. The percent removal of spores and spor eling biomass not settlement density has previously been shown to be influenced by substrate modulus in the range of 0.2 to 9 .4 MPa. Chaudhury et al. have reported that the percent removal of spores was negatively correlated with the elastic modulus of t he PDMS crosslinked networks It was the softest modulus that had the only statistically significant ( p < 0.05) percent removal of these marine organisms on the PDMS substrates 69 This behavior was also observed wit h the sporeling biomass on the PDMS substrates off force P c required for removal of a rigid disk from a thin elastomer film (t << a) attached to a rigid material. When the elastomeric adhesive film is thick (t >> a) the P c is described as the following,
97 where t, a, K, and G c represent the film thickness, contact radius, bulk modulus o f the film, and the critical fracture energy, respectively. In the instance the equation modified further 68 According to this equation the pull off force P c is related to the substrate modulus E, as well as the surface energy (which is equivalent to 1/ 2 G c ). equation was tested by Berglin et al. using pseudo barnacles and poly(butylmethacrylate co allylmethacrylate) crosslinked networks as the test surface. The attachment strength was positively correlated with stora ge m odulus as was the ) 1/2 It was postulated that t he higher chain mobility of the crosslinked network could enable interfacial slip at lower modulus, resulting in lower adhesion strength between the substrate and pseudobarnacles 67 However, the amphiphil ic samples show b ehavior that contradicts what was de scribed in the Chaudhury and Berglin paper and was is predicted by the Kendall equation. Substrates with higher modulus exhibited low er percent removal of sporeling biomass Furthermore t he spores were strongly adhered to the amphiphilic gels compared to the PDMSe samples. So much so, that the wall stress had to be incre ased from 52 kPa to 100 kPa to remove sporelings from the amphiphilic gels. These gels ability to strongly attach to these organisms may be due to the hydrophilicity of these samples. Ulva spores have been reported to adhere strong ly to hydrophilic samples 100 This was seen with self assembled monolayers with varying wettability. The wettability of CH 3 /OH terminated SAMs was found to influence percent removal of spores; whereby, the percent removal of spores increased with increasing hydrophobicity of the mixe d CH 3 /OH terminated self assembled monolayers ( SAMs )
98 If surface energy did play a role in the strength of attachment of the sporelings to the substrates, it was seen in the detachment strength assay (Figure 5 4). Captive bubble contact angle measurements show that the amphiphilic gels possess greater hydrophobicity as the crosslinking concentration in the feed composition is increased, it is not surprising then that the percent removal is higher for those highly crosslinked networks. However, the correlat ion between surface energy is low (R 2 value of 0.31). F igure 5 4 Percent re moval of sporeling biomas s on amphi philic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) and PDMSe. Asterisk (*) denotes statistical significanc e to PDMSe control (p < 0.05) The correlation R 2 between percent removal and the term (E ) 1/2 was found to be 0.57 which was higher than that for just modulus alone, which was 0.39. As stated previously, the network structure may have played a role in th e fouling abilities of these amphiphilic gels. The M c of the gels was found to decrease almost 50 fold, when the crosslinking agent molar ratio increased from 2.5 mmol/mmol to 75 mmol/mmol. The 0 20 40 60 80 100 PDMSe 2.5 5 10 25 50 75 % Removal Composition *
99 correlation R 2 between the M c and percent removal of sporeling biomas s and sporeling biomass was found to be 0.83 and 0.98, respectively (Figure 5 5). F igure 5 5 Relationship between percent removal of sporeling biomas s and M c (top). Correlation between percent removal of sporeling biomas s and substrate modulus ( bottom) R = 0.83 0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 % Removal M c (kg/mol) R = 0.3931 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 6 7 8 9 % Removal Modulus (MPa)
100 F igure 5 6. Relationship between percent removal of sporeling biomass s and M c (top) of crosslinked networks. Relationship between sporeling biomass and substrate modulus ( bottom ) of crosslinked networks The green algae adhere to a surface t hrough an adhesive composed of hydroxyproline rich proteins, that is secreted as a liquid. It has been stated that the adhesive behaves similarly to a hydrogel, swelling when exposed to water The green algae are secured to the surface when the adhesive p ad has cured 73 The substrate R = 0.9766 0.0E+00 5.0E+03 1.0E+04 1.5E+04 2.0E+04 2.5E+04 3.0E+04 0 10 20 30 40 50 60 Biomass (RFU) M c (kg/mol) R = 0.3383 0.0E+00 5.0E+03 1.0E+04 1.5E+04 2.0E+04 2.5E+04 3.0E+04 0 1 2 3 4 5 6 7 8 9 Biomass (RFU) Modulus (MPa)
101 network structure may have influenced the adhesion process for t he spores and sporeling biomass growth The hydrogels with a higher M c may have facilitated the p ene tration o f the sporeling adhesive into the substrate This may be the reason why the sporeling biomas s was greater on hydrogels with low MPTSDS/(TRIS+HEMA) molar ratio This may also explain why the sporeling biomass was easier to remove from the surfaces with higher crosslink density. Issues with these coatings during the assay arose due to delamination of coatings from glass slides. For this reason, the assay to test the percent removal of spores assay could not be completed. The chemical and mechanical i ntegrity of the samples were not affected by sea water exposure. These contact angles were measured using the captive air bubble technique were shown to be similar to contact angles obtained from water equilibrated samples. The contact angle measurements a re shown in in Table 5 2. Table 5 2 Contact angle measurements of amphiphilic crosslinked netwo rks equilibrated in sea water for 48h MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol ) Captive Air Bubble Contact Angle ( ) 2.5 23 2 5 24 2 10 25 2 25 27 2 50 28 3 75 31 6 Conclusions The assays indicate that the differences between the amphiphilic gels with varying crosslink density influenced the zoospore settlement density, as well as sporeling mass and attachment strength. The spore density of the common marine fouler Ulva linza increased as the elastic modulus o f the amphiphilic gels decreased This contradicts
102 data previously described, wh ereby, the spore density was lower on softer surfaces. This behavior was also observed with the sporeli ng biomas s growth Fu r thermore, t he attachment strength of s poreling biomas s does not follow the Kendall equation. The spore attachment increased with decreasing substrate elastic modulus. It seems that the M c plays a higher role in marine organism settlem ent than the modulus. Gels with a higher M c possessed higher Ulva settlement and attachment. Interestingly enough, the fouling behavior of spores and growth of sporelings mirrored that of the protein, bovine serum albumin. Similar to the BSA the differe nce in network structure most notably M c and mesh size, between gels may account for differences in antifouling abilities of the gels towards sporeling biomas s growth Reasons for this may stem from the ease at which biomolecules can diffuse into the netw ork structure due to increases in mesh size with lower crosslink density. These biomolecules may originate from the surrounding milieu or may be egested from the spores themselves. Protein adsorption has been reported to play a role in enhancing fouling of marine organisms.
103 CHAPTER 6 PEPTIDE CONJUGATED AMPHIPHILIC CROSSLINKED NETWORKS WITH TUNABLE MECHANICAL PROPERTIES Introduction Organogenesis can be emulated with the aid of biomaterials for the treatment of atherosclerosis, a disease of the vessel whe re the thickening and stiffening of the art erial wall adversely affects blood circulation 19 In order create an effective tissue engineered vascular graft (TEVG) the scaffold must direct the cells to regenerate tissue. In the case of TEVGs, it is necessary to induce physiologically functional endothelial and smooth muscle cells (SMCs). To accomplish this, the scaffold must exhibit relevant mechanical and chemical properties. The human aorta has been reported to exhibit a modulu s in the range of 0.3 1 MPa and the human saphenous vein has been reported to exhibit a modulus in the range of 0.03 2.51 MPa in the circumferential direction and 0.16 0.47 in the longitudinal direction 118 119 Synthetic materials, such as hydrogels, have a wide range of applications due to their tunable properties. These properties are dependen t upon feed composition, processing method, and polymer volume fraction. These crosslinked gels p ossess high permeability and low stiffness. Since these synthetic materials tend to resist non specific adsorption of bioactive molecules (i.e.: proteins or pe ptides), cell adhesion and spreading will not be possible without suitable modification in the form of physi adsorption or covalent bonding of bioactive molecules to the crosslinked network prior to cell seeding 120 Extrac ellul ar matrix (ECM) products (such as collagen, fibronectin, la minin, fibrinogen, and matrigel) are commonly used to adhere mammalian cells to synthetic polymers 29 The ECM is an important component of tissue because it provides a 3
104 dimensional structure of bioactive molecules. Cells adhere to a these ECM components via integrins. Integrins are involved in mediating adhesion between the ECM and the ce ll cytoskeleton. They act as signaling receptors (outside in signaling), relaying information about a substrate to the inside of the cell; thus, mediating cell adhesion, migration, and assemblage of the ECM. These processes are what make integrins essentia l for tissue regeneration 38 The presence and concentration of proteins or peptides physi adsorbed or covalently bonded to the crosslinked structures has been shown to influence cell adhesion, proliferation, and morphology. This phenomena may explain why t he type and amount of bioactive molecules of the ECM vary between the different types of tissue 121 During the developmental stages of a blood vessel formation the SMCs exhibit a n a migratory, highly proliferative state, and exhibit a fibroblast like morphology 29 When the contractile phenotype is expressed, SMCs exhibit a b i polar spindle shape 30 31 SMCs have the ability to contract and balance ECM production by synthesis and degradation of the matrix components when expressing the contractile phenotype. In order to obtain this phenotype in a vascular graft, the SMCs will need to be indu ced by environmental cues: chemical topograph ical and mechani cal Proteins have previously been grafted (via plasma, IR/gamma radiation, or chemical modification) or physi adsorbed to hydrogels with improved results for cell adhesion, morphology, and phenotype 74 75 Fibronectin alone was shown to not affect proli feration or induce the contractile phenotype; however, it has been shown to promote cell adhesion 40 32 39 Laminin has been shown to promote the contractile
1 05 phenotype by resisting cell proliferation in the presence of platelet derived growth factor (PDGF) a nd inducing high volume density of myofilaments (V v myo ) 40 32 The advantages of using peptide sequences as opposed to protein molecules include lower prevalence of conformational changes, improved control over ligand density, lower prevalence of immune respons e, and stability during sterilization 77 Surface coverage was found to also be greater for fibronectin adhesion peptide (FAP) tet hered poly(HEMA co 2 ) compared to laminin/ fibronectin, tethered poly(HEMA co 2 respectively) 78 The Arg Gly Asp (RGD) conjugated coatings have been previously shown to influence smooth muscle cell behavior. Higher adhesion of human corona ry artery SMCs was observed on PEG hydrogels comprised of RGD conjugated PEG (M w 3400 g/mol ) compared to laminin and fibronectin coated cover slips. There was decreased expression of contractile markers actin, calponin, and SM laminin coated glass cover slips compared to those on peptide incorporated hydrogels 42 This behavior was not expected, since laminin has been shown to enhance contractile properties of smooth muscle cells (SMC). Culturing cells on high modulus cover slips (E glass Pa) may have directed cells towards synthetic phenotype 40 Contraction of the SMCs was not observed on the RGD conjugated PEG hy drogels in response to exposure to carbachol, which was used to examine ligand induced contraction 42 Mann et al. found that SMCs adhere at a higher rate to glass slides coated with KQAGDV, compared to slides coated with RGDS, VAPG, RGES (a non adhesive control peptide), or the control glass slide. Cell migration for all peptide
106 conjugated glass slides decreased, except RGES, as the protein concentration increased from 0.2 nmol/cm 2 to 2.0 nmol/cm 2 Presence of functional peptides resulted in lower proliferation of SMCs; in addition, proliferation of SMCs was found to be d ependent upon peptide concentration 10 The mechanical properties have time and time again been shown to influence cell behavior. It is well known that cells can sense and respond to physical forces 43 Mechanical signals can be sent across the cell membrane via integrins. The amount of deformation at the adhesion site can c ontrol cell behavior 36 Amphiphilic gels were used as a substrate to grow SMCs. The effect of modulus on SMC behavior was studied on these gels that exhibit ed tunable mechanical propert ies by varying the di functional monomer concentration (MPTSDS). Amphiphilic crosslink ed networks that were fabricated were functionalized with Arg Gly A sp Ser (RGDS). Peptide conjugation to the hydrogels was performed by the reaction of the epoxy group o f glycidyl methacrylate to the peptide RGDS. Epoxides can react with nucleophilic groups (i.e., amino, hydroxyl, or thiol functional groups) on proteins and peptides, forming secondary amino bonds, ether groups, and thioether bonds 122 The effect of peptide concentration and substrate modulus on SMC behavior was studied by examining attachment and proliferation Experimental Section Materials Hexanol and 2 hydroxyethyl methacrylate (HEMA) were used as purchased from Fisher Scientific and Sigma Aldrich, resp ectively. 1,3 bis(3 methacryloxypropyl)tetrakis (trimeth ylsiloxy)disiloxane (MPTSDS) was purchased as is from Gelest, Inc. Glycidyl methacrylate and tris(trimethylsiloxy) 3 methacryloxypropyl silane (TRIS) were
107 purchased from and Sigma Aldrich Irgacure 29 59 was purchased from Ciba Specialty Chemicals. Deionized (DI) water (>17.8 cm ) obtained from Thermo Scientific Branstead EASYpure II system was used for the swelling and compression tests Quant iT PicoGreen dsDNA reagent, Lambda DNA standard, and 20 X TE buffer were obtained from Invitrogen. Triton X 100 was obtained from Sigma Aldrich. Synthesis of Amphiphilic Crosslinked Networks The proper amounts of monomer, initiator (0.1 mol % prepolymerization mixture), and diluent (29 mol % prepolymerization mixture) were transferred to glass bottles (Table 6 1) The prepolymerization mixtures were stirred on a stir plate using a stir bar. The prepolymerization mixtures were purged with nitrogen for approximately 2 minutes. The prepolymerization mixtures were be transferred by a pipette into molds, consisting of a PDMSe gasket sandwiched between two hexamethyldisilizane treated quartz crystal plates (held together with clamps). The prepolymerization mixtures were exposed to ultraviolet (UV) light ( BLAK RAY Lamp Model UVL 56, 365 nm 6.75 mW/cm 2 ) with the mold in a vertical position for approximately 70 minutes at room temperature.
108 Figure 6 1. Synthesis of amphiphilic crosslinked networks Table 6 1. Chemical composition of amphiphilic crosslinked netw orks MPTSDS/(TRIS+HEMA) Molar Ratio ( mmol/mmol ) Composition (m ol %) HEMA GMA MPTSDS TRIS 2.5 77.5 14.1 0.25 8.07 10 .0 77.5 14.1 0.99 7.32 75 .0 77.5 14.1 6.98 1.34 RGDS Functionalization of Crosslinked Networks Th e samples were exposed to a 400 L aliquot of a 0.05 mM, 0.5mM, and 5.0 mM RGDS/ 0.1 M Na 2 CO 3 solution for 24 hours To remove residual monomer and hexanol from the hydrogel s the hydrogel s were soaked in swelling solvents. The amphiphilic hydrogels were submerged in 50 v/v % ethanol in wate r solution for 24 hours, a 25 v/v % ethanol in water solution for 24 additional hours, followed by deionized (DI) water until characterization.
109 F igure 6 2. Synthesis of RGDS conjugated amphiphilic crosslinked networks Fourier Transform Infrared Spectro scopy Attenuated Total Reflectance (FTIR ATR) Dried samples were loaded into the ThermoScientific Nicolet ATR FTIR Spec trometer equipped with a germanium crystal. A spectrum was made from a scan number and scan resolution set at 32 and 4 cm 1 respective ly. The characteristic peak at 1645 cm 1 correspond ed to the amide bond of the peptide, RGDS. Fluorescence Microscopy The peptide conjugated crosslinked networks were label ed with fluorescein isothiocyanate (FITC) to determine the relative peptide concen tration. FITC solution were prepared from 10 0 m M carbonate solution (pH = 9.5) Each sample was coated with 500 L of FITC solution, to allow for a nucleophilic reac tion between the RGDS and FITC The sampl es were allowed to react for 24 h at room temperatu re The samples were washed 3 times and soaked overnight in DI water before exposure to a light source exhibiting an excitation wavelength of 494 nm and recording the intensity of the light at an emi ssion wavelength of 520 nm. Three spots from each gel wer e imaged. Image J was utilized to measure the fluo rescence intensity of the images taken of the hydrogels. Na 2 CO 3 pH = 9.5
110 Equilibrium Water Content Discs (8mm) were punch ed out of the peptide conjugated films and weighed w s The crosslinked networks were weighed w d afte r an equilibrium dry weight was obtained in a vacuum oven (28 in Hg) at room temperature The weight measurements were used to calculate the equilibrium water content EWC Compression Testing The elasti c moduli of th e peptide conjugated crosslinked networks were determined by compression testing using the TA.XT.plus Texture Analyzer. Discs (8mm) were punched out from swollen gels. Mechanical testing was performed in compression at a 1 mm/min strain rate in DI water a t room temperature. The compression modulus of the crosslinked networks was calculated from the linear region of the stress deflection curve. Cell Culture Human SMCs were cultured in smooth muscle cell media (SMCM) supplemented with 10% fetal bovine serum (FBS), 2 mM g lutamine, 500 units of penicill in and 100 2 until used for the cell experiments. The media was replaced every 2 3 days Cells were used between passage 6 and 8 42 123 Cell Proliferation The SMCs were seeded onto the sterili zed substra tes at a density of 2 .5 x 10 4 cells/mL in SMCM. Aliquots (1 mL ) of the cell suspension were transferred to each
111 were removed when the media was replaced. The cells were allowed to grow on the substrates for 1 3, and 7 days. A 1 v/v % Triton X 100 solution was used to lyse the cells. The cells were placed in the incubator for 10 minutes at 37 C at 5% CO 2 A 70 L aliquot of TE buffer was added to 30 L of the cell l ysate; 100 L PicoGreen solution was added to the cell lysate/TE solution as well as the dsDNA standards A dsDNA derived standard curve was created to determine the DNA concentration for each sample. The 96 well plate was placed in a microplate reader, which was used to excite the samples at 480 nm and record the emission wavelength at 520 nm. Statistical Analysis The data was reported in mean standard deviation. Analysis of variance used to make inferences about sample mea ns and to ma ke multiple comparisons between sample means A p value of 0.05 was used to determine statistical significance. Results The FTIR ATR spectra indicate that the siloxane, HEMA, and GMA monomers are incorporated into the network (Figure 6 3) It has been reported that the CH bending of the epox ide ring, symmetric ring stretching, and asymmetric stretching of the ring are observed at 758, 905, and 1244 cm 1 respectively 124 125 These pe aks are present in the spectra, which mean that the epoxides are available near the surface for reaction with the peptide, RGDS
112 F igure 6 3 FTIR ATR spectra of amphiphilic crosslinked networks with an MPTSDS/(TRIS+HEMA) molar ratio of 2.5, 10, and 75 mmol/mmol To make the amphiphilic hydrogels more suitable for cell growth, RGDS was conjugated to the surface. The FTIR ATR spectra of the non funct ionalized and functionalize d networks are shown in Figure 6 4 The RGDS peptide, which should exhibit an amide peak at 1645 cm 1 and 1590 cm 1 appears to be very small or absent according to the spectra. The amide peak at the surface may not have been det ected due depth of penetration, which was expected to be approximately 2 3 m. 2.5 10 75
113 Figure 6 4. FTIR ATR Spectra of RGDS conjugated amphiphilic crosslinked networks with an MPTSDS/(TRIS+HEMA) molar ratio of (top) 2.5 and (bottom) 75 mmol/mmol F urther analysis of peptide functionalization was performed using a using a fluorescent dye, fluorescein isothiocyanate (FITC) F luorescence microscopy was used to detect th e FITC labeled RGDS by emission of green visible light (520 nm) when exposed to ligh t with a wavelength of 494 nm 126 The fluorescence intensity generall y increased with increasing FITC RGDS exp osure on all gels aside from the least 5 mM 0.05 mM 0.5 mM 0 mM 5 mM 0.5 mM 0.05 mM 0 mM
114 crosslinked network as seen in Figure 6 5 A high fluorescence signal was obtained with gels exhibiting an MPTSDS/(TRIS+HEMA) ratio of 2.5 mmol/mmol compared to the other compositions. F igure 6 5. Fluorescence of FITC RGDS conjugated amphiphilic crosslinked networks Fluorescence signal normalized to gel with highest fluorescence signal amongst all hydrogels (top) and fluorescence normalized to the sample with highest fluorescence signal within the composition (bottom) 0 0.2 0.4 0.6 0.8 1 1.2 2.5 10 75 Fluorescence (au) MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol) 0.05 mM 0.5 mM 5 mM 0 0.2 0.4 0.6 0.8 1 2.5 10 75 Fluorescence (au) MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol) 0.05 mM 0.5 mM 5 mM RGDS Concentration RGDS Concentration
115 Differences in swelling abilities between gels may have also played a role in the fluorescence intensity emitted from the samples. The overall fluorescence signal decreased as the gel became more highly crosslinked. This FITC labeled RGDS molec ule, which has a molecular weight of approximately 833 g/mol, may have penetra ted the network with an MPTSDS/(TRIS+HEMA) molar ratio of 2.5 mmol/mmol. The gel with a n MPTSDS/(TRIS+HEMA) molar ratio of 75 mmol/mmol exhibited the lowest fluorescent intensity for all peptide concentrations compared to the others. Other methods, such as XPS, might be better suited to quantify the concentration of peptide on the surfaces of these gels due to the smaller penetration depth To ensure that the surface modification process did not affect the bulk properties of the network structure, swelling tests and mechanical tests were performed. The presence and con centration of peptide, RGDS, did not influence the water content of the amphiphilic hydrogels except for gel with an MPTSDS/(TRIS+HEMA) of 10 mmol/mmol exposed to a concentration of 0.05 mM RGDS as shown in Figure 6 6 This hydrogel also exhibited higher water content than the other gels The modulus of this sample with the same composition and RGDS concentration was found to be statistically significant modulus compared to the other gels with higher or lower RGDS functionalization (Figure 6 7) It should be mentioned that the gels functionalized with 0.05 mM RGDS were fabricated and characterized at a later date. How ever, differences between samples should not exist due to the similarities in sample size, test conditions, and test parameters
116 Figure 6 6 Swelling behavior of RGDS conjugated amphiphilic crosslinked networks of varying MPTSDS/(TR IS+HEMA) molar rat io Asteris k (*) denotes samples with statistical significance (p < 0.05). F igure 6 7. Compression modulus as a function of initial concentration of RGDS on amphiphilic crosslinked networks of varying MPTSDS/(TR IS+HEMA) molar ratio Asteris k (*) de notes statistically significance between gels with differing peptide concentration (p < 0.05) 0 5 10 15 20 25 30 2.5 10 75 EWC (%) MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol) 0 mM 0.05 mM 0.5 mM 5.0 mM 0 1 2 3 4 5 6 2.5 10 75 Modulus (MPa) MPTSDS/(TRIS+HEMA) Molar Ratio (mmol/mmol) 0 mM 0.05 mM 0.5 mM 5 mM RGDS Concentration RGDS Concentr ation
117 The functionalization of the crosslinked networks was not expected to influence the bulk properties, due to the modification of the gel taking place at the surfa ce. I f the peptide was conjugated into the bulk of the material the mechanical and swelling p roperties might be affected; due mainly to the molecular weight of the RGDS (M w of 433.42 g/mol) and the interaction between the peptide and the synthetic n etwork system. This was observed with PEG based hydrogels with peptides (YIGSR/RGDS/IKVAV) incorporated throughout the network. The type and concentration of peptide in a PEG based ne twork system was found to influence swelling, mesh size, and storage modulus, du e to ligand interaction with the polymer network (i.e.: via Van der Waals or ionic bonding). The YIGSR peptide in the PEG based hydrogels influenced the mesh size and swelling ratio, due to the hydrogen bonding between the phenolic hydroxyl group and the ether oxygen on the PEG. This interaction resulted in lower mesh size and reduced swelling, which in turn may have led to the higher elastic modulus. It was only at h of peptide incorporation that lead to differences in bulk properties (swelling ratio, mesh size, and storage modulus) between hydrogels with and without peptide incorporation 77 Proliferation of SMCs on the gels with varying modulus was assessed using a PicoGre en Assay. This assay utilizes a fluorescent nucleic acid stain to quantify double stranded DNA ( dsDNA) from lysed cells. SMCs were grown on hydrogels with varying compression modulus and peptide concentration (0.05, 0.5, and 5 mM RGDS) for 1, 3, and 7 days (Figure 6 8) It was hypothesized that the gels with the highest modulus would exhibit the high est density of cells; this positive correlation would be observed with peptide concentration as well.
118 F igure 6 8. SMC proliferation on TCPS and RGDS conjugated amphi philic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mm ol) after (a) 1, (b) 3, and (c) 7 days 0 2 4 6 8 10 0.05 0.5 5 DNA Content (ng/mL) RGDS Initial Concentration (mM) Day 1 2.5 10 75 TCPS 0 2 4 6 8 10 0.05 0.5 5 DNA Content (ng/mL) RGDS Initial Concentration (mM) Day 3 2.5 10 75 TCPS 0 5 10 15 20 25 0.05 0.5 5 DNA Content (ng/mL) RGDS Initial Concentration (mM) Day 7 2.5 10 75 TCPS Composition Composition Composition (a) (b) (c)
119 Peptide concentration does not have a significant impact on SMC density at day s 1 and 3. This behavior contradicts what has been seen in the literature in regards to protein concentration. Cell proliferation generall y increa ses with protein concentration at these time points or even earlier 40 39 However, SMC proliferation has been shown to dec rease with increased peptide concentration 10 Only at day 7, did the peptide concentr ation have an impact on cell density. At this time point SMC density general ly increased with increasing peptide concentration for all substrates. Moreover, t he SMC DNA content is higher on gels with higher modulus for all time points The gel with an MP TSDS/(TRIS+HEMA) molar ratio of 75 mmol/m mol has the highest cell density compared to the other crosslinked gels. It seems that the substrate modulus is positively correlated with cell adhesion which has been observed previously on PDMS materials with var ying modulus (0.05 1.79 MPa). The proliferation rate was greatest on softer siloxane coatings, with cell density increasing 20 fold from day 2 to day 7 79 However, SMCs proliferation was found to be positively correlated with substrate modulus on polyacrylamide and PEG hydrogels coatings exhibiting a moduli in the range of 0.03 0.08 MPa and 0.01 0.42 MPa, respectively 11 12
120 F igure 6 9 Proliferation of SMCs on conjugated amphi philic crosslinked networks of varying MPTSDS/(TRIS+HEMA) molar ratio (mmol/mmol) after 1, 3, and 7 days (peptide concentration = 5 mM). Interestingly enough, lower cell adhesion and growth was observed on the TCPS substrate compared to the hydrogel with an MPTSDS/(TRIS+HEMA) molar ratio of 75 mmol/mmol. TCPS with a modulus of approximately 3 GPa T he TCPS substrate exhibited the lowest cell density at day 1 and was comparab le to hydrogels with an MPTSDS/(TRIS+ HEMA) molar ratio of 2.5 and 10 mmol/mmol at day 3. At day 7, this substrate surpass ed those two gels in regards to cell density. Geometric cues from proteins have been shown to influence the proliferation of SMCs. Thakar et al. have shown that the bovine SMC shape on micropatterned ovals and circles affected the proliferation rate when the cell spreading area was kept constant 127 The proliferation rate of bovine SMCs grown on collagen coated channels (20 to 50 m) was affected by channel width. The SMCs grown on patterns containing 30 to 20 m wide channels proliferated at a statistically significant (p < 0.05) rate that 0 1 2 3 4 5 6 1 3 7 Cell Proliferation (Fold Increase) Time (Day) 2.5 10 75 TCPS Composition
121 was slower than the others Cells were elongated in the direction of the channels, had a lower cell spreading area, and exhibited le ss stress fiber formation on the patterns with smaller channel width (30 to 20 m). Interestingly enough, there was a d actin expression in SMCs grown on channels with smaller width (30 to 20 m), suggesting that the elongated morphology may not always induce contractile phenotype 128 The attachment of SMCs onto patterned and smooth hydrogels that were coated with fibronectin (50 g/mL) and not RGDS, with varying modulus was investigated (Figure 6 10) The a mphiphilic gels were topographically modified with the Sharklet AF TM pattern to assess cell adhesion. The fabrication method can be seen in the appendix These gels have been shown to adhere t o this topography and align in a way that may in fact induce SMC phenotype 129 Preliminary experiments indicate that the modulus and pattern does not influ ence the SMC attachment on pattern ed samples after 24 hours. The patterned samples exhibited higher attachment on gels with 10 and 75 mmol/mmol MPTSDS/(TRIS+HEMA) molar ratio but were not deemed statistically significant (p > 0.05). The TCPS substrate exh ibited the highest concentration of cells.
122 F igure 6 10 SMC attachment on smooth or topographically modified (Sharklet AF TM +1 SK 2x2 wafer no. 5816 ) amphiphilic hydrogels with M PTSDS/ (TRIS+HEMA) molar ratio of 2.5, 10, and 75 mmol/mmol after 24h Sm and SK denote samples that are smoot h and patterned, respectively A s teri sks (*) denotes statistical significance to the TCPS control (p < 0.05) Conclusions Amphiph i lic crosslinked networks were fabricated with tunable mechanical properties and then functionalized with the bioactive molecule RGDS, at the surface to promote cell adhesion and growth The functionalization of the surface was shown not to influence compression modulus or swelling behavior of most of the hydrogels, which was expected due to the modification occurring at the surface and not in the bulk. Fluorescence labeling of the peptide show a general increase in fluorescence intensity as the peptide concentration is increased. However, the differences in crosslink network resulted in v ast differences in peptide concentration due to FITC absorption into the gels Opacity of the least crosslinked samples may have also affected the fluorescence intensity obtained, resulting in a higher measured protein conjugation. 0 5 10 15 20 25 2.5 SK 2.5 Sm 10 SK 10 Sm 75 SK 75 Sm TCPS DNA Content (ng/mL) Composition
123 Intimal hyperplasia, whic h is the excessive growth of SMCs, is a common condition that diminishes the effectiveness of vascular grafts due to the de differentiation of SMCs into a synthetic phenotype 41 The following results demonstrate the importance of substrate modulus and chemical cues in attachment and proliferation rate of mammalian cells. The elastic modulus of the mat erial was shown to influence the attachment and proliferation of the SMCs on all days The cells were able t o sense the substrate stiffness once cell attachment was achieved through by ligand anchorage Gels with higher substrate stiffness resulted in high er cell attachment and cell proliferation Moreover, adhesion was generally enhanced with patterned hydrogel compositions compared to the smooth hydrogels, indicating that the topographically modified gels may be use f ul in facilitating cell attachment.
124 CHA PTER 7 CONCLUSIONS AND FUTURE WORK Conclusions The material cell/biomolecule interaction must be considered when constructing materials t hat are used to control attachment of biomolecules and unicellular organisms The purpose of this research was to dete rmine what role the physicochemical properties of the surface play in influencing biological responses of marine and mammalian organisms. The results support the notion that physical and chemical parameters of a substrate govern the att achment and prolifer ation of organisms in two model systems: Ulva linza and smooth muscle cells. Hydrogels were an appropriate choice as the substrate material to study the behavior of these organisms due to their easily tunable chemical and mechanical properties. Chemical a nd Mechanical Properties of Poly(HEMA co DEGDMA) Hydrogels Poly(2 hydroxyethyl methacrylate) hydrogels were fabricated with varying di functional monomer concentration and found to have a compression modulus ranging from 0.4 1.1 MPa The low swelling c %) and low molecular weight between crosslinks (M c < 180 g/mol) endow these hydrogels with an elastic modulus higher than what is reported for PEG based hydrogels. The surface energy between differe nt compositions did not deviate subs tantially between compositions; so the lack of variance between surface energy and compression modulus were presumed to lead to less than noticeable differences in bioadhesion of marine organisms. This was observed in the bovine serum adsorption assay. The re was no statistical difference or obvious trends correlating crosslink molar ratio and protein adsorption.
125 Chemical and Mechanical Properties of Amphiphilic Hydrogels A mphiphilic crosslinked networks containing siloxane monomers, were fabrica ted in ord e r to obtain a wide range in mod ulus This was the case with gels exhibiting an order of magnitude difference in compression modulus (0.4 4.7 MPa) by varying the molar crosslink ratio in the feed composition The di functional monomer concentration in th e feed composition w as found to influence the sessile drop contact angle and hysteresis. When the contact angles were measured in air the hydrogels exhibited a hydrophobic surface. When the hydrogels were equilibrated in water and the contact angles were m easured in an aqueous environment, the hydrophilic domains are dominant at the surface, as indicated by high surface energies. The captive bubble contact angle measurements indicate that the surfaces with varying MPTSDS/(TRIS+HEMA) molar ratio do not devia te substantially. The ability of the backbone chains and side groups to reorganize in order to minimize the interfacial tension was influenced by the crosslink concentration in the feed composition Differences between gels in an aqueous environment mainly lie in their network structure, which influences the mechanica l properties Thus, differences in biological responses seen between gels must be a result of these parameters. Fouling Properties of Amphiphilic Hydrogels A trend was observed between the hydr ogels with varying crosslink molar ratio and BSA adsorption. The R 2 value between protein adsorption versus surface energy (captive bu bble technique) and modulus E was 0.66 and 0.67, respectively. The hysteresis and M c were found to exhibit a higher correlation to protein adsorption, exhibiting R 2 value of 0.72 and 0.93, respectively However, the differences between gels w ere not deemed statistically significant. The protein adsorption assay proved to be
126 a good pr edictor for the marine organism attachment assays. A similar trend between crosslink molar ratio and accum ulation of spores and sporeling biomass growth was observ ed in the assays to determine AF and FR ab ilities of the gels Mo reover, the relative growth of sporeling biomas s did not positively correlate with (E ) 1/2 ; contradicting the results seen previously with elastomer coatings exhibiting a modulus ranging from (0.2 to 9.4 MPa) 69 The settlement of Ulva zoospores and sporeling s was negatively c orrelated with crosslin ker molar ratio Surface energy of the substrates did not differ substantially between substra tes, so other networks parameters (i.e., M c and elastic modulus) w ere postulated to pla y a large role in the biomass growth of sporelings. A high er correlation wa s obtained betw een sporeling biomass and M c (R 2 = 0.98) t han between sporeling biomass and modulus (R 2 = 0.34) This was also the case for the results obtained from the assay examining sporeling biomass growth on the hydrogels The p ercent removal of spor eling biomas s exhibited a high er correlation with M c (R 2 = 0.83) than with the substrate modulus (R 2 = 0.39) The fact that there was a higher correlation between M c and biofouling of sporelings than modulus and biofouling suggests that surface characteris tics may play a stronger role than the bulk properties. None of the coatings display FR properties towards sporelings compared to the PDMSe control. However, a ll of the amphiphilic coatings exhibited better antifouling properties towards spores compared to the PDMSe control, with one of the coatings exhibiting an 80% reduction in spore density compared to the PDMSe control. Sporeling biomass was lower on all amphiphilic coatings, except for the gel with 2.5 mmol/mmol crosslink ratio, compared to the PDMSe control
127 RGDS Functionalized Amphiphilic Hydrogels Hydrogels with similar mechanical and chemical properties were made with GMA incorporated into the network to promote a ring open ing reaction between GMA and the peptide, RGDS This was done in order t o promote adhesion between SMCs and the surface of the crosslinked networks. Determining the peptide concentration at the surface proved to be difficult to the opacity of the lesser crosslinked samples and the difference in swelling between samples. To mea sure the peptide concentration, FITC fluorescent probes were used. However, FITC absorption was present in the least crosslinked gels, resulting in substantially high fluorescence intensity. Attachment and proliferation of SMCs on hydrogels with varying R GDS initial concentration and substrate modulus was investigated. Cells were able to adhere and grow on the surface after 24 hours. The hydrogel with the highest DNA content at all time points was the gel exhibiting the highest substrate modulus. This was expected due to the inherent nature of SMCs to proliferate at hig her modulus, especially when the substrate modulus may resemble the stiffness observed in a diseased blood vessel. It would seem that this rapid proliferation is a result of cells expressing the synthetic phenotype. A positive correlation was seen between cell DNA content and the initial RGDS concentration only on Day 7. The higher initial peptide concentration may have facilitated higher levels of cell anchorage at the surface, which would re sult in enhanced pr oliferation Hydrogels with the Sharklet AF TM (+1 SK 2x2) pattern were also shown to promote enhanced cell attachment; higher attachment of cells was observed on patterned gels with an MPTSDS/(TRIS+HEMA) molar ratio of 10 and 75 mmol/mmol However, a statistically significant difference between smooth and patterned surfaces was not
128 observed for any of the compositions All of the gels, patterned and smooth, exhibited higher attachment of cells than the TCPS control. S tatistically significa nt cell attachment was found on the gels 2.5 Sm, 10 Sm, 10 Sk compared to TCPS. Further work will be done to elucidate what role these topographical features play in cell morphology and phenotype modulation. Future Work RGDS Functionalized Amphiphilic H ydrogels Additional chemical analysis at the surface should be performed to determine the surface coverage of the peptide, which can be done using XPS. Other assays to detect biomolecule conjugation to the surface of these hydrogels should be performed. F or example, the ninhydrin assay can be used to detect biomolecules with primary and secondary amines 130 E xperiments to investigate the SMC phenotype on the RGDS crosslinked netw orks sho uld be done in order to determine if the SMCs grown on these netwo rks are indeed in a mature contractile state. Reverse transcriptase polymerase chain reaction can be used to quantitatively assess differentiation markers (smooth muscle actin, caldesmon, a nd calponin) in the SMCs grown on these types of gels This assay could also be used to evaluate SMC phenotypic modulation on the topographically modified samples over a longer period of time that is, greater than 24 hours In addition, cell s preading and shape should be studied on topographically modified gels with varyi ng composition due to the importance of alignment of mature SMCs in the native blood vessel. Due to the wide and physiologically relevant range in stiffness, other cells can be cultured on these materials that are also present in the blood vessel (i.e., endothelial
129 cells and fibroblasts) Attachment of endothelial progenitor cells has proven to be problematic for tissue engineered vascular constructs. Finding ways to circumvent such problem s is necessary to prevent cardiovascular complications, such as thrombosis. Th is amphiphilic material can also be used to grow various types of stem cells, such as mesenchymal stem cells, with the intention of inducing differentiation into certain lineages based on the substrate modulu s. AF and FR Properties of Amphiphilic Hydrogels These amphiphilic coatings that performed best in the Ulva assays should also be examined for the foul resistant properties towards other marine organisms, such as barnacles, b ecause t ougher materials are necessary to avoid degradat ion by these hard fouler s Fouling resistant properties may be enhanced wit h topography (Sharklet TM ) as it did with PEGDMA hydrogels and PEG grafted PDMS coatings 129 Incorporation of higher concentrations of the siloxane monomers or inclusion of fluoropolymers, may also improve the FR properties of the gels. Due to the chemical heterogeneity of the surface, more surface analysi s should be completed in various environments Morphological changes can be studied using the AFM with these amphiphilic gels in a hydrated and dehydrated state Stark d ifferences in the morphology of the substrate may be seen in the hydrated state for the lower crosslinked networks Addi tionally, t he mole cular structure of these amphiphilic gels should be analyzed by sum frequency generation vibrational spectroscopy. Inferences about functional groups present and buried i nto the bulk can be made in air and in water using this instrument. Information gathered from this instrument can be used to support data obtained from contact angle measurements in both environments.
130 APPENDIX F ABRICATION OF T OPOGRAHPICALLY MODIF IED PHEMA BASED HYDROGELS Patterned hydrogels were fabricated by solvent ca sting of prepolymerization mixtures into molds that contained a silicon wafer with the negative of the desired pattern. Channels and the Sharklet AF TM pattern with varying feature size were replicated onto the hydrogel surface. The quality of the pattern w as assessed by light microscopy and scanning elect r on microscopy (SEM). For synthesis of the poly(HEMA co DEGDMA) hydrogels, HEMA and DEGDMA were added to glass bottles. Initiators, APS SMBS and DI water were added to the glass bottles The prepolymeriz ation mixtures were purged with nitrogen for approximately 2 minutes and then placed in a centrifuge (2000 RPM) to degas approximately 2 minutes. For synthesis of the amphiphilic gels, t he monomers diluent, and photo initiator were transferred to glass bo ttles and mixed until the photo initiator dissolved completely The prepolymerization mixtures were purged with nitrogen for approximately 2 minutes. The glass plates used in the mold for hydrogel polymerization were treated with HMDS via vapor deposition to create a hydrophobic surface for easy release of hydrogels. The topographically modified silicon wafers (Wafer no. 7477) with the negative of the desired pattern were cleaned with acetone and ethanol then treated with HMDS via vapor deposition for 5 mi nutes. Prior to mold assemblage, the silicon wafers were treated with the appropriate prepolymerization mixtures to wet the surface. The molds were then assembled. The molds consisted of a poly(dimethylsiloxane) elastomer (PDMSe) gasket and silicon wafer s andwiched between two hexamethyld isilazane treated glass or quartz plates, held together with paper clamps.
131 The prepolymerization mixtures for the amphiphilic gels were then transferred by a pipette into the molds (Figure A 1) The prepolymerization mixtur es were then exposed to ultraviolet (UV) light (BLAK RAY Lamp, Model UVL 56, 366nm) for 70 minutes at room temperature. To remove residual monomer and hexanol from the hydrogels, the hydrogels were soaked in swelling solvents. The amphiphilic hydrogels wer e submerged in 50 v/v % ethanol in water solution for 24 hours, a 25 v/v % ethanol in water solution for 24 additional hours, followed by DI water until characterization or protein ad s orption assays were performed. The prepolymerization mixture s for the HE MA gels were transferred by a pipette into a mold. The mold was then placed in the oven in a vertical position for 3 hours at 50 C. After curing, the hydrogel films were soaked in deionized water for 24 hours to remove residual monomer and initiator. F igure A 1. Scheme of mold used for hydrogel polymerization by UV or heat. Images were obtained using the Zeiss microscope (Figure A 2 and Figure A 3) The samples were imaged while in a hydrated state. Prior to obtaining SEM images, the samples were lyophilized and sputter coated with Au/Pd to prevent charging of the
132 sample by the electron beam. The SEM was operating under secondary electron mode and the voltage was set at 5 kV (Figure A 2 and Figure A 3) F igure A 2 Image of poly (HEMA co DEGDMA) hydrogel with the +2.8 SK 2x2 pattern using the SEM at 2000x magnification (left) and Zeiss optical microscope (right). The SEM was operating under secondary electron mode, using a voltage of 5 kV. F igure A 3. Image of the amphiphil ic gels with the +2.8 SK 2x2 pattern using the SEM at 2000x magnification(left) and Zeiss optical microscope (right).
133 LIST OF REFERENCES 1. Niklason LE, Gao J, Abbott WM, Hirschi KK, Houser S, Marini R, Langer R. Functional Arteries Grown in Vitro. Science 1999;284(5413):489 493. 2. Callow J, Callow M. Trends in the development of environmentally friendly fouling resistant marine coatings. Nat Commun 2011;2(244). 3. Schultz M, Bendick J, Holm E, Hertel W. Economic impact of biofouling on a naval surface ship. Biofouling 2011;27(1):87 98. 4. Kwon O, Nho Y, Lee Y. Radiation induced copolymerization of 2 hydroxyethyl methacrylate and polyethylene glycol methacrylate and its protein adsorption and bacterial attachment. J Ind Eng Chem 200 3;9(2):138 145. 5. Rasmussen K, Willemsen P, Ostgaard K. Barnacle settlement on hydrogels. Biofouling 2002;18(3):177 191. 6. Murosaki T, Noguchi T, Kakugo A, Putra A, Kurokawa T, Furukawa H, Osada Y, Gong J, Nogata Y, Matsumura K and others. Antifouling ac tivity of synthetic polymer gels against cyprids of the barnacle ( Balanus Amphitrite ) in vitro Biofouling 2009;25(4):313 320. 7. Murosaki T, Noguchi T, Hashimoto K, Kakugo A, Kurokawa T, Saito J, Chen Y, Furukawa H, Gong J. Antifouling properties of tough gels against barnacles in a long term marine environment experiment. Biofouling 2009;25(7):657 666. 8. Lai Y. Role of bulky polysiloxanylalkyl methacrylates in oxygen permeable hydrogel materials. J Appl Polym Sci 1995;56:317 324. 9. Martinelli E, Agostin i S, Galli G, Chiellini E, Glisenti A, Pettitt M, Callow M, Callow J, Graf K, Bartels F. Nanostructured films of amphiphilic fluorinated block copolymers for fouling release application. Langmuir 2008;24:13138 13147. 10. Mann B, West J. Cell adhesion pepti des alter smooth muscle cell adhesion, proliferation, migration, and matrix protein synthesis on modified surfaces and in polymer scaffolds. J Biomed Mater Res 2002;60:86 93. 11. Brown X, Bartolak Suki E, Williams C, Walker M, Weaver V, Wong J. Effect of s ubstrate stiffness and PDGF on the behavior of vascular smooth muscle cells: Implications for atherosclerosis. J Cell Physiol 2010;225:115 122. 12. Peyton S, Raub C, Keschrumrus V, Putnam A. The use of poly(ethylene glycol) hydrogels to investigate the imp act of ECM chemistry and mechanics on smooth muscle cells. Biomater 2006;27:4881 4893.
134 13. Prendergast G. Settlement and Behavior of Marine Fouling Organisms. In: Durr S, Thomason J, editors. Biofouling. West Sussex, UK: Blackwell Publishing Ltd; 2010. p 3 0 59. 14. Swain G, Schultz M. The testing and evaluation of non toxic antifouling coatings. Biofouling 1996;10(1 3):187 197. 15. Thomas K, Brooks S. The environmental fate and effects of antifouling paint biocides. Biofouling 2010;26(1):73 88. 16. Targett N, Stochaj W. Natural antifoulants and their analogs: applying nature's defense strategies to problems of biofouling control. In: Thompson M, Nagabhushanam R, Sarojini R, Fingerman M, editors. Recent developments in biofouling control: A.A. Balkema/Rotterd am; 1994. p 221 227. 17. Price R, Patchan M, Clare A, Rittschof D, Bonaventura J. Performance enhancement of natural antifouling compounds and their analogs through microencapsulation and controlled release. In: Thompson M, Nagabhushanam R, Sarojini R, Fin german M, editors. Recent Developments in Biofouling Control: A.A. Balkema/Rotterdam; 1994. p 321 334. 18. Bakus G, Wright M, Khan A, Ormsby B, Gulko D, Licuanan W, Carriazo E, Ortiz A, Chan D, Lorenzana D and others. Experiments seeking marine natural ant ifouling compounds. In: Thompson M, Nagabhushanam R, Sarojini R, Fingerman M, editors. Recent Developments in Biofouling Control: A.A. Balkema/Rotterdam; 1994. p 373 381. 19. Dee KC, Puleo DA, Bizios R. An Introduction To Tissue Biomaterial Interactions: J ohn Wiley & Sons, Inc.; 2002. p 1 13. 20. Ratner B, Hoffman A, Schoen F, Lemons J. Biomaterials Science, an introduction to materials in medicine. San Diego, CA: Elsevier Academic Press; 2004. 21. Wang X, Lin P, Yao Q, Chen C. Development of small diameter vascular grafts. World J Surg 2007;31:682 689. 22. Xue L, Greisler HP. Biomaterials in the development and future of vascular grafts. J Vasc Surg 2003;37(2):472 480. 23. Salacinski HJ, Goldner S, Giudiceandrea A, Hamilton G, Seifalian AM, Edwards A, Carso n RJ. The Mechanical Behavior of Vascular Grafts: A Review. J Biomater Appl 2001;15(3):241 278. 24. Kiyama H, Imazeki T, Kurihara S, Yoneshima H. Long term Follow up of Polyurethane Vascular Grafts for Hemoaccess Bridge Fistulas. Ann Vasc Surg 2003;17(5):5 16 521.
135 25. Kirkpatrick CJ, Unger RE, Krump Konvalinkova V, Peters K, Schmidt H, Kamp G. Experimental approaches to study vascularization in tissue engineering and biomaterial applications. J Mater Sci Mater Med 2003;14(8):677 681. 26. Chan Park MB, Shen J Y, Cao Y, Xiong Y, Liu Y, Rayatpisheh S, Kang GC W, Greisler HP. Biomimetic control of vascular smooth muscle cell morphology and phenotype for functional tissue engineered small diameter blood vessels. J Biomed Mater Res A 2009;88A(4):1104 1121. 27. Valar mathi MT, Davis JM, Yost MJ, Goodwin RL, Potts JD. A three dimensional model of vasculogenesis. Biomater 2009;30(6):1098 1112. 28. Campbell N, Reece J. Biology, 7th Edn. San Francisco: Pearson Education, Inc.; 2005. 29. Martins Green M, Bissell MJ. Cell EC M interactions in development. Semin Dev Biol 1995;6(2):149 159. 30. Worth NF, Rolfe BE, Song J, Campbell GR. Vascular smooth muscle cell phenotypic modulation in culture is associated with reorganisation of contractile and cytoskeletal proteins. Cell Moti l Cytoskel 2001;49(3):130 145. 31. Xu CY, Inai R, Kotaki M, Ramakrishna S. Aligned biodegradable nanofibrous structure: a potential scaffold for blood vessel engineering. Biomater 2004;25(5):877 886. 32. Hayward I, Bridle K, Campbell G, Underwood P, Campbe ll J. Effect of extracellular matrix proteins on vascular smooth muscle cell phenotype. Cell Bio Inter 1995;19(10):839 846. 33. McKenzie J. Protein interactions at material surfaces. In: Narayan R, editor. Biomedical Materials. New York, NY: Springer Scien ce+Business Media, LLC; 2009. p 215 237. 34. Moiseeva E. Adhesion receptors of vascular smooth muscle cells and their functions. Cardiovasc Res 2001;52:372 386. 35. Schnapp LM, Hatch N, Ramos DM, Klimanskaya IV, Sheppard D, Pytela R. The Vitronectin. J Biol Chem 1995;270(39):23196 23202. 36. Maloney JM, Walton EB, Bruce CM, Van Vliet KJ. In fluence of finite thickness and stiffness on cellular adhesion induced deformation of compliant substrata. Phys Rev E 2008;78(4):041923. 37. Peng X, Huang J, Xiong C, Fang J. Cell adhesion nucleation regulated by substrate stiffness: A Monte Carlo study. J Biomech 2012;45:116 122.
136 38. Bkel C, Brown NH. Integrins in Development: Moving on, Responding to, and Sticking to the Extracellular Matrix. Dev Cell 2002;3(3):311 321. 39. Naito M, Funaki C, Hayashi T, Yamada K, Kanichi A, Yoshimine N, Kuzuya F. Substra te bound fibrinogen, fibrin and other cell attachment promoting proteins as a scaffold for cultured vascular smooth muscle cells. Atherosclerosis 1992;96:227 234. 40. Hirst S, Twort C, Lee T. Differential effecs of extracellular matrix proteins on human ai rway smooth muscle cell proliferation and phenotype. Am J Respir Cell Mol Biol 2000;23:335 344. 41. Beamish J, He P, Kottke Marchant K, Marchant R. Molecular regulation of contractile smooth muscle cell phenotype: Implications for vascular tissue engineeri ng. Tissue Engineering 2010;16(5):467. 42. Beamish J, Fu A, Choi A, Haq N, Kottke Marchant K, Marchent R. The influence of RGD bearing hydrogels on the re expression of contractile vascular smooth muscle cell phenotype. Biomater 2009;30:4127 4135. 43. Engl er A, Sen S, Sweeney H, Discher D. Matrix elasticity directs stem cell lineage specification. Cell 2006;126:677 689. 44. Engler AJ, Griffin MA, Sen S, Bnnemann CG, Sweeney HL, Discher DE. Myotubes differentiate optimally on substrates with tissue like sti ffness. J Cell Bio 2004;166(6):877 887. 45. Munoz Pinto DJ, Bulick AS, Hahn MS. Uncoupled investigation of scaffold modulus and mesh size on smooth muscle cell behavior. Journal of Biomedical Materials Research Part A 2009;90A(1):303 316. 46. Engler A, Ric hert L, Wong J, Picart C, Discher D. Surface probe measurements of the elasticity of sectioned tissue, thin gels and polyelectrolyte multilayer films: correlations between substrate stiffness and cell adhesion. Surf Sci 2004:1 13. 47. Magin C, Finlay J, Ca llow M, Callow J, Brennan A. Antifouling performance of cross linked hydrogels: Refinement of an attachment model. Biomacromol 2011;12:915 922. 48. Wang R, Kreuzer H, Grunze M. The interaction of oligo(ethylene oxide) with water: a quantum mechanical study Phys Chem Chem Phys 2002;2:3613 3622. 49. Harder P, Grunze M, Dahint R, Whitesides G, Laibinis P. Molecular confirmation in oligo(ethylene glycol) terminated self assembled monolayers on gold and silver surfaces determines their ability to resist protein adsorption. J Phys Chem B 1998;102:426 436.
137 50. Ucar I, Cansoy E, Erbil H, Pettitt M, Callow M, Callow J. Effect of contact angle hysteresis on the removal of the sporelings of the green alga Ulva from the fouling release coatings synthesized from polyole fin polymers. Biointerphases 2010;5(3):75 84. 51. Carman ML, Estes TG, Feinberg AW, Schumacher JF, Wilkerson W, Wilson LH, Callow ME, Callow JA, Brennan AB. Engineered antifouling microtopographies correlating wettability with cell attachment. Biofouling 2006;22(1):11 21. 52. Guo Z, Liu W, Su B. Superhydrophobic surfaces: From natural to biomimetic to functional. J Colloid Interf Sci 2011;353:335 355. 53. Park D, Weinman C, Finlay J, Fletcher B, Paik M, Sundaram H, Dimitriou M, Sohn K, Callow M, Callow J and others. Amphiphilic surface active triblock copolymers with mixed hydrophobic and hydrophilic side chains for tuned marine fouling release properties. Langmuir 2010;26(12):9772 9781. 54. Joshi R, Goel A, Mannari V, Finlay J, Callow M, Callow J. Evaluat ing fouling resistance and fouling release performance of smart polyurethane surfaces: An outlook for efficient and environmentally benign marine coatings. J Appl Polym Sci 2009;114:3693 3703. 55. Li Y, Liu C, Yang J, Gao Y, Li X, Que G, Lu J. Anti biofoul ing properties of amphiphilic phosphorylcholine polymer films. Colloids Surf B Biointerf 2011;85(2):125 130. 56. Tan B, Hussain H, Chaw K, Dickinson G, Gudipati C, Birch W, Teo S, He C, Liu Y, Davis T. Barnacle repellant nanostructured surfaces formed by t he self assembly of amphiphilic block copolymers. Polym Chem 2010;1:276 279. 57. Gudipati C, Finlay J, Callow J, Callow M, Wooley K. The antifouling and fouling release performance of hyperbranched fluoropolymer (HBFP) poly(ethylene glycol) (PEG) composite coatings evaluated by adsorption of biomacromolecules and the green fouling alga Ulva. Langmuir 2005;21:3044 3053. 58. Feng S, Wang Q, Gao Y, Huang Y, Qing F. Synthesis and characterization of a novel amphiphilic copolymer capable as anti biofouling coati ng material. J Appl Polym Sci 2009;114:2071 2078. 59. Pieper R, Ekin A, Webster D, Casse F, Callow J, Callow M. Combinatorial approach to study the effect of acrylic polyol composition on the properties of crosslinked siloxane polyurethane fouling release coatings. J Coat Technol Res 2007;4(4):453 461. 60. Maldonado Codina C, Efron N. Dynamic wettability of pHEMA based hydrogel contact lenses. Ophthal Physiol Opt 2006;26:408 418.
138 61. Good R. Contact angle, wetting, and adhesion: a critical review. J Adhes S ci Technol 1992;6(12):1269 1302. 62. Chaudhury M, Whitesides G. Correlation between surface free energy and surface constitution. Science 1992;255:1230 1232. 63. Owens D, Wendt R. Estimation of the surface free energy of polymers. J Appl Polm Sci 1969;13:1 741 1747. 64. Baier R. The role of surface energy in thrombogenesis. Bull NY Acad Med 1972;48(2):257 272. 65. Baier R, Meyer A. Surface analysis of fouling resistant marine coatings. Biofouling 1992;6:165 180. 66. Lindner E. A low surface free energy appro ach in the control of marine biofouling. Biofouling 1992;6:193 205. 67. Berglin M, Lonn N, Gatenholm P. Coating modulus and barnacle bioadhesion. Biofouling 2003;19(Supplement):63 69. 68. Brady RJ, Singer I. Mechanical factors favoring release from fouling release coatings. Biofouling 2000;15(1 3):73 81. 69. Chaudhury M, Finlay J, Chung J, Callow M, Callow J. The influence of elastic modulus and thickness on the release of the soft fouling green alga Ulva linza (syn. Enteromorpha linza ) from poly(dimethylsi loxane) (PDMS) model networks. Biofouling 2005;21(1):41 48. 70. Young C, Wu J, Tsou T. Fabrication and characteristics of polyHEMA artificial skin with improved tensile properties. J Membrane Sci 1998;146:83 93. 71. Lou X, Munro S, Wang S. Drug release cha racteristics of phase separation pHEMA sponge materials. Biomaterials 2004;25(20):5071 5080. 72. Webster D, Chrisholm B. New directions in antifouling technology. West Sussex. UK: Blackwell Publishing Ltd; 2010. 366 387 p. 73. Humphrey AJ, Finlay JA, Petti tt ME, Stanley MS, Callow JA. Effect of Ellman's Reagent and Dithiothreitol on the Curing of the Spore Adhesive Glycoprotein of the Green Alga Ulva. The Journal of Adhesion 2005;81(7 8):791 803. 74. Gonen Wadmany M, Oss Ronen L, Seliktar D. Protein polymer conjugates for forming photopolymerizable biomimetic hydrogels for tissue engineering. Biomater 2007;28(26):3876 3886.
139 75. Akdemir ZS, Akakaya H, Kahraman MV, Ceyhan T, Kayaman Apohan N, Gngr A. Photopolymerized Injectable RGD Modified Fumarated Poly( ethylene glycol) Diglycidyl Ether Hydrogels for Cell Growth. Macromol Biosci 2008;8(9):852 862. 76. Gunn J, Turner S, Mann B. Adhesive and mechanical properties of hydrogels influence neurite extension. J Biomed Mater Res 2005;72A:91 97. 77. Zustiak S, Dur bal R, Leach J. Influence of cell adhesive peptide ligands on poly(ethylene glycol) hydrogel physical, mechanical and transport properties. Acta Mater 2010;6:3404 3414. 78. Bi J, Downs C, Jacob J. Tethered protein/peptide surface modified hydrogels. J Biom ater Sci Polym Edn 2004;15(7):905 916. 79. Brown X, Ookawa K, Wong J. Evaluation of polydimethylsiloxane scaffolds with physiologically relevant elastic moduli: interplay of substrate mechanics and surface chemistry effects on vascular smooth muscle cell r esponse. Biomater 2005;26:3123 3129. 80. Tomic S, Micic M, Dobic S, Filipovic J, Suljovrujic E. Smart poly(2 hydroxyethyl methacrylate/itaconic acid) hydrogels for biomedical application. Radiat Phys Chem 2010;79(5):643 649. 81. Cowling M, Hodgkiess T, Par r A, Smith M, Marrs S. An alternative approach to antifouling based on analogues of natural processes. Sci Total Environ 2000;258:129 137. 82. Arima T, Hamada T, McCabe J. The effects of cross linking agents on some properties of HEMA based resins. J Dent Res 1995;74(9):1597 1601. 83. Lou X, van Coppenhagen C. Mechanical characteristics of poly(2 hydroxyethyl methacrylate) hydrogels crosslinked with various difunctional compounds. Polym Int 2001;50:319 325. 84. Sperling L. Introduction to Physical Polymer S cience. Hoboken: Wiley; 2006. 85. Davis T, Hughlin M. Swelling of poly(2 hydroxyethyl methacrylate) gels in water/dioxan mixtures. Polym Commun 1987;28:218 220. 86. Refojo M, Yasuda H. Hydrogels from 2 hydroxyethyl methacrylate and propylene glycol monoacr ylate. J Appl Polym Sci 1965;9:2425 2435. 87. Peppas N, Moynihan H, Lucht L. The structure of highly crosslinked poly(2 hydroxyethyl methacrylate) hydrogels. J Biomed Mater Res 1985;19:397 411. 88. Urban M. Attenuated Total Reflectance Spectroscopy of Poly mers: Theory and Practice. Washington, DC: American Chemical Society; 1996.
140 89. Carr L, Cheng G, Xue H, Jiang S. Engineering the polymer backbone to strengthen nonfouling sulfobetaine hydrogels. Langmuir 2010;26(18):14793 14798. 90. Nguyen D. Determination of equilibrium surface energy of adsorbed polyvinyl alcohol layers in water at 25 degrees C. Colloids Surf A 1996;116:145 160. 91. Hamilton W. A technique for the characterization of hydrophilic solid surfaces. J Colloid Interf Sci 1972;40(2):219 222. 92. Jones M, McColl I, Grant D, Parker K, Parker T. Protein adsorption and platelet attachment and activation, on TiN, TiC, and DLC coatings on titanium for cardiovascular applications. J Biomed Mater Res 2000;52:413 421. 93. Hydroxyalkyl Methacrylate Polymers and Copolymers. Journal of Macromolecular Science, Part C: Polymer Reviews 1973;9(1):3 47. 94. Kiremitci M, Pesmen A, Pulat M, Gurhan I. Relationship of surface characteristics to cellular attachment in PU and PHEMA. J B iomater Appl 1993;7:250 264. 95. Perova T, Vij J, Xu H. Fourier transform infrared study of poly(2 hydroxyethyl methacrylate) PHEMA. Colloid Polym Sci 1997;275:323 332. 96. Lee HB, Jhon MS, Andrade JD. Nature of water in synthetic hydrogels. I. Dilatometry specific conductivity, and differential scanning calorimetry of polyhydroxyethyl methacrylate. Journal of Colloid and Interface Science 1975;51(2):225 231. 97. Bajpai AK, Shrivastava M. ADSORPTION DYNAMICS OF BOVINE SERUM ALBUMIN (BSA) ONTO BINARY INTERP ENETRATING POLYMER NETWORKS (IPNS) OF POLY(2 HYDROXYETHYL METHACRYLATE) (PHEMA). Journal of Macromolecular Science, Part A 2001;38(11):1123 1139. 98. Ying P, Jin G, Tao Z. Competitive adsorption of collagen and bovine serum albumin effect of the surface we ttability. Colloids and Surfaces B: Biointerfaces 2004;33(3 4):259 263. 99. Herrwerth S, Eck W, Reinhardt S, Grunze M. Factors that Determine the Protein Resistance of Oligoether Self Terminal Hydrophilicity, and Lateral Packing Density. Journal of the American Chemical Society 2003;125(31):9359 9366. 100. Finlay JA, Callow ME, Ista LK, Lopez GP, Callow JA. The Influence of Surface Wettability on the Adhesion Strength of Settled Spores of the Green Alga Entero morpha and the Diatom Amphora. Integrative and Comparative Biology 2002;42(6):1116 1122.
141 101. Marabotti I, Morelli A, Orsini LM, Martinelli E, Galli G, Chiellini E, Lien EM, Pettitt ME, Callow ME, Callow JA and others. Fluorinated/siloxane copolymer blends for fouling release: chemical characterisation and biological evaluation with algae and barnacles. Biofouling 2009;25(6):481 493. 102. Lee SH, Ruckenstein E. Surface restructuring of polymers. J Colloid Interf Sci 1987;120(2):529 536. 103. Karlsson J, Gat enholm P. Surface mobility of grafted hydrogels. Macromol 1999;32(22):7594 7598. 104. Yasuda H, Sharma A. Effect of orientation and mobility of polymer molecules at surfaces on contact angle and its hysteresis. J Polym Sci Polym Physics Edn 1981;19:1285 12 91. 105. Schmidt D, Brady RJ, Lam K, Schmidt D, Chaudhury M. Contact angle hysteresis, adhesion, and marine biofouling. Langmuir 2004;20:2830 2836. 106. Kurokawa T, Gong JP, Osada Y. Substrate Effect on Topographical, Elastic, and Frictional Properties of Hydrogels. Macromolecules 2002;35(21):8161 8166. 107. Holly F, Refojo M. Wettability of hydrogels I. Poly(2 hydroxyethyl methacrylate). J Biomed Mater Res 1975;9:315 326. 108. Wenzel RN. RESISTANCE OF SOLID SURFACES TO WETTING BY WATER. Industrial & Engine ering Chemistry 1936;28(8):988 994. 109. Otsuka H, Nagasaki Y, Kataoka K. Surface Characterization of Functionalized Polylactide through the Coating with Heterobifunctional Poly(ethylene glycol)/Polylactide Block Copolymers. Biomacromolecules 2000;1(1):39 48. 110. Extrand C, Kumagai Y. Experimental studies on wetting and contact angle hysteresis of soft polymer surfaces. Proceedings of the annual meeting Adhesion Society Meeting 1996;19:466 468. 111. Faibish RS, Yoshida W, Cohen Y. Contact Angle Study on P olymer Grafted Silicon Wafers. Journal of Colloid and Interface Science 2002;256(2):341 350. 112. Gerbig YB, Phani AR, Haefke H. Influence of nanoscale topography on the hydrophobicity of fluoro based polymer thin films. Applied Surface Science 2005;242(3 4):251 255. 113. Wang H J, Cao Y, Sun Y Y, Wang K, Cao C, Yang L, Zhang Y D, Zheng Z, Li D, Wang J Y and others. Is there an optimal topographical surface in nanoscale affecting protein adsorption and cell behaviors? Journal of Nanoparticle Research 2011;1 3(9):4201 4210.
142 114. Mequanint K, Patel A, Bezuidenhout D. Synthesis, Swelling Behavior, and Biocompatibility of Novel Physically Cross Linked Polyurethane block Poly(glycerol methacrylate) Hydrogels. Biomacromolecules 2006;7(3):883 891. 115. Lee CT, Smith 2004;44(2):524 536. 116. a F and others. Poly(ethylene glycol) Containing Hydrogel Surfaces for Antifouling Applications in Marine and Freshwater Environments. Biomacromolecules 2008;9(10):2775 2783. 117. Gray NL, Banta WC, Loeb GI. Aquatic Biofouling Larvae Respond to Differences in the Mechanical Properties of the Surface on which they Settle. Biofouling 2002;18(4):269 273. 118. Wagenseil JE, Mecham RP. Vascular Extracellular Matrix and Arterial Mechanics. Physiological Reviews 2009;89(3):957 989. 119. Wesly RL, Vaishnav RN, Fuch s JC, Patel DJ, Greenfield JC. Static linear and nonlinear elastic properties of normal and arterialized venous tissue in dog and man. Circulation Research 1975;37(4):509 520. 120. Hern DL, Hubbell JA. Incorporation of adhesion peptides into nonadhesive hy drogels useful for tissue resurfacing. J Biomed Mater Res 1998;39(2):266 276. 121. Kleinman HK, Philp D, Hoffman MP. Role of the extracellular matrix in morphogenesis. Curr Opin Biotech 2003;14(5):526 532. 122. Mateo C, Torres R, Fernndez Lorente G, Ortiz C, Fuentes M, Hidalgo A, Lpez Gallego F, Abian O, Palomo JM, Betancor L and others. Epoxy New Tool for Improved Immobilization of Proteins by the Epoxy Method. Biomacromolecules 2003;4(3):772 777. 123. Mann B, Gobin A, Tsai A, Schmedlen R, West J. Smooth muscle cell growth in photopolymerized hydrogels with cell adhesive and proteolyticaly degradable domains: synthetic ECM analogs for tissue engineering. Biomater 2001;22:3045 3051. 124. Nasirtabrizi M, Mohebalizadeh S, Jadid A. Glycidyl m ethacrylate polymers containing indole groups: Synthesis and characterization. Iranian Polym J 2011;20(7):579 586. 125. Bai Y X, Li Y F, Wang M T. Study on synthesis of a hydrophilic bead carrier containing epoxy groups and its properties for glucoamylase immobilization. Enzyme and Microbial Technology 2006;39(4):540 547.
143 126. Seo HS, Ko YM, Shim JW, Lim YK, Kook J K, Cho D L, Kim BH. Characterization of bioactive RGD peptide immobilized onto poly(acrylic ac id) thin films by plasma polymerization. Applied Surface Science 2010;257(2):596 602. 127. Thakar RG, Cheng Q, Patel S, Chu J, Nasir M, Liepmann D, Komvopoulos K, Li S. Cell Shape Regulation of Smooth Muscle Cell Proliferation. Biophys J 2009;96(8):3423 34 32. 128. Thakar RG, Ho F, Huang NF, Liepmann D, Li S. Regulation of vascular smooth muscle cells by micropatterning. Biochem Biophys Res Commun 2003;307(4):883 890. 129. Magin CM. Engineered microtopographies and surface chemistries direct cell attachment and function . United States -Florida: University of Florida; 2010. 130. Starcher B. A Ninhydrin Based Assay to Quantitate the Total Protein Content of Tissue Samples. Analytical Biochemistry 2001;292(1):125 129.
144 BIOGRAPHICAL SKETCH Angel Ej iasi was raised in Cedar Rapids, IA. She recei ved her Bachelor of Science degree in biomedical e ngineering in 2005 at the University of Iowa and a Master of Sc ience degree in materials science and e ngineering in 2009 at the University of Florida