nanoscale feature composite: an ensemble surface

NANOSCALE FEATURE COMPOSITE: AN ENSEMBLE SURFACE FOR ENHANCING CARDIOVASCULAR IMPLANT ENDOTHELIALIZATION by

Phat L. Tran _____________________ A Dissertation Submitted to the Faculty of the BIOMEDICAL ENGINEERING GRADUATE INTERDISCIPLINARY PROGRAM In Partial Fulfillment of the Requirements For the Degree of DOCTOR OF PHILOSOPHY In the Graduate College THE UNIVERSITY OF ARIZONA

2011

2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Phat L. Tran entitled Nanoscale Feature Composite: Enhancing cardiovascular implant endothelialization using nanowell trapped charged ligand-bearing nanoparticles surfaces and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy ________________________________________________Date: 12/15/2011 Jeong-Yeol Yoon ________________________________________________Date: 12/15/2011 Marvin J. Slepian ________________________________________________Date: 12/15/2011 Mark R. Riley ________________________________________________Date: 12/15/2011 Pak Kin Wong ________________________________________________Date: 12/15/2011 Xiaoyi Wu Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College. I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement. ________________________________________________ Date: 12/15/2011 Dissertation Director: Jeong-Yeol Yoon

3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the author.

SIGNED: Phat Le Tran

4

ACKNOWLEDGEMENTS

I would like to acknowledge the members of my dissertation committee: Dr. Marvin Slepian – thank you for constantly challenging me to become a better engineer, scientist, and researcher. Dr. Mark Riley – thank you for allowing me to use your lab’s equipment; and always willing to discuss data and everything else. Drs. Pak Kin Wong and Xiaoyi Wu – thank you for your valuable inputs and insights, and for being great role models. Finally, I would like to acknowledge my advisor, Dr. Jeong-Yeol Yoon. Thank you for your support and mentorship in my research and marriage. Thank you all for sharing your science with me. This undertaking marathon wouldn’t have been possible without your guidance. I am eternally grateful. To my family - Mom, Dad, and Brother, you are to blame for my Permanent Head Damage (PhD). When we left our war-torn country in search of a better life, you have instilled in me the importance of hard work, perseverance, and the will to imagine the possibilities. You have laid the foundation that allowed me to earn the first doctoral degree in our family. I couldn’t have asked for anything better. Cheers to my friends and colleagues, you have helped me through this process either by counsel or by deed. I thank each of you for your friendship, continued support, and unwavering belief in me. To many awesome undergraduate students that have helped culminate this dissertation, thank you. I hope the “eat free” policy I started in the Biosensorlab will live on. Thank you all for your collaboration. Finally, I would like to acknowledge the past and presence students, faculty, and staffs of BME-GIDP. Thank you all for your help in the past five years and for making every other minute in the program insightful and enjoyable. Once again, I want to express my sincere thanks to all of you and the ones I have forgotten to mention.

5

DEDICATION

To my mom, dad, brother, and my amazing wife Jane,

I am here today because of your hard work and perseverance. You have instilled in me the power and will to imagine the possibilities. This Permanent Head Damage (PhD) is yours as much as it is mine.

6

TABLE OF CONTENTS LIST OF FIGURES ................................................................................................................ 10 LIST OF TABLES .................................................................................................................. 12 ABSTRACT .......................................................................................................................... 13 1. CHAPTER 1 - ENGINEERED NANOSCALE SURFACE TEXTURES FOR CARDIOVASCULAR DISEASE APPLICATIONS ...................................................................... 15 Cardiovascular disease .................................................................................................. 15 Pathology ................................................................................................................... 15 Atherosclerotic coronary artery disease ................................................................... 16 Interventional approaches ............................................................................................ 17 Lifestyle changes........................................................................................................ 18 Thrombolytic treatment ............................................................................................ 18 Revascularization therapy ......................................................................................... 19 Treatment impacts – tissue response to implants........................................................ 21 A need for re-endothelialization of cardiovascular implants ....................................... 23 Engineering nanoscale features for studying cell-nanotopography interaction. ......... 24 Nanotopographical fabrication techniques ............................................................... 28 Nanoimprint lithography (NIL) ........................................................................................... 28 Electrospinning .................................................................................................................. 29 Dip-Pen Nanolithography (DPN) ........................................................................................ 30 Electron beam lithography (EBL) ....................................................................................... 31 Nanocontact printing ......................................................................................................... 32 Optical lithography............................................................................................................. 32 Size dependent self-assembly ............................................................................................ 33

The interaction between nanoscale features and cells ................................................ 34 The impact of nanoscale features on protein adsorption and cell adhesion............ 35 The impact of nanoscale features on cell alignment, migration, and proliferation . 37

7

TABLE OF CONTENTS - Continued The impact of nanoscale features on cell viability and functionality. ....................... 39 Emerging biochemical links between cell adhesivity and immobilized nanofeatures . 40 Overview and future outlooks ...................................................................................... 42 2.

CHAPTER 2 - RESEARCH OBJECTIVES ..................................................................... 44 Project Significance ....................................................................................................... 44 Summary and specific aims of the dissertation ............................................................ 45 I.

The fabrication of nanoscale protein array .................................................... 45 Overall approach ................................................................................................................ 45 Electron beam lithography technique ............................................................................... 45 The Size-Dependent Self-Assembly method ...................................................................... 46

II. The effect of nanofeatured ensemble surface on endothelial cells under minimal shear stress .................................................................................................. 47 Overall approach ................................................................................................................ 47 Bioreactor fabrication ........................................................................................................ 47 Shear-resistance of endothelial cells on test surfaces ....................................................... 48

III. The study of functional intactness or normalcy of endothelial cells adherent to nanofeatured ensemble surfaces ......................................................................... 48 Overall approach ................................................................................................................ 48 Apoptotic assays ................................................................................................................ 49 Cell surface profiling .......................................................................................................... 49 Endocytic mechanism ........................................................................................................ 50

3. CHAPTER 3 - THE FABRICATION OF SIZE-DEPENDENT SELF-ASSEMBLY (SDSA) PROTEIN NANOARRAY FOR FRET DETECTION OF OCTAMER-4 ........................................ 51 Overview ....................................................................................................................... 51 Introduction .................................................................................................................. 52 Materials and Methods ................................................................................................. 57 Substrate preparation and spin coating .................................................................... 57 E-beam lithography and resist development ............................................................ 57

8

TABLE OF CONTENTS - Continued Covalent attachment of antibodies ........................................................................... 58 Development of a protein nanoarray ........................................................................ 59 FRET detection ........................................................................................................... 61 SEM and AFM imaging ............................................................................................... 62 Results and Discussion .................................................................................................. 64 Characterization of protein nanoarrays made with size-dependent self-assembly . 64 FRET detection of Oct4 .............................................................................................. 67 Conclusion ..................................................................................................................... 75 4. CHAPTER 4 – AN ENSEMBLE SURFACE OF NANOSCALE FEATURES FOR ENDOTHELIAL CELL ADHESION ......................................................................................... 76 Overview ....................................................................................................................... 76 Introduction .................................................................................................................. 77 Materials and Methods ................................................................................................. 80 Results and Discussions................................................................................................. 84 Conclusion ..................................................................................................................... 96 5. CHAPTER 5 - FUNCTIONAL INTACTNESS OF ENDOTHELIAL CELLS ADHERENT TO ENSEMBLE NANOTOPOGRAPHICAL SURFACES ................................................................ 98 Overview ....................................................................................................................... 98 Introduction .................................................................................................................. 99 Materials and Methods ............................................................................................... 102 Results and Discussions............................................................................................... 106 Nanoscale featured surface modulates the expression of β1-integrins .................. 107 The effect of nanoscale features on HUVECs migration is αVβ3-integrin specific ... 109 Nanofeatured surface induced Platelets Endothelial Cell Adhesion Molecule–1. . 111 The influence of RGD peptide to endothelial cells viability .................................... 114 The effect of cytochalasin B on cell adhesion and internalization of nanoparticles115 Conclusion ................................................................................................................... 116

9

TABLE OF CONTENTS - Continued 6.

CHAPTER 6 - CONCLUSION ................................................................................... 118

7.

CHAPTER 7 – FUTURE DIRECTIONS ...................................................................... 120

APPENDICES .................................................................................................................... 126 APPENDIX A. ABBREVIATION ......................................................................................... 127 APPENDIX B. COVALENT ATTACHMENT OF PEPTIDES TO COOH-PARTICLES PROTOCOL128 APPENDIX C. PROTOCOL FOR FLOW STUDY ................................................................... 129 APPENDIX D. CYTOSKELETON & FOCAL ADHESION STAINING PROTOCOL .................... 130 APPENDIX E. PROTOCOL FOR FACS ANALYSIS ................................................................ 131 APPENDIX F. REPRINT PERMISSIONS .............................................................................. 132 REFERENCES .................................................................................................................... 136

10

LIST OF FIGURES Figure 1.1. Pathology of atherosclerosis coronary artery disease. ................................. 17 Figure 1.2. Immunofluorescent images of ECs on PDMS consisting of fibronectin islands reveal changes in the substrate coverage, cell density, and focal adhesion densities 14 days of cell culture. Reprinted with permission from ref. [76]. ...................................... 36 Figure 1.3. The extension of microfilament from cardiomyocites at different ridge sizes. Reprinted with permission from ref. [79]. ........................................................................ 38 Figure 1.4. Bone-specific minerals osteopontin and osteocalcin staining of hMSCs after 21 days showing high secretion of mineral proteins under random disorder array of nanopits. Reprinted with permission from ref. [65]. ....................................................... 40 Figure 1.5. Schematic of the biofunctionalized gold nanoparticle substrate, at spacing gradient, in contact with the cell membrane. Reprinted witht permission from ref.[86] 41 Figure 3.1. Schematic illustration of the protein nanoarray fabrication process using electron beam lithography to create micro- and nano-wells, onto which antibodyconjugated beads are self-assembled from the larger to the smaller beads. .................. 56 Figure 3.2. Snapshots of droplet manipulation for SDSA array. ....................................... 59 Figure 3.3. Experimental setup of three-axis droplet manipulator. ................................ 60 Figure 3.4. High density one component protein nanoarray. .......................................... 63 Figure 3.5. Multi-component SDSA protein nanoarray for the detection of Oct4. .......... 69 Figure 3.6. Detection of Oct4 under Fluorolog3 Spectrofluorometer. ............................ 72 Figure 4.1. Nanoparticle array and bioreactor concepts. ................................................ 78 Figure 4.2. Nanowell distribution on p-doped silicon substrata and cell adhesion model. ........................................................................................................................................... 80 Figure 4.3. The effect of different configurations of well patterns and well sizes on HUVEC adhesion without nanoparticles or adhesive ligand. ........................................... 84 Figure 4.4. Saturation of nanoparticles and Schematic of etched wells at different sizes to examine the effect of different charge density on cell adhesion. ............................... 86 Figure 4.5. Comparative effect of adding features to an underlying p-doped silicon surface using individual adhesive elements to enhance HUVEC adhesion. ..................... 88 Figure 4.6. Resistance to detachment with ensemble nanotextured surface................. 90 Figure 4.7. The effect of different shear stress conditions in relation to the orientation of HUVECs. ........................................................................................................................ 92 Figure 4.8. The effect of nanoparticle array on the orientation of HUVECs.................... 96 Figure 5.1. The effect of nanoscale featured surface on the expression of β1-integrins. ......................................................................................................................................... 107

11

LIST OF FIGURES - Continued Figure 5.2. The effect of nanoscale featured surfaces on HUVECs migration is αVβ3integrins specific. ............................................................................................................ 109 Figure 5.3. The effect of nanoscale textured surface on the expression of PECAM-1. .. 111 Figure 5.4. The involvement of RGD peptides nanowells textured surfaces hindered apoptosis. ........................................................................................................................ 114 Figure 5.5. The effect of cytochalasin B on cell adhesion and endocytosis of nanoparticles. ................................................................................................................. 115 Figure 5.6. Fluorescent image of ECs on PMMA surface. .............................................. 117

12

LIST OF TABLES Table 1.1. Summary of nanofabrication techniques for tissue engineering. ................... 27

13

ABSTRACT The establishment and maintenance of functional endothelial cells (ECs) on an engineered surface is central to tissue engineering. As the field advances, the role of cellular mechanisms, particularly the adhesive interaction between the surface of implantable devices and biological systems, becomes more relevant in both research and clinical practice. Knowledge of these interactions can address many fundamental biological questions and would provide key design parameters for medical implants. It has been shown that EC functionality and adhesivity, crucial for the re-endothelialization process, can be induced by nanotopographical modification. Therefore, the goal of this dissertation research was to develop an ensemble surface composing of nanoscale features for the enhancement of endothelial cell adhesion.

Without adhesion,

subsequent vital mechanism involved in cell alignment, elongation or spreading, proliferation, migration, and ECM proteins deposition will not occur. Experiments in support of this goal were broken down into three specific aims. The first aim was to characterize and develop a size-dependent self-assembly (SDSA) nanoarray of Octamer transcription factor 4 as a demonstration to the fabrication of nanoscale feature surface. This nanoparticle array platform was a pilot studied for the second aim, which was the development of an ensemble surface of nanoscale features for endothelial cell adhesion. The third aim was to evaluate and assess EC response to the ensemble surface.

14

Hence, we developed an ensemble surface composed of nanoscale features and adhesive elements for EC adhesivity. By using shear stress as a detachment force, we demonstrated greater cell retention by the ensemble surface than uniform controls. Adhesive interactions and cellular migration through integrin expressions, which are critical to tissue development and wound healing process was also observed. Furthermore, cell viability was relatively sustainable, as indicated by the low expression of apoptotic signaling molecules.

The findings presented within this dissertation

research can be applicable to blood-contact medical implants and possess the potential for future clinical translation.

15

1. CHAPTER 1 - ENGINEERED NANOSCALE SURFACE TEXTURES FOR CARDIOVASCULAR DISEASE APPLICATIONS Cardiovascular disease Cardiovascular disease (CVD) is the leading cause of death in the United States (US) and a worldwide epidemic. In 2007, American Heart Association has estimated that one in every 2.9 deaths in the US is due to CVD. In fact, 82.6 million American adults (more than 1 in 3) have 1 or more types of CVD.

The most common disease is the

atherosclerotic coronary artery disease, which accounts for about 50% of CVD associated deaths. Directly and indirectly, the price tag for CVD is projected to be about 1 trillion US dollars by 2030 [1]. Pathology Cardiovascular disease can lead to a variety of conditions, ranging from radiating discomfort to the arm and sweating to myocardial infarction (heart attack) and death. The primary impact of CVD is the narrowing of coronary arteries that supplies the heart with blood. This may result in insufficient or complete lack of blood flow to the myocardium, which leads to a lack of blood supply and impairing the heart’s ability to function. Regardless of the partial or complete blockage of the arteries, the heart will not be able to supply peripheral tissues and organs with nutrients and oxygen, resulting in mortality.

16

Atherosclerotic coronary artery disease Coronary artery disease (CAD) accounts for more than 50% of CVD’s fatality. The disease is a condition in which plaque builds up inside the arteries; therefore impairing the supply of oxygen-rich blood to the heart (Fig. 1.1). The buildup of plaque in the arteries (a.k.a. atherosclerosis) has long been thought to be the storage of cholesterol; but now the mechanism underlying CAD includes inflammatory components. Basically, arterial endothelial cells are triggered to release adhesion molecules by certain stimuli like injury or damage to the blood vessels and proinflammatory cytokines. These adhesion molecules promote blood leukocytes and platelets adhesion, which later increase extracellular matrix deposition and lesion formation.

As the process

progresses, calcification occurs, lipids accumulate, and lesion expansion, which is known as atherosclerotic plaque. When an area of the plaque is ruptured, it can block smaller arteries and potentially cause angina or heart attack.

17

Figure 1.1. Pathology of atherosclerosis coronary artery disease. A) Show a normal artery with normal blood flow. B) shows an artery with plaque buildup, which impairs blood flow. Source: http://www.nhlbi.nih.gov/health/health-topics/topics/cad/ last accessed 11/29/2011.

Interventional approaches Certain types of heart disease like genetic defects, cannot be prevented. However, other types of CVD can be modified through lifestyle changes, medicines and medical procedures. These interventional approaches are aim to relieve symptoms, reduce the risk of plaque buildup or blood clots formation, and widen or bypass clogged arteries with revascularization and grafting approaches.

18

Lifestyle changes An inactive lifestyle is one of the top risk factors for heart disease. Fortunately, a healthy lifestyle like regular exercise and healthy diets can strengthen the heart and lungs and maintain a healthy weight. Routine physical activity can improve the body’s use of oxygen, reduce excess weight, and lower the risk of contracting diabetes. A healthy diet can also prevent high blood pressure and reduce bad cholesterol (LDL). If smoking is part of your lifestyle, then you need to quit. The chemicals in tobacco harm blood cells and can damage the function of the heart and blood vessels. In fact, smoker has a higher risk of CAD and peripheral arterial disease. Thrombolytic treatment When changes in lifestyle are not enough to reduce the risk of CVD, a wide range of drugs are available to manage and to address variable areas of heart diseases. A patient with heart disease is usually prescribed anticoagulant agents or blood thinners to prevent blood clots and to decrease his/her chance of having a heart attack. Furthermore, the patient can also be asked to take cholesterol-lowering medications to control the cholesterol levels in the blood. Medications are also prescribed to lower high blood pressure. Most importantly, these medications help reduce the heart’s workload and relieve heart disease.

19

Revascularization therapy When changes in lifestyle and thrombolytic therapies are not enough to relieve heart disease, interventional procedures and surgeries are required to immediately restore proper blood flow. Percutaneous catheter intervention, commonly known as angioplasty is the deployment of a balloon, to open the narrowed arteries. The balloon is inflated to compress the plaque against the wall of the artery and restores blood flow. During the procedure, a small mesh tube called a stent can be placed at the site to support the opening of the artery and prevent collapse of the vessel. Traditionally, a stent is made out of bare metal. However, the metal stent is a foreign object to the body; therefore it triggers an inflammatory response and consequently results in platelet formation and reoccurrence of occlusion. To mitigate the problem, stents are coated with anti-inflammatory drugs to prevent the immune response. Unfortunately, one solution leads to another problem. The inflammatory drugs prevent the endothelial cells from forming a monolayer lining to the luminal side of the stent. To mimic the smooth surface of endothelial cells, hydrogel and heparin were paved on the stent to prevent the progression of clot formation. These approaches have shown to be very successful in the lab and clinical trials but have not been approved by the Food and Drug Administration (FDA). The other approach is to utilize a biodegradable stent. Unfortunately, this strategy does not work well in terms of mechanical support. The artery or vessel can collapse when the stent degrades.

20

When angioplasty and stent deployment fails, the doctor has to resort to coronary artery bypass grafting (CABG) using the patient’s saphenous veins or mammary artery. This approach is advantageous because the graft is autologous, which has adequate structural and material properties that are acceptable by the immune system. This procedure involves harvesting a vein from the leg and opening the chest to bypass the occluded artery. The primary goal of this bypass conduit is to provide blood to the heart muscle. A native vessel bypass conduit can remain patent for about 10 years. However, most patients still require revascularization procedures due to graft failure, which can be caused by thrombosis, intimal thickening, or atherosclerosis. Despite these potential limitations, a native graft is still the best choice for bypass conduit. Unfortunately, when the patient runs out of sufficient healthy vessels to serve as replacement, the doctor has to use alternative grafts as bypass conduits. Cadaver’s veins and animal vessels have been explored as alternatives, but they tend to have infection and rejection by the immune system. However, if these veins were to be decellularized and re-seeded with patient’s stem cells, then perhaps the infection and rejection by the immune systems can be negated. However, this latter is still in its infancy stage. The most immediate treatment available is to use an engineered graft as an alternative for bypass conduit. Expanded polytetrafluoroethylene (ePTFE) and polyethylene terephthalate (PET) have been shown to be a successful replacements with a diameter larger than 5 mm. Smaller diameter grafts are more likely to fail due to

21

restenosis.

Furthermore, smaller diameter graft experience low flow rates; thus

exposing greater risk of occlusion.

Treatment impacts – tissue response to implants In spite of all medical advances and interventional treatments, cardiovascular disease is still the leading cause of death in the US. Lifestyle modifications and pharmaceutical drugs can greatly relieve symptoms, prevent further progression of heart disease, and restore blood flow to ischemic regions of the heart. However, they have not yet been successful in re-opening occluded vessels. Implantation of coronary stents or bypassing with grafts is the other resort to treating heart disease but it leads to a cascade of tissue responses. The extent and timing of each response varies based upon the patient and the device’s specific characteristics like size, composition, and the location of implants. Moreover, the overall progression of tissue response seems to be generally consistent [2]. Following the revascularization treatment, an onset of inflammation responses involve platelets and fibrin deposition. Next, the infiltration of chronic inflammation, a dual role of macrophages as inflammatory mediators and wound healing regulators in the foreign body reaction take place. Finally, tissue repair and remodeling occurs, which involves the shrinkage of scar tissues, the deposition of extracellular matrix proteins and the re-endothelialization of the luminal surface [3]. In an effort to reduce tissues inflammations after implantation of stents or grafts, antiproliferative agents like sirolimus and paclitaxel, have been utilized and FDA approved.

22

Sirolimus is an antiproliferative agent that inhibits growth factor and cytokine-induced cell division. Paclitaxel is an anti-mitotic drug that was originally developed as a cancer therapy. The drugs are used to coat a stent known as the drug-eluting stent, and they to prevent cell proliferation; thus they suppresses vessel lesion and consequently in-stent restenosis. The drug-eluting stent has led to improvements in patient care by reducing rates of restenosis and decreasing the prevalence of revascularization procedures. Unfortunately, anti-proliferative drugs are not specific for any one cell type. They are able to suppress smooth muscle cells from proliferation, which leads to vessel lesion, but also affect other cell types at the location of implants; particularly endothelial cells that are crucial in the final remodeling stages of healing. The other approach to reduce tissue responses to implanted devices is to coat the surface of the implants with polymers or biomolecules, thus mimicking the microenvironment of cells and tissues. Hydrogels, heparin, and proteins have been investigated in conjunction with antiproliferative drugs to prevent neointimal hyperplasia and promote endothelial cell growth.

The goal in every case is to

beneficially impact vascular healing and long-term patency of implanted devices. One approach is the method of polymeric endoluminal gel paving. which was first investigated by Slepian MJ in 1996, as wall supports, barriers, and therapeutic biomaterials for local sustained drug delivery [4]. He was able to effectively limit the deposition of cells and proteins and arterial wall thrombogenicity by percutaneously applying hydrogel polymers onto vascular endoluminal surfaces. The other approach is

23

the use of heparin on stents to address thrombosis risks [5]. This anticoagulant has demonstrated safety in human trials, and has shown lower rates of thrombosis. However, no significant long-term benefits were found overall when restenosis of heparin-coated stents was compared to uncoated stents. The final popular approach is protein coatings. The overall goal is to coat the surface with biological agents that resembles extracellular matrix to enhance endothelial cell regeneration and healing. One prevalent biological agent is the integrin-binding RGD sequence. This peptide coating has been investigated in combination with polymer elution technology, and the combo has shown to reduce the extent of neointimal hyperplasia and regenerate layer of endothelial cells by capturing circulating endothelial progenitor cells [6]. Overall, these approaches provide encouraging data for pursuing additional biologic coating options for intravascular therapy. However, the best treatment to tissue response or subsequent thrombosis and restenosis is to re-grow the layer of endothelial cells.

A need for re-endothelialization of cardiovascular implants There is an unmet need for the regeneration of a functional layer of endothelial cells. This re-endothelialization process is the holy grail to long-term implants healing and patency. A confluent monolayer of endothelial cells acts as a selective barrier between circulating blood in the lumen and the surrounding tissue, releases vasoactive agents, and provides a non-thrombogenic surface by releasing heparan sulfate to activate antithrombin III, which inhibits several factors in the coagulation cascade. Without this

24

functional layer of endothelial cells, revascularization treatments like stents and grafts will potentially lead to thrombosis and restenosis.

Engineering nanoscale features for studying cell-nanotopography interaction. A single layer of endothelial cells (ECs) grow directly on the basement membrane, interfacing ECs and the tunica intimal layers. The basement membrane is one particular type of the extracellular matrix (ECM). There are many chemical, physical and geometrical features within the ECM. The matrix is largely composed of adhesive elements like integrins, cell-surface receptors, ligands, glycosaminoglycans, and fibers that make up the microenvironment interfacing cells and tissues. The physiological size of these elements is ranging from micro to nanoscale; the sizes are well capable of mediating the attachment and spreading of cells and regulating signal processing. Therefore, mimicking this microenvironment is crucial to understanding how surface topography can influence cell function and tissue development such that long-term function of medical implant devices can be engineered. Much of what is known today regarding how cells would react to structure and shape of their environment stemmed from a study by Harrison in 1911, where he grew cells on a spider web [7]. Later in 1952, the term contact guidance was used by Weiss and Garber to describe cell alignment to topography. Eight years later, Curtis and Varde demonstrated that cells react to topographical factors much like their environment [8].

25

Since then, researchers have been able to produce accurate micro and nanoscale substances to study cellular behavior. The borrowing of fabrication techniques from the microelectronics industry have facilitated research into cell response to the topography of their environment. The effects of topographical features on cells are becoming more popular including changes in cell adhesion, migration, proliferation, cytoskeletal organization, apoptosis, and gene expression. Cells can be influenced in similar ways by their micro and nanoscale environments, but the type of response is rather unique. At the micro scale, cells are can be guided by features in the same magnitude of size as themselves such as mechanical confinement and robust structures. At the nanoscale, features are far more smaller than elements surrounding cells; and therefore must influence cell behavior through signaling processes and mechanisms. The microenvironment surrounding the cells displays features of pores, fibers, and ridges at the nanometer range. The widespread development of patterning structural cues or deposition of adhesive elements at the nanoscale for studying cell and surface interactions is now becoming more routine. Nanotechnology is referred to a length scale of 1-100 nm in the physical realm [9]. However, considering the cellular structure of biomolecules involve in cell-cell or cell-surface interactions, it is appropriate to extend its length scale to the sub-micrometer range.

The response elicited with

nanotopography is, in fact, greater and the effects exerted on cells may range from

26

subtle to strong. In this chapter, we will discuss the working principle, capability, and limitations

of

nanotopographical

nanotopography interactions.

patterning

techniques

for

studying

cell-

However, this chapter of the dissertation is not to

compare and contrast these techniques but rather summarize some of the fabrication techniques, which led to our size-dependent self-assembly approach for the development of an ensemble surface that composed of nanoscale features suitable for cell-nanotopography adhesivity.

27

Table 1.1. Summary of nanofabrication techniques for tissue engineering. Resolutio n Limits (nm)

Topography and Pattern size

Nanoimprinting Lithography

2

wells/grooves , µm-cm

High aspect ratios and density

mold dependent

Electrospinning

3

nanofibers, cm

Direct patterning and uniform fibers, 3D capable Versatility and work under ambient condition Direct patterning and spatial conformation s high density and inexpensive

disorder patterns

Nanofabricatio n Techniques

Dip-pen 15 nanolithography

islands, µmmm

Electron beam lithography

1-2

wells/grooves , nm-mm

Nanocontact printing

35

line/grooves, µm-cm

Optical lithography

20

lines/grooves, mm-cm

Size-dependent Self-assembly

30

wells, µm-mm

Advantages

inexpensive, fast, and capable of patterning large areas. multiple components

Disadvantage s

Applications Cell adhesion, regenerative medicine Cell adhesion and alignment

Influence by humidity and ink compositions Time consuming and expensive

Cell adhesion and proliferation Cell adhesion, proliferation , and viablity

structural deformation and limited spatial resolution limited by the wavelength and mask

Cell adhesion

expensive to fabricate and time consuming

sensor capable and cell adhesion

Cell adhesion and alignment

28

Nanotopographical fabrication techniques The extracellular matrix at which cells interact often includes structures and textures at nanoscale. For instance, the microenvironment surrounding the cells displays features of pores, fibers, and ridges at the micro and nanometer range.

To mimic these

microenvironment features, many nanofabrication techniques have been invented with one common goal, which is to enhance and to modulate functional cells. Unfortunately, the functional part is still in the early stage of investigation. Fabrication at nanoscale can either be top-down, creating small features from a large block, or bottom-up, building from molecules into a composite feature. Since the birth of fabrication techniques from the microelectronics industry, researchers have been able to develop new techniques for producing accurate nanoscale features on substrates upon which to study cellular behavior (Table 1.1). In light of several excellent recent reviews on nanofabrication techniques [10–13], we will overview the advantages and disadvantages of some of the promising and heavily used fabrication techniques. Some of these techniques are nanoimprinting lithography, electrospinning, dip-pen nanolithography, electron beam lithography, nanocontact printing, optical lithography, and size-dependent self-assembly. Nanoimprint lithography (NIL) In nanoimprint lithography (a.k.a. hot embossing), a mould is pressed into a polymeric substrate at a controllable pressure and temperature, where the polymer is heated

29

above the glass transition point [14,15] to conform the desirable nanostructures. The nanostructures can be replicated in great quantity, resulting in an inexpensive expenditure. The technique also permits high aspect ratio features if the polymeric solution is of low viscosity. Features as small as 5 nm [16,17] and aspect ratios up to 100 nm width and 2 µm depth [18] have been demonstrated. In fact, Hua et al. [19] replicated feature sizes as small as 2 nm using single-walled carbon nanotubes as templates. On the other hand, NIL has two critical disadvantages, mold release and pattern transfer. The high density of nanofeatures dramatically increases the total surface area, leading to strong adhesion of the imprinted polymer to the mold. One solution to such problem is to coat the mold to lower the surface energy. The quality of the pattern transfer is often deteriorated due to the mismatching of thermal expansion between the mould and the polymeric substrate. Electrospinning Electrospinning is another direct patterning technique frequently used in tissue engineering, in which the polymer solution is pushed out of a nozzle under an applied voltage towards a grounded surface.. When the solvent evaporates, the polymer solidifies into disordered nanofibers surface with uniform features, roughness and chemistry. This process can be controlled by the polymer properties (i.e. molecular weight, viscosity, evaporation rate, surface tension, and conductivity) and process

30

parameters such as capillary diameter, flow rate, electric voltage, temperature, and humidity [20]. The resulting fibers can range from 3 nm [21] to several micrometers, thus mimicking the structure of the ECM [22,23]. Dip-Pen Nanolithography (DPN) Dip-pen nanolithography is a sequential technique that relies on the atomic force microscopy (AFM) technology [24] where an AFM tip is used like a fountain pen to selectively deposit ink in a predetermined pattern. Advantages of this technique are high degree of versatility and can be performed under ambient conditions. Although the precise transfer process of the ink is complex and heavily influenced by the ink compositions and humidity, DPN is able to write a pattern as small as 15 nm [25]. In a direct writing fashion, DPN deposits a variety of inks like adhesive ligands and nanoparticles or metal ions on a variety of surfaces [26]. With a massively parallel approach (in a 2-D DPN) to simultaneously align ~55,000 tips, pattern over 1 cm2 areas with 80-100 nm features can be patterned in less than 30 minutes [27]. As a powerful technique for nanoscale patterning, DPN has been used to investigate biological recognition at the single molecule level [26]. For instance, Lee et al. [28] used DPN to directly print nanoarray of αVβ3-integrins to study the molecular interaction between the vitronectin receptor and cells. The other patterning technique that stemmed from DPN is to use an AFM tip to selectively remove molecules at a given location (shaving) so the exposed area can be back filled with other molecules of interests [29].

31

Electron beam lithography (EBL) Electron beam lithography was first developed by the microelectronics industry to generate circuitry in the nanometer regime. It has been rapidly adopted by biological engineers to explore cellular behavior on nanotextured surfaces. In a typical EBL process, a pre-defined AutoCAD design is integrated into a nanopatterns generation systems, which can control the focused electron beam from the SEM to write patterns on an electron-sensitive resist (negative or positive) that was coated on a substrate. The electron beam will break the backbone of photoresists, which can be developed by solvents [30]. EBL writes in interconnected dots of focused electron beam such that a pattern can be spatially controlled by the beam size and the spacing between dots. However, the region of exposure is usually larger than the desired design because of the Monte Carlo affect, which is the electron scattering in the photoresist and the substrate.

To

minimize the scattering effect, one can choose to use higher voltage and thinner spin coated resists (less than 100 nm). Such modifications have allowed Vieu et al. to pattern 3-5 nm lines on PMMA resist [31]. Furthermore, Murray et al. demonstrated highest resolution of 1-2 nm by using metal halide resists and 100 keV electron beam [32]. However, EBL is time consuming for high density patterns.

32

Nanocontact printing Nanocontact printing (nCP) is an extension of the microcontact printing (µCP). Basically, a rigid polymeric stamp with exposing nanofeatures is used to transfer biological chemical or molecules onto a substrata, which is routinely used in studying cell-surface interactions [33]. nCP allows simultaneous patterning over the entire substratum, without relying on the use of an expensive fabrication tools. The stamp or printing mould is usually made out of soft Polydimethylsiloxane (PDMS) polymer.

It is

susceptible to structural collapse or deformation in an unintended way, resulting in inherently limited spatial resolution [34]. Due to the mechanical defects of the soft PDMS, there has been a need for a more rigid printing material that can provide higher aspect ratio but yet flexible enough to print biomolecules over uneven surfaces. Fortunately, Nakarnatsu et al. reported the use of hydrogen silsesquioxane (HSQ) as printing mould to transfer pattern with 35 nm line with onto silicon surface [35]. Optical lithography Optical lithography is an extension of the photolithography. Basically, the fabrication process involves exposing a light-sensitive resist polymer (photoresist) to light through a mask with desired features. The exposed photoresist is then developed in solvents to dissolve the exposed or unexposed regions. In traditional photolithography, fabrication is limited by the wavelength of the UV light, with feature size of a few hundred

33

nanometers to tens of micrometers [36]. In optical lithography, shorter wavelength from an excimer laser is used. The acquired feature sizes can be reduced from around 250 nm (UV light) to 20 nm with 157 nm deep [37]. Although optical lithography has the capability to push the resolution to nanometer scale, it can still be limited by the mask and expensive equipment. Size dependent self-assembly The size-dependent self-assembly (SDSA) technique was first proposed by Tremaine et al. [38] in 2009. It is an integration of EBL to etch features of different size, exposing an electrostatic substratum layer underneath the photoresist for the self-assembly of biomolecules conjugated particles [38–40]. The logic is that particles, varying in sizes, self-assemble to specific locations depending on the diameter matching to the surface nanopatterns.

With the advances in the manufacturing of nanoparticles and the

capability of EBL, SDSA can generate multiple component arrays in nanometer scale. However, molecules or particles can also self-assemble outside of the desired surface pattern. Furthermore SDSA is limited by the size of nanopatterns generated by EBL. In general, fabrication techniques overviewed above allow biological engineers to create nanoscale surfaces with respect to feature size, resolution, and architecture for biological studies. The influence of these nanoscale textured and featured surfaces on cellular behavior will be discussed in the following section.

34

The interaction between nanoscale features and cells The impact of nanofeatured surfaces on cell orientation, migration and cytoskeletal organization was first noted by Harrison in 1911 when he grew cells on a spider web and the cells followed the fibers of the web [7]. Later in 1964, it was proposed that cells react to the topography of their environment [8]. Since then, numerous studies have shown that many cell types react strongly to surface topography [41–43]. Interactions between cells and the engineered surface are central to tissue engineering. Knowledge of these interactions can address many fundamental biological questions and would provide key design parameters for medical implants [44]. It has been shown that cells associated with engineered surface may potentially restore the function of damaged tissues [12,45]. Previous studies have also shown that structural topography have strong effects on cell behavior [13], but less is known about how endothelial cells react to nanoscale textured surface. Endothelial cells are exquisitely responsive to nanofeatures since cells in their natural environment are surrounded by nanostructures. In fact, the basement membrane of the endothelial tissue is comprised of various extracellular matrix components like fibrous collagen, hyaluronic acid, proteoglycans, and fibronectin ranging from 10-300 nm in size [46,47]. These nanometer sized features in the form of pores, ridges, grooves, posts or fibers, when mimicked on biomaterials, promote endothelial cell adhesion [48,49]; making them more progressively useful for applications in tissue engineering [50–52].

Without adhesion, subsequent vital

35

mechanism involved in cell alignment, elongation or spreading, proliferation, migration, and ECM proteins deposition will not occur. Hence, nanotopographical surfaces that stimulate cellular adhesion [53–57], facilitate cell alignment [54,58–61], alter morphology [42,62,63], modulate proliferation [64–66] and migration [60,67–69] and monitor vitality [70,71] have been reported. The impact of nanoscale features on protein adsorption and cell adhesion It is generally thought that surface area increases with increasing topography, thus enhancing protein adsorption. Ample evidence have shown that protein adsorption, indicative of ECM deposition, increases on surface with nanoscale features [13,57,72– 74]. However, the apparent surface that cells can sense is determined by the size and dimensions of nanofeatures. In fact, Carpenter et al. [57] demonstrated that polymeric surface with nanofeatures can greatly influence surface energy, protein adsorption and cell adhesion. They found that vertical dimensions from 5 to 90 nm are best for protein adsorption, while 18 nm is best for cell adhesion. In another study, Kim et al. [75] fabricated nanostructures out of polyethylene glycol for protein adsorption and cell adhesion. They found proteins and cells preferred to adhere on nanostructured PEG surfaces, but the level of adhesion was significantly lower than tissue culture surface. The latter findings could possibly be caused by the high hydrophobicity of PEG nanostructure modified surface.

36

Figure 1.2. Immunofluorescent images of ECs on PDMS consisting of fibronectin islands reveal changes in the substrate coverage, cell density, and focal adhesion densities 14 days of cell culture. Reprinted with permission from ref. [76]. Cell adhesion is the very first mechanistic interaction between the cell and the engineered surface.

Without adhesion, subsequent mechanisms vital to cell

functionality would not exist. There is a common trend that a nanoscale surface that might more accurately mimic the ECM features can greatly enhance cell adhesion. However, various cell types respond differently to nanofeatured surface. For instance, Thomas Webster and colleagues have produced a rough poly (lactic-co-glycolic acid) (PLGA) surface by NaOH treatment and found that the surface enhanced rat aortic SMCs but decreased rat aortic ECs adhesion and proliferation. However, when the surface chemistry was removed by PDMS/PLGA casting methods, they found that SMCs and ECs increased in density on the nanostructured PLGA [77]. They also found the similar effect on nanostructured Titanium and CoCrMO [55,78].

37

In another study, Feinberg et al. have engineerd a high density endothelial cell on PDMS substrate by microcontact printing of fibronectin (Fig. 1.2). They found that microscale patterning of fibronectin into islands of focal adhesion size promotes the formation of EC monolayers similar to the in vivo like cell density and morphology. The impact of nanoscale features on cell alignment, migration, and proliferation There are many chemical, physical and geometrical features within the extracellular matrix. The matrix is largely composed of adhesive elements like integrins, cell-surface receptors, ligands, glycosaminoglycans, and fibers that make up the microenvironment interfacing cells and tissues. The physiological size of these elements is ranging from micro to nanoscale; well capable of mediating the attachment and spreading of cells and regulating signal processing.

38

Figure 1.3. The extension of microfilament from cardiomyocites at different ridge sizes. Reprinted with permission from ref. [79]. A wide range of cell types such as fibroblast, SMCs, and ECs have responded profoundly to the nanofeatured surfaces such as grooves, islands of proteins, pillars, and wells or pits.

Cells can adhere to ridges and reach the floor of the grooves on shallow

nanogratings. Also, cells polarized and elongated along with the ridges and grooves [53,63], but retarded in spreading in the direction of perpendicular patterns, resulting in a smaller cell size and lower proliferation rates [80]. On the other hand, cells cannot reach the groove with an increased depth. Broadening the spacing while keeping the ridge width constant would enable the cells to partially reach the groove (Fig. 1.3). If the grating depth is increased, the cells would be restricted onto the ride surface [81]. Cells on other geometrical cues like nanowells, nanopits, nanopillars seem to have

39

similar trends [43]. However, they appeared to be smaller and rounded with less organized cytoskeletons [63,82,83]. The impact of nanoscale features on cell viability and functionality. Nanofeatured surfaces have been shown to induce the cell phenotype and alter cell functionality. Dalby et al. first demonstrated in 2007 [65] that nanoscale disorder can stimulate hMSCs to produce bone minerals in vitro, without osteogenic supplements (Fig 1.4). They used EBL to generate nanopits at 120 nm diameter, 100 nm depth, and 300 nm center-center spacing. These nanopits had different symmetry and varying disorder. They were able to grow hMSCs on surface of nanopits for up to 28 days. They found that highly ordered nanopatterns produce low to negligible level of cell adhesion. However, hMSCs grown on disordered square array of nanopits with 50 nm displacement (DSQ50) for 21 days showed discrete areas of intense cell aggregation and early modulation of bone-specific minerals expression. By the 28 days, they identified complete mineralization at the discrete sites. Microarray and polymerase chain reaction confirmed DSQ50 was able to induce high upregulation of bone minerals than positive control, which was stimulated with a osteogenic supplement - dexamethasone.

40

Figure 1.4. Bone-specific minerals osteopontin and osteocalcin staining of hMSCs after 21 days showing high secretion of mineral proteins under random disorder array of nanopits. Reprinted with permission from ref. [65]. Most recently, McMurray et al. [84] (also from Dalby’s group) were able to transform nanopits array from the PMMA surface onto a polycaprolactone surface using hot embossing technique. They found that by reducing the level of disorder (DSQ50) to zero as possible (SQ), the resulting nanotopography induced a switch from osteogenic stimulation to a surface conductive to hMSC growth. Most importantly, they were able to retain hMSC phenotype and maintain cell growth over eight weeks. They also concluded that the nanopits surface is non-invasive by employing small interference RNA to repress cell signaling and metabolic pathways.

Emerging biochemical links between cell adhesivity and immobilized nanofeatures Cells interact with the surface through integrins.

Integrins are a family of

transmembrane receptors that bind to the ECM basement membrane through shared

41

RGD domains. Clustering of integrins initiates the assembly of adaptor proteins like vinculin, talin, paxillin, and focal adhesion kinase (FAK), components of the focal adhesion complex. FAK provokes cell movement and cytoskeletal contractility [83,85], which are indicative of alignment and migration. Recently, Lehnert et al. found that the limit of ECM geometry on micropatterned substrata for cell spreading and adhesion is 58 nm [72], thus showing the close relationship between integrin positioning and nanotopography.

Figure 1.5. Schematic of the biofunctionalized gold nanoparticle substrate, at spacing gradient, in contact with the cell membrane. Reprinted witht permission from ref.[86] Cell adhesion, gene expression, and cell viability can be affected by the nanoscale feature surface composed of immobilized nanoparticles. For example, Jang and Nam

42

[87] demonstrated cell adhesion was strongly affected by microarray of nanoparticles. They used a noncontact microarrayer to fabricate a nano-environment for cell-surface interaction studies.

They revealed a dramatic change in the F-action bundle and

network formation on nanoparticle-modified substrata compared to glass surfaces. The availability and distribution of ECM binding sites can dictate cell functionality and behavior. This dictation can be studied by immobilizing gold nanoparticles that were biofunctionalized with cyclic RGDfk on PEG surface at different spacing (Fig.1.5). Arnold et al.[86,88] found the spacing of gold nanoparticles coated with cyclic RGDfk peptide had strong influence on the formation and localization of the focal adhesion on cells. In fact, they found the strongest polarization of cell bodies occurred at the spacing ranging from 60 to 70 nm, which is also congruent with Lehnert et al. [72]. This suggests the universal limit of length scale or spacing for integrins attachment might exist.

Overview and future outlooks Although substantial evidence regarding the relationship between nanoscale feature surface and cellular behavior have been reported, the mechanistic connection in the relationship is not completely understood. Variation in cell types, disparity in nanoscale features and different experimental protocols sometime produce conflicting results. Furthermore, the combination of nanoscale features is unquantifiable and too difficult to optimize cell-surface or substratum interactions. However, some general trends have emerged in the relationship between nanoscale feature surface and cellular response.

43

The first establishes that nanoscale feature surfaces increase surface area and roughness; thus, altering protein adsorption and enhancing cell adhesion. The second confirms that the dimension of the nanoscale structure is a key design parameter for cell adhesion and subsequent alignment, migration, proliferation and differentiation. Finally, the third concerns about the effect of nanofeatured surfaces composed of immobilized nanoparticles on cell adhesion, gene expression, and cell viability. Advances in nanofabrication technologies have stimulated our understanding of cellular behavior on nanofeatured surfaces. However, the majority of the research thus far is still restricted to preliminary work and qualitative analyses. It is undisputable that nanotopographical features enhance cell adhesion and modulate alignment, migration, proliferation and differentiation but there are many questions which remain unsolved. For instance, is the cell health or function similar to in vivo on the engineered nanofeatured surface? On an engineer aspect, how would the cell behave on 3D materials that have been fabricated with nanoscale topographic features?

On a

traditional scientific frontier, what would be the epigenetic effect of cells impacted by the nanofeatured surfaces? Answers to these questions can address many fundamental biological questions and would provide key design parameters for medical implants.

44

2. CHAPTER 2 - RESEARCH OBJECTIVES Project Significance Cardiovascular disease (CVD) is the leading cause of death in United States. In fact, more than one in every three Americans have one or more types of CVDs. The most common type is the atherosclerotic coronary artery disease, which contributes to about 50% of CVD-associated deaths.

Despite advances in thrombolytic, revascularization, and

autogenous or synthetic grafts therapies, thrombosis and restenosis still occur. The occlusion, whether occluded partially or completely, has devastating consequences. The narrowing of the blood vessels obstructs the flow of blood through the circulation system. Patients typically present suddenly with an acute myocardial infarction, as a result of abrupt coronary thrombotic occlusion. If not treated promptly, this may result in death or the development of significant, permanent, incapacitating heart failure. Lack of endothelialization of the implant (stent) endoluminal surface is the central pathophysiologic mechanism.

Therefore, we are proposing a novel method of

enhancing endothelialization, of which may potentially be beneficial to the safety and effectiveness of stents and broaden cardiovascular implant device applicability, which are currently limited by surface thrombosis.

45

Summary and specific aims of the dissertation I. The fabrication of nanoscale protein array Overall approach A functionalized surface with rough topography and a certain order of patterning are features that the protein nanoarray can provide for cell adhesion. In this study, we utilize Electron Beam Lithography (EBL) to etch wells that are relative to the particle size and the size dependent self-assembly (SDSA) method to deposit protein conjugated particles.

Particles are carboxylated and negatively charged so that they can be

attracted toward and anchored to the etched wells, which carry positive charge. As a proof of concept, we conjugated Octamer transcription factor 4 (Oct4), which is responsible for the differentiation of stem cells to other cell lineages, and mouse antibody (as positive control, mIgG) to particles of 80 nm and 180 nm, respectively. This demonstrate the fabrication of multiple component nanoarray and subsequently follow the detection of the antibodies’ counterparts through fluorescence resonance energy transfer mechanism using spectrofluorometer and confocal through spectral image mode. Electron beam lithography technique Electron beam lithography (EBL) is a nanoscale lithography technique that can directly write on a negative or positive photoresist. It is an integration system composing of a scanning electron microscope (SEM) and nanopatterns generation system (NPGS). The

46

NPGS system can upload an autocad design and control the electron beam from the SEM, which can be assumed as heat, to etch or break the polymer. By controlling the dosage of the beam, one can control the magnitude and the size of the pattern. Furthermore, EBL writes in interconnected dots; therefore the pattern can also be controlled by manipulating the spacing between dots. Because the beam is made out of electrons, it can pattern nanoscale features with high resolution, which has been shown to be favorable toward cell and protein adhesion. The Size-Dependent Self-Assembly method The concept of size-dependent self-assembly is to generate multiple component arrays based on size and electrostatic properties. Basically, negatively charged particles will be added sequentially on the pattern of wells using a droplet vibrational manipulator such that larger particles can saturate the larger wells leaving smaller wells for smaller particles. The method is robust and provides a rough surface topography at nanoscale feature suitable for enhancing cell adhesion, particularly when the particle is functionalized with bioadhesive ligands. As a proof of concept, we used this SDSA method to illustrate the detection of proteins in soluble solution by conjugating protein counterparts to the particle.

47

II. The effect of nanofeatured ensemble surface on endothelial cells under minimal shear stress Overall approach To study the effect of nanofeatured ensemble surface on endothelial cell adhesion, we must first investigate the effect of the surface topography (wells) at different spatial configurations (x-y spacing) and well sizes in comparison with the uniform hydrophilic surface p-doped silicon ((+)S) and the bioinert PMMA (P) surface. Subsequently, we examined the effect of nanoparticles alone and bioadhesive ligands on the uniform surfaces of ((-)beads on (+)S) and ((-)beads on P) on cell adhesion. We also look at the effect of nanowells pattern alone (wells), nanoparticles in wells ((-)beads in wells), bioadhesive ligands in wells (RGD in wells), and the ligand conjugated nanoparticles (as the ensemble surface) in wells (RGD-(-)beads in wells) on endothelial cell adhesion. Finally, we used shear stress as a detachment force to study the responsiveness of endothelial cells on the ensemble surface. These experiments can potentially shed light to a greater surface endothelialization process, which is translatable to circulatory implant device. Bioreactor fabrication The effect of ensemble surface on the attachment of endothelial cells is tested through a bioreactor; thus mimicking the physiological flow condition. An acrylic biochamber is fabricated using a mill machine, allowing placement of a test coupon (surface + cells).

48

The biochamber has a defined flow path of 1 x 1 x 5 mm (HxWxL), allowing gas exchange, overflowing perfusion and sealing for placement in an incubator.

The

biochamber is connected to a pulsatile pump and a media reservoir to mimic blood flow. Shear-resistance of endothelial cells on test surfaces Endothelial cells grown in M199 w 15% FCS + endothelial growth factor and heparin (media) will be seeded on test coupons (5 x 105 cells/ml plating density) and incubated overnight.

EC-seeded test coupons will then be placed in the sterile bioreactor,

connected to a sterile loop (tubing, roller pump and media reservoir) and exposed to overflowing media at designated shear rates (1 and 1.5 dynes/cm2) in an incubator 37oC with a 95/5 O2/CO2 humidified environment. Identical coupons incubated under static conditions will serve as controls. At 4 and 72 hours, test coupons will be removed and the number of adherent EC will be determined. The number of adherent cells under flow conditions will be compared to number under static conditions at each time point and % adhesion will be calculated. Comparisons will be made via t-tests. III. The study of functional intactness or normalcy of endothelial cells adherent to nanofeatured ensemble surfaces Overall approach Functional intactness of endothelial cell grown on test surfaces will be examined hierarchically in terms of induction of apoptosis, expression of cell surface markers and the endocytosis of nanoparticles. Each of these states will be determined via immunohistochemical methods

49

utilizing either flow cytometry or direct staining methods. Endothelial cells grown on test coupon surfaces as in Aim 2 will be incubated for 4, 24, and 72 hrs. ECs will be harvested from coupons (trypsin), washed and resuspended in PBS (Ca/Mg free). Cells will then be exposed to designated test antibody, washed, paraformaldehyde fixed and analyzed via flow cytometry. Mean fluorescence intensity will be reported.

Apoptotic assays Apoptosis is a normal physiologic process called program cell death, which occurs during the stage of biological development as well as in maintaining tissue homeostasis. The earliest features of apoptosis is the loss of plasma membrane, thus exposing the membrane phospholipid phosphatidylserine (PS). Phycoerythrine conjugated annexin V has a high affinity for PS, which can be used to indicate early stages of apoptosis. The latest stages of apoptosis or necrosis is the loss of membrane integrity, thereby allowing 7-Amino Actinomycin D to permeate the nucleus and stain DNA. The apoptotic process was measured over time (4, 24, and 72 hrs) to monitor the normalcy of endothelial cells on different coupons (nanoscale feature surfaces + cells).

Cell surface profiling The behavior of endothelial cells adherent to different coupons is reflective to the expression of β1, αVβ3, αVβ5, CD31 (PECAM-1), and CD138 (Syndecan-1) cellular markers. These surface proteins will be used to evaluate the migration and adhesion of ECs on the ensemble

surface over 3 time points (4, 24, and 72 hrs). Integrin β1 is associated with the fibronectin receptor, which is primarily responsible for focal adhesions.

Integrin αVβ3 is an integral

membrane protein that binds to vitronectin and is usually accountable for migration. Integrin

50

αVβ5 is known to participate in cell surface mediated signaling. CD31 is the platelets endothelial cells adhesion molecule 1 (PECAM-1). This glycoprotein makes up a large portion of endothelial cell intercellular junctions for vascular cell adhesion and is a signaling molecule for neutrophil recruitment. CD138 is also known as Syndecan-1 or heparin sulfate proteoglycans which has antithrombotic properties.

Endocytic mechanism Cytochalasin B will be used to determine the properties or the effect of internalizing nanoparticles to ECs. Endothelial cells on test surfaces will be fixed with paraformaldehyde, washed and stained with a primary antibody. Secondary antibody i.e. mIgG-FITC and PhalloidinTRITC will then be added.

Preparations will be washed, mounted and examined via

epifluoresence microscopy. The number of endothelial cells before and after treatment of Cytochalasin B will be reported.

51

3. CHAPTER 3 - THE FABRICATION OF SIZE-DEPENDENT SELFASSEMBLY (SDSA) PROTEIN NANOARRAY FOR FRET DETECTION OF OCTAMER-4 This chapter have been published in the journal of Analytical and Bioanalytical Chemistry 398(2): 759-768, 2010. (Permission for reprint – Appendix F)

Overview An alternative approach for fabricating a protein array at nanoscale is suggested with a capability of characterization and/or localization of multiple components on a nanoarray.

Fluorescent micro- and nanobeads each conjugated with different

antibodies are assembled by size-dependent self-assembly (SDSA) onto nanometer wells that were created on a polymethyl methacrylate (PMMA) substrate by electron beam lithography (EBL). Antibody-conjugated beads of different diameters are added serially and electrostatically attached to corresponding wells through electrostatic attraction between the carboxylic groups of the beads and exposed p-doped silicon substrate underneath the PMMA layer. This SDSA method is enhanced by vibrated-wire-guide manipulation of droplets on the PMMA surface containing nanometer wells. Saturation rates of antibody-conjugated beads to the nanometer patterns are up to 97% under one component and 56% under two components nanoarrays. High density arrays (up to 40,000 wells) could be fabricated, which can also be multi-component. Target detection utilizes fluorescence resonance energy transfer (FRET) from fluorescent beads to

52

fluorescent secondary antibody of Octamer-4 (Oct4), which eliminates the need for multiple steps of rinsing. The 100 nm green beads are covalently conjugated with antiOct4 to capture Oct4 peptides; where the secondary anti-Oct4 (anti-gIgG) tagged with phycoerythrin via F(ab)2 fragment is then added to function as an indicator of Oct4 detection.

FRET signals are detected through confocal microscopes, and further

confirmed by Fluorolog3 spectrofluorometer. The success rate of detecting Oct4 is about one in every three beads (about 32% and 34% of saturated beads exhibits FRET under one and two component nanoarrays, respectively).

Introduction DNA microarray technologies have successfully been implemented in identifying specific genomic information from living organisms and have become routine practice these days. As microarray analysis is currently being applied to disease state monitoring, drug screening processes, proteomics and cell research, and clinical diagnosis [89–91], more emphasis is given to protein nanoarray technologies. T he ability to develop protein nanoarray with well-defined feature size, shape, and spatial configuration is crucial in enhancing signal-to-noise ratio such that signal can be identified individually. Latest techniques and approaches to the fabrication of protein nanoarrays at defined positions and spacing have involved: ink-jet and pipette deposition [92], dip-pen nanolithography (DPN) [93], electron beam lithography [30], and nanocontact printing [94]. However,

53

developing nanoarrays of multiple biomolecules and the ability to retain their full structure and function are still an obstacle to overcome. Proteins are relatively unstable when immobilized on a surface of an array than proteins that are dissolved in buffer. This difficulty is usually seen under sandwich immunoassay, where multiple steps of rinsing are involved such that protein’s structure and functionality is altered; thus reducing the ability to detect targets of interest. Therefore, an in situ detection method with limiting rinsing steps are highly warranted. Solving such obstacles will benefit the understanding of biomolecular interactions and dramatically increase the detection limit, preferably at the level of single molecule. While DNA sequencing studies provide a better understanding of the genome architecture and gene regulation, protein arrays provide a comprehensive knowledge of genes at a functional level. Proteins are involved in a wide range of biological functions such as catalyzing reactions in living organisms, translating information in cells, regulating biochemical activities, amplifying chemical products, providing mechanical supports, and most importantly, mediating biological defense mechanism [95]. The occurrence of diseases and various cancers take place at the proteomic level, where the expression and distribution of proteins are altered [96]. In response to these diseases, proteins commonly known as biomarkers are secreted at extremely low levels, especially at the early on-set of disease development. This is a serious problem because protein cannot be amplified like DNA. Therefore, methods with extreme sensitivity and high specificity are highly desired for clinical diagnosis and therapeutic applications.

54

Early detection of protein biomarkers is not only sought after for untreatable diseases or cancers detection but also for stem cell research. Transcriptional proteins involved in differentiating stem cells, like Octamer-4 (Oct4), are present at extremely low volume and in the early stage of cell development [97]. A protein array at nanoscale, which requires much less sample volume and potentially offers single molecular detection of important targets, is expected to play an important role in studying stem cells [98]. The rationale is that controlling the fate of cell differentiation would be critical to stem cell research and offer the possibility of curing untreatable diseases. Unfortunately, the development of protein nanoarrays and methods for detecting targets at the single molecule level are still in their infancy. Among the advances in nanopatterning (as mentioned above), DPN, which uses atomic force microscope equipment to deposit localized add-on materials with the cantilevers on a substrate is considered as a potential tool for patterning biomolecules on the nanoscale. However, only a limited number of types of proteins, typically two or three, have been patterned in nanometer scale [93]. Furthermore, high-density nanoarrays can also contribute to the success rate of studying proteins or detecting important targets. Recently a 55,000 pen array has been demonstrated for patterning large areas with 80 million dots [27]. However, efforts in multiple-component patterning showed significant complexity in both equipment modification and the process.

55

Besides the fabrication of a protein nanoarray, a method for single molecule biorecognition is also warranted. Until now, targets can be detected through electrical properties or optical modalities. While electrical detection may warrant recognition of a single molecule, its ability to study molecular interactions or structural integrities of a protein is limited. On contrary, Kang et al. has demonstrated the use of dual-color total internal reflection fluorescence microscopy (TRIFM) for detection of single molecules of DNA hybridization [99]. However, this deals with DNA not protein, but the concept may be geared toward proteomic study.

Furthermore, Huang and Chen have recently

illustrated the detection of single molecule by incorporating electrical properties with fluorescence detection. They have applied an electrical potential onto nanowires that were grafted with aptamers to modulate the fluorescence of fluorophores on the target complex [100]. The technique is novel but lacks the ability to study conformational dynamics and interactions of proteins.

Thus, characterization, visualization, and

detection of important proteins or their dynamic interactions constitute a critical step in biorecognition applications and studying of cell mechanisms.

56

Figure 3.1. Schematic illustration of the protein nanoarray fabrication process using electron beam lithography to create micro- and nano-wells, onto which antibodyconjugated beads are self-assembled from the larger to the smaller beads. Beads-COOH = carboxylated micro- and nano-beads. BSA = bovine serum albumin (to passivate the beads). Silicon wafer is p-doped with spin-coated layer of polymethyl methacrylate (PMMA). We have previously reported the assembly of particles on electron beam lithographic patterns [38,101]. In this study, we provide a new approach to the fabrication of protein nanoarrays which involves the vibrated-wire-guide droplet manipulation system, electron beam lithography (EBL) and size-dependent self -assembly (SDSA) of proteinconjugated beads, and fluorescence resonance energy transfer (FRET) to detect Octamer-4 (Oct4) transcriptional factor. The concept of SDSA is that the larger beads cannot sit in smaller wells while they occupy the larger wells, and smaller beads later fill in the smaller wells to generate a multi-component protein nanoarray. This method allows control of the location of each type of bead and thus control of the location of multiple proteins. For clarity, a schematic representation of the fabrication process is

57

shown in Fig. 3.1. Ultimately, this new concept of protein nanoarray has the potential to deliver efficient, near real-time, highly sensitive and selective analyses of transcriptional proteins in an effort to control stem cell lineages.

Materials and Methods Substrate preparation and spin coating A p-doped silicon wafer (p-type boron, 450-648 μm thick and 4-75 Ω-cm-1, Exsil, Inc., Prescott, AZ, USA), containing a positive surface charge, was cut into 1 cm 2 chips. Each chip was washed with acetone (Sigma-Aldrich, St. Louis, MO, USA) and isopropyl alcohol (IPA; Honeywell, Chandler, AZ, USA) and subsequently spin-coated with a photoresist, which was made by a 1:1 and 2:3 dilutions of 950 PMMA [poly(methyl methacrylate); Microchem, Newton, MA, USA] with C4 thinner (Microchem), resulting in about 100 and 80 nm layers of PMMA (measured by a profilometer) respectively. The resist was applied to the chip at 500 rpm for 5 s followed by 4000 rpm for 40 s. The chip was then placed on a hot plate at 180°C for 1 min to remove any excess residues and to facilitate resist adhesion.

The chip was then cooled to room temperature before e-beam

patterning. E-beam lithography and resist development A FEI Inspec S scanning electron microscope (SEM; FEI Company, Hillsboro, OR, USA) equipped with JC Nabity nanometer pattern generation system (NPGS; JC Nabity,

58

Bozeman, MT, USA) was used to etch patterns into the PMMA. Desirable patterns were obtained by the DesignCAD software. The pattern was etched with high voltage of 30 keV at about 10 pA with varying spot size from 1.5 to 3. Each line of pattern was separated by 1 µm and each well was separated by 1 µm center to center considering the limitation of the fluorescence and confocal microscopes’ resolution. After patterning, the etched array was developed with 1:3 methyl isobutyl ketone / isopropyl alcohol (MIBK/IPA) (Michrochem) developer for 60 s, then 30 s with IPA (Honeywell). Finally, the etched array was washed with deionized water and dried with nitrogen gas. Covalent attachment of antibodies Carboxylated, fluorescent polystyrene beads were covalently conjugated with antibodies of interest. The 180 nm glacial blue beads (excitation = 360 nm, emission = 425 nm, parking area = 17.5 Å2 per carboxyl group; catalog number FC02F from Bangs Laboratories, Fishers, IN, USA) were covalently conjugated with mouse immunoglobulin G (mIgG; catalog number I5381; Sigma-Aldrich, St. Louis, MO, USA). The 100 nm green beads (excitation = 458 nm, emission = 510 nm, carboxylated but unavailable parking area; catalog number F8803 from Molecular Probes, Eugene, OR, USA) were covalently conjugated with anti-Oct4 (goat polyclonal antibody; catalog number ab52014; Abcam, Cambridge, MA, USA). The full protocol of covalent antibody conjugation can be found from Bangs Laboratories or Molecular Probes. Basically, different sizes of beads were

59

resuspended in 50 mM 2-(N-morpholino)ethanesulfonic acid (MES; Sigma-Aldrich, St. Louis, MO, USA) buffer at pH 6.0 and linked with carbodiimide at room temperature for 15 minutes. The mixture was centrifuged at 14,000 rpm for 15 minutes then decanted supernatant and resuspended in 50 mM phosphate buffered saline (PBS; Sigma-Aldrich, St. Louis, MO, USA) at pH 7.4. Mouse IgG or anti-Oct4 was then covalently conjugated to the 180 nm glacial blue or 100 nm green beads, respectively and slowly rocked overnight at 4°C to facilitate proper orientation of antibodies. The solution was washed with PBS-BN (10 mM PBS at pH 7.4, 1% BSA, 0.05% sodium azide) and rotated with 40 mM hydroxylamine for 30 minutes at room temperature to assist in packaging antibodies. Finally, conjugated particles were washed and stored in PBS-BN. Development of a protein nanoarray

Figure 3.2. Snapshots of droplet manipulation for SDSA array. A) A droplet of beads is suspended and (B) a vibrating metal wire transports the droplet over the array. (C) Another droplet of beads is immobilized over the array.

60

A 0.5 µl of about 0.025% (w/v) solid content of non-antibody-conjugated and antibodyconjugated beads in PBS-BN were serially pipetted onto the protein chip by pipette tips. These droplets of bead suspension were transported to the array area (where the nanometer wells were patterned) using a vibrated metal wire (o.d. = 0.5 mm) (Fig. 3.2).

Figure 3.3. Experimental setup of three-axis droplet manipulator. The vibrator is mounted on a rapid-prototyped plastic spring such that the metal wire can vibrate to assist in generating high density arrays. The metal wire was connected to a microcontroller (Arduino Duemilanove, SparkFun Electronics, Boulder, Colorado) interfaced with a USB port that can be programmed to control the three-axis of the droplet manipulator. A Nintendo game pad was attached to the microcontroller so that x-, y- and z-movements of a metal wire could be made possible from the experimenter’s input. Details can be found elsewhere [102]. The

61

water contact angle of PMMA is ca. 70°. This is not enough to make “wire-guide” droplet manipulations [102]. Hence, the metal wire was vibrated to make necessary xand y-movements. This vibration provided sufficient energy for micro- and nanobeads to assemble into nanometer wells. Droplets travelled across the patterned area up to three times, followed by removal from that area. Additional droplet of 10 mM PBS was transported and moved over the same area to remove weakly bound particulates. Fig. 3.2 shows the snapshots of these droplet movements, and Fig. 3.3 shows the experimental setup of three-axis droplet manipulator. A complete movie is available for this “wire-guide” droplet manipulator system as a supplemental material. FRET detection Target biomolecules are Oct4 peptide (catalog number ab20650; Abcam) and anti-mIgGFITC (FITC = fluorescein isothiocyanate; catalog number F9006; Sigma). As yellow-dyeconjugated Oct4 is not available, sandwich immunoassay is attempted using yellow-dyeconjugated secondary antibodies. For the detection of Oct4 peptide, 0.5 µl of 0.5 mg/ml of anti-Oct4 (secondary) donkey polyclonal anti-goat IgG F(ab)2 fragment tagged with phycoerythrin (anti-gIgG-PE; catalog number 7004; Abcam) were applied over the array for about 3 mins, which contains 180 nm blue, mIgG-conjugated beads and/or 100 nm green, anti-Oct4-conjugated beads. PE is excited with green color and emits yellow (575 nm). The sequence of fluorescent energy transfer is: blue light source → green beads → yellow PE that is captured by the target. The chip was then rinsed with cold PBS buffer pH 7.4 and mounted in VectaShield mounting medium (catalog number H-1000; Vector

62

Laboratories, Burlingame, CA, USA). The FRET signal is detected by using the spectral analysis available from the C1 Si Laser Scanning Confocal Fluorescence Microscope system. Intensity signals, recorded before and after the presence of targets, were then compared and analyzed using MATLAB (The MathWorks, Inc., Natick, MA, USA). To confirm FRET from the above systems, Fluorolog3 spectrofluorometer (HORIBA Jobin Yvon Inc., Edison New Jersey, NJ, USA) was used at an increment and integration time of 1 second and bandpass of 1 nm for excitation and 1.5 nm for emission. Anti-mIgG-FITC was also used as a negative control, where the sequence of fluorescent energy transfer is: UV light source → blue beads → green FITC in target (FITC is excited with blue color and emits green, 520 nm).

Intensity signals were analyzed in the same manner

described above (using MATLAB). SEM and AFM imaging The Veeco Dimension 3100 atomic force microscope (AFM) was used to check the etched patterns. It was operated in tapping mode with integral gain of about 0.2 and amplitude of about 1.2 V. The SEM used for e-beam lithography (FEI Inspec S SEM) was also used to image the protein nanoarrays. Due to the sensitivity of PMMA to the electron beam, the Si chips needed to be sputter-coated with gold approximately 5-8 nm thick. The metal coating allowed the sample to be more conductive and provided a protective layer that allowed longer viewing time as well as enhanced signaling of the samples before the PMMA deteriorated.

63

Figure 3.4. High density one component protein nanoarray. A) The AFM image of the patterned wells capable of making nanoscale protein array. Each well is separated by 500 nm whereas each line of well is separated by 1 µM. The smallest well is about 100 nm. B) high density nanoarray containing 100 nm beads. Each bead is separated by 1 µM. C) Spectral image of 100 nm green beads. The image is digitally zoomed by the confocal microscope illustrating some FRET signals. The table below the image summarizes the number of beads involved. D) Raw spectrums collected from 3 spots in Fig. 3.4C. The table shows that spot with FRET has higher ratio of intensity (ROI) than spot without FRET (p