Citation
Endothelial cell adhesion on acrylate-based shape memory polymers for use in cardiovascular stents

Material Information

Title:
Endothelial cell adhesion on acrylate-based shape memory polymers for use in cardiovascular stents
Creator:
Govindarajan, Tina
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Doctor of philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Bioengineering, CU Denver
Degree Disciplines:
Bioengineering
Committee Chair:
Park, Daewon
Committee Members:
Shandas, Robin
Benninger, Richard
Mestroni, Luisa
Jacot, Jeffrey
Peña, Brisa

Notes

Abstract:
Although cardiovascular disease is the leading cause of death worldwide, current treatment methods continue to be limited. Balloon angioplasty, bare metal stents (BMS) and drug eluting stents (DES) provide minimally invasive methods for opening and supporting narrowed vessels, but complications such as restenosis and thrombosis from reduced biocompatibility resulted in vessel re-occlusion, limiting this approach. Shape memory polymers (SMPs) have shown promise as polymer stents due to their self-deployment capabilities and vascular biocompatibility. Prior research on SMPs for stent use has focused on the mechanical properties for stent deployment and implantation and to confirm low cytotoxicity. However, both in vitro studies and in vivo studies using animal models, have demonstrated that endothelialization of a device surface soon after implantation enhances the likelihood of device integration, leading to device success. Thus, to make SMPs a more viable option for cardiovascular stents, verifying the endothelial cell recruitment ability of the surface is an important step. This work aims to optimize the surface of a shape memory polymer previously developed in our group to encourage endothelialization for future use in cardiovascular stents. First, we optimized the SMP formulation to retain its shape memory properties while also encouraging endothelial cell adhesion and survival. The 80:20 weight percent ratio tBA:PEGDMA SMP was selected for further experiments, based on previous thermomechanical data as well as acquired endothelialization data. In vivo, healthy vascular ECs are typically elongated and aligned in the direction of flow; as a result, many studies involving implanted blood-contacting devices contain topographical cues that direct cells to adhere in an organized fashion. Topographical surface modifications provide a less resistive regulatory path compared to those of the biological or chemical variety. Grooved surfaces have demonstrated cell adhesion, alignment, and improved cell function on a variety of surfaces, including metal and polymer; as a result, microgrooves were introduced to the surface of the SMPs using metal printed molds. Metal printing offers a simple, cost-effective, reproducible, and robust method for mold fabrication that can be used to fabricate molds with surface features on the order of tens of microns. Microgrooved SMPs demonstrated increased cell adhesion, survival and alignment compared to their unpatterned analogues. To further optimize rapid cell adhesion, groove depth was increased, and shape memory was extended to the surface features. Through initial compression of the grooved SMP surface and subsequent surface recovery at physiological temperature after cell introduction, cells attach to seemingly flat surfaces prior to feature recovery. which increases cell adhesion. The microgrooves also continue to encourage cell alignment. These studies provide some preliminary data that may aid in the future use of these materials for cardiovascular stents.

Record Information

Source Institution:
Auraria Library
Holding Location:
University of Colorado Denver
Rights Management:
Copyright Tina Govindarajan. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
ENDOTHELIAL CELL ADHESION ON ACRYLATE-BASED
SHAPE MEMORY POLYMERS FOR USE IN CARDIOVASCULAR STENTS
By
TINA GOVINDARAJAN
B.S., University of Colorado at Boulder, 2012
A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Bioengineering Program
2019


This thesis for the Doctor of Philosophy degree by Tina Govindaraj an has been approved for the Bioengineering Program by
Daewon Park, Chair Robin Shandas, Advisor Richard Benninger Luisa Mestroni Jeffrey Jacot Brisa Pena
Date: May 18, 2019


Govindarajan, Tina (PhD, Bioengineering)
Endothelial Cell Adhesion on Acrylate Based Shape Memory Polymers for use in Cardiovascular Stents
Thesis directed by Professor Robin Shandas
ABSTRACT
Although cardiovascular disease is the leading cause of death worldwide, current treatment methods continue to be limited. Balloon angioplasty, bare metal stents (BMS) and drug eluting stents (DES) provide minimally invasive methods for opening and supporting narrowed vessels, but complications such as restenosis and thrombosis from reduced biocompatibility resulted in vessel re-occlusion, limiting this approach. Shape memory polymers (SMPs) have shown promise as polymer stents due to their self-deployment capabilities and vascular biocompatibility.
Prior research on SMPs for stent use has focused on the mechanical properties for stent deployment and implantation and to confirm low cytotoxicity. However, both in vitro studies and in vivo studies using animal models, have demonstrated that endothelialization of a device surface soon after implantation enhances the likelihood of device integration, leading to device success. Thus, to make SMPs a more viable option for cardiovascular stents, verifying the endothelial cell recruitment ability of the surface is an important step.
This work aims to optimize the surface of a shape memory polymer previously developed in our group to encourage endothelialization for future use in cardiovascular stents. First, we optimized the SMP formulation to retain its shape memory properties while also encouraging endothelial cell adhesion and survival. The 80:20 weight percent ratio tBA:PEGDMA SMP was
m


selected for further experiments, based on previous thermomechanical data as well as acquired endothelialization data.
In vivo, healthy vascular ECs are typically elongated and aligned in the direction of flow; as a result, many studies involving implanted blood-contacting devices contain topographical cues that direct cells to adhere in an organized fashion. Topographical surface modifications provide a less resistive regulatory path compared to those of the biological or chemical variety. Grooved surfaces have demonstrated cell adhesion, alignment, and improved cell function on a variety of surfaces, including metal and polymer; as a result, microgrooves were introduced to the surface of the SMPs using metal printed molds. Metal printing offers a simple, cost-effective, reproducible, and robust method for mold fabrication that can be used to fabricate molds with surface features on the order of tens of microns. Microgrooved SMPs demonstrated increased cell adhesion, survival and alignment compared to their unpatterned analogues.
To further optimize rapid cell adhesion, groove depth was increased, and shape memory was extended to the surface features. Through initial compression of the grooved SMP surface and subsequent surface recovery at physiological temperature after cell introduction, cells attach to seemingly flat surfaces prior to feature recovery, which increases cell adhesion. The microgrooves also continue to encourage cell alignment. These studies provide some preliminary data that may aid in the future use of these materials for cardiovascular stents.
The form and content of this abstract are approved. I recommend its publication.
Approved: Robin Shandas
IV


ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to many individuals who have contributed to my personal and professional development. First, I would like to express my heartfelt appreciation to my advisor, Dr. Robin Shandas, for his unwavering support of my training and research, as well as for his patience, motivation and encouragement and most importantly, for always pushing me to be a better scientist. I would also like to extend my thanks to my thesis committee: Dr. Daewon Park, Dr. Richard Benninger, Dr. Luisa Mestroni, Dr. Jeffrey Jacot, and Dr. Brisa Pena for their insight and support over the course of the dissertation. I am indebted to all of you.
Additionally, I am very grateful to current/former members of the Shandas Lab, specifically Jennifer Wagner, Kiran Dyamenahalli, Roopali Shah and Michael Zimkowski, who provided much of the foundational training that made this work possible. An additional shout-out to Jennifer Wagner and Brisa Pena -1 truly believe that I would not have made it this far without their support and assistance. There have also been many individuals who have assisted me with equipment training/usage, specifically Steven Lewis, Eric Wartchow, the folks at the NanoCharacterization Facility (NCF) at CU Boulder, Stephen Huddle, and of course, the Bioengineering Department at CU Denver (Anschutz. Thank you all.
To my friends, thank you for encouraging me, letting me vent when I needed to, and pushing me to be the best person I can be, in all aspects. To my family, my mother and younger brother, who continue to love, support, and lift me up, even on days when I may make it difficult, I cannot thank you enough. Finally, to my late father, who taught me more about life than he, or I, ever thought possible, thank you. I would not be the person I am today without you all and words cannot express how sincerely grateful I am for your presence in my life.
v


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION................................................................1
Cardiovascular Disease......................................................1
Current Stents and Limitations .............................................3
Endothelialization .........................................................6
Polymers ...................................................................8
Shape Memory Polymers ......................................................9
Polymer & Shape Memory Polymer Stents .....................................13
Surface Modification to Increase Biocompatibility..........................13
Methods for Surface Modification...........................................14
Surface Roughening.........................................................15
Surface Patterning.........................................................19
Chemical Modification of the Surface.......................................24
Surface Coatings and Films.................................................28
Attachment of Pharmaceuticals, Biopharmaceuticals or Biomolecules
to the Surface.............................................................30
Porous Surfaces to Facilitate Drug Delivery................................33
II. SHAPE MEMORY POLYMERS CONTAINING HIGHER ACRYLATE
CONTENT DISPLAY INCREASED ENDOTHELIAL CELL ATTACHMENT......................35
Introduction...............................................................35
Materials and Methods......................................................37
Results....................................................................41
vi


Discussion
52
III. SMP FORMULATION OPTIMIZATION FOR ENDOTHELIALIZATION:
INCREASED ACRYLATE CONTENT IN SHAPE MEMORY POLYMERS.................58
Introduction........................................................58
Materials and Methods...............................................59
Results.............................................................60
Discussion..........................................................67
IV. MICROGROOVES ENCOURAGE ENDOTHELIAL CELL ADHESION
AND ORGANIZATION ON SHAPE MEMORY POLYMER SURFACES...................70
Introduction........................................................70
Materials and Methods...............................................73
Results.............................................................77
Discussion..........................................................91
V. THERMOELASTIC RECOVERY OF MACROSCALE SHAPE MEMORY
POLYMER SURFACE FEATURES............................................97
Introduction........................................................97
Materials and Methods..............................................100
Results............................................................103
Discussion.........................................................113
VI. TEMPERATURE-ACTIVATED MICROGROOVES IMPROVE ENDOTHELIAL CELL ADHESION AND ALIGNMENT ON SHAPE
MEMORY POLYMER SURFACES............................................119
Introduction.......................................................119
vii


Materials and Methods
122
Results.................................................................125
Discussion..............................................................136
VII. CONCLUSIONS, LIMITATIONS AND FUTURE WORK...............................140
Conclusions.............................................................140
Limitations.............................................................143
Future Work.............................................................145
REFERENCES.....................................................................152
APPENDIX
A. FTIR of Original SMP Formulations......................................176
B. CellEvent Caspase 3/7 Green for Apoptosis of Original Formulations....178
C. Extended Live/Dead Study of Select SMPs - Day 7........................179
D. Scanning Electron Microscopy (SEM) of Cell-Adherent SMP................181
E. D31 Staining of Microgrooved and Unpatterned SMPs......................182
viii


LIST OF TABLES
TABLE
2.1. Tg, Tonset and Tg range for SMP Formulations...................43
4.1 Tg, Tonset and Tgrange for unpatterned and microgrooved 80:20 wt%
tBA:PEGDMA550, 750, 1000 SMPs .................................78
5.1. SMP Formulations used for this study. Three different weight percent
ratios were tested, with 3 variations of PEGDMA MW, for 9 total
formulations..........................................................101
5.2. Methods matrix for compression & recovery of SMP surface
protrusion............................................................103
5.3 Glass transition temperatures and compression and recovery ranges of
select SMPs investigated..............................................104
6.1 Glass Transition Temperature, Compression Temperature and
Microgroove Recovery Rate.............................................128
7.1. Glass Transition Temperature, Compression Temperature and
Microgroove Recovery Rate of 80:20 tBA:PEGDMA550 SMPs with microgroove widths of 50pm, 100 pm & 150 pm...........................147
IX


LIST OF FIGURES
FIGURE
1.1 Surface modification techniques commonly used to enhance endothelialization
and/or reduce thrombosis................................................15
1.2 Scanning electron microscope (SEM) image of RIE textured silicon
surface using plasma consisting of C12, CF4 and 02 gases................17
1.3 Atomic force microscopy (AFM) images of (a) untreated, (b) micro-roughened
and (c) nano-roughened polydimethylsiloxane (PDMS) films................19
1.4 SEM image of Silicon pillars formed via plasma processing................21
1.5 Isolated platelets in buffer adhering to both surfaces, but platelets in plasma
do not adhere to ion treated polymer surface............................25
2.1 SMP formulation matrix of the nine formulations used.....................38
2.2 (A) Storage Modulus (MPa) of (A) tBA:PEGDMA550 (B) Storage Modulus (MPa)
of tBA:PEGDMA750 (C) Storage Modulus (MPa) of tBA:PEGDMA1000............42
2.3 Atomic force microscopy (AFM) images of (a) tBA:PEGDMA550,
(b) IB A: PEGDM A75 0 and (c) tBA:PEGDMA1000............................43
2.4 (A) Representative AFM images of tBA:PEGDMA550 samples
(B) Roughness of tBA:PEGDMA550 (C) Roughness of tBA:PEGDMA750
(D) Roughness of tBA:PEGDMA1000......................................45-46
2.5 Live-Dead Analysis of SMP formulations with the lowest weight percent of
monomer (20 wt% tBA)....................................................47
2.6 Live-Dead Analysis of SMP formulations with equal weight percent monomer
(tBA) and crosslinker (PEGDMA)..........................................47
x


2.7 Live Dead Analysis of SMP formulations with highest weight percent (80 wt%)
monomer (tB A)..................................................................48
2.8 (A) Cell count of HUVECs present on each sample (B) Normalized EC
Attachment...................................................................49-50
2.9 Metabolic Activity/Cytocompatibility of SMPs....................................51
3.1 High tBA content SMP Formulations...............................................59
3.2 Glass transition temperature (Tg) of all SMPs...................................60
3.3 Contact angle of SMPs...........................................................61
3.4 SMP surface roughness measurements..............................................62
3.5 Endothelial cell adhesion on 85:15 wt% tBA:PEGDMA SMPs........................63
3.6 Endothelial cell adhesion on 90:10 wt% tBA:PEGDMA SMPs........................64
3.7 Endothelial cell adhesion on 95:5wt% tBA:PEGDMA SMPs............................64
3.8 Estimated adherent ECs on SMPs..................................................65
3.9 Cell metabolism of cell-adherent SMPs...........................................66
4.1 Storage modulus and tan delta curves for 80:20 wt% unpatterned and
microgrooved SMPs...............................................................77
4.2 SEM micrographs verified pattern transfer and surface feature presence, or lack
thereof, on unpatterned and micropattemed SMP surfaces .........................79
4.3 Roughness of unpatterned vs. microgrooved 80:20 wt% tBA:PEGDMA SMP
surfaces........................................................................80
4.4 Wettability of unpatterned vs. microgrooved SMP surfaces........................81
xi


4.5 Endothelial cell attachment at 1 day, 3 days and 7 days after cell introduction assessed using Live/Dead Cell Imaging and NucBlue Live Cell Stain to
mark nuclei.................................................................82
4.6 A) Approximate cell presence, live and dead, on all unpatterned vs. microgrooved SMPs B) Percentage increase in EC presence on unpattemed
and microgrooved SMPs.......................................................84
4.7 Approximate adherent ECs/SMP, live and dead, on all unpatterned vs.
microgrooved SMPs...........................................................86
4.8 Approximate adherent ECs/SMP, live only, on all unpattemed vs.
microgrooved SMPs...........................................................87
4.9 Endothelial cell alignment as measured by nuclei and actin fiber organization on 80:20 wt% tBA:PEGDMA550, 80:20 wt% tBA:PEGDMA750 and 80:20
wt% tBA:PEGDMA1000, 1 day and 7 days after cell introduction............89-90
5.1 Deformation retention of 50:50 tBA:PEGDMA550 vs. 80:20 tBA:PEGDMA550 .......105
5.2 Percent recovery of SMP protrusion at temperatures above Tg for A) 0.5mm
protrusion B) 1.0mm protrusion.........................................106-107
5.3 Percent recovery of SMP protrusion at physiological temperature for A) 0.5mm
protrusion B) 1.0mm protrusion ............................................108
5.4 Recovery time for A) 60 second compression B) 60 minute compression.....109-110
5.5 Recovery time at temperatures 25% above Tg and 50% above Tg.................Ill
5.6 Percent height compression for A) 0.5mm protrusion and B) the 1.0mm
protrusion Tg..............................................................112
6.1 Metal Mold Design and Printing..............................................123
xii


6.2 Surface manipulation of SMPs...................................................126
6.3 SEM micrographs confirm pattern transfer to the SMP surface with groove
widths and depths ranging from 55-65 pm (p ± SD)...............................129
6.4 Contact angle and groove depth of micropatterned SMP surfaces before
compression, after compression and after recovery..............................130
6.5 Atomic Force Microscopy of 80:20 wt% tBA:PEGDMA 550,750, 1000............ 131-132
6.6 Endothelial cell adhesion 24 hours after cell introduction assessed using
Live/Dead Cell Imaging...........................................................133
6.7 Endothelial cell adhesion counts and Live/Dead ratios, 24 hours after cell
Introduction assessed using Live/Dead Cell Imaging...............................133
6.8 Endothelial cell alignment as assessed by measuring nuclei and actin orientation.135
7.1 Figure summary of percentage endothelial cell adhesion, relative to ECs
introduced on Day 0 of the study, on all SMPs & surfaces.........................143
7.2 SEM of varying groove widths and depth of 80:20 wt% tBA:PEGDMA550 SMPs...........147
7.3 Cell adhesion on passive and temperature-responsive 50pm 80:20 wt%
tBA:PEGDMA550...........................................................148
7.4 Cell adhesion on passive and temperature-responsive 100pm 80:20 wt%
tBA:PEGDMA550...........................................................149
7.5 Cell adhesion on passive and temperature-responsive 150pm 80:20 wt%
xm
tBA:PEGDMA550
149


CHAPTERI
INTRODUCTION
“Portions of this chapter were previously published in Polymers, 2014, 6 and are included
with the permission of the copyright holder
Cardiovascular Disease
Cardiovascular disease (CVD), including atherosclerosis and related diseases, is one of the leading causes of death globally.1 According to the American Heart Association (AHA),
CVD accounts for approximately 1 in every 3 deaths in the United States while on a global scale, deaths from CVD totaled almost 18 million, and are expected to exceed 23 million by 2030.2 As a result, the AHA predicts that medical costs relating to CVD are projected to increase to 749 billion by 2035.2 Risk factors for CVD include smoking, physical inactivity, poor nutrition, obesity, and poor management of cholesterol, blood sugar and blood pressure. While lifestyle alterations and adjustments are the first line of treatment for CVD, more advanced therapies, such as pharmaceutical, medical device or surgical interventions are usually required.
Atherosclerosis often results from localized inflammatory response and can be characterized by plaque formation in blood vessels.1,3> 4 Atherosclerosis, like many inflammatory diseases, is the result of the usually beneficial leukocyte recruitment process becoming uncontrolled.5 Arterial luminal endothelial cells and leukocytes connect by expressing adhesion molecules such as selectins, intercellular adhesion molecule-I (ICAM-I) and capsular cell adhesion molecule-I (VCAM-I) as well as the corresponding receptor molecule. This expression event, coupled with a chemotactic gradient, encourages leukocyte recruitment to the site of the inflammation. For atherosclerosis, the inflammation site is the blood vessel itself. Atherosclerotic plaque may consist of fat, cholesterol, calcium and/or blood components.6 Plaque buildup causes
1


hardening of the blood vessel and limits blood flow to tissues, ultimately leading to an acute ischemic condition such as stroke or myocardial infarction.5,7
Studies in both animals and humans have identified endothelial dysfunction as one of the key events in both early and advanced atherosclerosis.5,8-10 One instigator of atherosclerosis is an excess presence of cholesterol in the blood, termed hypercholesterolemia, which causes changes in permeability of arterial endothelial cells to allow lipids into and aggregate in the arterial wall. Studies by Schwenke and Carew have shown that lipoprotein retention in the artery wall appears to have a greater impact on atherosclerotic lesion formation than the lipoprotein transfer rate.11 Adhesion molecules expressed by endothelial cells bind circulating monocytes that subsequently migrate into the sub-endothelial space and become foamy macrophages; in conjunction with oxidized lipid particles, foamy macrophages further enhance the accumulation of lipids in the subcellular space. Cells from the arterial wall may emit oxidative products that seed lipids and initiate oxidation. Generally, lipid oxidation takes place in two stages: 1) lipids are oxidized 2) monocytes are recruited and become macrophages, which have great oxidative capacity.11 Oxidized lipids can may up-regulate adhesion molecules on ECs and cause injury to the endothelium.8,12 All of these events contribute to vascular remodeling associated with atherosclerosis and lesion formation.7,13
Traditionally, surgical methods such as coronary artery bypass graft surgery (CABG) were used to treat coronary artery disease (CAD), which involved using arteries or veins from other parts of the body to bypass the narrowed coronary vessel. CABG, a surgical procedure and the most common type of open-heart surgery, is highly invasive and comes with a gamut of risks including wound infection, bleeding, reactions to anesthesia, fever, pain, stroke, heart attack and
2


even death. Recovery time from a CABG procedure may take 6-12 weeks, or longer, but CABG is still the preferred procedure for patients with severe CAD.14
Carotid Endarterectomy (CEA) is a surgical procedure to remove the plaque from the carotid arteries in the neck, restoring blood flow to the brain and preventing stroke.15 The first CEA was performed in the mid-1950s, but efficacy data regarding this procedure was not available until the 1990s.16 Since CEA is an invasive procedure performed under sedation, the risks associated with surgery are also present here, and thus CEA is only performed in patients with active plaque that may embolize or in patients with greater than 70% stenosis, as these patients appear to benefit most from surgical intervention; most patients, however, benefit from less invasive methods, which include pharmaceuticals and lifestyle alterations, to mitigate the risks.15"17
To mitigate the risks associated with surgical intervention, balloon angioplasty was conceptualized in the mid-1960s and implemented in the 1970s as a less invasive technique to open occluded vessels.18,19 A balloon is inserted via catheter on a guidewire to the site of the blockage and inflated to re-expand the lumen and restore blood flow to downstream organs and tissues.16 While balloon angioplasty was a significant step in improving treatment of arterial occlusion, limitations stemmed from re-narrowing of arteries caused by elastic recoil, delay in vascular remodeling or neointimal proliferation.
Current Stents and Limitations
Cardiovascular stents are expandable tubes used to treat narrow or weakened arteries that arise as a result of atherosclerosis and its resultant sequelae, such as coronary artery disease, peripheral artery disease, etc. These devices provide a minimally invasive means to mechanically support the damaged vessel which restores oxygenated blood flow to the organs and tissues
3


downstream.20 Stents were first developed and implanted in the 1980s, and have since experienced improvements in material, design and function.19 Although cardiovascular stents have saved countless lives, the device has many limitations, which drive continued research in the area.21 In particular, thrombosis and restenosis continue to be relatively important problems with current stents. Given that these issues arise from surface interactions, surface modification techniques are an active area of current research.
Descriptions of an ideal stent have been profiled across literature 22 The ideal stent
should:
1) have a low profile and the ability to be crimped onto a balloon that is mounted on a guidewire
2) sufficiently expand once the stent reaches the target area and is deployed by balloon expansion
3) demonstrate good radial strength and minimal recoil, so that the stent offers support to the vessel wall, regardless of stresses and maintains its integrity
4) be able to navigate through even the small diameter atherosclerotic vessels
5) be radiopaque and/or MRI compatible for ease of stent placement
6) demonstrate blood compatibility and thrombo-resistivity, or resistance to platelet adhesion and activation
7) have the potential for drug delivery
The first types of stents used were bare metal stents (BMSs), composed of a variety of metals and/or alloys such as stainless steel, cobalt-chromium and tantalum for balloon-expandable stents, or nitinol (nickel-titanium alloy) for self-expanding stents 20 316L Stainless Steel (SS) is the most common material used to fabricate metal stents because it is corrosion
4


resistant and has suitable mechanical properties. These stents provided the necessary mechanical support for the weakened vessel; however, an increased risk for thrombosis and/or restenosis in these devices may generate additional need for reintervention 6 to 12 months post-implantation.20, 23
Restenosis, or re-narrowing of the vessel, often results from excessive neointimal proliferation following balloon angioplasty or stent implantation due to vessel injury from the expansion.6,18,24 Additional causes of restenosis may include reduced compliance between the stent and the vessel and excessive tissue-remodeling response to the stent material.18 Thrombogenicity, one of the aforementioned issues associated with BMSs, refers to increased propensity of the device or material to generate a blood clot on the material surface.25 Thrombotic events often occur due to net electrical charge differences between blood components and the stent surface, as well as surface potential incompatibility between the metal and the contacting blood.22,26
As a proposed improvement on BMS, the first generation of drug-eluting stents (DESs) consisted of a metal backbone and a permanent, non-absorbable polymer coating to house a drug of choice.27 While DESs offered control and localization for drug release to the injured vessel, incidences of hypersensitivity, heart attack and even death remained problematic.20 An improved DES replaced the non-absorbable polymer coating with a non-thrombogenic, absorbable one. This absorbable coating served to encourage endothelialization through a directed drug release profile and reduced inflammatory response during polymer degradation. While these improved DESs decreased the occurrence of restenosis through release of anti-proliferative agents, late stage thrombosis still occurred.1,6’28-31 Late stage or late stent thrombosis (LST) can result from a variety of issues, ranging from the stenting procedure itself to early termination of
5


antiproliferative drugs loaded in the stent; these issues may cause increased local fibrin deposition and delayed healing.28,29 Furthermore, studies have shown that drug elution itself may result in inhibited endothelial cell proliferation.32 Endothelialization
Vascular endothelial cells compose the inner lining of all blood vessels in the human body. Endothelial cells were once considered a passive barrier between blood and tissue, but it is now known that these cells collectively form a dynamic organ system, creating a semi-permeable barrier between blood and tissue.33 Furthermore, endothelial cells have been shown to differentiate based on both internal and external factors, in order to meet the needs of the environment in which they function.34 This dynamic organ system collectively participates in a range of physiological and pathological processes including vasculature development and remodeling, vascular tone and blood fluidity control, movement of blood vessels and nutrients, as well as active players in atherosclerosis and tumor angiogenesis.34 The endothelium is also an active component in the prevention of intimal hyperplasia and thrombosis.32
Integrin binding to the extracellular matrix mediates endothelial cell adhesion and migration. Integrins are heterodimeric cell surface receptors that serve as anchors to the extracellular matrix. Integrins, which recognize specific extracellular matrix ligands to facilitate adhesion, also serve as vehicles of chemical and mechanical signals between cells and their environments. Numerous studies have investigated both the chemical and mechanical mechanisms associated with cell adhesion. Chemically, the focus has centered on the interaction between endothelial integrins and ECM matrix proteins such as fibronectin, laminin, and collagen. On the mechanical level, substrate stiffness has been shown to affect cell adhesive strength, contractility, focal adhesion formation, cell-cell interactions, etc.33
6


The process of endothelialization generally involves recruitment of endothelial cells as well as early and late endothelial progenitor cells (EPCs).35’36 The specific contribution of EPCs to the endothelialization process is still unclear, especially since these EPCs make up a very small percentage of circulating cells. Minor disturbances in endothelium that do not disturb the underlying basement membrane are primarily dealt with by migration of healthy endothelial cells from adjoining endothelium, whereas larger gaps in endothelium as well as endothelialization of implanted vascular grafts are primarily covered by endothelial cells in circulation.36 Initially, circulating hematopoietic stem cells and platelets adhere to the site of graft implantation; in some cases, these cells are replaced by the migration and spreading of endothelial cells, whereas in other instances endothelial cells replicate in an attempt to cover the exposed areas.35 Various methods and techniques to encourage endothelialization have been investigated and will be discussed later in this section.
It is now well known that a healthy, intact endothelium is required for protection against maladies such as hyperplasia and thrombosis; specifically, the endothelium is responsible for maintaining a homeostatic environment in the blood vessel.37 Many devices, biodegradable and otherwise, aim to encourage endothelialization of the surface to reap the benefits of the endothelium, which largely includes reduced device complication and rejection.35 Rapidly establishing a complete monolayer of endothelial cells on the luminal surface of a stent reduces and potentially eliminates many of the limitations associated with current stents and may guarantee long-term device success.38 Once cells adhere to a compatible scaffold, they can carry out regular functions, such as proliferation, migration, and differentiation, which may ultimately determine cellular survival or death.39
7


Polymers
Polymers are organic materials that have long chains held together by covalent bonds. Polymers are attractive materials for biomedical devices because they are highly tailorable and possess a wide range of properties, both chemical and physical, that may be attractive for biomedical device design. Polymers may be synthetic, which are attractive due to their mass producibility and serializability, but most synthetic polymers do not actively interact with their surroundings in their native form, and the path to regulatory approval can be challenging. Naturally delivered polymers on the other hand are, as the name suggests, naturally occurring, inside or outside of the body. These materials may integrate more easily with their surroundings but are more difficult to obtain in larger quantities and have relatively sub-optimal mechanical properties. Polymers are used in a variety of biomedical devices such as contact lenses, sutures, joint implants, gloves, wound dressings, as well as vascular grafts.40
Polymers are favorable materials for stents because they can achieve increased hemocompatibility with the proper selection of polymer components, polymerization and processing techniques.26 DESs have been shown to cause a delay in re-endothelialization, which may promote a thrombotic environment, as re-endothelialization is an important component in vessel healing. In addition, instances of very late stage thrombosis (LST) have also been seen with DESs, making polymer stents a potentially more appealing route for stent materials.41 Patients who receive DESs are often required to continue an anti-platelet regimen for 12 months to prevent adverse effects from the DESs and while these anticoagulants prevent thrombosis, they may carry a sustained risk of hemorrhage, or bleeding, and related side effects 24,42,43 Due to safety concerns with existing stent materials, current research in stent design is progressing
8


towards using biodegradable/bioabsorbable or biomimetic materials for polymer or metal stents as well as polymer coating-free DESs, among others.6,41,44,45 Shape Memory Polymers
Shape memory polymers are smart materials that can change their shape upon the application of a stimulus such as heat, light, infrared irradiation, humidity, immersion in water, electric field application, or the application of alternating magnetic fields, to name a few.46"50 To attest to their diversity, shape memory polymers (SMP’s) vary in chemical composition, mode of activation, and/or mode of degradation.51 SMP’s have advantages over their metallic, shape-memory alloy (SMA) counterparts in that SMP’s are very versatile and easily manipulated, as indicated above, and can handle strain deformations up to 800%, compared to only 8% for SMA.48,52 They also have low density, high frozen strain, low manufacturing costs, easy processing methods, wide shape transition, and biocompatibility.50 SMPs also have a higher mechanical stability than hydrogels, another polymer commonly used in the biomaterials field.48 SMA’s, while they do have benefits, especially in mechanical strength, also have drawbacks such as high material costs, limited thermo-mechanical property control, and limited resistance to fatigue.53
SMP’s are fabricated by processing the polymer using conventional methods into its permanent shape. The polymer is then deformed and fixed, or programmed, into its temporary shape. The permanent shape is recovered when the polymer is exposed to a stimulus and this recovery is a result of the shape memory effect (SME).48
SMP’s are governed heavily by the shape memory effect (SME). SME allows a polymer to undergo large amounts of strain without becoming permanently deformed; thus, the better the SME in a polymer, the more likely it is that the polymer will return to its original shape without
9


any evidence of prior deformation.54 It is also important to note that SMPs and shape-changing polymers are different; SMPs are driven by the SME which itself is driven by thermal transitions, whereas shape-changing polymers are driven by microscopic movements in the polymer resulting in macroscopic movement of the material.55 The SME is an effect of polymer structure, polymer morphology and applied processing and programming. Cyclic thermo-mechanical tests are used to validate the effectiveness of SME in polymers.48
SME is governed by a dual segment, or dual phase, system; one segment is elastic while the other segment is a transition segment that becomes softer when the stimulus is applied.52 These segments are dependent upon the chemical composition of the material and often exhibit strong interactions. Transition segments soften at a switching/transition temperature Tsw/Ttrans, either the glass transition temperature, Tg, or the melting temperature, Tm.52,55 Tg usually covers a broad range of temperatures whereas the range for Tm is more narrow, and thus Tg is more often used as Tsw or Ttrans.55 The Ttrans can be manipulated simply by altering the ratio of components, monomers and crosslinkers, that contribute to the make-up the polymer; other properties such as crosslinking density and mechanical properties also change with changes in this ratio.50,54
The permanent shape of the SMP is determined by the net-points and the chain segments that connect them.55 These chain segments and net-points may consist of either chemical crosslinks, which are often covalent bonds, or physical crosslinks, which are usually created by segregated domains. Crosslinks help keep the chains from slipping by acting like entanglements or anchors, helping the polymer retain its permanent shape. In addition, chemical crosslinks may help prevent the polymer from dissolving while still allowing the polymer to swell, which may be important for some applications.48 Domains with the highest thermal transition act as the
10


netpoints which determine the permanent shape, while domains with the second highest transition, the chain segments, act as molecular switches. These molecular switches are responsible for fixing the temporary, deformed shape48,55
Shape recovery is driven by entropy elasticity of switching domains. In the elastic state, the polymer is more amorphous and rubberier, and thus the polymer chains display a random coil formation and are at the state of higher entropy. In general, polymer segments reorient when the SMP is heated above Ttrans and again when there is an applied deformation to the polymer; when T>Ttrans, polymer segments exhibit entropically favorable random coil conformation. Property changes such as permeability, transparency, melting, and crystallization may be a result of the changing of the switching segments from a more oriented state to a more random conformation.55
SMP’s can be analyzed using a few different parameters. Shape fixity ratio (Rf) measures the ability of the switching domains in the SMP to fix a deformation when a mechanical stress is applied.52,55
(eu)
Rf = — x 100%
Em
au is the strain on unloading em is the max strain
The free strain recovery ratio (Rr) is a measure for the ability of the material to memorize and recover its permanent shape.48,52,55
Rr = v _ PJ x 100%
Em ~ Ep
eu is the strain on unloading £m is the max strain
is the strain of the system after recovery
11


The cyclic thermomechanical tests that were mentioned before result in stress and strain data, which can be plotted and the slope of the stress vs. strain curve yields the elastic modulus, E, which is a measure of the material’s stiffness, or its ability to deform elastically when a force is applied. The elastic modulus is important for materials such as SMPs that are used as medical devices and may need to mimic the properties of some native tissues or biological structures.48
There are several important considerations associated with changing even the slightest aspect of the monomer(s) and/or crosslinker(s) that make up a polymer; these considerations can be chemical, thermal or mechanical. Increased hard segment content is directly correlated to an increase in mechanical strength of the polymer. Increasing the number of hard segment blocks increases Tg which limits twisting and coiling of switching segments, controlling shape memory properties. Increased Tg may help with heat and oxidation resistance. Switching segments with higher molecular weights yield more crystalline solids. But, as the molecular weight of the switching segments decreases, the deformation recovery rate increases. Side groups on the monomers may function as physical crosslinks and help stabilize the material above Tg, depending upon the chemical nature of those side groups. Crosslinking is dependent upon initiator concentration, crosslinking temperature, and cure time and these can be adjusted to achieve maximum crosslinking.48
Yet another benefit of SMPs is that they can be modified to cater to specific purposes. Certain filler materials such as fibers or magnets can improve mechanical strength/increase modulus or enhance the magnetic properties, respectively. Their versatility, adaptability and ease of processing, and low cost have helped push these materials into the forefront of many research efforts.
12


Polymer & Shape Memory Polymer Stents
Polymers have caught the attention of the medical device industry due to their diversity and versatility. Polymers are less dense than metals and have higher flexibility than many other materials, which allows better matching of stent compliance with that of the local vessel.18 In addition, polymers are easy to manufacture and often have lower bulk material costs and processing costs.56,57 Polymers also possess a wide range of bulk properties, such as elasticity, conductivity, strength and degradability, which can provide the stent designer with a large palette of useful features.58 Thus, polymers can be easily and cost-effectively tailored to fit the needs of their application, making these materials appealing for use in the medical device industry.
Shape memory polymers (SMPs) have added advantages to those seen with conventional polymers. As stents are often delivered via catheter, SMP systems offer benefits for catheter storage and deployment, since the materials can pack tightly without becoming permanently deformed during the storage period. SMPs may also enhance the ease of delivery of many of these devices, and produce lower recovery forces, leading to minimally invasive procedures with reduced recovery times for patients.18 These added benefits, on top of the already present benefits of polymers, make SMPs potentially attractive materials for next-generation polymeric stents.
Surface Modification to Increase Biocompatibility
Despite the variety of materials and designs currently available for stents, there is still a need for a single material that has the desired mechanical properties while simultaneously achieving optimal biocompatibility.1,43 Biocompatibility refers to the reaction elicited by a material when it is inserted into the body; ideally, this reaction should be favorable and should not provoke a negative response such as an attack by the immune system on the foreign
13


material.59"62 Surface modification techniques strive to retain favorable bulk properties while changing the surface to cater to specific needs, often to enhance biocompatibility.29,63 Since shape memory is not a surface property, surface modifications should enhance biocompatibility without interfering with the shape memory capabilities of SMPs. Surface modifications that allow for improved blood contact (minimal thrombogenicity) while encouraging vascular wall healing via endothelial cell migration, anchorage and proliferation, are the focus of research goals in this area.1,21,59‘ 64‘65 In addition, surface modifications for drug release in an effort to eliminate the polymer coating are also being explored.1
One of the keys to success for many medical devices is successful wound healing, a process that begins at the surface of a material. Successful wound healing depends on a range of material properties, both surface and bulk, such as surface texture, surface chemistry, surface energy, crystallinity as well as leachable content and biocompatibility of the degradation products. In essence, biocompatibility is heavily dependent upon surface properties as well as interactions between the surface and cells and/or proteins, or between cells themselves.43,59,64-69 Protein adsorption may also play a factor in dictating the success or failure of blood-contacting devices; some proteins, such as albumin, can be beneficial for biocompatibility as albumin may decrease both platelet adhesion and binding of microorganisms that may elicit infection, but nonspecific proteins, such as fibrinogen and Immunoglobulin G (IgG), may increase platelet adhesion by instigating a host response.43 Methods for Surface Modification
Surface modifications should generally be thin, affecting only the topmost layer of the surface; thick layers may undesirably alter the bulk properties and have difficulty adhering to the surface, while overly thin layers are subject to erosion; despite these requirements, however,
14


there are a number of ways to modify the surface of a material to enhance its functionality.63-70 For polymer-derived stents, methods for modifying surfaces with the end goal of achieving improved blood compatibility, re-endothelialization, or both can be grouped into six major categories. These categories are surface roughening, surface patterning, chemical modification of the surface, surface coatings and films, attachment of pharmaceuticals or biopharmaceuticals to the surface, and the formation of porous surfaces to facilitate drug delivery, many of which are represented in Figure 1.1. Multiple techniques may be used to achieve the desired properties.1
a) Surface Roughening
b) Surface Patterning
c) Chemical Modification of the Surface
d) Attachment of Pharmaceuticals
harmaceuticals
e) Porous Surfaces to Facilitate Drug Delivery
Figure 1.1. Surface modification techniques commonly used to enhance endothelialization and/or reduce thrombosis.71
Surface Roughening
In general, surface roughening aims not only to increase the surface area of the material, but also to restrict cell movement, which contributes to enhanced cell attachment.43-72-74 Cells are still able to migrate on roughened surfaces, but no significant increases or decreases in migration have been noted compared to smooth surfaces.75 In addition, surface roughening modifies the
15


topology of the surface without chemical alteration, which may have benefits, depending upon the material and its desired use.76
For metals, roughening techniques such as sputtering with TiN or TiCE have been used to successfully enhance endothelial cell attachment. However, these cells express less endothelial nitric oxide synthase (eNOS), which may lead to increased endothelial cell dysfunction; this reduced eNOS activity has actually been shown to be characteristic of metals in general, modified or bare, presenting a reason for further research into non-metal implant materials.77 Microblasting followed by reactive ion etching on metal surfaces also produces roughened, high energy surfaces that may potentially improve cell attachment.42
For polymers, oxygen or argon plasma deposition increases surface roughness as well as hydrophilicity, both of which have been shown to enhance cell attachment; application of plasma deposition towards SMP-based stents may allow for enhanced wound healing and biocompatibility.43’57 Plasma processing alters the surface topography through melting and recrystallization processes, resulting in more ridges compared to the original surface, as displayed in Figure 1.2.62’78 Etching and sanding, both plasma- or chemical-based, as well as polishing and/or microblasting also serve to improve surface roughness.1,79
16


Figure 1.2. Scanning electron microscope (SEM) image of RIE textured silicon surface using plasma consisting of Cb, CF4 and O2 gases (scale bar = 200 nm). Reprinted with permission from Elsevier, 2001.62
Shadpour et al., roughened polymer surfaces using a slurry of alumina particles, with the intention of enhancing endothelial cell attachment without altering the chemical make-up of the polymer surface. This process, in addition to being used to roughen the surface and increase surface area, can also be used for patterning purposes, both of which encourage cell and biomolecule attachment.76 This method, which has been shown to increase cell attachment while modifying the polymer surface without disturbing the bulk, may be worth investigating for next-generation SMP stents due to the potential for increased biocompatibility.
Plasma- and chemical-based etching occurs when a surface is exposed to etching gas, which is often a type of plasma, and the top layer of the surface is changed through chain scission processes where old bonds are broken and new ones formed; more simply, etching degrades the polymer surface.56- 80- 81 This process also modifies the surface topography and affects surface wettability, potentially driving the surface to become more biocompatible.80- 81
17


Etching can also be performed prior to coating a material, to ensure that the coating adheres.81 Treatment with particular acids, which has an “etching effect”, may also encourage attachment and migration of endothelial cells, especially in polymeric hydrogels.82
Grafting of different length polymer chains can alter the surface roughness, particularly on a nanometer-scale. Roughening at this scale has been shown to enhance cell attachment and improve biocompatibility.83
Transfer printing, a common technique used for patterning, may also be used to roughen the polymer surface. The mold that houses the polymer during curing transfers the roughened features onto the surface during polymerization, as seen in Figure 1.3.84
Although many of these techniques have not yet been applied to SMPs, their use on polymers shows promise for the successful application to SMPs, granted that the methods continue to modify only the topmost layers of the material.
18


Figure 1.3. Atomic force microscopy (AFM) images of (a) untreated, (b) micro-roughened and (c) nano-roughened polydimethylsiloxane (PDMS) films. Reprinted with permission from IOP, 2009.84
Surface Patterning
Surface patterning offers a more organized means of roughening to alter the surface of a material. Patterning may quell non-specific protein-surface interactions, as these effects often lead to device failure.85 Such patterning techniques are often used to enhance endothelial cell
19


attachment, which in turn encourages vessel wall healing and promotes an anti-thrombotic environment.
Nanopillar arrays, formed by plasma processing as shown in Figure 1.4, provide a scaffold for cell proliferation or drug delivery.62 Patterning on metal surfaces, primarily on the nanometer scale, has been shown to promote more endothelial cell attachment compared to random nanopatteming or even patterning on the micron scale.1,86 These nanopattemed surfaces also encourage more endothelial cell attachment compared to smooth cell attachment which is desirable in vessel healing, support greater cell densities on the surface, and even enhance spreading of these endothelial cells.1,87 Cells in their native environment come into contact with features on the nano-scale, which could be the reason for enhanced cell attachment.83,88,89 Some patterning methods strive to mimic native endothelium for a biomimetic effect, in hopes of encouraging more rapid endothelialization and vessel healing, without the presence of plasma proteins or extracellular matrix.1,3> 85,88 Biomimetic patterning may have major implications for SMP stents in that increased biocompatibility can be obtained simply by polymerizing the stent inside of a native blood vessel, directly transferring native endothelial pattern onto the stent surface.
20


Figure 1.4. SEM image of Silicon pillars formed via plasma processing (scale bar = 20 pm). Reprinted with permission from Elsevier, 2001.62
Patterning can also be achieved through diblock copolymer grafts, which form nanometer-sized patterns on solid surfaces. Diblock copolymers can be either physically or chemically attached and form nano-sized domains when they undergo microphase separation. These patterns either encourage or discourage protein adsorption and/or cell adhesion, depending on the polymers involved. For this reason, diblock copolymers have been investigated with regard to surface energy or topography and are being explored for their potential uses in reference to bioactivity.90
Polymers that undergo phase separation such as the mixture of polystyrene and poly(4-bromostyrene) can produce a range of surface topographies just by varying the polymer concentrations and proportions9192 Changes in polymer ratio can yield variations in shape, such as pits, islands and ribbons for example, whereas changing the concentration of the polymer may alter the feature sizes. Cell spreading and proliferation differ based on feature height, with
21


shorter feature heights producing promising results in enhanced cell spreading and proliferation.92
With regard to polymer surface patterning, lithography is one of the more frequently used techniques, a technique common in the electronics field, mainly for patterning of silicon wafers.64 Patterns can include anything from dots and pillars to grooves and ridges, where grooves and ridges are the most studied, often due to the increased tendency of cells to attach and spread along those features.64,93 Lithography may even be used to create hierarchical patterns or tilted patterns, if desired.94 A few theories have attempted to predict why cells prefer to align along grooves and ridges, but different cells have different preferences with regard to size and shape of the formed pattern.64 Photolithography is commonly used on polymer surfaces and this technique selectively exposes surfaces to photoirradiation, creating a pattern on the surface 95-97 This allows for controlled topographical features, directing cell attachment.95,98 Lithographic techniques continue to be a prominent surface modification method for polymers and applying these methods to SMPs, particularly SMP-based stents, may also prove beneficial.
Microfluidic channels offer another means to direct cell adhesion via patterning. Proteins adsorb onto the surface after passing through elastomeric channels in solution form, and once adhered, these proteins, such as fibronectin and collagen, are used for selective cell adhesion.
This method can also be used to produce a patterned cell co-culture, if two different types of cells need to adhere to the same surface 99
Self-Assembled Monolayers (SAMs), a common chemical-based surface modification technique, have also been explored in creating patterns on biomaterial surfaces.64,100 SAMs encourage cell adhesion and orientation, qualities that are advantageous to stent biocompatibility,
22


by controlling protein adsorption onto the surface.101 SAMs are also used for microcontact printing, another method for patterning that is commonly used to encourage cell attachment.62
With regard to SMP-specific patterning techniques, methods in which balls (steel or lime glass) that make indentations on the surface have been explored. Different sized indentations can be made using different sized balls.102 In addition, wrinkling patterns on top of SMPs can be formed using the shape memory capabilities of the polymer itself and if the wrinkling is controlled, a number of surface properties that improve biocompatibility can be manipulated, including roughness, wetting, and bonding among others.103
Transfer printing involves the transfer of a pattern from a mold to a polymer substrate, resulting in a thin patterned film on the surface of the polymer.94 These films are usually polymers themselves and have the potential to encourage cell adhesion by introducing nanoscale patterns that favor cell attachment. Transfer printing can also create surfaces with hydrophobic and hydrophilic characteristics, directing cell attachment to certain areas.104 Zhao et al., determined that microtransfer molding using a PDMS mold creates micron-sized patterns, which may once again increase endothelial cell attachment.105
Similar to transfer printing, stencil-assisted printing involves using a stencil to imprint a desired pattern or structure onto the polymer surface. The patterns develop on the surface that is left uncovered by the stencil, thus directing cell attachment to these exposed areas. This technique does not require any chemical modification after the stencil has been manufactured, making it an appealing method to enhance material biocompatibility as well as a potential technique for surface patterning of SMPs.85
Nanopatterning through dip-pen lithography uses the tip of an atomic force microscope (AFM) to create a pattern on a material’s surface. This tip is dipped into a polymer solution and
23


touched to the surface of a material, altering the chemical makeup of that surface in an organized manner, creating a pattern.106 Depending upon the polymer that is applied to the surface, enhancement in blood compatibility and/or cell attachment can be achieved. The use of a heated tip to create patterns on the surface of SMPs has been explored, and may provide an avenue for patterning SMPs to encourage cell attachment.107 Indentations can also be made using a scanning force microscope (SFM), generally for analytical purposes, but there may be potential for surface modification here as well.108,109
In an effort to physically mimic the patterns found in native vessels, pre-polymer solutions were polymerized inside of a harvested, native blood vessel.94 The polymer adopts the surface features of the blood vessel on its own surface, but the main limitation of this method is that the vessel tissue had to be dissolved to remove the polymer, rendering reproducibility difficult. However, since SMPs acquire their permanent shapes during initial polymerization, this method applied to SMPs may be worth further investigation.
Although SMPs have been treated with only a few patterning techniques, the success associated with patterning polymer materials suggests that patterning SMPs with these techniques may have positive outcomes.
Chemical Modification of the Surface
Chemical modification techniques chemically alter the surface of a material without significantly affecting its bulk properties. Some examples of chemical modifications include chemical vapor deposition (CVD), plasma vapor deposition (PVD), grafting techniques, self-assembled monolayers (SAMs), among others.1,110
24


For metals, many chemical modification techniques, such as plasma immersion ion implantation (PHI) using acetylene, nitrogen or oxygen, are used to reduce corrosion, wear and metal leaching into the surrounding environment and even increase hardness of the material.29 PHI treatment of polymers has been shown to reduce thrombus formation and platelet aggregation by increasing hydrophilicity and protein adsorption onto the surface, as displayed in Figure 1.5.111
Stainless Steel Ion Treated Plasma Polymer
Figure 1.5. Isolated platelets in buffer adhering to both surfaces, but platelets in plasma do not adhere to ion treated polymer surface (scale bar = 20 pm). Reprinted with permission from PNAS, 2011.111
Chemical vapor deposition (CVD) utilizes plasma or other reactive chemicals to deposit thin films onto the surface of the material, slightly altering the surface to allow for deposition of the film.98-112 Due to the non-fouling properties associated with the deposited film, plasma-driven CVD techniques are popular for blood compatibility.112 One form of CVD has been used on stents and other blood contacting devices commercially and goes by the name parylene.
25


Parylene aids in biocompatibility as well as providing a means for sustained drug release from a porous matrix.113 However, this coating does not have functional groups to attach biomolecules, so further treatment with plasma or chemicals to introduce tethering molecules would be required for biomolecule attachment.
Low pressure plasma treatments that use electrons, ions, radicals, metastables or ultraviolet rays (UV) radiation elicit reactions at the surface of polymers.81 For plasma-based treatments, ammonia plasma treatment may encourage cell attachment more through the interaction of acidic groups on the plasma membrane surface and amine/amide groups on the surface of the polymer, which play an important role in endothelial cell adhesion and growth.43, 114 Some studies have shown that cornea cells showed enhanced attachment and growth on plasma-treated surfaces vs. untreated surfaces, and exploration into whether this applies to other types of cells may have merit.81 Studies prepared by Ho et al. found that polymer samples that undergo water vapor plasma treatment may elicit enhanced cell attachment compared to untreated samples, potentially due to the formation of hydroxyl groups on the surface, allowing for hydrogen bond formation between the surface and the cells.115,116
While research has been generally inconclusive about which surfaces are best for supporting cell adhesion and growth, surfaces that are mildly hydrophobic or mildly hydrophilic appear to support optimal cell development; these mild conditions may be achieved through plasma treatment using reaction gases containing organic compounds.64,117-119 Plasma treatment of polymer materials has positive effects on cell adhesion and development on the material surface, mainly by enhancement of hydrophilicity and wettability.70,74,117,120 Plasma deposition may even be used to reduce thrombogenicity 70
26


Plasma vapor deposition (PVD) techniques such as matrix-assisted pulsed laser evaporation, deposit organic and biological materials onto the surface of blood-contacting devices, altering the surface.1 Ionic plasma deposition has been shown to increase endothelial cell adhesion.121 Certain polymer surfaces exposed to N2 and O2 in Helium display enhanced attachment properties, with the extent of surface modification depending upon the polymer surface itself43,80 Other surfaces exposed to nitrogen gases have been known to exhibit reduced thrombotic properties.4,122 As with etching, plasma processes cause the formation of free radicals at the material surface, causing the formation of cross-links.57 These reactive surfaces can be used to encourage coverage with a thin film or can facilitate the attachment of (bio)molecules.
Photografting of polymers using high energy electrons, gamma radiation, ultraviolet (UV) light and visible light can change the surface of polymers to improve blood compatibility and enhance endothelialization.1,43 Bilek el al., found that treatment of a polymer surface with ions to create a free radical surface encourages protein immobilization while retaining protein structure, potentially enhancing biocompatibility.111 Photo-oxidation, a method to introduce hydrophilic groups to polymer surfaces in a controlled manner through the manipulation of photo-oxidation time and grafting time, has also been shown to be beneficial to endothelial cell development on the material surface.64
Chemical grafting methods, such as the grafting of polyethylene glycol (PEG) monoacrylates to the surface of a biomaterial, can reduce the attachment of erythrocytes through steric repulsion, thus decreasing the risk of thrombosis.123,124 PEG is largely hydrophilic and has a large exclusion volume, contributing to this effect and it is also non-toxic and non-immunogenic which are important components of a biocompatible material.123 Grafting copolymerization methods that graft hydrophilic polymers onto hydrophobic surfaces in an effort
27


to neutralize hydrophobicity may encourage cell adhesion.64 Plasma and ultraviolet grafting on polymer surfaces may also promote anti coagulation and antibacterial properties.21
Self-Assembled Monolayers (SAMs) modify the surfaces of materials to enhance hydrophobicity/hydrophilicity or to add reactive or functional groups to the surface. SAMs change the surface energy or wettability of the polymer surface through careful selection of the functional groups used for the monolayer, potentially increasing biocompatibility of the material within the vessel.125 SAMs offer the benefit of ease of fabrication, and the ability to control order and orientation, allowing for the exposure of a select group on the modified surface, creating the ability to cater the biocompatibility of the material to suit specific needs.70
Chemical modification techniques strive to alter the surface of the material in order to enhance the functionality of that material. Polymer substrates undergo exposure to these different techniques, resulting in a material with an improved surface and mostly unchanged bulk properties. If performed properly, these chemical modification techniques can be applied to SMPs allowing for a better surface without affecting the bulk.
Surface Coatings and Films
Surface coatings and films are additional ways to modify surfaces of both metals and polymers in an effort to increase biocompatibility. These techniques often do not involve direct attachment of chemical groups or chemical alteration of the surface the way conventional chemical modification techniques do, but still alter the surfaces for increased biocompatibility. A few coating and film techniques that have been shown to increase endothelial cell attachment or reduce blood coagulation and thrombosis are discussed.
With regard to wet coating/solvent coating of stents, dimethyl sulfoxide (DMSO) has been shown to prevent vascular smooth muscle cell activity on the stent surface, reducing
28


chances for restenosis while also preventing tissue factor activity, thus discouraging thrombosis.41 Studies show that DMSO does not exert toxicity to human vascular endothelial cells, further solidifying this technique as a potentially viable option for polymers and SMPs.126 Dip coating, used to form nanostructures on the surface of medical-grade polymers, creates superhydrophobic surfaces that prevent blood coagulation.127 Coating polymers with polyelectrolyte multilayers provides a good platform for endothelial cells on polymer surfaces.21
Langmuir-Blodgett (LB) films, consisting of highly ordered, densely packed structures of known thickness deposited and crosslinked to the surface of the polymer, also allow for cell adhesion, decreased platelet adhesion and enhanced hemocompatibility.43,128 These LB films can be deposited on polymer surfaces by chemically treating the polymer to attract the LB film and introducing the polymer to a Langmuir-Blodgett trough, allowing a monolayer to form prior to endothelial cell exposure.129 These LB films have not yet been studied extensively on three dimensional scaffolds, but implementing these films on three dimensional structures may be worth further investigation due to the increased biocompatibility offered by this technique. Layer-by-Layer (LbL) polymer films have been shown to reduce platelet adhesion on nitinol, a commonly used stent material.1 LbL deposition of chitosan on the surface of the polymer poly-1-lactide (PLLA), showed improved cell compatibility.130 Studies with diamond-like carbon (DLC) films have also displayed successful attempts to improve blood compatibility on polymer surfaces.25
29


Attachment of Pharmaceuticals, Biopharmaceuticals or Biomolecules to the Surface
The ability to attach a substance to the surface of a material while retaining its bulk properties is an appealing method of delivery for pharmaceuticals or biomolecules.58 Polymers usually have inert surfaces, so in those instances, the surface must be functionalized prior to attaching the bioactive molecule to the surface. As a surface technique, these methods can be applied to either polymers or SMPs, and while most of these techniques have been tested on polymers, there may be benefits to applying these techniques on SMPs. The bioactive compound can be attached through electrostatic interactions, ligand-receptor interactions, or covalent attachment, where covalent linkages are most common as this linkage is often the most stable.58
Chemical Vapor Deposition (CVD) is not only used to enhance biocompatibility, but is also used to create tethering groups on the surface of a polymer for proteins and other biomolecules to attach via covalent bonding.1,112,113 Some of these biomolecules help create a less thrombotic environment by immobilizing on the surface of polymers in the blood vessel. Plasma deposition techniques produce stable films that can aid in corrosion resistance and functionalization sites for the attachment of (bio)pharmaceuticals onto the surfaces of both metals and polymers.1
Wet chemical surface modification methods require chemical reagents to create reactive functional groups on the surface of a polymer, often without expensive equipment or methods, and can be done easily in a laboratory setting. Wet chemicals are able to accomplish deeper penetration of porous surfaces compared to energy-source-based modification techniques, creating a more stable and noncorrosive functionalized surface. If repeatability is desired however, this method may not be the ideal choice, as a wide range of reactive groups are generated, and the orientation of the biomolecule can be crucial for attachment. However, to
30


promote specificity, it may be possible to block some functional groups, allowing for the specific molecule, whether it be a molecule to enhance endothelialization or a protein to reduce thrombus formation, to attach successfully.58
Plasma treatment methods can introduce reactive groups to the surface of a normally inert polymer, allowing for the attachment of a desired bioactive compound. These methods do not require hazardous chemicals, yet still have the capability to modify the surface while imparting less degradation and roughness compared to wet chemical surface modification techniques.58 In addition, the film deposited on the polymer surface can be manipulated by changing the deposition rate, energy range and surface topography.131 Plasma pre-treatment has also been used prior to attaching collagen to a polymer nanofiber mesh, a method that showed increased cell attachment, spreading and viability.132 Thus, if a less corrosive method for biomolecule attachment is required, plasma treatment may be a favorable option, but great care must be taken to avoid contamination of the sample. Once the surface has been functionalized, the desired bioactive agent can be attached for purposes of enhanced cell attachment or thrombosis prevention.
Nitric oxide (NO) or thrombomodulin, both of which are integral to maintaining homeostasis in the blood vessel, can be released from the polymer backbone itself or attached to the surface.43 Gene-eluting stents are capable of delivering biologically active, therapeutic genes in an effort to reduce restenosis, accelerate re-endothelialization and reduce thrombosis. Another set of stents, termed biologically active stents, incorporate antibodies or proteins, such as CD34 antibodies, onto the surface to attract endothelial progenitor cells, or the Arg-Gly-Asp peptide sequence, which also attracts endothelial progenitor cells, speeding up the re-endothelialization
30 133
process. ’
31


In order to mimic naturally occurring conditions in the blood vessel, there has been some work in functionalizing the surface of the polymer with the arginine-glycine-aspartate (RGD) sequence, a protein commonly found in the native extracellular matrix (ECM).134,135 Hegemann et al., determined that this environment promotes endothelial cell attachment and growth.136 Similarly, other cell-adhesion peptides, such as glycine-arginine-glycine-aspartate (GRGD), immobilized on the surface of biomaterials have displayed enhanced endothelial cell attachment.137 Pre-absorbed proteins, such as fibronectin, laminin and gelatin, present on polymer surfaces have been shown to increase cell attachment, but may reduce cell proliferation.138,139. Endothelial cells may be seeded directly onto the material prior to implantation to ensure biocompatibility.128 To combat thrombotic events directly, adding lysine to the surface of a material has been shown to perform clot lysis, preventing blood coagulation.140
Layer-by-Layer (LbL) polymer films have also been used to effectively deliver nitric oxide (NO) donor to the site of vessel injury and can house DNA to be delivered to the vessel wall from the covering.1 This technique deposits both positively and negatively charged biomacromolecules as well, and can even be used for biodegradable polymers, due its mild preparation environment.66
Polypyrrole composites, an electrically-conducting polymer, containing heparin or sodium nitrate, have the ability to switch between oxidized and reduced states, and this switching ability controls the release of biological signaling agents such as growth factors, thus directing cell growth on a surface.141
Biomolecules attached to the surface of polymeric materials in a patterned manner may help control cell behavior or direct cell signaling.142 A range of biomolecules have been
32


immobilized on material surfaces, where the selection of biomolecule(s) is dictated by the nature of cells to be deposited on the surface.
The desire to attach (bio)pharmaceuticals and (bio)molecules to material surfaces is driven by the need provide localized delivery of the molecule or drug without changing the bulk properties of the delivery vehicle. Although many of these techniques require reactive surface groups, further investigation into functionalization of SMPs for (bio)molecule and (bio)pharmaceutical may be desirable, especially for localized drug and molecule delivery. Porous Surfaces to Facilitate Drug Delivery
As mentioned before, stents are commonly used as drug delivery vehicles to stimulate vessel healing and allow for better incorporation of the stent into the body without the use of oral anticoagulant drugs. The drugs used with the stents can be attached directly to the surface of the stent, as was discussed briefly above, or they can be incorporated into the surface of the stent using pores to house the drug until delivery.
Porous stents allow for drug incorporation without an additional polymer coating that is commonly found in drug eluting stents.1 A variety of surface modification techniques have been used to adjust surfaces for the purpose of housing drug for delivery. Etching of the polymer surface for long periods of time may cause pores to form, which can be used to house drug for localized delivery.80,143
The use of photolithography or soft lithography to create pores in polymer sample surfaces or to fabricate porous micro- or nano-particles for embedding onto a polymer surface for drug delivery has also been under investigation.144 Sandblasting has been shown to effectively create porous surfaces on metal stents.1 Aluminum coatings exposed to acidic solutions form ceramic aluminum oxide, resulting in nanoscale pores on that film for drug delivery.1,145 Acidic
33


treatment of stainless steel stents has also been successful in creating porous surfaces for drug elution.145 Stents with a porous hydroxyapatite coating have exhibited promising results for drug elution as well.1,146
A porous surface formed by carbon nanoparticles embedded in polymer has displayed promise as a means for localized drug delivery.147 Similarly, cobalt-chromium alloy stents covered with a porous carbon-carbon coating also showed potential in the arena of drug elution and enhanced cell attachment.147
Drug delivery from pores is not solely limited to the surface; research efforts have also looked into loading drug components into the bulk of SMPs and using shape memory capabilities for drug elution.148,149 This is particularly applicable to SMP stents, since loading a SMP stent with drug components allows for sustained and localized drug delivery.148 Drug release can be controlled by altering the co-monomer ratio, but efforts to maintain the shape memory effect must also be considered.149 Hydrogels, a class of polymer with swelling capabilities, have also been used for slow-release, drug delivery using a diffusion mechanism through pores that often penetrate the bulk material of the hydrogel.150,151 Porous surfaces may allow for localized delivery of a drug or molecule without the need for prior functionalization of the surface. If functionalization for molecular tethering is not an option, forming pores in the surface for drug incorporation may prove beneficial for a range of surfaces, and might soon be extended to SMPs.
34


CHAPTER II
SHAPE MEMORY POLYMERS CONTAINING HIGHER ACRYLATE CONTENT DISPLAY INCREASED ENDOTHELIAL CELL ATTACHMENT - Portions of this chapter were previously published in Polymers, 2017, 9 and are included with the
permission of the copyright holder.
Introduction
Stent materials have a greater chance of survival in vivo if endothelialization of the device occurs soon after device implantation.152 Rapid endothelialization is often characterized as significant cell presence within 24 hours of implantation, with full cellular confluence achieved after 3 to 7 days. 153 Although drug eluting stents (DES) have decreased incidence of restenosis, they typically do not achieve rapid endothelialization, which may limit long term utility.41,154-156 As such, focus has turned to modifying the surface characteristics of stents to promote natural endothelialization. A variety of surface modification techniques including physical, chemical, and biological methods have been evaluated on stent materials including metals and polymers.115,137,139,155,157,158 From a device development and regulatory perspective, it is simpler to utilize the topography rather than using chemical or biological modification; in other words, a simpler regulatory path means simpler manufacturing.
Endothelialization of materials for successful integration of implanted biomedical devices was studied as early as the 1970s.77,159 Endothelialization may occur by binding circulating endothelial progenitor cells or through endothelial cell migration from adjoining endothelium.152, 160 Once endothelial cells attach, usually within the first 24 hours after device implantation, healthy cells proliferate, forming and retaining a permanent endothelial barrier on the surface of the device, resulting in reduced risk of long-term device rejection. Thus, if surface characteristics
35


of the device allow quick recruitment and proliferation of the endothelial lining after implantation, the chances of post-implant problems should decrease.161 Rapid endothelialization also increases hemocompatibility, another imperative for successful integration of a cardiovascular stent.31 The potential for surface modification to enhance biocompatibility has led to increased interest in endothelialization studies, particularly for cardiovascular implants.161
Shape memory polymers (SMPs) are one class of materials that are being considered for use in implanted, blood-contacting devices.18 SMPs are smart materials that recover their original shape upon the application of an external stimulus.60,162-168 These smart plastics are initially fabricated into their permanent shape and then are deformed and fixed into a temporary shape. These materials recover their original shape upon exposure to a stimulus such as heat, light, humidity, electrical or magnetic fields, among others.66,164,168-170 Their ability to recover from large deformations makes SMPs appealing as materials for biomedical devices, since such recovery allows implantation of these devices using minimally invasive techniques.
Progress in SMP research is not limited to biomedical applications. Developments in information carriers for one-time identification, aerospace applications, smart textiles, polymer actuators and sensors, active assembly/disassembly are also notable applications of SMPs.46,166, ni-174 jqjgh anci iow temperature SMPs are being developed for extreme environments, such as jet propulsion and aerospace applications.175,176 Smart textile applications using SMPs range from aesthetic improvements such as appeal, color changing capabilities, soft display to functional applications such as comfort, controlled drug release, wound monitoring, emotion sensing, extreme environment protection, etc.171 SMP actuators may be employed in adjustable rotation rate heat engines or self-regulating sun protectors for buildings.46 Active assembly and
36


disassembly should simplify and automate processing procedures, resulting in high speed, low-cost disassembly, rendering parts useful for additional life cycles.174
Since the shape memory effect, which drives the shape memory capabilities in SMPs, is a result of polymer structure and processing,31,162 prior work in biomaterials has focused on tuning bulk properties to meet the requirements of various medical applications.47,60,163 Previous work from our group has primarily focused on thermomechanical properties such as shape recovery, the shape memory effect (SME), and modulus, as well as cytotoxicity, but the cytocompatibility of these acrylate-based SMPs has not been investigated.18,47,165,177 In addition, there have been very few studies evaluating surface characteristics of SMPs in the context of endothelial growth.
66, 169
The purpose of this study was to examine the relationship between polymer characteristics of a well-studied acrylate-based SMP and endothelial cell attachment and growth. This would represent the first step in evaluating the potential for these materials for bloodcontacting devices. Various compositions of SMP containing different weight percent ratios of tert-butyl acrylate (tBA) and poly(ethylene glycol) dimethacrylate (PEGDMA) were tested for endothelial cell attachment in vitro. The rapid endothelialization target for this study was high live cell presence after 24 hours and complete cell sheet formation 72 hours after cell seeding. Materials and Methods
Shape memory polymers were formulated using tert-butyl acrylate (tBA) and polyethylene glycol) dimethacrylate (PEGDMA) with average molecular weights (Mn) of 550, 750 and 1000 with polymerization facilitated by photoinitiator 2,2 - dimethoxy-2-phenylacetophenone (DMPA). All products were obtained from Sigma-Aldrich (St. Louis, MO, USA), except for PEGDMA 1000, which was obtained from Polysciences (Warrington, PA,
37


USA) and were used as received. A total of nine polymer solutions were prepared from these
monomer components (Figure 2.1).
<
00
CD
c
tf)
CD
CD
i_
O
Figure
Monomer mixtures were injected into molds composed of standard microscope slides (Thermo Fisher Scientific, Waltham MA, USA) separated by a 1.33mm silicone spacer (Mcmaster-Carr, Elmhurst, IL, USA) and cured under ultraviolet (UV) radiation of wavelength=365nm for 20 minutes, similar to previous methods.47 The samples were then removed from the molds and post-cured in an oven at 75°C overnight, similar to methods done previously.18
Samples were post-processed at 75°C in an oven overnight prior to use in characterization or cell attachment studies. Post-processing steps including annealing, which generated consistent physical properties and reduced material defects. Complete conversion of monomers was verified using Fourier transform infrared spectroscopy (FTIR), as has been done previously.165 FTIR samples were fabricated under similar conditions to those described above,
80:20 wt% tBA:PEGDMA550 80:20 wt% tBA:PEGDMA750 80:20 wt% tBA:PEGDMA1000
50:50 wt% tBA:PEGDMA550 50:50 wt% tBA:PEGDMA750 50:50 wt% tBA:PEGDMA1000
20:80 wt% tBA:PEGDMA550 20:80 wt% tBA:PEGDMA750 20:80 wt% tBA:PEGDMA1000
Increasing MW PEGDMA 2.1. SMP formulation matrix of the nine formulations used.
38


but were made thinner, approximately 0.005mm, to allow the IR signal to penetrate the sample for FTIR analysis.
Dynamic Mechanical Analysis (DMA) was performed using a TA Q800 DMA (TA Instruments, New Castle, DE, USA) and was used to verify glass transition temperature (Tg) of the various SMP formulations.178,179 All samples were cut into specimens with dimensions of 20mmx 5mmx 1mm for testing. Each sample was equilibrated to 0°C for 1 minute and heated to 100°C at a rate of 3°C/minute. Testing was conducted at a frequency of 1.0 Hz and cyclic strain control at 0.1% strain.
A Rame-Hart goniometer (Rame-Hart, Succasunna, NJ, USA) was used to obtain contact angle measurements and wettability of each SMP sample.180 Wettability of each formulation was measured by applying water droplets to each surface and measuring the angle that formed between the water droplet and the surface of the sample. Measurements were taken 10 seconds after the 5 pL water droplet was introduced to the surface of the SMP to maintain consistency. Contact angles were measured using DROPImage Advanced computer software (Rame-Hart, Succasunna, NJ, USA). Three different samples were analyzed per SMP formulation. Five drops were applied to each SMP sample surface and at least five measurements were taken per drop.
Surface topography, a measure the surface roughness of each SMP formulation, was obtained using atomic force microscopy (AFM).181 SMP fabrication molds were made using new glass microscope slides as done previously, which were cleaned using detergent and diH20, followed by ethanol and acetone and a final rinse using diH20 to remove any surface artifacts on the glass. These measures were taken to ensure that the surface features detected by the AFM were a result of the changes in weight percent or molecular weight of the PEGDMA. Images were obtained using a NanoSurf easyScan 2 (Nanomaterials Characterization Facility, University
39


of Colorado, Boulder). Image post-processing was completed using Gwyddion open source software (Gwyddion, Brno, Czech Republic). The root mean square roughness coefficient, Rq, measured by the standard deviation of the distribution of surface heights, also obtained from Gwyddion, provided quantitative information of the sample surface.182
Human umbilical vein endothelial cells (HUVECs), obtained from the endothelium of the umbilical vein, are a common cell model for angiogenesis and re-endothelialization studies. HUVECs are also robust, making them a favorable cell type for use in such studies and as a result, HUVECs were the chosen cell model for this re-endothelialization study.161
Prior to cell culture experiments, human umbilical vein endothelial cells (Lonza, Walkersville, MD, USA) were seeded in T-75 flasks using complete growth medium: EBM-2 Cell Culture Bullet Kit (Lonza, Walkersville, MD, USA). HUVECs were maintained in conditions of 37°C and 5% CO2 in a humidified incubator. Cells were washed with HEPES, 1M, buffer (Life Technologies, Carlsbad, CA, USA) prior to changing of the media. Media was changed every two to three days, and cells were passaged at 80-90% confluence. Cell passages two through six were used for cell seeding on SMP substrates. All experiments were conducted in triplicate.
To monitor general cell health and ensure that there were no signs of contamination present, HUVECs and HUVEC-SMP samples were observed under an inverted light microscope (Nikon, Melville, NY, USA) daily.
SMP samples were submerged in growth media and equilibrated to 37°C for 24 hours prior to cell seeding. HUVECs were then plated on 1cm diameter SMP substrates in 24-well plates and allowed to attach. Cells were seeded at a seeding density of 1 x 105 cells/mL per well. Daily monitoring of cell-adherent SMPs using transmission microscopy allowed qualitative
40


assessment of proper cell growth and absence of contamination. Cell viability was quantitatively assessed at two time points, approximately 24 hours after plating and again at approximately 72 hours. Complete cell medium was changed daily to ensure cells received consistent nourishment during the study. The Live/Dead Cell Imaging Kit (488/570) (Life Technologies, Carlsbad, CA, USA) was used to assess endothelial cell attachment and viability through fluorescent staining. Live cells, which were actively attached to the substrate, emit green fluorescence, while dead cells fluoresce red. Images were obtained using an EVOS FL Cell Imaging System (Life Technologies, Carlsbad, CA, USA). At least three images from three replicate experiments were used for cell attachment counting using ImageJ software (NUT).
PrestoBlue® Cell Viability Reagent (Life Technologies, Carlsbad, CA, USA), a plate-based resazurin assay, was added to cell-substrate samples and left to incubate for 2 hours. This cell viability reagent, when added to cells, exploits the reducing power of cells to quantitatively measure cell metabolism. This also provides an indirect measurement of cell proliferation and cytotoxicity.183 Media was removed from samples after the two-hour incubation period and fluorescence was measured on a Synergy 2 microplate reader (BioTek, Winooski, VT, USA).
The data were expressed as mean ± standard deviation. Statistical analysis was performed using MATLAB and significance was determined using ANOVA with a-level of significance set to 0.05. The Tukey’s Honest Significant Difference Test assessed the significance between individual samples if ANOVA determined significance of the sample set.
Results
Select bulk properties of SMP formulations were characterized using storage modulus and tan delta (8) curve data, obtained from dynamic mechanical analysis (DMA).179 The plateaus above and below Tg on the storage modulus curves represent the glassy and rubbery moduli,
41


respectively. As shown in Figures 2A-2C, the glassy regions are absent for most formulations containing 50 weight percent tBA or less. Rubbery modulus increases with increasing PEGDMA content, as shown previously.184
The peak of the tan delta curve was used to determine the activation temperature or glass transition temperature, Tg, which is the temperature at which the material can revert from its temporary shape back to its permanent shape. The onset of shape recovery, TonSet, which is the beginning of the shape recovery transition, as well as the Tg range, was calculated using methods described previously and are displayed in Table 2.1.18 DMA data for some of the samples has been analyzed by our group in previous experiments; our data agreed with prior results.18- 47-179-
184
Storage Modulus (MPa) of
Storage Modulus (MPa) of
tBA:PEGDMA750
20:80 tBA:PEGDMA
---50:50 tBA:PEGDMA
---80:20 tBA:PEGDMA
(U
O)
to
o
CO
Storage Modulus (MPa) of tBA:PEGDMA1000
20:80 wt% tBA:PEGDMA
--50:50 wt% tBA:PEGDMA
---80:20 wt% tBA:PEGDMA
10°-I---------i---------.--------1---------i---*-----.
0 20 40 60 80 100
10O ----'---T----â– ---T---â– ----1
0 20 40 60
100
Temperature (°C)
(B)
Temperature (°C)
(C)
Tan Delta of tBA:PEGDMA550 Tan Delta of tBA:PEGDMA750 Tan Delta of tBA:PEGDMA1000
Figure 2.2. (A) Storage Modulus (MPa) of tBAPEGDMA550 (B) Storage Modulus (MPa) of tBAPEGDMA750 (C) Storage Modulus (MPa) of tBAPEGDMAlOOO. Glassy regions are absent for the formulations containing 50 wt% tBA or less. (D) Tan delta of tBAPEGDMA550. (E) Tan delta of tBAPEGDMA750 (F) Tan delta of tBAPEGDMAlOOO. Tg increases with increasing tBA content and decreases with increasing PEGDMA MW.
42


Increased monomer (tBA) content resulted in increased Tg, so (Tg 80:20 wt% tBA:PEGDMA) > (Tg 50:50 wt% tBA:PEGDMA) > (Tg 20:80 wt% tBA:PEGDMA) for a given crosslinker.
Decreasing the molecular weight of the PEG component in PEGDMA also increased Tg, which indicates that samples containing PEGDMA550 had higher Tg than samples containing PEGDMA750, which in turn had higher Tg than samples containing PEGDMA1000, for a given weight percent ratio (Figure 2D-2F).
Table 2.1: Tg, Tonset and Tg range for SMP Formulations. Data presented as p ± SD.
Formulation Tg Tonset Tg Range
(°C) (°C) (°C)
20:80 tBA:PEGDMA1000 6 ± 2
20:80 tBA:PEGDMA750 11 ± 1 - -
20:80 tBA:PEGDMA550 25 ± 1 15 ± 2 19 ± 5
50:50 tBA:PEGDMA1000 10± 1 8 ± 3 3 ± 5
50:50 tBA:PEGDMA750 19 ± 2 12 ± 1 13 ±4
50:50 tBA:PEGDMA550 44 ± 1 25 ±3 38 ± 7
80:20 tBA:PEGDMA1000 44 ± 1 26 ±3 37 ±4
80:20 tBA:PEGDMA750 52 ± 1 35 ± 1 32 ± 1
80:20 tBA:PEGDMA550 60 ±3 47 ±3 24 ±2
Contact angle increased with increasing tBA content for a given crosslinker (Figures 3A-3C). Thus,
(CA 80:20 wt% tBA:PEGDMA) > (CA 50:50 wt% tBA:PEGDMA) > (CA 20:80 wt% tBA:PEGDMA formulations).
Specifically, water contact angles increased 11%-23% from the 20:80 wt% tBA:PEGDMA formulations to the 80:20 wt% 1BA:PEGDMA formulations and 7%-22% between the 50:50
43


wt% tBA:PEGDMA and the 80:20 wt% tBA:PEGDMA groups. Additionally, wettability decreased with increasing crosslinker length for a given weight percent of crosslinker, i.e. samples containing PEGDMA1000 were more hydrophobic than those containing PEGDMA550.
A Wettability of tBA:PEGDMA550
100
— 80
O
_0>
g 60 <
8 40
c o
° 20 0
SMP Formulation (tBA:PEGDMA)
***________
______»**
84.3
20:80 50:50 80:20
B Wettability of tBA:PEGDMA750
100
— 80
o
0
ra 60 <
ra 40
c
o
° 20
0
SMP Formulation (tBA:PEGDMA)
Q Wettability of tBA:PEGDMA1000
20:80 50:50 80:20
SMP Formulation (tBA:PEGDMA)
Figure 2.3. A. Wettability of tBA:PEGDMA550. B. Wettability of tBA:PEGDMA750, C. Wettability of tBA:PEGDMA1000. Wettability decreases (increasing hydrophobicity) with increasing tB A content and increasing crosslinker (PEGDMA) MW. Significance was determined using one-way ANOVA to determine significant differences between samples of a given crosslinker in addition to the Tukey’s Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p <0.001.
AFM imaging was used to assess the topographical features present on each SMP surface, quantified using the root mean square surface coefficient, Rq. As seen in Figures 2.4A-
44


D, roughness increased with increasing tBA content, so 80:20 wt% tBA:PEGDMA formulations were roughest while the 20:80 wt% tBA:PEGDMA formulations were smoothest for a given crosslinker. Roughness increased 73%-95% between the 80:20 wt% tBA:PEGDMA group and the 20:80 wt% tBA:PEGDMA group and increased 23%-68% from the 50:50 wt% tBA:PEGDMA formulation to the 80:20 wt% tBA:PEGDMA formulation. Additionally, samples containing PEGDMA1000 were rougher than those containing PEGDMA550 for a given weight percent ratio.
20:80 tBA:PEGDMA550 50:50 tBA:PEGDMA550 80:20 tBA:PEGDMA550
SMP Formulation (tBA:PEGDMA) SMP Formulation (tBA:PEGDMA)
45


0 Roughness of tBA:PEGDMA1000
39.02
20:80 50:50 80:20
SMP Formulation (tBA:PEGDMA)
Figure 2.4. A. Representative AFM images of tBA:PEGDMA550 samples. 3D AFM images depict increases in surface roughness as tBA increases. tBA:PEGDMA750 and tBA:PEGDMA1000 AFM images follow a similar trend. B. Root mean square roughness (Rq) of tBA:PEGDMA550. C. Root mean square roughness (Rq) of tBA:PEGDMA750. D. Root mean square roughness (Rq) of tBA :PEGDMA1000. Root mean square roughness (Rq) generally increases with increasing tBA content and increasing PEGDMA MW. Significance was determined using one-way ANOVA to determine significant differences between samples of a given crosslinker in addition to the Tukey’s Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p <0.001.
Cell viability, characterized as endothelial cell attachment on top of the SMP substrate, was monitored using both light and fluorescence microscopy. Results for SMP formulations containing the lowest amount of tBA (20 weight percent) are shown in Figure 2.5. These samples displayed little or no live FfUVEC presence 24 hours after cell seeding, but the presence of dead cells was prevalent indicating that few cells survived after 72 hours.
46


20:80 tBA: PEGDMA550 20:80 tBA: PEGDMA750 20:80 tBA: PEGDMA1000
24 hours after EC seeding
72 hours after EC seeding
Figure 2.5. Live-Dead Analysis of SMP formulations with the lowest weight percent of monomer (20 wt% tBA). These samples show little to no endothelial cell attachment and have a high presence of dead endothelial cells. Scale bar = 400 pm.
SMP formulations containing equal weight percent monomer and crosslinking agent, 50:50 wt% tBA:PEGDMA, displayed the greatest variability in endothelial cell viability. These formulations showed endothelial cell presence 24 hours after HUVEC introduction, but viability and cell attachment decreased 72 hours after cell introduction.
24 hours after EC seeding
72 hours after EC seeding
50:50 tBA: PEGDMA550 50:50 tBA: PEGDMA750 50:50 tBA: PEGDMA1000
47


Figure 2.6. Live-Dead Analysis of SMP formulations with equal weight percent monomer (tBA) and crosslinker (PEGDMA). There are endothelial cells present on the surface of all samples regardless of crosslinker length, but there is some variability based on the crosslinker used in the sample. Specifically, both PEGDMA550 and PEGDMA750 samples seem to support more HUVEC attachment compared to the PEGDMA1000 sample. Scale bar = 400 pm.
SMPs with the highest tBA content, 80 weight percent, showed the highest amount of endothelial cell attachment, displaying 4%-89% greater endothelial cell presence 24 hours after cell introduction and 33%-100% increased cell presence after 72 hours compared to the other formulations. These samples also had the highest ratio of live cells to dead cells.
80:20 tBA: PEGDMA550 80:20 tBA: PEGDMA750 80:20 tBA: PEGDMA1000
24 hours after EC seeding
72 hours after EC seeding
Figure 2.7. Live Dead Analysis of SMP formulations with highest weight percent (80 wt%) monomer (tBA). Endothelial cell attachment is indicated by the high number of living cells and the low number of dead cells present on the samples. Scale bar = 400 pm.
The 80:20 wt% tBA:PEGDMA1000 sample initially displayed less endothelial cell attachment compared to the other formulations with 80 weight percent monomer, but after 72 hours, cell presence increased, an indication of healthy endothelial cells. The 80:20 wt% tBA:PEGDMA750 formulation supported cell attachment 24 hours after HUVEC introduction, and was able to retain most cells after 72 hours. The final sample, 80:20 wt% tBA:PEGDMA550, displayed HUVEC attachment 24 hours after cell seeding, and was able to
48


retain cell attachment 72 hours after initial introduction. All samples containing 80:20 wt% tBA:PEGDMA had few dead cells present, if any.
We found that EC attachment occurred on samples containing at least 50 weight percent tBA, as seen in Figure 2.8A. However, even though these formulations displayed endothelial cell attachment, the samples did not display cell sheet formation after 72 hours. The largest ratio of EC coverage on a sample compared to the tissue culture plate control, which displayed a full endothelial cell sheet after 72 hours, is approximately 0.4, as displayed in Figure 2.8B.
A® Live & Dead Endothelial Cells iI!™!] Present on SMP Surfaces _
CO 120.0k -
C0
^ 100.0k-I
0
O
to
q3
o
"O
c
LU

£
I
£
/ / #

£
£
/ / / /

£
.<§>
&
<0
,9 ,& ,& r9 ,& ,& r9 /.
gs ft if Q? A/ ^ ^ Cs 'C/ Cs Cs ’’VO)
,Q> -O cv V <§>• <§> <£ <§>’
£
<§>’
q°
£
49


B Normalized EC Attachment
Figure 2.8. (A) cell count of HUVECs present on each sample, scaled to size of the SMP sample. Living endothelial cells are present on sample containing at least 50 wt% tBA. (B) endothelial cell count of each SMP sample normalized to endothelial cell count of control sample (TCPS). While ECs attach to SMP surfaces, full coverage of SMP samples is yet to be achieved.
50


Cell metabolism was measured daily for 72 hours, with results displayed in Figure 2.9. Resazurin, which is initially non-fluorescent, is reduced to a fluorescent resorufin when added to healthy cells. Increases or decreases in reduction provide insight into cell health, such as metabolism and cytotoxicity.183 Samples containing 20 wt% tBA did not show any signs of resazurin reduction, further confirming that if any cells were present on the samples, the cells were unhealthy, dying, or already dead. Most of the samples containing 50 wt% tBA and 80 wt% tBA showed signs of increasing resazurin reduction, which may be an indication of increased endothelial cell presence, and consequently, possible cell proliferation.
SJMetabolic Activity of Endothelial Cells £. 1 Attached to SMPs
Figure 2.9. Cytocompatibility of SMPs. There is evidence of increasing metabolic activity, prominently 72 hours after cell introduction, but some samples show evidence of metabolic
51


activity increasing just 48 hours after cell seeding. Samples that are cytotoxic have little metabolic activity compared to samples that are cytocompatible, confirming a lack of EC presence. Significance was determined using one-way ANOVA to determine significant differences between samples of a given wt% ratio and crosslinker combination, in addition to the Tukey’s Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.
Discussion
Current stent technologies have found broad clinical utility but continue to encounter issues such as restenosis and/or thrombosis, both of which may require subsequent reintervention to prevent further complications. While there has been extensive work on fine-tuning bulk mechanical properties ofSMPs and brief studies on cytotoxicity and biocompatibility for applications such as hernia meshes, embolic coils, and stent grafts from our group as well as others, little work has focused on surfaces, and more specifically, cytocompatibility and hemocompatibility of these materials 47,66,177,185,186
Previous work from our group has addressed bulk mechanical properties such as thermomechanical behavior, the shape memory effect, partially constrained and free recovery, biocompatibility and cytotoxicity of these tBA:PEGDMA SMPs. However, previous studies have not reported on surface properties or endothelial cell attachment on the surface of these materials.18,47,165,177,179 Endothelialization of implanted biomedical devices increases the likelihood of device integration due to improvements in hemocompatibilty and reduced risk of device rejection, which necessitates optimization of surfaces to encourage EC attachment.18,31,161
This study evaluated the ability of select acrylate-based shape memory polymers to attach and retain endothelial cells. Numerous studies have shown that interactions between a material’s surface and its surroundings play a notable role in dictating the success of an implanted biomedical device.4,5> 61,128 While previous studies from our lab have assessed bulk mechanical
52


properties of acrylate-based SMPs for stent use, no studies have evaluated surface characteristics of these SMPs in detail.
Our group has previously examined the activation temperature for the SMP formulations used in this study.18,47,178 Briefly, the 80:20 wt% tBA:PEGDMA1000 and the 50:50 wt% tBA:PEGDMA550 samples have glass transition temperatures closest to body temperature, with Tg’s close to 44°C for both formulations. The remaining 80:20 wt% formulations had higher Tg’s, but our group has shown that these formulations are still able to exhibit shape memory at physiological temperature.178 Prior data has shown that varying the crosslinking agent between 10% and 40% does not significantly impact Tg for related crosslinking agents.18 However, when crosslinker content exceeds 40% weight percent, the transition regime between glassy and rubbery becomes blurry and depletes shape memory ability, which may cause the larger differences in Tg displayed here.187 Additionally, glassy modulus and the transition between glassy to rubbery state is nearly non-existent for formulations containing 50 wt% or less, which also demonstrates the depletion of the shape memory ability. Increases in rubbery modulus, and reductions in stiffness, are seen as PEGDMA content increases, agreeing with prior data.18 All of the samples containing 20:80 wt% tBA:PEGDMA as well as the 50:50 wt% tBA:PEGDMA750 and the 50:50 wt% tBA:PEGDMA1000 formulations have considerably lower Tg’s and exhibit breakage when deformed at room temperature due to a higher rubbery modulus and reduced stiffness and thus would not exhibit the shape memory effect at 37°C.
SMP formulations with higher Tg’s are stiffer at physiological temperature. Specifically, the 20:80 wt% tBA:PEGDMA1000 sample, with a glass transition temperature of approximately 6°C, exhibited a storage modulus of 11.27MPa, whereas the 80:20 tBA:PEGDMA550 SMP, with a glass transition temperature of 60°C, had a storage modulus of 1194 MPa. This confirms
53


that the samples with Tg’s lower than 37°C are not as stiff as samples with glass transition temperatures greater than 37°C. The large difference in storage modulus between the two formulations provides additional insight into material properties at 37°C, which could be an important factor in material choice. For a device that is going to serve as a support mechanism, higher stiffness may be more desirable, as would be the case for a cardiovascular stent in a blood vessel.
Numerous studies have cited substrate stiffness as an important factor in determining cell attachment to a substrate.188,189 These studies found that stiffer samples often exhibit higher endothelial cell attachment compared to their softer counterparts.186 Our data affirm these prior results. Stiffer SMPs, or those with activation temperatures above 37°C, displayed greater endothelial cell attachment and viability. Specifically, all of the 80:20 wt% tBA:PEGDMA formulations as well as the 50:50 wt% tBA:PEGDMA550 sample exhibited endothelial cell attachment and retained attached cells for up to 72 hours. Thus, SMPs with Tg slightly higher than body temperature, and therefore increased stiffness, appear well suited for endothelial cell attachment, similar to results found in other studies.188
Contact angle measurements provided quantitative wettability data of each surface.180 Surfaces exhibiting moderate wettability have shown a higher affinity for cell attachment compared to surfaces with extreme wettability.161 Formulations with higher PEGDMA content are more hydrophilic, as are formulations with lower molecular weight PEG chains, i.e. samples containing PEGDMA550. PEGDMA is commonly used in hydrogels for its hydrophilic tendencies as well as its highly tunable material properties, which explains the smaller contact angles for samples containing higher amounts of PEGDMA.190 Since the formulations were based on weight percent ratios of tBA and PEGDMA, the greater hydrophilicity of samples
54


containing PEGDMA550 could be the result of increased PEG presence in the sample compared to a sample containing PEGDMA1000. Despite these trends in wettability, the differences in wettability did not appear significant enough to have a pronounced effect on endothelial cell attachment to these surfaces. This is analogous to results found in other studies where changing PEG length produced large variations in cell attachment.191,192
Surface roughness was also analyzed for each sample. Roughness increased with increasing tB A content and with molecular weight. Rougher surfaces also had higher contact angles, and have been shown to be more conducive to cell attachment, which is consistent with other studies.75,76,83 Even though neither surface roughness nor wettability is a sole deciding factor of HUVEC attachment to the SMP surfaces, both aspects have shown to affect endothelial cell attachment. It is often unclear which factor may play the dominant role in cell attachment, due to the complex nature of cell-surface interactions, as seen in other studies.116
HUVEC attachment was assessed using fluorescence microscopy. Formulations containing the lowest weight percent monomer, 20:80 wt% tBA:PEGDMA, displayed very little cell attachment within 24 hours, leading to minimal or no HUVEC presence 72 hours after cell introduction. Since these formulations have high PEG content, and PEG has been shown to resist protein and cell attachment, the presence of dead cells or lack of cells appears reasonable.144 However, recent studies have also indicated that some PEGDMA-based hydrogels may support cell adhesion of certain cell types, requiring the consideration of these low tBA materials.193 The absence of dead cells between 24 hours and 72 hours can be explained by the tendency of dead endothelial cells to detach because there is no active mechanism for dead cells to remain tethered to the surface of the SMP. The dead cells are then removed when samples are washed with buffer or cell culture medium is replenished.
55


Formulations containing equal amounts of tBA and PEGDMA displayed the greatest amount of variation in endothelial cell attachment. The 50:50 wt% tBA:PEGDMA1000 sample behaved more like formulations containing 20 weight percent tBA, supporting little HUVEC attachment initially and showing a decrease in HUVEC presence after 72 hours, indicating that this formulation may not support endothelial attachment. The 50:50 wt% tBA:PEGDMA750 and the 50:50 wt% tBA:PEGDMA550 samples behaved more similarly to formulations containing 80 weight percent tBA, displaying HUVEC attachment 24 hours after cell seeding and retaining a small number of attached cells after 72 hours.
The SMP formulations containing 80:20 wt% tBA:PEGDMA exhibited the greatest HUVEC attachment. These samples displayed cell attachment and retained endothelial cells 72 hours after initial cell seeding. Some samples even showed indications of increased EC presence, which may indicate cell proliferation. Thus, these formulations may be good candidates for use in implanted devices that require rapid endothelialization to succeed, such as cardiovascular stents. Further work, preferably in vivo, would be required, to confirm that these formulations would be good candidates for stent fabrication.
The SMP samples used in this study were solid surfaces, whereas stents are often perforated tubes, which have been fabricated by our group in previous work.183 Perforated SMP stents are easy and inexpensive to fabricate, unlike metals, and these perforated SMPs may experience greater endothelial cell surface coverage due to migration of ECs from adjoining healthy endothelium in addition to endothelial progenitor cell (EPC) attachment and should verified in subsequent studies. Due to the exploratory scope of this initial study, additional experiments evaluating the effect of surface roughness on endothelial cell attachment and viability were not included but will be conducted in future studies. Protein adsorption to the
56


sample surface from the culture medium has been shown to encourage cell attachment and SMPs that displayed cell attachment may have demonstrated more selective protein adsorption from the cell culture medium, allowing ECs to attach long enough to produce their own adhesion proteins, which should be verified in future work21,69‘116‘194 Finally, in vivo studies would help confirm the in vitro results, increasing interest in further investigating these materials for bloodcontacting devices.
57


CHAPTER III
SMP FORMULATION OPTIMIZATION FOR ENDOTHELIALIZATION: INCREASED ACRYLATE CONTENT IN SHAPE MEMORY POLYMERS
Introduction
We have previously demonstrated successful endothelial cell adhesion on SMPs with at least 50 wt% tBA up to 80 wt% tBA monomer content. The higher tBA content contributed to the increased stiffness, roughness and hydrophobicity of the SMP surface, which appears to contribute favorably to endothelial cell recruitment and retention. As such, in order to determine the optimal SMP formulation for EC attachment, and since previous data suggests that increased tBA in the SMP corresponds to increased cell adhesion, it is necessary to further increase the tBA content in the SMPs beyond 80 wt% and analyze the effect that this increase has on subsequent cell adhesion.195
Based on prior data, endothelial cells seem to prefer high tBA SMP formulations.195 Since the previously analyzed formulations contained maximum 80 weight percent tBA, it would be advantageous to further increase the tBA content in the SMPs in increments of 5 wt%, ranging from 85 wt% tBA to 95 wt% tBA, with PEGDMA comprising the remaining formulation weight. The increased tBA content may contribute to stiffer, rougher, and more hydrophobic surfaces; if the previous trend continues, these materials may provide a more optimal environment for endothelial cell adhesion. The effect of further increasing tBA monomer, not only on the material itself, but also on endothelial cell adhesion and survival, will be investigated in this follow-up study.
58


Materials and Methods
Shape memory polymers were again formulated using tert-butyl acrylate (tBA) and
polyethylene glycol) dimethacrylate (PEGDMA) with average molecular weights (Mn) of 550,
750 and 1000 with polymerization facilitated by photoinitiator 2,2 - dimethoxy-2-
phenylacetophenone (DMPA). All products were used as received. The formulations analyzed
during this part of the study are displayed in Figure 3.1.
<
CD
-s?
I
CD
C
'(/>
CO
o
c
Figure 3.1. High tBA content SMP Formulations. Formulations contain high weight percentages of tBA, ranging from 85-95 wt%.
DMA, AFM, contact angle as well as the cell adhesion study using Live/Dead Imaging
and cell metabolism, were performed similar to methods performed for the original SMP
formulations. Every effort to maintain method consistency was made so that the results could be
verified against original data. The data for the high tBA wt% were compared against those from
the original SMPs to determine the optimal SMP formulation for future studies and use.
95:5 wt% tBA:PEGDMA550 95:5 wt% tBA:PEGDMA750 95:5 wt% tBA:PEGDMA1000
90:10 wt% tBA:PEGDMA550 90:10 wt% tBA:PEGDMA750 90:10 wt% tBA:PEGDMA1000
85:15 wt% tBA:PEGDMA550 85:15 wt% tBA:PEGDMA750 85:15 wt% tBA:PEGDMA1000
Increasing MW PEGDMA
59


Results
The glass transition temperature (Tg) was measured for the additional SMP formulations and is presented in Figure 3.2. Since the increasing increments of tBA are smaller than previously studied, increases in Tg are less pronounced compared to the original formulations, as increases are approximately 10-20% for the high tBA formulations whereas the original formulations experienced increases ranging from 40-80% due to the large increments in tBA or PEGDMA.
Figure 3.2. Glass transition temperature (Tg) continues to increase with increasing tBA content. Significance was determined using one-way ANOVA to determine significant differences between samples of a given wt% ratio and crosslinker combination, in addition to the Tukey’s Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.
60


Notably, the increases in Tg are greater as the MW of PEGDMA also increases, due to the reduced effect of PEGDMA as a result of the reduced weight percent. These results agree with trend that was previously observed for SMPs containing lower tBA amounts.178-179-195
Hydrophobicity/ wettability of the SMPs was measured using the static contact angle method and assessed against the previously analyzed SMPs, as shown in Figure 3.3.
100-
o 00 o m o C>l m o •O o 00 o m o C>l m o •O o 00 o m o C>l m o •O
o o o m o ib" o> o o o m o ib o> o o o m o ib o>
m 00 00 o> m 00 00 o> m 00 00 o>
PEGDMA550 PEGDMA750 PEGDMA1000
SMP Formulation
Figure 3.3. Contact angle of SMPs. Increasing tBA monomer typically increases water contact angle, which indicates hydrophobicity increases as tBA content increases. Significance was determined using one-way ANOVA to determine significant differences between samples of a given wt% ratio and crosslinker combination, in addition to the Tukey’s Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.
61


Overall, the increase in weight percent tBA corresponds to a more hydrophobic SMP surface, which agrees with results obtained previously. Again, the increases in hydrophobicity are subtle, but still present with increasing tBA content.
To further characterize the surfaces of the high tBA SMPs, surface roughness was measured, again, using atomic force microscopy (AFM). Roughness was quantified using the root mean square surface coefficient, Rms which is displayed in Figure 3.4.
Figure 3.4. SMP surface roughness measurements. Increasing the weight percent of tBA has variable effects on surface roughness. There are clear increases in roughness between the highest tBA content SMP tested and the original SMP test samples, but trends are difficult to discern. Significance was determined using one-way ANOVA to determine significant differences between samples of a given wt% ratio and crosslinker combination, in addition to the Tukey’s Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.
62


Increasing the weight percent of tBA has variable effects on surface roughness. There are
generally clear increases in roughness between the highest tBA content SMP (95:5
tBA:PEGDMA) tested and the original SMP test samples (20-80 wt% tBA), but there do not
appear to be any clear trends correlating roughness and tBA content.
HUVECs were introduced to the high tBA SMPs and allowed to attach. Cell adhesion
was monitored for up to 72 hours after cell introduction. Cell viability was again assessed using
the Live/Dead Imaging Kit, where live cells are marked with green fluorescence and dead cells
fluoresce red. Cell viability for all formulations are shown in Figures 3.5-3.7.
85:15 wt% tBA:PEGDMA550 85:15 wt% tBAPEGDMA750 85:15 wt% tBA:PEGDMA1000
24
hours
after
EC
seeding
72
hours
after
EC
seeding
Live/Dead/Nuclei
Figure 3.5. Endothelial cell adhesion on 85:15 wt% tBA:PEGDMA SMPs. ECs appear to demonstrate similar adhesion tendencies to the 80:20 wt% tBA:PEGDMA SMPs.
63


90:10 wt% tBA:PEGDMA550 90:10 wt% tBA:PEGDMA750 90:10 wt% tBA:PEGDMA1000
24
hours
after
EC
seeding
72
hours
after
EC
seeding
Live/Dead/Nuclei
Figure 3.6. Endothelial cell adhesion on 90:10 wt% tBA:PEGDMA SMPs. ECs continue to adhere to the surface.
95:5 wt% tBA:PEGDMA550 95:5 wt% tBA:PEGDMA750 95:5 wt% tBA:PEGDMA1000
24
hours
after
EC
seeding
72
hours
after
EC
seeding
Live/Dead/Nuclei
Figure 3.7. Endothelial cell adhesion on 95:5wt% tBA:PEGDMA SMPs. ECs continue to adhere to the surface, but adhesion appears to decrease on these high tBA SMPs, which may be an indication of reduced surface compatibility.
64


Endothelial cell attachment to the 85:15 wt% tBA:PEGDMA SMPs appears to be similar to the attachment behavior of the ECs on 80:20 wt% tBA:PEGDMA, however as tBA content
increases, the surfaces display fewer adherent cells. This would need to be confirmed quantitatively however, as shown in Figure 3.8. Cell adhesion appears to be maximized on 80-85 wt% tBA:PEGDMA SMPs with less cell adhesion on surfaces with lower (50% tBA or less) or
higher (90% tBA or more).
140.0k
120.0k
CL
CO 100.0k
1/5
^ 80.0k "CD
= 60.0k CD _£Z
-I—'
-g 40.0k
c
LU
20.0k
0.0
Live - 24 Hours
Live - 72 Hours
Dead - 24 Hours
Dead - 72 Hours
L4W
u

I

.4
o
o
o
LLI
CL
<
00
O
oo
o
CM
o o O o o o o o O o o o o o O o o
LO LO O LO LO o LO LO o LO LO o LO LO O LO LO
LO O LO o LO o LO o LO O LO
< < t— < < T— < < T— < < T— < < t— < <
< < < < <
Q Q Q Q Q Q Q Q Q Q Q Q
O O O O O Q o O Q O O Q O O Q O O
LLI LU CD LU LU CD LU LU CD LU LU CD LU LU CD LU LU
CL CL LU CL CL LU CL CL LU CL CL LU CL CL LU CL CL
< < Q_ < < Q_ < < Q_ < < Q_ < < Q_ < <
00 00 < 00 00 < 00 00 < 00 00 < 00 00 < 00 00
m m m m CD
O O -*—* O O O O LO LO O O LO LO
CO CO o LO LO o Cxi Cxi LO •*— o •*— •*— LO LO LO
o o LO o o CM O O ■*“ LO LO ■*“ o o LO CD CD
CM CM o LO LO o 00 00 LO 00 00 o CD CD CD
LO CO 00 CD
c
o
O
W
CL
O
Figure 3.8. Estimated adherent ECs on SMPs. ECs appear to prefer the 80-85 wt% tBA SMPs. EC adhesion on SMPs containing greater than 90% tBA or 50 wt% or less still support cell adhesion, but not to the extent of the “optimal” formulations.
65


50k
Figure 3.9. Cell metabolism of cell-adherent SMPs. Cell metabolism typically increases for a given SMP, which may suggest proliferation.
Cell metabolism was measured daily for 72 hours, as shown in Figure 3.9 and compared to the cell metabolism of the original formulations. Again, there are increases in cell metabolism between day 1 and day 3 for most of the formulations, which is an indication of cell proliferation. It should be noted, however, that statistically significant increases in cell metabolism are less evident for the high tBA SMPs. Additionally, cell metabolism is still significantly lower compared to the “ideal” cell-treated TCPS.
66


Discussion
This study is an extension of a previous study, one that suggested tBA:PEGDMA based SMPs with greater monomer (tBA) content demonstrated improved endothelial cell adhesion capabilities compared to SMPs containing higher PEGDMA content. This work specifically aimed to optimize the tBA:PEGDMA-based SMP formulation for endothelial cell adhesion so that it may be potentially used for cardiovascular stent fabrication, due to the reduced complications that result from cell sheet formation on implanted blood-contacting devices. Thus, by further increasing the tBA content, the effect on cell adhesion should be assessed.
Increasing the tBA content in the SMPs continues to increase the glass transition temperature (Tg) of the SMPs. Even though the increase in tBA is small, there are significant differences in glass transition temperature between, for example the 85:15 wt% tBA:PEGDMA and the 90:10 wt% tBA:PEGDMA samples. Although the Tg is greater, shape memory at physiological temperature still occurs, but at a less rapid rate. The increase in Tg also resulted in increased stiffness which, if the trend follows, should also be favorable to cell adhesion.188,196 Studies by Shalali et al., have shown that endothelial cells prefer stiffer hydrogel scaffolds, by exhibiting elongated, spread out cell morphologies.196 Yeung et. al., found that adhesion to stiffer substrates may upregulate adhesion receptors.188 Therefore, the increased Tg and resulting increased stiffness would present a more favorable environment for cell adhesion.
Contact angle also continues to increase for high tBA SMPs. Similar to the increased Tg, the addition of more tBA creates a more hydrophobic surface. The original tBA samples demonstrated contact angles slightly under 90°, which is representative of a slightly hydrophilic surface, whereas the majority of these high tBA SMPs demonstrate contact angles greater than 90°, representing slightly hydrophobic surfaces. Interestingly, the roughness measurements for
67


these samples did not directly correlate with contact angle measurements. However, this may be due to the non-homogenous polymerization of the SMP, caused by unreacted tBA or tBA reacting with itself. Analysis of additional samples, or analysis using ATR-FTIR, which specifically analyzes surface species, is required.
Unlike contact angle, surface roughness did not appear to display any notable trends. Instead, the surface roughness, Rms of the formulations between 85 and 95 wt% tBA had high variation between formulations and weight percent ratios of a given formulation. This may be a result of the increased tBA amounts, which may have saturated the polymer and affected the polymerization reaction, leading to surface non-homogeneity. Since the contact angle measurements are conducted on a larger scale, these smaller effects may not have been as impactful during contact angle measurements, whereas the roughness measurements are conducted using a very fine tip on the end of the AFM cantilever, which is more sensitive to such variations. This would need to be confirmed, again with ATR-FTIR, XPS, ESCA or another surface species analysis technique.
Based on the cell adhesion studies using Live/Dead to analyze adherent cells, the preferred formulations for the tBA:PEGDMA SMPs are in the 80-85 wt% tBA range. Formulations with 50 wt% or less tBA can support cell adhesion as shown previously, but cell presence is not as prevalent as on other samples, and cells often appear rounded and unhealthy. Additionally, at greater than 90 wt% tBA, cell adhesion is also lower; thus, it appears that the surface environment created by the 15-20 wt% PEGDMA supports cellular adhesion and survival by optimizing stiffness, roughness and wettability.
SMP formulations containing high amounts of monomer or crosslinker appear be suboptimal for EC adhesion and survival. Of the original formulations, which included the 20:80
68


wt% tBA:PEGDMA data, these high PEGDMA formulations did not support EC adhesion or survival due to the toxic nature of the high PEGDMA. It is possible that the high tBA formulations also experienced a similar effect when formulations contained greater than 90 wt% tBA. The optimal nature of the 80 & 85 wt% is again evident when assessing the cell metabolism measurements. Cell metabolism typically increases over the course of the study, but fluctuations become prevalent as tBA increases.
Here, additional SMP formulations containing high weight percent ratios of tBA were assessed for cell adhesion and survival. While these surfaces were often stiffer and more hydrophobic, ECs appeared to prefer formulations containing 80-90 wt% tBA; formulations containing less or more tBA seemed to provide surfaces that were not ideal for cell adhesion and survival. Moving forward, SMPs with the 80:20 1BA:PEGDMA weight percent ratios will be used to further optimize the surfaces to encourage both endothelial cell adhesion and organization.
69


CHAPTER IV
MICROGROOVES ENCOURAGE ENDOTHELIAL CELL ADHESION AND ORGANIZATION ON SHAPE MEMORY POLYMER SURFACES - Portions of this chapter were previously published in ACS Applied Bio Materials, 2019, and are included
with the permission of the copyright holder.
Introduction
Although cardiovascular disease (CVD) is the leading cause of death globally, treatment methods continue to be limited.38,197-199 Initially, surgical intervention was the preferred method for opening occluded vessels, but the invasiveness of surgery and lack of support to arteries frequently resulted in vessel re-occlusion, limiting this approach. Balloon angioplasty and bare metal stents (BMS) provide a less invasive method for opening and supporting narrowed vessels; however, complications such as restenosis and thrombosis from reduced biocompatibility often prompt reintervention in many patients.153,154,156,157 Drug eluting stents (DES) reduce restenosis due to the addition of an anti-proliferative agent, but delayed re-endothelialization and resulting late stage thrombosis continue to be prevalent issues.41,154,155,160 Many of these limitations in current stent technologies are due to material noncompliance and reduced surface patency, which result in sub-optimal patient outcomes. These issues provide continued support for research into improving stent performance.32,137,200
Endothelialization of stents and other blood-contacting devices facilitates proper function and integration with surrounding tissue.157,161 However, reduced compatibility between stents and their respective physiological environments promotes poor endothelial cell recruitment and adhesion. Endothelialization decreases device rejection and eliminates the need for long term anti-platelet therapies due to the generation of an anti-thrombotic and anti-proliferative
70


environment created by healthy endothelial cells.77,157 Current in vitro cell seeding on devices prior to implantation has resulted in reduced thrombosis and improved integration, but these methods may not be feasible in practice due to their laborious and highly specialized nature.32,38,
201 In situ re-endothelialization of implanted medical devices would be the preferred approach.32,
161,201
Mechanical, chemical, and/or topographical alterations to stent materials may promote endothelialization and are thus important considerations in the design of such devices.35,202"204 Chemical, physical and biofunctional surface modification methods have been shown to encourage in situ endothelialization.65,71,94 In this regard, surface patterning may be particularly attractive since it not only promotes cell adhesion, but also encourages cell alignment for a variety of cell types, including endothelial cells, all with far less potential regulatory burden than that required for chemical or biological additions.62,205,206 Grooves and ridges are some of the most common features used to encourage cell attachment and promote cell alignment.207"209 Cell alignment is essential for proper biological and mechanical function in most native tissues, but many biomedical device materials do not contain the structural cues necessary to encourage such organization.210,211 In native blood vessel under healthy flow conditions, endothelial cells align along the direction of flow, whereas in unhealthy flow conditions, ECs demonstrate isotropic orientations, which may result in a more thrombogenic and inflammatory environment.212'214
Shape memory polymers (SMPs) are customizable, smart materials that can recover their original shape following deformation.58,215 For biomedical devices, SMPs have been used in cardiovascular, orthopedic, and dental applications.216"218 Our group has previously developed SMPs for stents, embolic coils, hernia meshes, etc.18,47,165, l77'l79‘195 Prior research on this family of SMPs for stent use has largely focused on the mechanical properties for stent deployment and
71


implantation and to confirm low cytotoxicity.18,165 We recently evaluated endothelial cell adhesion on the surface of these acrylate-based SMPs and found that certain formulations, specifically those containing a higher weight percent ratio of the acrylate monomer, encourage endothelial cells to adhere without requiring surface modification.195 However, cells on these surfaces were randomly oriented; a key next step is to evaluate whether axially oriented topographical surface modifications facilitate endothelial cell alignment.
To study this further, we investigated techniques for micropatteming the surface of these acrylate-based SMPs, and then evaluated the extent of endothelial cell attachment onto patterned and unpattemed surfaces. Since the unpatterned surfaces encourage cell adhesion, micropatteming should further optimize the surface for cell adhesion and survival. Shallow microgrooves were created using 3D metal printed molds, a technique that offers a scalable, cost-effective and practical approach to topographically pattern polymerizable materials. To the best of our knowledge, surface patterning of this family of SMPs for endothelial cell adhesion and organization has not yet been studied. In this investigation, three SMP formulations with varying crosslinker molecular weights were analyzed. These particular formulations, which contained high weight percent ratios of acrylate, previously demonstrated endothelial cell adhesion capabilities, but did not contain the topographical cues needed to encourage alignment. Unpatterned and microgrooved substrate surface properties were assessed using scanning electron microscopy (SEM), atomic force microscopy (AFM) and contact angle measurements. Endothelial cell adhesion and alignment were evaluated for up to seven days using fluorescence microscopy.
72


Materials and Methods
Micropatterned metal plates were fabricated using an EOSINT M270 3D metal printer (EOS, Munich, Germany). The length, width and height of the metal print were specified to match the dimensions of standard microscope slides using SolidWorks (Waltham, MA, EISA). The surface of the metal printed piece consisted of approximately 50-pm-width repeating, shallow grooves. This microgroove pattern is a result of the direct metal laser-sintering method (DMLS) used by the metal printer to manufacture parts. This pattern was subsequently used for micropatteming of the SMP surface via molding.
The components for shape memory polymer fabrication, monomer tert-butyl acrylate (tBA) and crosslinker poly(ethylene glycol) dimethacrylate (PEGDMA) with average molecular weights (Mn) of 550 and 750, were purchased from Sigma-Aldrich (St. Louis, MO, LISA). PEGDMA1000 was obtained from Polysciences (Warrington, PA, USA). Polymerization was facilitated by photoinitiator 2,2 - dimethoxy-2-phenylacetophenone (DMPA), also obtained from Sigma-Aldrich.
Three different monomer mixtures were used for this study (i.e., one mixture for each PEGDMA molecular weight) and all were an 80:20 wt% ratio of tBA:PEGDMA, similar to previous methods195. Monomer mixtures were injected into molds composed of a standard microscope slide (Thermo Fisher Scientific, Waltham MA, USA) and a metal plate, separated by a 1.33mm silicone spacer (Mcmaster-Carr, Elmhurst, IL, USA). The unpatterned SMP surfaces were created with a polished steel plate, while the metal printed piece was used for the microgrooved SMPs. The pre-polymers were then cured under ultraviolet (UV) radiation from a Dymax Model 200 Light Curing System (Dymax, Torrington, CT, USA) of wavelength 365nm for 10 minutes pulsed (30 seconds on, 30 seconds off) followed by 10 minutes uninterrupted
73


curing. The samples were then removed from the molds at 90°C and post-cured in an oven at 75°C overnight, similar to previous methods.195
Dynamic mechanical analysis (DMA) was then conducted on a Q100 DMA (TA Instruments, New Castle, DE, USA) to ensure that the thermomechanical bulk properties of the SMPs remained consistent between unpatterned and microgrooved samples. Three SMP samples per formulation and surface condition were cut into 20mm x 5mm x 1mm specimens, equilibrated to 0°C and ramped to 100°C at a rate of 3°C/min, as performed previously.178
The surfaces of the unpattemed and microgrooved SMPs were imaged using a JEOL ASM 6010LV (JEOL USA, Peabody, MA, USA). Prior to SEM imaging, the SMPs were sputter coated for 30 seconds using a Leica EM ACE200 (Leica Microsystems, IL, USA; EM Laboratory, Children’s Hospital Colorado). SEM micrographs were used to ensure that the unpatterned surfaces did not exhibit any organized surface features and to verify micropatterning and groove width of the microgrooved SMP surfaces.
Surface topography and surface roughness were obtained using atomic force microscopy (AFM). Five different samples per formulation were cleaned with ethanol and air-dried to remove any debris prior to imaging. Topographical data and images were obtained using a JPK AFM system (JPK, Berlin, Germany). Image post-processing was completed using Gwyddion open source software (Gwyddion, Brno, Czech Republic). The root mean square roughness coefficient, Rms, which provides quantitative information of the sample surface. Rms, measured by the standard deviation of the distribution of surface heights of the sample, was also obtained from Gwyddion.219
Contact angle measurements and wettability of SMP samples were obtained using a Kudos Precision Instruments DropMeter A60 (Manhattan, NY, USA). Wettability of each
74


formulation was measured by applying 10 pL water droplets to each surface and measuring the angle formed between the water droplet and the surface of the sample. Measurements were taken ten seconds after the water droplet was introduced to the surface of the SMP to maintain consistency. Contact angles were measured using SurfaceMeter Elements computer software (NBSI, Ningbo City, China). Five different samples were analyzed per surface, per SMP formulation. Five drops were applied to each SMP sample surface and five measurements were taken per drop.
Obtained from the endothelium of the umbilical vein, human umbilical vein endothelial cells (HUVECs) are a common cell model for angiogenesis and re-endothelialization studies. HUVECs were the chosen cell model for this re-endothelialization study because they are robust, which makes them a favorable cell type for use in such studies.161
Prior to cell culture experiments, HUVECs (Lonza, Walkersville, MD, USA) were seeded in T-75 flasks using complete growth medium: EGM-2 Cell Culture Bullet Kit (Lonza, Walkersville, MD, USA). HUVECs were maintained in standard cell culture conditions of 37°C and 5% CO2 in a humidified incubator. HUVECs and HUVEC-SMP samples were observed daily under a Nikon inverted light microscope (Nikon, Melville, NY, USA). Cells were washed with HEPES, 1M, buffer (Life Technologies, Carlsbad, CA, USA) prior to changing media. Media was changed every other day, and cells were passaged at 80-90% confluence. Cell passages four through seven were used for cell seeding on SMP substrates. At least three independent experiments were performed, and all experiments were conducted in triplicate.
SMP substrates were submerged in ethanol, air-dried and subsequently steam sterilized prior to HUVEC seeding. Steam sterilization has been previously used to sterilize acrylate-based SMPs successfully without disrupting shape memory capabilities or other material properties.165.
75


178 HUVECs were then plated at a seeding density 1 x 105 cells/mL per well on 10mm diameter SMP substrates in coverslip bottom 24-well plates and allowed to attach. Cell-adherent SMPs were monitored daily for proper cell growth and absence of contamination using transmission microscopy.
Cell viability was quantitatively assessed at three time points: 1 day, 3 days and 7 days after cell introduction. Endothelial cell attachment and viability was assessed using the Live/Dead Cell Imaging Kit (488/570) (Life Technologies, Carlsbad, CA, USA). Live cells, which were actively attached to the substrate, emit green fluorescence, while dead cells emit red fluorescence. Complete cell medium was changed every other day during the study. Images were obtained using a Zeiss Axiovert.Al inverted microscope (Zeiss, Thornwood, New York, USA). At least five images from three replicate experiments were used for cell attachment counting using ImageJ software (NIH, Bethesda, MD, USA).
After 1 and 7 days of sub-culture, cell-attached SMP samples were fixed using 4% paraformaldehyde in PBS for 10 minutes. Fixed samples were then submerged in 0.1% Triton-X for permeabilization prior to staining with Anti-CD31 (Abeam, Cambridge, MA, USA), which was used per manufacturer instructions. In addition to Anti-CD31, phalloidin and DAPI were also used according to manufacturer instructions. DAPI (4',6-Diamidino-2-Phenylindole, Dihydrochloride), (MilliporeSigma, Burlington, MA, USA) was used to visualize nucleus alignment, which was measured as 0° if the nucleus position was perpendicular to the groove while nuclei parallel to the groove were measured at 90°. The alignment angle was measured using ImageJ software (NIH, Bethesda, MD, USA). Filamentous F-actin was visualized using Alexa 568 Phalloidin Actin Stain (Thermo Fisher Scientific, Waltham, MA, USA) and
76


confirmed cell alignment. Fixed cell imaging was performed on a Zeiss AxioObserver inverted microscope (Zeiss, Thomwood, New York, USA).
The data were expressed as mean ± standard deviation (p ± SD), unless otherwise noted. Statistical analysis was performed using MATLAB (MathWorks, Natick, MA, USA) and significance was determined using a two-tailed t-test with a-level of significance of 0.05 when comparing unpatterned vs. micropattemed groups. When comparing more than two groups, analysis was conducted using a two-way ANOVA and the Tukey’s Honest Significant Difference Test assessed the significance between individual samples if ANOVA determined significance of the sample set.
Results
Dynamic mechanical analysis (DMA) confirmed the consistency of bulk properties, specifically the glass transition temperature as measured by the peak of the tan delta as well as the storage modulus, between unpatterned and micropattemed SMPs, similar to prior results (Figure 4.1).179-195 Data confirmed that microgroove introduction does not significantly affect Tg,
Tonset or Tg range, as demonstrated in Table 4.1.
80:20 tBA:PEGDMA550 80:20 tBA:PEGDMA750
80:20 tBA:PEGDMA1000
Temperature (°C)
------Unpattemed Storage Modulus (MPa) ---------Microgrooved Storage Modulus (MPa)
------Unpattemed Tan Delta ------Microgrooved Tan Delta
Figure 4.1. Storage modulus and tan delta curves for 80:20 wt% unpattemed and microgrooved SMPs. Both storage modulus and tan delta display overlap between unpatterned and microgrooved surfaces.
77


Table 4.1. Tg, Tonset and Tgrange (n = 3), for unpatterned and microgrooved 80:20 wt% tBA:PEGDMA550, 750, 1000 SMPs.
SMP Surface (X) (X) Tq Range (X)
550 Unpatterned 59 ± 1 48 ±2 21 ±3
550 Microgrooved 58 ±1 49 ±1 18 ± 1
750 Un patterned 52 ±1 37 ±2 30 ±3
750 Microgrooved 50 ±2 37 ±1 26 ±3
1000 Unpatterned 44 ±1 28 ±2 32 ±5
1000 Microgrooved 44 ±1 28 ±1 32 ± 1
Scanning electron micrographs qualitatively depicted pattern transfer to the surface of micropattemed SMPs as well as the lack of periodic, pattern-like surface features on unpatterned surfaces. As shown in Figure 4.2, the unpattemed surfaces lack repetitive surface features while the patterned surfaces exhibit repeating shallow microgrooves, confirming pattern transfer from the mold to the SMP surface. The microgroove widths, which ranged from 55pm to 60pm, were measured using ImageJ (NIH, Bethesda, MD, USA).
78


80:20 tBA:PEGDMA550
Unpatterned Microgrooved
Width: 60 |jm ± 6 Depth: 8 pm ± 2
80:20 tBA:PEGDMA750
Unpatterned Microgrooved
80:20 tBA:PEGDMA1000
Unpatterned Microgrooved
Width: 61 pm ± 10| Depth: 9 pm ± 2
Width: 56 pm ± 8 Depth: 9 pm dfc 2
Scale Bar: 10Qpm
Figure 4.2. SEM micrographs verified pattern transfer and surface feature presence, or lack thereof, on unpatterned and micropattemed SMP surfaces. Unpattemed surfaces exhibit topographical randomness. Micropattemed surfaces display shallow grooves with widths ranging from approximately 55 to 60 pm. Scale Bar = 100 pm.
Atomic Force Microscopy (AFM) was used to quantify the topography of each SMP surface, by measurement of surface roughness.121 The root mean square surface coefficient (Rms), which provides a quantitative measure of surface roughness, was obtained from AFM data. Micropattemed surfaces exhibited 11-14% higher roughness compared to unpatterned SMP surfaces, but surface roughness did not demonstrate statistically significant differences between unpatterned SMPs and their microgrooved analogues (Figure 4.3). The roughness of unpatterned
SMP samples, polymerized in glass molds, has been previously reported by our group and similar results for unpatterned surfaces were confirmed.195
79


600-
3 500 "I

i 4o°h


0
U)
3
O
300-
200-100-
] Unpatterned |||||||||||||||| Microgrooved
tBA:PEGDMA550 tBA:PEGDMA750 tBA:PEGDMA1000
Figure 4.3. Roughness of unpatterned vs. microgrooved 80:20 wt% tBA:PEGDMA SMP surfaces. Microgrooved surfaces exhibit 11%-14% increased roughness compared to unpatterned SMP surfaces. A 2-tailed t-test was used to determine significance between unpatterned and microgrooved surfaces, n = 5. While there is an increase in roughness between unpatterned and microgrooved surfaces, the differences are not statistically significant. Topographical views of unpatterned and microgrooved tBA:PEGDMA550, tBA:PEGDMA750 and tBA:PEGDMA1000 surfaces further confirm pattern transfer to microgrooved surfaces and lack of patterning on unpatterned surfaces.
80


Surface wettability was determined by measuring the contact angles formed between a 1OpL drop of diH20 and the SMP surface. Unpattemed SMP surfaces consistently exhibited contact angles close to 90°, indicating slightly hydrophobic surfaces. Micropatterning the SMP surfaces decreased wettability by 3-6%, as seen in other studies using different materials.220 Notably, significant differences between wettability of unpatterned vs microgrooved surfaces is confirmed and are shown in Figure 4.4.
Figure 4.4. Wettability of unpatterned vs. microgrooved SMP surfaces. Hydrophobicity is 3%-6% greater for microgrooved surfaces, indicating a small increase in hydrophobicity between unpatterned and microgrooved surfaces. A 2-tailed t-test determined significance between unpatterned and microgrooved surfaces, *** corresponds to p < 0.001. n=5.
Endothelial cell attachment and viability was assessed using Live/Dead imaging
fluorescence microscopy. Live cells are displayed in green, whereas dead cells fluoresce red.
NucBlue Live Cell ReadyProbes was used to mark cell nuclei as indicated in blue (Figure 4.5).
81


tBA:PEGDMA550 tBA:PEGDMA750 tBA:PEGDMA1000


Figure 4.5. Endothelial cell attachment at 1 day, 3 days and 7 days after cell introduction assessed using Live/Dead Cell Imaging and NucBlue Live Cell Stain to mark nuclei. Microgrooved SMPs exhibited greater cell attachment compared to unpattemed surfaces. There is also some evidence of cell organization in the direction of the grooves. Scale Bar: 200pm. Note: some of these SMPs, specifically the formulations containing PEGDMA750 & 1000, occasionally absorb the NucBlue stain; this is particularly evident on microgrooved surfaces, near the edges of the grooves.
Although all SMPs and topographies demonstrate endothelial cell adhesion and increases in living cells during the study, endothelial cell attachment is greater on micropattemed surfaces compared to their unpatterned analogues, for all SMP formulations, at all timepoints, after endothelial cell seeding, as shown in Figure 6A. The 80:20 wt% tBA:PEGDMA550 microgrooved surfaces displayed the highest endothelial cell adhesion, which was measured by counting live, surface-adherent cells, as displayed in Figure 6. It should be noted, however, that the 80:20 wt% tBA:PEGDMA1000 microgrooved surface consistently experienced the greatest increases in cell presence compared to its unpatterned analogue. Occasionally, increases in dead cells on micropattemed vs. unpatterned surfaces are present, but the percentage of dead cells on microgrooved surfaces remains equal to or lower than their unpattemed counterparts. Long-term cell cultures, up to seven days, depict similar trends as microgrooved surfaces and continue to show higher endothelial cell presence compared to unpattemed surfaces In addition to cell adhesion, live/dead data was also used to assess cell proliferation. Cell proliferation was estimated as the increase in cell presence on each surface relative to the initial day 1 cell counts and displayed as percentage increase in adherent cells (Figure 4.6B). Although the unpattemed surfaces initially demonstrated less cell adhesion, these same unpatterned SMP surfaces typically demonstrated greater increases in cell presence on day 3 and day 7 compared
83


to their microgrooved analogues. However, microgrooved surfaces ultimately demonstrated the
greatest overall cell adhesion compared to their respective unpatterned analogues.
£
E
m
ll
w
b
CL
5
CO
W
o
LU
c
a)
a>
â– a
<
200k i
Unpatterned Live Microgrooved Live Unpattemed DeadBI Microgrooved Dead
84


Figure 4.6. A) Approximate cell presence, both live and dead, on all unpatterned vs. microgrooved SMPs. Microgrooved surfaces often demonstrate statistically significant increases in cell presence compared to unpattemed SMP surfaces. Dead endothelial cell presence is typically comparable between unpatterned and microgrooved surfaces. B) Percentage increase in EC presence on unpatterned and microgrooved SMPs. Unpatterned surfaces demonstrate a higher percentage increase in cells, but cell presence remains highest on microgrooved surfaces compared to unpatterned ones. Significance between unpatterned and microgrooved surfaces was determined using a 2-tailed, unpaired t-test, * corresponds to p < 0.05, ** corresponds to p <
0.01, *** corresponds to p < 0.001. Significance between SMPs of varying crosslinker lengths and unpattemed and microgrooved surfaces was determined using a 2-way ANOVA, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.
To simplify comparison of the means of the unpatterned vs. microgrooved surfaces, specifically the groups that appear to be statistically similar, but demonstrate statistically significant differences, cell adhesion and viability was also presented as data ± standard error (p ± SEM), as shown in Figure 4.7.
85


Figure 4.7. Approximate adherent ECs/SMP, live and dead, on all unpattemed vs. microgrooved SMPs. Statistical significances are visually clearer with data presented as p ± SEM, since the focus is on the precision of the mean rather than the spread of the data. Significance between unpatterned and microgrooved surfaces was determined using a 2-tailed, unpaired t-test, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. Significance between SMPs of varying crosslinker lengths and unpattemed and microgrooved surfaces was determined using a 2-way ANOVA, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.
86


Un patterned
Microgrooved
200k-,

***
***
150k
100k
50k
0
Figure 4.8. Approximate adherent ECs/SMP, live only, on all unpatterned vs. microgrooved SMPs. Significance between unpatterned and microgrooved surfaces was determined using a 2-tailed, unpaired t-test, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. Significance between SMPs of varying crosslinker lengths and unpattemed and microgrooved surfaces was determined using a 2-way ANOVA, * corresponds to p < 0.05, ** corresponds top<0.01, *** corresponds to p < 0.001
Endothelial cell alignment was assessed by measuring nuclear orientation and alignment
angle relative to the groove, as displayed in Figure 4.9. For cells adhering to microgrooved surfaces, nucleus orientation angles of 90° indicate cellular position parallel to the direction of
the groove. Endothelial cells are more randomly oriented on unpattemed surfaces, whereas on
micropattemed surfaces, ECs orient themselves in the direction of the groove, depicted by
increases in the percentage of cells for orientation angles 80°-90°. Cells on unpattemed surfaces
87


Full Text

PAGE 1

E NDOTHELIAL CELL ADHESION ON ACRYLATE BASED SHAPE MEMORY POLYMERS FOR USE IN CARDIOVASCULAR STENTS By TINA GOVINDARAJAN B.S. , University of Colorado at Boulder , 2012 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Bioengineering Program 2019

PAGE 2

ii T his thesis for the Doctor of Philosophy degree by Tina Govindarajan has been approved for the Bioengineering Program b y Daewon Park, Chair Robin Shandas, Advisor Richard Benninger Luisa Mestroni Jeffrey Jacot Brisa Peña Date: May 18, 2019

PAGE 3

iii Govindarajan , Tina (PhD, Bioengineering) Endothelial Cell Adhesion on Acrylate Based Shape Memory Polymers f or u se i n Cardiovascular Stents Thes is directed by Professor Robin Shandas ABSTRACT Although cardiovascular disease is the leading cause of death worldwide, current treatmen t methods continue to be limited. Balloon angioplasty , bare metal stents (BMS) and drug eluting stents (DES) provide minimally invasive method s for opening and supporting narrowed vessels, but complications such as restenosis and thrombosis from reduced biocompatibility resulted in vessel re occlusion, limit ing this approach. Shape memory polymers (SMPs) have shown promise as polymer stents due t o their self deployment capabilities and vascular biocompatibility. Prior research on SMPs for stent use has focused on the mechanical properties for stent deployment and implantation and to confirm low cytotoxicity. However , both in vitro studies and in vivo studies using animal models, ha ve demonstrated that endothelialization of a device surface soon after implantation enhances the likelihood of device integration , leading to device success . Thu s, to make SMPs a more viable option for cardiovascular stents, verif y ing the endothelial cell recruitment ability of the surfac e is an important step . This work aims to optimize the surface of a shape memory polymer previously developed in our group to encourage endothelialization for future us e in cardiovascular stents. First, we optimiz ed the SMP formulation to retain its shape memory properties while also encouraging endothelial cell adhesion and survival . T he 80:20 weight percent ratio tBA:PEGDMA SMP was

PAGE 4

iv selected for further experiments , based on previous thermomechanical data as well as acquired endothelialization data. In vivo , healthy vascular ECs are typically elongated and aligned in the direction of flow ; as a result , many studies involving implanted blood contacting dev ices contain topographical cues that direct cells to adhere in an organized fashion. T opographical surface modification s provide a less resistive regulatory path compared to those of the biological or chemical variety. G rooved surfaces have demonstrated cell adhesion, alignment, and improved cell function on a variety of surfaces, including metal and polymer ; as a result, microgrooves were introduced to the surface of the SMPs using metal printed molds . Metal printing offers a simple, cost ef fective, reproducible , and robust method for mold fabrication that ca n be used to fabricate molds with surface features on the order of tens of microns. Microgrooved SMPs demonstrated increased cell adhesion, survival and alignment compared to their unpatt erned analogues. To further optimize rapid cell adhesion, groove depth was increased, and shape memory was extended to the surface features. Through initial compress ion of the grooved SMP surface and subsequent surface recover y at physiological temperature after cell introduction, cell s attach to seemingly flat surfaces prior to feature recovery. which increas es cell adhesion . The microgrooves also continu e to encourage cell alignment. Th ese studies provide some preliminary data that may aid in the future use of these materials for cardiovascular stents . The form and content of this abstract are approved. I recom mend its publication. Approved: Robin Shandas

PAGE 5

v A C KNOWLEDGEMENTS I would like to express my sincere gratitude to many individuals who have contributed to my personal and professional development . First, I would like to express my heartfelt appreciation to my advisor, Dr . Robin Shandas, for his unwavering support of my training and research , as we ll as for h is patience, motivation and encouragement and most importantly , for always pushing me to be a better scientist . I would also like to extend my thank s to my thesis committee: Dr. Daewon Park, Dr. Richard Benninger, Dr. Luisa Mestroni, Dr. Jeffrey Jacot, and Dr. Brisa Pe ñ a for their insight and support over the course of the dissertation . I am indebted to all of you . Additionally, I am very grateful to current/former members of the Shandas Lab, specifically Jennifer Wagner, Kiran Dyamenahalli, Roopali Shah and Michael Zimkowski , who provided much of the foundational training that ma de this work possible. An additional shout out to Jennifer Wagner and Brisa Peña I truly believe that I would not have made it this far without their support and assistance. There have also been many individuals who have assisted me with equipment training/usage, specifically Steven Lewis, Eric Wartchow , the folks at the Nano C haracterization Facility (NCF) at CU Boulder , Stephen Huddle, and of course , the Bioengineering Department at CU Denver|Anschutz . Thank you all. To my friends, thank you for encouraging me, letting me vent when I needed to , and pushing me to be the best person I can be , in all aspects . To my family, my mother and younger br other, who continue to love , support , and lift me up, even on days when I may make it difficult , I cannot thank you enough . F inally, to my late father, who taught me more about life than h e, or I, ever thought possible , thank you . I would not be the person I am today without you all and words cannot express how sincerely grateful I am for your presence in my life .

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ .............. 1 Cardiovascular Disease ................................ ................................ ................................ ....... 1 Current Stents and Limitations ................................ ................................ .......................... 3 Endothelialization ................................ ................................ ................................ .............. 6 Polymers ................................ ................................ ................................ ............................ 8 Shape Memory Polymers ................................ ................................ ................................ ... 9 Polymer & Shape Memory Polymer Stents ................................ ................................ ..... 1 3 Surface Modification to Increase Biocompatibility ................................ .......................... 13 Methods for Surface Modification ................................ ................................ .................... 14 Surface Roughening ................................ ................................ ................................ .......... 1 5 Surface Patterning ................................ ................................ ................................ ............. 1 9 Chemical Modification of the Surface ................................ ................................ .............. 2 4 Surface Coatings and Films ................................ ................................ .............................. 2 8 Attachment of Pharmaceuticals, Biopharmaceuticals or Biomolecules to the Surface ................................ ................................ ................................ .................... 30 Porous Surfaces to Facilitate Drug Delivery ................................ ................................ .... 3 3 II. SHAPE MEMORY POLYMERS CONTAINING HIGHER ACRYLATE CONTENT DISPLAY INCREASED ENDOTHELIAL CELL ATTACHMENT .......... 3 5 Introduction ................................ ................................ ................................ ....................... 3 5 Materials and Methods ................................ ................................ ................................ ...... 3 7 Results ................................ ................................ ................................ ............................... 4 1

PAGE 7

vii Discussion ................................ ................................ ................................ .......................... 5 2 III. SMP FORMULATION OPTIMIZATION FOR ENDOTHELIALIZATION: INCREASED ACRYLATE CONTENT IN SHAPE MEMORY POLYMER S .............. 5 8 Introduction ................................ ................................ ................................ ....................... 5 8 Materials and Methods ................................ ................................ ................................ ...... 5 9 Results ................................ ................................ ................................ ............................... 60 Discussion ................................ ................................ ................................ ......................... 6 7 IV. MICROGROOVES ENCOURAGE ENDOTHELIAL CELL ADHESION AND ORGANIZATION ON SHAPE MEMORY POLYMER SURFACES ................. 70 Introduction ................................ ................................ ................................ ....................... 70 Materials and Methods ................................ ................................ ................................ ...... 7 3 Results ................................ ................................ ................................ ............................... 7 7 Discussion ................................ ................................ ................................ ......................... 91 V. THERMOELASTIC RECOVERY OF MACROSCALE SHAPE MEMORY POLYMER SURFACE FEATURES ................................ ................................ ............... 9 7 Introduction ................................ ................................ ................................ ....................... 9 7 Materials and Methods ................................ ................................ ................................ .... 100 Results ................................ ................................ ................................ ............................. 10 3 Discussion ................................ ................................ ................................ ....................... 11 3 VI. TEMPERATURE ACTIVATED MICROGROOVES IMPROVE ENDOTHELIAL CELL ADHESION AND ALIGNMENT ON SHAPE MEMORY POLYMER SURFACES ................................ ................................ ............. 11 9 I ntroduction ................................ ................................ ................................ ..................... 11 9

PAGE 8

viii Materials and Methods ................................ ................................ ................................ .... 1 2 2 Results ................................ ................................ ................................ ............................. 12 5 Discussion ................................ ................................ ................................ ....................... 13 6 VII. CONCLUSIONS, LIMITATIONS AND FUTURE WORK ................................ ......... 1 40 Conclusions ................................ ................................ ................................ ..................... 1 40 Limitations ................................ ................................ ................................ ...................... 14 3 Future Work ................................ ................................ ................................ .................... 14 5 REFERENCES ................................ ................................ ................................ ............................ 1 5 2 APPENDIX A. FTIR of Original SMP Formulations ................................ ................................ .............. 1 7 6 B. CellEvent Caspase 3/7 Green for Apoptosis of Original Formulations .......................... 1 7 8 C. E xtended Live/Dead Study of Select SMPs Day 7 ................................ ....................... 1 7 9 D. Scanning Electron Microscopy (SEM) of Cell Adherent SMP ................................ ...... 1 81 E. D31 Staining of Microgrooved and Unpatterned SMPs ................................ .................. 1 8 2

PAGE 9

ix LIST OF TABLES TABLE 2.1. Tg, Tonset and Tg range for SMP Formulation s ................................ ................... 4 3 4.1 T g , T onset and T g range for unpatterned and microgrooved 80:20 wt% tBA:PEGDMA550, 750, 1000 SMPs ................................ ................................ ... 7 8 5.1. SMP Formulations used for this study. Three different weight percent ratios were tested, with 3 variations of PEGDMA MW, for 9 total formulations. ................................ ................................ ................................ ........ 101 5.2 . Met hods matrix for compression & recovery of SMP surface protrusion . ................................ ................................ ................................ ............ 10 3 5. 3 Glass transition temperatures and compression and recovery ranges of select SMPs investigated . ................................ ................................ ..................... 10 4 6.1 Glass Transition Temperature, Compression Temperature and Microgro ove Recovery Rate ................................ ................................ ................ 1 2 8 7.1. Glass Transition Temperature, Compression Temperature and Microgroove Recovery Rate of 80:20 tBA:PEGDMA550 SMPs with microgroove widths of 50µm, 100 µm & 150 µm ................................ ............... 14 7

PAGE 10

x LIST OF FIGURES FIGURE 1.1 Surface modification techniques commonly used to enhance endothelialization and/or reduce thrombosis ................................ ................................ ................................ ... 15 1.2 Scanning electron microscope (SEM) image of RIE textured silicon surface using plasma consisting of Cl2, CF4 and O2 gases ................................ .............. 1 7 1.3 Atomic force microscopy (AFM) images of (a) untreated, (b) micro roughened and (c) nano roughened polydimethylsiloxane (PDM S) films ................................ .......... 1 9 1. 4 SEM image of Silicon pillars formed via plasma processing ................................ ............ 2 1 1 .5 Isolated platelets in buffer adhering to both surfaces, but platelets in plasma do not adhere to ion treated polymer surface ................................ ................................ ..... 2 5 2 . 1 SMP formulation matrix of the nine formulations used ................................ .................... 3 8 2.2 (A) Storage Modulus (MPa) of (A) tBA:PEGDMA550 (B) Storage Modulus (MPa) of tBA:PEGDMA750 (C) Storage Modulus (MPa) of tBA:PEGDMA1000. .................... 4 2 2.3 Atomic force microscopy (AFM) images of (a) tBA: PEGDMA550 , (b) tBA: PEGDMA750 and (c) tBA: PEGDMA1000 ................................ ......................... 43 2 .4 (A) Representative AFM images of tBA:PEGDMA550 samples (B) Roughness of tBA:PEGDMA550 (C) Roug hness of tBA :PEGDMA750 (D) Roughness of t BA:PEGDMA1000 ................................ ................................ ....... 4 5 46 2.5 Live Dead Analysis of SMP formulations with the lowest we ight percent of monomer (20 wt% tBA) ................................ ................................ ................................ ..... 4 7 2.6 Live Dead Analysis of SMP formulations with equal weight percent monomer (tBA) and crosslinker (PEGDMA). ................................ ................................ ................... 4 7

PAGE 11

xi 2.7 Live Dead Analysis of SMP formulations with highest weight percent (80 wt%) monomer (tBA) ................................ ................................ ................................ .................. 4 8 2.8 (A) C ell count of HUVECs present on each sample (B) Normalized EC At tachment ................................ ................................ ................................ ................... 49 50 2.9 Metabolic Ac tivity/ Cytocompatibility of SMPs ................................ ................................ 5 1 3.1 High tBA content SMP Formulations ................................ ................................ ................ 5 9 3.2 Glass transition temperature (T g ) of all SMPs ................................ ................................ ... 60 3.3 Contact angle of SMPs ................................ ................................ ................................ ....... 6 1 3.4 SMP surface roughness measurements ................................ ................................ .............. 6 2 3.5 Endothelial cell adhesion on 85:15 wt% tBA:PEGDMA SMPs ................................ ....... 6 3 3.6 Endothelial cell adhesion on 90:10 wt% tBA:PEGDMA SMPs ................................ ....... 6 4 3.7 Endothelial cell adhesion on 95:5wt% tBA:PEGDMA SMPs ................................ ......... 6 4 3.8 Estimated adherent ECs on SMPs ................................ ................................ ...................... 6 5 3.9 Cell metabolism of cell adherent SMPs ................................ ................................ ............ 6 6 4.1 Storage modulus and tan delta curves for 80:20 wt% unpatterned and microgrooved SMPs ................................ ................................ ................................ .......... 7 7 4.2 SEM micrographs verified pattern transfer and surface feature presence, or lack thereof, on unpatterned and micropatterned SMP surfaces ................................ .............. 7 9 4.3 Roughness of unpatterned vs. microgrooved 80:20 wt% tBA:PEGDMA SMP surface s ................................ ................................ ................................ ............................. 80 4.4 Wettability of unpatterned vs. microgrooved SMP surfaces ................................ ............. 8 1

PAGE 12

xii 4.5 Endothelial cell attachment at 1 day, 3 days and 7 days after cell introduction assessed using Live/Dead Cell Imaging and NucBlue Live Cell Stain to mark nuclei ................................ ................................ ................................ ......................... 8 2 4.6 A) Approximate cell presence , live and dead, on all unpatterned vs. microgrooved SMPs B) Percentage increase in EC presence on unpatterned and microgrooved SMPs ................................ ................................ ................................ .... 8 4 4.7 Approximate adherent ECs/SMP, live and dead, on all unpatterned vs. microgrooved SMPs ................................ ................................ ................................ ........... 8 6 4.8 Approximate adherent ECs/SMP, live only , on all unpatterned vs. microgrooved S M Ps ................................ ................................ ................................ ........... 8 7 4.9 Endothelial cell alignment as measured by nuclei and actin fiber organization on 80:20 wt% tBA:PEGDMA 550, 80:20 wt% tBA:PEGDMA750 and 80:20 wt% tBA:PEGDMA1000 , 1 day and 7 days after cell introduction ............................ 8 9 90 5.1 Deformation retention of 50:50 tBA:PEGDMA550 vs. 80:20 tBA:PEGDMA55 0 ........ 10 5 5.2 Percent recovery of SMP protrusion at temperatures above T g for A) 0.5mm protrusion B) 1.0mm protrusion ................................ ................................ .............. 10 6 10 7 5.3 Percent recovery of SMP protrusion at physiological temperature for A) 0.5mm protrusion B) 1.0mm protrusion ................................ ................................ ..................... 10 8 5.4 Recovery time for A) 60 second compression B) 60 minute compression .............. 10 9 1 10 5.5 Recovery time at temperatures 25% above T g and 50% above T g ................................ .. 1 11 5.6 Percent height compression for A) 0.5mm protrusion and B) the 1.0mm protrusion T g ................................ ................................ ................................ .................... 11 2 6.1 ................................ ................................ ...................... 12 3

PAGE 13

xiii 6.2 Surface manipulation of SMPs ................................ ................................ ........................ 12 6 6.3 SEM micrographs confirm pattern transfer to the SMP surface with groove widths and depths ranging from 55 65 µ m (µ ± SD) ................................ ....................... 12 9 6.4 Contact angle and groove depth of micropatterned SMP surfaces before compression, after compression and after recovery ................................ ......................... 1 30 6.5 Atomic F o rce Microscopy of 80:20 wt% tBA:PEGDMA 550,750, 1000 ............... 1 31 13 2 6.6 ................................ ................................ ................................ .. 13 3 6.7 ................................ ..................... 13 3 6.8 ......... 13 5 7.1 Figure summary of percentage endothelial cell adhesion, relative to ECs introduced on Day 0 of the study, on all SMPs & surfaces ................................ ............. 14 3 7.2 SEM of varying groove widths and depth of 80:20 wt% tBA:PEGDMA550 SMPs ...... 14 7 7.3 Cell adhe sion on passive and temperature responsive 50µm 80:20 wt% tBA:PEGDMA550 ................................ ................................ ................................ ........... 14 8 7.4 Cell adhesion on passive and temperature responsive 100µm 80:20 wt% tBA:PEGDMA550 ................................ ................................ ................................ ........... 14 9 7.5 Cell adhesion on passive and temperature responsive 150µm 80:20 wt% tBA:PEGDMA550 ................................ ................................ ................................ ........... 14 9

PAGE 14

1 CHAPTER I INTRODUCTION Polymers, 2014, 6 and are included with the permission of the copyright holder Cardiovascular Disease Cardiovascular disease (CVD) , including atherosclerosis and related diseases, is one of the leading causes of death globally . 1 According to the American Heart Association (AHA) , CVD accounts for approximately 1 in every 3 deaths in the United States while on a global scale, deaths from CVD totaled almost 18 million, and are expected to exceed 23 million by 2030 . 2 As a result, the AHA predicts that medical costs relating to CVD are projected to increase to 749 billion by 2035 . 2 Risk factors for CVD include smoking, physical inactivity, poor nutrition, obesity, and poor management of cholesterol, blood su gar and blood pressure. While lifestyle alteration s and adjustments are the first line of treatment for CVD, more advanced therapies, such as pharmaceutical, medical device or surgical interventions are usually required. A therosclerosis often results from localized inflammatory response and can be characterized by plaque formation in blood vessels . 1, 3, 4 Atherosclerosis , like many inflammatory disease s , is the result of the usually beneficial leukocyte recruitment process becoming uncontrolled . 5 Arterial luminal endothelial cells and leukocytes connect by expressing adhesion molecules such as selectins, intercellular adhesion molecule I (ICAM I) and capsular cell adhesion molecule I (VCAM I) as well as the corresponding receptor molecule . This expression event, coupled with a chemotactic gradient, encourages leukocyte recruitment to the site of the inflammation. For atherosclerosis, the inflammation site is the blood vessel itself. Atherosclerotic plaque may consist of fat, cholesterol, calcium and/or blood c omponents . 6 Plaque buildup causes

PAGE 15

2 hardenin g of the blood vessel and limits blood flow to tissues, ultimately leading to an acute ischemic condition such as stroke or myocardial infarction . 5, 7 Studies in both animals and humans have identified endothelial dysfunction as one of the key events in both early and advanced atherosclerosis . 5, 8 10 One instigator of atherosclerosis is an excess presence of cholesterol in the blood, termed hypercholesterolemia, which causes changes in permeability of arterial endothelial cells to allow lipids into and aggregate in the arterial wall. Studies by Schwen ke and Carew have show n that lipoprotein retention in the artery wal l appears to have a greater i mpact on atherosclerotic lesion formation than the lipoprotein transfer rate . 11 Ad hesion molecules expressed by endothelial cells bind circulating monocytes that subsequen tly migrate into the sub endothelial space and become foamy macrophages ; in conjunction with oxidized lipid particles, foamy macrophages further enhance the accumulation of lipids in the subcellular space. Cells from the arterial wall may emit oxidative products that seed lipids and initiate oxidation . Generally, l ipid oxidation takes place in two stages: 1) lipids are oxidized 2) monocytes are recruited and become macrophages , which h ave great oxidative capacity . 11 Oxidized lip ids can may up regulate adhesion molecules on ECs and cause injury to the endothelium . 8, 12 All of these events contribute to vascular remodeling associated with atherosclerosis and lesion formation . 7, 13 Traditionally, surgical methods such as coronary artery bypass graft surgery (CABG) were used to treat coronary artery disease (CAD) , which involved using arteries or veins from other parts of the body to bypass the narrowed coronary vessel. CA BG , a surgical procedure and the most common type of open heart surgery, is highly invasive and comes with a gamut of risks including wound infection, bleeding, reactions to anesthesia, fever, pain, stroke, heart attack and

PAGE 16

3 even death. Recovery time from a CABG procedure may take 6 12 weeks, or longer , but CABG is still the preferred procedure for patients with severe CAD . 14 Carotid Endarterectomy (CEA) is a surgical procedure to remove the plaque from the carotid arteries in the neck, restoring blood flow to the brain and preventing stroke . 15 The first CEA was performed in the mid 1950s, but efficacy data regarding this procedure was not available until the 1990s . 16 Since CEA is an invasive procedure performed under sedat ion , the risks associated with surgery are also present here , and thus CEA is only performed in patients with active plaque that may embolize or in patients with greater than 70% stenosis, as these patients appear to benefit most from surgical intervention ; most patients , however, benefit from less invasive methods, which include pharmaceuticals and lifestyle alterations, to mitigate the risks . 15 17 To mitigate the risks associated with surgical intervention, balloon angioplasty was conceptualized in the mid 1960s and implemented in the 1970s as a less invasive technique to open occluded vessels . 18, 19 A balloon is inserted via catheter on a guidewire to the site of the blockage and inflated to re expand the lumen and restore blood flow to downstream organs and tissues . 16 While balloon angioplasty was a significant step in improv ing treatment of arterial occlusion, lim itation s stemm ed from re narrowing of arteries caused by elastic recoil, delay in vascular remodeling or neointimal proliferation . Current Stents and Limitations Cardiovascular stents are expandable tubes used to treat narrow or weakened arteries that arise as a result of atherosclerosis and its resultant sequelae, such as coronary artery disease, peripheral artery disease, etc. These devices provide a minimally invasive means to mechanically support the damaged vessel which restores oxygenated blood fl ow to the organs and tissues

PAGE 17

4 downstream . 20 Stents were first developed and implanted in the 1980s, and have since experienced improvements in material, design and function . 19 Although cardiovascular stents have saved countless lives, the device has many limitations, which drive continued research in the area . 21 In particular, thrombosis and restenosis continue to be relatively important problems wit h current stents. Given that these issues arise from surface interactions, surface modification techniques are an active area of current research. Descriptions of an ideal stent have been profiled across literature . 22 The ideal stent should : 1 ) have a low profile and the ability to be crimped onto a balloon that is mounted on a guidewire 2) sufficient ly expand once the stent reaches the target area and is deployed by balloon expansion 3) demonstrate good radial strength and minimal recoil, so that the stent offers support to the vessel wall, regardless of stresses and maintains its integrity 4) be able to navigate through even the small diameter atherosclerotic vessels 5) be radiopaque and/or MRI compatible for ease of stent placement 6) demonstrate blood compatibility and thrombo resistivity, or resistance to platelet adhesion and activation 7) have the potential for drug delivery The first types of stents used were bare metal stents (BMSs), composed of a variety of metals and/or alloy s such as stainless steel, cobalt chromium and tantalum for balloon expandable stents, or nitinol (nickel titanium alloy) for self expanding stents . 20 316L Stainless Steel (SS) is the most common material used to fabricate metal stent s because it is corrosion

PAGE 18

5 resistant and has suitable mechanical properties . These stents provided the necessary mechanical support for the weakened vessel; however, an increased risk for thrombosis and/or restenosis in these devices may generate additional need for reintervention 6 to 12 months post implantation . 20, 23 Restenosis, or re nar rowing of the vessel, often results from excessive neointimal proliferation following balloon angioplasty or stent implantation due to vessel injury from the expansion . 6, 18, 24 Additional causes of restenosis may inc lude reduced compliance between the stent and the vessel and excessive tissue remodeling response to the stent material . 18 Thrombogenicity, one of the aforem entioned issues associated with BMSs, refers to increased propensity of the device or material to generate a blood clot on the material surface . 25 Thrombotic events often occur due to net electrical charge differences between blood components and the stent surface, as well as surface potential incompatibility between the metal and the contacting blood . 22, 26 As a proposed improvement on BMS, the first generation of drug eluting stents (DESs) consisted of a metal backbone and a permanent, non absorbable polymer coating to house a drug of choice . 27 While DESs offered control and localization for drug release to the injured vessel, incidences of hypersensitivity, heart attack and even de ath remained problematic . 20 An improved DES replaced the non absorbable polymer coating with a non thrombogenic, absorbable one. This absorbable coating served to encourage endothelialization through a directed drug release profile an d reduced inflammatory response during polymer degradation. While these improved DESs decrease d the occurrence of restenosis through release of anti proliferative agents, late stage thrombosis still occurred . 1, 6, 28 31 Late stage or late stent thrombosis (LST) can result from a variety of issues, ranging from the stenting procedure itself to early termination of

PAGE 19

6 antiproliferative drugs loaded in the stent; these issues may cause increased local fibrin deposition and del ayed healing . 28, 29 Furthermore, studies have shown that drug eluti on itself may result in inhibited endothelial cell proliferation . 32 Endothelialization Vascular endothelial cells compose the inner lining of all blood vessels in the human body . Endothelial cells were once considered a passive barrier between blood and tissue, but it is now know n that these cells collectively form a dynamic organ system , creating a semi permeable barrier between blood and tissue . 33 Furthermore, e ndothelial cells have been shown to differentiate based on both internal and external factors, in order to meet the needs of the environment in which they function . 34 This dynamic organ system collectively participates in a range of physiological and pathological processes including vasculature development and remodeling, vascular tone and blood fluidity control, movement of blood vessels and nutrients, as well as active players in atherosclerosis and tumor angiogenesis . 34 The endothelium is also an active component in the prevention of intimal hyperplasia and thrombosis . 32 Integrin binding to the extracellular matrix mediates endothelial cell adhesion and migration. Integrins are heterodimeric cell surface receptors that serve as anchors to the extracellular matrix. Integrins, which recognize specific extracellular matrix ligands to facilitate adhesion, also serve as vehicles of chemical and mechanical signals between cells and their environments . Numerous s tudies have investigated both the chemical and mechanical mec hanisms associated with cell adhesion. Chemically, the focus has centered on the interaction between endothelial integrins and ECM matrix proteins such as fibronectin, laminin, and collagen. On the mechanical level, substrate stiffness has been shown to af fect cell adhesive strength, contractility, focal adhesion formation, cell cell interactions, etc. 33

PAGE 20

7 The process of endothelialization generally involves recruitment of endothelial cells as well as early and late endothelial progenitor cells (EPCs) . 35, 36 The specific con tribution of EPCs to the endothelialization process is still unclear, especially since these EPCs make up a very small percentage of circulating cells . Minor disturbances in endothelium that do not disturb the underlying basement membrane are primarily dea lt with by migration of healthy endothelial cells from adjoining endothelium, whereas larger gaps in endothelium as well as endothelialization of implanted vascular grafts are primarily covered by endothelial cells in circulation . 36 Init ially, c irculating hematopoietic stem cells and platelets adhere to the site of graft implantation; in some cases, these cells are replaced by the migrati on and spreading of endothelial cells, whereas in other instances endothelial cells replicate in an attempt to cover the exposed areas . 35 Various methods and techniques to encourage endothelialization have been investigated and will be discussed later in this section. It is now well known that a healthy, intact endothelium is required for protection against maladies such as hyperplasia and thrombosis; specifically, the endothelium is responsible for maintaining a homeostatic environment in the blood vessel . 37 Many devices, biodegradable and oth erwise, aim to encourage endothelialization of the surface to reap the benefits of the endothelium, which largely includes reduced device complication and rejection . 35 Rapidly establishing a complete monolayer of endothelial cells on the luminal surface of a stent reduces and potentially eliminates many of the limitations associated with current stents and may guarantee long term device success . 38 Once cells adhere t o a compatible s caffold , they can carry out regular functions, such as proliferation, migration, and differentiation, which may ultimately determine cellular survival or death . 39

PAGE 21

8 Polymers Polymers are organic materials that have long chains held together by covalent bonds. Polymers are attractive materials for biomedical devices because they are highly tailorable and possess a wide range of properties, both chemical and physical, that may be attractive for biomedical device d esign. Polymers may be synthetic, w hich are attractive due to their mass producibility and serializability, but most synthetic polymers do not actively interact with their surrounding s in their native form, and the path to regulatory approval can be challenging. Naturally delivered polymers on the other hand are, as the name suggests, naturally occurring, inside or outside of the body . These materials may integrate more easily with their surroundings but are more difficult to obtain in larger quantities and have relatively sub optimal mechanical properties . Polymers are used in a variety of biomedical devices such as contact lenses, sutures, joint implants, gloves, wound dressings, as well as vascular grafts . 40 P olymer s are favorable materials for stents because they can achieve increased hemocompatibility with the proper selection of polymer components, polymerization and processing techniques . 26 DES s have been shown to cause a delay in re endothelialization, which may promot e a thrombotic environment, as re endothelialization is an important component in vessel healing. In addition, instances of very late stage thrombosis (LST) have also b een seen with DESs, making polymer stents a potentially more appealing route for stent materials . 41 Patients who receive DESs are often required to continue an anti platelet regimen for 12 months to prevent adverse effects from the DESs and while these anticoagulants prevent thrombosis, the y may carry a sustained risk of hemorrhage, or bleeding, and related side effects . 24, 42, 43 Due to safety concerns with existing stent materials, current research in stent design is progressing

PAGE 22

9 towards using biodegradable/bioabsorbable or biomimetic materials for polymer or metal stents as well as polymer coating free DESs, among others . 6, 41, 44, 45 Shape Mem ory Polymers Shape memory polymers are smart materials that can change their shape upon the application of a stimulus such as heat, light, infrared irradiation, humidity, immersion in water, elec tric field application, or the application of alternating magnetic fields, to name a few . 46 50 To activation, and/or mode of degradatio n . 51 me indicated above, and can handle strain deformations up to 800%, compared to only 8% for SMA . 48, 52 They also have low density, high frozen strain, low manufacturing costs, easy processing methods, wide shape transition, and biocompatibility . 50 SMPs also have a higher mechanical stability than hydrogels, another polymer commonly used in the biomaterials field . 48 such as high material costs, limited thermo mechanical property control, and limited resistance to fatigue . 53 s into its permanent shape. The polymer is then deformed and fixed, or programmed, into its temporary shape. The permanent shape is recovered when the polymer is exposed to a stimulus and this recovery is a result of the shape memory effect (SME) . 48 effect (SME). SME allows a polymer to undergo large amounts of strain without becoming permanently deformed; thus, the better the SME in a polymer, the more likely it is that the polymer will return to its original shape without

PAGE 23

10 any evidence of prior defo rmation . 54 It is also important to note that SMPs and shape changing polymers are different; SMPs are driven by the SME which itself is driven by thermal transitions, whereas shape changing polymers are driven by microscopic movements in the polymer resulting in macroscopic mov ement of the material . 55 The S ME is an effect of polymer structure, polymer morphology and applied processing and programming. Cyclic thermo mechanical tests are used to validate the effectiveness of SME in polymers . 48 SME is governed by a dual segment, or dual phase, system; one segment is elastic while the other segment is a transition segment that becomes softer when the stimulus is applied . 52 These segments are dependent upon the chemical composition of the material and often exhibit strong interactions. Transition segm ents soften at a switching/transition temperature T sw /T trans , either the glass transition temperature, T g , or the melting temperature, T m . 52, 55 T g usually covers a broad range of temperatures whereas the range for T m is more narrow, and thus T g is more often used as T sw or T trans . 55 The T trans can be manipulated simply by altering the ratio of components, monomers and crosslinkers, that contribute to the make up the polymer; other properties such as crosslinking density and mechanical properties also change with changes in t his ratio . 50, 54 The permanent shape of the SMP is determined by the net points and the chain segments that connect them . 55 These chain segments and net points may consist of either chemical crosslinks, which are often covalent bonds, or physical crosslinks, which are usually created by segregated domains. Crosslinks help keep the chains from slipping by acting like entanglements or anchors, helping the polymer retain its permanent shap e . In addition, chemical crosslinks may help prevent the polymer from dissolving while still allowing the polymer to swell, which may be important for some applications . 48 Domains with the highest thermal transition act as the

PAGE 24

11 netpoints which determine the permanent shape, while domains with the secon d highest transition, the chain segments, act as molecular switches. These molecular switches are responsible for fixing the temporary, deformed shape . 48, 55 Shape recovery is driven by entropy elasticity of switching domains. In the elastic state, the polymer is more amorphous and rubberier , and thus the polymer chains display a random coil formation and are at the state of higher entropy. In general, polymer segments reorient when the SMP is heated above T trans and again when there is an applied deformation to the polymer; when T>T trans , po lymer segments exhibit entropically favorable random coil conformation. Property changes such as permeability, transparency, melting, and crystallization may be a result of the changing of the switching segments from a more oriented state to a more random conformation . 55 analyzed using a few different parameters. Shape fixity ratio (R f ) measures the ability of the switching domains in the SMP to fix a deformation when a mechanical stress is applie d . 52, 55 The free strain recovery ratio (R r ) is a measure for the ability of the material to memorize and recover its permanent shape . 48, 52, 55

PAGE 25

12 The cyclic thermomechanical tests that were mentioned before result in stress and strain data, which can be plotted and the slope of the stress vs. strain curve yields the elastic modulus, E, which is a measur is applied. The elastic modulus is important for materials such as SMPs that are used as medical devices and may need to mimic the properties of some native tissues or biologi cal structures . 48 There are several important considerations associated with changing even the slightest aspect of the monomer(s) and/or crosslinker(s) that make up a polymer; these considerations can be chemical, thermal or mechanical. Increased hard segment content is directly corr elated to an increase in mechanical strength of the polymer. Increasing the number of hard segment blocks increases T g which limits twisting and coiling of switching segments, controlling shape memory properties. Increased T g may help with heat and oxidati on resistance. Switching segments with higher molecular weights yield more crystalline solids. But, as the molecular weight of the switching segments decreases, the deformation recovery rate increases. Side groups on the monomers may function as physical c rosslinks and help stabilize the material above T g , depending upon the chemical nature of those side groups. Crosslinking is dependent upon initiator concentration, crosslinking temperature, and cure time and these can be adjusted to achieve maximum crossl inking . 48 Yet another benefit of SMPs is that they can be modified to cater to specific purposes. Certain filler materials such as fibers or magnets can improve mechanical strength/increase modulus or enhance the magnetic properties, respectively. Their versatility, adaptability and ease of p rocessing, and low cost have helped push these materials into the forefront of many research efforts.

PAGE 26

13 Polymer & Shape Memory Polymer Stents Polymers have caught the attention of the medical device industry due to their diversity and versatility. Polymers are less dense than metals and have higher flexibility than many other materials, which allows better matching of stent compliance with that of the local vessel . 18 In addition, polymers are easy to manufacture and often have lower bulk material costs and processing costs . 56, 57 Polymers also possess a wide range of bulk properties, such as elasticity, conductivity, strength and degradability, which can provide the s tent designer with a large palette of useful features . 58 Thus, polymers can be easily and cost effectively tailored to fit the needs of their application, making these materials appealing for use in the medical device industry. Shape memory polymers (SMPs) have added advan tages to those seen with conventional polymers. As stents are often delivered via catheter, SMP systems offer benefits for catheter storage and deployment, since the materials can pack tightly without becoming permanently deformed during the storage period . SMPs may also enhance the ease of delivery of many of these devices, and produce lower recovery forces, leading to minimally invasive procedures with reduced recovery times for patients . 18 These added benefits, on top of the already present benefits of polymers, make SMPs potentially attractive materials for next generation polymeric stents. Surface Modification to Increase Biocompatibility Despite the varie ty of materials and designs currently available for stents, there is still a need for a single material that has the desired mechanical properties while simultaneously achieving optimal biocompatibility . 1, 43 Biocompatibility refers to the reaction elicited by a material when it is inserted into the body; ideally, this reaction should be favorable and should not provoke a negative response such as an attack by the immune system on the foreign

PAGE 27

14 material . 59 62 Surface modification techniques strive to retain favorable bulk properties while changing the surface to cater to specific needs, often to enhance biocompatibility . 29, 63 Since shape memory is not a surface property, surface modifications should enhance biocompatibility without interfering with the shape memory capabilities of SMPs. Surface modifications that allow for improved blood contact (minimal thrombogenicity) while encouraging vascular wall healing via endothelial cell migration, anchorage and proliferation, are the focus of research goals in this area . 1, 21, 59, 64, 65 In add ition, surface modifications for drug release in an effort to eliminate the polymer coating are also being explored . 1 One of the keys to success for many medical devices is successful wound healing, a process that begins at the surface of a material. Successful wound healing depends on a range of material properties, both surface and bulk, such as surface texture, surface chemistry, surface energy, crystallinity as well as leachable content and biocompatibility of the degradation products. In essence, biocompatibility is heavily dependent upon surface properties as well as interactions between the surface and cells and/or proteins, or between cells themselves . 43, 59, 64 69 Protein adsorption may also play a factor in dictating the success or failure of blood contacting devices; some proteins, such as albumin, can be beneficial for biocompatibility as albumin may decrease both platelet adhesion and binding of microorganisms that may elicit infection, but non specific proteins, such as fibrinogen and Immunoglobulin G ( IgG ) , may increase platelet adhesion by instigating a host response . 43 Methods for Surface Modification Surface modifications should generally be thin, affecting only the topmost layer of the surface; thick layers may undesirably alter the bulk proper ties and have difficulty adhering to the surface, while overly thin layers are subject to erosion; despite these requirements, however,

PAGE 28

15 there are a number of ways to modify the surface of a material to enhance its functionality . 63, 70 For polymer derived stents, methods for modifying surfaces with the end goal of achieving improved blood compatibility, re endothelialization, or both can be grouped into six major categories. These categories are surface roughening, surface patterning, chemical modification of the surface, surface coatings and films, attachment of pharmaceuticals or biopharmaceuticals to the surface, and the formation o f porous surfaces to facilitate drug delivery, many of which are represented in Figure 1 . 1 . Multiple techniques may be used to achieve the desired properties . 1 Figure 1 . 1 . Surface modification techniques commonly used to enhance endothelialization and/or reduce thrombosis . 71 Surface Roughening In general, surface roughening aims not only to increase the surface area of the material, but also to restrict cell movement, which contributes to enhanced cell attachment . 43, 72 74 Cells are still able to migrate on roughened surfaces, but no significant increases or decreases in migration have been noted compared to smooth surfaces . 75 In addition, surface roughening modifies the

PAGE 29

16 topology of the surface without chemical alteration, which may have benefits, depending upon the material and its desired use . 76 For metals, roughening techniques such as sputtering with TiN or TiO 2 have been used to successfully enhance endothelial cell attachment. However, these cells express less endothelial nitric oxide synthase (eNOS), which may lead to increased endothelial cell dysfunction; this reduced eNOS activity has actually been shown to be characteristic of metals in general, modified or bare, presenting a reason for further research into non metal implant materials . 77 Microblasting followed by reactive ion etching on metal surfaces also produces roughened, high energy surfaces that may potentially improve cell attachment . 42 For polymers, oxygen or argon plasma deposition increases surface roughness as well as hydrophilicity, both of which have been shown to enhance cell attachment; application of plasma deposition towards SMP based stents may allow for enhanced wound healing and biocompatibility . 43, 57 Plasma processing alters the surface topography through melting and recrystallization processes, resulting in more ridges compared to the original surface, as displayed in Figure 1. 2 . 62, 78 Etching and sanding, both plasma or chemical based, as well as polishing and/or microblasting also serve to improve surface roughness . 1, 79

PAGE 30

17 Figure 1. 2 . Scanning electron microscope (SEM) image of RIE textured silicon surface using plasma consisting of Cl 2 , CF 4 and O 2 gases (scale bar = 200 nm). Reprinted with permission from Elsevier, 2001 . 62 Shadpour et al. , roughened polymer surfaces using a slurry of alumina particles, with the intention of enhancing endothelial cell attachment without altering the chemical make up of the polymer surface. This process, in addition to being used to roughen the surface and increase surface area, can also be used for patterning purposes, both of which encourage cell and biomolecule attachment . 76 This method, which has been shown to in crease cell attachment while modifying the polymer surface without disturbing the bulk, may be worth investigating for next generation SMP stents due to the potential for increased biocompatibility. Plasma and chemical based etching occurs when a surface is exposed to etching gas, which is often a type of plasma, and the top layer of the surface is changed through chain scission processes where old bonds are broken and new ones formed; more simply, etching degrades the polymer surface . 56, 80, 81 This process also modifies the surface topography and affects surface wettability, potentially driving the surface to become more biocompatible . 80, 81

PAGE 31

18 Etching can also be performed prior to coating a material, to ensure that the coating adheres . 81 Treatme and migration of endothelial cells, especially in polymeric hydrogels . 82 Grafting of different length polymer chains can alter the surface roughness, particularly on a nanometer scale. Ro ughening at this scale has been shown to enhance cell attachment and improve biocompatibility . 83 Transfer printing, a common tec hnique used for patterning, may also be used to roughen the polymer surface. The mold that houses the polymer during curing transfers the roughened features onto the surface during polymerization, as seen in Figure 1. 3 . 84 Although many of these technique s have not yet been applied to SMPs, their use on polymers shows promise for the successful application to SMPs, granted that the methods continue to modify only the topmost layers of the material.

PAGE 32

19 Figure 1. 3 . Atomic force microscopy (AFM) images of (a) untreated, (b) micro roughened and (c) nano roughened polydimethylsiloxane (PDMS) films. Reprinted with permission from IOP, 2009 . 84 Surface Patterning Surface patterning offers a more organized means of roug hening to alter the surface of a material. Patterning may quell non specific protein surface interactions, as these effects often lead to device failure . 85 Such patterning techniques are often used to enhance endothelial cell

PAGE 33

20 attachment, which in turn encourages vessel wall healing and promotes an anti thrombotic environment. Nanopillar arrays, formed by plasma processing as shown in Figure 1. 4 , provide a scaffold for cell proliferation or drug delivery . 62 Patterning on metal surfaces, primarily on the nanometer scale, has been shown to promote more endothelial cell attachment compared to random nanopatterning or even patterning on the micron scale . 1, 86 These nanopatterned surfaces also enc ourage more endothelial cell attachment compared to smooth cell attachment which is desirable in vessel healing, support greater cell densities on the surface, and even enhance spreading of these endothelial cells . 1, 87 Cells in their native environment come into contact with features on the nano scale, which could be the reason for enhanced cell attachment . 83, 88, 89 So me patterning methods strive to mimic native endothelium for a biomimetic effect, in hopes of encouraging more rapid endothelialization and vessel healing, without the presence of plasma proteins or extracellular matrix . 1, 3, 85, 88 Biomimetic patterning may have major implications for SMP stents in that increased biocompatibility can be obtained simply by polymerizing the stent inside of a native blood vessel, directly transferring native endothelial pattern onto the stent surface.

PAGE 34

21 Figure 1. 4 . Reprinted with permission from Elsevier, 2001 . 62 Patterning can also be achieved through diblock copol ymer grafts, which form nanometer sized patterns on solid surfaces. Diblock copolymers can be either physically or chemically attached and form nano sized domains when they undergo microphase separation. These patterns either encourage or discourage protei n adsorption and/or cell adhesion, depending on the polymers involved. For this reason, diblock copolymers have been investigated with regard to surface energy or topography and are being explored for their potential uses in reference to bioactivity . 90 Polymers that undergo phase separation such as the mixture of polystyrene and poly(4 bromostyrene) can produce a range of surface topographies just by varying the polymer concentrations and proportions . 91, 92 Changes in polymer ratio can yield variations in shape, such as pits, islands and ribbons for example, whereas changing the concentrat ion of the polymer may alter the feature sizes. Cell spreading and proliferation differ based on feature height, with

PAGE 35

22 shorter feature heights producing promising results in enhanced cell spreading and proliferation . 92 With regard to polymer surface patterning, lithography is one of the more frequently used techniques, a technique common in the electronics field, mainly for patterning of silicon wafers . 64 Patterns can include anything from dots and pillars to g rooves and ridges, where grooves and ridges are the most studied, often due to the increased tendency of cells to attach and spread along those features . 64, 93 Lithography may even be used to cre ate hierarchical patterns or tilted patterns, if desired . 94 A few theories have attempted to predict why cells prefer to align along grooves and ridges, but different cells have different preferences with regard to size and shape of the formed pattern . 64 Photolithography is commonly used on polymer surfaces and this techniqu e selectively exposes surfaces to photoirradiation, creating a pattern on the surface . 95 97 This allows for controlled topographical features, directing cell attachment . 95, 98 Lithographic techniques continue to be a prominent surface modification method for polymers and applying these methods to SMPs, particularly SMP based stents, may also prove beneficial. Microfluidic channels offer another means to direct cell adhesion via patterning. Proteins adsorb onto the surface after passing through elastomeric channels in solution form, and once adhered, these proteins, such as fibronectin and collagen, are used for selective cell adhesion. This method can als o be used to produce a patterned cell co culture, if two different types of cells need to adhere to the same surface . 99 Self Assembled Monolayers (SAMs), a common chemical based surface modification technique, have also been explored in creating patterns on biomaterial surfaces . 64, 100 SAMs encourage cell adhesion and orientation, qualities that are advantageous to stent biocompatibility,

PAGE 36

23 by controlling protein adsorption onto the surface . 101 SAMs are also used for microcontact printing, another method for patterning that is commonly used to encourage cell attachment . 62 With regard to SMP specific patterning techniques, methods in which balls (steel or lime glass) that make indentations on the surface have been explored. Different sized indentations can be made using different sized balls . 102 In addition, wrinkling patterns on top of SMPs can be formed using the sha pe memory capabilities of the polymer itself and if the wrinkling is controlled, a number of surface properties that improve biocompatibility can be manipulated, including roughness, wetting, and bonding among others . 103 Transfer printing involves the transfer of a pattern from a mold to a polymer substrate, resulting in a thin patterned film on the surface of the polymer . 94 These films are usually polymers themselves and have the potential to encourage cell adhesion by introducing nanoscale patterns that favor cell attachment. Transfer printing can also create surfaces with hydrophobic and hydrophilic characteristics, directing cell attachment to certain areas . 104 Zhao et al. , determined that microtransfer molding using a PDMS mold creates micron sized patterns, which may once again increase endothelial cell attachment . 105 Similar to transfer printing, stencil assisted printing involves usi ng a stencil to imprint a desired pattern or structure onto the polymer surface. The patterns develop on the surface that is left uncovered by the stencil, thus directing cell attachment to these exposed areas. This technique does not require any chemical modification after the stencil has been manufactured, making it an appealing method to enhance material biocompatibility as well as a potential technique for surface patterning of SMPs . 85 Nanopatterning through dip pen lithography uses the tip of an atomic force microscope

PAGE 37

24 touched to t he surface of a material, altering the chemical makeup of that surface in an organized manner, creating a pattern . 106 Depending upon the polymer that is applied to the surface, enhancement in blood compatibility and/or cell attachment can be achieved. The use of a heated tip to create patterns on the surface of SMPs has been explored, and may provide an avenue for patterning SMPs to encourage cell attachment . 107 In dentations can also be made using a scanning force microscope (SFM), generally for analytical purposes, but there may be potential for surface modification here as well . 108, 109 In an effort to physically mimic the patterns found in native vessels, pre polymer solutions were pol ymerized inside of a harvested, native blood vessel . 94 The polymer adopts the surface features of the blood vessel on its own surface, but the main limitation of this method is that the vessel tissue had to be dissolved to remove the polymer, rendering reproducibility di fficult. However, since SMPs acquire their permanent shapes during initial polymerization, this method applied to SMPs may be worth further investigation. Although SMPs have been treated with only a few patterning techniques, the success associated with p atterning polymer materials suggests that patterning SMPs with these techniques may have positive outcomes. Chemical Modification of the Surface Chemical modification techniques chemically alter the surface of a material without significantly affecting its bulk properties. Some examples of chemical modifications include chemical vapor deposition (CVD), plasma vapor deposition (PVD), grafting techniques, self assembled monolayers (SAMs), among others . 1, 110

PAGE 38

25 For metals, many chemical modification techniques, such as plasma immersion ion implantation (PIII) using acetylene, nitrogen or oxygen, are used to reduce corrosion, wear and meta l leaching into the surrounding environment and even increase hardness of the material . 29 PIII treatment of polymers has been shown to reduce thrombus formation and platelet aggregation by increasing hydrophilicity and protein adsorption onto the su rface, as displayed in Figure 1. 5 . 111 Figure 1. 5 . Isolated platelets in buffer adhering to both surfaces, but platelets in plasma do not adhere to ion treated polymer surface ( scale bar = ) . Reprinted with permission from PNAS, 2011 . 111 Chemical vapor deposition (CVD) utilizes plasma or other reactive chemicals to deposit thin f ilms onto the surface of the material, slightly altering the surface to allow for deposition of the film . 98, 112 Due to the non fouling properties associated with the deposited film, plasma driven CVD techniques are popular for blood compatibility . 112 One form of CVD has been used on stents and other blood contacting devices commercially and goes by the name parylene.

PAGE 39

26 Parylene aids in biocompatibility as well as providing a means for sustained drug release from a porous matrix . 113 However, this coating does not have functional groups to attach biomolecules, so further treatment with plasma or chemicals to introduce tethering molecules would be required for biomolecule attachment. Low pressure plasma treatments that use electrons, ions, radicals, metastables or u ltraviolet rays ( UV ) radiation elicit reactions at the surface of polymers . 81 For plasma based treatments, ammonia plasma treatment may encourage cel l attachment more through the interaction of acidic groups on the plasma membrane surface and amine/amide groups on the surface of the polymer, which play an important role in endothelial cell adhesion and growth . 43, 114 Some studies have shown that cornea cells showed enhanced attachment and growth on plasma treated surfaces vs. untreated surfaces, and exploration into whether this applies to other types of cells may have merit . 81 Studies prepared by Ho et al. found that polymer samples that undergo water vapor plasma treatment may elicit enhanced cell attachment compared to untreated samples, potentially due to the formation of hydroxyl groups on the surface, allowing for hydrogen bond formation between the surface and the cell s . 115, 116 While research has been generally inconclusive about which surfaces are best for supporting cell adhesion and growth, surfaces that are mildly hydrophobic or mildly hydrophilic appear to support optimal cell development; these mild conditions may be achieved through plasma treatment using reaction gases containing organic compound s . 64, 99, 117 119 Plasma treatment of polymer materials has positive effects on cell adhesi on and development on the material surface, mainly by enhancement of hydrophilicity and wettability . 70, 74, 117, 120 Plasma deposition may even be used to reduce thrombogenicity . 70

PAGE 40

27 Plasma vapor depos ition (PVD) techniques such as matrix assisted pulsed laser evaporation, deposit organic and biological materials onto the surface of blood contacting devices, altering the surface . 1 Ionic plasma deposition has been shown to increase endothelial cell adhesion . 121 Certain polymer surfaces exposed to N 2 and O 2 in Helium display enhanced attachment properties, with th e extent of surface modification depending upon the polymer surface itself . 43, 80 Other surfaces exposed to nitrogen gases have been known to exhibit reduced thrombotic properties . 4, 122 As with etch ing, plasma processes cause the formation of free radicals at the material surface, causing the formation of cross links . 57 These reactive surfaces can be used to encourage coverage with a thin film or can facilitate the attachment of (bio)molecules. Photografting of polymers using high energy electrons, gamma radiation, ultra violet (UV) light and visible light can change the surface of polymers to improve blood compatibility and enhance endothelialization . 1, 43 Bilek et al. , found that treatment of a polymer surface with ions to create a free radical surface encourages protein immobilization while retaining protein structure, potentially enhancing biocompatibility . 111 Photo oxidation, a method to introduce hydrophilic groups to polymer surfaces in a controlled manner through the manipulation of photo oxidation time and grafting tim e, has also been shown to be beneficial to endothelial cell development on the material surfac e . 64 Chemical grafting methods, such as the grafting of polyethylene glycol (PEG) monoacrylates to the surface of a biomaterial, can reduce the attachment of erythrocytes through steric repulsion, thus decreasing the risk of thrombosis . 123, 124 PEG is largely hydrophilic and has a large exclusion volume, contributing to this effect and it is also non toxic and non immunogenic which are important com ponents of a biocompatible material . 123 Grafting copolymerization methods that graft hydrophilic polymers onto hydrophobic surfaces in an effort

PAGE 41

28 to neutralize hydrophobicity may encourage cell adhesion . 64 Plasma and ultraviolet grafting on polymer surfaces may also promote an ticoagulation and antibacterial properties . 21 Self Assembled Monolayers (SAMs) modify the surfaces of materials to enhance hydrophobicity/hydrophilicity or to add reactive or functional groups to the surface. SAMs change the surface energy or wettability of the polymer surface through careful selection of the functional groups used for the monolayer, potentially increasing biocompatibility of the material within the vessel . 12 5 SAMs offer the benefit of ease of fabrication, and the ability to control order and orientation, allowing for the exposure of a select group on the modified surface, creating the ability to cater the biocompatibility of the material to suit specific needs . 70 Chemical modifica tion techniques strive to alter the surface of the material in order to enhance the functionality of that material. Polymer substrates undergo exposure to these different techniques, resulting in a material with an improved surface and mostly unchanged bul k properties. If performed properly, these chemical modification techniques can be applied to SMPs allowing for a better surface without affecting the bulk. Surface Coatings and Films Surface coatings and films are additional ways to modify surfaces of bot h metals and polymers in an effort to increase biocompatibility. These techniques often do not involve direct attachment of chemical groups or chemical alteration of the surface the way conventional chemical modification techniques do, but still alter the surfaces for increased biocompatibility. A few coating and film techniques that have been shown to increase endothelial cell attachment or reduce blood coagulation and thrombosis are discussed. With regard to wet coating/solvent coating of stents, dimethy l sulfoxide (DMSO) has been shown to prevent vascular smooth muscle cell activity on the stent surface, reducing

PAGE 42

29 chances for restenosis while also preventing tissue factor activity, thus discouraging thrombosis . 41 Studies show that DMSO does not exert toxicity to human vascular endothelial cells, further solidifying this technique as a potentially viable option for polymers and SMPs . 126 Dip coating, used to form nanostructures on the surface of medical grade polymers, creates superhydrophobic surfaces that prevent blood coagulation . 127 Coating polymers with polyelectrolyte multilayers provides a good platform for endothelial cells on polymer surfaces . 21 Langmuir Blodgett ( LB) films, consisting of highly ordered, densely packed structures of known thickness deposited and crosslinked to the surface of the polymer, also allow for cell adhesion, decreased platelet adhesion and enhanced hemocompatibility . 43, 128 These LB films can be deposited on polymer surfaces by chemically treating the polymer to attract the LB film and introducing the polymer to a Langmuir Blodgett trough, allowing a monolayer to form prior to endothelial cell exposure . 129 These LB films have not yet been studied extensively on three dimensional scaffolds, but implementing these films on three dimensional structures may be worth further investigation due to the increas ed biocompatibility offered by this technique. Layer by Layer (LbL) polymer films have been shown to reduce platelet adhesion on nitinol, a commonly used stent material . 1 LbL deposition of chitosan on the surface of the polymer poly l lactide (PLLA), showed improved cell compatibility . 130 Studies with diamond like carbon (DLC) fil ms have also displayed successful attempts to improve blood compatibility on polymer surfaces . 25

PAGE 43

30 Attachment of Pharmaceuticals, Biopharmaceuticals or Biomolecules to the Surface The ability to attach a substance to the surface of a material while retaining its bulk properties is an appealing method of delivery for pharmaceuticals or biomolecules . 58 Polymers usually have inert surfaces, so in those instances, the surface must be functionalized prior to attaching the bioactive molecule to the surface. As a surface technique, these methods can be applied to either polymers or SMPs, and while most of th ese techniques have been tested on polymers, there may be benefits to applying these techniques on SMPs. The bioactive compound can be attached through electrostatic interactions, ligand receptor interactions, or covalent attachment, where covalent linkage s are most common as this linkage is often the most stable . 58 Chemical Vapor Deposition (CVD) is not only used to enhance biocompatibility, but is also used to create tethering groups on the surface of a polymer for proteins and other biomolecules to attach via covalent bo nding . 1, 112, 113 Some of these biomolecules help create a less thrombotic environment by immobilizing on the surface of polymers in the blood vessel. Plasma deposition techniques produce stable films that can aid in corrosion resistance and funct ionalization sites for the attachment of (bio)pharmaceuticals onto the surfaces of both metals and polymers . 1 Wet chemical surface modification methods require chemical reagents to create reactive functional groups on the surface of a polymer, often without expensive equipment or methods, and can be done easily in a laboratory setting. Wet chemicals are able to ac complish deeper penetration of porous surfaces compared to energy source based modification techniques, creating a more stable and noncorrosive functionalized surface. If repeatability is desired however, this method may not be the ideal choice, as a wide range of reactive groups are generated, and the orientation of the biomolecule can be crucial for attachment. However, to

PAGE 44

31 promote specificity, it may be possible to block some functional groups, allowing for the specific molecule, whether it be a molecule to enhance endothelialization or a protein to reduce thrombus formation, to attach successfully . 58 Plasma treatment methods can introduce reactive groups to the surface of a normally inert polymer, allowing for the attachment of a desired bioactive compound. These methods do not require hazardous chemicals, yet still have the capability to modify the surface while imparting less degradation and roughness compared to wet chemical surface modification techniques . 58 In addition, the film deposited on the polymer surface can be manipulated by c hanging the deposition rate, energy range and surface topography . 131 Pl asma pre treatment has also been used prior to attaching collagen to a polymer nanofiber mesh, a method that showed increased cell attachment, spreading and viability . 132 Thus, if a less corrosive method for biomolecule attachment is required, plasma treatment may be a favorable option, but great care must be taken to avoid contamination of the sample. Once the surface has been functionalized, the desired bioactive agent c an be attached for purposes of enhanced cell attachment or thrombosis prevention. Nitric oxide (NO) or thrombomodulin, both of which are integral to maintaining homeostasis in the blood vessel, can be released from the polymer backbone itself or attached t o the surface . 43 Gene eluting stents are capable of delivering biologically active, therap eutic genes in an effort to reduce restenosis, accelerate re endothelialization and reduce thrombosis. Another set of stents, termed biologically active stents, incorporate antibodies or proteins, such as CD34 antibodies, onto the surface to attract endoth elial progenitor cells, or the Arg Gly Asp peptide sequence, which also attracts endothelial progenitor cells, speeding up the re endothelialization process . 30, 133

PAGE 45

32 In order to mimic naturally occurring conditions in the blood vessel, there has been some work in functionalizing the surface of the polymer with the arginine glycine aspartate ( RGD) sequence, a protein commonly found in the native extracellular matrix (ECM) . 134, 135 Hegemann et al. , determined that this environment promotes endothelial cell attachment and growth . 136 Sim ilarly, other cell adhesion peptides, such as glycine arginine glycine aspartate (GRGD), immobilized on the surface of biomaterials have displayed enhanced endothelial cell attachment . 137 Pre absorbed proteins, such as fibronectin, laminin and gelatin, present on polymer surfaces have been shown to increase cell attachment, but may reduc e cell proliferation . 138, 139 . Endothelial cells may be seeded directly onto the material prior to implantation to ensure biocompatibility . 128 To combat thrombotic events directly, adding lysine to the surface of a material has been shown to perform clot lysis, preventing blood coagulation . 140 Layer by Layer (LbL) polymer films have also been used to effectively deliver nitric oxide (NO) donor to the site of vessel injury and can hou se DNA to be delivered to the vessel wall from the covering . 1 This technique deposits both positively and nega tively charged biomacromolecules as well, and can even be used for biodegradable polymers, due its mild preparation environment . 66 Polypyrrole composites, an electrically conducting polymer, containing heparin or sodium nitrate, have the ability to switch between oxidized and reduced states, and this switching ability controls the release of biological signaling agents such as growth factors, thus directing cell growth on a surface . 141 Biomolecules attached to the surface of polymeric materials in a patterned manner may help control cell behavior or direct cell signaling . 142 A range of biomolecules have been

PAGE 46

33 immobilized on material surfaces, where the selection of biomolecule(s) is dictated by the nature of c ells to be deposited on the surface. The desire to attach (bio)pharmaceuticals and (bio)molecules to material surfaces is driven by the need provide localized delivery of the molecule or drug without changing the bulk properties of the delivery vehicle. Al though many of these techniques require reactive surface groups, further investigation into functionalization of SMPs for (bio)molecule and (bio)pharmaceutical may be desirable, especially for localized drug and molecule delivery. Porous Surfaces to Facili tate Drug Delivery As mentioned before, stents are commonly used as drug delivery vehicles to stimulate vessel healing and allow for better incorporation of the stent into the body without the use of oral anticoagulant drugs. The drugs used with the stents can be attached directly to the surface of the stent, as was discussed briefly above, or they can be incorporated into the surface of the stent using pores to house the drug until delivery. Porous stents allow for drug incorporation without an additional polymer coating that is commonly found in drug eluting stents . 1 A variety of surface modification techniques have been used to adjust surfaces for the purpose of housing drug for delivery. Etching of the polymer surface for long periods of time may cause pores to form, which can be used to house drug for localized delivery . 80, 143 The use of photolithography or soft lithography to create pores in polymer sample surfaces or to fabricate porous micro or nano particles for embedding onto a polymer surface for drug delivery has a lso been under investigation . 144 Sandblasting has been shown to effectively create porous surfaces on met al stents . 1 Aluminum coatings exposed to acidic solutions form ceramic aluminum oxide, resulting in nanoscale pores on that film for drug delivery . 1, 145 Acidic

PAGE 47

34 treatment of stainless s teel stents has also been successful in creating porous surfaces for drug elution . 145 Stents with a porous hydroxyapatite coating have exhibited promising results for drug elution as well . 1, 146 A porous surface formed by carbon nanoparticles embedded in polymer has displayed promise as a means for localized drug delivery . 147 Similarly, cobalt chromium alloy stents covered with a porous carbon carbon coating also showed potential in the arena of drug elution and enhanced cell attachment . 147 Drug delivery from pores is not solely limited to the surface; research efforts have also looked into loading drug components into the bulk of SMPs and using shape memory capabilities for drug elution . 148, 149 This is particularly app licable to SMP stents, since loading a SMP stent with drug components allows for sustained and localized drug delivery . 148 Drug release can be controlled by altering the co monomer ratio, but efforts to maintain the shape memory effect must also be considered . 149 Hydrogels, a class of polymer with swelling capabilities, have also been used for slow release, drug delivery using a diffusion mechanism through pores that often penetrate the bulk material of the hydrogel . 150, 151 Porous surfaces may allow for localized delivery of a drug or molecule without the need for prior functionalization of the surface. If functionalization for molecular tethering is not an option, forming pores in the surface for drug incorporation may prove beneficial for a range of surfaces, and might soon be extended to SMPs.

PAGE 48

35 CHAPTER II SHAPE MEMORY POLYMERS CONTAINING HIGHER ACRYLATE CONTENT DISPLAY INCREASED ENDOTHELIAL CELL ATTACHMENT Portions of t his chapter were previously published in Polymers, 2017, 9 and are included with the permission of the copyright holder . Introduction Stent materials have a greater chance of survival in vivo if endothelialization of the device occurs soon after device imp lantation . 152 Rapid endothelialization is often characterized as significant cell presence within 24 hours of im plantation, with full cellular confluence achieved after 3 to 7 days . 153 Although drug eluting stents (DES) have decreased incidence of restenosis, they typically do not achieve rapid endothelialization, which may limit long term utility . 41, 154 156 As such, focus has turned to modifying the surface characteristics of stents to promote natural endothelialization. A variety of surface modification techniques including physical, chemical, and biological methods have been evaluated on stent materials including metals and polymers . 115, 137, 139, 155, 157, 158 From a device development and regula tory perspective, it is simpler to utilize the topography rather than using chemical or biological modification; in other words, a simpler regulatory path means simpler manufacturing. Endothelialization of materials for successful integration of implanted biomedical devices was studied as early as the 1970 s . 77, 159 Endothelialization may occur by binding circulating endothelial progenitor cells or through endothelial cell migration from adjoining endothelium . 152, 160 Once endothelial cells attach, usually within the first 24 hours after device implantation, healthy cells proliferate, forming and retaining a permanent endothelial barrier on the surface of the device, resulting in reduced risk of long term device rejection. Thus, if surface characteristics

PAGE 49

36 of the device allow quick recruitment and proliferation of the endothelial lining after implantation, the chances of post implant problems should decrease . 161 Rapid endothelialization also increases hemocompatibility, another imperative for successful integration of a cardiovascular stent . 31 The potential for surface modification to enhance biocompatibility has led to increased interest in endothelialization studies, particularly for cardiovascular implants . 161 Shape memory polymers (SMPs) are one class of materials that are being considered for use in implanted, blood contacting devices . 18 SMPs are smart materials that recover their original shape upon the application of an external stimulus . 60, 162 168 These smart plastics are initially fabricated into their permanent shape and then are deformed and fixed into a temporary shape. These materials recover their original shape upon exposure to a stimulus such as heat, light, humidity, electrical or magnetic fields, among others . 66, 164, 168 170 Their ability to recover from large deformations makes SMPs appealing as materials for biomedical devices, since such recovery allows implantation of these devices using minimally invasive techniques. Progress in SMP research is not limited to biomedical applications. Developments in information carriers for one time identification, aerospace applications, smart textiles, polymer actuators and sensors, active assembly/disassembly are also notable applications of SM Ps . 46, 166, 171 174 High and low temperature SMPs are being developed for extreme environments, such as jet propulsion and aerospace applications . 175, 176 Smart textile applications using SMPs range from aesthetic improvements such as appeal, color changing capabilities, soft display to functional applications such as comfort, controlled drug release, wound monitoring, emotion sensing, extreme environment protection, etc. 171 SMP actuators may be employed in adjustable rotation rate heat engines or self regulating sun protectors for buildings . 46 Active assembly and

PAGE 50

37 disassembly should simplify and au tomate processing procedures, resulting in high speed, low cost disassembly, rendering parts useful for additional life cycles . 174 Since the shape memor y effect, which drives the shape memory capabilities in SMPs, is a result of polymer structure and processing , 31, 162 prior work in biomaterials has focused on tuning bulk properties to meet the requirements of various medical applications . 47, 60, 163 Previous work from our group has primarily focused on thermomechanical properties such as shape recovery, the shape memory effect (SME), and modulus, as well as cytotoxicity, but the cytocompatibility of these acrylate based SMPs has not been investigated . 18, 47 , 165, 177 In addition, there have been very few studies evaluating surface characteristics of SMPs in the context of endothelial growth . 66, 169 The purpose of this study was to examine the relationship between polymer characteristics of a well studied acrylate based SMP and endothelial cell attachment and growth. This would represent the first step in evaluating the potential for these materials for blood contacting devices. Various compositions of SMP containing different weight percent ratios of tert butyl acrylate (tBA) and poly(ethylene glycol) dimethacrylate (PEGDMA) were tested for endothelial cell attachment in vitro . The rapid en dothelialization target for this study was high live cell presence after 24 hours and complete cell sheet formation 72 hours after cell seeding. Materials and Methods Shape memory polymers were formulated using tert butyl acrylate (tBA) and poly(ethylene glycol) dimethacrylate (PEGDMA) with average molecular weights (M n ) of 550, 750 and 1000 with polymerization facilitated by photoinitiator 2,2 dimethoxy 2 phenylacetophenone (DMPA). All products were obtained from Sigma Aldrich (St. Louis, MO, USA), exce pt for PEGDMA1000, which was obtained from Polysciences (Warrington, PA,

PAGE 51

38 USA) and were used as received. A total of nine polymer solutions were prepared from these monomer components (Figure 2. 1). Figure 2.1 . SMP formulation matrix of the nine formulations used. Monomer mixtures were injected into molds composed of standard microscope slides (Thermo Fisher Scientific, Waltham MA, USA) separated by a 1.33mm silicone spacer (Mcmaster Carr, Elmhurst, IL, USA) and cured under ultraviolet (UV) radia tion of wavelength=365nm for 20 minutes, similar to previous methods . 47 The samples we re then removed from the molds and post cured in an oven at 75°C overnight, similar to methods done previously . 18 Samples were post processed at 75°C in an o ven overnight prior to use in characterization or cell attachment studies. Post processing steps including annealing, which generated consistent physical properties and reduced material defects. Complete conversion of monomers was verified using Fourier tr ansform infrared spectroscopy (FTIR), as has been done previously . 165 FTIR samples were fabricated under similar condi tions to those described above,

PAGE 52

39 but were made thinner, approximately 0.005mm, to allow the IR signal to penetrate the sample for FTIR analysis. Dynamic Mechanical Analysis (DMA) was performed using a TA Q800 DMA (TA Instruments, New Castle, DE, USA) and wa s used to verify glass transition temperature (T g ) of the various SMP formulations . 178, 179 All samples were cut into specimens wi th dimensions of 20mm×5mm×1mm for testing. Each sample was equilibrated to 0°C for 1 minute and heated to 100°C at a rate of 3°C/minute. Testing was conducted at a frequency of 1.0 Hz and cyclic strain control at 0.1% strain. A Ramé Hart goniometer (Ramé H art, Succasunna, NJ, USA) was used to obtain contact angle measurements and wettability of each SMP sample . 180 Wettability of each formulation was measured by applying water droplets to each surface and measuring the angle that formed between the water droplet and the surface of the sample. Measurements were taken 10 seconds after the 5 µL wa ter droplet was introduced to the surface of the SMP to maintain consistency. Contact angles were measured using DROPImage Advanced computer software (Ramé Hart, Succasunna, NJ, USA). Three different samples were analyzed per SMP formulation. Five drops we re applied to each SMP sample surface and at least five measurements were taken per drop. Surface topography, a measure the surface roughness of each SMP formulation, was obtained using atomic force microscopy (AFM) . 181 SMP fabrication molds were made using new glass microscope slides as done previously, which were cleaned using detergent and diH 2 O, followed by ethanol and acetone and a final rinse using diH 2 O to remove any surface artifacts on the glass. These measures were taken to ensure that the surface features detected by the AFM were a result of the changes in weight percent or molecular we ight of the PEGDMA. Images were obtained using a NanoSurf easyScan 2 (Nanomaterials Characterization Facility, University

PAGE 53

40 of Colorado, Boulder). Image post processing was completed using Gwyddion open source software (Gwyddion, Brno, Czech Republic). The root mean square roughness coefficient, R q , measured by the standard deviation of the distribution of surface heights, also obtained from Gwyddion, provided quantitative information of the sample surface . 182 Human umbilical vein endothelial cells (HUVECs), obtained from the endothelium of the umbilical vein, are a common cell model for angiogenesis and re endothelialization studies. HUVECs are also robust, making them a favorable cell type for use in such studies and as a result, HUVECs were the chosen cell model for this re endothelialization study . 161 Prior to cell culture experiments, human umbilical vein endothelial cells (Lonza, Walkersville, MD, USA) were seeded in T 75 flasks using comp lete growth medium: EBM 2 Cell Culture Bullet Kit (Lonza, Walkersville, MD, USA). HUVECs were maintained in conditions of 37°C and 5% CO 2 in a humidified incubator. Cells were washed with HEPES, 1M, buffer (Life Technologies, Carlsbad, CA, USA) prior to ch anging of the media. Media was changed every two to three days, and cells were passaged at 80 90% confluence. Cell passages two through six were used for cell seeding on SMP substrates. All experiments were conducted in triplicate. To monitor general cell health and ensure that there were no signs of contamination present, HUVECs and HUVEC SMP samples were observed under an inverted light microscope (Nikon, Melville, NY, USA) daily. SMP samples were submerged in growth media and equilibrated to 37°C for 24 hours prior to cell seeding. HUVECs were then plated on 1cm diameter SMP substrates in 24 well plates and allowed to attach. Cells were seeded at a seeding density of 1× 10 5 cells/mL per well. Daily monitoring of cell adherent SMPs using transmission micro scopy allowed qualitative

PAGE 54

41 assessment of proper cell growth and absence of contamination. Cell viability was quantitatively assessed at two time points, approximately 24 hours after plating and again at approximately 72 hours. Complete cell medium was chan ged daily to ensure cells received consistent nourishment during the study. The Live/Dead Cell Imaging Kit (488/570) (Life Technologies, Carlsbad, CA, USA) was used to assess endothelial cell attachment and viability through fluorescent staining. Live cell s, which were actively attached to the substrate, emit green fluorescence, while dead cells fluoresce red. Images were obtained using an EVOS FL Cell Imaging System (Life Technologies, Carlsbad, CA, USA). At least three images from three replicate experime nts were used for cell attachment counting using ImageJ software (NIH). PrestoBlue® Cell Viability Reagent (Life Technologies, Carlsbad, CA, USA), a plate based resazurin assay, was added to cell substrate samples and left to incubate for 2 hours. This cel l viability reagent, when added to cells, exploits the reducing power of cells to quantitatively measure cell metabolism. This also provides an indirect measurement of cell proliferation and cytotoxicity . 183 Media was removed from samples after the two hour incubation period and fluorescence was measured on a Synergy 2 microplate reader (BioTek , Winooski, VT, USA). The data were expressed as mean ± standard deviation. Statistical analysis was performed level of significance set Test assessed the significance between individual samples if ANOVA determined significance of the sample set. Results Select bulk properties of SMP formulations were characterized using storage modulus mechanical analysis (DMA) . 179 The plateaus above and below T g on the storage modulus curves represent the glassy and rubbery moduli,

PAGE 55

42 respectively. As shown in Figures 2A 2C, the glassy regions are absent for most formulations containing 50 weight percent tBA or less. Rubbery modulus increases with increasing PEGDMA content, as shown previously . 184 The peak of the tan delta curve was used to determine the activation temperature or glass transition temperature, T g , which is the temperature at which the material can revert from its temporary shape back to its permanent shape. The onset of shape recover y, T onset , which is the beginning of the shape recovery transition, as well as the T g range, was calculated using methods described previously and are displayed in Table 2. 1 . 18 DMA data for some of the samples has been analyzed by our group in previous experiments; our data agreed with prior results . 18, 47, 179, 184 Figure 2. 2 . (A) Storage Modulus (MPa) of tBA:PEGDMA 550 (B) Storage Modulus (MPa) of tBA:PEGDMA750 (C) Storage Modulus (MPa) of tBA:PEGDMA1000. Glassy regions are absent for the formulations containing 50 wt% tBA or less. (D) Tan delta of tBA:PEGDMA550. (E) Tan delta of tBA:PEGDMA750 (F) Tan delta of tBA:PE GDMA1000. T g increases with increasing tBA content and decreases with increasing PEGDMA MW.

PAGE 56

43 Increased monomer (tBA) content resulted in increased T g , so (T g 80:20 wt% tBA:PEGDMA) > (T g 50:50 wt% tBA:PEGDMA) > (T g 20:80 wt% tBA:PEGDMA) for a given crosslinker. Decreasing the molecular weight of the PEG component in PEGDMA also increased T g , which indicates that samples containing PEGDMA550 had higher T g than samples containing PEGDMA750, which in turn had higher T g than samples containi ng PEGDMA1000, for a given weight percent ratio (Figure 2D 2F). Table 2 .1 : T g , T onset and T g range for SMP Formulations. Data presented as µ ± SD. Contact angle increased with increasing tBA content for a given crosslinker (Figures 3A 3C). Thus, (CA 80:20 wt% tBA:PEGDMA) > (CA 50:50 wt% tBA:PEGDMA) > (CA 20:80 wt% tBA:PEGDMA formulations). S pecifically, water contact angles increased 11% 23% from the 20:80 wt% tBA:PEGDMA formulations to the 80:20 wt% tBA:PEGDMA formulations and 7% 22% between the 50:50 Formulation T g (°C) T onset (°C) T g Range (°C) 20:80 tBA:PEGDMA1000 6 ± 2 20:80 tBA:PEGDMA750 11 ± 1 20:80 tBA:PEGDMA550 25 ± 1 15 ± 2 19 ± 5 50:50 tBA:PEGDMA1000 10 ± 1 8 ± 3 3 ± 5 50:50 tBA:PEGDMA750 19 ± 2 12 ± 1 13 ± 4 50:50 tBA:PEGDMA550 44 ± 1 25 ± 3 38 ± 7 80:20 tBA:PEGDMA1000 44 ± 1 26 ± 3 37 ± 4 80:20 tBA:PEGDMA750 52 ± 1 35 ± 1 32 ± 1 80:20 tBA:PEGDMA550 60 ± 3 47 ± 3 24 ± 2

PAGE 57

44 wt% tBA:PEGDMA and the 80:20 wt% tBA:PEGDMA groups. Additionally, wettability decreased wit h increasing crosslinker length for a given weight percent of crosslinker, i.e. samples containing PEGDMA1000 were more hydrophobic than those containing PEGDMA550. Figure 2. 3 . A. Wettability of tBA:PEGDMA550. B. Wettability of tBA:PEGDMA750, C. Wettability of tBA:PEGDMA1000. Wettability decreases (increasing hydrophobicity) with increasing tBA content and increasing crosslinker (PEGDMA) MW. Significance was determined using on e way ANOVA to determine significant differences between samples of a individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds t o p < 0.001 . AFM imaging was used to assess the topographical features present on each SMP surface, quantified using the root mean square surface coefficient, R q . As seen in Figures 2. 4A A B C

PAGE 58

45 D, roughness increased with increasing tBA content, so 80:20 wt% tBA :PEGDMA formulations were roughest while the 20:80 wt% tBA:PEGDMA formulations were smoothest for a given crosslinker. Roughness increased 73% 95% between the 80:20 wt% tBA:PEGDMA group and the 20:80 wt% tBA:PEGDMA group and increased 23% 68% from the 50:5 0 wt% tBA:PEGDMA formulation to the 80:20 wt% tBA:PEGDMA formulation. Additionally, samples containing PEGDMA1000 were rougher than those containing PEGDMA550 for a given weight percent ratio . A B C

PAGE 59

46 Figure 2. 4 . A. Representative AFM images of tBA:PEGDMA550 samples. 3D AFM images depict increases in surface roughness as tBA increases. tBA:PEGDMA750 and tBA:PEGDMA1000 AFM images follow a similar trend. B. Root mean square roughness (Rq) of tBA:PEGDMA550. C. Root mean square roughness (Rq) of tBA:PEGDMA750. D. Root mean square roughness (Rq) of tBA:PEGDMA1000. Root mean square roughness (Rq) generally increases with increasing tBA content and increasing PEGDMA MW. Significance was determined using one way ANOVA to determine significant differences between samples of a individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. Cell viability, characterized as endothelial cell attachment on top of the SMP substrate, was monitored using both light and fluorescence microscopy. Results for SMP formulations containing the lowest amount of tBA (20 weight percent) are shown in Figure 2 . 5. These samples displayed little or no live HUVEC presence 24 hours after cell seeding, but the presence of dead cells was prevalent indicating that few cells survived after 72 hours. D

PAGE 60

47 Figure 2. 5. Live Dead Analysis of SMP formulations with the lowest weight percent of monomer (20 wt% tBA). These samples show little to no endothelial cell attachment and have a high presence of dead endothelial cells. Scale bar = 400 µm. SMP formulations containing equal weight percent monomer and crosslinking a gent, 50:50 wt% tBA:PEGDMA, displayed the greatest variability in endothelial cell viability. These formulations showed endothelial cell presence 24 hours after HUVEC introduction, but viability and cell attachment decreased 72 hours after cell introductio n.

PAGE 61

48 Figure 2. 6. Live Dead Analysis of SMP formulations with equal weight percent monomer (tBA) and crosslinker (PEGDMA). There are endothelial cells present on the surface of all samples regardless of crosslinker length, but there is some variability based on the crosslinker used in the sample. Specifically, both PEGDMA550 and PEGDMA750 samples seem to support more HUVEC attachment compared to the PEGDMA1000 sample. Scale bar = 400 µm. SMPs with the highest tBA content, 80 weight percent, showed the highest amount of endothelial cell attachment, displaying 4% 89% greater endothelial cell presence 24 hours after cell introduction and 33% 100% increased cell presence after 72 hours compared to the other formulatio ns. These samples also had the highest ratio of live cells to dead cells. Figure 2. 7. Live Dead Analysis of SMP formulations with highest weight percent (80 wt%) monomer (tBA). Endothelial cell attachment is indicated by the high number of living cells an d the low number of dead cells present on the samples. Scale bar = 400 µm. The 80:20 wt% tBA:PEGDMA 1000 sample initially displayed less endothelial cell attachment compared to the other formulations with 80 weight percent monomer, but after 72 hours, cell presence increased, an indication of healthy endothelial cells. The 80:20 wt% tBA:PEGDMA750 formula tion supported cell attachment 24 hours after HUVEC introduction, and was able to retain most cells after 72 hours. The final sample, 80:20 wt% tBA:PEGDMA550, displayed HUVEC attachment 24 hours after cell seeding, and was able to

PAGE 62

49 retain cell attachment 72 hours after initial introduction. All samples containing 80:20 wt% tBA:PEGDMA had few dead cells present, if any. We found that EC attachment occurred on samples containing at least 50 weight percent tBA, as seen in Figure 2. 8A. However, even though these formulations displayed endothelial cell attachment, the samples did not display cell sheet formation after 72 hours. The largest ratio of EC coverage on a sample compared to the tissue culture plate control, which displayed a full endothelial cell sh eet after 72 hours, is approximately 0.4, as displayed in Figure 2. 8B. A

PAGE 63

50 Figure 2. 8 . (A) cell count of HUVECs present on each sample, scaled to size of the SMP sample. Living endothelial cells are present on sample containing at least 50 wt% tBA. (B) endothelial cell count of each SMP sample normalized to endothelial cell count of contro l sample (TCPS). While ECs attach to SMP surfaces, full coverage of SMP samples is yet to be achieved. B

PAGE 64

51 Cell metabolism was measured daily for 72 hours, with results displayed in Figure 2. 9. Resazurin, which is initially non fluorescent, is reduced to a fl uorescent resorufin when added to healthy cells. Increases or decreases in reduction provide insight into cell health, such as metabolism and cytotoxicity . 183 Samples containing 20 wt% tBA did not show any signs of resazurin reduction, further confirmin g that if any cells were present on the samples, the cells were unhealthy, dying, or already dead. Most of the samples containing 50 wt% tBA and 80 wt% tBA showed signs of increasing resazurin reduction, which may be an indication of increased endothelial cell presence, and consequently, possible cell proliferation. Figure 2. 9 . Cytocompatibility of SMPs. There is evidence of increasing metabolic activity, prominently 72 hours after cell introduction, but some samples show evidence of metabolic

PAGE 65

52 activity inc reasing just 48 hours after cell seeding. Samples that are cytotoxic have little metabolic activity compared to samples that are cytocompatible, confirming a lack of EC presence. Significance was determined using one way ANOVA to determine significant diff erences between samples of a given wt% ratio and crosslinker combination, in addition to the p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. Discussion Current stent technologies have found broad clinical utility but continue to encounter issues such as restenosis and/or thrombosis, both of which may require subsequent reintervention to prevent further complications. While there has been extens ive work on fine tuning bulk mechanical properties ofSMPs and brief studies on cytotoxicity and biocompatibility for applications such as hernia meshes, embolic coils, and stent grafts from our group as well as others, little work has focused on surfaces, and more specifically, cytocompatibility and hemocompatibility of these materials . 47, 66, 177, 185, 186 Previous work from our group has addressed bulk mechanical properties such as thermomechanical behavior, the shape memory effect, partially constrained and free recovery, biocompatibility and cytotoxicity of these tBA:PEGDMA SMPs. However, previous studies have not reported on surface properties or endothelial cell attachment on the surface of these materials . 18, 47, 165, 177, 179 Endothelialization of implanted biomedical devices increases the likelihood of devic e integration due to improvements in hemocompatibilty and reduced risk of device rejection, which necessitates optimization of surfaces to encourage EC attachment . 18, 31, 161 This study evaluated the ability of select acrylate based shape memory polymers to attach surface and its surround ings play a notable role in dictating the success of an implanted biomedical device . 4, 5, 61, 128 While previous studies from our lab have assessed bulk mechanical

PAGE 66

53 properties of acrylate based SMPs for stent use, no studies have evaluated surface characterist ics of these SMPs in detail. Our group has previously examined the activation temperature for the SMP formulations used in this study . 18, 47, 178 Briefly, the 80:20 wt% tBA:PEGDMA1000 and the 50:50 wt% tBA:PEGDMA550 samples have glass trans ition temperatures closest to body temperature, with T g T g physiological tem perature . 178 Prior data has shown that varying the crosslinking agent between 10% and 40% does not significantly impact T g for related crosslinking agents . 18 However, wh en crosslinker content exceeds 40% weight percent, the transition regime between glassy and rubbery becomes blurry and depletes shape memory ability, which may cause the larger differences in T g displayed here . 187 Additionally, glassy modulus and the transition between glassy to rubbery state is nearly non existent for formulations containing 50 wt% or less, which also demonstrates the depletion of the shape memo ry ability. Increases in rubbery modulus, and reductions in stiffness, are seen as PEGDMA content increases, agreeing with prior data . 18 All of the samples c ontaining 20:80 wt% tBA:PEGDMA as well as the 50:50 wt% tBA:PEGDMA750 and the 50:50 wt% tBA:PEGDMA1000 formulations have considerably lower T g breakage when deformed at room temperature due to a higher rubbery modulus and reduced stiffness an d thus would not exhibit the shape memory effect at 37°C. SMP formulations with higher T g the 20:80 wt% tBA:PEGDMA 1000 sample, with a glass transition temperature of approximately 6°C, exhibited a storage modulus of 11.27MPa, whereas the 80:20 tBA:PEGDMA550 SMP, with a glass transition temperature of 60°C, had a storage modulus of 1194 MPa. This confirms

PAGE 67

54 that the samp les with T g temperatures greater than 37°C. The large difference in storage modulus between the two formulations provides additional insight into material properties at 37°C, which could b e an important factor in material choice. For a device that is going to serve as a support mechanism, higher stiffness may be more desirable, as would be the case for a cardiovascular stent in a blood vessel. Numerous studies have cited substrate stif fness as an important factor in determining cell attachment to a substrate . 188, 189 These studies found that stiffer samples often exhibit higher endothelial cell attachment compared to their softer coun terparts . 186 Our data affirm these prior results. Stiffer SMPs, or those with activation temperatures above 37°C, displayed grea ter endothelial cell attachment and viability. Specifically, all of the 80:20 wt% tBA:PEGDMA formulations as well as the 50:50 wt% tBA:PEGDMA550 sample exhibited endothelial cell attachment and retained attached cells for up to 72 hours. Thus, SMPs with T g slightly higher than body temperature, and therefore increased stiffness, appear well suited for endothelial cell attachment, similar to results found in other studies . 188 Contact angle measurements provided quantitative wettability data of each surface . 180 Surfaces exhibiting moderate wettability have shown a higher affinity for cell attachment compared to surfaces with extreme wettability . 161 Formulations with higher PEGDMA content are more hydrophilic, as are formulations with lower molecul ar weight PEG chains, i.e. samples containing PEGDMA550. PEGDMA is commonly used in hydrogels for its hydrophilic tendencies as well as its highly tunable material properties, which explains the smaller contact angles for samples containing higher amounts of PEGDMA . 190 Since the formulations were based on weight percen t ratios of tBA and PEGDMA, the greater hydrophilicity of samples

PAGE 68

55 containing PEGDMA550 could be the result of increased PEG presence in the sample compared to a sample containing PEGDMA1000. Despite these trends in wettability, the differences in wettabili ty did not appear significant enough to have a pronounced effect on endothelial cell attachment to these surfaces. This is analogous to results found in other studies where changing PEG length produced large variations in cell attachment . 191, 192 Surface roughness was also analyzed for each sample. Roughnes s increased with increasing tBA content and with molecular weight. Rougher surfaces also had higher contact angles, and have been shown to be more conducive to cell attachment, which is consistent with other studies . 75, 76, 83 Even though neither surface roughness nor wettability is a sole deciding factor of HUVEC attachment to the SMP surfaces, both aspects have shown to affect endothelial cell attachment. It is often unclear which factor may play the dominant role in cell attachment, due to the complex nature of cell surface inte ractions, as seen in other studies . 116 HUVEC attachment was assessed using fluorescence microscopy. Formulations containing the lowest weight percent monomer, 20:80 wt% tBA:PEGDMA, displayed very little cell attachment within 24 hours, leading to minimal or no HUVEC presence 72 hours after cell introduction. Since these formulations have high PEG content, and PEG has been shown to resist protein and cell attachment, the presence of dead cells or lack of cells appears reasonable . 144 However, recent studies have also indicated that some PEGDMA based hydrogels may support cell adhesion of certain cell types, requiring the consideration of these low tBA materials . 193 The absence of dead cells between 24 hours and 72 hours can be explained by the tendency of dead endothelial cells to detach because there is no active mechanism for dead cells to remain tethered to the surface of the SMP. The dead cells are then removed when samples are washed with buffer or cell culture medium is replenished.

PAGE 69

56 Formulations containing equal amounts of tBA and PEGDMA displayed the greatest amount of variation in endothelial cell attachment. The 50:50 wt% tBA:PEGDMA1000 sample behaved more like formulations containing 20 weight percent tBA, supporting little HUVEC attachment initially and showing a decrease in HUVEC presence after 72 hours, indicating that this formulation may not support endothelial attachment. The 50:50 wt% tBA:PEGDMA750 a nd the 50:50 wt% tBA:PEGDMA550 samples behaved more similarly to formulations containing 80 weight percent tBA, displaying HUVEC attachment 24 hours after cell seeding and retaining a small number of attached cells after 72 hours. The SMP formulations con taining 80:20 wt% tBA:PEGDMA exhibited the greatest HUVEC attachment. These samples displayed cell attachment and retained endothelial cells 72 hours after initial cell seeding. Some samples even showed indications of increased EC presence, which may indic ate cell proliferation. Thus, these formulations may be good candidates for use in implanted devices that require rapid endothelialization to succeed, such as cardiovascular stents. Further work, preferably in vivo , would be required, to confirm that these formulations would be good candidates for stent fabrication. The SMP samples used in this study were solid surfaces, whereas stents are often perforated tubes, which have been fabricated by our group in previous work . 183 Perforated SMP stents are easy and inexpensive to fabricate, unlike metals, and these perforated SMPs may experience greater endoth elial cell surface coverage due to migration of ECs from adjoining healthy endothelium in addition to endothelial progenitor cell (EPC) attachment and should verified in subsequent studies. Due to the exploratory scope of this initial study, additional exp eriments evaluating the effect of surface roughness on endothelial cell attachment and viability were not included but will be conducted in future studies. Protein adsorption to the

PAGE 70

57 sample surface from the culture medium has been shown to encourage cell at tachment and SMPs that displayed cell attachment may have demonstrated more selective protein adsorption from the cell culture medium, allowing ECs to attach long enough to produce their own adhesion proteins, which should be verified in future work . 21, 69, 116, 194 Finally, in vivo studies would help confirm the in vitro results, increasing interest in further investigating these materials for blood contacting devices.

PAGE 71

58 CHAPTER III SMP FORMULATION OPTIMIZATION FOR ENDOTHELIALIZATION: INCRE ASED ACRYLATE CONTENT IN SHAPE MEMORY POLYMERS Introduction We have previously demonstrated successful endothelial cell adhesion on SMPs with at least 50 wt % tBA up to 80 wt % tBA monomer content. The higher tBA content contribute d to the increased stiffness, r oughness and hydrophobicity of the SMP surface, which appears t o contribute favorably to endothelial cell recruitment and retention. As such, in order to determine the optimal SMP formulation for EC attachment, and since previous data suggests that increas ed tBA i n the SMP corresponds to increased cell adhesion, it is necessary to further increase the tBA content in the SMPs beyond 80 w t% and analyze the effect that this increase has on subsequent cell adhesion . 195 Based on prior data, en dothelial cells seem to prefer high tBA SMP formulations . 195 Since the previously analyzed formulations contained maximum 80 weight percen t tBA, it would be advantageous to further increase the tBA content in the SMPs in increment s of 5 wt%, ranging from 85 wt% tBA to 95 wt% tBA , with PEGDMA comprising the remaining formulation weight. The increased tBA content may contribute to stiffer, rougher, and more hydrophobic surfaces ; if the previous trend continues , these materials may provide a more optimal environment for endothelial cell adhesion . The effect of further increasing tBA monome r, not only on the material itself, but also on endothelial cell adhesion and survival , will be investigated in this follow up study.

PAGE 72

59 Materials and Methods Shape memory polymers were again formulated using tert butyl acrylate (tBA) and poly(ethylene gl ycol) dimethacrylate (PEGDMA) with average molecular weights (M n ) of 550, 750 and 1000 with polymerization facilitated by photoinitiator 2,2 dimethoxy 2 phenylacetophenone (DMPA). All products were used as received. The formulations analyzed during this part of the study are displayed in Figure 3. 1. Figur e 3. 1 . High tBA content SMP Formulations. Formulations contain high weight percen tage s of tBA, ranging from 85 95 wt%. DMA, AFM, contact angle as well as the cell adhesion study using Live/Dead Imaging and cell metabolism , were performed similar to methods performed for the original SMP formulations . Every effort to ma intain method consistency was made so that the results could be verified against original data. The data for the high tBA wt% were compared against those from the original SMPs to determine the optimal SMP formulation for future studies and use.

PAGE 73

60 Results The glass transition temperature (T g ) was measured for the additional SMP formulations and is presented in Figure 3. 2. Since the increasing increments of tBA are smaller than previously studied, increases in T g are less pronounced compared to the original formulations , as increases are approximately 10 20% for the high tBA formulations whereas the original formulations experienced increases ranging from 40 80% due to the large increments in tBA or PEGDMA . Figure 3. 2 . Glass t ransition t emperature (T g ) continues to increase with increasing tBA content. Significance was determined using one way ANOVA to determine significa nt differences Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0 .001.

PAGE 74

61 Notably, the increases in T g are greater as the MW of PEGDMA also increases, due to the reduced effect of PEGDMA as a result of the reduced weight percent. These results agree with trend that was previously observed for SMPs containing lower tBA amounts . 178, 179, 195 Hydrophobicity / wettability of the SMPs was measured using the static contact angle method and assessed against the previously analyzed SMPs, as shown in Figure 3 .3 . Figure 3 .3 . Contact angle of SMPs. Increasing tBA monomer typically increases water contact angle, which indicates hydropho bicity increases as tBA content increases. Significance was determined using one way ANOVA to determine significant differences between samples of a Difference Proc edure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.

PAGE 75

62 Overall, the increase in weight percent tBA corresponds to a more hydrophobic SMP surface , which agrees with results obtained previously . Again, the increase s in hydrophobicity are subtle , but still present with increasing tBA content. To further characterize the surfaces of the high tBA SMPs, surface roughness was measured, again, using atomic force microscopy (AFM). Roughness was quantified using the root mean square surface coefficient, R ms which is displayed in Figure 3.4. Figure 3.4 . SMP surface roughness measurements. Increasing the weight percent of tBA has variable effects on surface roughness. There are clear increases in roughness between the highest tBA content SMP tested and the original SMP test samples, but trends are difficult to discern. Significance was determined using one way ANOVA to determi ne significant differences Honest Significance Difference Procedure between individual samples. * corresponds to p < 0.05, ** corresponds to p < 0.01, *** correspo nds to p < 0.001.

PAGE 76

63 Increasing the weight percent of tBA has variable effects on surface roughness. There are generally clear increases in roughness between the highest tBA content SMP (95:5 tBA:PEGDMA) tested and the original SMP test samples (20 80 wt% tBA ) , but there do not appear to be any clear trends correlating roughness and tBA content. HUVECs were introduced to the high tBA SMPs and allowed to attach . Cell adhesion was monitored for up to 72 hours after cell introduction. Cell viability was again assessed using the Live/Dead Imaging Kit, where live cells are marked with green fluorescence and dead cells fluoresce red. Cell viability for all formulations are shown in Figures 3.5 3.7 . Figure 3.5 . Endothelial cell adhesion on 85:15 wt% tBA:PEGDMA SMPs. ECs appear to demonstrate similar adhesion tendencies to the 80:20 wt% tBA:PEGDMA SMPs.

PAGE 77

64 Figure 3.6. Endothelial cell adhesion on 90:10 wt% tBA:PEGDMA SMPs. ECs continue to adhere to the surface . Figure 3.7. Endothelial cell adhesion on 95:5wt% tBA:PEGDMA SMPs. ECs continue to adhere to the surface, but adhesion appears to decrease on these high tBA SMPs, which may be an indication of reduced surface compatibility.

PAGE 78

65 Endothelial cell attachment to the 85:15 wt% tBA:PEGDMA SMPs appears to be similar to the attachment behavior of the ECs on 80:20 wt% tBA:PEGDM A, however as tBA content increases, the surfaces display fewer adherent cells. This would need to be confirmed q uantitatively however, as shown in Figur e 3.8. Cell adhesion appears t o be maximized on 80 85 wt% tBA:PEGDMA SMPs with less cell adhesion on surfaces with lower (50% tBA or less) or higher (90% tBA or more) . Figure 3.8 . Estimated adherent ECs on SMPs. ECs appear to prefer t he 80 85 wt% tBA SMPs. EC adhesion on SMPs containing greater than 90% tBA or 50 wt% or less still support cell

PAGE 79

66 Figure 3.9 . Cell metabolism of cell adherent SMPs . Cell metabolism typically increases for a given SMP, which may suggest proliferation. Cell metabolism was measured daily for 72 hours , as shown in Figure 3.9 and compared to the cell metabolism of the original formulations. Again, there are increases in cell metabolism between day 1 and day 3 for most of the formulations , which is an indication of cell proliferation. It should be noted, however, that statistically significant increases in cell metabolism are less evident for the high tBA SMPs. Additionally, ce ll metabolism is still significantly lower compared to treated TCPS.

PAGE 80

67 Discussion This study is an extension of a previous study, one that suggested tBA:PEGDMA based SMPs with greater monomer (tBA) content demonstrated improved endothelial cell adhesion capabilities compared to SMPs containing higher PEGDMA content. This work specifically aimed to optimize the tBA:PEGDMA based SMP formulation for endothelial cell adhesion so that it may be potentially used for cardiovascular sten t fabrication , due to the reduced complications that result from cell sheet formation on implanted blood contacting devices . Thus, b y further increasing the tBA content, the effect on cell adhesion should be assessed . Increasing the tBA content in the SMPs continues to increase the glass transition temperature (T g ) of the SMPs. Even though the increase in tBA is small , there are significant differences in glass transition temperature between , for example the 85:15 wt% tBA:PEGDMA and the 90:10 wt% tBA:PEGDMA samples. Although the T g is greater, shape memory at physiological temperature still occurs, but at a less rapid rate . The increase in T g also resulted in increased stiffness which , if the trend follows, should also be favorable to cell adhesion . 188, 196 Studies by Shalali et al., have shown that endothelial cells prefer stiffe r hydrogel scaffolds, by exhibiting elongated, spread out cell morphologies . 196 Yeung et. al., found that adhesion to stiffer substrates may upregulate adhesion receptors . 188 Therefore, the increased T g and resulting increased stiffness would present a more favorable environment for cell adhesion. Contact angle also continues to increase for high tBA SMPs. Similar to the increase d T g , the addition of more tBA creates a more hydrophobic surface . The original tBA samples demonstrated contact angles slightly under 90°, which is representative of a slightly hydroph il ic surface, whereas the majority of these high tBA SMPs demonstrate contact angles greater than 90°, representing slightly hydrophobic surfaces. Interestingly, the roughness measurements for

PAGE 81

68 these samples did not directly correlate with contact angle measu rements. However, this may be due to the non homogenous polymerization of the SMP, caused by unreacted tBA or tBA reacting with itself. Analysis of additional samples, or analysis using ATR FTIR, which specifically analyzes surface species, is required. Unlike contact angle, surface roughness did not appear to d isplay any notable trends . Instead, the surface roughness , R ms of the formulations between 85 and 95 wt% tBA had high variation between formulations and weight percent ratios of a given formulation. This may be a result of the increased tBA amounts, which may have saturated the polymer and affected the polymerization reactio n, leading to surface non homogeneity. Since the contact angle measurements are conducted on a larger scale, these smaller effects may not have been as impactful during contact angle measurements , whereas the roughness measurements are conducted using a very fine tip on the end of the AFM canti lever, which is more sensitive to such variations. This would need to be confirmed, again with ATR FTIR , XPS , ESCA or another surface species analysis techniqu e . Based on the cell adhesion studies using Live/Dead to analyze adherent cells , the preferred formulations for the tBA:PEGDMA SMPs are in the 80 85 wt% tBA range . Formulations with 50 wt% or less tBA can support cell adhesion as shown previously , but cell presence is not as prevalent as on other samples , and cells often appear round ed and unhealth y. Additionally , at greater than 90 wt% tBA, cell adhesion is also lower; thus, it appears that the surface environment created by the 15 20 wt% PEGDMA supports cellular adhesion and survival by optimizing stiffness, roughness and wettabilit y . SMP formulations containing high amounts of monomer or crosslinker appear be sub optimal for EC adhesion and survival. Of the original formulations, which included the 20:80

PAGE 82

69 wt% tBA:PEGDMA data , these high PEGDMA formulations did not support EC adhesion or survival due to the toxic nature of the high PEGDMA . It is possible that the high tBA formulations also experienced a similar effect when formulations contained greater than 90 wt% tBA. The optimal nature of the 80 & 85 wt% is again evident when assessing the cell metabolism measurements. Cell metabolism typicall y increases over the course of the study, but fluctuations become prevalent a s tBA increases. Here, additional SMP formulations containing high weight percent ratios of tBA were assessed for cell adhesion and survival. While the se surfaces were often stiffer and more hydrophobic, ECs appeared to prefer formulations containing 80 90 wt% tBA ; formulations containing less or more tBA seemed to provide surfaces that were not ideal for cell adhesion and survival . Moving forward , SMPs with the 80:20 tBA:PEGDMA weight percent ratios will be used to further optimize the surfaces to encourage both endothelial cell adhesion and organization .

PAGE 83

70 CHAPTER IV MICROGROOVES ENCOURAGE ENDOTHELIAL CELL ADHESION AND ORGANIZATION ON SHAPE MEMORY POLYMER SURFACES Portions of this chapter were previously publ ished in ACS Applied Bio Materials , 201 9 , and are included with the permission of the copyright holder. Introduction Although cardiovascular disease (CVD) is the leading cause of death globally, treatment methods continue to be limited . 38, 197 199 Init ially, s urgical intervention was the preferred method for opening occluded vessels, but the invasiveness of surgery and lack of support to arteries frequently resulted in vessel re occlusion, limiting this approach. Balloon angioplasty and bare metal stents (BMS) provide a less invasive method for opening and supporting narrowed vessels; however, complications such as restenosis and thrombosis from reduced biocompatibility often prompt reintervention in many patients . 153, 154, 156, 157 Drug eluting stents (DE S) reduce restenosis due to the addition of an anti proliferative agent, but delayed re endothelialization and resulting late stage thrombosis continue to be prevalent issues . 41, 154, 155, 160 Many of these limitations in current stent technologies are due to material noncompliance and reduced surface patency , which result in sub optimal patient outcomes . These issues provide continue d support for research into improving stent performance . 32, 137, 200 Endothelialization of stents and other blood contacting devices facilitates proper function and integration with surrounding tissue . 157, 161 However, reduced compatibility between stents and their respective physiological environments promotes poor endothelial cell recruitment and adhesion. Endothelialization decreases device rejection and eliminates the need for long term anti platelet therap ies due to the generation of an anti thrombotic and anti proliferative

PAGE 84

71 environment created by healthy endothelial cells . 77 , 157 Current in vitro cell seeding on devices prior to implantation has resulted in reduced thrombosis and improved integration, but these methods may not b e feasible in practice due to their laborious and highly specialized nature . 32, 38, 201 In situ re endothelialization of implanted medical devices would be the preferred approach . 32, 161, 201 Me chanical, chemical, and/or topographical alterations to stent materials may promote endothelialization and are thus important considerations in the design of such devices . 35, 202 204 Chemical, physical and biofunctional surface modification methods have been shown to encourage in situ endothelialization . 65, 71, 94 In this regard, surface patterning may be particularly attractive since it not only promotes cell adhesion, but also encourages cell alignment for a variety of cell types, including endothelial cells, all with far less potential regulatory burden than that required for chemical or biological additions . 62, 205, 206 Grooves and ridges are some of the most common features used to encourage cell attachment and promote cell alignment . 207 209 Cell alignment is essential for proper biological and mechanical function in most native tissues, but many biomedical device materials do not contain the structural cues necessary to encourage such organization . 210, 211 In native blood vessel under healthy flow conditions, endothelial cells align along the direction of flow , whereas in unhealthy flow conditions, ECs demonstrate isotropic orientations, which may result in a more thrombogenic and inflammatory environment . 212 214 Shape memory polymers (SMPs) are customizable, smart materials that can recover their original shape following deformation . 58, 215 For biomedical dev ices, SMPs have been used in cardiovascular, orthopedic, and dental applications . 216 218 Our group h as previously developed SMPs for stents, embolic coils, hernia meshes, etc. 18, 47, 165, 177 179, 195 Prior research on this family of SMP s for stent use has largely focused on the mechanical properties for stent deployment and

PAGE 85

72 implantation and to confirm low cytotoxicity . 18, 165 We recently evaluated endothelial cell adhesion on the surface of these acrylate based SMPs and found that certain formulations, speci fically those containing a higher weight percent ratio of the acrylate monomer, encourage endothelial cells to a dhere without requiring surface modification . 195 However, cells on these surfaces were randomly oriented; a key next step is to evaluate whether axially oriented topographical surface modifications facilitate endothelial cell alignment. To study this further, we investigated techniques for micropatterning the surface of these acrylate based SMPs, and then evaluated the extent of endothelial cell attachment onto patterned and unpatterned surfaces. Since the unpatterned surfaces encourage cell adhes ion, micropatterning should further optimize the surface for cell adhesion and survival. Shallow microgrooves were created using 3D metal printed molds, a technique that offers a scalable, cost effective and practical approach to topographically pattern po lymerizable materials. To the best of our knowledge, surface patterning of this family of SMPs for endothelial cell adhesion and organization has not yet been studied. In this investigation, three SMP formulations with varying crosslinker molecular weights were analyzed. These particular formulations, which contained high weight percent ratios of acrylate, previously demonstrated endothelial cell adhesion capabilities, but did not contain the topographical cues needed to encourage alignment. Unpatterned and microgrooved substrate surface properties were assessed using scanning electron microscopy (SEM), atomic force microscopy (AFM) and contact angle measurements . Endothelial cell adhesion and alignment were evaluated for up to seven days using fluorescence microscopy.

PAGE 86

73 Materials and Methods Micropatterned metal plates were fabricated using an EOSINT M270 3D metal printer (EOS, Munich, Germany). The length, width and height of the metal print were specified to match the dimensions of standard microscope slide s using SolidWorks (Waltham, MA, USA). The surface of the metal printed piece consisted of approximately 50 width repeating, shallow grooves. This microgroove pattern is a result of the direct metal laser sintering method (DMLS) used by the metal printe r to manufacture parts. This pattern was subsequently used for micropatterning of the SMP surface via molding. The components for shape memory polymer fabrication, monomer tert butyl acrylate (tBA) and crosslinker poly(ethylene glycol) dimethacrylate (P EGDMA) with average molecular weights (Mn) of 550 and 750, were purchased from Sigma Aldrich (St. Louis, MO, USA). PEGDMA1000 was obtained from Polysciences (Warrington, PA, USA). Polymerization was facilitated by photoinitiator 2,2 dimethoxy 2 phenylace tophenone (DMPA), also obtained from Sigma Aldrich. Three different monomer mixtures were used for this study (i.e., one mixture for each PEGDMA molecular weight) and all were an 80:20 wt% ratio of tBA:PEGDMA, similar to previous methods 195 . Monomer mixtures were injected into molds composed of a standard microscope slide (Thermo Fisher Scientific, Waltham MA, USA) and a metal plate, separated by a 1.33mm silicone spacer (Mcmaster Carr, Elmhurst, IL, USA). The unpatterned SMP surfaces were created with a polished steel plate, while the metal printed piece was used for the microgrooved SMPs. The pre polymers were then cured under ultraviolet (UV) radiation from a Dymax Model 200 Light Curing System (Dymax, Torrington, CT, USA) of wavelength 365nm for 10 minutes pulsed (30 seconds on, 30 seconds off) followed by 10 minutes uninterrupted

PAGE 87

74 curing. The samples wer e then removed from the molds at 90°C and post cured in an oven at 75°C overnight, similar to previous methods . 195 Dynamic mechanical analysis (DMA) was then conducted on a Q100 DMA (TA Instruments, New Castle, DE , USA) to ensure that the thermomechanical bulk properties of the SMPs remained consistent between unpatterned and microgrooved samples. Three SMP samples per formulation and surface condition were cut into 20mm x 5mm x 1mm specimens, equilibrated to 0°C a nd ramped to 100°C at a rate of 3°C/min, as performed previously . 178 The surfaces of the unpatterned and microgrooved SMPs were imaged using a JEOL ASM 6010LV (JEOL USA, Peabody, MA, USA). Prior to SEM imaging, the SMPs were sputter coated for 30 seconds using a Leica EM ACE200 (Leica Microsystems, IL, USA; EM La unpatterned surfaces did not exhibit any organized surface features and to verify micropatterning and groove width of the microgrooved SMP surfaces. Surface topography a nd surface roughness were obtained using atomic force microscopy (AFM). Five different samples per formulation were cleaned with ethanol and air dried to remove any debris prior to imaging. Topographical data and images were obtained using a JPK AFM system (JPK, Berlin, Germany). Image post processing was completed using Gwyddion open source software (Gwyddion, Brno, Czech Republic). The root mean square roughness coefficient, R ms , which provides quantitative information of the sample surface. R ms , measured by the standard deviation of the distribution of surface heights of the sample, was also obtained from Gwyddion . 219 Contact angle measurements and wettabi lity of SMP samples were obtained using a Kudos Precision Instruments DropMeter A60 (Manhattan, NY, USA). Wettability of each

PAGE 88

75 formulation was measured by applying 10 µL water droplets to each surface and measuring the angle formed between the water droplet and the surface of the sample. Measurements were taken ten seconds after the water droplet was introduced to the surface of the SMP to maintain consistency. Contact angles were measured using SurfaceMeter Elements computer software (NBSI, Ningbo City, Chi na). Five different samples were analyzed per surface, per SMP formulation. Five drops were applied to each SMP sample surface and five measurements were taken per drop. O btained from the endothelium of the umbilical vein , h uman umbilical vein endothelial cells (HUVECs) are a common cell model for angiogenesis and re endothelialization studies. HUVECs were the chosen cell model for this re endothelialization study because they are robust , which mak es them a favorable cell type for use in such studies . 161 Prior to cell culture experiments, HUVECs (Lonza, Walkersv ille, MD, USA) were seeded in T 75 flasks using complete growth medium: E G M 2 Cell Culture Bullet Kit (Lonza, Walkersville, MD, USA). HUVECs were maintained in standard cell culture conditions of 37°C and 5% CO 2 in a humidified incubator. HUVECs and HUVEC SMP samples were observed daily under a Nikon inverted light microscope (Nikon, Melville, NY, USA). Cells were washed with HEPES, 1M, buffer (Life Technologies, Carlsbad, CA, USA) prior to changing media. Media was changed every other day, and cells were p assaged at 80 90% confluence. Cell passages four through seven were used for cell seeding on SMP substrates. At least three independent experiments were performed, and all experiments were conducted in triplicate. SMP substrates were submerged in ethanol, air dried and subsequently steam sterilized prior to HUVEC seeding . Steam sterilization has been previously used to sterilize acrylate based SMPs successfully without disrupting shape memory capabilities or other material properties . 165,

PAGE 89

76 178 HUVECs were then plated at a seed ing density 1× 10 5 cells/mL per well on 10mm diameter SMP substrates in coverslip bottom 24 well plates and allowed to attach. Cell adherent SMPs were monitored daily for proper cell growth and absence of contamination using transmission microscopy. Cell viability was quantitatively assessed at three time points: 1 day, 3 days and 7 days after cell introduction. Endothelial cell attachment and viability was assessed using the Live/Dead Cell Imaging Kit (488/570) (Life Technologies, Carlsbad, CA, USA). Liv e cells, which were actively attached to the substrate, emit green fluorescence, while dead cells emit red fluorescence. Complete cell medium was changed every other day during the study. Images were obtained using a Zeiss Axiovert . A1 inverted microscope ( Zeiss, Thornwood, New York, USA). At least five images from three replicate experiments were used for cell attachment counting using ImageJ software (NIH, Bethesda, MD, USA). After 1 and 7 days of sub culture, cell attached SMP samples were fixed using 4% paraformaldehyde in PBS for 10 minutes. Fixed samples were then submerged in 0.1% Triton X for permeabilization prior to staining with Anti CD31 (Abcam, Cambridge, MA, USA), which was use d per manufacturer instructions. In addition to Anti CD31, phalloidin and DAPI were also used according to manufacturer instructions. DAPI (4',6 Diamidino 2 Phenylindole, Dihydrochloride), (MilliporeSigma, Burlington, MA, USA) was used to visualize nucleus alignment, which was measured as 0° if the nucleus position was perpendicular to the groove while nuclei parallel to the groove were measured at 90°. The alignment angle was measured using ImageJ software (NIH, Bethesda, MD, USA). Filamentous F actin was visualized using Alexa 568 Phalloidin Actin Stain (Thermo Fisher Scientific, Waltham, MA, USA) and

PAGE 90

77 confirmed cell alignment. Fixed cell imaging was performed on a Zeiss AxioObserver inverted microscope (Zeiss, Thornwood, New York, USA). The data were expre ssed as mean ± standard deviation ( µ ± SD) , unless otherwise noted. Statistical analysis was performed using MATLAB (MathWorks, Natick, MA, USA) and significance was determined using a two tailed t level of significance of 0.05 when comparing u npatterned vs. micropatterned groups. When comparing more than two groups, analysis was conducted using a two Difference Test assessed the significance between individual samples if ANOVA determined significance of the sample set. Results Dynamic mechanical analysis (DMA) confirmed the consistency of bulk properties, specifically the glass transition temperature as measured by the peak of the tan delta as well as the storage modulus, between unpatterned and mi cropatterned SMPs , similar to prior results (Figure 4. 1) . 179, 195 Data confirmed that microgroove introduction does not significantly affect T g , T onset or T g range, as demonstrated in Table 4. 1. Figure 4.1 . Storage modulus and tan delta curves for 80:20 wt% unpatterned and microgrooved SMPs. Both storage modulus and tan delta display overlap between unpatterned and microgrooved surfaces.

PAGE 91

78 Table 4.1 . T g , T onset and T g range ( n = 3 ) , for unpatterned and microgrooved 80:20 wt% tBA:PEGDMA550, 750, 1000 SMPs. Scanning electron micrographs qualitatively depicted pattern transfer to the surface of micropatterned SMPs as well as the lack of periodic, pattern like surface features on unpatte rned surfaces. As shown in Figure 4. 2, the unpatterned surfaces lack repetitive surface features while the patterned surfaces exhibit repeating shallow microgrooves, confirming pattern transfer from the mold to the SMP surface. The microgroove widths, whic measured using ImageJ (NIH, Bethesda, MD, USA).

PAGE 92

79 Figure 4. 2 . SEM micrographs verified pattern transfer and surface feature presence, or lack thereof, on unpatterned and micropatterned SMP surfaces. Unpatterned surfaces exhibit topographical randomness. Micropatterned surfaces display shallow grooves with widths ra nging from approximately 55 to 60 m. Scale Bar = 100 µm. Atomic Force Microscopy (AFM) was used to quantify the topography of each SMP surface, by measurement of surface roughness . 121 The root mean square surface coefficient (R ms ), which provides a quan titative measure of surface roughness, was obtained from AFM data. Micropatterned surfaces exhibited 11 14% higher roughness compared to unpatterned SMP surfaces, but surface roughness did not demonstrate statistically significant differences between unpat terned SMPs and their microgrooved analogues (Figure 4. 3). The roughness of unpatterned SMP samples, polymerized in glass molds, has been previously reported by our group and similar results for unpatterned surfaces were confirmed . 195

PAGE 93

80 Figure 4. 3 . Roughness of unpatterned vs. microgrooved 80:20 wt% tBA:PEGDMA SMP surfaces. Microgrooved surfaces exhibit 11% 14% increased roughness compared to unpatterned SMP surfaces. A 2 tailed t test was used to determine significance between unpatterned and microgrooved surfaces, n = 5. While there is an increase in roughne ss between unpatterned and microgrooved surfaces, the differences are not statistically significant. Topographical views of unpatterned and microgrooved tBA:PEGDMA550, tBA:PEGDMA750 and tBA:PEGDMA1000 surfaces further confirm pattern transfer to microgroov ed surfaces and lack of patterning on unpatterned surfaces.

PAGE 94

81 Surface wettability was determined by measuring the contact angles formed between a contact angles close t o 90°, indicating slightly hydrophobic surfaces. Micropatterning the SMP surfaces decreased wettability by 3 6%, as seen in other studies using different materials . 220 Notably, significant differences between wettability of unpatterned vs microgrooved surfaces is confirmed and are shown in Figure 4. 4. Figure 4. 4 . Wettability of unpatterned vs. microgrooved SMP surfaces. Hydrophobicity is 3% 6% greater for microgrooved surfaces, indicating a small increase in hydrophobicity between unpatterned and microgrooved surfaces. A 2 tailed t test determined significance between unpatterned and microgrooved surfaces, *** corresponds to p < 0.001. n=5. Endothelial cell attachment and viability was assessed using Live/Dead imaging fluorescence microscopy. Live cells are displayed in green, whereas dead cells fluoresce red. NucBlue Live Cell ReadyProbes was used to mark cell nuclei as indicated in blue (Fi gure 4. 5).

PAGE 95

82

PAGE 96

83 Figure 4. 5 . Endothelial cell attachment at 1 day, 3 days and 7 days after cell introduction assessed using Live/Dead Cell Imaging and NucBlue Live Cell Stain to mark nuclei. Microgrooved SMPs exhibited greater cell attachment compared to unpatterned surfaces. There is also some evidence of cell organization in the direction of the grooves. Scale Bar: 200 m. Note: some of these SMPs, specifically the formulations containing PEGDMA 750 & 1000, occasionally absorb the NucBlue stain; this is particularly evident on microgro oved surfaces, near the edges of the grooves. Although all SMPs and topographies demonstrate endothelial cell adhesion and increases in living cells during the study, endothelial cell attachment is greater on micropatterned surfaces compared to their unp atterned analogues, for all SMP formulations, at all timepoints, after endothelial cell seeding, as shown in Figure 6A. The 80:20 wt% tBA:PEGDMA550 microgrooved surfaces displayed the highest endothelial cell adhesion, which was measured by counting live, surface adherent cells, as displayed in Figure 6. It should be noted, however, that the 80:20 wt% tBA: P EGDMA 1000 microgrooved surface consistently experienced the greatest increases in cell presence compared to its unpatterned analogue. Occasionally, incre ases in dead cells on micropatterned vs. unpatterned surfaces are present, but the percentage of dead cells on microgrooved surfaces remains equal to or lower than their unpatterned counterparts. Long term cell cultures, up to seven days, depict similar tr ends as microgrooved surfaces and continue to show higher endothelial cell presence compared to unpatterned surfaces In addition to cell adhesion, live/dead data was also used to assess cell proliferation. Cell proliferation was estimated as the increase in cell presence on each surface relative to the initial day 1 cell counts and displayed as percentage increase in adherent cells (Figure 4. 6 B ). Although the unpatterned surfaces initially demonstrated less cell adhesion, these same unpatterned SMP surfaces typically demonstrated greater increases in cell presence on day 3 and day 7 compared

PAGE 97

84 to their microgrooved analogues. However, microgrooved surfaces ultimately demonstrated the greatest overall cell adhesion compared to their respective unpatterned analogues.

PAGE 98

85 Figure 4. 6 . A) Approximate cell presence , both live and dead, on all unpatterned vs. microgrooved SMPs. Microgrooved surfaces often demonstrate statistically significant increases in cell presence compared to unpatterned SMP surfaces. Dead endothelial cell presence is typically comparable between unpatterned and microgrooved surfaces. B) Percentage increase in EC presence on unpatterne d and microgrooved SMPs. Unpatterned surfaces demonstrate a higher percentage increase in cells, but cell presence remains highest on microgrooved surfaces compared to unpatterned ones. Significance between unpatterned and microgrooved surfaces was determi ned using a 2 tailed, unpaired t test, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. Significance between SMPs of varying crosslinker lengths and unpatterned and microgrooved surfaces was determined using a 2 way ANOV A, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. To simpl i fy compar ison of the means of the unpatterned vs. microgrooved surfaces, specifically the groups that appear to be statistically similar, but demonstrate statistically significant differences, cell adhesion and viability was also presented as data ± standard error (µ ± SEM), as shown in Figure 4.7.

PAGE 99

86 Figure 4. 7 . Approximate adherent ECs/SMP , live and dead, on all unpatterned vs. microgrooved SMPs. Statistical significances are visual ly clearer with data presented as µ ± SEM, since the focus is on the precision of the mean rather than the spread of the data. Significance between unpatterned an d microgrooved surfaces was determined using a 2 tailed, unpaired t test, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. Significance between SMPs of varying crosslinker lengths and unpatterned and microgrooved surface s was determined using a 2 way ANOVA, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001.

PAGE 100

87 Figure 4. 8 . Approximate adherent ECs/SMP, live only , on all unpatterned vs. microgrooved SMPs. Significance between unpatterned and microgrooved surfaces was determined using a 2 tailed, unpaired t test, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001. Significance between SMPs of varying crosslinker lengths and unpatterned and microgrooved surfaces was determined using a 2 way ANOVA, * corresponds to p < 0.05, ** corresponds to p < 0.01, *** corresponds to p < 0.001 Endothelial cell alignment was assessed by measuring nuclear orientation and alignment angle relative to the groove, as displayed in Fi gure 4. 9 . For cells adhering to microgrooved surfaces, nucleus orientation angles of 90° indicate cellular position parallel to the direction of the groove. Endothelial cells are more randomly oriented on unpatterned surfaces, whereas on micropatterned surfaces, E Cs orient themselves in the direction of the groove, depicted by increases in the percentage of cells for orientation angles 80° 90°. Cells on unpatterned surfaces

PAGE 101

88 are more isotropically distributed, demonstrating a lack of organization. Nucleus orientatio n, or lack thereof, was further verified by normalizing to a gaussian fit function.

PAGE 102

89

PAGE 103

90

PAGE 104

91 Figure 4. 9 . Endothelial cell alignment as measured by nuclei and actin fiber organization on 80:20 wt% tBA:PEGDMA550, 80:20 wt % tBA:PEGDMA750 and 80:20 wt% tBA:PEGDMA1000 , 1 day and 7 days after cell introduction . Increases in nuclei and actin orientation at 80° 90° on microgrooved surfaces indicate that cells display higher organization on microgrooved surfaces compared to their unpatterned counterparts at all time points . Nucleus orientation was confirmed by normalizing data to a gaussian fit function. Scale bar: 100µm. Discussion Despite advances in cardiovascular stent technologies, issues such as restenosis and thrombosis continue to occur, requiring reintervention to prevent further complications. Rapid re endothelialization may eliminat e , or at least reduce , the need for antiproliferative drugs by promoting a local anti inflammatory setting; several studies have shown that the presence of a well functioning endothelium promotes an anti thrombotic and anti stenotic local environment . 212, 221 Thus, rapidly restoring the endothelial lining could greatly improve device success . 222 Numerous methods have been implemented to accomplish rapid endothelial restoration, but the time cost and complexity of some methods limit their utility. We present a straightforward and customizable strate gy for microgroove implementation on shape memory polymer surfaces to optimize endothelial cell adhesion by encouraging organized attachment. This exploratory study examined the effect of introducing microgrooves to the surface of acrylate based shape memory polymers to increase endothelial cell adhesion and organization. Previous studies from our group have examined thermomechanical behavior, the shape memory effect and recovery, biocompatibility and cytotoxicity of these tBA:PEGDMA SMPs . 1 8, 165, 178, 179, 184 Recently, our group demonstrated endothelial growth onto a family of unmodified acrylate based SMP surfaces, but adherent cells lacked spatial organization . 195 In this previous study, the SMP formulations with weight percent ratios of 80:20 tBA:PEGDMA demonstrated improved cell adhesion and survival compared to the formulations containing greater amounts o f PEGDMA; as such, these formulations were selected for surface optimization through

PAGE 105

92 microgroove introduction. Studies have shown that endothelial cells adhering to microgrooved surfaces demonstrate increased viability, migration and organization relative to unpatterned surfaces . 160, 223 Dynamic mechanical analysis (DMA) data confirmed that pattern introduction had little effect on the bulk properties of the SMP. SMPs are attractive materials for s tent fabrication due to their low cost, ease of fabrication, customizability, drug loading potential, and compliance matching ability, among others . 18, 224 Previous studies by our group have demonstrated the ther momechanical utility of these SMPs for use in cardiovascular stents . 18 Here, we aimed to limit modification of the material to the surface, to ensure that th e bulk properties, such as modulus and activation temperature, were maintained and could potentially be used for safe and minimally invasive stent delivery . 18, 47, 179 While the shape memory effect was not extensively investigated here, these formulations were selected, in part, for their previously validated shape memory capabilities. Ensuring that the changes introduced here did not change thermomechanical properties of the bulk material was therefore important 18, 178, 179, 195 Although the fabrication method was slightly modified compared to previous work by our group, there were no significant differences in T g , T on set and T g range between unpatterned and microgrooved SMPs, confirming minimal effect on thermomechanical properties. Successful pattern transfer from a metal printed mold to the shape memory polymer surface via cast molding was confirmed using scanning e lectron micrographs of the polymer surfaces. Molding methods allow for pattern introduction while limiting chemical variation between the groove, ridge and sides compared to etching or embossing 73, 76 The primary advantage of using the 3D metal printing approach is that it offers a rapid, reproducible, and customizable method for mold fabrication, joining the growing list of low cost/high volume

PAGE 106

93 manufacturing techniques . 94, 225 Current topographical patterning methods, many of which originated in the semiconductor industry, produce highly precise and complex surface features, which have utility in the study of cell fun ction, but the high cost and specialization associated with these methods often reduces translational and scaling capability . 85, 105, 226 Surfaces were also quantita tively measured by assessing roughness and wettability. Surface roughness was obtained from AFM data and showed that roughness was lower on unpatterned surfaces compared to microgrooved ones, consistent with previous studies . 73, 227 The unpatterned SMPs contained random surface features, which increased roughness compared to SMPs fabricated in smooth glass mo lds, but the addition of microgrooves resulted in the highest observed surface roughness . 195 Pattern in troduction has been shown to increase the contact points on a surface, which would contribute to increased surface roughness; further, rougher surfaces have been shown to be more favorable to cell adhesion . 75, 83 Micropatterned surfaces also displayed larger water contact angles, indicating an increase in hydrophobicity, consistent with related studies . 227, 228 Higher hydrophobicity has been shown to increase cell attachment for certain cell types , including endothelial cells . 69, 229 Pareta et. al, found that endothelial cells preferentially adhere to a surface at a specific surface energy, which is influenced by wettability . 121 Increases in both roughness and hydrophobicity contribute to increased pres ence in cell adhesion to the microgrooved surfaces, as confirmed by the data presented here, but individual contributions are difficult to discern . 75, 230 Several groups have investigated endothelial cell adhesion on SMPs; for example, the EC adhesion capabilities of thermoplastic, degradable block copolymers, electrospun SMP scaffolds, etc., have been investigated and it has been shown th at these materials typically promote angiogenesis . 60, 231, 232 Our group previously demonstrated endothelial cell adhe sion to

PAGE 107

94 unpatterned , acrylate based SMPs with varying compositions . 195 Here, we selected the most promi sing of previously evaluated materials in our lab and further optimized the surface topographically through the introduction of microgrooves using a molding process. Microgroove introduction to the surface allows for the topographical modification while st ill maintaining the cell adhesion capabilities of the base material . The data confirmed that microgrooved surfaces increase cell adhesion , similar to results seen on other materials, including metals as well as polymers . 121, 160, 222, 233 Microgrooved SMPs initially displayed 20 30% higher cell presence o ne day after cell seeding compared to their unpatterned analogues. The continued increase in cell presence up to seven days after initial cell introduction, which was notably greater on unpatterned surfaces compared to their microgrooved counterparts, furt her confirmed cytocompatibility of both unpatterned and microgrooved surfaces . The improved functionality of endothelial cells on microgrooved surfaces most likely contributed to the greater cell adhesion initially, and subsequent sustained survival. Lu et . al., demonstrated similar results on titanium surfaces, in which microgrooved titanium displayed higher initial cell presence . 223 At the end of the study time period, the 80:20 wt% tBA:PEGDMA550 microgrooved SMP displayed the highest cell adhesion of the formulations and surfaces examined. The higher cell adhesion to this specific s urface may be a result of the combined benefits of material stiffness and microgroove addition. Our group previously found that endothelial cells prefer stiffer SMP surfaces, such as the one found on 80:20 tBA:PEGDMA550, and the addition of microgrooves ma y have further optimized the surface for cell adhesion and viability . 195 Notably, the 80:20 wt% tBA:PEG DMA1000 SMPs experienced the greatest increases in cell adhesion between unpatterned and microgrooved surfaces, which further supports the motivation for microgroove addition to the SMP surface. Additionally,

PAGE 108

95 Huang et. al., determined that endothelial cell s adhering to microgrooved titanium oxide surfaces displayed reduced adhesiveness to monocytes and platelets, promoting an anti inflammatory and anti thrombotic phenotype, an added benefit to microgroove introduction . 213 In addition to increased cell adhesion and viability, microgrooved surfaces also encouraged cell organization. Alignment was assessed by measuring the orientation angle of the nuclei . 211 Microgrooved surfaces appear to encourage endothelial cell alignment through contact guidance, complementing results from other work . 93, 234, 235 previously implemented on a wide range of materials and demonstrated successful endothelial cell organization compared to unpatterned surfaces . 211, 222, 236 Prior studies on other SMPs, specifically those with microwrinkled surfaces, have indicated that the addition of groove like nano and micro topographies also encourages cell alignment . 237 239 Other s tudies have shown that microtextured surfaces may contain local differences in surface energy, directing the spatial arrangement and conformation of adsorbed cell adhesion proteins, which dictates cell attachment to a surface . 230 To further improve cell alignment, it may be valuable to reduce groove width to more accurately match elongated cell size or increase groove depth ; these and other changes will be in vestigated in future work . 236, 240 In addition to improving cell alignment, further decreasing groove width may also generate more surface contact points by increasing material surface area, improving cell adhesion as well as proliferation . 39 Further, implementing a combination of micro and nano grooves, or a combination of other topographies such as wrinkles, waves, pits, etc., may provide added benefits compared to using a single surface feature dimension alone . 241 Recent studies by Brasch et. al., have found that the orientation of cellular Golgi bodies may be a better indicator of cell adhesion, which may also provide insight into cell orientation

PAGE 109

96 preferences . 237 It should be noted that the d imensions of the artifact micropattern were dictated by the specific model of the printer material deposition head and thus may be unique to the specific metal printer used; use of a different printer mat erial deposition head may generate microgrooves of varying widths and/or depths or other topographies. This study was terminated 7 days after EC introduction, but cell survival beyond this time point, up to 14 or 21 days will be of additional interest. Ass essment of EC adhesion on SMPs of varied composition, but similar activation temperatures, may provide additional insight into the nuances of cell adhesion on these materials and will be investigated in future work. In depth studies regarding the effect of programming these materials on subsequent EC adhesion would also increase the clinical relevance of this work. Finally, in vivo studies , as well as verifying the effect of steam sterilization on SMP thermomechanical properties, are required to validate th e results from these in vitro studies for implementation in implanted, blood contacting devices. The benefits of cell organization are not limited to endothelial cells in the vascular environment. Muscle tissue, nerve tissue, smooth muscle cells in vasc ular tissue and periodontal tissue all require cell alignment to maintain proper cellular function . 205, 234, 240 The generic, yet customizable, nature of this approach may allow for the extension of this technique to a variety of materials and cell types for tissue engineerin g.

PAGE 110

97 CHAPTER V THERMOELASTIC RECOVERY OF MACROSCALE SHAPE MEMORY POLYMER SURFACE FEATURES Introduction Shape memory polymers (SMPs) recover their original shape from a programmed shape upon application of a stimulus such as heat, light, electric fields, alternating magnetic fields, immersion in water, voltage, etc. 48, 108, 164, 242, 243 This feature makes SMPs promising materials for a range of applications, biomedical and otherwise, due to thei r ease of fabrication, highly tunable material properties and lower fabrication cost . 47, 244, 245 SMPs are highly versatile and can be made into a range of shapes using a variety of methods, which allows for specific tailoring of these materials to a desired application . 203 Mention of shape memory in SMPs first occurred in the 1940s, where it was first called the form of heat shrinkage tu bing, while the term SMP was not formally coined until the mid 1980s . 169, 177 SMP research gain ed prevalence in the 1990s and continues to be of interest due to SMP utility in a broad range of applications. Shape memory polymers have been developed for numerous applications including biomedical materials and devices, information carriers, aerospace applications, smart textiles, polymer actuators, sensors, active assembly, among others . 18, 46, 166, 167, 169, 171, 172 The glass transition temperature (T g ) is the temperature at which the SMP transitions from its hard, glassy state to its soft, rubbery state. T g can be manipulated by changes in polymer chemistry or structure and can vary over a wide range of temperatures . 163, 177 The ability of SMPs recover their original shape is referred to as the sha pe memory effect (SME). The SME for

PAGE 111

98 thermoresponsive SMPs involves fabricating the polymer into its original desired shape, deforming the SMP at a temperature above the glass transition temperature (T g ), T=T high , lowering the temperature below T g , to T=T lo w , while deformed, and finally, recovering the original shape at a temperature above T g, T=T high , while unconstrained . 108 The SMP can undergo multiple cycles of programming and recovery and may be programmed into a different shape in each cycle . 164 The shape memory effect is not an intrinsic property of the material, but rathe r a result of polymer fabrication and processing. A wide range of polymers, from thermosets to thermoplastics, may exhibit shape memory, as long as these polymers contain sufficient crosslinks for shape memory or physical crosslinks in the form of ordered crystalline regions . 164, 177, 246 It should be noted, however, that most polymers exhibit some degree of shape memory; however, the efficacy and ability to control that shape memory varies . 215 SMPs are not the only materials that exhibit the SME; select metals and ceramics are also capable of shape memory, but SMPs have lower manufacturing costs, lower stresses and higher recoverable strains close to 100% under proper temperature and load conditions . 1 85, 247 Thermal recovery of SMPs also relies on the programming conditions of the SMP. Specifically, it has been found that thermoplastic polymers are more sensitive to temperatu re and deformation programming conditions than thermosets . 215 Collins et. al., found that thermoplastic SMPs exhibit lower shape recovery with increased deformation and load hold time . 245 Thermal recovery of SMPs also relies on the programming conditions of the SMP. Specifically, it has been found that thermoplastic polymers are more sensitive to temperature and deformation programming conditions than thermosets . 215 Collins et. al., found that thermoplastic SMPs exhibit lower shape recovery with increased deformation and load hold time . 245

PAGE 112

99 Much of the research related to SMPs has widely focused on bulk properties and the SME for delivery and de ployment of minimally invasive biomedical devices. For example, our group and related groups ha ve investigated the effects of sterilization on SMP thermomechanical properties, the use of SMPs to improve hernia mesh delivery, and cardiovascular stent deform ation and deployment, among others . 18, 165, 177 179, 248 Yakacki et. al studied t he relationship between network structure, rubbery modulus, glass transition, free strain and constrained strain recovery . 184 Nguyen et al., developed a model to predict shape memory performance from uniaxial tension results . 185 Other groups h ave investigated free and constrained shape recovery, changes in color, transparency, permeability, elasticity, and enthalpy due to the shape memory effect, SMPs for use in self tightening sutures and cell compatibility with SMP surfaces for intended use i n implanted biomedical devices . Yakacki et. al studied the relationship between network structure, rubbery modulus, glass transition, free strain and constrained strain recovery . 184 Nguyen et al., developed a model to predict shape memory performance from uniaxial tension results . 185 Other groups have investigated free and constrained shape recovery, changes in color, transparency, permeability, elasticity, and enthalpy due to the shape memory effect, SMPs for use in self tightening sutures and cell compatibility with SMP surfaces for intended use in implanted biomedical de vices . 55, 57, 162 Some work has also been conducted on analyzing the surface properties of acrylate based SMPs. Wornyo et. al used nanoindentation to investigate shape memory properties at the submicron level . 1 08 Additional work with respect to nanoindentation and its effects on indentation strain rate, backbone chemistry, and processing techniques has also been conducted Some work has also been conducted on analyzing the surface properties of acrylate based SMPs. Wornyo et. al used nanoindentation to investigate shape memory properties at the submicron

PAGE 113

100 level . 108 Additional work with respect to nanoindentation an d its effects on indentation strain rate, backbone chemistry, and processing techniques has also been conducted . 249 251 Yang et. al., investigated the nanoscale strain storage and nanoindentation recovery using heated AFM probe tips . 107, 252 Indentation and recovery using a spherical indenter on the surface of nitinol surfaces has also been investigated . 253 Groups have also used SMP surface topography to direct cell alignment and encourage cell sheet formation . Indentation and recovery using a spherical indenter on the surface of nitinol surfaces has also been investigated . 253 Groups have also used SMP surface topography to direct cell alignment and enco urage cell sheet formation . 203, 210, 254 In this work, we investigate the compression and recovery of a macro scale protrusion on the surface of the acrylate based SMP. Much o f the previous work has focused on the mechanics of the bulk material, while few studies have considered surface deformation and recovery of acrylate based SMP surface features, particularly for use in biomedical devices that may benefit from recoverable s deformation and recovery of surface features on acrylate based SMPs. By understanding how surface features present on this acrylate based SMP can deform and recover, we hope to e xtend this knowledge to smaller scale features that will be consequential on the cellular level. These features can then be introduced to minimally invasive, implantable biomedical devices, with the guarantee that the features would recover and perform aft er device deployment. Materials and Methods Shape memory polymers were fabricated using tert butyl acrylate (tBA) and poly(ethylene glycol) dimethacrylate (PEGDMA) with average molecular weights of M n =550, 750, 1000 and photoinitiator 2,2 dimethoxy 2 phe nylacetophenone (DMPA). All products were obtained from Sigma Aldrich (St. Louis, MO, USA) except PEGDMA1000, which was obtained

PAGE 114

101 from Polysciences (Warrington, PA, USA) and were used as received. Pre polymer mixtures are presented in Table 5. 1. Table 5. 1 : SMP Formulations used for this study. Three different weight percent ratios were tested, with 3 variations of PEGDMA MW, for 9 total formulations. Pre polymer mixtures were injected into molds composed of standard microscope slides (Thermo Fisher Scienti fic, Waltham, MA, USA) on one side and concavity well microscope slides (Electron Microscopy Sciences, Hatfield, PA, USA) on the other side. The cavity microscope slide contained 1 cavity measuring 15mm in diameter and either 1mm or 0.5mm in depth. The cav ity containing face of the slide was placed towards the inside of the mold. Pre polymer solution was then injected into the mold and polymerized under an ultraviolet (UV) lamp at radiation of 365 nm wavelength. The samples were irradiated for 10 minutes un der pulsed conditions (30 seconds on, 30 seconds off) followed by uninterrupted irradiation for 10 minutes, for a total of 20 minutes.

PAGE 115

102 Solid SMPs, which contained a single semi spherical protrusion, were then placed in a dry oven at 75°C for two to three d ays prior to compression experiments. Post processing the SMP samples in a dry oven anneals the SMP and reduces defects. The glass transition temperature (T g ) for the SMPs was determined using a TA Q800 DMA, as has been presented previously by our group. B riefly, the samples were cut into specimens with dimensions of 20mm×5mm×1mm for testing. It should be noted that these samples were flat and did not contain any surface features. Each sample was equilibrated to 0°C for 1 minute and heated to 100°C at a rat e of 3°C/minute. Testing was conducted at a frequency of 1.0 Hz and cyclic strain control at 0.1% strain. Once the glass transition temperature of each sample was obtained, the temperature range for the surface recovery experiments was determined. SMP samples were cut using a razor blade into approximately 20mm x 20mm square specimens with the protrusion at the center of the sample, for compression testing. All edges were then sanded using 600 grit sandpaper to reduce imperfections from cutting. Pr ior to compression testing, the height, diameter(s) of the protrusion, and length and width of the sample were measured using digital calipers. Samples were placed in a temperature controlled environment and heated or cooled to either 25% or 50% above T g , placing the SMP in the rubbery region. Samples were then moved to the Mark10 ESM1500 tensile tester and compressed at a rate of 20mm/min up to a load of approximately 3.5 kN using a 4.448 kN load cell at a temperature either 25% or 50% below T g . Sample hea ting was accomplished using Kapton® insulate flexible heaters (Omega Engineering Inc, CT, USA) attached to the compression grips.

PAGE 116

103 The samples were compressed at 3.5 kN for 60 seconds (25% above T g or 50% above T g ) or 60 minutes (25% above T g ). The samples were then removed and the same measurements that were taken before compression were taken immediately after load removal. The compressed SMP sample was then placed in the temperature controlled chamber again and heated to either 25% or 50% above T g or 37° C for surface recovery. The height of the recovered surface feature was measured after recovery of the surface feature had taken place. Four samples of each formulation were tested, and each experiment was repeated five times. Table 5.2 : Methods matrix f or compression & recovery of SMP surface protrusion Re s ults The glass transition temperature (T g ) for thermoplastic SMPs is the temperature at which the shape memory effect (SME), is instigated, when the SMP transitions from its hard, glassy state to its soft, rubbery state. The glass transition temperatures are presented in Table 5. 3. Our group has previously studied glass transition temperature of these materials and our data agrees with previous results . 18, 177 Briefly, glass transition temperature increases with increasing tBA content and decreasing crosslinker MW. Additionally, after verifying the glass transition

PAGE 117

104 temperature, the temperature recovery was also determined, as display ed in Table 5. 3. Compression temperature, T low , was set at room temperature, or 25°C. Table 5.3 : Glass transition temperatures and compression and recovery ranges of select SMPs investigated. The 80:20 wt% tBA:PEGDMA formulations were the only formulati ons of the ones tested that exhibited successful compression and recovery. Specifically, these samples deformed when the load was applied, remained in their temporary shape after the load was removed and did not fully recover their permanent shape until te mperatures were increased beyond T low . The compressed state of the 80:20 tBA:PEGDMA550 specimen is displayed in Figure 5. 1.

PAGE 118

105 The 50:50 tBA:PEGDMA550 formulation was compressible and deformed when the load was applied, but did not retain the compressed sha pe after the load was removed, as shown in Figure 5. 1. Shape recovery occurred almost immediately after the load was removed, even though the temperature during compression and immediately following compression was below T g . The sample also experienced cracking due to load application. The remainder of the formulations were not able to remain in their temporary shape at temperatures below T g and most of these samples cracked or broke when compressed. Figure 5.1 . Deformati on retention of 50:50 tBA:PEGDMA550 vs. 80:20 tBA:PEGDMA550. The 50:50 tBA:PEGDMA550 sample recovers its original shape almost immediately after the load is removed. However, the 80:20 tBA:PEGDMA550 sample retains its deformed state, unconstrained, long af ter the load is removed. Protrusion height was measured prior to compression and after recovery. These measurements were then used to calculate percent recovery, . At recovery temperatures 25% above T g , specimens from all the formulations exhibited close to 100% recovery, regardless of protrusion height or compression times, as shown in Figures 5. 2A

PAGE 119

106 and 5. 2B. For the 80:20 tBA:PEGDMA formulations, there is a small decrease in % recovery as the PEGDMA c rosslinker length increases. Specifically, Recovery is still, at minimum, 97%. Additionally, recovery appears to be independent of protrusion height for a given compression and recovery condition. A

PAGE 120

107 Figure 5.2 . Percent recovery of SMP protrusion at temperatures above T g for A) 0.5mm protrusion B) 1.0mm protrusion. Recovery is close to 100% for all the samples, regardless of formulation, protrusion heig ht or recovery temperature greater than T g . Percent recovery was also measured after the samples were recovered at 37°C, since the intent is to use these materials for implanted medical devices, which would need to exhibit shape recovery at physiological temperatures. Percent recovery was at least 94%, regardless of compression time or SMP formulation, indicating that all of the samples are able to recover from their deformed shape at 37°C, as shown in Figures 5. 3A and 5. 3B. Additionally, for the 80:20 tBA :PEGDMA formulations, percent recovery is typically higher when the samples are compressed for 60 minutes vs. 60 seconds. Again, recovery is not dependent on surface feature size when recovery occurs at physiological temperature. B

PAGE 121

108 Figure 5.3 . Percent recovery of SMP protrusion at physiological temperature for A) 0.5mm protrusion B) 1.0mm protrusion. Recovery is lower compared to recovery at temperatures above T g due to the lower recovery temperature, but recovery is still greater than 90%. A B

PAGE 122

109 Recovery time was defined as the time required for the SMP surface feature to recover to its original height following exposure to temperatures above T g or to physiological temperature. Surface feature size did not appear to have a significant effect on recovery time for surfaces recovered at temperatures 25% above T g , since recovery time did not differ greatly between the 0.5mm protrusion and the 1.0mm p rotrusion. Recovery time, however, did show small increases when compression time increased from 60 seconds to 60 minutes and these increases became more significant as the protrusion height was increased. Recovery time monotonically increased as PEG length decreased for samples recovered at physiological temperature, since the T g of those samples lay farther from 37°C. Additionally, as protrusion height doubled from 0.5mm to 1.0mm, recovery time also approx imately doubled, as shown in Figures 5. 4A and 5. 4B. When compression time was increased from 60 seconds to 60 minutes, recovery time also increased. A

PAGE 123

110 Figure 5. 4. Recovery time for A) 60 second compression B) 60 minute compression. Compression is generally comparable for 60 second vs. 60 minute compression. Recovery time approximately doubles for the 1.0mm protrusion vs. the 0.5mm protrusion. Recovery time was also measured for compressed samples recovered at temperatures 50% above T g . However, as displayed in Figure 5. 5, there are few differences between recovery times for samples recovered at 25% above T g vs. samples recovered at 50% above T g , indicating that the higher recovery temperature does not seem to have a significant effect on the recovery time. B

PAGE 124

111 Figure 5. 5 . Recovery time at temperatures 25% above T g and 50% above T g . Recovery times are mostly identical regardless of whether compression and recovery occur at temperatures 25% above T g or at temperatures 50% above T g . Percent compression was measured as the difference between the original height of the protrusion and the height of the protrusion after the load was applied. Samples containing longer PEGDMA chains were more likely to experience a greater change in height when load was applied, as seen in Figure 5. 6. The percent compression for the 80:20 tBA:PEGDMA formulations seems to follow the following trend:

PAGE 125

112 The maximum percent compression was approximately 20% for the 0.5mm protrusion and 30% for the 1.0mm protrusion, indicating that percent compression does not scale directly with protrusion height, but does still increase. A B

PAGE 126

113 Figure 5. 6. Percent height compression for A) 0.5mm protrusion and B) the 1.0mm protrusion. Percent compression is typically lowest for the 60s compression at 25% above T g . However, percent compression exhibits less of a difference between sample s that are compressed for 60min at 25% above T g and samples that are compressed for 60sec at 50% above T g . Discussion This study examined the compression and recovery of surface features on select acrylate based shape memory polymers (SMPs). While previous studies have primarily focused on bulk shape memory, few studies, if any, have investigated compression and recovery of surface features on an acrylate based SMP, specifically for use in biomedical devices. By investigating surface compression and recover y, we hope to ex pand our understanding of the surface thermomechanics, so that soon, it will be possible to fabricate SMPs with controlled surface features for use in implanted medical devices that may benefit from shape memory surface features. SMPs are promising materials for minimally invasive implanted biomedical devices because they exhibit high strain recovery. Employing SMPs for deformation and recovery of surface features on implanted biomedical devices may increase the probability of device succes s. Recent studies have shown that patterned surfaces on cardiovascular stents may encourage endothelial cell attachment and alignment on the stent surface, increasing the likelihood of stent success . 210, 254 Originally, most device surfaces contain ed s urface patterns that c ould not be recovered after deformation, and thus c ould not be controlled. SMPs, however, allow for controlled recovery of the surface features, in addition to controlling deployment of the device itself. This may reduce friction and sticking that would result from the patterned surface being in a deformed state f or extended periods of time and allow for full recovery of the device, down to the surface features.

PAGE 127

114 Of the formulations tested, only the 80:20 tBA:PEGDMA formulations successfully exhibited shape memory of the surface feature. Specifically, samples mad e from these formulations could be compressed and remained deformed until introduced to temperatures greater than T low . Most f ormulations containing greater than 20% PEGDMA did not exhibit shape memory of the surface feature. Prior studies have shown that when the crosslinker concentration exceeds 40%, the transition regime widens, blurring the transition from the glassy to rubbery region, which depletes the shape memory ability . 187 It then follows that formulations may not exhibit shape memory of the surface feature, which agrees with our data. It has been confirmed that increasing the crosslinker increase s the crosslinking density and as a result, increases the rubbery modulus. The relationship between crosslinker concentration and rubbery modulus has been described by , where is the rubbery modulus, is the gas constant, is the abso lute temperature at which is measured and is the crosslinking density . 187 The increased crosslinking density and rubbery modulus contribute to the softer nature of the SMP, as these SMPs do not contain the crystalline regions that are required for shape memory. The higher rubbery modulus also decreases the stiffness, which reduces the resistance to material breakage under load, making the materials too brittle for biomedical device use. Recovery of the SMP surface feature at room temperatures greater than T g , whether recovery temperature was 25% or 50% above T g , was near 100% for all formulations. At these elevated temperatures, mobility of the molecular chains drives the SMP to regain its original configuration. Thus, when the SMP is exposed temperatures at least 25% greater , near full recovery of the

PAGE 128

115 regime . 255 Additionally, at temperatures 50% above T g , sample fracture was not observed, indicating that, if desired, deformation and recovery at temperatures 50% above T g is possible as these temperatures are within the range where chain integrity is maintained . 184 This study confirmed that for tBA:PEGDMA SMPs , high amounts of tBA are required to successfully exhibit shape m emory , even for surface features on a smaller scale. Recovery time for the surface features appears to be unaffected by protrusion height or formulation when recovery occurs at temperatures 25% or 50% above T g . At these higher temperatures, the SMP has entered the rubbery regime and the molecular chains are able to quickly recover to their original configuration. It has been previously reported that samples deformed and programmed at lower temperatures begin sh ape recovery at lower temperatures, but since these SMPs are in the rubbery regime at both 25% and 50% above T g , exposing the SMP to temperatures higher than 25% above T g for deformation does not seem to display any additional benefit . 255 We have previously shown that increasing the monomer, tBA, increases T g to temperatures that are significantly greater than physiological temperature . 18, 256 Data from current and previous work indicates that if it is desirable to have 100% surface feature recovery in a short time frame, it may be n ecessary to formulate an SMP with a T g of around 30°C, placing the 25% above T g temperature point at physiological temperature. This would guarantee almost complete surface feature recovery within 10 15 seconds. However, an additional monomer or crosslinker would need to be added to the formulation to retain shape memory capabilities while lowering T g to around 30°C. In addition, to avoid accidental deployment, the device would need to be held in cold storage, which would pose a major disadvantage for this model . 177 Other methods to induce faster SMP recovery at temperatures below T g , including indire ct heating using irradiation or

PAGE 129

116 lowering T g by infusing the SMP with lower MW molecules into the SMP, have been presented elsewhere . 164 On the other hand, if a delay in complete surface feature recovery after device deployment is desirable, as it may be the case when shape memory governs the device deployment in addition to surface feature recovery , the high tBA formulations presented in this study are promising candidates. Recovery of SMP surface features to their original form at 37°C is possible, even though physiological temperature lies below T g for most of the shape memory formulations. However, the time required for these surface features to recover their original shape varies based on how far T g is from physiological temperature, since increasing T g has been shown to delay recovery during heat ing . 47 It was previously mentioned that all polymers exhibit some degree of shape memo ry, which may contribute to the surface feature recovery, even though T g s of the SMPs tested were all above physiological temperature. It should be noted that it is recommended that the temperature range for the transition from temporary to permanent shape be in a narrow window for efficient storage of programmed shape and rapid recovery of permanent shape . 245, 247 The decreased crosslinker length for the 80:20 tBA:PEGDMA formulations resulted in a stiffer material that is more resistant to compression, even when the material is in the rubbery regime at temperatures great er than T g . Formulations containing longer PEGDMA chains experienced greater compression, due to the rubbery nature of PEGDMA, making PEGDMA a desirable component in many hydrogels . 257 Also, for samples that were compressed for 60 regime . 245 When the SMP surface feature is compressed, the molecular chains become aligned and pressed into the base of the sample and these chains remain compressed while the temperature is below T g . When the compressed sample becomes exposed to temperatures greater

PAGE 130

117 than T g , the SMP returns to its rubbery regime, and the molecular chains regain their entropic form, similar to bulk material compression . 162, 164 The 50:50 tBA:PEGDMA550 formulation was not resistant to compression, as deformation when the load was applied was evident. However, the increased crosslinker content increased the elastic entropy, caus ing the SMP feature to recover immediately after the load was removed. The higher crosslinker content increased the amount of amorphous chain segments, which can be stretched when a load is applied, but as soon as this external force is removed, the SMP re verts to its original form. Since the compressed height was measured after the load was removed, it may appear as if the 50:50 tBA:PEGDMA550 samples were resistant to deformation, when instead, they exhibit a higher degree of elastic entropy due to the hig her number of amorphous soft segments . 48 While the compression and recovery experiments conducted for this study were completed on a larger scale to be consequential for most biological applications, on a macro scale, the results obtained from this study should be scalable to measurements in the tens to hundreds of microns range, which does have an impact on a cellular scale. We found that scaling the surface feature height down by 50% also cut recovery time approximately in half at physiological temperature. This relationship should continue to surface feature sizes in the tens of microns, where the material, and not molecular interactions, govern s thermomechanics. This study confirmed that deformation and recovery of a surface feature is possible, in addition to deformation of the bulk mater ial which was previously verified. The next step would involve scaling the surface features down to heights and widths that would be influential on a cellular scale. Additionally, fabrication of a dual deploying SMP, one that would exhibit bulk recovery al most immediately followed by delayed surface recovery, minimizing sticking and

PAGE 131

118 friction should also be conducted. If the mechanical robustness of metal is still required, i t may also be possible to coat a metal stent with the surface patterned SMP, which w ould include both the mechanical integrity of the metal backbone as well as the controlled surface properties of the shape memory polymer.

PAGE 132

119 CHAPTER VI TEMPERATURE ACTIVATED MICROGROOVES IMPROVE ENDOTHELIAL CELL ADHESION AND ALIGNMENT ON SHAPE MEMORY POLYMER SURFACE S Portions of this chapter may be submitted to Journal of Biomaterials Research Part B , 2019, and are included with the permission of the copyright holder Introduction Implanted cardiovascular stents are the primary treatment opti on for blood vessels weakened by cardiovascular disease (CVD) . 154 Stents provide mechanical support to the diseased vessel, restoring blood flow downstream. However, current stents , both bare metal stents (BMS) and drug eluting stents (DE S), often exhibit limitations due to reduced compatibility between the stent and its surrounding environment, leading to instances of restenosis or thrombosis that may require subsequent reintervention . 258 Endothelial cell recruitment, or endothelialization , is a promising approach to reduce stent rejection. Studies conducted as early as the 1970s have shown an increased success rate of implanted biomedical devices whose surface s can rapidly instigate endoth elialization . 77, 152, 159 Rapid re endothelialization is often described as major cell presence within 24 hours and full cell coverage within 3 7 days of device implantation . 153 A complete, healthy endothelium creates a non thrombogenic and non proliferative environment, which reduces the necessity for antiproliferative agents . 77, 157 In vitro endothelialization prior to stent implantation has been shown to increase patency of these grafts, but these procedures can be laborious and costly, a nd often limited to specialty centers . 32 Cell material surface interactions are crucial to the formation of multicellular organisms and disease progression, as well as important to the survival of implanted biomedical devices ,

PAGE 133

120 particularly in cases that necessitate endothelializat ion . 115, 259 Mechanical, chemical, and/or topographical environmental cues can have profound effects on the fate of a n implanted device . 202 204 Oftentimes, the unadulterated surface of a biomaterial is noncompliant with the surrounding physiological environment, prompting the requirement for a new biomaterial or modification of the current biomaterial surface. Numerous su rface modification methods including chemical, physical and biofunctional, have been used to encourage in situ endothelialization . 65, 71, 94 Surface patterning specifically has shown promise in encouraging cell alignment for a variety of cell types, including endothelial cells . 62, 205 Structural organization of cells for tissue formation is one of the main challenges for current in vitro models . 205 Cellular and extracellular matrix (ECM) organization is essential for biological and mechanical function in most native tissues, but many artificial scaffolds do not contain the structural cues required to encourage such organization. As a result, numerou s patterns and patterning methods, both on the micro and nano scale, have been implemented to encourage endothelial cells to exhibit the elongated, athero resistant form on metal and polymer surfaces alike , a s these topographical features provide the cont act guidance necessary for cell elongation and organization . 212, 213, 223 Studies of various metal stent surfaces with microgrooves have demonstrated increases in aortic endothelial cell migration at least two fold compared to their smooth analogues . 160 E ndothelial cells in vivo align in the direction of blood flow, and as such, endothelial cells exhibiting alignment in vitro demonstrate improved function compared to their randomly oriented counterparts . 210 Increasing interest in using patterned surfaces to encourage cell alignment has been driven by improved function of elongated and organized endothelial cells . 212, 213 Previous studies have shown that if the depth of the surface features is too great, cel ls

PAGE 134

121 conform to the plateaus of the groove and may not populate the bottom of the surface feature, leading to incomplete cell coverage . 87 Most in vitro endothelialization studies are limited by physically static substrates that were unable to change or adapt during cell culture . 203 Stimuli responsive substrates have become a more attractive option to overcome some of the limitations set by passive surfaces. Temperature responsive surfaces have been investigated for controlled cell alignme nt, cell seeding, cell sheet formation , as well as to gain a broader understanding of cell material interactions, among other applications . 203, 210, 254, 260 263 Shape memory polymers (SMPs) have recently been investigated for use in response to the limitations of static substrates are active materials that recover their original shape when exposed to a stimulus . 55, 133, 213, 246 Thermo responsive SMPs are promising materials for implanted biomedical devices because shape memory can be instigated by bo dy heat, eliminating the need for additional equipment to instigate shape memory. Our group has investigated the thermomechanical properties and some potential applications of acrylate based SMPs for use in hernia meshes, cardiovascular stents as well as o ther applications . 18, 47, 165 Using SMPs for fabrication of the complete stent would allow for improved matching of mechanical properties to surrounding tissue and the option to customize the surface for improved biocompatibility . 126, 264 Three SMP formulations containing monomer tert Butyl acrylate (tBA ) and varying molecular weights of crosslinker poly(ethylene glycol) dimethacrylate (PEGDMA) were mold casted using a metal printed mold to generate grooved surfaces as well as temperature responsive grooved surfaces were analyzed using transmission microscopy, scanning electron microscopy and contact angle methods. Dynamic mechanical analysis (DMA) provided the glass transition temperature (T g ), which dictated the

PAGE 135

122 compression temperature as well as surface r ecovery time. Endothelial cell attachment and alignment w ere measured using fluorescence m ethods . Temperature responsive s hape memor y microgrooved surfaces often displayed higher endothelial cell presence compared to passive microgrooved surfaces. Cell ali gnment was confirmed on both thermo responsive and passive microgrooved surfaces, but significant differences in alignment between passive and temperature responsive surfaces were not observed. Materials and Methods The metal face of the fabrication mold was designed using SolidWorks (Waltham, MA, USA) and fabricated using an EOSINT M270 metal printing system (EOS, Munich, Germany) . The surface of the metal print was customized to contain a trapezoidal repeating patte rn, generating plateaus and grooves of equal width on the mold surface. Each face of the trapezoid measured 50 µ m width and 25mm length to match the length of a standard microscope slide, as shown in Figure 6. 1. The ends of the metal print were intentional ly unpatterned for ease of handling and SMP removal. The metal print, which was composed of maraging steel, was polished using steel peening and edges of the part were sanded to ensure uniform polymer synthesis.

PAGE 136

123 Pre polymer formulations contained monomer tert Butyl acrylate (tBA), crosslinkers poly (ethylene glycol) dimethacrylate (PEGDMA) (Mn = 550, 750 (Sigma Aldrich) , and 1000 (Polysciences)) and 2,2 Dimethoxy 2 phenylacetonphenone (DMPA). All items were received from Sigma Aldrich unless otherwise indi cated and used as received. Pre polymer solutions were injected into molds consisting of a clean standard glass microscope slide, 1.33mm silicone spacer, and the custom metal printed part. The mold was then placed in a UV light chamber, with the glass end facing towards the UV lamp to allow for UV curing, under wavelength = 365nm for 20 minutes, similar to previous methods . 265 The samples were rapidly and carefully removed from the molds at approximately 90°C, to ensure that the SMP could be removed in the absence of anti adhesives and with minimal damage to the surface of the SMP. Once the SMPs were removed from t he molds, they were post processed at 75°C for

PAGE 137

124 2 4 48 hours to anneal the material to ensure consistent mechanical properties and reduced material defects. A TA Q800 DMA (TA Instruments, New Castle, DE, USA) was used to confirm the glass transition tempera ture (T g ) of the micropatterned SMPs. Our group has previously reported the glass transition temperature values for these materials and data obtained was in agreement with previous values . 178, 195 The glass transition temperatures were then used to determine the compression temperature, which was set at 25% above T g , for cell attachment experiments. Glass transition temperatures and temperatures for compression and recovery are reported in Table 6. 1 . Scanning electron microscopy (SEM) using a JEOL ASM 6010LV (JEOL USA, Peabody, MA, USA ; ) was used to verify qualitative and quantitative pattern transference to the surface of the SMP. SMP s were sputter coate d using a Hospital Colorado) for 30 seconds prior to imaging. M icrographs were then used to confirm patterning on the surface of the SMPs. Grooved surfaces were then measured at random locations using ImageJ (NIH, Bethesda, MD, USA) . Fifty measurements were taken per SMP formulation to assess quality of pa t tern transference. Wettability of unpatterned and micropatterned SMP surfaces was obtained via contact angle using a Kudos Precisio n Instruments DropMeter A60 (Manhattan, NY, USA). A 10µL water droplet was applied to each surface and the angle formed between the water droplet and the surface of the sample was measured. Measurements were taken 10 seconds after the water droplet was int roduced to the surface of the SMP to maintain consistency. Contact angles were measured using SurfaceMeter Elements computer software (NBSI, Ningbo City, China). Contact angles less than 90° indicate a hydrophilic surface whereas contact angle values great er than 90°

PAGE 138

125 are present on hydrophobic surfaces. Five different samples were analyzed per surface, per SMP formulation. Five drops were applied to each SMP sample surface and five measurements were taken per drop. Recovery times for the SMPs were measured using a Point Gray Research Digital Camera (FLIR Integrated Imaging Solutions, Inc. Richmond, BC, Canada) and stopwatch. Micropatterned SMPs were heated to 25% above T g and subsequently compressed on a Mark 10 Mechanical testing device (Mark 10 Corporation, Copaigue, NY, USA) for 5 mins. The SMPs were then placed on a metal platform heated by Kapton ® Heaters (Omega Engineering, INC., Norwalk, CT, USA) at 37°C and time re quired for the surface pattern to recover was measured. Ten different samples per formulation were each measured three times to minimize error. Atomic Force Microscopy (AFM) was used to obtain information regarding the topography and roughness of the SMP surfaces . Samples were rinsed with ethanol and air dried to remove any debris prior to imaging. A JPK AFM system (JPK, Berlin, Germany) was used to obtain t opographical data and image s . Image post processing was completed using Gwyddion open source softwa re (Gwyddion, Brno, Czech Republic). As part of the post processing, t he root mean square roughness coefficient, R ms , provided quantitative information of the sample surface. R ms , measured by the standard deviation of the distribution of surface heights of the sample, was also obtained from Gwyddion . 219 Prior to cell seeding, all samples were sterilized using ultraviolet (UV) light for 1 hour. After sterilization, o ne set of micropatterned SMPs, which contained three samples of the same SMP formulation, were heated and surface compressed while another identical set of micropatterned samples were not surface manipulated in any way , as shown in Figure 6 . 2 . To ensure consistency, both sets of samples were heated and cooled prior to cell introduction.

PAGE 139

126 Figure 6.2. Surface manipulation of SMPs. The surface of one set of SMPs is unmanipulated, while the surface of another set of SMPs is compressed prior to cell introduction and allowed to recover at physiological temperature after cell introduction. HUVECs were then plated on 1cm diameter SMP substrates in 24 well plates and allowed to attach. Cells were seeded at a seeding density of 1× 10 5 cells/mL pe r well. Cell adherent SMPs were checked using transmission microscopy to ensure an absence of contamination, prior to formal assessment of cell viability, approximately 24 hours after cell seeding . The Live/Dead Cell Imaging Kit (488/570) ( Life Technologi es, Carlsbad, CA, USA ) was used to assess endothelial cell attachment and viability. Live cells, which are actively attached to the substrate and emit green fluorescence, while dead cells fluoresce red. Images were obtained using an AxioVert A1 (Zeiss , Tho rnwood, NY, USA ). Each experiment was conducted in triplicate. At least ten images from five replicate experiments were used for cell counting using ImageJ software (NIH , Bethesda, MD, USA ). After live cell imaging, cell adherent SMPs were washed with PBS and fixed in 4% paraformaldehyde for 10 minutes. Fixed cell adherent samples were then permeabilized using 0.25% Triton X in PBS prior to stain application. Samples were then stained with Ale xa Fluor 568 Phalloidin ( Thermo Fisher Scientific, Waltham MA, USA ) for marking actin exoskeleton

PAGE 140

127 fibers as well as DAPI (4',6 Diamidino 2 Phenylindole, Dihydrochloride) to mark nuclei for orientation measurements. The data were expressed as mean ± standard deviation ( µ ± SD), unless otherwise noted . Statistical analysis was performed using MATLAB (MathWorks, Natick, MA, USA) and significance was determined using a two tailed t level of significance of 0. 05 when comparing passive vs. temperature responsive groups. A two way ANOVA was used when comparing greater than 2 groups, the significance between individual samples if ANOVA determined signific ance of the sample set. Results Dynamic mechanical analysis (DMA) was used to confirm the glass transition temperature as measured by the peak of the tan delta , similar to previous methods 195 . Glass transition temperature (T g ) increases with decreasing crosslinker molecular weight, as shown in Table 6. 1 and agree s with previous results . 18, 195 The glass transition temperature was also used to determine the compression tempe rature, T c , which was set approximately 25% above T g and as a result also increased with increasing PEGDMA MW .

PAGE 141

128 Table 6. 1 . Glass Transition Temperature, Compression Temperature and Microgroove Recovery Rate S canning electron micrographs were used to qualitatively assess pattern transfer to the surface of the SMP from the mold. SEM confirmed microgroove pattern transfer to SMP surfaces , as shown in Figure 6. 3 . Furthermore, micrographs were also used to qualitatively measure groove width a nd depth to ensure that the groove measurements were maintained from previously specified values. Groove widths range from 55 60 µm, similar to the shallow groove width that was previously used, while groove depths measured slightly greater, ranging from 5 5 65 µm. So, w hile the groove widths are comparable to the specified dimensions, groove depths are larger. SMP Formulation, 50 um micropattern Glass Transition Temp (T g ) (°C) T onset (°C) Compression Temperature (25% Above T g ) (°C) Storage Temp (°C) Recovery Temp (°C) µGroove Recovery Rate at 37°C (s) 80:20 wt% tBA:PEGDMA 1000 44 ± 1 32 ± 2 55 25 37 25 ± 6 80:20 wt% tBA:PEGDMA 750 50 ± 1 38 ± 1 63 25 37 38 ± 9 80:20 wt% tBA:PEGDMA 550 58 ± 1 50 ± 1 73 25 37 60 ± 11

PAGE 142

129 Figure 6. 3 . SEM micrographs confirm pattern transfer to the SMP surface with groove widths and depths ranging from 55 65 µ m (µ ± SD). C were used to measure wettability . The original, static SMP surfaces exhibited hydrophobic surfaces, as demonstrated by the contact angles ranging from 110° 113°. However, after compression, contact angles to 98° 101°, decreasing hydrophobicity approximately 10% compared to the original sur face. After microgroove recovery, contact angles also recover to the original values, around 112° 113°. Statistically significant differences in contact angle are present between the original surface and the compressed surface as well as the compressed sur face and the recovered surface for all formulations. In addition, there are also significant differences in wettability between the original and recovered surfaces, which are greater for the tBA:PEGDMA550 and tBA:PEGDMA750 (p < 0.001) compared to the tBA:P EGDMA1000 (p < 0.05).

PAGE 143

130 Prior to assessing cell adhesion, surface recovery of the previously deformed SMPs was measured. All SMP formulations recover their original shape a t physiological conditions as determined by measuring groove depth prior to compressi on and after recovery . Recovery after material deformation has been investigated for these materials specifically on the bulk level to ensure that the shape memory effect leads to device deployment for devices such as cardiovascular stents and hernia meshe s . 18, 178, 179 Here, it is confirmed that the shape memory capabilities extend to the surface of the material as measured by the groove depth, as shown in Figure 6. 4 . Moreover , like the bulk recovery data, t he SMPs with the higher T g s also demonstrated longer recovery times, which was measured as the time required for the micropattern to regain its original configuration . Figure 6. 4 . Contact angle and groove depth of micropatterned SMP surfaces before compression, after compression and after recovery. Original surface c ontact angle is similar across the variations in SMP formulation . W hen the surface pattern is compressed, the wettability decreases approximately 10% , but recovers to the original value after compression recovery.

PAGE 144

131 Atomic Force Microscopy (AFM) provides both qualitative and quantitative information regarding a material surface. Here, AFM was conducted on the original, passive surface (no surface manipulation), the compressed surface, as well as the recovered surface . There are variations in roughness between the formulations, but since the surface feature is unchanged, roughness is also similar across the surfaces (Figure 6. 5). When th e surface is compressed, roughness also decreases as a result of the load application. The differences in roughness are not statistically significant between formulations, which suggests that the roughness measured is a result of pattern introduction to th e surface.

PAGE 145

132 Figure 6. 5 . Atomic Force Microscopy of 80:20 wt% tBA:PEGDMA 550,750, 1000. Roughness is generally similar across all SMPs , for each surface condition, due to the surface feature (microgroove) being the major contributing factor of the surface roughness for these patterned surfaces . The reduction on roughness for compressed surfaces is a result of compressive load, but roughness recovers to nearly its original value for all SMPs. Endothelial cell a dhesion was assessed using Live/Dead imaging fluorescence microscopy , 24 hours after cell introduction . Live cells fluoresce green, whereas dead cells are marked in red , as shown in Figure 6 .6 . Cell adhesion appears to increase on temperature responsive surfaces compared to passive surfaces, as demonstrated by the increased presence of marked red, but the increases in dead cells are not statistically significant. Cell a dhesion was also quantified by counting adherent cells. Both tBA:PEGDMA550 group and tBA:PEGDMA750 group displayed cell adhesion increases of approximately 20 25% between temperature responsive and passive surfaces. Alternatively, the tBA:PEGDMA1000 formul ation experienced a smaller, statistically insignificant 5% increase between temperature responsive and passive substrates.

PAGE 146

133 Figure 6 .6 . F igure 6. 7 .

PAGE 147

134 Endothelial cell alignment was assessed by measuring nuclear and actin orientation and alignment angle relative to the groove, as displayed in Figure 6. 8 . S pecifically, the angle formed between the nucleus/actin fiber and the horizontal provided the me tric for quantification of cell orientation. N ucleus and actin fiber orientation angles of 90° indicate cellular position is parallel to the direction of the gr oove. Endothelial cells demonstrate small increases, in orientation parallel to the grooves as depicted by increases in the percentage of cells for orientation angles 80° 90°. Nucleus and actin fiber orientation was further verified by normalizing to a gau ssian fit function.

PAGE 148

135

PAGE 149

136 Figure 6. 8 . Discussion This study proposes using temperature responsive , shape memory surface microgrooves to increase endothelial cell a dhesion compared to passive microgrooved surfaces on select acrylate based shape memory polymers . Microgrooves serve as topographical cues to encourage endothelial cells to align axially to the groove, which should aid in mimic king cell organization in vivo . Due to the initial low profile of shape memory surfaces and gradual recovery of the grooved surface at physiological temperature , cells populate more of the surface compared to a conventionally grooved surface with high profile surface features . Once the surfa ce has fully recovered to its original shape, the grooves assist with maintaining cell alignment. Increasing endothelial cell adhesion and maintaining cellular alignment should improve the utility of these materials, potentially bring ing the m one step clos er for use in stent fabrication. Similar studies have investigated using temperature activated SMPs to investigate cell alignment, cell sheet formation, cell delivery, but to the best of our knowledge, few studies have focused on the utility of using tempe rature responsive surfaces to improve cell adhesion on acrylate based SMP materials . 210, 262 Endothelialization of the material surface is one strategy to optimize cardiovascular material pe rformance in vivo . 266 A healthy endothelium provides an antiproliferative and anticoagulant surface, which prevents rest enosis inducing intimal hyperplasia as well as thrombosis; as such, in situ endothelialization is an attractive method to improve device integration . 267 Initially the shape memory effect was widely investigated for device deployment or drug delivery fr om the bulk material, but to fully optimize these SMPs for cardiovascular

PAGE 150

137 stents, rapid endothelialization is required . 18, 47, 177 Here, we again used a facile and inexpensive method, mold casting using a metal printed mold, for increasing cell adhesion using shape memory . 265 3D Metal printing provides a simple, cheap , reproducible and robust method for mold fabrication while sti ll generating patterns on the tens of microns level . 225, 268 There is some variability in the specified surface feature, but the dimensions are largely maintained, and subsequently trans ferred to the SMP using the casting method which is commonly used for UV crosslinked polymers, as seen in the scanning electron micrographs . 269 Many methods previously used to encourage surface patterning often require special equipment or training, limiting their practice and scalability . 85, 226 Pattern addition to the SMP surface appeared to be the major contributing factor in roughness and contact angle measurements. We previously showed that introduction of shallow microgrooves to the SMP surface generated rougher surfaces with small increases in contact angle . 265 These microgrooved surfaces demonstrate even higher roughness and hydroph obicity, which may contribute to the increased cell adhesion on the microgroove plateaus, similar to results in related studies which found correlations between increased surface roughness and improved cell adhesion for various cell types . 270 272 While there are variations in surface roughness between the formulations, the differences are not statistically significant and the va lues are close in range, which may suggest that the roughness is primarily influenced by the pattern. Notably , contact angle did demonstrate statistically significant differences between original & compressed as well as compressed & recovered surfaces. During compression, the surface features are compressed to 80% of their original profile, which reduced hydropho bicity

PAGE 151

138 by 10%; however, the hydrophobicity observed on these microgrooved surfaces were similar to or slightly greater than the hydrophobicity observed on shallow microgrooved surfaces, which continued to provide an attractive surface for the initial cell adhesion . 265, 273 However, as previously determined, contributions from wettability and/or roughness are difficult to discern in terms of their specific contribution to cell adhesion . 75, 230 Temperature responsive surfaces demonstrated increased cell adhesion compared to passive surfaces. The greater cell presence on the temperature responsive tBA:PEGDMA550 surface may largely be a result of the slow surface feature recovery time of the formulations examined ; t he delayed surface feature recovery provided a low profile surface for cell adhesion prior to surface recovery. Notably, the tBA:PEGDMA550 and tBA:PEGDMA750 temperature responsive surfaces displ ayed statistically significant increases in cell adhesion compared to their passive analogues, whereas the tBA:PEGMDA1000 surface did not experience a significant increase; further differences in wettability of the original and recovered tBA:PEGDMA550 and tBA:PEGDMA750 surfaces demonstrated greater statistically significant differences (p < 0.001) , whereas the differences in wettability of the original and recovered tBA:PEGMDA1000 surface were statistical ly significan t, but to a lesser degree (p < 0.05) . Wh ile the tBA:PEGDMA1000 surface recovered most rapidly, which may have contributed to the comparable cell adhesion result between recovered and passive surfaces, there may be a correlation between the increased cell adhesion on recovered surfaces and change s in surface wettability due to surface manipulation. Cell alignment, as measured by calculating the orientation angle between the nuclei and actin fibers relative to the groove direction, did not vary significantly between temperature activated and passive surfaces. Since the main difference between the groups was surface

PAGE 152

139 manipulation and not surface feature dimension, the insignificant differences in alignment are not surprising . However, these grooves are higher profile compared to the shallow, sta tic microgrooves that were used in the previous study and unlike other studies that have investigated the correlation between cell depth and alignment, the endothelial cells did not demonstrate superior alignment on the high profile grooves . 274 Since the width of the groove plateau (~50µm) was wider than the width of an individual endothelial cell (~10 20µm), optimal alignment may require a reduction in microgroove width . During this exploratory investigation, temperature responsive surfaces demonstrated increased c ell adhesion compared to passive surfaces of the same surface feature dimension . Here, we selected one surface feature with sp ecific dimensions , 50µm width and depth, and chose to focus on the effect of shape memory on cell adhesion. To further understand the benefits and limitations of using temperature responsive surfaces, additional topographies such as varied groove dimensions and shapes , pits, islands, waves, etc., should be investigated. S ince alignment did not significantly improve compared to previous work, further optimization of the current surface feature dimension, which will include reducing the groove platea u width to better match the width of elongated ECs as well as reducing groove depth to give endothelial cells the opportunity to create cell cell contact while still maintaining enough depth to utilize the shape memory capabilities of the material, should be considered.

PAGE 153

140 CHAPTER VII CONCLUSIONS, LIMITATIONS AND FUTURE WORK Conclusions Th is work ha s demonstrated that these acrylate based SMPs encourage endothelial cell adhesion without further chemical or bioactive surface modification. The initial stud ies were used to optimize the SMP surface by varying the components that compose the SMP , tBA monomer and PEGDMA crosslinker , to determine the best surface for cell adhesion. Previous work has demonstrated biocompatibility primarily throug h cytotoxicity testing of similar and select acrylate based SMPs ; the materials were evaluated for changes i n cell ular morphology such as granulation, crenation, rounding, detachment and/or lysis. Ma terial evaluation which was performed according to ISO sta ndards, found that these materials are non reactive and thus do not cause cell lysis or a reduction in cell growth . 165, 1 79 The SMPs ranged from high tBA based materials to high PEGDMA based materials and were assessed for material properties as well as cell adhesion. It was determined that the SMPs containing greater tBA content, which were stiffer, mildly wettable ( contact angles near 90°) and relatively rough er , encouraged optimal EC adhesion. Formulations containing high amounts of tBA, i.e., greater than 90 wt%, limited EC adhesion , while formulations containing 50 wt% tBA or less did not support sustained EC adhe sion . Further, the 80:20 wt% tBA:PEGDMA SMPs have previously been investigated for thermo mechanical properties and device delivery and appear to be promising candidates for cardiovascular stent use. Thus, for surface optimization studies, wherein the SMP s urface was topographically modified for cell adhesion, 80:20 wt% tBA:PEGDMA formulations were selected .

PAGE 154

141 Since most endot helial cells exhibit some form of organization in native blood vessels, the surface of the SMP was patterned using metal printed molds containing shallow microgrooves, approximately 50µm width and 10µm depth . Since microgroove introduction was limited to the surface, the bulk properties , such as glass transition temperature, were similar to unpatterned SMPs. Microgroo ve additio n increase d roughness about 10% , but values were not statistically significant ; alternatively, hydrophobicity of the surfaces did experience statistically significant increases between unpatterned and microgrooved surfaces with 3 6% increase in c ontact angle . The increased hydrophobicity and roughness most likely contributed to the increased cell adhesion on microgrooved surfaces and comparable survival , with adherent cells also demonstrating increase s in alignment , parallel to the microgroove. Adding topographical cues to the SMP surfac e provide d an improved surface for cell a dhesion without introducing any chemical or bioactive agents that may cause side effects either from the agent itself or from its depletion. Compression and re covery testing of select acrylate based polymer surface features was conducted for potential use in biomedical devices. While these studies were conducted on a scale larger than what would be consequential at the cell level, the data should extend to small er feature sizes, especially on the micron scale. One of the limitations of many biomedical devices is that the scaffolds are stationary; using SMPs provides a more dynamic surface . Of the samples tested, only the 80:20 wt% tBA:PEGDMA formulations exhibite d shape deformation and recovery. When exposed to temperatures at least 25% greater than glass transition temperature, percent recovery and recovery time were mostly unaffected by formulation, protrusion size and compression time. Surface recovery was near ly 100%, regardless of the recovery temperature, however, the time required to recover the surface feature varied. All of the 80:20 wt%

PAGE 155

142 tBA:PEGDMA formulations recovered at physiological temperature, but SMPs with higher T g s exhibited a longer recovery tim e , which may be preferred if a delayed deployment is desired. These initial data show promise for use of SMPs to introduce controlled surface features. A ctive surfaces have become more attractive due to their ability to change with their surroundings ; as such, extend ing the shape memory aspect to the surface , specifically to the surface features , should further improve material functionality . Since micr og rooves were previously used to encourage en dothelial cell adhesion and alignment, microgrooves of the same width, approximately 50µm, were then given an increased depth , creating distinct plateaus and crevices, so that the grooves could be temperature activated using shape memory. By adding this feature and using the shape memory effect of the material , we were able to further increase cell adhesion over more of the SMP surface by initially generating low profile surface features prior to recovery , as shown in Figure 7.1 . These high er profile microgrooved surfaces demonstrated increased roughness and hydrophobicity compared to the previous shallow microgrooved surface which may have contributed to some of the initial increases in cell adhesion . C ell alignment , however, did not show significant differences between passive and temperature active surfaces, nor did cell alignment increase significantly compared to the previous shallow microgrooves .

PAGE 156

143 Figure 7.1 . Figure s ummary of percentage e ndothelial cell adhesion , relative to ECs introduced on Day 0 of the study, on all SMP s & surfaces . EC adhesion shows improvements as low profile microgrooves are introduced (Aim 2) and finally as higher profile microgrooves are used to improve cell adhesion via shape memory (Aim 3 ) . Limitations Although these acrylate based SMPs have shown promise in encouraging endothelial cells on these material s , there are limitations to this work. T he native extracellular matrix provides more than just a scaffold for endothelial cells ; it also produces fibrous proteins, which include collagen, laminin, and elastin as well as glycosaminoglycans (GAGs) , which provide a grounding substance to resist compressive forces . 275 W hile these SMP s attempt t o mimic in vivo

PAGE 157

144 condition s and provide support for damaged vessels, the materials may experience limited cell adhesion and function without the presence of these molecules . 18 It may be beneficial to add these molecules to the SMP surface in future studies to determine if there is an advantage to incorporating fibrous proteins and GAG s improved cell adhesion. However, the primary motivation of limiting surface modifications to topographical methods is to reduce regulatory burden, which would allow these materials and devices to reach clinical utility more rapidly ; therefore, the cost benefit tradeoff of incorp orating these molecules is a major consideration that must be addressed prior to clinical implementation . The complexity of cell biomaterial surface interactions can also be a limiting factor for many of the studies conducted here. Wettability and roughn ess as well as some brief analyses of surface stiffness were the major surface characteristics analyzed ; however, there are at least a couple more surface characteristics that may influence cell adhesion , including, but not limited to, chemical species and surface charge, for example, which would need to be verified using ATR FTIR, XPS or ESCA for chemical species analysis or SPM or AFM for surface charge measurements . 40, 276 While metal printing offers a facile, robust and relatively inexpensive method for mold fabrication, the surface features of the mold could not be smaller than the 50 µ m feature dimension used for these studies , due to the printhead nozzle used to fabricate the mold. However, with advancements in 3D metal printing technolog y and with 3D printing becoming more mainstre am, the ease of generating smaller surface features will potentially increase while the cost will likely decrease , allowing for smaller, yet customizable surface features . Additionally, s ome experimental aspects also resulted in limitations. The Live/Dead assay used to quantify cell adhesion would often cause cells to become rounded and apoptotic,

PAGE 158

145 particularly at early timepoints. This, at times, resulted in fewer adherent cells for fixed cell analyses, such as the actin & nuclei orientation studies. On th e material side, these SMPs were often difficult to image following fluorescent staining , as many of the antibodies and conjugated fluorophores are readily absorbed by the material . This was particularly challeng ing when attempting to stain and image focal adhesions ; a lthough the focal adhesion complexes are considered to be large , the individual components that are targeted by specific antibodies (vinculin, paxillin, FAK, etc.) are small, and thus would with the background generated by dye absorption . F ocal adhesion marking and imaging would have provided further confirmation of cell organization, since these protein complexes physically connect the cellular cytoskeleton to the surrounding extracellular matrix, and thus play an import ant role during cell adhesion . However, the absorption capabilities of the SMP proved to be dominant in this case , which essentially washed out any signal emitted by the antibodies on the focal adhesions . 277 Future Work Although this work has made progress in potentially using these materials for implanted blood contacting devices, these studies present preliminary dat a and, for these materials to be seriously considered for use in cardiovascular stents, additional work is required. Since most blood contacting devices, are implanted in the blood vessel under flow conditions, all surfaces should be examined for their ability to re tain shape and recruit and maintain cell adhesion under physiological flow conditions. Prior to cell introduction, all surfaces, but primarily the surfaces that contain high profile microgrooves, should be subjected to flow conditions to ensure that flow i s not significantly disturbed, which may lead to endothelial cell dysfunction. Further, analysis under flow , using a bioreactor or similar setup, should allow for the measurement of cell

PAGE 159

146 adhesion from circulating endothelial progenitor cells vs. cell recru itment from adjoining healthy endothelium . 278 These studies focused heavily on SMP surface characterization and optimization for endothelial cell adhesion and later alignment, but additional work that focuses more on the effect of these surfaces on the endot helial cells should be conducted. Specifically, protein regulation of cells on the varying surfaces should be analyzed and may provide some additional insight into which features are more beneficial to cell adhesion and which surface features may be more i nhibitive, since studies have shown that stiffer or rougher surfaces may upregulate or downregulate certain proteins that influence cell adhesion. In addition to the current microgrooved surface, other surface features such as pillars, waves, or more custo mized surface features may be applied and verified for cell adhesion , particularly in environments that may not require grooved features for alignment . A brief study was performed using only the 80:20 tBA:PEGDMA 550 formulation, but instead of varying the surface feature shape, the dimensions of the grooves were varied. Again, pattern transfer to the surface up to 150µm was confirmed using SEM , which also provided quantitative measurements of the groove width and depth. Similar to the 50µm grooved SMPs, the depths of the grooves were a few to several microns greater than the widths of the groove plateaus.

PAGE 160

147 Figure 7. 2 . SEM of varying groove widths and depth of 80:20 wt% tBA:PEGDMA550 SMPs. SEM confirm s pa ttern transfer to the surface of SMP as well as provides qualitative measurements of the surface feature, primarily width and depth of grooves. Similar methods to those used previously were employed to compress and recover the surface and cell adhesion to these temperature responsive surfaces were compared to passive, microgrooved surfaces. Table 7.1. Glass Transition Temperature, Compression Temperature and Microgroove Recovery Rate o f 80:20 tBA:PEGDMA550 SMPs with microgroove widths of 50µm, 100 µm & 150 µm Feature Width

PAGE 161

148 Unlike the previous study however, preliminary data was collected up to 7 days after cell introduction . A s shown in Figure s 7. 3 7. 5 , there is evidence of cell survival on both passive and temperature responsive surfaces. Furthermore, endothelial cells on temperature responsive surfaces appear to proliferate and provide greater coverage compared to cells on passive surfaces. Further, on microgrooved surfaces up to 100µm, cells appear to demonstrate high survival , eventually populating the groove plateaus. However, for the 150µm groove surfaces , ce ll adhesion appears to be less optimal on both the passive and temperature responsive surfaces , which suggests that ECs prefer microgrooved surfaces up to 100µm . The results presented here are again, preliminary, and additional replicates of this experimen t are required to confirm these findings before fi nal conclusions can be drawn . Figure 7. 3 . Cell adhesion on passive and temperature responsive 50µm 80:20 wt% tBA:PEGDMA550 . Cell presence appears to be greater on the temperature responsive surface and continues to increase at a greater rate compared to the passive surface, up to 7 days. Scale bar = 500µm.

PAGE 162

149 Figure 7. 4 . Cell adhesion on passive and temperature responsive 100µm 80:20 wt% tBA:PEGDMA55 0. Similar to the 50µm µgrooved surface, cell presence appears to be greater on the temperature responsive surface and continues to increase at a greater rate compared to the passive surface, up to 7 days. However, cells do not appear to adhere to the crevices of the grooves on either surface. Scale bar = 500µm. Figure 7. 5 . Cell adhesion on passive and temperature responsive 150µm 80:20 wt% tBA:PEGDMA550. Cell presence appears to be comparable between the temperature responsive

PAGE 163

150 surface and the pas sive surface up to 7 days after cell seeding. Similar to the 100µm grooved surface, cells do not appear to adhere to the crevices of the grooves on either surface. Scale bar = 500µm. Use of t hese micro grooved surfaces, both the shallow , static microg rooved surfaces and the deeper , temperature responsive surfaces, may be applicable to other cell types that would benefit from cell alignment such as muscle cells or Schwann cells in neurons . 205 As such, these temperature active surfaces may be applicable to other devices or even further investigated to deliver drugs, bioactive agents or even cells for wound healing and tissue regeneration , some of which has been investigated by other groups . 261, 279 Many of the studies conducted here were on the shorter spectrum , ranging from 24 hours to 7 days. However, extended time studies up to 14 day s, 28 days , or beyond may provide additional information regarding EC survival on these surfaces. Extended time point studies may also provide some v erification as to how well th ese materials perform in a nutrient rich environment long term . While extended studies may be necessary to bring these materials to clinical relevance , depending on the experiment length, the y can be expensive and time consumin g to conduct due to the cost of the materials and cell culture maintenance required during the study. Finally, prior to attaining clinical relevance, animal studies must be conducted to confirm in vitro results. Rodent s are a good initial animal model to obtain preliminary data as a high throughput screening tool; furthermore, rodents are useful to study mechanicals of cardiovascular physiology and disease since genetic manipulation is simpler . 280 282 However, rabbit or porcine models would also be required prior to clinical trials to generate more clinically relevant data, since these models are more similar, both anatomically and physiologically, to humans . 280, 283 Animal studies, specifically in the porcine model, have been conducted on some of these

PAGE 164

151 materials in the hernia mesh form, which have confirmed biocompatibility and even demonstrated reduced scar tissue formation compared to control materials . 179 Once data from animal studies is acquired, the se materials may be tested in a clinical trial setting. It should be noted that related SMPs have been commercialized, specifically the Medusa Multi Coil (MMC) from EndoShape, Inc., which is an FDA approved embolization device with synthetic fibers to promote thrombogenicity . 284 When compared with metal coils, George et. al., found that use of the MMC ha s the potential for significant advantages over the use of metal coils, but longer term follow up is required . 28 4 Nonetheless , these studies provide some clinical support for implementing the SMPs used in this work in animal studies a s well as subsequent clinical studies .

PAGE 165

152 REFERENCES 1. Martinez, A. W.; Chaikof, E. L., Microfabrication and nanotechnology in stent design. Wiley Interdiscip Rev Nanomed Nanobiotechnol 2011, 3 (3), 256 68. 2. Benjamin, E. J.; Virani, S. S.; Callaway, C. W.; Chamberlain, A. M.; Chang, A. R.; Cheng, S.; Chiuve, S. E.; Cushman, M.; Delling, F. N.; Deo, R.; Ferranti, S. D. d.; Ferguson, J. F.; Fornage, M.; Gillespie, C.; Isasi, C. R.; Jiménez, M. C.; Jordan, L. C.; Judd, S. E.; Lackland, D.; Lichtman, J. H.; Lisabeth, L.; Liu, S.; Longenecker, C. T.; Lutsey, P. L.; Mackey, J. S.; Matchar, D. B.; Matsushita, K.; Mussolino, M. E.; Nasir, Ritchey, M. D.; Rodriguez, C. J.; Roth, G. A.; Rosamond, W. D.; Sampson, U. K. A.; Satou, G. M.; Shah, S. H.; Spartano, N. L.; Tirschwell, D. L.; Tsao, C. W.; Voeks, J. H.; Willey, J. Z.; Wilkins, J. T.; Wu, J. H.; Alger, H. M.; Wong, S. S.; Muntner, P., Heart Disease and Stroke Stati stics—2018 Update: A Report From the American Heart Association. 2018, 137 (12), e67 e492. 3. Miller, D. C.; Thapa, A.; Haberstroh, K. M.; Webster, T. J., Endothelial and vascular smooth muscle cell function on poly(lactic co glycolic acid) with nano structured surface features. Biomaterials 2004, 25 (1), 53 61. 4. Chandy, T.; Das, G. S.; Wilson, R. F.; Rao, G. H., Use of plasma glow for surface engineering biomolecules to enhance bloodcompatibility of Dacron and PTFE vascular prosthesis. Bioma terials 2000, 21 (7), 699 712. 5. Reape, T. J.; Groot, P. H., Chemokines and atherosclerosis. Atherosclerosis 1999, 147 (2), 213 25. 6. Khan, W.; Farah, S.; Domb, A. J., Drug eluting stents: developments and current status. J. Control. Release 2012, 161 (2), 703 12. 7. Bergheanu, S. C.; Bodde, M. C.; Jukema, J. W. J. N. H. J., Pathophysiology and treatment of atherosclerosis. 2017, 25 (4), 231 242. 8. Ross, R., Rous Whipple Award Lecture. Atherosclerosis: a defense mechanism gone awry. The American journal of pathology 1993, 143 (4), 987 1002. 9. Erl, W.; Weber, P. C.; Weber, C., Monocytic cell adhesion to endothelial cells stimulated by oxidized low density lipoprotein is mediated by distinct endothelial ligands. Atherosclerosis 1998, 136 (2), 297 303. 10. Sitia, S.; Tomasoni, L.; Atzeni, F.; Ambrosio, G.; Cordiano, C.; Catapano, A.; Tramontana, S.; Perticone, F.; Naccarato, P.; Camici, P.; Picano, E.; Cortigiani, L.; Bevilacqua, M.; Milazzo, L.; Cusi, D.; Barlassina, C .; Sarzi Puttini, P.; Turiel, M., From endothelial dysfunction to atherosclerosis. Autoimmunity Reviews 2010, 9 (12), 830 834.

PAGE 166

153 11. Berliner, J. A.; Navab, M.; Fogelman, A. M.; Frank, J. S.; Demer, L. L.; Edwards, P. A.; Watson, A. D.; Lusis, A. J., Atherosclerosis: Basic Mechanisms. 1995, 91 (9), 2488 2496. 12. Jeng, J. R.; Chang, C. H.; Shih Ming, S.; Hui Chong, C., Oxidized low density lipoprotein enhances monocyte endothelial cell binding against shear stress induced detachment. Biochimica et B iophysica Acta (BBA) Molecular Cell Research 1993, 1178 (2), 221 227. 13. Sakakura, K.; Nakano, M.; Otsuka, F.; Ladich, E.; Kolodgie, F. D.; Virmani, R., Pathophysiology of Atherosclerosis Plaque Progression. Heart, Lung and Circulation 2013, 22 (6), 399 411. 14. NIH NHLBI Coronary Artery Bypass Grafting. https://www.nhlbi. nih.gov/health topics/coronary artery bypass grafting (accessed 7 January). 15. Howell, S. J., Carotid endarterectomy. BJA: British Journal of Anaesthesia 2007, 99 (1), 119 131. 16. Erickson, K. M.; Cole, D. J., Carotid artery disease: stenting vs endart erectomy. BJA: British Journal of Anaesthesia 2010, 105 (suppl_1), i34 i49. 17. NIH NHLBI Carotid Endarterectomy, also known as Carotid Artery Surgery https://www.nhlbi.nih.gov/health topics/carotid endarterectomy (accessed 07 January). 18. Yakacki, C. M.; Shandas, R.; Lanning, C.; Rech, B.; Eckstein, A.; Gall, K., Unconstrained recovery characterization of shape memory polymer networks for cardiova scular applications. Biomaterials 2007, 28 (14), 2255 63. 19. Iqbal, J.; Gunn, J.; Serruys, P. W., Coronary stents: historical development, current status and future directions. Br. Med. Bull. 2013, 106 (1), 193 211. 20. Grabow, N.; Martin, D. P.; Sch mitz, K. P.; Sternberg, K., Absorbable polymer stent technologies for vascular regeneration. 2010, 85 (6), 744 751. 21. Boura, C.; Menu, P.; Payan, E.; Picart, C.; Voegel, J. C.; Muller, S.; Stoltz, J. F., Endothelial cells grown on thin polyelectrol yte mutlilayered films: an evaluation of a new versatile surface modification. Biomaterials 2003, 24 (20), 3521 30. 22. Mani, G.; Feldman, M. D.; Patel, D.; Agrawal, C. M., Coronary stents: a materials perspective. Biomaterials 2007, 28 (9), 1689 710. 23. Tran, H. S.; Puc, M. M.; Hewitt, C. W.; Soll, D. B.; Marra, S. W.; Simonetti, V. A.; Cilley, J. H.; DelRossi, A. J., Diamond like carbon coating and plasma or glow discharge treatment of mechanical heart valves. J. Invest. Surg. 1999, 12 (3), 133 40.

PAGE 167

154 24. Kolodgie, F. D.; Nakazawa, G.; Sangiorgi, G.; Ladich, E.; Burke, A. P.; Virmani, R., Pathology of atherosclerosis and stenting. Neuroimaging Clin. N. Am. 2007, 17 (3), 285 vii. 25. Alanazi, A.; Nojiri, C.; Noguchi, T.; Kido, T.; Komatsu, Y.; Hirakuri, K.; Funakubo, A.; Sakai, K.; Fukui, Y., Improved blood compatibility of DLC coated polymeric material. ASAIO J. 2000, 46 (4), 440 3. 26. Peng, T.; Gibula, P.; Yao, K. D.; Goos en, M. F., Role of polymers in improving the results of stenting in coronary arteries. Biomaterials 1996, 17 (7), 685 94. 27. van der Hoeven, B. L.; Pires, N. M.; Warda, H. M.; Oemrawsingh, P. V.; van Vlijmen, B. J.; Quax, P. H.; Schalij, M. J.; van der Wall, E. E.; Jukema, J. W., Drug eluting stents: results, promises and problems. Int. J. Cardiol. 2005, 99 (1), 9 17. 28 . Joner, M.; Finn, A. V.; Farb, A.; Mont, E. K.; Kolodgie, F. D.; Ladich, E.; Kutys, R.; Skorija, K.; Gold, H. K.; Virmani, R., Pathology of drug eluting stents in humans: delayed healing and late thrombotic risk. J. Am. Coll. Cardiol. 2006, 48 (1) , 193 202. 29. Chu, P. K., Enhancement of surface properties of biomaterials using plasma based technologies. Surface and Coatings Technology 2007, 201 (19 20), 8076 8082. 30. Costa, M. A.; Simon, D. I., Molecular basis of restenosis and drug eluting ste nts. Circulation 2005, 111 (17), 2257 73. 31. Huang, N.; Yang, P.; Leng, Y. X.; Ying Chen, J.; Wang, J.; Wan, G.; Sun, H.; Wu, X.; Sha Zhao, A., Improving Blood Compatibility of Cardiovascular Devices by Surface Modification . 2007; Vol. 342 343, p 801 804. 32. de Mel, A.; Jell, G.; Stevens, M. M.; Seifalian, A. M., Biofunctionalization of biomaterials for accelerated in situ endothelialization: a review. Biomacromolecules 2008, 9 (11), 2969 79. 33. Reinhart King, C. A., Chapter 3 Endothelial Cell Adhesion and Migration. In Angiogenesis In Vitro Systems , 2008; pp 45 64. 34. dela Paz, N. G.; D'Amore, P. A., Arterial versus venous endothelial cells. Cell Tissue Res. 2009, 335 (1), 5 16. 35. Melchiorri, A. J. ; Hibino, N.; Yi, T.; Lee, Y. U.; Sugiura, T.; Tara, S.; Shinoka, T.; Breuer, C.; Fisher, J. P., Contrasting Biofunctionalization Strategies for the Enhanced Endothelialization of Biodegradable Vascular Grafts. Biomacromolecules 2015, 16 (2), 437 44 6. 36. Yoder, M. C., Human Endothelial Progenitor Cells. 2011 .

PAGE 168

155 37. Anderson, D. E. J.; Glynn, J. J.; Song, H. K.; Hinds, M. T., Engineering an endothelialized vascular graft: a rational approach to study design in a non human primate model. PLoS One 201 4, 9 (12), e115163 e115163. 38. Melchiorri, A. J.; Hibino, N.; Fisher, J. P., Strategies and Techniques to Enhance the In Situ Endothelialization of Small Diameter Biodegradable Polymeric Vascular Grafts. 2013, 19 (4), 292 307. 39. Heath, D. E., Promoti ng Endothelialization of Polymeric Cardiovascular Biomaterials. 2017, 218 (8), 1600574. 40. Temenoff, J. S.; Mikos, A. G., Biomaterials The Intersection of Biology and Materials Science . Pearson Prentice Hall: 2008; p 478. 41. Luscher, T. F.; Steffel, J.; Eberli, F. R.; Joner, M.; Nakazawa, G.; Tanner, F. C.; Virmani, R., Drug eluting stent and coronary thrombosis: biological mechanisms and clinical implications. Circulation 2007, 115 (8), 1051 8. 42. Pypen, C. M.; Plenk, H., Jr.; Ebel, M. F.; Svagera, R.; Wernisch, J., Characterization of microblasted and reactive ion etched surfaces on the commercially pure metals niobium, tantalum and titanium. J. Mater. Sci. Mater. Med. 1997, 8 (12), 781 4. 43. Pavithra, D.; Doble, M., Biofilm formation, bacterial adhesion and host response on polymeric implants -issues and prevention. Biomed. Mater. 2008, 3 (3), 034003. 44. Ormiston, J. A.; Serruys, P. W.; Regar, E.; Dudek, D.; Thuesen, L.; Webster, M. W. I.; Onuma, Y.; Garcia Garcia, H. M.; McGreevy, R.; Veldhof, S., A bioabsorbable everolimus eluting coronary stent system for patients with single de novo coronary artery lesions (ABSORB): a prospective open label trial. The Lancet 2008, 371 (9616), 899 907. 45. Di Mario, C.; Ferrante, G., Biodegradable drug eluting stents: promises and pitfalls. The Lancet 2008, 371 (9616), 873 874. 46. Behl, M.; Kratz, K.; Noechel, U.; Sauter, T.; Lendlein, A., Temperature memory polymer actuators. Proc. Natl. Acad . Sci. U. S. A. 2013, 110 (31), 12555 9. 47. Yakacki, C. M.; Shandas, R.; Safranski, D.; Ortega, A. M.; Sassaman, K.; Gall, K., Strong, Tailored, Biocompatible Shape Memory Polymer Networks. Adv Funct Mater 2008, 18 (16), 2428 2435. 48. Lendlein, A.; Kelch, S., Shape Memory Polymers. 2002, 41 (12), 2034 2057. 49. Abrahamson, E. R.; Lake, M. S.; Munshi, N. A.; Gall, K., Shape Memory Mechanics of an Elastic Memory Composite Resin. 2003, 14 (10), 623 632.

PAGE 169

156 50. Karaaslan, M. A.; Tshabalala, M. A.; Busc hle Diller, G., Semi interpenetrating polymer network hydrogels based on aspen hemicellulose and chitosan: Effect of crosslinking sequence on hydrogel properties. 2012, 124 (2), 1168 1177. 51. Safranski, D. L.; Gall, K., Effect of chemical structure and crosslinking density on the thermo mechanical properties and toughness of (meth)acrylate shape memory polymer networks. Polymer 2008, 49 (20), 4446 4455. 52. Sun, L.; Huang, W. M.; Wang, C. C. ; Zhao, Y.; Ding, Z.; Purnawali, H., Optimization of the shape memory effect in shape memory polymers. 2011, 49 (16), 3574 3581. 53. Nair, D. P.; Cramer, N. B.; Scott, T. F.; Bowman, C. N.; Shandas, R., Photopolymerized Thiol Ene Systems as Shape Mem ory Polymers. Polymer (Guildf) 2010, 51 (19), 4383 4389. 54. Voit, W.; Ware, T.; Gall, K., Radiation crosslinked shape memory polymers. Polymer 2010, 51 (15), 3551 3559. 55. Behl, M.; Razzaq, M. Y.; Lendlein, A., Multifunctional shape memory polymers. Adv Mater 2010, 22 (31), 3388 410. 56. Hegemann, D.; Brunner, H.; Oehr, C., Plasma treatment of polymers for surface and adhesion improvement. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 2003, 208 , 281 286. 57. Farè, S.; Valtulina, V.; Petrini, P.; Alessandrini, E.; Pietrocola, G.; Tanzi, M. C.; Speziale, P.; Visai, L., In vitro interaction of human fibroblast s and platelets with a shape memory polyurethane. 2005, 73A (1), 1 11. 58. Goddard, J. M.; Hotchkiss, J. H., Polymer surface modification for the attachment of bioactive compounds. Progress in Polymer Science 2007, 32 (7), 698 725. 59. Angelova, N.; Hunkeler, D., Rationalizing the design of polymeric biomaterials. Trends Biotechnol. 1999, 17 (10), 409 21. 60. Lendlein, A.; Behl, M.; Hiebl, B.; Wischke, C., Shape memory polymers as a technology platform for biomedical applications. Exp ert Rev. Med. Devices 2010, 7 (3), 357 79. 61. Hersel, U.; Dahmen, C.; Kessler, H., RGD modified polymers: biomaterials for stimulated cell adhesion and beyond. Biomaterials 2003, 24 (24), 4385 4415. 62. Craighead, H. G.; James, C. D.; Turner, A. M. P. , Chemical and topographical patterning for directed cell attachment. Current Opinion in Solid State and Materials Science 2001, 5 (2), 177 184.

PAGE 170

157 63. Chu, P. K.; Chen, J. Y.; Wang, L. P.; Huang, N., Plasma surface modification of biomaterials. Materials S cience and Engineering: R: Reports 2002, 36 (5), 143 206. 64. Ma, Z.; Mao, Z.; Gao, C., Surface modification and property analysis of biomedical polymers used for tissue engineering. Colloids Surf B Biointerfaces 2007, 60 (2), 137 57. 65. Prasad, C. K.; Muraleedharan, C. V.; Krishnan, L. K., Bio mimetic composite matrix that promotes endothelial cell growth for modification of biomaterial surface. J. Biomed. Mater. Res. A 2007, 80 (3), 644 54. 66. Helmus, M. N.; Gibbons, D. F.; Cebon, D., Biocompatibi lity: meeting a key functional requirement of next generation medical devices. Toxicol. Pathol. 2008, 36 (1), 70 80. 67. Arima, Y.; Iwata, H., Effect of wettability and surface functional groups on protein adsorption and cell adhesion using well defined m ixed self assembled monolayers. Biomaterials 2007, 28 (20), 3074 82. 68. van Wachem, P. B.; Beugeling, T.; Feijen, J.; Bantjes, A.; Detmers, J. P.; van Aken, W. G., Interaction of cultured human endothelial cells with polymeric surfaces of different wettabilities. Biomaterials 1985, 6 (6), 403 408. 69. van Wachem, P. B.; Hogt, A. H.; Beugeling, T.; Feijen, J.; Bantjes, A.; Detmers, J. P.; van Aken, W. G., Adhesion of cultured human endothelial cells onto methacrylate polymers with varying surface wettability and charge. Biomaterials 1987, 8 (5), 323 8. 70. Ratner, B. D., Surface modification of polymers: chemical, biological and surface analytical challenges. Biosens. Bioelectron. 1995, 10 (9), 797 804. 71. Govindarajan, T.; Shandas, R., A Survey of Surface Modification Techniques for Next Generation Shape Memory Poly mer Stent Devices . 2014; Vol. 6, p 2309 2331. 72. Lyu, S. P.; Cernohous, J. J.; Bates, F. S.; Macosko, C. W., Interfacial Reaction Induced Roughening in Polymer Blends. Macromolecules 1999, 32 (1), 106 110. 73. Curtis, A.; Wilkinson, C., Topographical control of cells. Biomaterials 1997, 18 (24), 1573 83. 74. Meredith, J. C.; Sormana, J. L.; Keselowsky, B. G.; Garcia, A. J.; Tona, A.; Karim, A.; Amis, E. J., Combinatorial characterization of cell interactions with polymer surfaces. J. Biomed. Mate r. Res. A 2003, 66 (3), 483 90. 75. Lampin, M.; Warocquier, C.; Legris, C.; Degrange, M.; Sigot Luizard, M. F., Correlation between substratum roughness and wettability, cell adhesion, and cell migration. J. Biomed. Mater. Res. 1997, 36 (1), 99 108.

PAGE 171

158 7 6. Shadpour, H.; Allbritton, N. L., In situ roughening of polymeric microstructures. ACS Appl Mater Interfaces 2010, 2 (4), 1086 93. 77. Yeh, H. I.; Lu, S. K.; Tian, T. Y.; Hong, R. C.; Lee, W. H.; Tsai, C. H., Comparison of endothelial cells grown on different stent materials. J. Biomed. Mater. Res. A 2006, 76 (4), 835 41. 78. Noeske, M.; Degenhardt, J.; Strudthoff, S.; Lommatzsch, U., Plasma jet treatment of five polymers at atmospheric pressure: surface modifications and the relevance for adhesion. International Journal of Adhesion and Adhesives 2004, 24 (2), 171 177. 79. Ikada, Y., Surface modification of polymers for medical app lications. Biomaterials 1994, 15 (10), 725 736. 80. Avram, M.; Avram, M. A.; Bragaru, A.; Ghiu, A.; Iliescu, C. In Plasma Surface Modification of Polymer Substrates for Selective Hydrophobic Control , 2007 International Semiconductor Conference, Oct. 15 2007 Sept. 17 2007; 2007; pp 91 94. 81. Oehr, C., Plasma surface modification of polymers for biomedical use. Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 2003, 208 , 40 47. 82. McAuslan, B. R.; Johnson, G., Cell responses to biomaterials. I: Adhesion and growth of vascular endothelial cells on poly(hydr oxyethyl methacrylate) following surface modification by hydrolytic etching. J. Biomed. Mater. Res. 1987, 21 (7), 921 35. 83. Chung, T. W.; Liu, D. Z.; Wang, S. Y.; Wang, S. S., Enhancement of the growth of human endothelial cells by surface roughness a t nanometer scale. Biomaterials 2003, 24 (25), 4655 4661. 84. Ranjan, A.; Webster, T. J., Increased endothelial cell adhesion and elongation on micron patterned nano rough poly(dimethylsiloxane) films. Nanotechnology 2009, 20 (30), 305102. 85. Falconnet, D.; Csucs, G.; Grandin, H. M.; Textor, M., Surface engineering approaches to micropattern surfaces for cell based assays. Biomaterials 2006, 27 (16), 3044 63. 86. Khang, D.; Lu, J.; Yao, C.; Haberstroh, K. M.; Webster, T. J., The role of n anometer and sub micron surface features on vascular and bone cell adhesion on titanium. Biomaterials 2008, 29 (8), 970 83. 87. Martinez, E.; Engel, E.; Planell, J. A.; Samitier, J., Effects of artificial micro and nano structured surfaces on cell beha viour. Ann Anat 2009, 191 (1), 126 35.

PAGE 172

159 88. Flemming, R. G.; Murphy, C. J.; Abrams, G. A.; Goodman, S. L.; Nealey, P. F., Effects of synthetic micro and nano structured surfaces on cell behavior. Biomaterials 1999, 20 (6), 573 88. 89. Yim, E. K.; Rea no, R. M.; Pang, S. W.; Yee, A. F.; Chen, C. S.; Leong, K. W., Nanopattern induced changes in morphology and motility of smooth muscle cells. Biomaterials 2005, 26 (26), 5405 13. 90. Yu, Q.; Zhang, Y.; Chen, H.; Zhou, F.; Wu, Z.; Huang, H.; Brash, J. L., Protein adsorption and cell adhesion/detachment behavior on dual responsive silicon surfaces modified with poly(N isopropylacrylamide) block polystyrene copolymer. Langmuir 2010, 26 (11), 8582 8. 91. Karim, A.; Slawecki, T. M.; Kumar, S. K.; Do uglas, J. F.; Satija, S. K.; Han, C. C.; Russell, T. P.; Liu, Y.; Overney, R.; Sokolov, J.; Rafailovich, M. H., Phase Separation Induced Surface Patterns in Thin Polymer Blend Films. Macromolecules 1998, 31 (3), 857 862. 92. Dalby, M. J.; Yarwood, S. J.; Riehle, M. O.; Johnstone, H. J.; Affrossman, S.; Curtis, A. S., Increasing fibroblast response to materials using nanotopography: morphological and genetic measurements of cell response to 13 nm high polymer demixed island s. Exp. Cell Res. 2002, 276 (1), 1 9. 93. Wilkinson, C. D. W.; Riehle, M.; Ma, W.; Gallagher, J.; Curtis, A., The Use of Materials Patterned on a Nano and Micro Metric Scale in Cellular Engineering . 2002; Vol. 19, p 263 269. 94. del Campo, A.; Arzt, E., Fabrication approaches for generating complex micro and nanopatterns on polymeric surfaces. Chem. Rev. 2008, 108 (3), 911 45. 95. Nie, Z.; Kumacheva, E., Patterning surfaces with functional polymers. Nature Materials 2008, 7 , 277 . 96. Xu, C.; Yang, F.; Wang, S.; Ramakrishna, S., In vitro study of human vascular endothelial cell function on materials with various surface roughness. J. Biomed. Mater. Res. A 2004, 71 (1), 154 61. 97. Kane, R. S.; Takayama, S.; Ostuni, E.; Ingb er, D. E.; Whitesides, G. M., Patterning proteins and cells using soft lithography. Biomaterials 1999, 20 (23 24), 2363 76. 98. Lahann, J.; Balcells, M.; Rodon, T.; Lee, J.; Choi, I. S.; Jensen, K. F.; Langer, R., rm for Patterning Proteins and Mammalian Cells onto a Broad Range of Materials. Langmuir 2002, 18 (9), 3632 3638.

PAGE 173

160 99. Yamato, M.; Konno, C.; Utsumi, M.; Kikuchi, A.; Okano, T., Thermally responsive polymer grafted surfaces facilitate patterned cell see ding and co culture. Biomaterials 2002, 23 (2), 561 567. 100. Zhang, M.; Desai, T.; Ferrari, M., Proteins and cells on PEG immobilized silicon surfaces. Biomaterials 1998, 19 (10), 953 60. 101. Mrksich, M.; Whitesides, G. M., Using self assembled monolayers to understand the interactions of man made surfaces with proteins and cells. Annu. Rev. Biophys. Biomol. Struct. 1996, 25 , 55 78. 102. Liu, N.; Xie, Q.; Huang, W. M.; Phee, S. J.; Guo, N. Q., Formation of micro protrusion arrays atop shape memory polymer. Journal of Micromechanics and Microengineering 2008, 18 (2). 103. Zhao, Y.; Huang, W. M.; Fu, Y. Q., Formation of micro/nano scale wrinkling patterns atop shape memory polymers. Journal of Micromechanics and Microengineering 2011, 21 (6). 104. Zheng, Z.; Azzaroni, O.; Zhou, F.; Huck, W. T. S., Topography Printing to Locally Control Wettability. J. Am. Chem. Soc. 2006, 128 (24), 7730 7731. 105. Zhao, X. M.; Xia, Y.; Whitesides George, M., Fabrication of three dimensional micro structures: Microtransfer molding. Advanced Materials 1996, 8 (10), 837 840. 106. Shi, J., Micro and nano patterning of polymers. Chinese Science Bulletin 2004, 49 (14). 107. Yang, F.; Wornyo, E.; Gall, K.; King, W. P., Nanoscale indent formation in shape memory polymers using a heated probe tip. Nanotechnology 2007, 18 (28). 108. Wornyo, E.; Gall, K.; Yang, F.; King, W., Nanoindentation of shape memory polymer networks. Polymer 2007, 48 (11), 3213 3225. 109. Hinz, M.; Kleiner, A.; Hild, S.; Marti, O.; Dürig, U.; Gotsmann, B.; Drechsler, U.; Albrecht, T. R.; Vettiger, P., Temperature dependent nano indentation of thin polymer films with the scanning force microscope. European Polymer Journal 2004, 40 (5), 957 964. 110. Lee, H.; Dellatore, S. M.; Miller, W. M.; Messersmith, P. B., Mussel inspired surface chemistry for mul tifunctional coatings. Science 2007, 318 (5849), 426 30. 111. Bilek, M. M.; Bax, D. V.; Kondyurin, A.; Yin, Y.; Nosworthy, N. J.; Fisher, K.; Waterhouse, A.; Weiss, A. S.; dos Remedios, C. G.; McKenzie, D. R., Free radical functionalization of sur faces to prevent adverse responses to biomedical devices. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (35), 14405 10.

PAGE 174

161 112. biomedical applications. Surface and Coatings Techno logy 1998, 98 (1), 1102 1106. 113. Sharif, F.; Hynes, S. O.; Cooney, R.; Howard, L.; McMahon, J.; Daly, K.; Crowley, J.; Barry, F.; O'Brien, T., Gene eluting stents: adenovirus mediated delivery of eNOS to the blood vessel wall accelerates re endot helialization and inhibits restenosis. Mol. Ther. 2008, 16 (10), 1674 80. 114. Lu, A.; Sipehia, R., Antithrombotic and fibrinolytic system of human endothelial cells seeded on PTFE: the effects of surface modification of PTFE by ammonia plasma treatment a nd ECM protein coatings. Biomaterials 2001, 22 (11), 1439 46. 115. Lee, J. H.; Park, J. W.; Lee, H. B., Cell adhesion and growth on polymer surfaces with hydroxyl groups prepared by water vapour plasma treatment. Biomaterials 1991, 12 (5), 443 8. 116. Lee, J. H.; Jung, H. W.; Kang, I. K.; Lee, H. B., Cell behaviour on polymer surfaces with different functional groups. Biomaterials 1994, 15 (9), 705 11. 117. Dekker, A.; Reitsma, K.; Beugeling, T.; Bantjes, A.; Feijen, J.; Kirkpatrick, C. J.; van Aken, W. G., Surface modification of hydrophobic polymers for improvement of endothelial cell surface interactions. Clin. Mater. 1992, 11 (1), 157 162. 118. Tsuda, Y.; Kikuchi, A.; Yamato, M.; Sakurai, Y.; Umezu, M.; Okano, T., Control of cell ad hesion and detachment using temperature and thermoresponsive copolymer grafted culture surfaces. J. Biomed. Mater. Res. A 2004, 69 (1), 70 8. 119. Vladkova, T. G., Surface Engineered Polymeric Biomaterials with Improved Biocontact Properties. Internationa l Journal of Polymer Science 2010, 2010 , 1 22. 120. Egitto, F. D.; Matienzo, L. J., Plasma modification of polymer surfaces for adhesion improvement. IBM Journal of Research and Development 1994, 38 (4), 423 439. 121. Pareta, R. A.; Reising, A. B.; Miller, T.; Storey, D.; Webster, T. J., Increased endothelial cell adhesion on plasma modified nanostructured polymeric and metallic surfaces for vascular stent applications. Biotechnol. Bioeng. 2009, 103 (3), 459 71. 122 . Chu, C. F.; Lu, A.; Liszkowski, M.; Sipehia, R., Enhanced growth of animal and human endothelial cells on biodegradable polymers. Biochim. Biophys. Acta 1999, 1472 (3), 479 85. 123. Feng, Y.; Zhao, H.; Behl, M.; Lendlein, A.; Guo, J.; Yang, D., Gr afting of poly(ethylene glycol) monoacrylates on polycarbonateurethane by UV initiated polymerization for improving hemocompatibility. J. Mater. Sci. Mater. Med. 2013, 24 (1), 61 70.

PAGE 175

162 124. Liang, C. C.; Park, A. Y.; Guan, J. L., In vitro scratch assay: a c onvenient and inexpensive method for analysis of cell migration in vitro. Nat. Protoc. 2007, 2 (2), 329 33. 125. Chaudhury, M. K., Self assembled monolayers on polymer surfaces. Biosens. Bioelectron. 1995, 10 (9), 785 788. 126. Camici, G. G.; Steffel, J.; Akhmedov, A.; Schafer, N.; Baldinger, J.; Schulz, U.; Shojaati, K.; Matter, C. M.; Yang, Z.; Luscher, T. F.; Tanner, F. C., Dimethyl sulfoxide inhibits tissue factor expression, thrombus formation, and vascular smooth m uscle cell activation: a potential treatment strategy for drug eluting stents. Circulation 2006, 114 (14), 1512 21. 127. Sun, T.; Tan, H.; Han, D.; Fu, Q.; Jiang, L., No platelet can adhere -largely improved blood compatibility on nanostructured superh ydrophobic surfaces. Small 2005, 1 (10), 959 63. 128. Tirrell, M.; Kokkoli, E.; Biesalski, M., The role of surface science in bioengineered materials. Surface Science 2002, 500 (1), 61 83. 129. Pakalns, T.; Haverstick, K. L.; Fields, G. B.; McCarthy, J. B.; Mooradian, D. L.; Tirrell, M., Cellular recognition of synthetic peptide amphiphiles in self assembled monolayer films. Biomaterials 1999, 20 (23 24), 2265 79. 130. Zhu, Y.; Gao, C.; He, T.; Liu, X.; Shen, J., Layer by Layer Assembly To Modify Poly(l lactic acid) Surface toward Improving Its Cytocompatibility to Human Endothelial Cells. Biomacromolecules 2003, 4 (2), 446 452. 131. Wang, Y. X.; Robertson, J. L.; Spillman, W. B., Jr.; Claus, R. O., Effects of the chemical structure and the sur face properties of polymeric biomaterials on their biocompatibility. Pharm. Res. 2004, 21 (8), 1362 73. 132. Mao, C.; Qiu, Y.; Sang, H.; Mei, H.; Zhu, A.; Shen, J.; Lin, S., Various approaches to modify biomaterial surfaces for improving hemocompatib ility. Adv. Colloid Interface Sci. 2004, 110 (1 2), 5 17. 133. Xu, H.; Nguyen, K. T.; Brilakis, E. S.; Yang, J.; Fuh, E.; Banerjee, S., Enhanced endothelialization of a new stent polymer through surface enhancement and incorporation of growth factor delivering microparticles. J. Cardiovasc. Transl. Res. 2012, 5 (4 ), 519 27. 134. Mann, B. K.; West, J. L., Cell adhesion peptides alter smooth muscle cell adhesion, proliferation, migration, and matrix protein synthesis on modified surfaces and in polymer scaffolds. J. Biomed. Mater. Res. 2002, 60 (1), 86 93.

PAGE 176

163 135. Larsen, C. C.; Kligm an, F.; Kottke Marchant, K.; Marchant, R. E., The effect of RGD fluorosurfactant polymer modification of ePTFE on endothelial cell adhesion, growth, and function. Biomaterials 2006, 27 (28), 4846 55. 136. Hegemann, D.; Brunner, H.; Oehr, C., Plasma Trea tment of Polymers to Generate Stable, Hydrophobic Surfaces. Plasmas and Polymers 2001, 6 (4), 221 235. 137. McMillan, R.; Meeks, B.; Bensebaa, F.; Deslandes, Y.; Sheardown, H., Cell adhesion peptide modification of gold coated polyurethanes for vascula r endothelial cell adhesion. J. Biomed. Mater. Res. 2001, 54 (2), 272 83. 138. He, W.; Ma, Z.; Yong, T.; Teo, W. E.; Ramakrishna, S., Fabrication of collagen coated biodegradable polymer nanofiber mesh and its potential for endothelial cells growth. Bi omaterials 2005, 26 (36), 7606 15. 139. Balcells, M.; Edelman, E. R., Effect of pre adsorbed proteins on attachment, proliferation, and function of endothelial cells. J. Cell. Physiol. 2002, 191 (2), 155 61. 140. Chen, H.; Yuan, L.; Song, W.; Wu, Z.; Li, D., Biocompatible polymer materials: Role of protein surface interactions. Progress in Polymer Science 2008, 33 (11), 1059 1087. 141. Garner, B.; Hodgson, A. J.; Wallace, G. G.; Underwood, P. A., Human endothelial cell attachment to and growth on polypyrrole heparin is vitronectin dependent. J. Mater. Sci. Mater. Med. 1999, 10 (1), 19 27. 142. Ito, Y., Surface micropatterning to regul ate cell functions. Biomaterials 1999, 20 (23), 2333 2342. 143. Ye, Y. W.; Landau, C.; Willard, J. E.; Rajasubramanian, G.; Moskowitz, A.; Aziz, S.; Meidell, R. S.; Eberhart, R. C., Bioresorbable microporous stents deliver recombinant adenovirus gen e transfer vectors to the arterial wall. Ann. Biomed. Eng. 1998, 26 (3), 398 408. 144. Kim, S.; Kim, J. H.; Jeon, O.; Kwon, I. C.; Park, K., Engineered polymers for advanced drug delivery. Eur. J. Pharm. Biopharm. 2009, 71 (3), 420 30. 145. Wieneke, H .; Dirsch, O.; Sawitowski, T.; Gu, Y. L.; Brauer, H.; Dahmen, U.; Fischer, A.; Wnendt, S.; Erbel, R., Synergistic effects of a novel nanoporous stent coating and tacrolimus on intima proliferation in rabbits. Catheter. Cardiovasc. Interv. 2003, 60 ( 3), 399 407. 146. Costa, J. R., Jr.; Abizaid, A.; Costa, R.; Feres, F.; Tanajura, L. F.; Abizaid, A.; Maldonado, G.; Staico, R.; Siqueira, D.; Sousa, A. G.; Bonan, R.; Sousa, J. E., 1 year results of the hydroxyapatite polymer free sirolimus elu ting stent for the treatment of single de novo coronary lesions: the VESTASYNC I trial. JACC Cardiovasc. Interv. 2009, 2 (5), 422 7.

PAGE 177

164 147. Bhargava, B.; Reddy, N. K.; Karthikeyan, G.; Raju, R.; Mishra, S.; Singh, S.; Waksman, R.; Virmani, R.; Somaraj u, B., A novel paclitaxel eluting porous carbon carbon nanoparticle coated, nonpolymeric cobalt chromium stent: evaluation in a porcine model. Catheter. Cardiovasc. Interv. 2006, 67 (5), 698 702. 148. Wache, H. M.; Tartakowska, D. J.; Hentrich, A.; Wagn er, M. H., Development of a polymer stent with shape memory effect as a drug delivery system. J. Mater. Sci. Mater. Med. 2003, 14 (2), 109 112. 149. Wischke, C.; Behl, M.; Lendlein, A., Drug releasing shape memory polymers the role of morphology, processing effects, and matrix degradation. Expert opinion on drug delivery 2013, 10 (9), 1193 205. 150. Langer, R., Polymer controlled drug delivery systems. Acc. Chem. Res. 1993, 26 (10), 537 542. 151. Pillai, O.; Panchagnula, R., Polymers in drug delivery. Curr. Opin. Chem. Biol. 2001, 5 (4), 447 451. 152. McFarland, C. D.; Mayer, S.; Scotchford, C.; Dalton, B. A.; Steele, J. G.; Downes, S., Attachment of cultured human bone cells to novel polymers. J. Biomed. Mater. Res. 1999, 44 (1), 1 11. 153. Pislaru, S. V.; Harbuzariu, A.; Agarwal, G.; Witt, T.; Gulati, R.; Sandhu, N. P.; Mueske, C.; Kalr a, M.; Simari, R. D.; Sandhu, G. S., Magnetic forces enable rapid endothelialization of synthetic vascular grafts. Circulation 2006, 114 (1 Suppl), I314 8. 154. Stefanini, G. G.; Holmes, D. R., Jr., Drug eluting coronary artery stents. N. Engl. J. Med. 2 013, 368 (3), 254 65. 155. Waterhouse, A.; Wise, S. G.; Yin, Y.; Wu, B.; James, B.; Zreiqat, H.; McKenzie, D. R.; Bao, S.; Weiss, A. S.; Ng, M. K.; Bilek, M. M., In vivo biocompatibility of a plasma activated, coronary stent coating. Biomaterials 2012, 33 (32), 7984 92. 156. Hamid, H.; Coltart, J., 'Miracle stents' -a future without restenosis. McGill J. Med. 2007, 10 (2), 105 11. 157. Wise, S. G.; Waterhouse, A.; Michael, P.; Ng, M. K., Extracellular matrix molecules facilitating vascular bio integration. J Funct Biomater 2012, 3 (3), 569 87. 158. Bhattacharya, V.; McSweeney, P. A.; Shi, Q.; Bruno, B.; Ishida, A.; Nash, R.; Storb, R. F.; Sauvage, L. R.; Hammond, W. P.; Wu, M. H., Enhanced endothelialization and microvessel formation in polyester grafts seeded with CD34(+) bone marrow cells. Blood 2000, 95 (2), 581 5.

PAGE 178

165 159. Herring, M.; Gardner, A.; Glover, J., A single staged technique for seeding vascular grafts with autogenous endothelium. Surgery 1978, 84 (4), 498 504. 160. Sprague, E. A.; Tio, F.; Ahmed, S. H.; Granada, J. F.; Bailey , S. R., Impact of parallel micro engineered stent grooves on endothelial cell migration, proliferation, and function: an in vivo correlation study of the healing response in the coronary swine model. Circ. Cardiovasc. Interv. 2012, 5 (4), 499 507. 161. C amci Unal, G.; Nichol, J. W.; Bae, H.; Tekin, H.; Bischoff, J.; Khademhosseini, A., Hydrogel surfaces to promote attachment and spreading of endothelial progenitor cells. J. Tissue Eng. Regen. Med. 2013, 7 (5), 337 47. 162. Lendlein, A.; Langer, R., B iodegradable, elastic shape memory polymers for potential biomedical applications. Science 2002, 296 (5573), 1673 6. 163. F., E. F.; G., L.; M., F.; D., M., Shape Memory Materials for Biomedical Applications. Advanced Engineering Materials 2002, 4 (3), 91 104. 164. Behl, M.; Lendlein, A., Shape memory polymers. Materials Today 2007, 10 (4), 20 28. 165. Yakacki, C. M.; Ly ons, M. B.; Rech, B.; Gall, K.; Shandas, R., Cytotoxicity and thermomechanical behavior of biomedical shape memory polymer networks post sterilization. Biomed. Mater. 2008, 3 (1), 015010. 166. Pretsch, T.; Ecker, M.; Schildhauer, M.; Maskos, M., Switc hable information carriers based on shape memory polymer. J. Mater. Chem. 2012, 22 (16), 7757 7766. 167. Zhao, Q.; Qi, H. J.; Xie, T., Recent progress in shape memory polymer: New behavior, enabling materials, and mechanistic understanding. Progress in Polymer Science 2015, 49 50 , 79 120. 168. Sun, L.; Huang, W. M.; Ding, Z.; Zhao, Y.; Wang, C. C.; Pur nawali, H.; Tang, C., Stimulus responsive shape memory materials: A review. Materials & Design 2012, 33 , 577 640. 169. Xie, T., Recent advances in polymer shape memory. Polymer 2011, 52 (22), 4985 5000. 170. Pretsch, T., Review on the functional determin ants and durability of shape memory polymers. Polymers 2010, 2 (3), 120 158. 171. Jinlian, H.; Harper, M.; Guoqiang, L.; Samuel, I. I., A review of stimuli responsive polymers for smart textile applications. Smart Materials and Structures 2012, 21 (5), 053001. 172. Mirtschin, N.; Pretsch, T., Programming of One and Two Step Stress Recovery in a Poly(ester urethane). Polymers 2017, 9 (3).

PAGE 179

166 173. Bothe, M.; Pretsch, T., Bidirectional actuation of a thermoplastic polyurethane elastomer. Journal of Materials Chemistry A 2013, 1 (46), 14491 14497. 174. Sun, L.; Huang, W. M.; Lu, H. B.; Wang, C. C.; Zhang, J. L., Shape memory technology for active assembly/disassembly: fundamentals, techniques and example applications. 2014, 34 (1), 78 93. 175. Xiao, X.; Kong, D.; Qiu, X.; Zhang, W.; Liu, Y.; Zhang, S.; Zhang, F.; Hu, Y.; Leng, J., Shape memory polymers with high and low temperature resistant properties. 2015, 5 , 14137. 176. Yanju, L.; Haiyang, D.; Liwu, L.; Jinsong, L., Shape memory polymers and their composites in aerospace applications: a review. Smart Materials and Structures 2014, 23 (2), 023001. 177. Gall, K.; Yakacki, C. M.; Liu, Y.; Shandas, R.; Willett, N.; Ans eth, K. S., Thermomechanics of the shape memory effect in polymers for biomedical applications. J. Biomed. Mater. Res. A 2005, 73 (3), 339 48. 178. Zimkowski, M. M.; Rentschler, M. E.; Schoen, J.; Rech, B. A.; Mandava, N.; Shandas, R., Integrating a n ovel shape memory polymer into surgical meshes decreases placement time in laparoscopic surgery: an in vitro and acute in vivo study. J. Biomed. Mater. Res. A 2013, 101 (9), 2613 20. 179. Zimkowski, M. M.; Rentschler, M. E.; Schoen, J. A.; Mandava, N.; Shandas, R., Biocompatibility and tissue integration of a novel shape memory surgical mesh for ventral hernia: in vivo animal studies. J. Biomed. Mater. Res. B Appl. Biomater. 2014, 102 (5), 1093 100. 180. Decker, E. L.; Frank, B.; Suo, Y.; Garoff, S., Physics of contact angle measurement. Colloids and Surfaces A: Physicochemical and Engineering Aspects 1999, 156 (1), 177 189. 181. Westra, K. L.; Thomson, D. J., Effect of tip shape on surface roughness measurements from atomic force microscopy images o f thin films. 1995, 13 (2), 344 349. 182. Gadelmawla, E. S.; Koura, M. M.; Maksoud, T. M. A.; Elewa, I. M.; Soliman, H. H., Roughness parameters. Journal of Materials Processing Technology 2002, 123 (1), 133 145. 183. O'Brien, J.; Wilson, I.; Orton, T.; Pognan, F., Investigation of the Alamar Blue (resazurin) fluorescent dye for the assessment of mammalian cell cytotoxicity. Eur. J. Biochem. 2000, 267 (17), 5421 6. 184. Yakacki, C. M.; Willis, S.; Luders, C.; Gall, K., Deformation Limits in Shape Memory Polymers. Advanced Engineering Materials 2008, 10 (1 2), 112 119.

PAGE 180

167 185. Nguyen, T. D.; Yakacki, C. M.; Brahmbhatt, P. D.; Chambers, M. L., Modeling the relaxation mechanisms of amorphous shape memory polymers. Adv Mater 2010, 22 (31), 3411 23. 186. Haugh, M. G.; Murphy, C. M.; McKiernan, R. C.; Altenbuchner, C.; O'Brien, F. J., Crosslinking and mechanical properties significantly influence cell attachment, proliferation, and migration within collagen glycosaminoglycan scaffolds. Tissue engineering. Part A 2011, 17 (9 10), 1201 8. 187. Lakhera, N.; Yakacki, C. M.; Nguye n, T. D.; Frick, C. P., Partially constrained recovery of (meth)acrylate shape memory polymer networks. Journal of Applied Polymer Science 2012, 126 (1), 72 82. 188. Yeung, T.; Georges, P. C.; Flanagan, L. A.; Marg, B.; Ortiz, M.; Funaki, M.; Zahir, N.; Ming, W.; Weaver, V.; Janmey, P. A., Effects of substrate stiffness on cell morphology, cytoskeletal structure, and adhesion. Cell Motil. Cytoskeleton 2005, 60 (1), 24 34. 189. Discher, D. E.; Janmey, P.; Wang, Y. L., Tissue cells feel and respond to the stiffness of their substrate. Science 2005, 310 (5751), 1139 43. 190. B?ckstr?m, S.; Benavente, J.; Berg, R. W.; Stibius, K.; Larsen, M. S.; Bohr, H.; H¨¦lix Nielsen, C., Tailoring Properties of Biocompatible PEG DMA Hydrogels with UV Light %J Materials Sciences and Applications. 2012, Vol.03No.06 , 7. 191. Wang, G.; Bai, Y.; Ma, X.; Wang, W.; Yin, Q.; Du, Z., Effects of the PEG length of polycarboxylate based terpolyme rs on their dispersion properties. J. Mol. Liq. 2017, 225 , 333 338. 192. Pich, A.; Berger, S.; Ornatsky, O.; Baranov, V.; Winnik, M. A., The influence of PEG macromonomers on the size and properties of thermosensitive aqueous microgels. Colloid and Polymer Science 2008, 287 (3), 269 275. 193. Zant, E.; Grijpma, D. W., Synth etic Biodegradable Hydrogels with Excellent Mechanical Properties and Good Cell Adhesion Characteristics Obtained by the Combinatorial Synthesis of Photo Cross Linked Networks. Biomacromolecules 2016, 17 (5), 1582 1592. 194. Dichek, D. A.; Neville, R. F. ; Zwiebel, J. A.; Freeman, S. M.; Leon, M. B.; Anderson, W. F., Seeding of intravascular stents with genetically engineered endothelial cells. Circulation 1989, 80 (5), 1347 53. 195. Govindarajan, T.; Shandas, R., Shape Memory Polymers Containing Highe r Acrylate Content Display Increased Endothelial Cell Attachment. Polymers 2017, 9 (11).

PAGE 181

168 196. Tafazzoli Shadpour, M.; Haghighipour, N.; Omidvar, R.; Safshekan, F., Regulation of Endothelial Cell Adherence and Elastic Modulus by Substrate Stiffness AU J alali, Sharareh. Cell Communication & Adhesion 2015, 22 (2 6), 79 89. 197. Delgado Roche, L.; Alfonso Hernández, D., New Alternatives for Atherosclerosis Treatment Based on Immunomodulation. ISRN Vascular Medicine 2012, 2012 , 1 6. 198. Herrington, W.; L acey, B.; Sherliker, P.; Armitage, J.; Lewington, S., Epidemiology of Atherosclerosis and the Potential to Reduce the Global Burden of Atherothrombotic Disease. Circ. Res. 2016, 118 (4), 535 46. 199. Kivimäki, M.; Steptoe, A., Effects of stress on the d evelopment and progression of cardiovascular disease. Nature Reviews Cardiology 2017, 15 , 215. 200. Kounis, N. G.; Koniari, I.; Roumeliotis, A.; Tsigas, G.; Soufras, G.; Grapsas, N.; Davlouros, P.; Hahalis, G., Thrombotic responses to coronary stent s, bioresorbable scaffolds and the Kounis hypersensitivity associated acute thrombotic syndrome. J. Thorac. Dis. 2017, 9 (4), 1155 1164. 201. Avci Adali, M.; Stoll, H.; Wilhelm, N.; Perle, N.; Schlensak, C.; Wendel, H. P., In vivo tissue engineering: mimicry of homing factors for self endothelialization of blood contacting materials. Pathobiology 2013, 80 (4), 176 81. 202. Wong, J. Y.; Lea ch, J. B.; Brown, X. Q., Balance of chemistry, topography, and mechanics at the cell biomaterial interface: Issues and challenges for assessing the role of substrate mechanics on cell response. Surface Science 2004, 570 (1), 119 133. 203. Davis, K. A.; Burke, K. A.; Mather, P. T.; Henderson, J. H., Dynamic cell behavior on shape memory polymer substrates. Biomaterials 2011, 32 (9), 2285 93. 204. Lee, E. M.; Smith, K.; Gall, K.; Boyan, B. D.; Schwartz, Z., Change in surface roughness b y dynamic shape memory acrylate networks enhances osteoblast differentiation. Biomaterials 2016, 110 , 34 44. 205. Li, Y.; Huang, G.; Zhang, X.; Wang, L.; Du, Y.; Lu, T. J.; Xu, F., Engineering cell alignment in vitro. Biotechnol. Adv. 2014, 32 (2), 3 47 65. 206. Liang, C.; Hu, Y.; Wang, H.; Xia, D.; Li, Q.; Zhang, J.; Yang, J.; Li, B.; Li, H.; Han, D.; Dong, M., Biomimetic cardiovascular stents for in vivo re endothelialization. Biomaterials 2016, 103 , 170 182. 207. Wang, D.; Liu, M.; Gu, S.; Zhou, Y.; Li, S., Microtopography Attenuates Endothelial Cell Proliferation by Regulating MicroRNAs %J Journal of Biomaterials and Nanobiotechnology. 2017, Vol.08No.03 , 13.

PAGE 182

169 208. Shangwu, C.; Naoki, K.; Guoping, C., Biomimetic Assembly of Vascular En dothelial Cells and Muscle Cells in Microgrooved Collagen Porous Scaffolds. 2017, 23 (6), 367 376. 209. Moffa, M.; Sciancalepore, A. G.; Passione, L. G.; Pisignano, D., Combined Nano and Micro Scale Topographic Cues for Engineered Vascular Constructs b y Electrospinning and Imprinted Micro Patterns. 2014, 10 (12), 2439 2450. 210. Isenberg, B. C.; Tsuda, Y.; Williams, C.; Shimizu, T.; Yamato, M.; Okano, T.; Wong, J. Y., A thermoresponsive, microtextured substrate for cell sheet engineering with defined structural organization. Biomaterials 2008, 29 (17), 2565 72. 211. Aubin, H .; Nichol, J. W.; Hutson, C. B.; Bae, H.; Sieminski, A. L.; Cropek, D. M.; Akhyari, P.; Khademhosseini, A., Directed 3D cell alignment and elongation in microengineered hydrogels. Biomaterials 2010, 31 (27), 6941 6951. 212. Anderson, D. E.; Hinds, M . T., Endothelial cell micropatterning: methods, effects, and applications. Ann. Biomed. Eng. 2011, 39 (9), 2329 45. 213. Huang, N. F.; Lai, E. S.; Ribeiro, A. J.; Pan, S.; Pruitt, B. L.; Fuller, G. G.; Cooke, J. P., Spatial patterning of endothelium modulates cell morphology, adhesiveness and transcriptional signature. Biomaterials 2013, 34 (12), 2928 37. 214. Gray, B. L.; Lieu, D. K.; Collins, S. D.; Smith, R. L.; Barakat, A. I., Microchannel Platform for the Study of Endothelial Cell Shape and Function. Biomed. Microdevices 2002, 4 (1), 9 16. 215. Xiao, R.; Yakacki, C. M.; Guo, J.; Frick, C. P.; Nguyen, T. D., A predictive parameter for the shape memory behavior of thermoplastic polymers. Journal of Polymer Science Part B: Polymer Physics 20 16, 54 (14), 1405 1414. 216. Chan, B. Q. Y.; Low, Z. W. K.; Heng, S. J. W.; Chan, S. Y.; Owh, C.; Loh, X. J., Recent Advances in Shape Memory Soft Materials for Biomedical Applications. ACS Applied Materials & Interfaces 2016, 8 (16), 10070 10087. 217. Hardy, J. G.; Palma, M.; Wind, S. J.; Biggs, M. J., Responsive Biomaterials: Advances in Materials Based on Shape Memory Polymers. Adv Mater 2016, 28 (27), 5717 24. 218. Hager, M. D.; Bode, S.; Weber, C.; Schubert, U. S., Shape memory polymers: Past, present and future developments. Progress in Polymer Science 2015, 49 50 , 3 33. 219. Young, P. L.; Brackbill, T. P.; Kandlikar, S. G., Estimating Roughness Parameters Resu lting From Various Machining Techniques for Fluid Flow Applications. 2007, (4272X), 827 836.

PAGE 183

170 220. Gengec, N. A.; Gulsuner, H. U.; Erbil, H. Y.; Tekinay, A. B., Selective adsorption of L1210 leukemia cells/human leukocytes on micropatterned surfaces pre pared from polystyrene/polypropylene polyethylene blends. Colloids Surf B Biointerfaces 2014, 113 , 403 11. 221. Kakinoki, S.; Takasaki, K.; Mahara, A.; Ehashi, T.; Hirano, Y.; Yamaoka, T., Direct surface modification of metallic biomaterials via tyros ine oxidation aiming to accelerate the re endothelialization of vascular stents. 2018, 106 (2), 491 499. 222. Pacharra, S.; Ortiz, R.; McMahon, S.; Wang, W.; Viebahn, R.; Salber, J.; Quintana, I., Surface patterning of a novel PEG functionalized poly l lactide polymer to improve its biocompatibility: Applications to bioresorbable vascular stents. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2018, 0 (0). 223. Lu, J.; Rao, M. P.; MacDonald, N. C.; Khang, D.; Webster, T. J., Improved endothelial cell adhesion and proliferation on patterned titanium surfaces with rationally designed, micrometer to nanometer features. Acta Biomater. 2008, 4 (1), 192 201. 224. Yakacki, C. M., Shape Memory and Shape Changing Polymers. Polymer Rev iews 2013, 53 (1), 1 5. 225. Duda, T.; Raghavan, L. V., 3D Metal Printing Technology. IFAC PapersOnLine 2016, 49 (29), 103 110. 226. Sarker, B.; Walter, C.; Pathak, A., Direct Micropatterning of Extracellular Matrix Proteins on Functionalized Polyacryla mide Hydrogels Shows Geometric Regulation of Cell Cell Junctions. ACS Biomaterials Science & Engineering 2018, 4 (7), 2340 2349. 227. Moroni, L.; Lee, L. P., Micropatterned hot embossed polymeric surfaces influence cell proliferation and alignment. J. Biomed. Mater. Res. A 2009, 88 (3), 644 53. 228. Kwok, D. Y.; Neumann, A. W., Contact angle measurement and contact angle interpretation. Adv. Colloid Interface Sci. 1999, 81 (3), 167 249. 229. Wang, X.; Cooper, S., Adhesion of endothelial cells and endothelial progenitor cells on peptide linked polymers in shear flow. Tissue engineering. Part A 2013, 19 (9 10), 1113 1121. 230. Braber, E. T. d.; Ruijter, J. E. d.; Ginsel, L. A.; Recum, A. F. v.; Jansen, J. A., Orientation of ECM protein deposition, fibroblast cytoskeleton, and attachment complex components on silicone microgrooved surfaces. J. Biomed. Mater. Res. 1998, 40 (2), 291 300.

PAGE 184

171 231. Ruder, C.; Sauter, T.; Becker, T.; Kratz, K.; Hiebl, B.; Jung, F.; Lendlein, A.; Zohlnhofer, D., Viability, proliferation and adhesion of smooth muscle cells and human umbilical vein endothelial cells on electrospun polymer scaffolds. Clin. Hemorheol . Microcirc. 2012, 50 (1 2), 101 12. 232. Rickert, D.; Moses, M. A.; Lendlein, A.; Kelch, S.; Franke, R. P., The importance of angiogenesis in the interaction between polymeric biomaterials and surrounding tissue. Clinical hemorheology and microcircula tion. 2003, 28 (3), 175. 233. Palmaz, J. C.; Benson, A.; Sprague, E. A., Influence of Surface Topography on Endothelialization of Intravascular Metallic Material. J. Vasc. Interv. Radiol. 1999, 10 (4), 439 444. 234. Meyle, J.; Gültig, K.; Brich, M.; Hämmerle, H.; Nisch, W., Contact guidance of fibroblasts on biomaterial surfaces. J. Mater. Sci. Mater. Med. 1994, 5 (6), 463 466. 235. Zhou, L.; Lai, Y.; Huang, W.; Huang, S.; Xu, Z.; Chen, J.; Wu, D., Biofunctio nalization of microgroove titanium surfaces with an antimicrobial peptide to enhance their bactericidal activity and cytocompatibility. Colloids and Surfaces B: Biointerfaces 2015, 128 , 552 560. 236. Charest, J. L.; Bryant, L. E.; Garcia, A. J.; King, W . P., Hot embossing for micropatterned cell substrates. Biomaterials 2004, 25 (19), 4767 75. 237. Brasch, M. E.; Passucci, G.; Gulvady, A. C.; Turner, C. E.; Manning, M. L.; Henderson, J. H., Nuclear position relative to the Golgi body and nuclear ori entation are differentially responsive indicators of cell polarized motility. PLoS One 2019, 14 (2), e0211408. 238. Yang, P.; Baker, R. M.; Henderson, J. H.; Mather, P. T., In vitro wrinkle formation via shape memory dynamically aligns adherent cells. S oft Matter 2013, 9 (18). 239. Baker, R. M.; Brasch, M. E.; Manning, M. L.; Henderson, J. H., Automated, contour based tracking and analysis of cell behaviour over long time scales in environments of varying complexity and cell density. Journal of the Ro yal Society, Interface 2014, 11 (97), 20140386 20140386. 240. Rajnicek, A.; McCaig, C., Guidance of CNS growth cones by substratum grooves and ridges: effects of inhibitors of the cytoskeleton, calcium channels and signal transduction pathways. J. Cell Sc i. 1997, 110 ( Pt 23) , 2915 24. 241. Sales, A.; Ende, K.; Diemer, J.; Kyvik, A. R.; Veciana, J.; Ratera, I.; Kemkemer, R.; Spatz, J. P.; Guasch, J., Cell Type Dependent Integrin Distribution in Adhesion and Migration Responses on Protein Coated Mic rogrooved Substrates. ACS Omega 2019, 4 (1), 1791 1800.

PAGE 185

172 242. Cho, J. W.; Kim, J. W.; Jung, Y. C.; Goo, N. S., Electroactive Shape Memory Polyurethane Composites Incorporating Carbon Nanotubes. Macromolecular Rapid Communications 2005, 26 (5), 412 416. 243. Leng, J.; Lan, X.; Liu, Y.; Du, S., Shape memory polymers and their composites: Stimulus methods and applications. Progress in Materials Science 2011, 56 (7), 1077 1135. 244. Arrieta, J. S.; Diani, J.; Gilormini, P., Cyclic and monotonic testing of free and constrained recovery properties of a chemically crosslinked acrylate. 2014, 131 (2). 245. Collins, D. A.; Yakacki, C. M.; Lightbody, D.; Patel, R. R.; Frick, C. P., Sha pe memory behavior of high strength amorphous thermoplastic poly(para phenylene). Journal of Applied Polymer Science 2016, 133 (3), n/a n/a. 246. Nelson, B. A.; King, W. P.; Gall, K., Shape recovery of nanoscale imprints in a ymer. Applied Physics Letters 2005, 86 (10). 247. Ratna, D.; Karger Kocsis, J., Shape memory polymer system of semi interpenetrating network structure composed of crosslinked poly (methyl methacrylate) and poly (ethylene oxide). Polymer 2011, 52 (4), 1063 1070. 248. Choi, J.; Ortega, A. M.; Xiao, R.; Yakacki, C. M.; Nguyen, T. D., Effect of physical aging on the shape memory behavior of amorphous networks. Polymer 2012, 53 (12), 2453 2464. 249. Wang, L.; Yan, Y. B.; Yang, Q. Q.; Yu, J.; Guo, Z. X., Polyamide 66/organoclay nanocomposite fibers prepared by electrospinning. Journal of Materials Science 2012, 47 (4), 1702 1709. 250. Pathak, S.; Gregory Swadener, J.; Kalidindi, S. R.; Courtland, H. W.; Jepsen, K. J.; Goldman, H. M., Measuring the dy namic mechanical response of hydrated mouse bone by nanoindentation. J. Mech. Behav. Biomed. Mater. 2011, 4 (1), 34 43. 251. Wright Charlesworth, D. D.; Miller, D. M.; Miskioglu, I.; King, J. A., Nanoindentation of injection molded PLA and self reinforc ed composite PLA after in vitro conditioning for three months. J. Biomed. Mater. Res. A 2005, 74 (3), 388 96. 252. Yang, F.; Wornyo, E.; Gall, K.; King, W. P., Thermomechanical formation and recovery of nanoindents in a shape memory polymer studied using a heated tip. Scanning 2008, 30 (2), 197 202. 253. Zhang, Y.; Cheng, Y. T.; Grummon, D. S., Shape memory surface s. Applied Physics Letters 2006, 89 (4).

PAGE 186

173 254. Ebara, M.; Uto, K.; Idota, N.; Hoffman, J. M.; Aoyagi, T., Shape memory surface with dynamically tunable nano geometry activated by body heat. Adv Mater 2012, 24 (2), 273 8. 255. Yakacki, C. M.; Ortega, A . M.; Frick, C. P.; Lakhera, N.; Xiao, R.; Nguyen, T. D., Unique Recovery Behavior in Amorphous Shape Memory Polymer Networks. Macromolecular Materials and Engineering 2012, 297 (12), 1160 1166. 256. Yakacki, C. M.; Nguyen, T. D.; Likos, R.; Lamell, R.; Guigou, D.; Gall, K., Impact of shape memory programming on mechanically driven recovery in polymers. Polymer 2011, 52 (21), 4947 4954. 257. Lin Relationships of Photopolymerizable Poly(ethylene glycol) Dimethacrylate Hydrogels. Macromolecules 2005, 38 (7), 2897 2902. 258. Suzuki, T.; Kopia, G.; Hayashi, S.; Bailey, L. R.; Llanos, G.; Wilensky, R.; Klugherz, B. D.; Papandreou, G.; Narayan, P.; Leon, M. B.; Yeung, A. C.; Tio, F.; Tsao, P. S.; Falotico, R.; Carter, A. J., Stent based delivery of sirolimus reduces neointimal formation in a porcine coronary mod el. Circulation 2001, 104 (10), 1188 93. 259. Lehnert, D.; Wehrle Haller, B.; David, C.; Weiland, U.; Ballestrem, C.; Imhof, B. A.; Bastmeyer, M., Cell behaviour on micropatterned substrata: limits of extracellular matrix geometry for spreading and a dhesion. J. Cell Sci. 2004, 117 (Pt 1), 41 52. 260. Uto, K.; Aoyagi, T.; DeForest, C. A.; Hoffman, A. S.; Ebara, M., A Combinational Effect of "Bulk" and "Surface" Shape Memory Transitions on the Regulation of Cell Alignment. Advanced healthcare materials 2017, 6 (9). 261. Wang, J.; Brasch, M. E.; Bak er, R. M.; Tseng, L. F.; Peña, A. N.; Henderson, J. H., Shape memory activation can affect cell seeding of shape memory polymer scaffolds designed for tissue engineering and regenerative medicine. J. Mater. Sci. Mater. Med. 2017, 28 (10), 151. 262. Davi s, K. A.; Luo, X.; Mather, P. T.; Henderson, J. H., Shape memory polymers for active cell culture. Journal of visualized experiments : JoVE 2011, (53). 263. Li, W.; Gong, T.; Chen, H.; Wang, L.; Li, J.; Zhou, S., Tuning surface micropattern feature s using a shape memory functional polymer. RSC Advances 2013, 3 (25), 9865 9874. 264. Ruckenstein, E.; Gourisankar, S. V., A surface energetic criterion of blood compatibility of foreign surfaces. J. Colloid Interface Sci. 1984, 101 (2), 436 451. 265. Go vindarajan, T.; Shandas, R., Microgrooves encourage endothelial cell adhesion and organization on shape memory polymer surfaces. ACS Applied Bio Materials 2019 .

PAGE 187

174 266. Weng, Y.; Chen, J.; Tu, Q.; Li, Q.; Maitz, M. F.; Huang, N., Biomimetic modification o f metallic cardiovascular biomaterials: from function mimicking to endothelialization in vivo. Interface Focus 2012, 2 (3), 356 65. 267. Weng, Y.; Chen, J.; Tu, Q.; Li, Q.; Maitz, M. F.; Huang, N., Biomimetic modification of metallic cardiovascular bi omaterials: from function mimicking to endothelialization in vivo. Interface focus 2012, 2 (3), 356 365. 268. Gupta, M., 3D Printing of Metals. Metals 2017, 7 (10), 403. 269. Liu, D.; Xiang, T.; Gong, T.; Tian, T.; Liu, X.; Zhou, S., Bioinspired 3D Multilayered Shape Memory Scaffold with a Hierarchically Changeable Micropatterned Surface for Efficient Vascularization. ACS Applied Materials & Interfaces 2017, 9 (23), 19725 1 9735. 270. Leclerc, A.; Tremblay, D.; Hadjiantoniou, S.; Bukoreshtliev, N. V.; Rogowski, J. L.; Godin, M.; Pelling, A. E., Three dimensional spatial separation of cells in response to microtopography. Biomaterials 2013, 34 (33), 8097 8104. 271. Nguyen, A. T.; Sathe, S. R.; Yim, E. K. F., From nano to micro: topographical scale and its impact on cell adhesion, morphology and contact guidance. Journal of Physics: Condensed Matter 2016, 28 (18), 183001. 272. De Luca, A. C.; Zink, M.; Weidt, A.; Mayr, S. G.; Ma rkaki, A. E., Effect of microgrooved surface topography on osteoblast maturation and protein adsorption. 2015, 103 (8), 2689 2700. 273. Xu, L. C.; Siedlecki, C. A., Effects of surface wettability and contact time on protein adhesion to biomaterial surface s. Biomaterials 2007, 28 (22), 3273 3283. 274. Clark, P.; Connolly, P.; Curtis, A. S.; Dow, J. A.; Wilkinson, C. D., Topographical control of cell behaviour: II. Multiple grooved substrata. Development 1990, 108 (4), 635 44. 275. Rhodes, J. M.; Simons , M., The extracellular matrix and blood vessel formation: not just a scaffold. J. Cell. Mol. Med. 2007, 11 (2), 176 205. 276. Hartkamp, R.; Biance, A. L.; Fu, L.; Dufrêche, J. F.; Bonhomme, O.; Joly, L., Measuring surface charge: Why experimental cha racterization and molecular modeling should be coupled. Current Opinion in Colloid & Interface Science 2018, 37 , 101 114. 277. Duperret, E. K.; Ridky, T. W., Focal adhesion complex proteins in epidermis and squamous cell carcinoma. Cell cycle (Georgetown, Tex.) 2013, 12 (20), 3272 3285.

PAGE 188

175 278. Hagensen, M. K.; Raarup, M. K.; Mortensen, M. B.; Thim, T.; Nyengaard, J. R.; Falk, E.; Bentzon, J. F., Circulating endothelial progenitor cells do not contribute to regeneration of endothelium after murine arterial injury. Cardiovasc. Res. 2012, 93 ( 2), 223 31. 279. Neffe, A. T.; Hanh, B. D.; Steuer, S.; Lendlein, A., Polymer networks combining controlled drug release, biodegradation, and shape memory capability. Adv Mater 2009, 21 (32 33), 3394 8. 280. Row, S.; Swartz, D. D.; Andreadis, S. T., A nimal models of cardiovascular disease as test beds of bioengineered vascular grafts. Drug Discov. Today Dis. Models 2017, 24 , 37 45. 281. Perkins, L. E. L., Animal models of vascular stenting. Drug Discov. Today Dis. Models 2017, 24 , 31 36. 282. Chamberlain, J.; Wheatcroft, M.; Arnold, N.; Lupton, H.; Crossman, D. C.; Gunn, J.; Francis, S., A novel mouse model of in situ stenting. Cardiovasc. Res. 2010, 85 (1), 38 44. 283. Joner, M.; Nakazawa, G.; Finn, A. V.; Quee, S. C.; Coleman, L.; Acampado, E.; Wilson, P. S.; Skorija, K.; Cheng, Q.; Xu, X.; Gold, H. K.; Kolodgie, F. D.; Virmani, R., Endothelial Cell Recovery Between Comparator Polymer Based Drug Eluting Stents. J. Am. Coll. Cardiol. 2008, 52 (5), 333. 284. George, J. C.; Var ghese, V.; Kovach, R., The Medusa Multi Coil Versus Alternative Vascular Plugs for Iliac Artery Aneurysm Embolization (MVP EMBO) Study. J. Invasive Cardiol. 2016, 28 (1), 23 9.

PAGE 189

176 APPENDIX A FTIR of Original SMP Formulations Figure A1. 1 . FTIR of tBA monomer and PEGDMA crosslinker only. Figure A1. 2 . FTIR of SMPs only. 0 1 2 3 4 5 6 7 400 900 1400 1900 2400 2900 3400 3900 Relative Intensity Wavenumber, cm 1 FTIR Spectra of SMPs 20:80 1000 20:80 750 20:80 550 50:50 1000 50:50 750 50:50 550 80:20 1000 80:20 750 80:20 550 -0.5 0.5 1.5 2.5 3.5 4.5 5.5 6.5 7.5 400 900 1400 1900 2400 2900 3400 3900 Relative Intensity Wavenumber, cm 1 FTIR Spectra of Monomers tBA monomer PEGDMA crosslinker

PAGE 190

177 Figure A1. 3 . FTIR of SMP formulations as well as tBA monomer and PEGDMA crosslinker. Here, we see that the intensity of the alkene peak at approx. 1500 cm 1 is reduced , while the intensity of the alkane group at approx. 3000 cm 1 increases in polymerized SMPs . This likely indicate s that polymerization is successful due to the conversion of the C=C bond to a C C bond . 0 1 2 3 4 5 6 7 400 900 1400 1900 2400 2900 3400 3900 4400 Relative Intensity Wavenumber, cm 1 FTIR Spectra of Monomers and SMPs tBA monomer PEGDMA crosslinker 20:80 1000 20:80 750 20:80 550 50:50 1000 50:50 750 50:50 550 80:20 1000 80:20 750 80:20 550 Ether C= O

PAGE 191

178 APPENDIX B CellEvent Caspase 3/7 Green for Apoptosis of Original Formulations Figure B1. 1 Caspase 3/7 Green shows minimal apoptosis on cell adherent SMPs. 80:20 wt% tBA:PEGDMA550 50:50 wt% tBA:PEGDMA550 20:80 wt% tBA:PEGDMA550

PAGE 192

179 APPENDIX C Extended Live/Dead Study of Select SMPs Day 7 Figure C 1.1 Cell adhesion on select SMPs on Day 7. Live/Dead data indicates that ECs are still present on SMPs up to 7 days after cell introduction.

PAGE 193

180 Figure C 1.2 Cell adhesion on select SMPs up to 7 days after cell introduction. Cell adhesion is present on

PAGE 194

181 APPENDIX D Scanning Electron Microscopy (SEM) of Cell Adherent SMP 80:20 wt% tBA:PEGDMA550 50:50 wt% tBA :PEGDMA550 80:20 wt% tBA:PEGDMA750 50:50 wt% tBA:PEGDMA750 50:50 wt% tBA:PEGDMA1000 80:20 wt% tBA:PEGDMA1000

PAGE 195

182 APPENDIX E CD31 Staining of Microgrooved and Unpatterned SMPs Figure F 1 .1 . CD31, a glycoprotein expressed on the intercellular junctions of mature endothelial cells confirmed the presence of endothelial cells on the SMP surface. Additionally, the expression of CD31 on the cell surface confirmed the connectivity between endotheli al cells, an indication of cell sheet formation.