Lymphocyte and Macrophage Interactions in the Response

Home , Giant cell

LYMPHOCYTE AND MACROPHAGE INTERACTIONS IN THE RESPONSE

TO BIOMATERIAL SURFACES

DAVID T. CHANG

Submitted in partial fulfillment of the requirements

For the degree of Doctor of Philosophy

Dissertation Advisor: Dr. James M. Anderson

Department of Biomedical Engineering

CASE WESTERN RESERVE UNIVERSITY

August, 2008 CASE WESTERN RESERVE UNIVERSITY

SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

______

candidate for the ______degree *.

(signed)______(chair of the committee)

______

(date) ______

*We also certify that written approval has been obtained for any proprietary material contained therein.

List of Tables ...... iii

List of Figures...... v

Acknowledgements...... viii

List of Abbreviations ...... ix

Glossary ...... xii

Abstract...... xv

Chapter I. Introduction...... 1 Biological Response to Biomaterials...... 1 Protein Adsorption and Complement...... 5 Cellular Adhesion on Biomaterials...... 7 Macrophage Responses to Biomaterial Surfaces...... 10 Lymphocyte Responses to Biomaterials: Lymphocyte/Macrophage Interactions...... 16 Inflammatory Mediators in Inflammation, Foreign Body Reaction, and Wound Healing ...... 20 Mathematical Modeling...... 27 Model Biomaterial Surfaces ...... 30 Significance...... 45 Hypothesis and Specific Aims...... 46 References...... 47

Chapter II. Lymphocyte/Macrophage Interactions: Biomaterial Surface- Dependent Cytokine, Chemokine, and Matrix Protein Production ...... 60 Abstract...... 60 Introduction...... 61 Materials and Methods...... 64 Results...... 69 Discussion...... 81 References...... 88

Chapter III. Paracrine and Juxtacrine Lymphocyte Enhancement of Adherent Macrophage and Foreign Body Giant Cell Activation ...... 93 Abstract...... 93 Introduction...... 94 Materials and Methods...... 96 Results...... 99 Discussion...... 106

i References...... 111

Chapter IV. The Effect of Biomaterial Surface Chemistry on Adherent Lymphocyte Interactions with Macrophages and Foreign Body Giant Cells...... 113 Abstract...... 113 Introduction...... 114 Materials and Methods...... 117 Results...... 120 Discussion...... 131 References...... 142

Chapter V. IFN-γ Production from Lymphocyte Interactions with Biomaterial- Adherent Macrophages and Foreign Body Giant Cells ...... 147 Abstract...... 147 Introduction...... 148 Materials and Methods...... 150 Results...... 152 Discussion...... 155 References...... 170

Chapter VI. Dynamic Systems Model for Lymphocyte Interactions with Macrophages at Biomaterial Surfaces ...... 175 Introduction...... 175 Experimental Methods...... 178 Model Development...... 179 Results...... 195 Discussion...... 203 References...... 212

Chapter VII. Conclusions and Future Directions...... 216 Summary and Conclusions ...... 216 Future Directions...... 224 References...... 228

Appendix I. Dynamic Systems Model Main Code ...... 230

Appendix II. Dynamic Systems Model Output Function ...... 246

Appendix III. Dynamic Systems Model Differential Equations Function ...... 250

Works Cited ...... 253

ii List of Tables

Table 1.1: Adhesion Molecules 10

Table 1.2: T Lymphocyte Subpopulations and Functions 19

Table 1.3: Important Mediators Involved in Inflammation and Wound Healing 23

Table 1.4: Mechanical and Physical Properties of PET 32

Table 1.5: Composition and Contact Angles for PET-based Surfaces 44

Table 2.1: Minimum Detection Limit for the Utilized Protein Detection Techniques 68

Table 2.2: Signal Intensity Ratings and Material Variability Indices 72

Table 4.1: Adherent Lymphocyte Densities (cells/mm2) on Biomaterial Surfaces with Varying Monocyte Concentrations 121

Table 4.2: Adherent Macrophage and FBGC Densities (cells/mm2) on Different Surface Chemistries as a Function of Monocyte Concentration Plated 121

Table 4.3: Ratio of Adherent Lymphocyte Densities to Adherent Macrophage/FBGC Densities on PET-based Surfaces 123

Table 4.4: Density of Adherent Lymphocyte Subtypes (cells/mm2) on Biomaterial Surfaces over Varying Lymphocyte to Monocyte Co- culture Ratios 124

Table 4.5: Interactions of Adherent Lymphocyte (cells/mm2) at Biomaterial Surfaces as a Function of Lymphocyte to Monocyte Co-culture Ratios 129

Table 4.6: FBGC- or Macrophage-Adherent lymphocytes Normalized to the Respective Adherent FBGC or Macrophage Population 129

Table 4.7: Molecular Mediators of Potential Lymphocyte Interactions with Adherent Monocytes, Macrophages, and Foreign Body Giant Cells 134

Table 5.1: IFN-γ Production (pg/mL) from Lymphocyte and Monocyte Cultures Exposed to Biomaterial Surfaces over 3, 7, 10 days 155

iii Table 5.2: Modulators of Lymphocyte Activation and Proliferation 159

Table 5.3: Findings for Biomaterial Lymphocyte Activation Studies 164

Table 5.4: Modulators of Monocyte and Macrophage Activation 166

Table 6.1: Dynamic Systems Model Parameters 182

Table 6.2: Model Nomenclature 183

Table 6.3: Model Parameter Names and Values used in Simulations 192

Table 6.4: Model Parameters for Estimation 193

Table 6.5: Objective Function Scaling Factors for each Biomaterial 195

Table 6.6: Initial Parameter Values and Final Parameter Estimates for All 5 Material Surfaces 198

Table 6.7: Summary of Estimated Parameter Values 200

iv List of Figures

Figure 1.1: Host reaction to implanted biomaterials. 1

Figure 1.2: Variation of cells at the implant site over the course of inflammation and wound healing. 4

Figure 1.3: Complement pathway. 6

Figure 1.4: Chemical structure of poly(ethylene terephthalate). 32

Figure 1.5: Chemical structure of poly(benzyl N,N-diethyldithiocarbamate- co-styrene) 37

Figure 1.6: Preparation of the iniferter. 37

Figure 1.7: General photograft copolymerization reaction. 41

Figure 1.8: Vinyl monomers in photograft copolymers. 41

Figure 1.9: Final PET-based photograft copolymerized surfaces. 42

Figure 1.10: Chemical staining of biomaterial surfaces by toluidine blue and rose bengal for confirmation of charged surfaces. 45

Figure 2.1: (A) Macrophage/FBGC adhesion and (B) percent fusion from high monocyte co-cultures over 10 days. 70

Figure 2.3: IL-8 production from (A) low and (B) high monocyte co-cultures over 10 days. MIP-1β production from (C) low and (D) high monocyte co-cultures over 10 days. 77

v Figure 2.5: Ratio of MMP-9/TIMP-1 in co-cultures over (A) 3 days, (C) 7 days, and (E) 10 days. Ratio of MMP-9/TIMP-2 in co-cultures over (B) 3 days, (D) 7 days, and (F) 10 days. 80

Figure 3.1: IL-1β production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. 100

Figure 3.2: TNF-α production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. 101

Figure 3.3: IL-6 production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. 102

Figure 3.4: IL-8 production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. 103

Figure 3.5: MIP-1β production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. 104

Figure 3.6: IL-10 production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. 105

Figure 3.7: Lymphocyte effects on macrophage/FBGC production of MMP- 9 over (A) 3, (B) 7, and (C) 10 days of culture; TIMP-1 over (A) 3, (B) 7, and (C) 10 days of culture; and TIMP-2 over (A) 3, (B) 7, and (C) 10 days of culture. 106

Figure 4.1: Adherent lymphocyte density (A) and adherent macrophage density (B) as a function of the ratio of lymphocyte/monocyte co-culture concentration after 3 days 122

Figure 4.2: Adherent lymphocyte density as a function of the adherent macrophage and FBGC density after 3 days of culture. 124

Figure 4.3: Types of adherent lymphocytes on the biomaterial surfaces as a function of the ratio of lymphocyte/macrophage co-culture concentrations. A) Total numbers of adherent CD4+ T helper cells, CD8+ T suppressor cells, and CD56+ NK cells and B) percentages of each subtype contributing to the total lymphocyte population on the biomaterial surfaces. 125

Figure 4.5: Types of adherent lymphocyte interactions at the surface of the biomaterial as a function of the ratio of lymphocyte/monocyte co-culture concentrations. A) Total macrophage-, foreign body giant cell-, or direct biomaterial-adherent lymphocyte densities and the B) percentages of each of the types of adherent lymphocyte interactions. 130

Figure 5.1: Quantification of IFN-γ production from lymphocyte and monocyte single and co-cultures over A) 3, B) 7, and C) 10 days. 154

Figure 6.1: Schematic of the direct lymphocyte/monocyte co-culture system. 180

Figure 6.2: Overall system diagram. 181

Figure 6.3: Representative quantitative simulations provided by the developed model. 197

Figure 6.4: Modeling results from parameter estimation for (A) PET, (B) BDEDTC, (C) PAAm, (D) PAANa, (E) DMAPAAmMeI. 203

vii Acknowledgements

I would like to thank Dr. James Anderson for being a great mentor and

encouraging and pushing me to achieving my goals.

I would also like to acknowledge my committee members: Professor

Roger Marchant, Professor Gerald Saidel, Professor Edward Greenfield, and Professor Horst von Recum for their guidance in my research. Their

inputs have been invaluable.

My deepest appreciation to the members of the Anderson lab: Jasmine

Patel, Jacqueline Jones, Analiz Rodriguez, Matt MacEwan, Sarah

MacEwan, Bill Brodbeck, Amy McNally, Bonnie Berry, Erica Colton, and

Gabriela Voskerician for their time and generous contributions to my work

Finally, my family and friends have been unfailing in their support and

encouragement. Without them, I could not have succeeded.

viii List of Abbreviations

ANOVA Analysis of Variance APC Antigen Presenting Cells AS Autologous Serum ATR Attenuated Total Reflectance ATR-FTIR Attenuated Total Reflectance Fourier Transform Infrared Analysis Bc-1 Type 1 B cell Bc-2 Type 2 B cell BDEDTC poly(benzyl N,N-diethyldithiocarbamate-co-styrene) BDNF Brain-Derived Neurotrophic Factor BLC B-lymphocyte Chemoattractant ˚C Degrees Celsius CD Cluster of Differentiation ck β 8-1 Chemokine β 8-1 CR Complement Receptor DC-STAMP Dendritic Cell-Specific Transmembrane Protein DMAPAAmMeI N-methiodide of dimethylamino propylacrylamide ECM Extracellular Matrix EGF Epidermal Growth Factor ELISA Enzyme-Linked ImmunoSorbent Assay ENA Epithelial-Derived Neutrophil Activating Protein eV Electron Volt FAK Focal Adhesion Kinase FBGC Foreign Body Giant Cells FBR Foreign Body Reaction FGF Fibroblast Growth Factor Flt-3 Ligand Fms-like Tyrosine Kinase-3 Ligand (STK-1 Ligand) GCP Granulocyte Chemotactic Protein G-CSF Granulocyte-Colony Stimulating Factor GDNF Glial Cell-Derived Neurotrophic Factor GM-CSF Granulocyte Macrophage Colony Stimulating Factor GPa Gigapascals GPC Gas Permeation Chromatography GRO Growth Related Oncogene HGF Hepatic Growth Factor HLA Human Leukocyte Antigen ICAM Intercellular Adhesion Molecule IFN Interferon Ig Immunoglobulin IGF Insulin-Like Growth Factor

ix IGFBP Insulin-Like Growth Factor Binding Protein IgG Immunogobulin G IL Interleukin IP-10 Interferon - Inducible Protein - 10 IPN Interpenetrating Polymer Network LFA Leukocyte Function-Associated Antigen LIF Leukemia Inhibitory Factor LPS Lipopolysaccharide LTB4 Leukotriene B4 LVAD Left Ventricular Assist Device MCP Monocyte Chemotactic Protein MCSF Macrophage Colony Stimulating Factor M-CSFR Macrophage Colony Stimulating Factor Receptor MDC Macrophage-Derived Chemokine MeV Megaelectron Volt MHC Major Histocompatibility Complex MIF Macrophage Migration Inhibitory Factor MIG Monokine Induced by Gamma Interferon MIP Macrophage Inflammatory Protein MMP Matrix Metalloproteinase MPa Megapascals mRNA Messenger Ribonucleic Acid NAP Neutrophil Activating Protein NF-kB Nuclear Factor - kB NK Natural Killer NK1 Type 1 Natural Killer Cells NK2 Type 2 Natural Killer Cells NMR Nuclear Magnetic Resonance NO Nitric Oxide NT Neurotrophin PAAm Polyacrylamide PAANa Sodium Salt of Polyacrylic Acid PARC Pulmonary and Activation-Regulated Chemokine PBS Phosphate Buffered Saline PDGF Platelet-Derived Growth Factor PEG Polyethylene Glycol PEO Polyethylene Oxide PET Polyethylene Terephthalate PlGF Placenta Growth Factor PMN Polymorphonuclear Leukocyte RANTES Regulated Upon Activation, Normal T-cell Expressed and Secreted

x RNASE A Ribonuclease A ROI Reactive Oxygen Intermediates ROS Reactive Oxygen Species SCF Stem Cell Factor SCFR Stem Cell Factor Receptor SDF Stromal Cell-derived Factor SEM Standard Error of the Mean SFM Serum Free Medium STAT Signal Transducer and Activator of Transcription TARC Thymus and Activation-Regulated Chemokine Tc1 Type 1 T cytotoxic Tc2 Type 2 T cytotoxic TCPS Tissue Culture Polystyrene TCR T Cell Receptor Tg Glass Transition Temperature TGF-β Transforming Growth Factor Th1 Type 1 T Helper Lymphocyte Th2 Type 2 T Helper Lymphocyte TIMP Tissue Inhibitors of Metalloproteinases TLR Toll-like Receptor Tm Melting Point Temperature TNF Tumor Necrosis Factor TNFR Tumor Necrosis Factor Receptor TRAP Tartrate-Resistant Acid Phosphatase UV Ultraviolet VEGF Vascular Endothelial Growth Factor VLA Very Late Activation Antigens

xi Glossary

C γ-chemokines with only two cysteines: one N-terminal cysteine and one cysteine downstream CC β-chemokines that have two adjacent cysteines near their amino terminus. CCL1 I-309 CCL11 Eotaxin, Eotaxin-1 CCL12 MCP-5 CCL13 MCP-4 CCL15 MIP-1δ CCL17 TARC CCL18 PARC, MIP-4 CCL19 MIP-3β CCL2 MCP-1 CCL20 MIP-3α CCL22 MDC CCL23 MIP-3, ck δ-8 CCL24 Eotaxin-2 CCL26 Eotaxin-3 CCL3 MIP-1α CCL4 MIP-1β CCL5 RANTES CCL7 MCP-3 CCL8 MCP-2 CCL9/CCL10 MIP-1γ CD11a Integrin αL, α chain of LFA-1 and associates with CD18 to form receptor for CD50, CD54, or CD102 CD11b Complement receptor-3, Integrin αM chain; Associates with CD18 to form receptor for C3bi, fibrinogen, CD54, CD102 CD11c Integrin αX chain; Together with CD18 forms adhesion receptor for fibrinogen and C3bi CD137 4-IBB; receptor for 4-IBBL in T cell costimulation CD14 LPS receptor CD15 Hapten X CD152 CTLA-4; Ligand for CD80 and CD86 to inhibit T cell activation CD154 CD40 Ligand that induces activation CD16 FcγRIII CD18 β2 integrins; β chain of LFA-1 CD19 Forms a noncovalent complex with CD21, CD81, Leu13 in modulating B cell receptor signal transduction

xii CD2 Ligand for CD48 and CD58 (LFA-3) CD23 FcεRII; low affinity IgE receptor CD25 α chain of IL-2 recepter; Upregulated upon activation; Associates with CD122 and CD132 to form high affinity receptor for IL-2 CD28 T cell antigen that interacts with CD80 and CD86; functions in costimulations CD29 VLA β chain CD3 Pan T cell Marker; Forms signal transduction complex for TCR CD35 C3b Receptor; Facilitates phagocytosis CD4 Serves as receptor for MHC II in recognition of antigen CD40 Induces cell activation and/or differentiation upon binding its ligand, CD154 CD44 Extracellular matrix receptor type III; Receptor for fibronectin, laminin, collagen, and chemotactic cytokine osteopontin CD45 Leukocyte Common Antigen CD45RA Expressed on naïve T cells CD45RO Expressed on memory T cells CD50 Intercellular adhesion molecule-3; Ligand for activated LFA-1 CD51/CD61 α and β Integrin Subunits for Vitronectin Receptor CD54 ICAM-1; Functions as ligand for LFA-1 and Mac-1 CD56 Neural cell adhesion molecule-1 expressed predominantly on NK cells CD69 Early activation antigen; Earliest inducible cell surface glycoprotein acquired during activation CD71 Transferrin Receptor CD8 Forms receptor for MHC I CD80 Also called B7-1 and BB1; Interacts with CD28 (T cell costimulation) or CD152 (T cell inhibition) CD86 Also called B7-2; Interacts with CD28 (costimulation) or CD152 (inhibition) CD95 FAS, Apo-1; Cross linked CD95 induces apoptosis CDw49b/CD29 Very Late Antigen (VLA-2) CDw49d/CD29 VLA-4 Receptor

CX3C δ-chemokine with three amino acids between the two cysteines

CX3CL1 Fractalkine CXC α-chemokines with two N-terminal cysteines separated by one amino acid CXCL1 GRO-α CXCL10 IP-10 CXCL2 MIP-2α, GRO-β CXCL3 MIP-2β, GRO-γ

xiii CXCL4 Platelet Factor-4 CXCL5 ENA-78 CXCL7 NAP-2 CXCL8 IL-8 CXCL9 Mig IBB Also known as CD137; Receptor on T cells involved in T cell costimulation IBBL Costimulatory ligand for IBB receptor ICOS Costimulatory ligand for CD28 in T cell activation

xiv Lymphocyte and Macrophage Interactions in the Response to Biomaterial Surfaces

Abstract

DAVID T. CHANG

Synthetic polymers provide a wide array of physical properties and chemical compositions that can be exploited as components in prosthetics, devices, and tissue- engineered constructs. This requires elucidating the biological response to biomaterials and the relationship between surface characteristics and cellular behavior. Lymphocytes transiently appear at implant sites and have been observed clinically to be reactive.

However, the role of lymphocytes in the tissue response and the mechanisms involved are still unclear. This research aimed to gain insight into the interactions between lymphocytes and macrophages at biomaterial surfaces and explore the relationship between surface properties and lymphocyte/macrophage interactions. The studies addressed the hypothesis that biomaterial surface chemistries modulate the direct and indirect lymphocyte interactions with macrophages and foreign body giant cells. PET- based photograft copolymerized surfaces with hydrophobic, hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic characteristics were utilized as model surfaces. Quantification of select inflammatory mediators showed that hydrophilic/neutral and hydrophilic/anionic surfaces promoted pro-inflammatory responses and a reduced potential for ECM degradation relative to other surfaces.

xv Proteomic cytokine arrays provided a useful tool for assessing the role of soluble

mediators in the response to biomaterials and identified potential fusion factors. We

subsequently showed the capability of lymphocytes, through direct and indirect

mechanisms, to enhance macrophage activation and production of pro-inflammatory mediators. Hydrophilic/neutral and hydrophilic/anionic surfaces were shown to be highly but distinctly activating. These surfaces also promoted distinct lymphocyte subset

adhesion to macrophages and FBGCs compared to hydrophobic PET suggesting

induction of differential macrophage and FBGC phenotypes on varying biomaterial

surfaces. IFN-γ production in direct and indirect co-cultures rather than individual

cultures suggested nonspecific mechanisms of lymphocyte activation. Direct cell-cell

interactions enhanced IFN-γ production more so on the hydrophilic/anionic surfaces than

on any other surface, indicating this surface as highly lymphocyte activating. These

findings provide insight into lymphocyte and macrophage interactions in response to

biomaterial surfaces along with evidence consistent with the hypothesis that distinct

surface chemistries modulate lymphocyte and macrophage interactions. Finally, we

present a mechanistic model which provides a tool for further analysis of lymphocyte and

macrophage interactions in response to biomaterial surfaces and a step toward

quantitative predictability of biomaterial-dependent processes.

xvi Chapter I

Introduction

Biological Response to Biomaterials

Biocompatibility is a major concern when implanting any type of biomaterial,

biomedical device, or tissue-engineered constructs. It is immediately following

implantation that the host tissue undergoes a sequence of events: injury, cellular

infiltration, acute inflammation, chronic inflammation, granulation tissue formation, foreign body reaction (FBR), and fibrous capsule formation which Figure 1.1 illustrates.1

The extent of these responses determines the implant’s biocompatibility.

Figure 1.1: Host reaction to implanted biomaterials.2

- 1 - Chapter I

Regardless of the implanted material, the process of implantation creates an injury which initiates material interaction with the blood leading to protein adsorption to the biomaterial surface and development of a blood-based transient provisional matrix that forms on and around the biomaterial. The provisional matrix is the initial thrombus/blood clot at the tissue/material interface. The injury to vascularized connective tissue not only initiates the inflammatory responses (innate immunity), it also leads to thrombus formation involving activation of the extrinsic and intrinsic coagulation systems, the complement system, the fibrinolytic system, the kinin-generating system, and platelets.

From a wound healing perspective, blood protein deposition on a biomaterial surface is described as provisional matrix formation. The provisional matrix furnishes structural, biochemical, and cellular components to the processes of wound healing and foreign body reaction. The presence of mitogens, chemoattractants, cytokines, growth factors, and other bioactive agents within the provisional matrix provides for a rich milieu of activating and inhibiting substances capable of modulating macrophage activity, along with the proliferation and activation of other cell populations in the inflammatory and wound healing responses. The provisional matrix may be viewed as a naturally derived, biodegradable sustained release system in which bioactive agents are released to control subsequent phases of wound healing.

Injury also results in infiltration of cells to the site of the implant setting off inflammatory and wound healing events. Figure 1.2 shows the temporal cellular variation at the implant site over the course of the tissue reaction. Acute inflammation occurs over a short period of days where polymorphonuclear cells (PMNs) predominate

- 2 - Chapter I

and act to phagocytose and eliminate pathogens or foreign particles. Mast cell degranulation with histamine release and fibrinogen adsorption is known to mediate acute inflammatory responses to implanted biomaterials.3,4 Interleukin-4 (IL-4) and

interleukin-13 (IL-13) also are released from mast cells in a degranulation process and can play significant roles in determining the extent and degree of the subsequent development of the foreign body reaction.5,6 Biomaterial-mediated inflammatory

responses may be modulated by histamine-mediated phagocyte recruitment and

phagocyte adhesion to implant surfaces facilitated by adsorbed host fibrinogen. Both H1

and H2 histamine receptor antagonists greatly reduce the recruitment of

monocytes/macrophages and neutrophils on implant surfaces.

PMNs of the acute inflammatory phase are replaced by monocytes, macrophages, plasma cells, and lymphocytes as the response progresses to chronic inflammation. The cells continue the process of attempting to eliminate the foreign material. This phase of the reaction can last days, weeks, months, even years depending on the type and persistence of inflammatory stimulus. This chronic inflammatory response to biomaterials is usually of short duration and is confined to the implant site. With biocompatible materials, early resolution of the acute and chronic inflammatory responses occurs with the chronic inflammatory response composed of mononuclear cells usually lasting no longer than two weeks. Concurrent to the events in the region around the implant is the series of events at the material/tissue interface called the foreign body reaction. The fusion of macrophages to form foreign body giant cells (FBGCs) occurs in an attempt to remove the offending foreign material and can persist for the lifetime of the implant. FBGCs have been shown to mediate degradation of biomaterial surfaces

- 3 - Chapter I through the release of reactive oxygen intermediates (ROIs, oxygen free radicals), degradative enzymes, and acid into the privileged zone between the cell membrane and biomaterial surface.7-10 Chronic inflammation also has been used to describe the foreign body reaction.

As the tissue around the implanted material progresses to the healing response, granulation tissue forms involving neovascularization and progressive increase in fibroblast activity. Fibroblasts proliferate and synthesis of extracellular matrix increases to support the healing and formation of new tissue at the implant site. Granulation tissue is the precursor to fibrous capsule formation and granulation tissue is separated from the implant or biomaterial by the cellular components of the foreign body reaction; a one- to two-cell layer of monocytes, macrophages, and foreign body giant cells. Isolation of the material with the activities at the material interface occurs with fibrous encapsulation.

Figure 1.2: Variation of cells at the implant site over the course of inflammation and wound healing.1

- 4 - Chapter I

Protein Adsorption and Complement

Within seconds of implanting a biomaterial, biomedical device, or tissue- engineered construct, proteins from the blood and interstitial fluid will begin to adsorb

onto the surface. Some of these proteins include: fibrinogen, fibronectin, vitronectin, immunoglobulin, albumin, von Willebrand factor, and complement factors (e.g. C3b).

The Vroman effect along with material surface property determines the concentration and

distribution of plasma proteins adsorbed onto the surface.11-14 Additionally, material

surface property dictates the conformation as well as the biological reactivity of the adsorbed protein.12,14-16

The complement system has long been recognized as a major host defense system

for the interaction and removal of foreign substances in vivo. Complement activation and

its subsequent reactions have been identified as causing adverse side-effects when

blood/material interactions occur with devices such as hemodialyzers, oxygenators,

catheters, prostheses, stents, vascular grafts, and other devices and materials. In

blood/material interactions, there is tight cross-talk between the different cascade systems

and platelets and leukocytes in the induction of clotting and inflammation.

The components of the complement system are plasma proteins that undergo a

series of reactions in order to destroy foreign cells. The complement system shown in

Figure 1.3 can proceed through a classical as well as an alternative pathway depending on

the activator. Each pathway starts with a different factor but both terminate in the

membrane attack complex, which when formed in the cell membrane, results in cell lysis.

The classical pathway begins with the C1 complex while the alternative pathway begins

with C3 factor binding onto the target surface. Both then proceed to the formation of

- 5 - Chapter I

their respective C3 convertases. These convertases then catalyze the formation of the C5

convertase resulting in the terminal pathway. Biomaterials are capable of activating the

complement system through both pathways. Complement activation was shown to occur

primarily through the classical pathway when it occurred simultaneously to the deposition of a protein film on a material surface; however, the alternative pathway occurred through C3b deposition subsequently and on top of the adsorbed protein layer.17

Figure 1.3: Complement pathway.18

Cellular interactions with surface-adsorbed protein modulate cellular responses

such as adhesion, morphology, growth, differentiation, and activation. Therefore, different surface chemistries can elicit varying cellular behaviors based on the composition and conformation of the adsorbed protein layer. This has been shown for behaviors including adhesion19-24, differentiation25,26, activation20,27,28, and apoptosis21,22 on biomaterial surfaces.

- 6 - Chapter I

Cellular Adhesion on Biomaterials

Cells most likely do not interact directly with the foreign material but rather they

recognize the implanted material via the layer of adsorbed proteins on the surface.12,15

The adsorbed blood protein-modified material surface is the substrate with which the

recruited monocytes/macrophages encounter and interact. The various plasma and

extracellular matrix proteins deposited onto the surface give cells a means of attaching

via surface receptors. These interactions are mediated by receptor-ligand interactions.

The integrin family of receptors is a major set of adhesion molecules mediating cell-

extracellular matrix as well as intercellular interactions.29,30 These adhesion molecules

allow cells to migrate through the extracellular matrix and mediate signal transduction

between the cell and its environments so that the cell can respond to its environment.29

These transmembrane receptors are heterodimers consisting of α and β subunits with specificity for specific amino acid sequences. The partnering of different subunit chains confers variability to the specificity and function of the receptor.31 Integrins are able to

bind extracellular matrix and plasma proteins including: fibrinogen, fibronectin, complement 3b, immunoglobulin.32,33

Monocytes/Macrophages express integrins with 3 different types of β chains, β1,

β2, β3. In monocytes/macrophages there are three β1 integrins, four β2 integrins, and

one β3 integrin. β1 integrins include α4/β1 and α5/β1 which bind fibronectin and α6/β1

which binds laminin. Of the β2 integrins, there are αL/β2, αM/β2, and αD/β2 which are

specific for intercellular adhesion molecules (ICAMs) and αX/β2 which binds

complement fragment C3bi and fibrinogen. αM/β2 also interacts with a variety of other

ligands such as fibrinogen, C3bi, and Factor X. Finally, monocytes/macrophages express

- 7 - Chapter I

αV/β3 of the β3 integrins which bind to vitronectin along with other RGD containing

extracellular proteins.31

Initial monocyte adhesion has been shown to be achieved through β2 integrins, in

particular αM/β2 (Mac-1, CD11b/CD18), by binding to various adsorbed protein ligands

including fibrinogen, fibronectin, IgG, and complement fragment iC3b.21,34 While some

efforts have focused on complement components providing the initial adhesion ligands, others have focused on fibrinogen as being the principal adsorbed protein ligand. In particular, Mac-1 interactions with fibrinogen epitopes was a primary mediator of phagocyte accumulation on implanted biomaterial surfaces.27 Recent studies have

suggested that multiple protein ligands may participate in the receptor-ligand binding and monocyte adhesion. In particular, complement activation on fibrinogen-adsorbed surfaces has been suggested as the primary adhesion event.35 β1 integrins have been

determined to play a role in the subsequent adhesion and IL-4 induced macrophage

fusion to form foreign body giant cells. This phenomenon is time-dependent as β1

integrins are not initially detected on adherent monocytes but begin to appear during

macrophage development and are strongly expressed on fusing macrophages and foreign

body giant cells with increased culture time.36 β1 integrins appear to be involved in

adhesion through fibronectin adsorption.37

McNally et al. demonstrated expression and co-localization of α3, α5 or αV with

β1 on fusing macrophages/FBGC at day 7 of culture as well as the strong co-localization

of αM and αX with β2 in FBGC and at macrophage fusion interfaces. Therefore, IL-4-

induced FBGC are characterized by the expression of αM/β2, αX/β2, α5/β1, α5/β1,

α2/β1, and α3/β1, which indicates the potential interactions of complement C3b

- 8 - Chapter I

fragments, fibrin, fibrinogen, fibronectin, factor X, and vitronectin at sites of biomaterial

implantations.38

Receptors for the Fc region of antibodies can also function as adhesive molecules.

Serum amyloid P, which functions as an opsonin, has been shown to adsorb in high amounts on surfaces and interacts with Fcγ receptors to promote monocyte adhesion.39

Few studies, compared to monocyte/macrophage adhesion, have investigated lymphocyte interaction with biomaterial surfaces. Lymphocytes have been shown to adhere in a protein and biomaterial-dependent manner.40-42 Moreover, material surface

chemistries can dictate differential lymphocyte subpopulation adhesion.43 Brodbeck et al.

more recently showed visualization of biomaterial-adherent lymphocytes by scanning

electron microscopy and optical microscopy.44 Integrin receptors, as in

monocyte/macrophage adhesion, appear to be important for lymphocyte adhesion.

Fibronectin adsorption on a biomaterial surface facilitates human peripheral lymphocyte

attachment through β1 integrin.45 Similar to monocyte and macrophages described above, integrin receptors for complement factors exist on T-lymphocytes46 and have been shown to be up-regulated upon activation47. Thus, there is a possibility that adsorption of complement factor could play a role in lymphocyte attachment as well. Examples of integrin and immunoglobulin molecules, their tissue distribution, and ligands are shown in Table 1.1.

- 9 - Chapter I

Table 1.1: Adhesion Molecules32

Tissue Name Distribution Ligand Monocytes, T cells, α :β L 2 macrophages, ICAMs (LFA-1, CD11a/CD18) neutrophils, dendritic cells

Neutrophils, Integrins α :β ICAM-1, iC3b, m 2 monocytes, (Mac-1, CR3, CD11b/CD18) fibrinogen Bind to cell-adhesion macrophages molecules and extracellular matrix αx:β2 Dendritic cells, (CR4, p150.95, macrophages, iC3b CD11c/CD18) neutrophils

α :β Monocytes, 5 1 Fibronectin (VLA-5, Cd49d/CD29) macrophages

CD2 (LFA-2) T cells LFA-3 Lymphocytes, Immunoglobulin ICAM-1 activated vessels, LFA-1, Mac-1 Superfamily dendritic cells ICAM-2 Resting vessels LFA-1 Ligand for integrins ICAM-3 Naïve T cells DC-SIGN Role in cell adhesion Lymphocytes, LFA-3 antigen-presenting CD2 cells LFA: Leukocyte function-associated antigen; CR: Complement receptor; VLA: Very late activation antigens; ICAM: Intercellular adhesion molecules

Macrophage Responses to Biomaterial Surfaces

Macrophage Recruitment

The progression of events in inflammation and the foreign body response requires the extravasation and migration of monocytes/macrophages to the implant site. The guided movement of monocytes/macrophages occurs in response to chemokines and other chemoattractants. Chemokines are cytokines that have chemoattractive properties and consist of 4 major families: CC, CXC, C, and CX3C based on the spacing between

- 10 - Chapter I

the first 2 terminal cysteine residues.48 Chemokines are not only involved in

orchestrating cellular migration in inflammation and wound healing but play roles in hematopoiesis, angiogenesis, tumor metastasis, lymphocyte differentiation, and lymphocyte homing.48-50

Following blood-material interactions, platelets and the clot release

chemoattractants, such as transforming growth factor (TGF-β), platelet-derived growth

factor (PDGF), CXCL4 (Platelet Factor, PF4), leukotriene (LTB4), and interleukin (IL)-

1, that can direct macrophages to the wound site.51 In addition, mast cell degranulation

and release of histamine has been shown to play an integral role in recruiting phagocytes,

including macrophages, to the site of the implanted biomaterial.4 The assembly of

macrophages at the implant site leads to further propagation of chemoattractive signals.

Macrophage production of PDGF, tumor necrosis factor (TNF-α), IL-6, granulocyte- colony stimulating factor (G-CSF), and granulocyte macrophage colony stimulating factor (GM-CSF) call more macrophages to the wound site.51 Rhodes et al. showed

expression of CCL2 (Monocyte chemotactic protein, MCP-1) in exudate macrophages

surrounding implanted polyethylene materials.52 CCL2 (MCP-1) along with CCL5

(regulated upon activation, normal T-cell expressed and secreted, RANTES), CCL3

(macrophage inflammatory protein, MIP-1α), CCL4 (MIP-1β), CCL7 (MCP-3), CCL8

(MCP-2), and CCL13 (MCP-4) are chemokines known to attract monocytes/macrophages.48,53 However, CCL2 (MCP-1) was shown not to influence the

recruitment of monocytes to subcutaneous implant sites.54 Utilizing our in vitro culture

system of human blood-derived monocytes/macrophages we have demonstrated

production of CCL4 (MIP-1β), CCL2 (MCP-1), CCL13 (MCP-4), CCL22 (MDC) by

- 11 - Chapter I

biomaterial-adherent macrophages.55 Some of these products have the potential to recruit

additional macrophages to the biomaterial-tissue interface. Once at the implant site or

biomaterial surface, the macrophages can then adhere and engage in the subsequent events of the foreign body reaction.

Apoptosis

At the biomaterial surface, induction of apoptosis can occur through cytokine signaling (e.g. TNF-α) as well as disruption of cellular adhesion (e.g. integrin signaling).

Biomaterial surface chemistry has been shown to influence apoptosis of adherent macrophages both in vitro and in vivo.21,56-58 The induction of apoptosis was mediated by

TNF-α.

Integrins allow cellular interactions with adsorbed proteins on biomaterial

surfaces but are also important in regulating cell death which is necessary for cell

detachment and tissue remodeling.59 Anoikis is a term for apoptosis induced by cell

detachment from its supportive matrix.60 When a cell is properly adhered to a surface,

focal adhesion kinase (FAK) mediates survival signaling. Disruption of adhesion signals promotes anoikis.61 Caspase-3, a protein involved in apoptosis signaling, is activated by

neutrophils under shear stress and leads to detachment from biomaterials.62 Caspases are

involved in cleaving gelsolin, a protein involved in regulating actin polymerization, and

this in turn disrupts adhesion.61 Monocytes initially adhere to most surfaces quite well,

but fail to maintain adhesion over time.36 Materials that do not promote adhesion lead to

cell detachment and anoikis.

- 12 - Chapter I

Macrophage Fusion

In response to implanted biomaterial surfaces, monocytes adhere, differentiate

into macrophages, and subsequently fuse to form foreign body giant cells. The molecular

mediators in inducing fusion as well as the specific fusion mechanisms have not yet been

elucidated. IL-4 and IL-13 were discovered to be inducers of foreign body giant cell

formation. In vitro addition of IL-4 and IL-13 antibodies led to FBGC formation.63,64

Neutralization of IL-4 in vivo inhibited fusion while recombinant IL-4 addition resulted

in enhanced FBGC formation.65 Mannose receptor expression was found to be involved

in the IL-4-mediated fusion process.66 Additional molecules involved in the fusion

process include macrophage chemotactic protein (MCP)-1 (CCL2),54 dendritic cell-

specific transmembrane protein (DC-STAMP),67 and β1 integrin receptors.36 Fusion also

requires the presence of a surface that supports fusion as particular adsorbed proteins

such as fibronectin and vitronectin facilitate the fusion process.68,69 Therefore, this

suggests that surfaces that promote FBGC formation may preferentially adsorb these

proteins.

Phenotypic Characteristics of Foreign Body Giant Cells

Foreign body giant cells display an antigenic phenotype similar to monocytes and macrophages, reflecting the fact that FBGCs are formed from the fusion of monocyte-

derived macrophages.70 Current medical implantations such as arthroplasties along with

testicular and breast implants have provided a means of studying the characteristics of

foreign body giant cells in the human response to implanted materials. Foreign body

giant cells in tissue removed from human implant surgeries have been shown in situ to

express macrophage-associated membrane molecules such as CD45 (leukocyte common

- 13 - Chapter I

antigen), CD13, CD14, CD15A (Hapten X), CD37, CD39, CD43, and HLA-DR;

receptors such as CD16 (FcRIII), CD31 (FcRII), CD35 (C3b receptor), CD71 (transferrin

receptor); and adhesion molecules such as CD11a,b,c, CD18 (leukocyte function

associated, LFA, antigen family), CD54 (ICAM-1), and CD44; while CD68 was strongly

stained in the cytoplasm of FBGCs.70-73 Additionally, α and β integrin subunits for the

vitronectin receptor (CD51/CD61), very late antigen receptor (VLA-2, CDw49b/CD29)

and VLA-4 receptor (CDw49d/CD29) were detected on foreign body giant cells.74

Expression of osteoclast markers tartrate-resistant acid phosphatase (TRAP), and

vitronectin receptor were also found on FBGCs derived from tissue surrounding total

joint arthroplasties.75 Osteoclasts are phenotypically different multinucleated giant cells

found in bone. The foreign body giant cells formed near bone interfaces or in joint

capsules would be expected to differ from those formed in soft tissues as the surrounding

environmental signals would be dissimilar.

Foreign body giant cells have the potential to be responsive to cellular signals via

cell surface receptor expression as well as actively participate in the inflammatory

response through the production of cytokines. Foreign body giant cells derived from human arthroplasties were shown to express cytokine receptors on the cell membrane such as gp130, IL-1R type 1, IL-2Rα, IL-2Rγ, IL-6R, TNFR, M-CSFR, and SCFR, while receptors for IL-4 and GM-CSF were weakly detected and receptors for IL-3 and IL-8 were not present.70,76 FBGCs, derived from a murine model injected with nitrocellulose particles, showed production of IL-1α and TNF-α only during the first month of the

foreign body reaction and subsequently initiated the production of transforming growth

factor-β (TGF-β), but did not show the production of macrophage inflammatory proteins

- 14 - Chapter I

(MIP).77 Although derived from a non-human source, the study shows the capability for

FBGCs to influence the tissue response. Like macrophages, foreign body giant cells show the ability to engage in cell-cell as well as cell-matrix interactions.

As mentioned previously, materials implanted into varying types of tissues would be expected to evoke varying responses. The described investigations into the phenotypic characteristics of foreign body giant cells involved responses to particulate materials. The tissue surrounding total joint prosthesis often is exposed to wear particles of micrometer sizes, including polyethylene, acrylic cement, and metal, which can lead to a foreign body reaction. Macrophages are capable of phagocytosing very small particles

(< 5 μm) while larger particle sizes (> 10 μm) induce the formation of foreign body giant cells. Obviously in soft tissue environments with minimal mechanical forces, implants will tend to stay intact and macrophages and foreign body giant cells will adhere and respond to material surfaces. Therefore, tissue location as well as material form and size, are factors that can influence the foreign body reaction to materials introduced into the body.

Effector Functions

Monocytes/macrophages play an integral part in not only the inflammatory and wound healing response at the implant site but also the tissue response at the tissue/material interface. Monocytes migrate from the circulation to the implantation site, adhere to the surface of a biomaterial, and differentiate into macrophages to initiate the foreign body reaction. In response to the foreign stimulus, the macrophages are capable of directing the tissue response at the tissue/material interface through production of soluble factors, generating degradative agents, attempting to phagocytose, and fusing to

- 15 - Chapter I

form FBGCs.1,78-81 FBGCs have been shown to be responsible for causing oxidative

damage and stress cracking to implanted materials. FBGCs were removed from

implanted surfaces and evidence of surface pitting and damage were shown to exist only

in areas where the cells were located.7 The physical damage can ultimately lead to failure

of the implanted material, device, or construct.

Lymphocyte Responses to Biomaterials: Lymphocyte/Macrophage Interactions

Investigation into the role of monocytes and macrophages in response to

biomaterials led to the discovery that IL-464,65 and IL-1363 are potent inducers of foreign body giant cell formation from adherent monocytes/macrophages. IL-4 and IL-13 are cytokines secreted predominately by lymphocytes. The presence of lymphocytes at the implant site during the chronic inflammatory phase as well as lymphokine participation in macrophage fusion implicates the lymphocyte as playing a critical role in the foreign body response. The lymphocyte population is comprised of B lymphocytes (B cells), natural killer (NK) cells, and T lymphocytes (T cells). B cells are involved in the recognition of foreign substances and producing antibodies for the elimination of the antigens. NK cells are known for mediating killing of tumor cells as well as cells infected by intracellular pathogens through inducing apoptosis. T lymphocytes, which comprise the largest percentage, are categorized into CD8+ and CD4+ subpopulations with CD4+ T cells further divided into Type 1 T helper (Th1) and Type 2 T helper (Th2) subsets. CD8+ cells and Th1 CD4+ cells are involved in cell-mediated immunity while

CD4+ Th2 cells are involved in the humoral (antibody-mediated) response. CD8+ cells, also known as cytotoxic/suppressor cells, destroy cells in a similar manner as NK cells in

- 16 - Chapter I

creating pores in the cell membranes of target cells and signaling to the cell to undergo

apoptosis. Th1 CD4+ cells activate macrophages to destroy intracellular pathogens while

Th2 CD4+ cells initiate a B cell antibody-mediated response to antigen.

The normal immune response to foreign pathogens such as bacteria and viruses is well known and described. Initiation of the immune response occurs with recognition of the foreign material by phagocytic or antigen presenting cells (APCs) such as the macrophage. These cells process the antigen and present it to other immune cells such as

T lymphocytes through major histocompatibility complex molecules (MHC). The T cells recognize the presented antigen and MHC molecules through the T cell receptor (TCR).

Two classes of MHC molecules are expressed on host cell surfaces. Type I MHC molecules exist on virtually all cells in the body and present antigen from the cytosol

while Type II MHC molecules are present only on a subset of cells such as B and T cells,

dendritic cells, macrophages, and other APCs. CD4+ T “helper” cells specifically

recognize class II MHC molecules and once activated, secrete the autocrine IL-2 which

mediates activation and replication of lymphocytes. These cells can then communicate

with macrophages and phagocytic cells to destroy intracellular pathogens or stimulate B

cells to produce antigen-specific antibodies depending on whether the T cells

differentiate into Th1 or Th2 cells, respectively. T lymphocytes can also directly mediate

cytotoxicity through the CD8+ subpopulation which specifically recognizes class I MHC

molecules. Although the TCR/MHC interactions are necessary, they are insufficient for

stimulation. A costimulatory signal is still required for cellular responsiveness.

Examples of this second signal include B7, CD40, and 4-IBBL molecules on APCs

engaging with the CD28, CD40 ligand, and 4-IBB molecules of T cells, respectively.

- 17 - Chapter I

The T cell subpopulations and their functions are described in Table 1.2. Lymphocytes are generally activated through direct interactions with antigen presenting cells (APC)

(e.g. macrophages, dendritic cells). However, lymphocytes can also be activated through non-contact mechanism such as by macrophage-derived cytokines (e.g. IL-15 and IL-18) and lymphokines (e.g. IL-2 and IL-21).82,83

Prior research has primarily focused on the role of macrophages at the tissue/material interface. Although lymphocytes have been shown to adhere to biomaterial surfaces40,41,43,84, there have been relatively few studies determining lymphocyte activity and cellular interactions at the tissue and material surface interface.

Lymphocyte interactions with monocytes/macrophages at the material surface have been shown in vitro to enhance the adhesion and fusion of macrophages primarily through paracrine-mediated mechanisms.44 Moreover, these interactions have been shown to be material surface chemistry dependent.85 Recently, Rodriguez et al. showed the lack of T lymphocyte activation by the lack of proliferation and lack of cell surface activation markers (CD69 and CD25).86 Although we have some evidence that lymphocytes do play a role at the implant surface and have the capability of modulating the foreign body response, the specific roles, cellular interactions, both direct and indirect, and the effect of biomaterial surface chemistry remain unclear.

- 18 - Chapter I

Table 1.2: T lymphocyte Subpopulations and Functions32

CD8+ T Cells CD4+ T cells Cytotoxic (killer) Cells Th1 Th2 Immunity Response Cell-mediated Humoral Cell-mediated Receptor TCR TCR TCR Target Cell Molecule MHC I MHC II MHC II ● Recognize the presence ● Recognize the presence of ● Drives B cell proliferation of intracellular pathogens intracellular pathogens

● Create pores in target ● Activates macrophages to ● Stimulates a antigen- cell membranes destroy intracellular pathogens specific B cell antibody Functions response

● Induce target cell ● Can also induce B cells to apoptosis produce opsonizing antibody

Perforin IFN-γ IFN-γ CD40 ligand IL-4 IL-3 Effector Molecules Granzymes TNF-β GM-CSF Fas ligand IL-5 GM-CSF Fas ligand TNF-α TNF-α IL-3 CD40 ligand IL-10

Lymphocyte responses to synthetic materials have been observed clinically. A

reduction in circulating CD4+ T lymphocyte populations occurs due to apoptosis which in patients implanted with left ventricular assist device (LVAD).87 Left ventricular assist

device (LVAD) patients have also shown the development of B cell hyperreactivity and

immune dysfunction leading to heightened risk of infection and potential autoimmune

disorders due to biomaterial-activated T cells.88-90 Schuster et al. demonstrated the

elevated presence of anti-HLA antibodies and soluble CD40L in LVAD patients indicating B cell activation.88 Additionally, Katzin et al. found that lymphocytes in the

fluid and tissue surrounding silicone gel breast implants were predominantly T cells, and

- 19 - Chapter I

relative to peripheral blood, a greater percentage of the T cells were HLA-DR+ and

CD29+ indicating a state of immune activation.90

Natural killer (NK) cells have also been found to be adversely affected by exposure to synthetic materials. Patients undergoing hemodialysis exhibited reduced numbers of circulating NK cells after dialysis as well as subsequent decreased NK cell activity in vitro.91,92 Moreover, hemodialysis patients acquire an increased susceptibility to and incidence of viral infections.93 An in vivo study of an implanted Dacron prosthesis showed reduced levels of cytotoxic NK cells, large granular lymphocytes, leukocyte and lymphocyte counts.94 These examples provide further motivation for investigation of

lymphocyte responses to implanted biomaterial surfaces. The mechanisms of these

responses have not been fully elucidated.

Unlike the situation with metallic orthopaedic prosthetics, metallic debris/ions are

capable of complexing with serum proteins creating haptens recognizable by lymphocytes resulting in a cell-mediated type IV immune reaction.95 There has been no

evidence of synthetic polymer particles complexing with serum proteins. The specific

mechanisms for lymphocyte activation when exposed to synthetic polymers are unclear.

Inflammatory Mediators in Inflammation, FBR, and Wound Healing

Macrophage activation

Macrophages are activated upon interacting with biomaterial surfaces to produce a multitude of cytokines, chemokines, matrix metalloproteinases (MMPs), and tissue inhibitors of metalloproteinases (TIMP). Monocyte/macrophage cytokine responses to biomaterials have been examined both in vitro and in vivo. Studies have examined

- 20 - Chapter I

monocyte/macrophage response to various biomaterials using inflammatory cytokines

such as TNF-α, IL-1β, IL-6, IL-8 as indicators of cellular activation.96-106 Additionally,

material surface dependent cytokine/chemokine production has become evident.

Variations in surface topography can alter macrophage cytokine/chemokine secretion.107

Brodbeck et al. examined IL-1α, IL-1β, IL-6, IL-8, IL-10, TNF-α, and TGF-β and showed biomaterial surface chemistry dependent macrophage-derived cytokine mRNA expression in vitro and in vivo.108,109 Quantification of mRNA expression does not

necessarily reflect quantity of the final secreted products. Thus, other studies have taken

a proteomic approach. Jones et al. utilized proteomic cytokine array technology to screen for a multitude of inflammatory mediators to investigate cytokine production profiles induced by different biomaterial surface chemistries. Identified targets such as IL-1β, IL-

6, IL-8, IL-10, MMP-9, TIMP-1, TIMP-2, were quantified by ELISA and showed variability with material surface chemistry.55,110 Moreover, they found that based on the

cytokine profiles that adherent macrophages underwent a phenotypic switch from

classically activated to a more alternatively activated phenotype over time.

Macrophages show heterogeneous activation states depending on the activation

signal. IFN-γ and microbial products [e.g. lipopolysaccharide (LPS)] induce macrophages to be classically activated to secrete pro-inflammatory cytokines such as

high IL-1, TNF, IL-6, IL-12 but low anti-inflammatory IL-10 while IL-4 and IL-13

induce an alternatively activated phenotype characterized by high levels of IL-1ra, IL-10 and low IL-12.111

- 21 - Chapter I

Lymphocyte activation

Upon activation, the particular lymphocyte subpopulations can secrete distinct

soluble products. The cytokine profile for CD4+ Th1 cells includes IL-2, IFN-γ, and

TNF-β while the secreted products from CD4+ Th2 lymphocytes include IL-4, IL-5, IL-

6, IL-10, and IL-13.112 These are the two major classifications. However, the production

of these 2 cytokine profiles (i.e. Th1-Th2) extends to CD8+ T cells, NK cells, as well as

B lymphocytes upon differentiation into CD8+ T cytotoxic type 1 and type 2 (Tc1-Tc2),

NK type 1 and type 2 (NK1-NK2), and B cell type 1 and type 2 (Bc1-Bc2),

respectively.112,113

Functions of Inflammatory Mediators

In addition to indicating activation types and levels, inflammatory mediators are

capable of guiding the tissue response to implanted biomaterials. These signals help to

direct the action of other cell types during the inflammatory and wound healing phases.

Macrophage- and lymphocyte-derived cytokines and chemokines have been shown to

influence adherent cell behaviors. There is evidence to suggest that lymphocytes play a

role in the host response to biomaterials. For instance, macrophages can fuse to become

foreign body giant cells when stimulated by IL-4 and IL-13 which are known

lymphokines.63,64 In addition, osteopontin114 and macrophage chemotactic protein

(MCP)-1 54 have been shown to play a role in macrophage fusion to form FBGCs. Many

macrophage-derived cytokines and chemokines are capable of attracting and activating

different cell types. For instance, IL-8 is produced in response to biomaterial surfaces

and is capable of chemoattracting neutrophils and T cells. Other products such as MMPs

and TIMPs modulate the extracellular matrix (ECM) potentially influencing the

- 22 - Chapter I progression and resolution of wound healing leading to fibrous capsule formation surrounding the implanted material. Table 1.3 provides a description and functions of inflammatory mediators that are involved in the inflammatory and wound healing processes. These factors influence many cellular processes and behavior such as recruitment, activation, adhesion, protein/cytokine production, surface molecules/receptors expression, and fusion.

Table 1.3: Important Mediators Involved in Inflammation and Wound Healing 115

Factor Receptor Cellular Sources Functions IL-1β IL-1R1 Monocytes • Pro-inflammatory/pro-wound healing Macrophages • Activates both inflammatory cells (lymphocytes Dendritic cells and monocytes) and wound healing cells B lymphocytes (fibroblasts) NK cells • IL-1 stimulates thymocyte proliferation by inducing IL-2 release, B-cell maturation and proliferation, and fibroblast growth factor activity. • Modulates production of cytokines, chemokines, growth factors, extracellular matrix proteins • Enhances adhesion molecules and receptors IL-2 IL-2R T lymphocytes • T-cell proliferation B lymphocytes • Can stimulate B cells, monocytes, lymphokine- NK cells activated killer cells, and NK cells. IL-4 IL-4R type 1 T lymphocytes • Induces macrophage fusion with IL-4Rγ Mast cells • Role in B-cell activation IL-4R type 2 Eosinophils • Induces expression of class II MHC molecules on with IL-13Rα resting B-cells • Enhances both secretion and cell surface expression of IgE and IgG1 • Regulates expression of low affinity Fc receptor for IgE on (CD23) on both lymphocytes and monocytes • Regulates helper T cell differentiation into Th2 type IL-6 IL-6R with α Monocytes • Pro-inflammatory/anti-wound healing and β chains Macrophages • Stimulates the proliferation and differentiation of Fibroblasts B-lymphocytes Endothelial • Increases neutrophil production. T and B cells • Involved in T cell activation, growth and differentiation • Role in hematopoiesis and the proliferation of progenitors

- 23 - Chapter I

Factor Receptor Cellular Sources Functions IL-8 CXCR1 Monocytes • Proinflammatory chemokine CXCR2 Lymphocytes • Chemotactic activity for neutrophils and also Granulocytes activates neutrophil functions Fibroblasts • Chemotactic factor that attracts basophils and T- Endothelial cells, but not monocytes • Monocyte intracellular calcium influx, respiratory bust, adhesion • Decrease fibroblasts collagen mRNA expression • Proliferation of endothelial cells IL-13 IL-13Rα1 T lymphocytes • Promotes macrophage fusion IL-13Rα2 Mast cells • Synergizes with IL-2 in regulating interferon-γ IL-4Rα NK cells synthesis • Regulates B cell IgE secretion • Modulates Th2 cell development MIP-1β CCR5 Monocytes • Monokine with inflammatory and chemokinetic CCR8 Dendritic cells properties T lymphocytes • Capacity to regulate the trafficking and activation B lymphocytes state of select subgroups of inflammatory cells NK cells (including macrophages, lymphocytes, eosinophils, Neutrophils dendritic cells, and NK cells) Eosinophils • Hematopoietic cell development, lymphocyte Basophils differentiation and trafficking, immune modulation, Endothelial bone remodeling, and wound healing • Chemoattractant/adhesive effects for monocytes, CD4+ lymphocytes, thymocytes, dendritic cells • Recruitment of monocytes and specific subsets of T and B cells to the inflamed site • MIP-1 attracts CD4+, CD8+, and double positive thymocyte subsets, and induces mobilization of intracellular calcium, phosphorylation of protein tyrosine, and activation of the mitogen-activated protein kinase (MAPK) pathway in these cells. These activities suggest that MIP-1 may play a role in trafficking and/or development of lymphoid progenitors in primary lymphoid tissues • Proliferation and activation of CD56+ (NK) cells TNF-α P60 Macrophages • Pro-inflammatory/anti-wound healing P80 Monocytes • Binds to TNFRSF1A/TNFR1 and T cells TNFRSF1B/TNFBR. B cells • Induce cell death of tumor cell lines. Astrocytes • Binding and transmigration of lymphocytes across Fibroblasts endothelial cells Basophils • Induces expression of macrophage MHC antigens NK cells (class I more than class II), IL-1, GM-CSF, M- CSF, cytosolic calcium levels MCP-1 CCR2 Monocyte • Chemokine for monocytes, activated memory Macrophage (CD45RO+) T lymphocytes, and NK cells Fibroblast • Enhances the proliferative response of naïve T cells Neutrophils to anti-CD3 and to B7-1-induced costimulation • Polarize naïve T cells toward Th2 responses when challenged by antigen

- 24 - Chapter I

Factor Receptor Cellular Sources Functions MDC CCR4 Monocytes • Chemotactic for thymocytes, dendritic cells, Macrophages monocytes, IL-2 activated NK cells, Th2 T cells Dendritic cells and CLA+ T cells B cells T cells IL-10 IL-10Rα Monocytes • Inhibited expression of MHC class II antigens, IL-10Rα T cells (naïve and CD54 (ICAM-1), CD80 (B7), and CD86 (B7.2) on memory) monocytes NK cells • Downregulation of these stimulatory or B cells costimulatory molecules significantly affected the T cell-activating capacity of monocyte APCs • Dominant suppressive effects on the production of proinflammatory cytokines by monocytes and neutrophils and downregulates the expression of activating and costimulatory molecules on monocytes and dendritic cells (macrophage deactivating factor) • Inhibits monocyte production of metalloproteinases and increasing TIMP production • Inhibits the synthesis of a number of cytokines, including IFN-γ, IL-2, IL-3, TNF and GM-CSF produced by activated macrophages and by helper T cells. • Strongly inhibited cytokine production and proliferation of T cells and T cell clones activated in the presence of APCs due mostly to its downregulatory effects on APC functions • Inhibition of neutrophil cytokine and chemokine production, generation of superoxide anions, and survival • Enhanced expression of MHC class II antigens and survival of resting mouse B cells • Inhibitory effects on NK cells TGF-β TGF-βR Numerous • Strong suppressor of activation of T cells and of antibody secretion by B cells • Effects on chemotaxis (fibroblasts, lymphocytes, monocytes, macrophages, and neutrophils) • Ability to control cellular differentiation, apoptosis, and extracellular matrix production • Enhanced expression of extracellular matrix proteins, suppression of expression of matrix- degrading proteases, induction of expression of protease inhibitors, and increases expression of integrin receptors • Blocks antibody production by B cells • Depresses activity of NK cells • Inhibits generation of lymphokine activated killer cells and cytotoxic T cells • Inhibits respiratory burst of macrophages • Induces MHC II expression

- 25 - Chapter I

Factor Receptor Cellular Sources Functions Leptin OB-R Adipose • Proinflammatory cytokines increase leptin levels, Placenta whereas leptin regulates the production of several pro- and anti-inflammatory cytokines • In vivo, leptin levels are acutely increased by TNF and IL-1 • In vitro, leptin has been shown to modulate cytokine production by macrophages and T cells PDGF PDGF-Rαα Fibroblast • Potent stimulant of proliferation of mesenchymal PDGF-Rαβ Endothelial cells and stem cells PDGF-Rββ Monocyte • promotion of T cell proliferation Macrophage • Angiogenesis • Promotes proliferation of connective tissue- synthesizing cells and modifying matrix Eotaxin-2 CCR3 Unknown • Chemokine that can also induce inflammatory mediator production including the release of histamine and LTC4 • Chemotaxis of human eosinophils, resting T lymphocytes, basophils GRO CXCR1 Monocytes • Chemokine that shares structural features and many CXCR2 Macrophages biological activities with IL-8 Endothelial • Recruitment and activation of neutrophils, Neutrophils, lymphocytes and monocytes lymphocytes • Chemoattractant for neutrophils, basophils, Fibroblasts eosinophils, monocytes, lymphocytes Others • Role in angiogenesis, wound healing, tumorigenesis, Apoptosis I-309 CCR8 Activated T • Chemotactic for monocytes and Th2 differentiated lymphocytes T cells in vitro Monocytes IP-10 CXCR3 Endothelial • Chemotactic for activated T cells and activated NK Monocytes cells, not resting T cells Macrophages • Acts on CD4+ and CD8+ cells and will T cells preferentially chemoattract Th1 lymphocytes Neutrophils • Induce adhesion of activated T cells to ICAM and Fibroblasts VCAM Others • Antiangiogenic and antitumor activity in vivo NAP-2 CXCR1 Megakaryocytes • Active against endothelium and connective tissue CXCR2 T cells as well as leukocytes and lymphocytes Monocytes • Chemotaxis and activation of Neutrophils Neutrophil • Chemotaxis and activation of Fibroblasts Granulocytes RANTES CCR1 Macrophages • Chemotaxis and activation of T cells, eosinophils, CCR3 Epithelial basophils, monocytes, dendritic cells and neurons CCR4 Endothelial CCR5 ENA-78 CXCR1 Fibroblasts • Involved in neutrophil activation Macrophages • Chemoattractant for neutrophils Monocytes • Promotes neutrophil adhesion Neutrophils Platelets

- 26 - Chapter I

Factor Receptor Cellular Sources Functions PARC unknown Induced in • Chemotactic for both activated (CD3+) T cells and monocytes by nonactivated (CD14–) lymphocytes, but not for TH2 associated monocytes or granulocytes cytokines • May be involved in B cell migration into B cell Dendritic cells follicles in lymph nodes • Attracts naive T lymphocytes toward dendritic cells and activated macrophages in lymph nodes, • Chemotactic activity for naive T cells, CD4+ and CD8+ T cells • Preferentially attracts naïve resting T cells (CD45RA+) • May play a role in both humoral and cell-mediated immunity responses EGF • Proliferation of fibroblasts HGF • Promotes wound closure • Stimulates VEGF for angiogenesis Angiogenin • Promotes angiogenesis IGFBP • Enhances IGF activity • IGF and IGFBP - enhancing reepithelialization and granulation tissue deposition FGF • Angiogenic factor • Stimulates fibroblasts TIMP-1 -- Monocytes • Complexes with metalloproteinases (such as Macrophages collagenases) and irreversibly inactivates them Fibroblast • Mediates erythropoiesis in vitro; but, unlike IL-3, it is species-specific, stimulating the growth and differentiation of only human and murine erythroid progenitors. Known to act on MMP-1, MMP-2, MMP-3, MMP-7, MMP-8, MMP-9, MMP-10, MMP-11, MMP-12, MMP-13 and MMP-16. Does not act on MMP-14. TIMP-2 -- Monocytes • Complexes with metalloproteinases (such as Macrophages collagenases) and irreversibly inactivates them Fibroblast MMP-9 -- Monocytes • Cleaves most if not all of the constituents of the Macrophages extracellular matrix Fibroblasts

Mathematical Modeling

A mathematical model can be developed to describe how a biological system behaves based on mathematical expressions. Mechanisms or processes that influence the system are incorporated as parameters into equations that describe how the system operates. In this way, simulations would then provide quantitative predictions on how

- 27 - Chapter I

certain factors or perturbations can affect the biological system. If we were able to

determine how biomaterial properties influenced or correlated to model parameters, we

would be able to provide specific biomaterial inputs to the model to predict the biological

responses. This capability has not yet been reached due to the limited knowledge of the

biological response to biomaterial surfaces as well as limited experimental data.

Mechanistic models offer another method of analyzing a complex system. For instance, a

mechanistic model could be used to integrate all the important and relevant processes

occurring in a complex biological system and allow us to examine the dependence or importance of certain parameters relative to others. In experimental design, the more complex the experiment (i.e. the more parameters or variables are present), the harder it becomes to isolate the important factors. The models can also be utilized to test hypotheses as well as generate hypotheses for experimental validation.

Many models have been developed such as those describing receptor-ligand dynamics, cellular behaviors, cellular interactions, as well as complex biological systems.

Models relevant to the host response to biomaterials include those that describe cellular adhesion and fusion, cell to cell interactions, and the inflammatory response as a system.

Cellular adhesion was described by Zhu et al., in terms of multiple and concurrent ligand- receptor binding.116 N’Dri et al. utilized ligand-receptor binding in adhesion to substrate

and also modeled the rolling and migration characteristic of leukocyte movement in

response to stimulus.117 Bigerelle et al. developed a kinetic model to describe the

adhesion of osteoblasts onto metallic surfaces with varying compositions, surface chemistries, and topographies.118 This model is relevant to adhesion on biomaterial

- 28 - Chapter I

surfaces but involves a cell type that proliferates rather than adheres and fuses as in the case of macrophages.

Other models of interest describe lymphocyte kinetics in terms of proliferation and cell turnover of subpopulations119, cell to cell communication via autocrine and

paracrine mediated signaling in directing cell growth120, cell to cell interaction through

adhesive molecules121, and dendritic cell and cytotoxic T lymphocyte interactions on

cytotoxic lymphocyte response.122 However, these models are not relevant to the process of foreign body giant cell formation and do not involve lymphocyte interactions with the cells in the foreign body reaction. Additionally, models of the complex acute systemic inflammatory response to stimuli and various stresses such as infection and shock states have been proposed.123-126 Again, these models do not describe the series of events that

occur at biomaterial surfaces.

Models describing the foreign body response are few. Kao et al. and Zhao et al.

modeled the fusion process based on probability and fusion kinetics to describe the

FBGC density and size distribution on biomaterial surfaces in vivo over time.127-129

These models address adhesion and fusion parameters but do not account for other

cellular processes that occur at the biomaterial surface such as apoptosis and detachment.

Recently, our laboratory has established a model for our in vitro monocyte culture system

to describe the process of monocyte adhesion, differentiation to macrophages, and fusion

into foreign body giant cells (i.e. foreign body reaction).130 This model, however, does

not include interactions of lymphocytes with the events of the foreign body reaction.

- 29 - Chapter I

Model Biomaterial Surfaces

The surface chemistry and structure determine the biological response to

biomaterials and devices. Surface modification has been utilized to attempt to reduce the

foreign body reaction while maintaining advantageous bulk properties of a material. In

addition to the foreign body reaction, surface modification of materials can also be

utilized in order to produce the type of cell-material interaction desired such as in the design of tissue-engineered constructs. Polyethylene terephthalate (PET) is a material that has a wide variety of uses in biomedical applications and is commonly utilized as

arterial graft as well as in tissue reconstruction. Its mechanical and physical properties

make it an extremely attractive polymer. However, PET does have its limitations as a

biomaterial. As mentioned previously, surface modification methods can be utilized to

take advantage of its bulk properties while improving the surface chemistry and structure

of PET through increasing hydrophilicity as well as attaching biologically reactive

groups. These techniques are designed to enhance PET biocompatibility and biological

functionality for biomedical applications. Dr. Takehisa Matsuda, Professor of biomedical

engineering and graduate school of medicine at Kyushu University in Fukuoka, Japan,

has provided us with PET-based photograft copolymerized surfaces. These model

surfaces present distinct hydrophobic/hydrophilic properties as well as hydrophilic

materials presenting non-ionic, anionic, and cationic properties. This group of materials

has been used in our laboratory to probe and characterize cellular interactions with

biomaterials. These surfaces have demonstrated biomaterial-dependent macrophage

adhesion, fusion, and mRNA cytokine expression. These various photograft

copolymerized surfaces will be utilized to examine lymphocyte/macrophage interactions

- 30 - Chapter I

in the foreign body reaction and the effect biomaterials have on those interactions.

Moreover, these model surfaces will allow further investigation of the capability of

surfaces to enhance or reduce the foreign body reaction at the material/surface interface

to ultimately determine biological design criteria for the development of new biomaterials or tissue-engineered constructs.

PET (Base Polymer)

Polyethylene terephthalate (PET) is a polyester biomaterial with a structure which produces a crystalline polymer with a high melting point (Tm = 267 °C) and high tensile

strength.131 Figure 1.4 shows the chemical structure of PET. Table 1.4 lists the

mechanical and physical properties of PET. PET has excellent fiber forming properties

and can be fabricated into many different forms such as knit, velour, woven fabrics,

fabric tubes, and nonwoven felts. Its many biomedical applications include utilization as

arterial grafts, fixation of implants, hernia repairs, ligament reconstruction, and other

tissue reconstruction.131 PET, like many polymers, has a wide array of applications

because of their favorable bulk properties with high strength-to-weight ratio, good

corrosion resistance, and relatively low production costs.132 Two common commercially

available forms of PET are Dacron®, produced as a fiber and used as large diameter

arterial grafts, and Mylar®, PET produced as a thin film. In terms of biological

responses, Dacron®, is a polymer that facilitates high and rapid protein adsorption,

relatively high thombogenicity and moderate cell adhesivity.133 The surface chemistry

and structure determine the biological response to biomaterials and devices. Despite

advantageous bulk properties, the unmodified PET surface can be inadequate.

Limitations include insufficient cell adhesion and growth on PET for desired cell

- 31 - Chapter I

attachment on materials as well as blood incompatibility due to being moderately

inflammatory and thrombogenic.133,134 Therefore, surface modification techniques are

useful for preserving the bulk characteristics of the polymer while providing appropriate surface properties for the desired application.

O O

(CH2)2 O C C O

Figure 1.4: Chemical structure of poly(ethylene terephthalate).

Table 1.4: Mechanical and Physical Properties of PET135

Properties PET

Tensile Strength (MPa) 59-72 Tensile Modulus (Gpa) 2.8-4.1 Elongation (%) 50-300 Tg (°C) 69-82 Tm (°C) 265-270 Water Absorption (%) 0.1-0.2 Water Contact Angle (°) 73-78

General Surface Modification Techniques of PET

Modification of PET and other material surfaces can be performed using two

basic methods: (1) chemically or physically changing the surface chemistry or structure

or (2) overcoating the surface with a composition different from the bulk material. There

has been much research into PET surface modifications as well as the biological

responses to these surfaces.

- 32 - Chapter I

Modification of Original Surface

The original PET surface is hydrophobic and absent of functional groups. To

improve the surface, the material can be modified chemically or physically to add

reactive groups (e.g., -OH, -NH2, -COOH, -SH, or –CH=CH2). An example of chemical

modification is plasma treatment which is desirable because it is a clean and dry

technique. This technique utilizes a dissociated gaseous environment with elements such

as ions, free radicals, electrons, atoms, molecules, and photons in order to ablate or etch

the surface thereby modifying the surface with polar functional groups.131 Plasma glow

discharge treatment has been shown to produce carbonyl and carboxyl acid groups on the

material surface.136,137 A hydrophilic surface has been created on PET with plasma

treatment under argon showing excellent biocompatibility.138 Moreover, adhesion enhancement for applications such as for tissue engineered constructs can be achieved using atmospheric plasma treatment.132 Finally, UV and ozone treatment have also been

utilized for the surface modification of PET. The material surface becomes oxidized

upon absorption of UV light and ozone producing surface chemistries such as carboxylic

acid end-groups, terminal vinyl groups, and phenols.139

Noncovalent Coatings

One of the easiest methods of modifying a polymer surface is through

noncovalent interactions such as hydrophobic or electrostatic forces. Hydrophobic

interactions have been used to create surfaces that prevent protein adsorption. This can

be accomplished by altering the hydrophobic surface of the PET material and exposing a

hydrophilic moiety such as carbohydrates or other immunogenically inert molecules. For

instance, polyethylene glycol (PEG), a hydrophilic, biologically inert polymer, has been

- 33 - Chapter I

utilized in amphiphilic diblock and triblock designs to coat the surface in order to

increase PET biocompatibility.140 Physical and electrostatic adsorption is a highly simple

procedure with low cost; however, there is a relatively low loading level with relatively

high leakage of adsorbed molecules.131

Covalent Coatings

Various methods exist for covalent attachments on PET surfaces including radiation and photografting, plasma treatment, chemical grafting, and biological

immobilization. Unlike adsorption methods of modification, covalent binding is a more

complex procedure with high cost; however, this method produces low to no release of bound molecules.131

Radiation Grafting and Photografting

Radiation grafting can be performed using ionizing radiation (γ-radiation) or UV

radiation (photografting). The mechanism of grafting involves the breaking of chemical bonds on the surface of the material leading to free radicals, peroxides, or other reactive

species. These reactive groups are then used to graft desired monomers onto the surface.

Gamma sources typically have energies of approximately 1 MeV whereas UV radiation

sources are low at (<6 eV). The process involves the exposure of the material to the

radiation in the presence of air or oxygen allowing the formation of radicals. Then, the

heating of the material with the monomer initiates the graft polymerization.

UV radiation has been utilized to attach hydrogels to the surface of PET. For

instance, an interpenetrating polymer network (IPN) of poly(acrylamide-co-ethylene

glycol) hydrogel was covalently bonded to the surface of PET in order to increase the

hydrophilicity and lubricity of the material.141 Also, UV photografting of PAAm on PET

- 34 - Chapter I

surfaces with immobilization of biological molecules such as collagen or galactose ligands have shown to increase the adhesion of desired cells for tissue-engineered

constructs.142,143

Plasma Treatment

In addition to the plasma induced oxidation of the material surface, plasma can

also modify a material surface through deposition. The deposition process is not

completely understood but it is believed that the reactive gaseous environment creates

free radicals and other reactive species on the surface and in the gas phase. Reactive

species on the surface can polymerize with monomers or molecules in the gas phase to

create a new material surface. In addition, gas phase polymerization can settle and

deposit onto the surface.131 Plasma polymerization can improve on the oxidation of the

surface provided by plasma treatment alone. The hydrophilicity produced by the polar

functional groups is sometimes short-lived due to instability. Grafting a hydrophilic

group such as acrylic acid through plasma polymerization can stabilize the hydrophilicity of the material surface and provide greater protein immobilization and cellular adhesion.144

Chemical and Biological Grafting

The covalent bonding of molecules can be achieved by chemical reaction with the surface. This can only be achieved if the surface has exposed reactive groups. PET is an inert solid polymer so grafting of chemical groups or biomolecules can only occur after

chemical modification. These alterations can be performed by techniques such as

chemical reactions, plasma, radiation, UV, or ozone treatment, as described above.

Afterwards, many different molecules could be immobilized onto the surface to create a

- 35 - Chapter I

coating. Various biologically functional molecules have been grafted onto PET surfaces

each with specific functions of interest. Examples of attachment on PET include heparin to prevent thrombosis,136 RGD peptides for increased cellular adhesion on materials,145

and PEG for increased biocompatibility.140

Photograft Copolymerized Surfaces

Matsuda and colleagues have developed and reported a novel photograft

copolymerization technique able to control regional surface modification to the

micrometer scale.146-148 Regional control is achieved by immobilization of radical

initiators in the targeted surface regions with subsequent irradiation for copolymerization

of chemistries for desired surface properties. The mechanism occurs via living radical

polymerization. Using this method, sheets of Mylar® PET were photograft

copolymerized to yield distinct hydrophobic, hydrophilic, anionic, and cationic surface

characteristics.

Iniferter

One of the most crucial components of photografting is the use of an initiator for

radical generation after UV exposure. The method developed by Matsuda et al., which

builds off of the radical polymerization method described by Otsu et al., utilizes a

dithiocarbamate derivative, specifically poly(benzyl N,N-diethyldithiocarbamate-co-

styrene) (BDEDTC), as the initiator.147 Figure 1.5 shows the chemical structure of

BDEDTC.

- 36 - Chapter I

CH2 CH CH2 CH lm

CH2 S C N

Figure 1.5: Chemical structure of poly(benzyl N,N-diethyldithiocarbamate-co-styrene) BDEDTC (1:m = 1:1).

These compounds, described as iniferters, not only initiate the radical polymerization, but also act as the transfer and reaction terminating agent.149 The photoreactive polymer is prepared by the process depicted in Figure 1.6.

Cl S

S Reaction 1 + Na S C N

CH2 S C N

Vinylbenzyl N,N-diethyldithiocarbamate Vinylbenzyl chloride Sodium N,N- diethyldithiocarbamate

Initiator

Styrene

CH2 CH CH2 CH lm

CH2 S C N

Poly(benzyl N,N-diethyldithiocarbamate-co-styrene) BDEDTC

Figure 1.6: Preparation of the iniferter. (l:m = 1:1)

- 37 - Chapter I

Reaction 1 is the formation of the photoreactive monomer, vinylbenzy N,N- diethyldithiocarbamate from vinylbenzyl chloride and sodium N,N- diethyldithiocarbamate. The formation of the iniferter, BDEDTC, results from the radical polymerization of styrene to the photoreactive monomer product from Reaction 1.

Iniferter Characterization

Nuclear Magnetic Resonance (1H NMR) was utilized in characterizing the amount

of copolymerization. The spectrum provides the molar percentage of certain constituents such as the content of the photoreactive moiety in the iniferter. The percentage of

photoreactive moieties was dependent on the reaction time. Iniferters with photoreactive

content constituting 25 mol% were initially made to investigate the iniferter technique in architectural design146 and 49 mol% of the final polymer for laboratory large-scale

production.147

Gas Permeation Chromatography (GPC) was used to estimate the molecular

weight of the polymers. BDEDTC which were meta and para in structure were initially

made with molecular weights 51,400146 and then 37,000147 for the final polymer.

Living Radical Polymerization

Living radical polymerization is a type of addition polymerization where chain

termination and transfer are eliminated and the rate of chain initiation dominates the rate

of chain propagation. Through this type of chain growth, the chain lengths, molecular

weights, and rate of polymer growth, and chain end groups can be controlled. The

similar resulting chain lengths mean that the polydispersity is low. The general form of

the polymerization proceeds as described in the following reaction:

R* X—Y + nM X—(M) —Y n

- 38 - Chapter I

Essentially, monomers add in between the X—Y bond of the iniferter. In this way, the

molecular weight as well as the end groups of the resulting polymer can be controlled.

The molecular weight of the polymers produced via this type of polymerization increases with reaction time. Iniferters such as the BDEDTC follow the mechanisms of living

radical polymerization.148-150 These iniferters are considered to proceed by this form of

polymerization due to evidence showing that conversion and viscosity of the forming

polymers increase with reaction time150 as well as polymer mass increasing during UV

irradiation and stabilizing during withdrawal of UV.148,151 The polymerization is induced

only during times of photo-irradiation. Initiators capable of promoting this form of addition are weak; bimolecular termination is negligible and termination occurs via the primary radical.149

Reaction Process

The iniferters described above are immobilized on the surface by coating onto

PET films. Regional control can be achieved by selective coating as well as selective irradiation. To produce a homogeneously modified surface, the entire sheet of PET film is coated. The general photograft copolymerization reaction on the surface of a sheet of

PET is described in Figure 1.7.147

The polymer film is placed in a reaction cell in the presence of the monomer

solution in a nitrogen atmosphere. The BDEDTC iniferter becomes reactive with

exposure to UV forming a benzyl radical along with a dithiocarbamyl radical. The

benzyl radical initiates polymerization with the vinyl monomers as shown in Figure 1.7.

As described above in living radical polymerization, the reaction proceeds as long as

irradiation continues. Once the desired chain length and molecular weight is achieved,

- 39 - Chapter I

UV exposure ceases. The polymerization stops with the dithiocarbamyl radical from the iniferter end capping the resulting photografted copolymer. The treated film is subsequently rinsed to remove all unwanted monomers and polymers and then dried

under vacuum. The specific vinyl monomers used to produce the surfaces displaying

distinct surface characteristics are: acrylamide (AAm), sodium salt of acrylic acid

(AANa), and N-methiodide of dimethylamino propylacrylamide (DMAPAAmMeI). The

structures are shown in Figure 1.8. Finally, Figure 1.9 displays the chemistries of the 5

various modified surfaces. PET provides the base material. BDEDTC coating gives a

hydrophobic material while photograft copolymerization with PAAm, PAANa, and

DMAPAAmMeI provides surfaces with hydrophilic, hydrophilic/anionic, and

hydrophilic/cationic properties, respectively.

- 40 - Chapter I

Figure 1.7: General photograft copolymerization reaction.

O O O I-- + - + N N NH2 O Na

H Acrylamide Sodium Salt of Methiodide of Acrylic Acid Dimethylaminopropyl-acrylamide

Figure 1.8: Vinyl monomers in photograft copolymers.

- 41 - Chapter I

S BDEDTC CH2 S C N

CH CH CH CH Coating 2 2

Mylar Film PET

NH PAAm 2 O C S

CH2 C C S C N n

CH CH CH CH Coating 2 2

Mylar Film PET

PAANa O- Na+

O C S

CH2 C C S C N n

CH CH CH CH Coating 2 2

Mylar Film PET

- + I N(CH3)3

DMAPAAmMeI C

N O C S

CH2 C C S C N n

CH CH CH CH Coating 2 2

Mylar Film PET

Figure 1.9: Final PET-based photograft copolymerized surfaces.

- 42 - Chapter I

Characterization of Photograft Co-polymerized Surfaces

Modified PET-based surfaces were characterized via attenuated total reflectance

Fourier transform infrared analysis (ATR-FTIR), contact angle, and surface staining in

order to confirm surface chemistries as well as surface properties as previously

described.55

Attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectral

analysis were performed utilizing a Nexus 870 FTIR bench coupled to a Continuum

microscope (Thermo Nicolet, Madison, WI) with an attenuated total reflectance (ATR)

slide-on attachment containing a germanium crystal. Spectra resolution of 2.0 cm-1 was obtained for 32 scans over an area of 150 μm x 150 μm. Analysis of three samples of each material over five distinct areas was conducted to investigate sample and surface uniformity. The spectra for PET contained the following peaks along with corresponding bonds: 1715 cm-1 (C=O stretch), 1400-1600 cm-1 (C=C stretch), 1248 cm-1 (C-O stretch),

1100-1125 cm-1 (C-O stretch), 790-1023 cm-1 (C-H bend), and 724 cm-1 (C-H out-of-

plane bend). In addition to the PET peaks, the spectra for BDEDTC surfaces included

3026 cm-1 (aromatic C-H stretch), 1516 & 1410 cm-1 (aromatic C=C stretch), 1210 cm-1

(C-N stretch), and 700 cm-1 (C-H rock). The PAAm surface showed peaks corresponding

to the presence of the amide group: 3361 cm-1 and 3189 cm-1 (N-H stretch with two peaks for an unsubstituted amide), 1681 cm-1 (C=O stretch), and 1650 cm-1 (N-H bend). The

PAANa and DMAPAAmMeI surface showed similar spectra to PET. The polyacrylic

acid groups of PAANa do not contribute any additional functional groups to those found

in PET. The DMAPAAmMeI surface, however, should include peaks indicating the

presence of the amide group.

- 43 - Chapter I

Advancing water contact angles were measured by sessile drop method utilizing a

goniometer (Edmund Scientific, Barrington, NJ) at room temperature (22˚C). Contact

angles allowed determination of hydrophilicity/hydrophilicity. The measured contact angles were determined to be 71.9˚ for the unmodified PET, 76.5˚ for the BDEDTC- coated surface, and 37.2˚, 20.7˚, 16.4˚ for the PAAm, PAANa, and DMAPAAmMeI photograft copolymerized surfaces, respectively as shown in Table 1.5. The contact angles for PET and BDEDTC confirmed their hydrophobic nature while the contact angles for PAAm, PAANa, and DMAPAAmMeI indicated their hydrophilicity.

Table 1.5: Composition and Contact Angles for PET-based Surfaces

Surface Photograft Advancing Property Base Coating chemistry Copolymer Contact Angle (˚) PET Hydrophobic/neutral PET ------71.9 ± 1.2

BDEDTC Hydrophobic/neutral PET BDEDTC --- 76.5 ± 2.9

PAAm Hydrophilic/neutral PET BDEDTC PAAm 37.2 ± 1.8

PAANa Hydrophilic/anionic PET BDEDTC PAANa 20.7 ± 4.0

DMAPAAmMeI Hydrophilic/cationic PET BDEDTC DMAPAAmMeI 16.4 ± 1.6

PET, polyethylene terephthalate; BDEDTC, poly(benzyl N,N-diethydithiocarbamate-co-styrene); PAAm Polyacrylamide; PAANa, sodium salt of polyacrylic acid; DMAPAAmMeI, methiodide of poly(dimethylaminopropylacrylamide)

All surfaces were incubated at room temperature with 1 milliliter of 1.0 w/v%

aqueous solutions of toluidine blue and rose bengal (Sigma-Aldrich, St. Louis, MO) for 4

hours, washed with distilled, deionized water, and dried overnight prior to visualization.

Toluidine blue is a basic dye that reacts with negatively-charged ions while rose bengal,

an acidic dye, reacts with positively-charged ions. The anionic PAANa surfaces were

positively stained by toluidine blue but negative for rose bengal staining while cationic

- 44 - Chapter I

DMAPAAmMeI surfaces were stained positively by immersion in rose bengal but negative for toluidine blue staining (Figure 1.10). All other surfaces were negative for

surface staining after the 4 hour exposure to either rose bengal or toluidine blue which

confirmed their neutral properties.

Figure 1.10: Chemical staining of biomaterial surfaces by toluidine blue and rose bengal for confirmation of charged surfaces.

Significance

The design of novel biomaterials, devices, tissue-engineered constructs,

prostheses necessitates an understanding of the host-mediated response to biomaterials.

The inflammatory, wound healing, and foreign body responses can all act to determine

whether an implant such as a tissue-engineered scaffold will fail or be able to perform its

intended functions. Numerous studies have focused on macrophages which play an integral part in the foreign body reaction and guiding the host response to an implant.

Lymphocytes transiently appear at implant sites, can become reactive as shown in clinical applications of biomaterial implantations, and play a role in influencing macrophage

- 45 - Chapter I

behavior in vitro. These findings prompt further investigation of lymphocyte interactions with macrophages as we currently do not have a clear mechanistic understanding.

Many studies have shown that biomaterial surface chemistries can influence cellular responses. Gaining a mechanistic understanding of these observations is vital to ensure the proper application of biomaterials as an enabling technology. There are many different ways biomaterials can be utilized. Some applications require minimization of the foreign body reaction and subsequent formation of fibrous capsule (e.g. prosthetics, drug delivery devices, etc); however some applications may require the actions of macrophages and the foreign body reaction (e.g. biodegradable systems). Therefore,

model photograft copolymerized surfaces displaying hydrophobic, hydrophilic, anionic,

and cationic chemistries were utilized to probe the lymphocyte response and interactions

and to gain insight into how material surface chemistries can modulate the lymphocyte

behavior. The findings contribute to developing biological design criteria for the design

of new materials or constructs. A mathematical model describing the in vitro

lymphocyte/monocyte direct co-culture system provides a step towards quantitative

predictability of behavior and interactions of lymphocyte and macrophage in response to

biomaterial surfaces.

Hypothesis and Specific Aims

The overall hypothesis is that molecularly-engineered surfaces displaying distinct

surface chemistries modulate lymphocyte paracrine and juxtacrine interactions with

macrophages and foreign body giant cells. The hypothesis is addressed by the following

research aims:

- 46 - Chapter I

Specific Aim 1: To investigate quantitatively the effect of biomaterial surface

chemistries on cytokine, chemokines, and production from lymphocytes and

macrophages in direct co-culture.

Specific Aim 2: To determine how lymphocytes influence the production of

macrophage-derived inflammatory mediators in response to biomaterial surfaces

and evaluate how material surface chemistries can influence the response

Specific Aim 3: To evaluate quantitatively the adherence of lymphocytes on biomaterial

surfaces and characterize the effect of adherent macrophages and foreign body

giant cells on lymphocyte adhesion behavior

Specific Aim 4: To develop a mechanistic model for quantitative predictability of

lymphocyte, monocytes, macrophage, and foreign body giant cell behavior and

interactions on biomaterial surfaces

References

1. Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81-110.

2. Anderson JM, Rodriguez A, Chang DT. Foreign body reaction to biomaterials. Semin Immunol 2008;20(2):86-100.

3. Zdolsek J, Eaton JW, Tang L. Histamine release and fibrinogen adsorption mediate acute inflammatory responses to biomaterial implants in humans. J Transl Med 2007;5:31.

4. Tang L, Jennings TA, Eaton JW. Mast cells mediate acute inflammatory responses to implanted biomaterials. Proc Natl Acad Sci U S A 1998;95(15):8841-6.

5. Keegan AD. IL-4. In: Oppenheim JJ, Feldman M, editors. Cytokine Reference. San Diego, CA: Academic Press; 2001.

6. McKenzie ANJ, Matthews DJ. IL-13. In: Oppenheim JJ, Feldman M, editors. Cytokine Reference. San Diego, CA: Academic Press; 2001.

- 47 - Chapter I

7. Zhao Q, Topham N, Anderson JM, Lodoen G, Payet CR. Foreign-body giant cells and polyurethane biostability: in vivo correlation of cell adhesion and surface cracking. J Biomed Mater Res 1991;25(2):177-83.

8. Zhao QH, McNally AK, Rubin KR, Renier M, Wu Y, Rose-Caprara V, Anderson JM, Hiltner A, Urbanski P, Stokes K. Human plasma alpha 2-macroglobulin promotes in vitro oxidative stress cracking of Pellethane 2363-80A: in vivo and in vitro correlations. J Biomed Mater Res 1993;27(3):379-88.

9. Henson PM. The immunologic release of constituents from neutrophil leukocytes. II. Mechanisms of release during phagocytosis, and adherence to nonphagocytosable surfaces. J Immunol 1971;107(6):1547-57.

10. Henson PM. The immunologic release of constituents from neutrophil leukocytes. I. The role of antibody and complement on nonphagocytosable surfaces or phagocytosable particles. J Immunol 1971;107(6):1535-46.

11. Vroman L. The life of an artificial device in contact with blood: initial events and their effect on its final state. Bull N Y Acad Med 1988;64(4):352-7.

12. Eskin S, Horbett T, McIntire L, Mitchell R, Ratner B, Schoen F, Yee A. Some Background Concepts. In: Ratner BD, Hoffman AS, Schoen FJ, Lemons JE, editors. Biomaterial Sciences: An Introduction to Materials in Medicine. San Diego: Elsevier Academic Press; 2004. p 237-291.

13. Ziats NP, Pankowsky DA, Tierney BP, Ratnoff OD, Anderson JM. Adsorption of Hageman factor (factor XII) and other human plasma proteins to biomedical polymers. J Lab Clin Med 1990;116(5):687-96.

14. Horbett TA. Principles underlying the role of adsorbed plasma proteins in blood interactions with foreign materials. Cardiovascular Pathology 1999;2(3 (Suppl.)):137S-148S.

15. Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of biomaterial-cell interactions by adsorbed proteins: a review. Tissue Eng 2005;11(1-2):1-18.

16. Xu LC, Siedlecki CA. Effects of surface wettability and contact time on protein adhesion to biomaterial surfaces. Biomaterials 2007;28(22):3273-83.

17. Andersson J, Ekdahl KN, Lambris JD, Nilsson B. Binding of C3 fragments on top of adsorbed plasma proteins during complement activation on a model biomaterial surface. Biomaterials 2005;26(13):1477-85.

18. Gorbet MB, Sefton MV. Biomaterial-associated thrombosis: roles of coagulation factors, complement, platelets and leukocytes. Biomaterials 2004;25(26):5681- 703.

- 48 - Chapter I

19. Allen LT, Tosetto M, Miller IS, O'Connor DP, Penney SC, Lynch I, Keenan AK, Pennington SR, Dawson KA, Gallagher WM. Surface-induced changes in protein adsorption and implications for cellular phenotypic responses to surface interaction. Biomaterials 2006;27(16):3096-108.

20. Brevig T, Holst B, Ademovic Z, Rozlosnik N, Rohrmann JH, Larsen NB, Hansen OC, Kingshott P. The recognition of adsorbed and denatured proteins of different topographies by beta2 integrins and effects on leukocyte adhesion and activation. Biomaterials 2005;26(16):3039-53.

21. Brodbeck WG, Colton E, Anderson JM. Effects of adsorbed heat labile serum proteins and fibrinogen on adhesion and apoptosis of monocytes/macrophages on biomaterials. J Mater Sci Mater Med 2003;14(8):671-5.

22. Collier TO, Anderson JM. Protein and surface effects on monocyte and macrophage adhesion, maturation, and survival. J Biomed Mater Res 2002;60(3):487-96.

23. Jenney CR, Anderson JM. Adsorbed serum proteins responsible for surface dependent human macrophage behavior. J Biomed Mater Res 2000;49(4):435-47.

24. Lee MH, Ducheyne P, Lynch L, Boettiger D, Composto RJ. Effect of biomaterial surface properties on fibronectin-alpha5beta1 integrin interaction and cellular attachment. Biomaterials 2006;27(9):1907-16.

25. Keselowsky BG, Collard DM, Garcia AJ. Surface chemistry modulates focal adhesion composition and signaling through changes in integrin binding. Biomaterials 2004;25(28):5947-54.

26. Keselowsky BG, Collard DM, Garcia AJ. Integrin binding specificity regulates biomaterial surface chemistry effects on cell differentiation. Proc Natl Acad Sci U S A 2005;102(17):5953-7.

27. Hu WJ, Eaton JW, Ugarova TP, Tang L. Molecular basis of biomaterial-mediated foreign body reactions. Blood 2001;98(4):1231-8.

28. Shen M, Garcia I, Maier RV, Horbett TA. Effects of adsorbed proteins and surface chemistry on foreign body giant cell formation, tumor necrosis factor alpha release and procoagulant activity of monocytes. J Biomed Mater Res A 2004;70(4):533-41.

29. Delon I, Brown NH. Integrins and the actin cytoskeleton. Curr Opin Cell Biol 2007;19(1):43-50.

30. Giancotti FG, Ruoslahti E. Integrin signaling. Science 1999;285(5430):1028-32.

31. Berton G, Lowell CA. Integrin signalling in neutrophils and macrophages. Cell Signal 1999;11(9):621-35.

- 49 - Chapter I

32. Janeway CA, Travers P, Walport M, Shlomchik M. Immunobiology: The Immune System in Health and Disease. New York: Garland Publishing; 2001. 1-708 p.

33. Garcia AJ. Get a grip: integrins in cell-biomaterial interactions. Biomaterials 2005;26(36):7525-9.

34. McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 1994;91(21):10119-23.

35. Nilsson B, Ekdahl KN, Mollnes TE, Lambris JD. The role of complement in biomaterial-induced inflammation. Mol Immunol 2007;44(1-3):82-94.

36. McNally AK, Anderson JM. Beta1 and beta2 integrins mediate adhesion during macrophage fusion and multinucleated foreign body giant cell formation. Am J Pathol 2002;160(2):621-30.

37. Kao WJ, Lee D, Schense JC, Hubbell JA. Fibronectin modulates macrophage adhesion and FBGC formation: the role of RGD, PHSRN, and PRRARV domains. J Biomed Mater Res 2001;55(1):79-88.

38. McNally AK, Macewan SR, Anderson JM. alpha subunit partners to beta1 and beta2 integrins during IL-4-induced foreign body giant cell formation. J Biomed Mater Res A 2007;82(3):568-74.

39. Kim JK, Scott EA, Elbert DL. Proteomic analysis of protein adsorption: serum amyloid P adsorbs to materials and promotes leukocyte adhesion. J Biomed Mater Res A 2005;75(1):199-209.

40. Yokoyama M, Nakahashi T, Nishimura T, Maeda M, Inoue S, Kataoka K, Sakurai Y. Adhesion behavior of rat lymphocytes to poly(ether)-poly(amino acid) block and graft copolymers. J Biomed Mater Res 1986;20(7):867-78.

41. Groth T, Altankov G, Klosz K. Adhesion of human peripheral blood lymphocytes is dependent on surface wettability and protein preadsorption. Biomaterials 1994;15(6):423-8.

42. Groth T, Klosz K, Campbell EJ, New RR, Hall B, Goering H. Protein adsorption, lymphocyte adhesion and platelet adhesion/activation on polyurethane ureas is related to hard segment content and composition. J Biomater Sci Polym Ed 1994;6(6):497-510.

43. Maeda M, Kimura M, Inoue S, Kataoka K, Okano T, Sakurai Y. Adhesion behavior of rat lymphocyte subpopulations (B cell and T cell) on the surface of polystyrene/polypeptide graft copolymer. J Biomed Mater Res 1986;20(1):25-35.

44. Brodbeck WG, Macewan M, Colton E, Meyerson H, Anderson JM. Lymphocytes and the foreign body response: lymphocyte enhancement of macrophage adhesion and fusion. J Biomed Mater Res A 2005;74(2):222-9.

- 50 - Chapter I

45. Groth T, Zlatanov I, Altankov G. Adhesion of human peripheral lymphocytes on biomaterials preadsorbed with fibronectin and vitronectin. J Biomater Sci Polym Ed 1994;6(8):729-39.

46. Wagner C, Hansch GM. Receptors for complement C3 on T-lymphocytes: relics of evolution or functional molecules? Mol Immunol 2006;43(1-2):22-30.

47. Wagner C, Hansch GM, Stegmaier S, Denefleh B, Hug F, Schoels M. The complement receptor 3, CR3 (CD11b/CD18), on T lymphocytes: activation- dependent up-regulation and regulatory function. Eur J Immunol 2001;31(4):1173-80.

48. Esche C, Stellato C, Beck LA. Chemokines: key players in innate and adaptive immunity. J Invest Dermatol 2005;125(4):615-28.

49. Campbell DJ, Kim CH, Butcher EC. Chemokines in the systemic organization of immunity. Immunol Rev 2003;195:58-71.

50. Gerard C, Rollins BJ. Chemokines and disease. Nat Immunol 2001;2(2):108-15.

51. Broughton G, 2nd, Janis JE, Attinger CE. The basic science of wound healing. Plast Reconstr Surg 2006;117(7 Suppl):12S-34S.

52. Rhodes NP, Hunt JA, Williams DF. Macrophage subpopulation differentiation by stimulation with biomaterials. J Biomed Mater Res 1997;37(4):481-8.

53. Charo IF, Ransohoff RM. The many roles of chemokines and chemokine receptors in inflammation. N Engl J Med 2006;354(6):610-21.

54. Kyriakides TR, Foster MJ, Keeney GE, Tsai A, Giachelli CM, Clark-Lewis I, Rollins BJ, Bornstein P. The CC chemokine ligand, CCL2/MCP1, participates in macrophage fusion and foreign body giant cell formation. Am J Pathol 2004;165(6):2157-66.

55. Jones JA, Chang DT, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Proteomic analysis and quantification of cytokines and chemokines from biomaterial surface-adherent macrophages and foreign body giant cells. J Biomed Mater Res A 2007;83(3):585-96.

56. Brodbeck WG, Shive MS, Colton E, Nakayama Y, Matsuda T, Anderson JM. Influence of biomaterial surface chemistry on the apoptosis of adherent cells. J Biomed Mater Res 2001;55(4):661-8.

57. Brodbeck WG, Patel J, Voskerician G, Christenson E, Shive MS, Nakayama Y, Matsuda T, Ziats NP, Anderson JM. Biomaterial adherent macrophage apoptosis is increased by hydrophilic and anionic substrates in vivo. Proc Natl Acad Sci U S A 2002;99(16):10287-92.

- 51 - Chapter I

58. Jones JA, Dadsetan M, Collier TO, Ebert M, Stokes KS, Ward RS, Hiltner PA, Anderson JM. Macrophage behavior on surface-modified polyurethanes. J Biomater Sci Polym Ed 2004;15(5):567-84.

59. Damsky CH, Ilic D. Integrin signaling: it's where the action is. Curr Opin Cell Biol 2002;14(5):594-602.

60. Reddig PJ, Juliano RL. Clinging to life: cell to matrix adhesion and cell survival. Cancer Metastasis Rev 2005;24(3):425-39.

61. Frisch SM, Screaton RA. Anoikis mechanisms. Curr Opin Cell Biol 2001;13(5):555-62.

62. Shive MS, Brodbeck WG, Anderson JM. Activation of caspase 3 during shear stress-induced neutrophil apoptosis on biomaterials. J Biomed Mater Res 2002;62(2):163-8.

63. DeFife KM, Jenney CR, McNally AK, Colton E, Anderson JM. Interleukin-13 induces human monocyte/macrophage fusion and macrophage mannose receptor expression. J Immunol 1997;158(7):3385-90.

64. McNally AK, Anderson JM. Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 1995;147(5):1487-99.

65. Kao WJ, McNally AK, Hiltner A, Anderson JM. Role for interleukin-4 in foreign- body giant cell formation on a poly(etherurethane urea) in vivo. J Biomed Mater Res 1995;29(10):1267-75.

66. McNally AK, DeFife KM, Anderson JM. Interleukin-4-induced macrophage fusion is prevented by inhibitors of mannose receptor activity. Am J Pathol 1996;149(3):975-85.

67. Yagi M, Miyamoto T, Sawatani Y, Iwamoto K, Hosogane N, Fujita N, Morita K, Ninomiya K, Suzuki T, Miyamoto K and others. DC-STAMP is essential for cell- cell fusion in osteoclasts and foreign body giant cells. J Exp Med 2005;202(3):345-51.

68. McNally AK, Jones JA, Macewan SR, Colton E, Anderson JM. Vitronectin is a critical protein adhesion substrate for IL-4-induced foreign body giant cell formation. J Biomed Mater Res A 2007.

69. Keselowsky BG, Bridges AW, Burns KL, Tate CC, Babensee JE, LaPlaca MC, Garcia AJ. Role of plasma fibronectin in the foreign body response to biomaterials. Biomaterials 2007;28(25):3626-31.

- 52 - Chapter I

70. Athanasou NA, Quinn J. Immunophenotypic differences between osteoclasts and macrophage polykaryons: immunohistological distinction and implications for osteoclast ontogeny and function. J Clin Pathol 1990;43(12):997-1003.

71. Doussis IA, Puddle B, Athanasou NA. Immunophenotype of multinucleated and mononuclear cells in giant cell lesions of bone and soft tissue. J Clin Pathol 1992;45(5):398-404.

72. Al-Saffar N, Revell PA, Kobayashi A. Modulation of the phenotypic and functional properties of phagocytic macrophages by wear particles from orthopaedic implants. J Mater Sci Mater Med 1997;8(11):641-8.

73. Abbondanzo SL, Young VL, Wei MQ, Miller FW. Silicone gel-filled breast and testicular implant capsules: a histologic and immunophenotypic study. Mod Pathol 1999;12(7):706-13.

74. Quinn JM, Athanasou NA, McGee JO. Extracellular matrix receptor and platelet antigens on osteoclasts and foreign body giant cells. Histochemistry 1991;96(2):169-76.

75. Kadoya Y, al-Saffar N, Kobayashi A, Revell PA. The expression of osteoclast markers on foreign body giant cells. Bone Miner 1994;27(2):85-96.

76. Neale SD, Athanasou NA. Cytokine receptor profile of arthroplasty macrophages, foreign body giant cells and mature osteoclasts. Acta Orthop Scand 1999;70(5):452-8.

77. Hernandez-Pando R, Bornstein QL, Aguilar Leon D, Orozco EH, Madrigal VK, Martinez Cordero E. Inflammatory cytokine production by immunological and foreign body multinucleated giant cells. Immunology 2000;100(3):352-8.

78. Anderson JM. Multinucleated giant cells. Curr Opin Hematol 2000;7(1):40-7.

79. Smetana K, Jr. Multinucleate foreign-body giant cell formation. Exp Mol Pathol 1987;46(3):258-65.

80. Murch AR, Grounds MD, Marshall CA, Papadimitriou JM. Direct evidence that inflammatory multinucleate giant cells form by fusion. J Pathol 1982;137(3):177- 80.

81. Anderson JM. Inflammatory reaction: the nemesis of implants. In: Zilla P, Greisler H, editors. Tissue Engineering of Vascular Prosthetic Grafts. Austin: R.G. Landes Company; 1999. p 197-206.

82. Monaco C, Andreakos E, Kiriakidis S, Feldmann M, Paleolog E. T-cell-mediated signalling in immune, inflammatory and angiogenic processes: the cascade of events leading to inflammatory diseases. Curr Drug Targets Inflamm Allergy 2004;3(1):35-42.

- 53 - Chapter I

83. Hallett WH, Murphy WJ. Positive and negative regulation of Natural Killer cells: therapeutic implications. Semin Cancer Biol 2006;16(5):367-82.

84. Bergman AJ, Zygourakis K. Migration of lymphocytes on fibronectin-coated surfaces: temporal evolution of migratory parameters. Biomaterials 1999;20(23- 24):2235-44.

85. MacEwan MR, Brodbeck WG, Matsuda T, Anderson JM. Student Research Award in the Undergraduate Degree Candidate category, 30th Annual Meeting of the Society for Biomaterials, Memphis, Tennessee, April 27-30, 2005. Monocyte/lymphocyte interactions and the foreign body response: in vitro effects of biomaterial surface chemistry. J Biomed Mater Res A 2005;74(3):285-93.

86. Rodriguez A. T Cell Interactions in the Foreign Body Response to Biomaterials. Dissertation. Case Western Reserve University, 2007.

87. Ankersmit HJ, Edwards NM, Schuster M, John R, Kocher A, Rose EA, Oz M, Itescu S. Quantitative changes in T-cell populations after left ventricular assist device implantation: relationship to T-cell apoptosis and soluble CD95. Circulation 1999;100(19 Suppl):II211-5.

88. Schuster M, Kocher A, John R, Hoffman M, Ankersmit J, Lietz K, Edwards N, Oz M, Itescu S. B-cell activation and allosensitization after left ventricular assist device implantation is due to T-cell activation and CD40 ligand expression. Hum Immunol 2002;63(3):211-20.

89. Itescu S, John R. Interactions between the recipient immune system and the left ventricular assist device surface: immunological and clinical implications. Ann Thorac Surg 2003;75(6 Suppl):S58-65.

90. Katzin WE, Feng LJ, Abbuhl M, Klein MA. Phenotype of lymphocytes associated with the inflammatory reaction to silicone gel breast implants. Clin Diagn Lab Immunol 1996;3(2):156-61.

91. Bender BS, Curtis JL, Nagel JE, Chrest FJ, Kraus ES, Briefel GR, Adler WH. Analysis of immune status of hemodialyzed adults: association with prior transfusions. Kidney Int 1984;26(4):436-43.

92. Grooteman MP, Nube MJ, van Limbeek J, Schoorl M, van Houte AJ. Lymphocyte subsets in dialyser eluates: a new parameter of bioincompatibility? Nephrol Dial Transplant 1996;11(6):1073-8.

93. Raij L, Kay NE. Effect of hemodialysis membranes on human lymphocyte natural killer function. Contrib Nephrol 1987;59:17-25.

94. Jurkowski MK, Bobek-Billewicz B, Cwiklinska-Jurkowska M, Jurkowski P. Effect on peripheral blood natural killer cytotoxic cell activity in rats after

- 54 - Chapter I

intraperitoneal implantation of double veloured polyester (Dacron) prosthesis. Arch Immunol Ther Exp (Warsz) 2004;52(2):129-36.

95. Goodman SB. Wear particles, periprosthetic osteolysis and the immune system. Biomaterials 2007;28(34):5044-8.

96. Miller KM, Anderson JM. Human monocyte/macrophage activation and interleukin 1 generation by biomedical polymers. J Biomed Mater Res 1988;22(8):713-31.

97. Hwang JJ, Jelacic S, Samuel NT, Maier RV, Campbell CT, Castner DG, Hoffman AS, Stayton PS. Monocyte activation on polyelectrolyte multilayers. J Biomater Sci Polym Ed 2005;16(2):237-51.

98. Pueyo ME, Darquy S, Capron F, Reach G. In vitro activation of human macrophages by alginate-polylysine microcapsules. J Biomater Sci Polym Ed 1993;5(3):197-203.

99. DeFife KM, Yun JK, Azeez A, Stack S, Ishihara K, Nakabayashi N, Colton E, Anderson JM. Adhesion and cytokine production by monocytes on poly(2- methacryloyloxyethyl phosphorylcholine-co-alkyl methacrylate)-coated polymers. J Biomed Mater Res 1995;29(4):431-9.

100. Daniels AU, Barnes FH, Charlebois SJ, Smith RA. Macrophage cytokine response to particles and lipopolysaccharide in vitro. J Biomed Mater Res 2000;49(4):469-78.

101. Granchi D, Ciapetti G, Filippini F, Stea S, Cenni E, Pizzoferrato A, Toni A. In vitro cytokine production by mononuclear cells exposed to bone cement extracts. Biomaterials 2000;21(17):1789-95.

102. Ma N, Petit A, Yahia L, Huk OL, Tabrizian M. Cytotoxic reaction and TNF-alpha response of macrophages to polyurethane particles. J Biomater Sci Polym Ed 2002;13(3):257-72.

103. Sethi RK, Neavyn MJ, Rubash HE, Shanbhag AS. Macrophage response to cross- linked and conventional UHMWPE. Biomaterials 2003;24(15):2561-73.

104. Wagner VE, Bryers JD. Monocyte/macrophage interactions with base and linear- and star-like PEG-modified PEG-poly(acrylic acid) co-polymers. J Biomed Mater Res A 2003;66(1):62-78.

105. Chellat F, Grandjean-Laquerriere A, Le Naour R, Fernandes J, Yahia L, Guenounou M, Laurent-Maquin D. Metalloproteinase and cytokine production by THP-1 macrophages following exposure to chitosan-DNA nanoparticles. Biomaterials 2005;26(9):961-70.

- 55 - Chapter I

106. Epstein NJ, Bragg WE, Ma T, Spanogle J, Smith RL, Goodman SB. UHMWPE wear debris upregulates mononuclear cell proinflammatory gene expression in a novel murine model of intramedullary particle disease. Acta Orthop 2005;76(3):412-20.

107. Refai AK, Textor M, Brunette DM, Waterfield JD. Effect of titanium surface topography on macrophage activation and secretion of proinflammatory cytokines and chemokines. J Biomed Mater Res A 2004;70(2):194-205.

108. Brodbeck WG, Nakayama Y, Matsuda T, Colton E, Ziats NP, Anderson JM. Biomaterial surface chemistry dictates adherent monocyte/macophage cytokine expression in vitro. Cytokine 2002;18(6):311-19.

109. Brodbeck WG, Voskerician G, Ziats NP, Nakayama Y, Matsuda T, Anderson JM. In vivo leukocyte cytokine mRNA responses to biomaterials are dependent on surface chemistry. J Biomed Mater Res A 2003;64(2):320-9.

110. Jones JA, McNally AK, Chang DT, Qin LA, Meyerson H, Colton E, Kwon IL, Matsuda T, Anderson JM. Matrix metalloproteinases and their inhibitors in the foreign body reaction on biomaterials. J Biomed Mater Res A 2008;84(1):158-66.

111. Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 2004;25(12):677-86.

112. Santana MA, Rosenstein Y. What it takes to become an effector T cell: the process, the cells involved, and the mechanisms. J Cell Physiol 2003;195(3):392- 401.

113. Mosmann TR, Li L, Sad S. Functions of CD8 T-cell subsets secreting different cytokine patterns. Semin Immunol 1997;9(2):87-92.

114. Tsai AT, Rice J, Scatena M, Liaw L, Ratner BD, Giachelli CM. The role of osteopontin in foreign body giant cell formation. Biomaterials 2005;26(29):5835- 43.

115. Oppenheim J, Feldman M, Durum S, Hirano T, Vilcek J, Nicola N, editors. Cytokine reference: a compendium of cytokines and other mediators of host defense. Volume 1. San Diego: Academic Press; 2001. 2000 p.

116. Zhu C, Williams TE. Modeling concurrent binding of multiple molecular species in cell adhesion. Biophys J 2000;79(4):1850-7.

117. N'Dri NA, Shyy W, Tran-Son-Tay R. Computational modeling of cell adhesion and movement using a continuum-kinetics approach. Biophys J 2003;85(4):2273- 86.

- 56 - Chapter I

118. Bigerelle M, Anselme K. A kinetic approach to osteoblast adhesion on biomaterial surface. J Biomed Mater Res A 2005;75(3):530-40.

119. Macallan DC, Asquith B, Irvine AJ, Wallace DL, Worth A, Ghattas H, Zhang Y, Griffin GE, Tough DF, Beverley PC. Measurement and modeling of human T cell kinetics. Eur J Immunol 2003;33(8):2316-26.

120. Bajzer Z, Vuk-Pavlovic S. Modeling positive regulatory feedbacks in cell-cell interactions. Biosystems 2005;80(1):1-10.

121. Simon SI, Chambers JD, Sklar LA. Flow cytometric analysis and modeling of cell-cell adhesive interactions: the neutrophil as a model. J Cell Biol 1990;111(6 Pt 1):2747-56.

122. Ludewig B, Krebs P, Junt T, Metters H, Ford NJ, Anderson RM, Bocharov G. Determining control parameters for dendritic cell-cytotoxic T lymphocyte interaction. Eur J Immunol 2004;34(9):2407-18.

123. Vodovotz Y, Clermont G, Chow C, An G. Mathematical models of the acute inflammatory response. Curr Opin Crit Care 2004;10(5):383-90.

124. Kumar R, Clermont G, Vodovotz Y, Chow CC. The dynamics of acute inflammation. J Theor Biol 2004;230(2):145-55.

125. Chow CC, Clermont G, Kumar R, Lagoa C, Tawadrous Z, Gallo D, Betten B, Bartels J, Constantine G, Fink MP and others. The acute inflammatory response in diverse shock states. Shock 2005;24(1):74-84.

126. Vodovotz Y. Deciphering the complexity of acute inflammation using mathematical models. Immunol Res 2006;36(1-3):237-45.

127. Zhao OH, Anderson JM, Hiltner A, Lodoen GA, Payet CR. Theoretical analysis on cell size distribution and kinetics of foreign-body giant cell formation in vivo on polyurethane elastomers. J Biomed Mater Res 1992;26(8):1019-38.

128. Kao WJ, Zhao QH, Hiltner A, Anderson JM. Theoretical analysis of in vivo macrophage adhesion and foreign body giant cell formation on polydimethylsiloxane, low density polyethylene, and polyetherurethanes. J Biomed Mater Res 1994;28(1):73-9.

129. Kao WJ, Hiltner A, Anderson JM, Lodoen GA. Theoretical analysis of in vivo macrophage adhesion and foreign body giant cell formation on strained poly(etherurethane urea) elastomers. J Biomed Mater Res 1994;28(7):819-29.

130. Jones JA. Biomaterials and the Foreign Body Reaction: Surface Chemistry Dependent Macrophage Adhesion, Fusion, Apoptosis, and Cytokine Production. Dissertation. Case Western Reserve University, 2007.

- 57 - Chapter I

131. Abramson S, Alexander H, Best S, Bokros J, Brunski J, Colas A, Cooper S, Curtis J, Haubold A, LL H. Classes of materials used in medicine. In: Ratner BD HA, Schoen FJ, Lemons JE, editor. Biomaterial Science: An Introduction to Materials in Medicine. San Diego: Elsevier Academic Press; 2004. p 67-225.

132. Shenton M, Lovell-Hoare M, Stevens G. Adhesion enhancement of polymer surfaces by atmospheric plasma treatment. J Phys D: Appl Phys 2001;34:2754- 2760.

133. Marchant R, Wang I. Physical and chemical aspects of biomaterials used in humans. In: RS G, editor. Implantation Biology. Boca Raton: CRC Press; 1994. p 13-39.

134. Chinn J, Sauter J, Phillips R, Jr, Kao W, Anderson J, Hanson S, Ashton T. Blood and tissue compatibility of modified polyester: thrombosis, inflammation, and healing. J Biomed Mater Res 1998;39(1):130-40.

135. Ratner BD HA, Schoen FJ, Lemons JE, editor. Biomaterial Science: An Introduction to Materials in Medicine. 2nd ed. San Diego: Elsevier Academic Press; 2004.

136. Kim Y, Kang I, Huh M, Yoon S. Surface characterization and in vitro blood compatibility of poly(ethylene terephthalate) immobilized with insulin and/or heparin using plasma glow discharge. Biomaterials 2000;21(2):121-30.

137. Canonico P, Cassinelli C, Morra M. Surface-modified polyester fabrics. Med Device Technol 1997;8(3):16-21.

138. Piglowski J, Gancarz I, Staniszewska-Kus J, Paluch D, Szymonowicz M, Konieczny K. Influence of plasma modification on biological properties of poly(ethylene terephthalate). Biomaterials 1994;15(11):909-16.

139. Walzak M, Flynn S, Foerch R, Hill J, Karbashewski E, Lin A, Strobel M. UV and ozone treatment of polypropylene and poly(ethylene terephthalate). In: Mittal K, editor. Polymer Surface Modification: Relevance to Adhesion. Utrecht: VSP; 1995. p 253-72.

140. Nakayama Y, Miyamura M, Hirano Y, Goto K, Matsuda T. Preparation of poly(ethylene glycol)-polystyrene block copolymers using photochemistry of dithiocarbamate as a reduced cell-adhesive coating material. Biomaterials 1999;20:963-970.

141. Park S, Bearinger J, Lautenschlager E, Castner D, Healy K. Surface modification of poly(ehtylene tererphthalate) angioplasty balloons with a hydrophilic poly(acrylamide-co-ethylene glycol) interpenetrating polymer network coating. J Biomed Mater Res 2000;53(5):568-76.

- 58 - Chapter I

142. Tan W, Teo G, Liao K, Leong K, Mao H, Chan V. Adhesion contact dynamics of primary hepatocytes on poly(ethylene terephthalate) surface. Biomaterials 2005;26(8):891-8.

143. Ying L, Yin C, Zhuo R, Leong K, Mao H, Kang E, Neoh K. Immobilization of galactose ligands on acrylic acid graft-copolymerized poly(ehtylene terephthalate) film and its application to hepatocyte culture. Biomacromolecules 2003;4(1):157- 65.

144. Gupta B, Plummer C, Bisson I, Frey P, Hilborn J. Plasma-induced graft polymerization of acrylic acid onto poly(ethylene terephthalate) films: characterization and human smooth muscle cell growth on grafted films. Biomaterials 2002;23(3):863-71.

145. Biltresse S, Attolini M, Marchand-Brynaert J. Cell adhesive PET membranes by surface grafting of RGD peptidomimetics. Biomaterials 2005;26(2):4576-87.

146. Nakayama Y, Matsuda T. Surface Macromolecular Architectural Designs Using Photo-Graft Copolymerization Based on Photochemistry of Benzyl N,N- Diethyldithiocarbamate. Macromolecules 1996;29:8622-8630.

147. Nakayama Y, Anderson JM, Matsuda T. Laboratory-scale mass production of a multi-micropatterned grafted surface with different polymer regions. J Biomed Mater Res 2000;53(5):584-91.

148. Nakayama Y, Matsuda T, Irie M. A novel surface photo-graft polymerization method for fabricated devices. Asaio J 1993;39(3):M542-4.

149. Otsu T. Iniferter Concept and Living Radical Polymerization. Journal of Polymer Science: Part A: Polymer Chemistry 2000;38:2121-2136.

150. Otsu T, Yamashita K, K T. Synthesis, reactivity, and role of 4-vinylbenzyl N,N- diethyldithiocarbamate as a monomer-iniferter in radical polymerization. Macromolecules 1986;19:287-290.

151. Nakayama Y, Matsuda T. Photo induced surface heparin immobilization. Asaio J 1993;39(3):M754-7.

- 59 - Chapter II

Lymphocyte/Macrophage Interactions: Biomaterial Surface- Dependent Cytokine, Chemokine, and Matrix Protein Production

Abstract

The role of lymphocytes in the biological response to synthetic polymers is poorly understood despite the transient appearance of lymphocytes at the biomaterial implant site. To investigate cytokines, chemokines, and extracellular matrix (ECM) proteins produced by lymphocytes and macrophages in response to biomaterial surfaces, human peripheral blood monocytes and lymphocytes were co-cultured on polyethylene terephthalate (PET)-based material surfaces displaying distinct hydrophobic, hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic chemistries. Antibody array screening showed the majority of detected proteins are inflammatory mediators that guide the early inflammatory phases of wound healing. Proteomic ELISA quantification of IL-2, TNF-α, IL-1β, IL-6, IL-10, IL-8, MIP-1β, TGF-β2, TIMP-1, TIMP-2, and MMP-

9 and adherent cell analysis were performed after 3, 7, and 10 days of culture. IL-2, IL-5, and IFN-γ were not detected in any co-cultures by either protein cytokine array and/or

ELISA suggesting lack of lymphocyte activation. The hydrophilic/neutral surfaces increased IL-8 relative to the hydrophobic PET surface (p<0.05). The hydrophilic/anionic surfaces promoted increased TNF-α over hydrophobic and cationic surfaces and increased MIP-1β compared to hydrophobic surfaces (p<0.05). Since enhanced macrophage fusion was observed on hydrophilic/anionic surfaces, the production of these cytokines likely plays an important role in the fusion process. The hydrophilic/cationic surface promoted IL-10 production and increased matrix

- 60 - Chapter II

metalloproteinase (MMP)-9/tissue inhibitor of MMP (TIMP) relative to

hydrophilic/neutral and anionic surfaces (p<0.05). These results suggest hydrophilic/neutral and anionic surfaces promote pro-inflammatory responses and reduced degradation of the ECM, whereas the hydrophilic/cationic surfaces induce an anti-inflammatory response and greater MMP-9/TIMP with an enhanced potential for

ECM breakdown. The study also underscores the usefulness of protein arrays in assessing the role of soluble mediators in the inflammatory response to biomaterials.

Introduction

The development of novel biomaterials, biomedical devices, or tissue-engineered

constructs necessitates a thorough understanding of the biological responses to implanted

materials. Lymphocytes and macrophages both exist at the implant site, but lymphocytes

appear transiently at the implant site and predominantly during the chronic inflammation

phase.1,2 To date, the majority of studies center around investigating the role of

macrophages in inflammation, wound healing, and the foreign body reaction subsequent

to biomaterial implantation. Meanwhile, lymphocyte activities and interactions with macrophages in response to synthetic polymers are unclear.

Although the lymphocyte role in the biological response to biomaterials is poorly understood, there is evidence that lymphocytes can participate in this response.

Lymphocytes have been shown to adhere to biomaterial surfaces in vitro.3-6 In

lymphocyte/macrophage co-cultures, adherent lymphocytes are predominantly associated

with macrophages rather than the biomaterial surface indicating molecular interactions

between the two cells (i.e. juxtacrine interaction).7 IL-4 and IL-13, known lymphokines,

- 61 - Chapter II

have been demonstrated to participate in macrophage fusion to form foreign body giant

cells (FBGCs).8-10 In vitro lymphocyte/macrophage interactions at the material surface have been shown to enhance the adhesion and fusion of macrophages as well as stimulate lymphocyte proliferation primarily through paracrine-mediated mechanisms.7 Therefore,

both direct (juxtacrine) and indirect (paracrine) mechanisms of lymphocyte/macrophage

interactions may play an integral part in the inflammatory and wound healing events that

occur at the implant site.

The lymphocyte population, consisting of T lymphocytes (T cells), B

lymphocytes (B cells), and natural killer (NK) cells, responds to stimuli utilizing various

mechanisms of action. B cells are involved in the recognition of foreign substances and

producing antibodies for the elimination of the antigens. NK cells are known for

mediating the killing of cells by inducing apoptosis. T lymphocytes, which comprise the

largest percentage, are divided into cytotoxic (CD8+) and T helper (CD4+)

subpopulations. The CD8+ cells destroy cells in a similar manner as NK cells. CD4+ T

cells are further separated into type 1 (Th1) and type 2 (Th2) T helper subsets. These T

cells can communicate and direct other cell types either directly or indirectly through

soluble factors (i.e. cytokines) to induce a variety of responses.

T lymphocytes are capable of engaging in juxtacrine cell-cell interactions with

macrophages in immune activation. Macrophages can act as antigen presenting cells

(APC) to initiate an immune response by phagocytosing, processing, and presenting

foreign materials to lymphocytes. As a result, stimulated T lymphocytes can secrete

interleukin-2 (IL-2), which mediates activation and proliferation of lymphocytes.

Cellular activation can also result in secretion of varying effector molecules. For

- 62 - Chapter II

instance, Th1 type T cells produce interferon (IFN)-γ, IL-2, and tumor necrosis factor

(TNF)-β while Th2 type T cells produce IL-4, IL-5, IL-10, and IL-13.11 Contact-

mediated activation of macrophages by activated lymphocytes triggers production of

reactive oxygen species (ROS), nitric oxide (NO), IL-1β, and TNF-α.11,12

T lymphocytes and macrophages are not only capable of activating each other but

they are also able to induce immune suppressive effects. For instance, T cell receptor

(TCR) activation by macrophage major histocompatibility complex (MHC) without a

secondary co-stimulatory signal can render the T lymphocyte unresponsive (anergy); as a

result, the T cells fail to proliferate.13 Additionally, macrophages can drive the

differentiation of regulatory T cells capable of actively suppressing immune responses

via the expression of inhibitory cell surface molecules and/or generation of IL-10 and

transforming growth factor (TGF)-β.14 These are mechanisms for active suppression of the immune response to self and foreign antigen and establishing peripheral tolerance.

Material surfaces have been shown to be capable of dictating lymphocyte and

macrophage behavior. Marques et al. demonstrated that in mixed populations of

monocytes/macrophages and lymphocytes, starch-based polymers and poly-L-lactide

induced different levels of macrophage activation as measured by cytokine secretion.15

Trinidade et al. did not find a synergistic macrophage and lymphocyte cytokine response to orthopedic biomaterials; however, their investigation focused primarily on IL-6 and

TNF-α production. Previous reports from our laboratory have shown that material surface chemistries are capable of modulating surface-adherent macrophage activation as measured by cytokine, chemokine, and matrix protein production.16,17 Using our

lymphocyte/macrophage co-cultures system, our laboratory has demonstrated that

- 63 - Chapter II

material surface chemistries can dictate macrophage adhesion, fusion, and lymphocyte

proliferation. In addition, paracrine interactions between lymphocytes and macrophages

via soluble factors were shown to stimulate macrophage fusion.18 We hypothesized that

biomaterial surface chemistries could modulate the production of inflammatory mediators

from lymphocytes and macrophages in response to interactions with biomaterials. This

study investigated the cytokines, chemokines, and extracellular matrix proteins produced

by lymphocyte/macrophage interactions using a protein array as a general screening

approach in response to biomaterial surfaces displaying distinct hydrophobic,

hydrophilic/neutral, and hydrophilic/charged surfaces. Biomaterial-dependent

differences in soluble mediator production were identified and correlated with macrophage fusion. This study illustrates the feasibility of using protein arrays to dissect

the complex interaction of lymphocytes and monocytes with biomaterial surfaces.

Materials and Methods

Biomaterial Surfaces

The polyethylene terephthalate (PET)-based photograft copolymerized surfaces

displaying distinct hydrophobic, hydrophilic/neutral, hydrophilic/anionic, and

hydrophilic/cationic characteristic surface chemistries were synthesized as described

previously.19 Sheets of Mylar® PET were modified with poly(benzyl N,N-

diethyldithiocarbamate-co-styrene) (BDEDTC) coating for the hydrophobic surface.

Polyacrylamide (PAAm), polyacrylic acid (PAANa), and N-methiodide of dimethylamino propylacrylamide (DMAPAAmMeI) provided hydrophilic/neutral,

- 64 - Chapter II

hydrophilic/anionic and hydrophilic/cationic surfaces, respectively. These surfaces were

characterized, confirmed, and prepared for cell culture as described previously.16

Monocyte and Lymphocyte Isolation and Culture

Peripheral blood monocytes and lymphocytes were isolated from healthy adult blood donors by a non-adherent centrifugation method utilizing a Percoll gradient as previously described.20 Separate monocyte and lymphocyte populations were washed twice with phosphate-buffered saline (PBS) (Invitrogen, Grand Island, NY) containing magnesium chloride and calcium chloride, resuspended in serum free medium (SFM)

(Invitrogen, Grand Island, NY) supplemented with L-glutamine, antibiotics, and antimycotics, and kept at 4˚C prior to plating. The isolated lymphocyte population was composed of, on average, 51.3% ± 4.1% CD4+ T lymphocytes, 30.8% ± 5.2% CD8+ T lymphocytes, 10.7% ± 1.1% NK cells, 5.6% ± 0.5% B lymphocytes, and 1.6% ± 0.1% contaminating monocytes while the isolated monocyte population was composed of, on average, 60.5% ± 9.1% monocytes and 39.5 ± 9.1% contaminating lymphocytes, as determined by flow cytometry. The lymphocyte and monocyte populations were suspended in supplemented SFM containing 20% autologous serum (AS) for plating.

Lymphocyte and monocytes were co-cultured in 1 mL of SFM containing 20% AS on the biomaterial surfaces using either 1.5 x 106 cells of the isolated lymphocyte population

alone (low monocyte culture) for a lymphocyte:monocyte ratio of approximately

1.0:0.015, or together with 1.0 x 106 cells of the isolated monocyte population for a

lymphocyte:monocyte ratio of approximately 1.0:0.3 (high monocyte culture). The

cultures were then incubated at 37˚C and 5% CO2 for periods of 3, 7, and 10 days. Non-

adherent cells and supernatant were collected on days 3, 7, and 10 and separated by

- 65 - Chapter II centrifugation at 10,000 rpm for 5 minutes. Supernatants were stored at -80˚C until cytokine analysis.

Determination of Adherent Cell Densities and Fusion

After culture periods of 3, 7, and 10 days, adherent cells were fixed, stained, and analyzed as previously described.18 The density of adherent monocytes, macrophages, and foreign body giant cells were determined by counting nuclei from 5 representative

20x objective fields and expressed as cells/mm2. Fusion of macrophages were expressed as percent fusion and calculated by dividing the number of nuclei contained within the multinucleated giant cells by the total number of nuclei located within the field of view.

Protein Array Screening of Cytokines/Chemokines

Supernatant collected on days 3 and 10 from lymphocyte/monocyte co-cultures on the different biomaterial surface were screened with the RayBio® Human Cytokine

Antibody Array V (RayBiotech, Inc., Norcross, GA) which detects 79 cytokines, chemokines, and growth factors simultaneously. The screening was performed according to the manufacturer’s protocol. Detection was performed by a BioRad Versadoc

Chemiluminescence Imaging System with corresponding Versadoc Chemiluminescence

Software (BioRad Laboratories, Inc., Hercules, CA). Membranes were stored at -20˚C or

-80˚C for future reference. Analysis of results involved the intensity rating of positive signals from 0 to 4 representing no signal to strong signal, respectively. Media control signal ratings were subtracted from supernatant signal ratings. Since detection sensitivities for each cytokine vary as illustrated in Table 2.1, comparisons between cytokines were not done. The variation in the level of each individual cytokine across material surface chemistries was determined from 3 different donors. The level of

- 66 - Chapter II

variation (variation index) was calculated for each cytokine by taking the average of the

absolute intensity rating differences of that particular cytokine produced in response to

the various biomaterial surfaces (BDEDTC, PAAm, PAANa, and DMAPAAmMeI) using

the following formula (averaged over 3 different experiments):

⎛ BDEDTC − PAAm + BDEDTC − PAANa + BDEDTC − DMAPAAmMeI +⎞ ⎜ ⎟ ⎜ PAAm − PAANa + PAAm − DMAPAAmMeI + PAANa − DMAPAAmMeI ⎟ Variation Index = ⎝ ⎠ 6

Select cytokines based on strength of signal, importance, and material variability were

then chosen for further quantification by ELISA.

In addition, to determine whether individual cytokines may be preferentially over-

or under-produced on PAANa compared to the other biomaterial surfaces, a PAANa

comparative index was generated. This was calculated by subtracting the signal rating of

each soluble factor as determined by the cytokine array on the individual biomaterial surfaces from that produced in response to PAANa and averaging the result using the following formula (for 3 separate experiments):

[]()PAANa − BDEDTC + (PAANa − PAAm)+ (PAANa − DMAPAAmMeI) Comparative Index = 3

A negative number indicates the factor is under-produced in supernatants from cells

exposed to PAANa compared to the other biomaterial surfaces and a positive number

indicates the factor is over-produced in supernatants from cells exposed to PAANa

compared to the other biomaterial surfaces.

- 67 - Chapter II

Table 2.1: Minimum Detection Limit for the Utilized Protein Detection Techniques

Cytokines Array (pg/mL)a ELISA (pg/mL)b IL-1β 100 1 IFN-γ 100 -- IL-2 25 7 IL-3 100 -- IL-4 1 -- IL-5 1 -- IL-6 1 0.7 IL-7 100 -- IL-8 1 3.5 IL-10 10 3.9 IL-12p40p70 1 -- IL-13 100 -- TNF-α 10 15.6 TNF-β 1000 -- IP-10 10 -- TGF-β2 1000 7 MIP-1β 10 4 TIMP-1 100 80 TIMP-2 1 11 MMP-9 1000 156 aRaybiotech, Inc., Norcross, GA bR&D Systems, Minneapolis, MN

ELISA Quantification of Cytokine Production

After screening by cytokine antibody array, targeted cytokines were quantified by

ELISA (R&D systems, Minneapolis, MN) and the assay was performed according to

manufacturer’s instructions. Microplate color intensity was measured by an EL808 ultra

microplate reader with KC Junior software (Bio-Tek Instruments, Inc., Winooski,

Vermont). For MMP-9, TIMP-1, and TIMP-2, the data provided was used to determine

molar concentrations and calculate molar ratios of MMP-9 to TIMP-1 (MMP-9/TIMP-1) and MMP-9 to TIMP-2 (MMP-9/TIMP-2).21-25 These measured quantities were

- 68 - Chapter II

compared across the experimental conditions: material surface chemistry, co-culture

ratio, and time. Table 2.1 lists the cytokines being quantified by ELISA with their

respective sensitivities.

Statistical Analysis

All data are expressed as an average ± the standard error of the mean (SEM) of at

least 3 replicate experiments with 3 different donors to account for donor variability. All

data were analyzed by ANOVA with Tukey’s test for pair-wise comparisons at a 95%

confidence level utilizing Minitab (Minitab Inc., State College, PA).

Results

Adhesion and Fusion Analysis

The lymphocyte/monocyte co-cultures on the biomaterial surfaces displayed

material-dependent monocyte/macrophage/FBGC adhesion as well as macrophage/FBGC

fusion. The adhesion and fusion analysis of high monocyte co-cultures are shown in

Figure 2.1. Examination of surface adherent cells showed that PAAm, the

hydrophilic/neutral surface, minimized adhesion relative to all surfaces. Adhesion on the

hydrophilic/neutral surface (PAAm) was significantly less than on the hydrophobic

surface, BDEDTC on day 3 (p < 0.05), significantly less than the hydrophobic

(BDEDTC) and the hydrophilic/cationic (DMAPAAmMeI) surfaces on day 7 (p < 0.01), and less than the hydrophobic surface (PET) at day 10 (p < 0.05).

The hydrophilic/anionic (PAANa) surfaces increased macrophage fusion

compared to the other surfaces. The percentage of fusion on PAANa was significantly

- 69 - Chapter II

greater than on PET, BDEDTC, and DMAPAAmMeI until day 7 (p < 0.05) and greater

than on BDEDTC and DMAPAAmMeI on day 10 (p < 0.05).

Figure 2.1: (A) Macrophage/FBGC adhesion and (B) percent fusion from high monocyte co-cultures over 10 days. Data represents the mean ± standard error of the mean, n = 4. *Significance relative to PET (p < 0.05). ^Significance relative to BDEDTC (p < 0.05). #Significance relative to BDEDTC and DMAPAAmMeI (p < 0.05). †Significance relative to PET, BDEDTC, and DMAPAAmMeI (p < 0.05).

Protein Array Analysis

Since different biomaterial surface chemistries are capable of eliciting different biological responses, a general screen of 79 inflammatory cytokines, chemokines, growth factors, and matrix proteins produced by the cell populations on the various biomaterial surfaces was performed using a cytokine array. This approach was undertaken as an attempt to identify cytokines that could potentially account for differences in the

- 70 - Chapter II

biomaterial-dependent cellular responses. Analysis revealed 43 proteins that were

produced and/or demonstrated material variability. Table 2.2 shows a summary of the

analysis for the high monocyte co-culture of 43 proteins which includes the average

signal ratings for the media control, the average rating of the co-culture supernatants from

all materials, and a measure of the biomaterial dependence of the cytokine, the variability

index (see Materials and Methods). The presented ratings take into account the level in the autologous serum (i.e. the level of protein in the control media was subtracted out).

Signal ratings ranged from -1.5 to 4 and variability indices ranged from 0 to 1.3. A negative signal rating indicates the factor was present at lower concentrations in the co- cultures on biomaterials than that detected in autologous serum. The low monocyte co- culture revealed similar findings (data not shown). Based on the analysis shown in Table

2.2, targets that were produced at high levels (signal rating > 2) and those believed to be

important cytokines with material variability were chosen for further quantification.

These included TNF-α, IL-1β, IL-6, IL-10, IL-8, MIP-1β, TIMP-1, and TIMP-2.

Although several other proteins satisfied the criteria, not all were able to be further

analyzed in this study. As shown in Table 2.2, 24 of the proteins were present at some

level in autologous control serum. Soluble factors not detected in either the high or low

monocyte co-cultures include (with variable sensitivities): GCSF, IL-1β, IL-2, IL-3, IL-4,

IL-5, IL-7, IL-12p40p70, IL-13, IL-15, IFN-γ, MIG, SCF, SDF-1, TNF-β, IGF-1,

Oncostatin M, Thrombopoietin, VEGF, BDNF, BLC, ck β 8-1, Eotaxin, Eotaxin-3, FGF-

4, FGF-6, FGF-7, FGF-9, Flt-3 Ligand, Fractalkine, GCP-2, IGFBP-4, IL-16, MIF, NT-3,

NT-4, and Osteoprotegerin. Several of these minimum detection limits are included in

Table 2.1.

- 71 - Chapter II

Table 2.2: Signal Intensity Ratings and Material Variability Indices*

Day 3† Day 10† PAANa Media Signal Variability Signal Variability Comparative Protein£ Control Rating Index Rating Index Index# ENA-78 0.00 1.08 0.39 1.58 0.72 0.11 GM-CSF 0.00 0.08 0.17 0.00 0.00 0.00 GRO 1.00 0.92 0.39 0.42 0.72 -0.11 GRO-α 0.00 1.33 0.33 0.92 0.83 -0.78 I-309 0.00 0.83 0.78 1.08 1.17 -0.11 IL-1β 0.00 2.42 0.94 2.58 0.72 -0.33 IL-6 0.50 3.92 0.17 2.83 0.22 0.22 IL-8 1.33 2.92 0.39 1.92 0.17 0.11 IL-10 0.33 0.42 0.61 0.33 0.33 0.00 MCP-1 2.17 0.08 0.39 0.00 0.33 0.00 MCP-2 0.33 0.08 0.17 -0.50 0.33 -0.22 MCP-3 0.00 0.92 0.72 0.17 0.22 0.22 MCSF 0.00 0.00 0.00 0.08 0.17 0.33 MDC 0.00 1.00 0.33 0.42 0.50 -0.11 MIP-1β 0.33 0.00 0.44 0.33 0.33 0.44 MIP-1δ 1.00 0.33 0.56 -0.83 0.33 -0.22 RANTES 2.33 -0.67 0.00 -1.50 1.00 0.67 TARC 0.00 0.92 0.39 0.25 0.39 -0.33 TGF-β1 0.00 0.00 0.00 0.08 0.17 -0.11 TNF-α 0.17 0.67 0.89 0.25 0.50 0.56 TNF-β 0.33 -0.67 0.00 0.08 0.17 -0.11 EGF 2.50 -0.25 0.17 0.08 0.72 0.33 Angiogenin 2.50 -1.00 0.00 0.17 0.67 0.22 PDGF-BB 3.17 -0.83 0.67 0.17 0.78 -0.22 Leptin 2.67 -1.17 0.78 -0.25 0.72 -0.56 Eotaxin-2 0.33 1.50 0.22 1.50 0.67 0.22 GDNF 0.33 -0.33 0.44 0.17 0.33 0.67 HGF 1.17 -0.33 0.33 1.00 0.78 0.00 IGFBP-1 0.00 0.08 0.17 0.33 0.56 0.44 IGFBP-2 1.50 -0.33 0.00 0.08 0.50 -0.11 IGFBP-3 0.00 0.00 0.00 0.17 0.33 -0.22 IP-10 0.00 0.08 0.17 0.50 0.78 -0.67 LIF 0.00 0.08 0.17 0.00 0.00 0.00 LIGHT 0.00 0.08 0.17 0.00 0.00 0.00 MCP-4 0.00 0.08 0.17 0.00 0.00 0.00 MIP-3α 0.00 1.00 1.22 1.00 0.89 -0.44 NAP-2 2.33 -0.33 0.00 -0.25 0.39 0.33 PARC 2.83 -1.00 0.33 -0.17 1.11 0.22 PlGF 0.00 0.08 0.17 0.00 0.00 0.00 TGF-β2 1.67 -0.92 0.17 0.25 0.17 0.11 TGF-β3 0.00 0.08 0.17 0.00 0.00 0.00 TIMP-1 2.00 -0.67 0.00 -0.25 0.61 -0.11 TIMP-2 2.67 -0.08 0.17 0.17 0.33 0.22 *Data represents the mean of 3 donors †Signal Ratings represent the average of intensity ratings over all materials. Variability Index indicates the extent of material differences. All values are minus media control. #Represents a signal ratings comparison between PAANa and all other surfaces £ENA = epithelial-derived neutrophil activating protein; GRO = growth related oncogene; GM-CSF = granulocyte macrophage colony stimulating factor; IL = interleukin; MCP = macrophage chemotactic protein; MCSF = macrophage colony stimulating factor; MDC = macrophage-derived chemokine; MIP = macrophage inflammatory protein; RANTES = regulated upon activation, normal T-cell expressed, and secreted; TARC = thymus and activation-regulated chemokine; TNF = tumor necrosis factor; EGF = epidermal growth factor; PDGF = platelet-derived growth factor; GDNF = glial cell-derived neurotrophic factor; HGF = hepatic growth factor; IGFBP = insulin-like growth factor binding protein; IP = interferon-

- 72 - Chapter II

inducible protein; LIF = leukemia inhibitory factor; NAP = neutrophil activating protein; PARC = pulmonary and activation-regulated chemokine; PlGF = placenta growth factor; TGF = transforming growth factor; TIMP = tissue inhibitor of matrix metalloproteinases

ELISA Analysis

After cytokine array screening, production of TNF-α, IL-1β, IL-6, IL-10, IL-8,

MIP-1β, TIMP-1, and TIMP-2 by the lymphocyte/macrophage co-cultures were quantified by ELISA. Additionally, IL-2, TGF-β2, and MMP-9 were selected for

quantification because they were proteins of importance in lymphocyte proliferation (IL-

2), wound healing (TGF-β), or are targets of proteins detected in array analysis (MMP-9).

Two co-culture ratios (high and low monocyte) were utilized in order to examine the

effect of depleted monocytes/macrophages on lymphocyte/macrophage interactions and

the production of these various factors. IL-1β, IL-10, TGF-β2, MIP-1β, MMP-9, TIMP-

1, and TIMP-2 were present in autologous serum containing media controls at

concentrations of 24 ± 7 pg/mL, 4 ± 1 pg/mL, 57 ± 2 pg/mL, 24 ± 1 pg/mL, 44 ± 14

ng/mL, 32 ± 4 ng/mL, and 20 ± 2 ng/mL, respectively. These amounts were subtracted

from the quantities in the media exposed to the co-cultures to determine the produced

quantities. No IL-2 was detected (sensitivity < 7 pg/mL) while TGF-β2 was found to be

minimally produced (sensitivity < 7 pg/mL) (data not shown).

Cytokines

Figure 2.2 shows the quantity of inflammatory cytokines, IL-1β, IL-6, TNF-α,

and IL-10, produced over time. In general, the low monocyte and high monocyte co-

cultures showed similar material trends although there was variability in the amount of

cytokines produced by high and low monocyte co-cultures on the different material

surfaces. The hydrophilic PAAm surface tended to induce an increased production of IL-

- 73 - Chapter II

1β relative to other surfaces, hydrophobic PET and BDEDTC in particular, in both co-

culture ratios [Figure 2.2(A, B)]. IL-1β production was initially higher on PAAm and

PAANa surfaces and decreased by day 7 while all other surfaces remained relatively

constant. IL-6 on all surfaces remained relatively constant over time.

TNF-α levels were highest on the hydrophilic/anionic (PAANa) surfaces across

all time points and in both co-culture ratios as shown in Figure 2.2(E, F). The levels on

PAANa were significantly greater than hydrophobic (PET and BDEDTC) and

hydrophilic/cationic (DMAPAAmMeI) surfaces in low monocyte co-cultures at initial

day 3 and high monocyte co-cultures at day 7 (p < 0.005). Levels were increased on the

hydrophilic/neutral (PAAm) surfaces compared to PET, BDEDTC and DMAPAAmMeI

although the difference was not statistically significant. Over time, however, TNF-α levels on these two surfaces (PAAm and PAANa) decreased while the other surfaces produced lower levels that remained relatively constant or slightly increased.

Whereas IL-1β, TNF-α, and IL-6 either decreased over time or remained the same, IL-10 production generally increased with time as shown in Figure 2.2(G, H).

Significant differences were most evident in the high monocyte co-cultures. At day 7, the

IL-10 level was significantly higher on the hydrophilic/cationic surface,

DMAPAAmMeI, relative to the hydrophilic/neutral surface (p < 0.01). By day 10,

DMAPAAmMeI showed greater production of IL-10 than both hydrophilic/neutral (p <

0.05) and hydrophilic/anionic surfaces (p < 0.005).

- 74 - Chapter II

Figure 2.2: IL-1β production from (A) low and (B) high monocyte co-cultures over 10 days. IL-6 production from (C) low and (D) high monocyte co-cultures over 10 days. TNF-α production from (E) low and (F) high monocyte co-cultures over 10 days. IL-10 production from (G) low and (H) high monocyte co-cultures over 10 days. Data represents the mean ± standard error of the mean, n = 4. *Significance relative to PAANa (p < 0.05). #Significance relative to PAAm and PAANa (p < 0.05). †Significance relative to PET, BDEDTC, and DMAPAAmMeI (p < 0.05).

- 75 - Chapter II

Chemokines

IL-8 was produced at the highest concentration amongst all proteins quantified.

IL-8 production on the various materials across time is shown in Figure 2.3A and 2.3B.

IL-8 levels were produced at a level approximately 10 fold greater in high monocyte co- cultures relative to low monocyte co-cultures. The hydrophilic/neutral surface, PAAm, in high monocyte co-cultures, in particular, showed increased IL-8 levels relative to the hydrophobic surfaces over the entire culture period with significance compared to PET after 10 days of culture (p < 0.05).

MIP-1β, on the other hand, began decreasing after the initial day 3 time point over the time course regardless of culture ratio (Figure 2.3C and 2.3D). The quantity in high monocyte co-cultures declined at a much faster rate than in the low monocyte co-cultures.

MIP-1β levels in the high monocyte co-cultures were initially greater at day 3 than those in low monocyte co-cultures by a factor of two to eight depending on the surface. By 10 days, MIP-1β levels on all materials were lower in high monocyte co-cultures than in low monocyte co-cultures. The highest quantity of MIP-1β was seen on the hydrophilic/anionic surface, PAANa, over time and material. Levels on PAAm were slightly increased compared to hydrophobic and cationic surfaces although not statistically significant. The trend was the same for both co-culture ratios. However, in low monocyte co-cultures, MIP-1β production after 3 days was significantly higher on

PAANa relative to PET and significantly higher than PET and BDEDTC after 7 days (p <

0.05).

- 76 - Chapter II

Figure 2.3: IL-8 production from (A) low and (B) high monocyte co-cultures over 10 days. MIP-1β production from (C) low and (D) high monocyte co-cultures over 10 days. Data represents the mean ± standard error of the mean, n = 4. *Significance relative to PET (p < 0.05). †Significance relative to PET and BDEDTC (p < 0.05).

Extracellular Matrix Proteins

MMP-9 production increased after day 3 over the 10 day culture period and

diverged according to material surface [Figure 2.4(A, B)]. Both TIMP-1 and TIMP-2

mirrored MMP-9 in regards to increased production after day 3 as demonstrated in Figure

2.4, C – D and E – F, respectively. TIMP-1 also had a small amount produced at day 3

(10 ng/mL for the low monocyte co-culture and 100 ng/mL for the high monocyte co-

culture) while TIMP-2 was minimally produced in both co-cultures. DMAPAAmMeI

diminished TIMP-1 and TIMP-2 production at later time points showing significantly in

high monocyte co-cultures after 7 days (Figure 4D and 4F).

- 77 - Chapter II

Figure 2.4: MMP-9 production from (A) low and (B) high monocyte co-cultures over 10 days. TIMP-1 production from (C) low and (D) high monocyte co-cultures over 10 days. TIMP-2 production from (E) low and (F) high monocyte co-cultures over 10 days. Data represents the mean ± standard error of the mean, n = 4. *Significance relative to PAAm (p < 0.05). †Significance relative to PAAm and PAANa (p < 0.05).

- 78 - Chapter II

TIMPs bind and inhibit MMPs in a 1:1 stoichiometric ratio.26 Molar concentrations of MMP-9, TIMP-1, and TIMP-2 were calculated and subsequently used to determine MMP-9/TIMP-1 and MMP-9/TIMP-2 in order to quantitatively show the balance between matrix metalloproteinase and its inhibitor (TIMPs). These ratios are presented at 3, 7, and 10 day time points in Figure 2.5. In both low and high monocyte co-cultures, MMP-9/TIMP-1 [Figure 2.5(A, C, E)] and MMP-9/TIMP-2 [Figure 2.5(B,

D, F)] both increased with time on PET, BDEDTC, the most on DMAPAAmMeI, and least if at all on PAAm and PAANa surfaces. Initially, at day 3, MMP-9/TIMP-1 was close to 1 (< 5) on all surfaces and in general, increased on DMAPAAmMeI over time at a rate higher than on all other surfaces. By day 7 in low monocyte co-cultures, MMP-

9/TIMP-1 on PAAm and PAANa were at the lowest levels while the ratio on

DMAPAAmMeI was significantly higher (p < 0.05). By day 10, both co-cultures demonstrated lower ratios on PAAm and PAANa with those on PAAm significantly lower compared to DMAPAAmMeI (p < 0.05). Like MMP-9/TIMP-1, MMP-9/TIMP-2 also increased with time and showed an identical material-dependent trend. That is, after day 3, the ratio on hydrophilic/neutral (PAAm) and hydrophilic/anionic (PAANa) surfaces tended to be lower while the hydrophilic/cationic (DMAPAAmMeI) surface showed the highest levels. The ratio on DMAPAAmMeI in low monocyte co-cultures was significantly higher than both PAAm and PAANa at day 7 (p < 0.05).

- 79 - Chapter II

Figure 2.5: Ratio of MMP-9/TIMP-1 in co-cultures over (A) 3 days, (C) 7 days, and (E) 10 days. Ratio of MMP-9/TIMP-2 in co-cultures over (B) 3 days, (D) 7 days, and (F) 10 days. Data represents the mean ± standard error of the mean, n = 4. *Significance relative to PAAm (p < 0.05). †Significance relative to PAAm and PAANa (p < 0.05).

Finally, since PAANa induced significantly greater macrophage fusion than on

other surfaces, the protein arrays were utilized to identify soluble factors that could

account for the enhanced fusion. A comparison between the levels of these factors in

supernatants from lymphocyte and monocyte co-cultures exposed to the fusion–

promoting hydrophilic/anionic surface, PAANa, and those of the other biomaterials was

determined using a PAANa comparative index (see Materials and Methods and Table 2).

Cytokines and chemokines MIP-1β, RANTES, TNF-α, GDNF, and IGFBP-1 appeared to

- 80 - Chapter II

be over-produced in supernatants from co-cultures exposed to PAANa compared to other

substrates and positively correlated with macrophage fusion while the soluble factors

GRO-α, Leptin, IP-10, and MIP-3α appeared negatively correlated with fusion. These differences were verified in the subsequent ELISA analysis for two factors, TNF-α and

MIP-1β (Figures 2.2 and 2.3). Thus, the use of cytokine arrays is a feasible approach to identify soluble factors that may be important for biomaterial-dependent processes such as macrophage fusion.

Discussion

The objectives of this study were to examine the response of lymphocyte/macrophage interactions to biomaterials and the effect of material surface

chemistry on these interactions. Our findings show that in response to biomaterial

surfaces, cytokines, chemokines, as well as matrix proteins are produced, that are capable

of guiding cellular behavior and the development of the extracellular matrix surrounding

the implantation site, and biomaterial surface chemistry influences the particular cytokine

profiles produced by the inflammatory cells. Additionally, the concentration of

monocytes in the lymphocyte/macrophage co-cultures can alter the level of inflammatory

mediators produced in this experimental system.

Cell derived proteins from lymphocyte/macrophage co-cultures include cytokines,

chemokines, growth factors, and matrix proteins with pleiotropic effects on multiple

inflammatory and wound healing cell types. These include inflammatory cytokines such

as IL-1β, IL-6, IL-10, and TNF-α as well as chemokines (e.g. ENA-78, NAP-2, IL-8,

MCP, GRO, MIP-1β, IP-10, PARC) that target inflammatory effector cells such as

- 81 - Chapter II

neutrophils, monocyte/macrophages, and lymphocytes. Many of these chemokines are

capable of attracting more than one cell type. IL-8 chemotactically attracts neutrophils and T lymphocytes and also activates neutrophils. MIP-1β is known as a chemotactic protein for monocytes/macrophages, but it is also capable of chemotactically targeting

CD4+ T lymphocytes.27 MCP-1 has also been shown to play a role in the biological

response to biomaterials in promoting macrophage fusion to form foreign body giant

cells.28 In addition, many of the cytokines and chemokines are capable of influencing

lymphocyte behaviors. I-309, IL-8, MCPs, MDC, MIP-1β, MIP-3α, RANTES, PARC,

IP-10, and TARC can all chemoattract specific lymphocyte subpopulations such as T

lymphocytes.29 Cytokines such as IL-1β and IL-6 are capable of enhancing T

lymphocyte activation and proliferation with IL-1β augmenting IL-2 production and IL-6

increasing responsiveness to IL-2.30,31 Chemokines such as MIP-1β, MCP-1, and

RANTES have been shown to induce activation and proliferation of NK cells.32

Moreover, RANTES can activate and induce proliferation of T cells independent of other stimuli.33 Nonetheless, T lymphocyte activation was not observed in our studies as

neither IL-2 nor IFN-γ was detected in the co-cultures.

The majority of the soluble factors produced over the 10 days of culture act

primarily to direct the early part of the normal wound healing response. As expected,

TNF-α and IL-1β were initially high and decreased over the 10 day period. However,

IL-10, a cytokine suppressing the inflammatory response, generally increased over the 10

day cultures suggesting dampening of the immune response after 10 days. These results

mirror time-dependent resolution of the inflammatory response in normal wound

healing.34 Additionally, MMP-9, TIMP-1, and TIMP-2 all increased in production after 3

- 82 - Chapter II

days of culture. MMP-9 readily digests extracellular matrix molecules such as gelatin,

elastin, and collagens and is inhibited by the TIMPs.35 During the inflammatory phase,

proteases and MMPs are elaborated in order to degrade the damaged extracellular matrix

and facilitate cellular migration through the matrix. In general, the balance between

protease and inhibitor tilted towards protease action and matrix breakdown. IL-1β and

TNF-α, which were produced at high levels at day 3, induce MMP transcription.34 Thus,

the increase in MMP-9 after day 3 is consistent with the natural wound healing

progression. The continued production of MMP-9 over its inhibitor was also likely due

to the lack of TGF-β in the co-cultures. TGF-β, a major growth factor in the later stages

of the wound healing response, upregulates TIMPs while decreasing MMP production

but was not detected in our system.34 Therefore, lymphocytes and macrophages in

culture demonstrated the capability of guiding the inflammatory phase of the wound

healing response.

IL-2 and IFN-γ were undetected at minimum detection limits of less than 7 pg/mL

(ELISA) and 100 pg/mL (cytokine array), respectively while considerable levels of IL-10

were found. The lack of IL-2 suggests the absence of classic T lymphocyte activation

and proliferation in response to synthetic polymers. Macrophages are capable of rendering T cells unresponsive, or anergic, as a result of TCR/MHC interaction in the absence of costimulation.13 IL-10 inhibits the activation of macrophages and

downregulates macrophage co-stimulatory activity.36,37 Furthermore, IL-10 can directly

inhibit T cell activation and induce T cell anergy.13,38 Therefore, the results from this in

vitro study suggest that the lack of T cell activation as determined by cytokine release

may be mediated by IL-10 inhibition either directly or indirectly. Our findings are

- 83 - Chapter II

similar to those described by Hoves, et al. They found substantial levels of IL-10, low

levels of IL-2 and IFN-γ, and minimal T lymphocyte proliferation in in vitro co-cultures,

and showed that human monocyte-derived macrophages produced anergic T cells.14 The absence (as determined by protein cytokine array) of additional lymphokines such as IL-4 and IL-5 further suggest lymphokine suppression and lack of T lymphocyte activation.

As shown above, material surface chemistries are capable of dictating cellular behaviors such as adhesion, fusion, and protein production. Hydrophilic/neutral PAAm surfaces diminished adhesion while all other surfaces, hydrophobic and hydrophilic/charged, promoted adhesion. Overall cellular (macrophage/FBGC) adhesion on PAANa also diminished over time as macrophages fused into FBGCs. The adhesion results confirm previous material surface chemistry dependency shown in our laboratory.39 Although adhesion on PAAm was inhibited, fusion was still capable of

occurring. The amount of fusion on PAAm was higher in these co-cultures compared to

fusion in adherent monocyte/macrophage cultures.16 This is likely due to the presence of

lymphocytes in culture since lymphocytes have been shown to facilitate fusion.7

Additionally, hydrophilic/anionic PAANa increased macrophage fusion while hydrophobic and hydrophilic/cationic surfaces inhibited fusion. The PAANa fusion promotion confirms previous findings in our laboratory.18

The protein array analysis showed many cytokines exhibited potential material- dependent production based on a calculated variability index. Furthermore, the array analysis identified soluble factors that could account for the biomaterial-dependent

differences in fusion. MIP-1β, TNF-α, and RANTES were found to positively correlate

with fusion and could be proteins that are produced and also possibly utilized during

- 84 - Chapter II

macrophage fusion to form FBGCs. On the other hand, GRO-α, a chemokine that can

attract monocytes, along with IP-10 and MIP-3α negatively correlated with macrophage

fusion suggesting that this chemokine may be preferentially utilized during the fusion process. Quantification by ELISA confirmed the correlation of MIP-1β and TNF-α with

fusion on the biomaterial surfaces. These results indicate that the use of protein arrays is

a feasible approach to identify soluble factors that may be important for biomaterial-

dependent processes such as macrophage fusion. However, the variability in the

detection sensitivities for the protein targets on the array (from 1 pg/mL to 2000 pg/mL)

somewhat limits its usefulness since biomaterial-dependent changes in signal intensity

could be due to either trivial or rather large changes in the concentration of the molecules.

Therefore, more precise quantification of molecules of interest by ELISA helps

substantiate the observed differences.

The hydrophilic/neutral (PAAm) and hydrophilic/anionic (PAANa) surfaces were

both shown to promote a pro-inflammatory response albeit through different cytokine

secretion profiles while the hydrophilic/cationic surfaces (DMAPAAmMeI) promoted a

more anti-inflammatory response. Despite the inhibition of cellular adhesion on PAAm,

higher levels of IL-8, IL-1β, and TNF-α, and lower levels of IL-10 were induced on

PAAm compared to other surfaces. TNF-α and MIP-1β were over-produced in response

to PAANa with significantly less IL-10 secreted. DMAPAAmMeI induced greater IL-10

production and relatively low levels of pro-inflammatory mediators such as IL-1β, TNF-

α, and MIP-1β compared to the other surfaces. IL-10 suppresses inflammatory cytokines

specifically inhibiting cytokines such as IL-1β, IL-6, IL-8, TNF-α, MIP-1β in

monocytes/macrophages in addition to its effects on co-stimulation mentioned above.40,41

- 85 - Chapter II

Aside from its effects on metalloproteinases, TNF-α also induces cellular apoptosis. Brodbeck et al. demonstrated that TNF-α was responsible for biomaterial- mediated spontaneous apoptosis.42 Furthermore, hydrophilic/neutral and hydrophilic/anionic biomaterial surfaces were shown to induce a higher level of adherent macrophage apoptosis compared to hydrophobic and cationic surfaces both in vitro and in vivo.43,44 The results from this study are consistent with these findings in that TNF-α was produced at the highest levels on both hydrophilic/neutral (PAAm) and hydrophilic/anionic (PAANa) surfaces.

MMPs degrade the damaged extracellular matrix and also facilitate cellular migration through the matrix. These actions ultimately facilitate the progression of the immune response to injury from inflammatory to the proliferative and ultimately the remodeling phases. The actions of the MMPs are inhibited and regulated by the TIMPs in a 1:1 stoichiometric ratio,26 and the balance between protease and inhibitor determines the degradative state of the ECM. The hydrophilic/neutral (PAAm) and hydrophilic/anionic (PAANa) surfaces were both shown to minimize MMP-9/TIMP suggesting a reduction in degradation of extracellular matrix while the hydrophilic/cationic surfaces (DMAPAAmMeI) promoted greater MMP-9/TIMP suggesting the potential for greater extracellular matrix breakdown. Based upon the cumulative actions of the inflammatory mediators, hydrophilic/neutral and hydrophilic/anionic surfaces induce a pro-inflammatory state with a potentially inhibited progression from the inflammatory phase. Conversely, hydrophilic/cationic surfaces induce an anti-inflammatory state while possibly promoting progression through the wound healing phases.

- 86 - Chapter II

In comparing the high monocyte to low monocyte co-cultures, the ratios of MIP-

1β and TIMP-2 production were less than the ratio of monocytes (high:low) on all

materials. TNF-α showed both higher and lower ratios depending on the surface. These

ratio differences suggest that cell numbers and/or lymphocyte/macrophage interactions

can affect the amounts of soluble factors produced. Levels of MIP-1β were markedly

reduced in high monocyte co-cultures relative to low monocyte co-cultures as time

progressed. This could be due to enhanced lymphocyte-mediated activation of

macrophages resulting in increased utilization or breakdown of the chemokine or a

reflection of lymphocyte-mediated inhibition of chemokine production. MIP-1β, a

known chemokine, may also be involved in the fusion process since macrophages are

required to come together in close proximity. Increased macrophage activation in higher

macrophage co-cultures would then lead to increased utilization of this factor. TIMP-2

production also demonstrated a similar decreased response. Unlike MIP-1β, which is a

chemokine that targets monocytes/macrophages, TIMP-2 is a matrix protein that acts on

the extracellular matrix. In this case, inhibited production is more likely than increased

utilization. TNF-α levels were also reduced in high monocyte co-cultures compared to

low monocyte co-cultures on PAANa and earlier time points on PAAm where the

concentrations were relatively high. On the other hand, enhanced production of TNF-α

in high monocyte-co-cultures relative to low monocyte co-cultures was evident on PET,

BDEDTC, and DMAPAAmMeI where TNF-α levels were comparatively low. This

variable response suggests that there may be a feedback mechanism leading to inhibition

of the specific cytokine production. Therefore, the cellular interactions at the

tissue/material interface and the subsequent response are potentially influenced by

- 87 - Chapter II

material surface chemistries as well as lymphocyte or macrophage numbers. In contrast,

Trindade et al. suggested that there was a lack of synergism between the lymphocytes and

macrophages with respect to IL-6 and TNF-α production in response to particles

composed of orthopedic materials.45 Our findings prompt further investigation into possible lymphocyte-mediated monocyte/macrophage activation, inflammatory mediator production, and material dependency of these interactions and responses.

This study contributes to understanding lymphocyte/macrophage interactions that occur at the implant site of a biomaterial. The inability of T lymphocytes to produce IL-2 or IFN-γ when interacting with macrophages indicates the lack of classic T lymphocyte activation. However, the results suggest lymphocytes may modulate the production of soluble factors by macrophages. Our findings provide quantitative evidence to support our hypothesis that material surface chemistry is capable of modulating cytokine, chemokine, and matrix protein production from lymphocyte/macrophage interactions.

More specifically, material surface characteristics or chemistries can dictate the

production profile of different inflammatory mediators. Further studies are necessary to better understand the importance of juxtacrine and paracrine-mediated

lymphocyte/macrophage interactions and the material surface-dependent effects on those

interactions.

References

1. Anderson JM. Inflammation and the foreign body response. Problems in General Surgery 1994;11(2):147-60.

2. Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81-110.

- 88 - Chapter II

3. Maeda M, Kimura M, Inoue S, Kataoka K, Okano T, Sakurai Y. Adhesion behavior of rat lymphocyte subpopulations (B cell and T cell) on the surface of polystyrene/polypeptide graft copolymer. J Biomed Mater Res 1986;20(1):25-35.

4. Yokoyama M, Nakahashi T, Nishimura T, Maeda M, Inoue S, Kataoka K, Sakurai Y. Adhesion behavior of rat lymphocytes to poly(ether)-poly(amino acid) block and graft copolymers. J Biomed Mater Res 1986;20(7):867-78.

5. Groth T, Altankov G, Klosz K. Adhesion of human peripheral blood lymphocytes is dependent on surface wettability and protein preadsorption. Biomaterials 1994;15(6):423-8.

6. Bergman AJ, Zygourakis K. Migration of lymphocytes on fibronectin-coated surfaces: temporal evolution of migratory parameters. Biomaterials 1999;20(23- 24):2235-44.

7. Brodbeck WG, Macewan M, Colton E, Meyerson H, Anderson JM. Lymphocytes and the foreign body response: lymphocyte enhancement of macrophage adhesion and fusion. J Biomed Mater Res A 2005;74(2):222-9.

8. McNally AK. Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 1995;147(5):1487-99.

9. DeFife KM, Jenney CR, McNally AK, Colton E, Anderson JM. Interleukin-13 induces human monocyte/macrophage fusion and macrophage mannose receptor expression. J Immunol 1997;158(7):3385-90.

10. Kao WJ, McNally AK, Hiltner A, Anderson JM. Role for interleukin-4 in foreign- body giant cell formation on a poly(etherurethane urea) in vivo. J Biomed Mater Res 1995;29(10):1267-75.

11. Romagnani S. Th1/Th2 cells. Inflamm Bowel Dis 1999;5(4):285-94.

12. Burger D, Dayer JM. The role of human T-lymphocyte-monocyte contact in inflammation and tissue destruction. Arthritis Res 2002;4 Suppl 3:S169-76.

13. Roncarolo MG, Bacchetta R, Bordignon C, Narula S, Levings MK. Type 1 T regulatory cells. Immunol Rev 2001;182:68-79.

14. Hoves S, Krause SW, Schutz C, Halbritter D, Scholmerich J, Herfarth H, Fleck M. Monocyte-derived human macrophages mediate anergy in allogeneic T cells and induce regulatory T cells. J Immunol 2006;177(4):2691-8.

15. Marques AP, Reis RL, Hunt JA. Cytokine secretion from mononuclear cells cultured in vitro with starch-based polymers and poly-L-lactide. J Biomed Mater Res A 2004;71(3):419-29.

- 89 - Chapter II

16. Jones JA, Chang DT, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Proteomic analysis and quantification of cytokines and chemokines from biomaterial surface-adherent macrophages and foreign body giant cells. J Biomed Mater Res A 2007;83(3):585-96.

17. Jones JA, McNally AK, Chang DT, Qin LA, Meyerson H, Colton E, Kwon IL, Matsuda T, Anderson JM. Matrix metalloproteinases and their inhibitors in the foreign body reaction on biomaterials. J Biomed Mater Res A 2008;84(1):158-66.

18. MacEwan MR, Brodbeck WG, Matsuda T, Anderson JM. Student Research Award in the Undergraduate Degree Candidate category, 30th Annual Meeting of the Society for Biomaterials, Memphis, Tennessee, April 27-30, 2005. Monocyte/lymphocyte interactions and the foreign body response: in vitro effects of biomaterial surface chemistry. J Biomed Mater Res A 2005;74(3):285-93.

19. Nakayama Y, Anderson JM, Matsuda T. Laboratory-scale mass production of a multi-micropatterned grafted surface with different polymer regions. J Biomed Mater Res 2000;53(5):584-91.

20. McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 1994;91(21):10119-23.

21. Olson MW, Gervasi DC, Mobashery S, Fridman R. Kinetic analysis of the binding of human matrix metalloproteinase-2 and -9 to tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2. J Biol Chem 1997;272(47):29975-83.

22. Xue M, Thompson PJ, Clifton-Bligh R, Fulcher G, Gallery ED, Jackson C. Leukocyte matrix metalloproteinase-9 is elevated and contributes to lymphocyte activation in type I diabetes. Int J Biochem Cell Biol 2005;37(11):2406-16.

23. Mott JD, Thomas CL, Rosenbach MT, Takahara K, Greenspan DS, Banda MJ. Post-translational proteolytic processing of procollagen C-terminal proteinase enhancer releases a metalloproteinase inhibitor. J Biol Chem 2000;275(2):1384- 90.

24. Osthues A, Knauper V, Oberhoff R, Reinke H, Tschesche H. Isolation and characterization of tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2) from human rheumatoid synovial fluid. FEBS Lett 1992;296(1):16-20.

25. Stetler-Stevenson WG, Liotta LA, Kleiner DE, Jr. Extracellular matrix 6: role of matrix metalloproteinases in tumor invasion and metastasis. Faseb J 1993;7(15):1434-41.

26. Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res 2003;92(8):827-39.

- 90 - Chapter II

27. Schall TJ, editor. The Cytokine Handbook. New York: Academic Press; 1994. 419-460 p.

28. Kyriakides TR, Foster MJ, Keeney GE, Tsai A, Giachelli CM, Clark-Lewis I, Rollins BJ, Bornstein P. The CC chemokine ligand, CCL2/MCP1, participates in macrophage fusion and foreign body giant cell formation. Am J Pathol 2004;165(6):2157-66.

29. Ward SG, Westwick J. Chemokines: understanding their role in T-lymphocyte biology. Biochem J 1998;333 ( Pt 3):457-70.

30. Houssiau FA, Coulie PG, Van Snick J. Distinct roles of IL-1 and IL-6 in human T cell activation. J Immunol 1989;143(8):2520-4.

31. Le JM, Fredrickson G, Reis LF, Diamantstein T, Hirano T, Kishimoto T, Vilcek J. Interleukin 2-dependent and interleukin 2-independent pathways of regulation of thymocyte function by interleukin 6. Proc Natl Acad Sci U S A 1988;85(22):8643-7.

32. Maghazachi AA, Al-Aoukaty A, Schall TJ. CC chemokines induce the generation of killer cells from CD56+ cells. Eur J Immunol 1996;26(2):315-9.

33. Bacon KB, Premack BA, Gardner P, Schall TJ. Activation of dual T cell signaling pathways by the chemokine RANTES. Science 1995;269(5231):1727-30.

34. Broughton G, 2nd, Janis JE, Attinger CE. The basic science of wound healing. Plast Reconstr Surg 2006;117(7 Suppl):12S-34S.

35. Nagase H, Visse R, Murphy G. Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc Res 2006;69(3):562-73.

36. Schwartz RH. Models of T cell anergy: is there a common molecular mechanism? J Exp Med 1996;184(1):1-8.

37. Ding L, Linsley PS, Huang LY, Germain RN, Shevach EM. IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the up-regulation of B7 expression. J Immunol 1993;151(3):1224-34.

38. Groux H, Bigler M, de Vries JE, Roncarolo MG. Interleukin-10 induces a long- term antigen-specific anergic state in human CD4+ T cells. J Exp Med 1996;184(1):19-29.

39. Brodbeck WG, Nakayama Y, Matsuda T, Colton E, Ziats NP, Anderson JM. Biomaterial surface chemistry dictates adherent monocyte/macophage cytokine expression in vitro. Cytokine 2002;18(6):311-19.

- 91 - Chapter II

40. de Waal Malefyt R, Abrams J, Bennett B, Figdor CG, de Vries JE. Interleukin 10(IL-10) inhibits cytokine synthesis by human monocytes: an autoregulatory role of IL-10 produced by monocytes. J Exp Med 1991;174(5):1209-20.

41. Groux H, Cottrez F. The complex role of interleukin-10 in autoimmunity. J Autoimmun 2003;20(4):281-5.

42. Brodbeck WG, Shive MS, Colton E, Ziats NP, Anderson JM. Interleukin-4 inhibits tumor necrosis factor-alpha-induced and spontaneous apoptosis of biomaterial-adherent macrophages. J Lab Clin Med 2002;139(2):90-100.

43. Brodbeck WG, Patel J, Voskerician G, Christenson E, Shive MS, Nakayama Y, Matsuda T, Ziats NP, Anderson JM. Biomaterial adherent macrophage apoptosis is increased by hydrophilic and anionic substrates in vivo. Proc Natl Acad Sci U S A 2002;99(16):10287-92.

44. Brodbeck WG, Shive MS, Colton E, Nakayama Y, Matsuda T, Anderson JM. Influence of biomaterial surface chemistry on the apoptosis of adherent cells. J Biomed Mater Res 2001;55(4):661-8.

45. Trindade MC, Lind M, Sun D, Schurman DJ, Goodman SB, Smith RL. In vitro reaction to orthopaedic biomaterials by macrophages and lymphocytes isolated from patients undergoing revision surgery. Biomaterials 2001;22(3):253-9.

- 92 - Chapter III

Paracrine and Juxtacrine Lymphocyte Enhancement of Adherent Macrophage and Foreign Body Giant Cell Activation

Abstract

Lymphocytes have been shown to be involved in modulating monocyte and macrophage behavior in the foreign body reaction. Lymphocyte effects on biomaterial- adherent macrophage and foreign body giant cell (FBGC) behavior were further investigated by culturing monocytes alone or together with lymphocytes, either in direct

co-cultures or indirectly in transwells, on a series of polyethylene terephthalate (PET)-

based photograft co-polymerized material surfaces displaying distinct hydrophobic,

hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic chemistries. After

periods of 3, 7, and 10 days, cytokine production was quantified by ELISA and

normalized to adherent macrophage/FBGC density to yield a measure of adherent

macrophage/FBGC activation. Interactions with lymphocytes enhanced adherent

macrophage and FBGC production of pro-inflammatory IL-1β, TNF-α, IL-6, IL-8, and

MIP-1β on the hydrophobic and hydrophilic/cationic surfaces but had no effect on anti-

inflammatory IL-10 production indicating lymphocytes promote a pro-inflammatory

response to biomaterials. Lymphocytes also did not significantly influence MMP-9,

TIMP-1, and TIMP-2 production. Interactions through indirect (paracrine) signaling

showed a significant effect in enhancing adherent macrophage/FBGC activation at early

time points while interactions via direct (juxtacrine) mechanisms dominated at later time

points. Biomaterial surface chemistries differentially affected the observed responses as

hydrophilic/neutral and hydrophilic/anionic surfaces evoked the highest levels of

- 93 - Chapter III

activation relative to the other surfaces but did not facilitate lymphocyte enhancement of adherent macrophage/FBGC activation.

Introduction

Biomaterials introduced into the body elicit a biological response involving

degradative and phagocytic activities leading to potential implant damage and failure.

Macrophages lie at the center of this foreign body reaction as the precursor cells that fuse

to form foreign body giant cells (FBGC). Once activated, macrophages guide the tissue response through the production of inflammatory mediators including cytokines, chemokines, and extracellular matrix (ECM)-modifying proteins. Lymphocytes appear transiently during the chronic inflammatory response and have the opportunity to interact with the macrophages at the tissue/material interface.1 We have shown previously that

lymphocytes are capable of adhering to biomaterial surfaces primarily through contact with macrophages/FBGCs in co-cultures; however, lymphocytes enhanced monocyte adhesion and macrophage fusion to form FBGCs through indirect or paracrine signaling.2

Therefore, both direct (juxtacrine) interactions via cell-cell adhesion mechanisms and

indirect (paracrine) interactions, mediated by cytokine and chemokine mechanisms, participate in lymphocyte and macrophage interactions at the tissue/material interface.

However, we have yet to gain a complete understanding of how lymphocytes and

macrophages interact at the implant site and affect macrophage behavior.

The lymphocyte population, which includes T lymphocytes (T cells), natural

killer (NK) cells, and B lymphocytes (B cells), participate in the host defense against

foreign pathogens. T cells are divided into primarily CD8+ cytotoxic T cells and CD4+

- 94 - Chapter III

type 1 and type 2 helper T cells (Th1 and Th2, respectively). Upon activation, these

lymphocytes produce distinct cytokines and perform specific roles. CD8+ T cells and

NK cells mediate cellular killing. Th1 and NK cells produce interferon (IFN)-γ while

Th2 cells secrete IL-4, IL-5, IL-10, IL-13.3 B cells respond when activated by producing

antigen-specific antibodies. Therefore, lymphocytes utilize direct and indirect

mechanisms in responding to stimuli and communication with each other as well as with

other cell types.

Lymphocytes are capable of activating or enhancing the activation of

macrophages through antigen-dependent and antigen-independent direct cell-cell contact

and indirect interactions. T-cell receptor (TCR)-activated T lymphocytes and cytokine-

activated T lymphocytes are capable of activating macrophages albeit resulting in differential production of inflammatory mediators.4 Membrane TNF-α, CD69, and the

CD40-CD40L signaling pathways appear to be involved in the contact-mediated activation.4 Lymphocyte stimulation of macrophages can also occur by non-contact

mechanisms through secretion of soluble factors. T cell-derived IL-4, IL-13, and IFN-γ or natural killer cell-derived IFN-γ can induce distinct forms of macrophage activation.3,5,6 These different stimuli, contact or cytokine-mediated, induce activated

macrophages to secrete a multitude of cytokines and chemokines such as IL-1β, TNF-α,

IL-6, IL-8, IL-10 with the profiles dependent on stimuli.4,5,7

Monocyte/macrophage cytokine responses to biomaterials have been examined

both in vitro and in vivo. Many studies have shown differential monocyte/macrophage

responses to various biomaterials using inflammatory cytokines such as TNF-α, IL-1β,

IL-6, and IL-8 as indicators of cellular activation.8-12 Recent proteomic investigations in

- 95 - Chapter III

our laboratory have shown material-dependent production of cytokines such as IL-1β, IL-

6, IL-10; chemokines such as IL-8, MIP-1β, MCP-1; and ECM-remodeling proteins such

as MMP-9, TIMP-1, and TIMP-2 from adherent macrophages/FBGCs populations and co-cultures with lymphocytes.13,14

To further investigate lymphocyte effects on adherent macrophage behavior, monocytes were cultured alone or together with lymphocytes on polyethylene

terephthalate (PET)-based photograft co-polymerized material surfaces displaying

distinct hydrophobic, hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic

chemistries. We hypothesized that lymphocytes differentially modulate the production of macrophage-derived inflammatory mediators in response to biomaterial surfaces. This study examined the role of paracrine (indirect) and juxtacrine (direct) lymphocyte

interactions on macrophage activation as measured by macrophage-derived cytokines, chemokines, and ECM-modifying proteins. In addition, varying biomaterial surface

chemistries were utilized to explore how these interactions were influenced by surface

properties. The results from this investigation provide additional insight into the role

lymphocytes play in the foreign body reaction and the interfacial interaction mechanisms

that exist between lymphocytes and macrophages at the biomaterial implant site.

Materials and Methods

Biomaterial Surfaces

Synthesis of the PET-based photograft co-polymerized surfaces is described

elsewhere.15 Briefly, sheets of Mylar® PET were initially coated with poly(benzyl N,N- diethyldithiocarbamate-co-styrene) (BDEDTC) and then photograft co-polymerized with

- 96 - Chapter III

acrylamide (PAAm), sodium salt of acrylic acid (PAANa), and N-methiodide of dimethylamino propylacrylamide (DMAPAAmMeI) to provides surfaces with distinct

hydrophobic, hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic surface

characteristics, respectively. Biomaterial surfaces were punched into 1.5 cm diameter

disks and sterilized via a 100% ethanol wash prior to use. The material samples were

placed into 24-well tissue culture polystyrene (TCPS) plates (Fisher Scientific,

Pittsburgh, PA) and secured in place at the bottom of the well under sterile conditions by

silicone rings that were subjected to 5 minutes of sonication and autoclave-sterilization.

Prior to cell seeding, surfaces in each well were washed with phosphate-buffered saline

containing magnesium chloride and calcium chloride (Invitrogen, Grand Island, NY)

(PBS++) to remove any residual ethanol from sterilization.

Monocyte and Lymphocyte Isolation and Culture

Monocytes and lymphocytes were isolated from non-medicated peripheral blood

of healthy adult volunteers utilizing a non-adherent centrifugation method as previously

described.16 The monocyte and lymphocyte populations were separately suspended in serum free medium (SFM) (Invitrogen, Grand Island, NY) supplemented with L- glutamine, antibiotics, and antimycotics along with 20% autologous serum (AS) for plating. 5 x 105 monocytes were plated on each of the biomaterial surfaces and incubated

at 37˚C and 5% CO2. After 2 hours, non-adherent cells were removed, surfaces were

washed once with warm PBS++, and wells were replenished with 1 mL of supplemented

SFM containing 20% AS with and without 1.5 x 106 lymphocytes. The lymphocyte

population was added both directly to the adherent monocytes (direct culture) and in 0.02

μm Anapore membrane transwell inserts (Nunc, Naperville, IL) separated from the

- 97 - Chapter III

adherent monocytes (indirect culture). Supernatants, collected after 3, 7, and 10 days,

were centrifuged at 3000 rpm and cell-free supernatants were stored at -20˚C for analysis.

Determination of Adherent Cell Densities

Adherent cells were washed twice with PBS++ to remove non-adherent cells and

residual media, fixed with 100% methanol for 5 minutes, air-dried, and stained with May-

Grünwald/Giemsa. The staining process involved the addition of May-Grünwald reagent

for 5 minutes followed by a PBS++ wash and then addition of the Giemsa reagent. After

incubation for 15 minutes at room temperature, the surfaces were finally washed twice

with dH20 and allowed to air-dry overnight. The adherent monocyte, macrophage, and

foreign body giant cell density (cells/mm2) was determined by counting nuclei from 5

fields under 20X objective using optical microscopy. Cells containing greater than 3

nuclei were considered foreign body giant cells.

Determination of Macrophage/FBGC Activation

Cytokine production was quantified by utilizing ELISA (R&D Systems,

Minneapolis, MN) for IL-1β, TNF-α, IL-6, IL-8, IL-10, MMP-9, TIMP-1, and TIMP-2,

performed according to manufacturer’s protocol, along with an EL808 ultra microplate

reader and KC Junior software (Bio-Tek Instruments, Inc., Winooski, Vermont). To

represent cellular activation, the concentration of cytokines produced was divided by the

adherent macrophage/FBGC density which yields the amount of cytokines produced per

adherent cell.

Statistical Analysis

All results were obtained by repeating experiments at least 4 times utilizing

different donors (n ≥ 4) and presented as an average ± the standard error of the mean

- 98 - Chapter III

(SEM). All data were analyzed by Minitab statistical software (Minitab Inc., State

College, PA) and comparisons made utilizing ANOVA and the Tukey post hoc test to determine significant differences.

Results

IL-1β, TNF-α, IL-6, IL-8, MIP-1β, IL-10, MMP-9, TIMP-1 and TIMP-2, were quantified in terms of production of these inflammatory mediators per adherent cell.

Adherent macrophages and foreign body giant cells on various biomaterial surfaces were introduced to lymphocytes both directly and indirectly in order to examine the effects of the two mechanisms of interaction on adherent cell activation as measured by inflammatory mediator production. We previously demonstrated material-dependent production of these cytokines, chemokines, and ECM-modifying proteins by adherent macrophages and foreign body giant cells.14 In all cultures, regardless of the presence of lymphocytes or period of culture time, the hydrophilic/neutral surface, PAAm, induced the highest level of activation which is consistent with our previous work.14 This is demonstrated in Figures 3.1-3.7. The hydrophilic/anionic surface, PAANa, also showed increased levels of activation in terms of TNF-α and MIP-1β production (Figures 3.2 and

3.5).

Pro-inflammatory Mediators

IL-1β production per cell was highest on the hydrophilic/neutral surface, PAAm, over all time points and cultures as shown in Figure 3.1(a-c). The presence of lymphocytes, interacting via soluble factors, showed a general trend towards increased activation at day 3 and day 7 of culture. This increase over adherent macrophage/FBGC

- 99 - Chapter III

cultures and direct co-culture with lymphocytes was significant only on the hydrophobic

PET surface [Figure 3.1(d-e)]. The hydrophobic surface, BDEDTC, and the hydrophilic/cationic surface, DMAPAAmMeI, showed trends of increased activation

only during the early time period of the cultures.

Figure 3.1: IL-1β production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. Results on the PAAm surface were removed for demonstrating lymphocyte effects on the other surfaces over (D) 3, (E) 7, and (F) 10 days of culture. Data represents the mean ± standard error of the mean, n = 4-5. **Significance compared to direct and monocyte-only cultures (p < 0.05).

Although TNF-α was most highly produced from adherent macrophages and

FBGCs on the hydrophilic/neutral and hydrophilic/anionic surfaces, lymphocytes did not

alter activation on these two surfaces [Figure 3.2(a-c)]. Lymphocytes did not

significantly influence activation until day 10 on the two hydrophobic surfaces, PET and

- 100 - Chapter III

BDEDTC, as illustrated in Figure 3.2(d-f). At that time, TNF-α production per cell was increased in direct lymphocyte co-cultures compared to both the monocyte-only culture and indirect co-culture (p < 0.01).

Figure 3.2: TNF-α production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. Results on the PAAm and PAANa surfaces were removed for demonstrating lymphocyte effects on the other surfaces over (D) 3, (E) 7, and (F) 10 days of culture. Data represents the mean ± standard error of the mean, n = 4-5. **Significance compared to indirect and monocyte-only cultures (p < 0.01).

IL-6 and IL-8 material-dependent activation were similar to IL-1β in that the hydrophilic/neutral surface, PAAm, induced the greatest activation [Figure 3.3 and 3.4

(a-c)]. These were similar culture trends early (day 3) with indirect co-cultures increasing IL-6 and IL-8 production per cell significantly on DMAPAAmMeI over the other two culture types and also significantly increasing IL-8 production per cell relative

- 101 - Chapter III to adherent macrophage/FBGC cultures on PET. By day 10, direct lymphocyte/macrophage co-cultures on BDEDTC showed increased IL-6 and IL-8 over the monocyte culture without lymphocytes (p < 0.05). Additionally at day 10, direct co- cultures on PET induced a higher level of IL-8 production per cell compared to the other

two culture types (p < 0.01).

Figure 3.3: IL-6 production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. Results on the PAAm surface were removed for demonstrating lymphocyte effects on the other surfaces over (D) 3, (E) 7, and (F) 10 days of culture. Data represents the mean ± standard error of the mean, n = 4-5. **Significance compared to direct and monocyte-only cultures (p < 0.01). *Significance compared to monocyte-only culture (p < 0.05).

- 102 - Chapter III

Figure 3.4: IL-8 production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. Results on the PAAm surface were removed for demonstrating lymphocyte effects on the other surfaces over (D) 3, (E) 7, and (F) 10 days of culture. Data represents the mean ± standard error of the mean, n = 4-5.

In terms of MIP-1β production, the material-dependent activation was the same as for TNF-α where PAAm and PAANa induced a higher level of MIP-1β production over all other surfaces as shown in Figure 3.5(a-c). In Figure 3.5(d-f), we see similar culture and time-dependent trends as the other pro-inflammatory cytokines and chemokines with indirect co-cultures significantly increasing IL-8 production per cell on the PET surface on day 7.

- 103 - Chapter III

Figure 3.5: MIP-1β production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. Results on the PAAm and PAANa surfaces were removed for demonstrating lymphocyte effects on the other surfaces over (D) 3, (E) 7, and (F) 10 days of culture. Data represents the mean ± standard error of the mean, n = 4-5. **Significance compared to direct and monocyte-only cultures (p < 0.05).

Anti-inflammatory Cytokine

IL-10 was the only anti-inflammatory mediator quantified. The data in Figure

3.6(a-c) shows that once again, PAAm was highly activating. However, unlike the pro- inflammatory cytokines and chemokines, IL-10 production per cell was not significantly affected by the presence of lymphocytes [Figure 3.6(d-f)].

- 104 - Chapter III

Figure 3.6: IL-10 production: lymphocyte effects on macrophages/FBGCs over (A) 3, (B) 7, and (C) 10 days of culture. Results on the PAAm surface were removed for demonstrating lymphocyte effects on the other surfaces over (D) 3, (E) 7, and (F) 10 days of culture. Data represents the mean ± standard error of the mean, n = 4-5.

Matrix metalloproteinases and Tissue Inhibitors

As shown in Figure 3.7, other than higher levels of MMP-9, TIMP-1, and TIMP-2 production on hydrophilic/neutral surfaces, lymphocyte containing co-cultures did not significantly influence production of these extracellular matrix-modifying proteins from adherent macrophages and FBGCs.

- 105 - Chapter III

Figure 3.7: Lymphocyte effects on macrophage/FBGC production of MMP-9 over (A) 3, (B) 7, and (C) 10 days of culture; TIMP-1 over (D) 3, (E) 7, and (F) 10 days of culture; and TIMP-2 over (G) 3, (H) 7, and (I) 10 days of culture. Data represents the mean ± standard error of the mean, n = 4-5. *Significance compared to direct culture (p < 0.05).

Discussion

Lymphocyte involvement in the biological response to biomaterial implants has received little attention despite their presence at the implant site which allows opportunity to interact with macrophages as well as other cell types. Lymphocyte effects on macrophage behavior in the foreign body reaction are now becoming identified.

- 106 - Chapter III

Brodbeck et al. previously demonstrated that lymphocytes can increase the level of

monocyte adhesion on the biomaterial surface as well as macrophage fusion to form

FBGCs.2 In light of these observations, we hypothesized that lymphocytes increased

adherent macrophage activation. The objectives of the current investigation were to

determine the effect lymphocytes have on macrophage activation as measured by the

production of inflammatory mediators and how the response could be influenced by

biomaterial surface chemistry.

Lymphocytes were cultured directly and indirectly with a population of adherent

monocytes, macrophages, and FBGCs. Initially at day 3, increases in macrophage/FBGC

activation were mediated by lymphocyte-derived soluble factors. The enhancement via

indirect (paracrine) mechanisms was observed in IL-1β, IL-6, and IL-8 production.

Direct co-cultures allowing both engagement of direct cell-cell contact and indirect

interactions between lymphocytes and adherent macrophages/FBGCs failed to result in

increased activation compared to cultures without lymphocytes. These results suggest

that initial cell-cell interactions may inhibit the action of the soluble signal. As time progressed, indirect soluble factors no longer significantly increased or altered adherent

cellular activation. Lymphocyte-derived IFN-γ and TNF are capable of activating and

enhancing macrophage production of cytokines such as IL-1, TNF, and IL-6.5 CD8+ T

cells, CD4+ Th1 cells, and NK cells are capable of producing IFN-γ and TNF when given

appropriate stimulation.3,6 Previous investigations of inflammatory mediators from

lymphocyte/macrophage co-cultures by protein cytokine array failed to detect IFN-γ, but

did detect TNF-α.13 Preliminary results utilizing ELISA with higher sensitivity

(minimum detection of 8 pg/mL) show that IFN-γ is produced in lymphocyte/macrophage

- 107 - Chapter III

co-cultures (unpublished data) suggesting that the failure to detect IFN-γ by protein array

(minimum detection of 100 pg/mL) could be due to limitations in the sensitivity of the

array. However, the specific cellular source of IFN-γ in the lymphocyte population is

currently unknown.

Direct cell-cell contact between the lymphocytes and macrophages over the 10

days of culture resulted in enhancement of pro-inflammatory cytokines TNF-α, IL-6, and

IL-8 although initially there was no such increase despite significant involvement of

soluble factors in the indirect culture system. This time-varying response in direct co-

cultures could possibly be due to macrophage or FBGC phenotypic changes over the 10

days of culture. This is the first evidence of lymphocyte contact-mediated modulation of

biomaterial surface-adherent macrophage and FBGC behavior. Not only are activated T

cells capable of influencing monocyte/macrophage responses through soluble factors, but

they are also able to induce monocyte/macrophage production of cytokines and

chemokines via membrane molecules such as membrane TNF-α, CD69, CD40, and LFA-

1.4,17,18 T cells can be activated by TCR-dependent and TCR-independent, or cytokine-

stimulated, mechanisms. The route of T cell activation dictates the contact-mediated

monocyte/macrophage cytokine profile. For instance, lymphocytes activated via the TCR

induced both TNF-α and IL-10 while lymphocytes activated via cytokines such as IL-6,

TNF-α, IL-2 or IL-15 induced TNF-α but not IL-10.4,17,19 Due to lymphocyte contact-

mediated induction of greater levels of only TNF-α and not IL-10 in the lymphocyte/macrophage co-cultures on biomaterial surfaces in this study, we suggest that the observed responses are most probably due to cytokine-activated T lymphocytes.

- 108 - Chapter III

Lymphocytes appear to promote and perpetuate significant pro-inflammatory

responses as the results show lymphocyte enhancement of pro-inflammatory cytokines

and chemokines with no significant effect on IL-10, an anti-inflammatory cytokine

known to suppress immune and inflammatory responses. T cells could be the effector

population responsible for the observed responses through secretion of soluble factors

such as IFN-γ as well as providing contact-mediated signals.3 The lack of detection of

other lymphokines in similar lymphocyte/macrophage co-cultures suggest that the T

lymphocytes could be in a state of hyporesponsiveness.13 If that is the case, NK cells

could be an alternative source of the soluble mediator(s).20 Despite being

hyporesponsive, T cell are still capable of increasing monocyte/macrophage activation

via contact mechanisms.4 The lymphocytes in the lymphocyte/macrophage co-cultures

on these PET-based surfaces, showed no influence on MMP-9, TIMP-1, or TIMP-2

production. MMP-9, TIMP-1, and TIMP-2 analyses were selected for this investigation since these are matrix-modifying proteins detected previously with biomaterial-adherent monocytes/macrophages/FBGCs.21

Material surface chemistries are known to elicit variable cellular responses. In

this study, material surface chemistries induced varying levels of macrophage activation.

Hydrophilic/neutral surfaces were the most highly activating surfaces while

hydrophilic/anionic surfaces showed instances of high activation for particular cytokines.

These results were consistent with our previous findings exploring material surface

effects on cytokine, chemokine, and ECM-modulating protein production from adherent

macrophages and FBGCs.14,21 The presented results further emphasize the fact that

macrophage/FBGC activation does not correlate with adherent cellular density.

- 109 - Chapter III

The current findings demonstrate that activation for pro-inflammatory cytokines and chemokines is enhanced by lymphocytes in both direct and indirect co-cultures but at different time points. Hydrophilic/neutral and hydrophilic/anionic surfaces induced the highest levels of activation per cell but the presence of lymphocytes on these surfaces did not significantly influence inflammatory mediator production when compared to cultures without lymphocytes. It is possible that the cells were maximally activated to produce the pro-inflammatory cytokines; if so, the presence of lymphocytes would have no further effect.

Overall, interactions through indirect soluble factor mechanisms show significant effects on adherent macrophage and FBGC activation at early time points while interactions through direct cell-cell mechanisms dominate at later time points. These direct and indirect lymphocyte interactions are pro-inflammatory as they enhance pro- inflammatory cytokines but do not influence IL-10, an anti-inflammatory cytokine, or particular ECM-modulating proteins. The results from this investigation show lymphocyte enhancement of adherent macrophage/FBGC activation via both direct and indirect mechanisms of interaction and further demonstrate lymphocyte involvement in the inflammatory and foreign body responses. Moreover, our findings support our hypothesis that biomaterial material chemistries differentially affect lymphocyte modulation of adherent macrophage and FBGC-derived cytokines and chemokines.

Further investigations are ongoing for the specific lymphocyte-derived molecules, soluble and membrane-bound, which mediate the observed macrophage and FBGC responses on biomaterial surfaces. Our findings contribute to gaining a better understanding of the biomaterial-dependent and independent cellular and molecular interactions occurring in

- 110 - Chapter III

the foreign body reaction at the surface of implanted biomaterials in order to provide

insight for the design of biomaterials with improved biocompatibility and optimal functional application.

References

1. Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81-110.

2. Brodbeck WG, Macewan M, Colton E, Meyerson H, Anderson JM. Lymphocytes and the foreign body response: lymphocyte enhancement of macrophage adhesion and fusion. J Biomed Mater Res A 2005;74(2):222-9.

3. Santana MA, Rosenstein Y. What it takes to become an effector T cell: the process, the cells involved, and the mechanisms. J Cell Physiol 2003;195(3):392- 401.

4. Monaco C, Andreakos E, Kiriakidis S, Feldmann M, Paleolog E. T-cell-mediated signalling in immune, inflammatory and angiogenic processes: the cascade of events leading to inflammatory diseases. Curr Drug Targets Inflamm Allergy 2004;3(1):35-42.

5. Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 2004;25(12):677-86.

6. O'Connor GM, Hart OM, Gardiner CM. Putting the natural killer cell in its place. Immunology 2006;117(1):1-10.

7. Burger D, Dayer JM. Cytokines, acute-phase proteins, and hormones: IL-1 and TNF-alpha production in contact-mediated activation of monocytes by T lymphocytes. Ann N Y Acad Sci 2002;966:464-73.

8. Hwang JJ, Jelacic S, Samuel NT, Maier RV, Campbell CT, Castner DG, Hoffman AS, Stayton PS. Monocyte activation on polyelectrolyte multilayers. J Biomater Sci Polym Ed 2005;16(2):237-51.

9. DeFife KM, Yun JK, Azeez A, Stack S, Ishihara K, Nakabayashi N, Colton E, Anderson JM. Adhesion and cytokine production by monocytes on poly(2- methacryloyloxyethyl phosphorylcholine-co-alkyl methacrylate)-coated polymers. J Biomed Mater Res 1995;29(4):431-9.

- 111 - Chapter III

10. Ma N, Petit A, Yahia L, Huk OL, Tabrizian M. Cytotoxic reaction and TNF-alpha response of macrophages to polyurethane particles. J Biomater Sci Polym Ed 2002;13(3):257-72.

11. Sethi RK, Neavyn MJ, Rubash HE, Shanbhag AS. Macrophage response to cross- linked and conventional UHMWPE. Biomaterials 2003;24(15):2561-73.

12. Marques AP, Reis RL, Hunt JA. Cytokine secretion from mononuclear cells cultured in vitro with starch-based polymers and poly-L-lactide. J Biomed Mater Res A 2004;71(3):419-29.

13. Chang DT, Jones JA, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Lymphocyte/macrophage interactions: biomaterial surface-dependent cytokine, chemokine, and matrix protein production. J Biomed Mater Res A 2008.

14. Jones JA, Chang DT, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Proteomic analysis and quantification of cytokines and chemokines from biomaterial surface-adherent macrophages and foreign body giant cells. J Biomed Mater Res A 2007;83(3):585-96.

15. Nakayama Y, Anderson JM, Matsuda T. Laboratory-scale mass production of a multi-micropatterned grafted surface with different polymer regions. J Biomed Mater Res 2000;53(5):584-91.

16. McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 1994;91(21):10119-23.

17. Brennan FM, Foey AD, Feldmann M. The importance of T cell interactions with macrophages in rheumatoid cytokine production. Curr Top Microbiol Immunol 2006;305:177-94.

18. Parry SL, Sebbag M, Feldmann M, Brennan FM. Contact with T cells modulates monocyte IL-10 production: role of T cell membrane TNF-alpha. J Immunol 1997;158(8):3673-81.

19. Sebbag M, Parry SL, Brennan FM, Feldmann M. Cytokine stimulation of T lymphocytes regulates their capacity to induce monocyte production of tumor necrosis factor-alpha, but not interleukin-10: possible relevance to pathophysiology of rheumatoid arthritis. Eur J Immunol 1997;27(3):624-32.

20. Lodoen MB, Lanier LL. Natural killer cells as an initial defense against pathogens. Curr Opin Immunol 2006;18(4):391-8.

21. Jones JA, McNally AK, Chang DT, Qin LA, Meyerson H, Colton E, Kwon IL, Matsuda T, Anderson JM. Matrix metalloproteinases and their inhibitors in the foreign body reaction on biomaterials. J Biomed Mater Res A 2008;84(1):158-66.

- 112 - Chapter IV

The Effect of Biomaterial Surface Chemistry on Adherent Lymphocyte Interactions with Macrophages and Foreign Body Giant Cells

Abstract

Lymphocytes have been shown to be capable of adhering onto protein preadsorbed surfaces as well as synthetic polymer surfaces in hemodialysis and column cell separation. Although these earlier findings provide some insight into lymphocyte interactions with biomaterials, they are not completely relevant to conditions relating to

biomaterials after implantation and during the foreign body reaction. In this study,

lymphocytes were co-cultured with different concentrations of monocytes on a set of

hydrophobic, hydrophilic/neutral, and hydrophilic/anionic biomaterial surfaces. After 3

days, surface adherent cells were analyzed utilizing immunofluorescence and phase

contrast imaging. The aim was to quantitatively evaluate lymphocyte adhesion as well as characterize the effects of adherent macrophages and biomaterial surface chemistries on these interactions. We found that all surfaces showed limited direct biomaterial-adherent lymphocytes regardless of the presence of macrophages or foreign body giant cells

(FBGC). In situations with higher levels of adherent macrophages and FBGCs, greater

than 90% of the adherent lymphocytes were interacting with adherent macrophages and

FBGCs. With minimal macrophages and FBGCs, there were still greater than 55%

macrophage- and FBGC-adherent lymphocytes. CD4+ and CD8+ T lymphocytes

comprised over 95% of identified adherent lymphocytes while CD56+ NK cells were

minimally present. The hydrophilic/anionic surface promoted higher levels of

macrophage- and FBGC- adherent lymphocytes but was nonselective for lymphocyte

- 113 - Chapter IV

subtype interactions. The hydrophilic/neutral surface was more selective for CD4+ T

lymphocyte interactions with adherent macrophages and FBGCs but showed low

lymphocyte adhesion. The hydrophobic surface also evoked low levels of lymphocyte

adhesion and interaction despite the capability to support abundant levels of adherent

macrophages but was more selective for CD8+ T lymphocyte interactions. These results

demonstrate that biomaterial surface chemistries can influence the number as well as type

of adherent lymphocyte interactions.

Introduction

Implantation of synthetic biomaterials elicits a foreign body reaction consisting of

monocyte adhesion, differentiation to macrophages, and subsequent macrophage fusion

to form foreign body giant cells (FBGC). Biomaterials induce adherent macrophage

activation and secretion of signaling factors to guide the inflammatory and wound healing

response. Lymphocytes transiently appear at the implant site during the inflammatory

response1 and there is evidence to indicate that they play a role in the tissue response to

implanted biomaterials. Lymphocytes have been shown to influence macrophage

behavior at biomaterial surfaces in vitro through enhancement of monocyte/macrophage adhesion, macrophage fusion, and macrophage activation.2,3 These effects are mediated

by indirect and direct lymphocyte interactions with monocytes, macrophages, and

FBGCs, but we still do not have a full mechanistic understanding of these interactions.

In addition to interactions occurring between lymphocytes and adherent macrophages

and/or FBGCs, knowledge on lymphocyte interactions with the biomaterial surface itself

is limited.

- 114 - Chapter IV

The fact that synthetic materials dictate varied cellular responses based on their

surface characteristics further illustrates the complexity in designing or selecting a

particular biomaterial that would be conducive to or potentially contribute to the desired

functionality of the therapeutic device. Certain surface properties can elicit more intense

levels of adherent macrophage activation compared to others or evoke varying cytokine

profiles. The biomaterial-dependent as well as biomaterial-independent responses must

be thoroughly understood for the proper design and utilization of biomaterials as an

enabling technology in biomedical applications.

Lymphocytes are white blood cells that perform a variety of actions in the

immune system to protect the body from foreign antigen. Subtypes of lymphocytes

include T lymphocytes (T cells), B lymphocytes (B cells), and natural killer cells (NK).

Lymphocytes within the same class can be capable of performing several functions and

thus have more specific nomenclature based on their differentiating characteristics. T

cells can directly kill infected cells through induction of apoptosis. These cells possess

CD8 markers on their cell surface and are referred to as CD8+ T suppressor/cytotoxic

cells. T cells carrying CD4 markers are known for their ability to affect other cell types

by producing cytokines and chemokines. CD4+ T cells can be type 1 T helper cells

(Th1) which secrete IFN-γ and TNF-β or type 2 T helper cells (Th2) which produce IL-4,

IL-5, IL-10, and IL-13. B cells contribute to the immune response by generating

antibodies that are specific to the antigen. NK cells are capable of mediating cell death as

well as secreting cytokines such as IFN-γ. The cell surface marker CD56 differentiates

NK cells from the other white blood cells. All of these cell types circulate in the peripheral blood surveying the body for foreign invaders.

- 115 - Chapter IV

Lymphocytes are capable of adhering onto surfaces composed of synthetic

materials and are affected by surface properties. The use of hemodialysers composed of

different synthetic polymers showed differential lymphocyte elution profiles after

hemodialysis.4,5 Similarly, various polymers have shown the capability to retain specific

subtypes of lymphocytes in the context of in vitro column separation methods.6,7

Additionally, lymphocyte adhesion on materials pre-adsorbed with proteins such as

fibronectin and vitronectin showed the differential effects of biomaterial and adsorbed

protein.8-10 More recently, Ito et al. demonstrated that an electrically charged polymer

could control lymphocyte adhesion.11 However, these results provide limited information

on lymphocyte adhesion behavior on biomaterial surfaces after implantation.

In this study, we investigated lymphocyte behavior on biomaterial surfaces in situations with minimal and abundant adherent macrophages. We hypothesized that lymphocytes will adhere to biomaterial surface and that adhesion is enhanced by the presence of adherent macrophages. To explore this, human peripheral blood lymphocytes were exposed to different biomaterial surfaces in vitro. We quantitatively evaluated lymphocyte subsets on the surfaces after a specified culture period and characterized the effect of adherent macrophages and biomaterial surface chemistries on the lymphocyte adhesion behavior. This study provides additional knowledge in regard to lymphocyte behavior at biomaterial surfaces and the effects biomaterial surface characteristics have on cell-cell and cell-material interactions.

- 116 - Chapter IV

Materials and Methods

Materials

24-well tissue culture polystyrene (TCPS) plates were acquired from Fisher

Scientific (Pittsburgh, PA). Macrophage serum-free medium (SFM) was obtained from

Invitrogen, Grand Island, NY. Mouse anti-human CD14 (clone M5E2) and mouse anti-

human CD8 (clone UCH-T4) were obtained from Santa Cruz Biotechnology, Inc. (Santa

Cruz, CA), mouse anti-human CD56 (clone 123C3) from Abcam, Inc. (Cambridge, MA), goat anti-human CD4 from R&D Systems (Minneapolis, MN), and normal goat and

mouse control IgG from Santa Cruz Biotechnology, Inc. Other reagents were obtained as

follows: RNase A from EMD Biosciences, Inc. (La Jolla, CA), Alexa Fluor 594 donkey

anti-mouse and anti-goat IgG from Invitrogen, YO-YO-1 from Molecular Probes

(Eugene, OR), donkey serum from Sigma (St. Louis, MO), and Gel/Mount from Biomeda

(Foster City, CA).

Preparation of Biomaterial Surfaces

Mylar® PET surfaces were modified as described previously 12 with surface

characterization described in Chapter I. Initially, PET surfaces were coated with

poly(benzyl N,N-diethyldithiocarbamate-co-styrene) (BDEDTC). Next, UV exposure in the presence of acrylamide (PAAm) and sodium salt of acrylic acid (PAANa) provided

grafted material substrates with distinct hydrophilic/neutral and hydrophilic/anionic surface chemistries, respectively. Unmodified PET provided a hydrophobic/neutral surface. The PET-based photograft copolymerized surfaces were cut into 1.5 cm diameter disks for sterilization by 100% ethanol prior to insertion into 24-well TCPS plates. Rings cut from silicone tubing of O.D. 5/8 and I.D. 3/8 (Cole Parmer, Vernon

- 117 - Chapter IV

Hills, IL) were sonicated for 5 minutes in 100% ethanol, autoclave sterilized, and used to

secure the material surfaces at the bottom of the well. Phosphate-buffered saline

containing magnesium chloride and calcium chloride (Invitrogen) (PBS++) was utilized to

wash surfaces and remove any residual ethanol on the material surfaces.

Monocyte and Lymphocyte Cell Culture

Monocytes and lymphocytes were isolated from peripheral blood of healthy non-

medicated donors by a non-adherent method as previously described13 and suspended

separately in SFM containing L-glutamine, antibiotics, antimycotics, and 20% autologous

serum (AS). As determined by flow cytometry previously (Chapter II),14 the isolated

monocyte populations contained, on average, 60.5% ± 9.1% monocytes and 39.5 ± 9.1%

contaminating lymphocytes while the isolated lymphocyte population consisted of, on

average, 51.3% ± 4.1% CD4+ T lymphocytes, 30.8% ± 5.2% CD8+ T lymphocytes,

10.7% ± 1.1% NK cells, 5.6% ± 0.5% B lymphocytes, and 1.6% ± 0.1% contaminating

monocytes. Monocytes were plated onto the PET, PAAm, and PAANa surfaces at 5 x

105 (100%), 2.5 x 105 (50%), and 0.5 x 105 (10%) cells in 0.25 mL SFM with 20% AS,

++ and allowed to incubate for 2 hours at 37˚C and 5% CO2. After a subsequent PBS wash to remove non-adherent cells, 1.5 x 106 lymphocytes (100%) in 0.5 mL SFM with

20% AS were added to each well with lymphocyte/monocyte ratios designated as

100%/100% (1/1.0), 100%/50% (1/0.5), and 100%/10% (1/0.1). Additionally, 1.5 x 106 lymphocytes in 0.5 mL SFM with 20% AS were plated onto PET surfaces (1/0). After 3 days, adherent cells were washed twice with PBS++ and fixed with acetone for 2 minutes

at -20˚C.

- 118 - Chapter IV

Lymphocyte Subtype Identification by Immunofluorescence

Adherent cells on material surfaces were prepared for immunofluorescent staining

by fixation with acetone for 2 minutes at -20˚C and air-dried. Samples were stored at 4˚C prior to staining. PBS++ was utilized as the diluent and wash buffer in all required

staining steps. Samples were first treated with RNase A (100 μg/mL) for 1.5 hours at

37˚C and then washed 3 times for 5 minutes each. Next, nonspecific sites were blocked

with 10% donkey serum for 1 hour at 37˚C. Then, primary detecting antibodies for

CD14, CD8, CD4, CD56 and isotype-matched control IgG were applied at 15 μg/mL

diluted with 1% donkey blocking serum for 1 hour at 37˚C. After 4 washes at 5 minutes

each, a secondary staining solution of YO-YO-1 (0.1 μmol/L diluted at 1:10,000 in

PBS++) and either Alexa Fluor 594 donkey anti-mouse or anti-goat IgG (20 μg/mL

diluted at 1:100 in PBS++) was added for 1 hour at 37˚C. Samples were then washed 3 times for 10 minutes each. Surfaces were removed from the 24-well plates, transferred to glass slides, and mounted under glass coverslips using Gel/Mount. Samples were imaged by fluorescence microscopy (Olympus IX71) using the Olympus Microsuite software with settings that blacken residual background fluorescence from corresponding nonspecific control antibodies. Positively detected cells were quantified in 10 fields under 40X objective. Utilizing YO-YO-1 to visualize nuclei and phase contract imaging of the same field, localization and physical interactions of identified cells were evaluated.

Statistical Analysis

Multiple human donors were utilized to account for donor variability. All results were presented as an average ± the standard error of the mean (SEM) (n = 3). Statistical analysis was performed utilizing Minitab statistical software (Minitab Inc., State College,

- 119 - Chapter IV

PA) and statistically significant differences were determined by ANOVA and the Tukey

post hoc test.

Results

Differential Effects on Lymphocyte Adhesion

We utilized an immunocytochemical staining approach with fluorescent

microscopy to quantitatively evaluate lymphocyte adhesion on model biomaterial surfaces. The surfaces were exposed to various initial quantities of monocytes in order to explore lymphocytes adhesion on biomaterials in the presence of minimal and abundant macrophages and FBGCs. All surfaces were examined after a period of 3 days sufficient for monocyte to macrophage development with possible foreign body giant cell development. The major types of lymphocytes in the cell isolations included T cells and

NK cells. Anti-CD4 antibodies were used to identify T helper cells, anti-CD8 to identify

T suppressor cells, and anti-CD56 to identify NK cells. Because B cells were such a small percentage of the isolated lymphocytes, they were not pursued.

Examination of the total adherent lymphocyte density in the different co-cultures on each of the biomaterials showed significant lymphocyte adhesion in lymphocyte/monocyte co-culture ratios of 1/0.5 and 1/1.0 on the hydrophilic/anionic

(PAANa) surface compared to the other surfaces (Figure 4.1A and Table 4.1). The difference from the other two surfaces was more than 4.5 fold. However, the PAANa surface was similar to the PAAm surface in that both evoked 10 fold increases in adherent lymphocyte density from 1/0.1 to 1/1.0 co-cultures while PET elicited an increase of less than 2 fold. In co-cultures with initial low monocyte levels, very few

- 120 - Chapter IV lymphocytes were adherent (< 50/mm2). These adherent lymphocytes were interacting with adherent macrophages and FBGCs in densities shown in Figure 4.1B and Table 4.2.

The ratios of adherent lymphocyte densities to adherent macrophage and FBGC densities

(Table 4.3) further illustrate the biomaterial differences. PAANa evoked the highest ratios, but PAAm also had similar higher ratios. PET ratios were extremely low (> 3 fold differences compared to the other two surfaces).

Table 4.1: Adherent Lymphocyte Densities (cells/mm2) on Biomaterial Surfaces with Varying Monocyte Concentrations

Surface 1/0.1 1/0.5 1/1.0

PET 36.9 ± 29.3 29.2 ± 8.6 57.5 ± 27.9 PAANa 23.2 ± 13.4 232.7 ± 66.9 259.3 ± 42.5 PAAm 2.6 ± 2.6 16.3 ± 13.8 29.2 ± 15.5 Mean ± SEM, n=3

Table 4.2: Adherent Macrophage and FBGC Densities (cells/mm2) on Different Surface Chemistries as a Function of Monocyte Concentration Plated

Adherent 1/0.1 1/0.5 1/1.0 Surface Cells Macrophage 48.1 ± 14.1 467.1 ± 174.8 629.4 ± 57.3 PET FBGC 0 ± 0 1.7 ± 1.7 4.3 ± 4.3 Total 48.1 ± 14.1 468.8 ± 174.7 633.7 ± 61.0 Macrophage 7.7 ± 7.7 123.6 ± 33.5 572.7 ± 114.1 PAANa FBGC 0 ± 0 32.4 ± 9.6 68.7 ± 23.8 Total 7.7 ± 7.7 156.1 ± 39.6 641.4 ± 99.0 Macrophage 6.9 ± 6.9 28.3 ± 6.8 83.3 ± 29.5 PAAm FBGC 0 ± 0 8.2 ± 5.8 14.5 ± 3.1 Total 6.9 ± 6.9 36.5 ± 6.0 97.7 ± 29.7 Mean ± SEM, n=3

- 121 - Chapter IV

350 A ** ** 300

250 ) 2 200

150 (cells/mm 100

Adherent Lymphocyte Density Lymphocyte Adherent 0 PET PET PET PET PAAm PAAm PAAm PAANa PAANa PAANa 1/0 1/0.1 1/0.5 1/1.0

800 B 700 ) 2 600

500

400 Macrophage

300 FBGC

200 Density (cells/mm * 100

Adherent Macrophage and FBGC FBGC and Macrophage Adherent 0 PET PET PET PET PAAm PAAm PAAm PAANa PAANa PAANa 1/0 1/0.1 1/0.5 1/1.0 Ratio of Lymphocyte/Monocyte Co-culture Concentrations

Figure 4.1: Adherent lymphocyte density (A) and adherent macrophage density (B) as a function of the ratio of lymphocyte/monocyte co-culture concentration after 3 days. Lymphocytes were co-cultured with different monocyte concentration for 3 days on each of the biomaterial surfaces. Results are expressed as the mean of 3 experiments ± the standard error of the mean. *Significant difference relative to other surfaces (p < 0.05). **Significance relative to other surfaces and 1/0.1 co-culture ratio. (p < 0.05).

- 122 - Chapter IV

Table 4.3: Ratio of Adherent Lymphocyte Densities to Adherent Macrophage/FBGC Densities on PET-based Surfaces

Surface 1/0.1 1/0.5 1/1.0 PET 0.8 0.1 0.1 PAANa 3.0 1.5 0.4 PAAm 2.7 0.5 0.3

Monocytes and macrophages are known to readily adhere to biomaterial surfaces and were identified by anti-CD14 antibodies. As expected in co-cultures of 1/0 and 1/0.1 lymphocyte/monocyte ratios, the number of adherent macrophages after 3 days was minimal or below 50/mm2. Despite the increase in the initial plating of monocytes,

PAAm evoked the lowest level of adherent macrophage and FBGC density after 3 days illustrated most clearly in 1/0.5 and 1/1.0 co-culture concentration ratios (> 5 fold decrease relative to other surfaces). As the initial plating of monocytes increased, the

PET and PAAm surfaces increased adherent densities ~10 fold over 1/0.1 to 1/1.0 co- cultures while PAANa evoked a 70 fold increase in macrophage and FBGC adherence.

These two surfaces both promoted macrophage/FBGC adhesion. The 3 fold difference in adhesion seen in 1/0.5 co-cultures can at least be partially explained by the induction of macrophage fusion on PAANa. As Figure 4.1B shows, FBGC formation occurred more on hydrophilic surfaces (PAANa and PAAm) but minimally on PET (< 5%). Although the percentage of FBGCs (~20% in 1/0.5 and ~15% in 1/1.0 co-culture) was similar between PAANa and PAAm, fusion occurred at the highest level on PAANa in terms of number and size of FBGCs.

Figure 4.2 illustrates how the adhesion of lymphocytes depends on the actual density of adherent macrophages and FBGCs on the biomaterial surface. Adherent

- 123 - Chapter IV lymphocyte density increased as the adherent macrophage/FBGC density increased but to a limit. Upon reaching a density of 100 macrophages and FBGCs/mm2, the limit was essentially approached.

350 * 300 *

250 PET )

2 PAANa 200 PAAm PET 150

(cells/mm PAANa 100 PAAm

50 Adherent Lymphocyte Density Density Lymphocyte Adherent

0 0 100 200 300 400 500 600 700

Adherent Macrophage and FBGC Density (cells/mm2)

Figure 4.2: Adherent lymphocyte density as a function of the adherent macrophage and FBGC density after 3 days of culture. Data represents the mean ± the standard error of the mean (n = 3). *Significant increase compared to the other two surfaces and 1/0.1 co-culture (p < 0.05).

Table 4.4: Density of Adherent Lymphocyte Subtypes (cells/mm2) on Biomaterial Surfaces over Varying Lymphocyte to Monocyte Co-culture Ratios

Lymphocyte Material 1/0.1 1/0.5 1/1.0 Subtype CD8+ 5.2 ± 5.2 8.6 ± 1.7 38.6 ± 20.8 PET CD4+ 31.8 ± 24.2 19.8 ± 10.8 17.2 ± 7.0 CD56+ 0 ± 0 0.9 ± 0.9 1.7 ± 0.9 CD8+ 6.9 ± 6.9 62.7 ± 11.2 89.3 ± 24.5 PAANa CD4+ 16.3 ± 15.0 169.2 ± 56.3 170.0 ± 52.3 CD56+ 0 ± 0 0.9 ± 0.9 0 ± 0 CD8+ 6.9 ± 0 1.7 ± 1.7 5.2 ± 1.5 PAAm CD4+ 2.6 ± 2.6 14.6 ± 14.6 24.0 ± 15.2 CD56+ 0 ± 0 0 ± 0 0 ± 0 Mean ± SEM, n=3

- 124 - Chapter IV

) 2 300 A

250

200 CD56 CD4 150 CD8

100

0 Adherent Lymphocyte Density (cells/mm Density Lymphocyte Adherent PET PET PAANa PAAm PET PAANa PAAm PET PAANa PAAm

1/0 1/0.1 1/0.5 1/1.0

100% B 90% 80% 70% 60% CD56 50% CD4 40% CD8 30% 20% 10% 0% Percent of Adherent Lymphocytes Adherent of Percent PET PET PET PET PAAm PAAm PAAm PAANa PAANa PAANa Isolation Lymphocyte 1/0 1/0.1 1/0.5 1/1.0 1/0 Ratio of Lymphocyte/Monocyte Co-culture Concentrations

- 125 - Chapter IV

Adherent Lymphocyte Subtypes and Interactions

As shown in Figure 4.3A and Table 4.4, adherent lymphocytes included CD4+ and CD8+ T cells and CD56+ natural killer cells. CD4+ and CD8+ T cells were found on all surfaces while CD56+ cells were only found on PET and PAANa surfaces.

Although CD56+ cells were capable of adhering and associating with macrophages, they accounted for less than 5% of the detected adherent lymphocytes which was over 2 fold lower than the percentage of CD56+ NK cells initially plated (Figure 4.3B). The majority (> 95%) of the adherent lymphocytes were CD4+ and CD8+ T cells. The percentages of CD4+ and CD8+ T lymphocytes on the PAANa surface were essentially the same as those from the original lymphocyte isolation. On the hydrophilic/neutral

PAAm surface, the percentage of CD8+ cells was less than half that of the lymphocyte isolation that was initially plated. PET showed less than half the lymphocyte isolation percentage of CD8+ T cells in co-cultures with low levels of adherent macrophages.

However, as the adherent macrophage density increased, the level of CD8+ T lymphocytes increased 2 fold over the percentage in the lymphocyte isolation while

CD4+ T lymphocytes decreased 2 fold from the lymphocyte isolation.

After identifying the lymphocyte subtypes on the biomaterial surfaces, we examined to what they were adhering (i.e. whether they were adhering to adherent macrophages and foreign body giant cells or interacting solely with the biomaterial surface). Utilizing the nucleic acid stain YO-YO-1 to visualize nuclei and overlaid phase contrast images of the same field to visualize the cells and their morphology, identified lymphocytes could be localized. Representative results are shown in Figure 4.4 in co- cultures after a period of monocyte to macrophage development and potential FBGC

- 126 - Chapter IV development (3 days). On all surfaces where the monocyte isolation was plated, there were initial 2 hour incubations for monocyte adhesion. Visualization of PET surfaces after the initial two hour incubation for the monocyte isolation showed that although lymphocytes were present as contaminants in the monocyte isolation and during the initial exposure to the PET surfaces, these lymphocytes were not adherent to monocytes or the surface (data not shown).

After 3 days of culture, lymphocytes exposed to the biomaterial surfaces with or without the presence of monocytes, macrophages, or foreign body giant cells were found adhering directly and solely to the biomaterial surface (i.e. biomaterial-adherent).

However, as Figure 4.5A and Table 4.5 shows, direct biomaterial-adherent lymphocytes were low (< 50 lymphs/mm2) regardless of the monocyte concentration plated.

Lymphocytes at the biomaterial surface predominantly adhered to macrophage or FBGCs

(> 90%) when there were significant numbers of adherent macrophages and FBGCs in co-culture (Figure 4.5B). When levels of adherent macrophages and FBGCs were low, the percentage of biomaterial-adherent lymphocytes was higher. This was further supported by the PAAm surface at higher monocyte concentration co-cultures. Despite abundant monocytes exposed to the PAAm surface (e.g. 1/0.5 and 1/1.0 co-cultures), macrophage adhesion was inhibited compared to the other surfaces (Figure 4.1B). The percentage of biomaterial-adherent lymphocytes was highest on PAAm compared to the other two monocyte, macrophage, and FBGC adhesion promoting surfaces. For instance, there were approximately 40% biomaterial-adherent lymphocytes in the 1/0.5 co-cultures.

- 127 - Chapter IV

A B

FBGC FBGC

FBGC

C D

FBGC

Figure 4.4: Immunofluorescent identification of lymphocyte subtypes and their interactions on the biomaterial surfaces for A) CD14+ monocytes/macrophages, B) CD8+ T suppressor cells, C) CD4+ T helper cells, and D) CD56+ natural killer cells after 3 days of co-culture. Cells were identified by staining with anti-CD14 antibodies for monocytes/macrophages, anti-CD8 antibodies for T suppressor cells, anti- CD4 antibodies for T helper cells, and anti-CD56 antibodies for natural killer cells. Utilizing YO-YO-1 (green, nuclei) and an overlaid phase contrast image, identified cells (red) were analyzed for their interactions with macrophages or foreign body giant cells. Representative images are from the hydrophilic/anionic surface (PAANa) where both macrophages and FBGCs are present. Solid arrows indicate macrophage or foreign body giant cell-adherent lymphocytes while dashed arrows indicate biomaterial-adherent lymphocytes. Scale bars, 50 μm.

Lymphocytes were capable of adhering to both adherent macrophages as well as adherent FBGCs as shown in Figure 4.5. Cell-associated lymphocytes on hydrophobic

(PET) and hydrophilic/neutral (PAAm) surfaces were virtually all adhering to macrophages since these surfaces do not readily facilitate foreign body giant cell formation. Although FBGCs do form on PAAm, the numbers are few. On the other

- 128 - Chapter IV

hand, the hydrophilic/anionic surface (PAANa) evoked abundant FBGC formation. On

PAANa, not only did the lymphocytes interact with macrophages, they also were

adhering to FBGCs. The number and percentage of FBGC-adherent lymphocytes

increased with the number of monocytes initially plated in co-culture. Up to 40% of the

adherent lymphocytes on PAANa were observed to be interacting with FBGCs despite

FBGCs accounting for < 25% of the adherent macrophages and FBGC. On PAAm, < 5% of adherent lymphocytes were observed to be interacting with FBGCs although 15-25% of adherent macrophages and FBGCs were FBGCs.

Table 4.5: Interactions of Adherent Lymphocyte (cells/mm2) at Biomaterial Surfaces as a Function of Lymphocyte to Monocyte Co-culture Ratios

Type of Material 1/0.1 1/0.5 1/1.0 Adhesion FBGC 0 ± 0 0 ± 0 0 ± 0 PET Macrophage 16.3 ± 10.1 29.2 ± 8.6 55.0 ± 25.3 Biomaterial 20.6 ± 19.3 0 ± 0 2.6 ± 2.6 FBGC 1.7 ± 1.7 81.6 ± 26.6 96.2 ± 31.8 PAANa Macrophage 15.5 ± 11.8 113.3 ± 36.8 158.9 ± 63.6 Biomaterial 6.0 ± 4.8 37.8 ± 22.5 4.3 ± 3.1 FBGC 0 ± 0 0 ± 0 1.7 ± 1.7 PAAm Macrophage 2.6 ± 2.6 10.3 ± 9.1 21.5 ± 9.1 Biomaterial 0 ± 0 6.0 ± 4.8 6.0 ± 4.8 Mean ± SEM, n=3

Table 4.6: FBGC- or Macrophage-Adherent lymphocytes Normalized to the Respective Adherent FBGC or Macrophage Population

1/0.5 1/1.0 PET PAANa PAAm PET PAANa PAAm FBGC-adherent lymphocytes:FBGCs 0.0 2.5 0.0 0.0 1.4 0.1

Macrophage-adherent lymphocytes:Macrophages 0.1 0.9 0.4 0.1 0.3 0.3

- 129 - Chapter IV ) 2 300 A 250

200 Biomaterial-adherent 150 Macrophage-adherent FBGC-adherent 100

0 Adherent Density Lymphocyte (cells/mm PET PET PET PET PAAm PAAm PAAm PAANa PAANa PAANa 1/0 1/0.1 1/0.5 1/1.0

100% 90% B 80% 70% * 60% Macrophage-adherent 50% FBGC-adherent 40% 30% 20% 10% 0% Percent of Adherent Lymphocytes PET PET PET PET PAAm PAAm PAAm PAANa PAANa PAANa 1/0 1/0.1 1/0.5 1/1.0 Ratio of Lymphocyte/Monocyte Co-culture Concentrations

Table 4.6 shows the adherent lymphocyte population normalized to the number of respective adherent FBGC or macrophage populations. On the PAANa surface, the normalized FBGC-adherent lymphocyte population was more than 2.5x greater than the normalized macrophage-adherent lymphocyte population in the 1/0.5 co-culture ratio and

- 130 - Chapter IV

more than 5x greater in the high initial monocyte co-culture ratio. On the PET and

PAAm surfaces, the normalized macrophage-adherent lymphocytes were greater than

normalized FBGC-adherent lymphocytes.

Discussion

This study demonstrated that lymphocytes are capable of adhering directly to biomaterial surfaces, but to a limited extent. Increased lymphocyte adhesion on biomaterial surfaces is mediated by both the presence of adherent macrophages and foreign body giant cells as well as surface chemistry. In the co-cultures, lymphocytes were capable of interacting with adherent macrophages as well as foreign body giant cells. Additionally, biomaterial surfaces chemistries influenced the number as well as type of lymphocyte interactions at the tissue/material interface.

When we investigated lymphocytes at the surface of our biomaterials, we found that in situations where there are minimal monocytes, the numbers of adherent lymphocytes were very low. This was the same on all surfaces. As the numbers of monocytes increased, so too did the number of adherent lymphocytes. What we discovered was that the number of lymphocytes directly adhering to the biomaterial surfaces did not increase significantly. The fact that lymphocytes adhered directly to our biomaterial surfaces was in agreement to previous studies on lymphocytes interactions with biomaterial surfaces. Groth et al, however, found that surface wettability could influence direct lymphocyte adhesion on biomaterials.8 In our study, our hydrophilic and

hydrophobic surfaces showed no significant difference in direct lymphocyte adhesion.

The disparate results could possibly be due to differences in the adsorbed protein layer

- 131 - Chapter IV

that the lymphocytes were exposed to on the surface of the biomaterial. Groth et al pre-

adsorbed specific proteins such as fibrinogen, fibronectin, and vitronectin while we utilized autologous serum containing biological proteins that the biomaterial would encounter upon introduction into the body.

The increase in adherent lymphocyte density as we increased initial monocyte concentrations was primarily due to interactions between lymphocytes and adherent macrophage and foreign body giant cells. This suggests that these cell-cell interactions dominate and are potentially much stronger than the lymphocyte interactions with the material. Lymphocytes do not have direct degradative or phagocytic capabilities like macrophages so they would not be expected to adhere strongly to material substrates like macrophages. Instead they depend on their ability to be mobile and migrate to perform their duties in the immune response.15,16 Lymphocytes can be in either adherent or non-

adherent states.17 The non-adherent state allows lymphocytes to circulate through the

vasculature. The majority of the lymphocytes isolated from the peripheral blood would tend to not directly adhere to biomaterials unless the biomaterial surfaces actually facilitated lymphocyte adhesion. The adsorbed protein layer on the biomaterials in this study may be inadequate for stable or strong lymphocyte binding. When it is necessary

for lymphocytes to migrate out into the tissue, home to peripheral sites, engage antigen

presenting cells (e.g. macrophages), or perform effector cell functions, lymphocytes must

then exist in an adherent state. Lymphocytes possess receptors (e.g. integrins) that allow

binding to vasculature and extracellular matrix (ECM) proteins for migration purposes as

well as to other cells which is required for cellular activation as well as for performing effector functions in elimination of foreign antigen.15-18 Integrins have low affinity for

- 132 - Chapter IV

ligands unless they are activated. Binding of cell surface receptors such as the T cell

receptor (TCR) or chemokine receptors, generates a signal to the lymphocyte causing an

increase in integrin affinity through conformational changes as well as avidity through an

increase in integrin clustering.16,17 The presence of biomaterial-adherent macrophage and production of cytokines and chemokines provides potential TCR interactions as well as chemokine receptor ligation.19 The lymphocyte adhesion to adherent macrophages and

foreign body giant cells we observed in this study do have functional significance as we

have previously shown that direct lymphocyte interactions can lead to enhanced adherent

macrophage and foreign body giant cell activation.3 These specific molecular

interactions are currently unknown.

The observation that lymphocytes adhere to macrophages is not surprising as

lymphocyte interactions with macrophages are well described in the immune response.

Although little is known about the specific molecular mechanisms of interaction in

response to biomaterial surfaces, lymphocyte and macrophage interactions in the immune

response and in chronic inflammatory states provide some insight.20,21 Potential

interactions include the pairing of integrin receptors and ligands as well as the binding of

TCR and major histocompatibility complex (MHC). Other molecular interactions may

exist as well and are described in Table 4.7.

- 133 -

Table 4.7: Molecular Mediators of Potential Lymphocyte Interactions with Adherent Monocytes, Macrophages, and FBGCs

Molecule Cell Type Expression Molecular Cell Type Function/ Cell Type Function/Effect on modulators Interactions Effect on MO, MΦ, FBGC Lymphocyte CD40L Lymphocyte T cell CD40/CD40L - T cell - Monocyte - Induces monocyte - T cell Activation activation - NK cell (Cytokine and TCR) - Th1 cell - Monocyte - IL-1β production24

CD40 - Monocyte GM-CSF, IL- - Anti-CD3- - Monocyte - IL-1 production25 3, or IFN-γ22,23 activated CD4+ T cell

- 134 - Activated - Monocyte - IL-12 production T cells

- PHA- - Monocyte - Induced interstitial activated - Macrophage collagenase, CD4+ T cell stromelysin, and TF protein and activity27

- Soluble - Monocyte - Induced CD54, CD40L MHC class II, CD86, and CD40 expression and production of TNF-α, IL-1β, IL-6, and IL-823 - NK cell - Activation as - Macrophage - Activated as measured by measured by CD6928 phagocytosis28

- CD40L - Monocyte - Co-stimulated TNF- transfected α and IL-6, enhanced cells IL-822

Molecule Cell Type Expression Molecular Cell Type Function/ Cell Type Function/Effect on modulators Interactions Effect on MO, MΦ, FBGC Lymphocyte 2B4 (CD244) - NK cell 2B4/CD4829 - NK cell - Proliferation - Monocyte - LPS stimulated 2B4/CD5830 - Macrophage macrophages directly CD48 (murine) - Macrophage - IFN-γ29 lysed29 CD58 (human) - Enhanced - Transcription of cytotoxicity29 NKG2D ligands: ULBP1,2,3 induced - Increased upon activation29 expression of activating receptors NKp30, - 135 NKp44, NKG2D, NKp46, 2B429

- Decreased perforin29

- NK 2B4 and NKG2D recruited to immunological synapse29 NKp30 - NK cells NKp30/NKp30L - Mediating NKp44 NKp46/NKp46L APC lysis31 NKp46 (Natural cytotoxicity receptors)

NKp30L - APC NKp44L NKp46L

Molecule Cell Type Expression Molecular Cell Type Function/ Cell Type Function/Effect on modulators Interactions Effect on MO, MΦ, FBGC Lymphocyte NKG2D - NK, CD8+ T NKG2D/MICA - Activating cells, NKT cells, NKG2D/MICB signal for target activated killing macrophages30

Membrane TNF-α - T cell TCR mTNF-α/TNFR - Enhanced LPS Activation stimulated monocyte- TNFR - Monocytes derived TNF-α32 - Macrophages - Multinucleated Giant cells CD69 - T cell T cell - Indicative of - Stimulates IL-1β - 136 activation lymphocyte production33,34 activation CD2 (LFA2) - T cell CD2/LFA3 - Intercellular - Intercellular adhesion adhesion LFA-3 - Lymphocyte - APCs - Involved in T - Partially induces IL- cell activation 1β34 and influences Th1/Th2 differentiation as part of immunological synapse20 Mac-1, CR3, - Monocyte Chemokine Mac-1/ICAM-1 - Intercellular - Intercellular CD11b/CD18 - Macrophage and TCR adhesion adhesion ligation

Molecule Cell Type Expression Molecular Cell Type Function/ Cell Type Function/Effect on modulators Interactions Effect on MO, MΦ, FBGC Lymphocyte LFA-1 - Monocytes Chemokine LFA-1/ICAM-1 - Intercellular - Intercellular (CD11a/CD18) - T cells and TCR adhesion adhesion - Macrophages ligation - Directs T cell - Induced IL-1β34 ICAM-1 - Lymphocytes activation and (soluble CD3 not - Multinucleated differentiation mobilized CD3 giant cells of Th1 vs Th2 stimulated T cell)33 - Macrophage as part of immunological synapse20

- Activating - 137 self APC lysis31

ICAM-3 - Naïve T cells ICAM-3/DC- - Intercellular - Intercellular SIGN adhesion adhesion DC-SIGN - Macrophage Fas (APO- - Macrophages Activation35 - Induce activated 1/CD95) macrophages35

Fas ligand - T cells Activation35 TCR - T cell TCR/MHC - Recognize - Antigen Antigen Presentation MHC - APCs Maturation - MHC I - Multinucleated and activation - Activation to - MHC II giant cells36 produce cytokines (e.g. Th1, Th2)

- Proliferation

Molecule Cell Type Expression Molecular Cell Type Function/ Cell Type Function/Effect on modulators Interactions Effect on MO, MΦ, FBGC Lymphocyte B7 - APCs Activation20 B7/CD28 - Costimulation - B7-1 (CD80) - Multinucleated for activating T - B7-2 (CD86) giant cells37 cells38

CD28 - T cells - Development - ICOS of regulatory T - CTLA-4 cells38

- Direct T cell activation and differentiation as part of - 138 immunological synapse20

Chapter IV

In co-cultures on PET surfaces after 2 hours for initial monocyte adherence in the presence of lymphocytes, minimal lymphocytes were observed to be interacting to monocytes. This suggests that biomaterial-adherent monocytes may not have the adequate phenotype for engaging in strong interactions with lymphocytes. This could be due to lack of maturation and activation after 2 hours. Maturation of monocytes leads to upregulation of the originally low levels of T cell interacting surface molecules such as

MHC molecules.39 An interesting finding was that lymphocytes were capable of adhering to foreign body giant cells in our in vitro cultures. The greater normalized

FBGC-adherent lymphocytes relative to normalized macrophage-adherent lymphocytes on the hydrophilic/anionic surface indicated that not only were they capable of interacting, lymphocytes on the hydrophilic/anionic surface either prefer interactions with

FBGCs compared to macrophages or engage in stronger or more stable connections. We can speculate that the interactions may be similar to those between lymphocytes and macrophages since FBGCs are formed from the fusion of macrophages and have been shown to possess antigenic similarities.36 The specific mechanisms of lymphocyte interaction with FBGCs as well as whether there is any functional significance to the interactions are currently unknown. The molecular interactions described in Table 4.7 may also be applicable to lymphocyte interactions with FBGCs as well.

Biomaterial surface properties made a significant impact on the interactions occurring on the surface. The hydrophilic/anionic surface (PAANa) evoked the highest level of adherent lymphocytes, the majority of which were adherent to macrophages and

FBGCs. If we compare the hydrophilic/anionic PAANa surface with the hydrophobic

PET surfaces, we see that lymphocyte adhesion is not simply a function of the number of

- 139 - Chapter IV

adherent macrophages or FBGCs as PET shows at least as much if not more adherent

macrophages. The ratio of adherent lymphocyte density to adherent macrophage and

FBGC density is significantly lower for PET. If we also compare these surfaces to the

hydrophilic/neutral PAAm surface, we see that despite much lower adherent macrophage

and FBGC densities, the PAAm surface elicited, like PAANa, similarly higher ratios of

adherent lymphocyte density to adherent macrophage and FBGC density relative to the

PAAm surface. These results indicate that adherent macrophage and FBGC phenotype

and secreted soluble products play a role. We demonstrated previously that the PAAm

and PAANa surfaces are highly activating surfaces in terms of macrophage cytokine and

chemokine production.3 These inflammatory mediators, as we discussed above, can lead

to increased avidity and affinity of integrins for stronger or more stable lymphocyte

interactions with adherent macrophages. In addition to increased production of

inflammatory cytokines and chemokines, macrophage activation also leads to

upregulation of cell surface molecules (e.g. chemokine receptors and MHC) with enhanced capability for antigen presentation and interactions with lymphocytes.40

CD4+ and CD8+ T lymphocytes were the predominant lymphocyte subtypes adhering directly to biomaterial surfaces and to macrophages and foreign body giant cells. The lack of CD56+ NK cells on the biomaterial surfaces does not rule out the fact that NK cells may be interacting at the surface. The few NK cells found on the surfaces were all interacting with adherent macrophages. These interactions could be merely transient or non-specific. Other studies have shown a predilection for CD3+ T lymphocyte to adhere onto surfaces pre-adsorbed with particular extracellular matrix proteins.9 The biomaterial surfaces utilized in this study demonstrated differences in

- 140 - Chapter IV

terms of the percentage of CD4+ and CD8+ T lymphocytes. The PAANa surface was

non selective for CD4+ and CD8+ T lymphocytes as the percentages of each were similar

to those from the lymphocyte isolation. PAAm, on the other hand, was less selective for

CD8+ T cells as the percentage of these CD8+ T cells found on the surface were less than half that of the lymphocyte isolation we plated initially. PET showed lower selectivity for adherent CD8+ cells with minimal adherent macrophage and FBGC density but became more selective for CD8+ T lymphocytes interactions as the adherent macrophage

and FBGC density increased. In previous studies involving column cell separation and

dialysis, various material surfaces were capable of selectively retaining lymphocyte

subpopulations.5,7 We have shown here that biomaterial surfaces can also promote

certain types of interactions: PET and PAAm for CD8+ T lymphocyte and CD4+ T

lymphocyte interactions with macrophages, respectively. Again, this could be the result

of differences in adherent macrophage responses on the different biomaterial surfaces.

The different activation levels and cytokine profiles evoked from adherent macrophages

and FBGCs on each of the surfaces can result in different cellular targets, behavior, and

interactions.41 These different lymphocytes can perform vastly different functions. In

the immune response, CD8+ T cells possess cytotoxic functions through perforin and Fas

pathways.42 CD4+ T cells, on the other hand, can mediate cellular as well as humoral

immune responses depending on their cytokine production profile.20 However, their role

in response to biomaterial surfaces is less clear. It is also possible that these different

cellular interactions contribute to the differences in the cytokine/chemokine response

elicited from the different surfaces.

- 141 - Chapter IV

Although synthetic biomaterials have already been utilized as biomedical

implants, the full capability of biomaterials as components in therapeutic applications

such as tissue engineering, regenerative medicine, or biomedical devices has yet to be

determined. We are still in the process of gaining a complete mechanistic understanding

the complex signaling, cellular behavior and interactions that occur when these materials

enter the body. During the foreign body reaction, lymphocytes have the opportunity to

engage in direct interactions with both biomaterial and the adherent macrophages and

foreign body giant cells. What we found was that all surfaces showed limited direct

biomaterial-adherent lymphocytes regardless of the presence of macrophages or FBGCs.

T lymphocytes and NK cells adhered to macrophage and foreign body giant cells to a

much greater extent than the surface itself. The number and type of lymphocyte

interactions with macrophage and/or FBGCs were dependent on biomaterial surfaces.

The results from this investigation help to broaden our understanding of the interactions

that occur at surfaces of biomaterials as well as how biomaterials influence cell-cell and

cell-material interactions. The goal is that this information will provide us with tools for

modulating biomaterial-mediated responses as well as potentially directing biological responses.

References

1. Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81-110.

- 142 - Chapter IV

3. Chang DT, Colton E, Anderson JM. Paracrine and juxtacrine lymphocyte enhancement of adherent macrophage and foreign body giant cell activation. J Biomed Mater Res A 2008.

4. Castiglione A, Pagliaro P, Romagnoni M, Beccari M, Cavaliere G, Veneroni G, Sorgato G. Flow cytometric analysis of leukocytes eluted from haemodialysers. Nephrol Dial Transplant 1991;6 Suppl 2:31-5.

5. Grooteman MP, Nube MJ, van Limbeek J, Schoorl M, van Houte AJ. Lymphocyte subsets in dialyser eluates: a new parameter of bioincompatibility? Nephrol Dial Transplant 1996;11(6):1073-8.

6. Yokoyama M, Nakahashi T, Nishimura T, Maeda M, Inoue S, Kataoka K, Sakurai Y. Adhesion behavior of rat lymphocytes to poly(ether)-poly(amino acid) block and graft copolymers. J Biomed Mater Res 1986;20(7):867-78.

7. Maeda M, Kimura M, Inoue S, Kataoka K, Okano T, Sakurai Y. Adhesion behavior of rat lymphocyte subpopulations (B cell and T cell) on the surface of polystyrene/polypeptide graft copolymer. J Biomed Mater Res 1986;20(1):25-35.

8. Groth T, Altankov G, Klosz K. Adhesion of human peripheral blood lymphocytes is dependent on surface wettability and protein preadsorption. Biomaterials 1994;15(6):423-8.

9. Groth T, Zlatanov I, Altankov G. Adhesion of human peripheral lymphocytes on biomaterials preadsorbed with fibronectin and vitronectin. J Biomater Sci Polym Ed 1994;6(8):729-39.

10. Bergman AJ, Zygourakis K. Migration of lymphocytes on fibronectin-coated surfaces: temporal evolution of migratory parameters. Biomaterials 1999;20(23- 24):2235-44.

11. Ito T, Iwasaki Y, Narita T, Akiyoshi K, Ishihara K. Controlled adhesion of human lymphocytes on electrically charged polymer surface having phosphorylcholine moiety. Science and Technology of Advanced Materials 2003;4(2):99-104.

12. Nakayama Y, Anderson JM, Matsuda T. Laboratory-scale mass production of a multi-micropatterned grafted surface with different polymer regions. J Biomed Mater Res 2000;53(5):584-91.

13. McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 1994;91(21):10119-23.

14. Chang DT, Jones JA, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Lymphocyte/macrophage interactions: Biomaterial surface-dependent cytokine, chemokine, and matrix protein production. J Biomed Mater Res A 2008.

- 143 - Chapter IV

15. von Andrian UH, Mackay CR. T-cell function and migration. Two sides of the same coin. N Engl J Med 2000;343(14):1020-34.

16. Hogg N, Laschinger M, Giles K, McDowall A. T-cell integrins: more than just sticking points. J Cell Sci 2003;116(Pt 23):4695-705.

17. Miyamoto YJ, Andruss BF, Mitchell JS, Billard MJ, McIntyre BW. Diverse roles of integrins in human T lymphocyte biology. Immunol Res 2003;27(1):71-84.

18. Ley K, Laudanna C, Cybulsky MI, Nourshargh S. Getting to the site of inflammation: the leukocyte adhesion cascade updated. Nat Rev Immunol 2007;7(9):678-89.

19. Jones JA, Chang DT, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Proteomic analysis and quantification of cytokines and chemokines from biomaterial surface-adherent macrophages and foreign body giant cells. J Biomed Mater Res A 2007;83(3):585-96.

20. Santana MA, Rosenstein Y. What it takes to become an effector T cell: the process, the cells involved, and the mechanisms. J Cell Physiol 2003;195(3):392- 401.

21. Monaco C, Andreakos E, Kiriakidis S, Feldmann M, Paleolog E. T-cell-mediated signalling in immune, inflammatory and angiogenic processes: the cascade of events leading to inflammatory diseases. Curr Drug Targets Inflamm Allergy 2004;3(1):35-42.

22. Alderson MR, Armitage RJ, Tough TW, Strockbine L, Fanslow WC, Spriggs MK. CD40 expression by human monocytes: regulation by cytokines and activation of monocytes by the ligand for CD40. J Exp Med 1993;178(2):669-74.

23. Kiener PA, Moran-Davis P, Rankin BM, Wahl AF, Aruffo A, Hollenbaugh D. Stimulation of CD40 with purified soluble gp39 induces proinflammatory responses in human monocytes. J Immunol 1995;155(10):4917-25.

24. Burger D. Cell contact-mediated signaling of monocytes by stimulated T cells: a major pathway for cytokine induction. Eur Cytokine Netw 2000;11(3):346-53.

25. Wagner DH, Jr., Stout RD, Suttles J. Role of the CD40-CD40 ligand interaction in CD4+ T cell contact-dependent activation of monocyte interleukin-1 synthesis. Eur J Immunol 1994;24(12):3148-54.

26. Shu U, Kiniwa M, Wu CY, Maliszewski C, Vezzio N, Hakimi J, Gately M, Delespesse G. Activated T cells induce interleukin-12 production by monocytes via CD40-CD40 ligand interaction. Eur J Immunol 1995;25(4):1125-8.

27. Mach F, Schonbeck U, Bonnefoy JY, Pober JS, Libby P. Activation of monocyte/macrophage functions related to acute atheroma complication by

- 144 - Chapter IV

ligation of CD40: induction of collagenase, stromelysin, and tissue factor. Circulation 1997;96(2):396-9.

28. Scott MJ, Hoth JJ, Stagner MK, Gardner SA, Peyton JC, Cheadle WG. CD40- CD154 interactions between macrophages and natural killer cells during sepsis are critical for macrophage activation and are not interferon gamma dependent. Clin Exp Immunol 2004;137(3):469-77.

29. Nedvetzki S, Sowinski S, Eagle RA, Harris J, Vely F, Pende D, Trowsdale J, Vivier E, Gordon S, Davis DM. Reciprocal regulation of human natural killer cells and macrophages associated with distinct immune synapses. Blood 2007;109(9):3776-85.

30. Hallett WH, Murphy WJ. Positive and negative regulation of Natural Killer cells: therapeutic implications. Semin Cancer Biol 2006;16(5):367-82.

31. Poggi A, Prevosto C, Zancolli M, Canevali P, Musso A, Zocchi MR. NKG2D and natural cytotoxicity receptors are involved in natural killer cell interaction with self-antigen presenting cells and stromal cells. Ann N Y Acad Sci 2007;1109:47- 57.

32. Parry SL, Sebbag M, Feldmann M, Brennan FM. Contact with T cells modulates monocyte IL-10 production: role of T cell membrane TNF-alpha. J Immunol 1997;158(8):3673-81.

33. Manie S, Kubar J, Limouse M, Ferrua B, Ticchioni M, Breittmayer JP, Peyron JF, Schaffar L, Rossi B. CD3-stimulated Jurkat T cells mediate IL-1 beta production in monocytic THP-1 cells. Role of LFA-1 molecule and participation of CD69 T cell antigen. Eur Cytokine Netw 1993;4(1):7-13.

34. Isler P, Vey E, Zhang JH, Dayer JM. Cell surface glycoproteins expressed on activated human T cells induce production of interleukin-1 beta by monocytic cells: a possible role of CD69. Eur Cytokine Netw 1993;4(1):15-23.

35. Ashany D, Song X, Lacy E, Nikolic-Zugic J, Friedman SM, Elkon KB. Th1 CD4+ lymphocytes delete activated macrophages through the Fas/APO-1 antigen pathway. Proc Natl Acad Sci U S A 1995;92(24):11225-9.

36. Athanasou NA, Quinn J. Immunophenotypic differences between osteoclasts and macrophage polykaryons: immunohistological distinction and implications for osteoclast ontogeny and function. J Clin Pathol 1990;43(12):997-1003.

37. Bainbridge JA, Revell PA, Al-Saffar N. Costimulatory molecule expression following exposure to orthopaedic implants wear debris. J Biomed Mater Res 2001;54(3):328-34.

38. Abbas AK. The control of T cell activation vs. tolerance. Autoimmun Rev 2003;2(3):115-8.

- 145 - Chapter IV

39. Friedl P, den Boer AT, Gunzer M. Tuning immune responses: diversity and adaptation of the immunological synapse. Nat Rev Immunol 2005;5(7):532-45.

40. Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 2004;25(12):677-86.

41. Luttikhuizen DT, Harmsen MC, Van Luyn MJ. Cellular and molecular dynamics in the foreign body reaction. Tissue Eng 2006;12(7):1955-70.

42. Mosmann TR, Li L, Sad S. Functions of CD8 T-cell subsets secreting different cytokine patterns. Semin Immunol 1997;9(2):87-92.

- 146 - Chapter V

IFN-γ Production from Lymphocyte Interactions with Biomaterial-Adherent Macrophages and Foreign Body Giant Cells

Abstract

The lymphocyte response to biomaterials is still unclear although there are some

clinical examples of lymphocyte mediated consequences to implanted synthetic polymers. In vitro studies have shown lymphocyte enhancement of monocyte adhesion, macrophage fusion, and macrophage/FBGC activation by lymphocyte-derived soluble factors. In order to investigate biomaterial-induced lymphokine production, lymphocytes and monocytes were cultured alone or together, directly or separated by a porous membrane transwell, on polyethylene terephthalate (PET)-based photograft copolymerized material surfaces displaying distinct hydrophobic, hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic chemistries. After periods of 3, 7, and 10 days, secreted IFN-γ, IL-4, and IL-13 were quantified by ELISA. IFN-γ was produced in direct and indirect co-cultures while essentially no IFN-γ was produced in lymphocyte and monocyte-only cultures indicating that lymphocytes are activated by interactions with adherent macrophages and FBGCs, not direct biomaterial contact. Moreover, macrophage-derived cytokines are required for induction of IFN-γ production. Direct lymphocyte interactions with adherent macrophages/FBGCs enhanced IFN-γ production relative to indirect co-cultures and elicited material surface chemistry-dependent IFN-γ production. Neither IL-4 nor IL-13, which mediate down-regulated responses, was produced in any of the cultures during the 10 day culture period. IFN-γ provides one

- 147 - Chapter V

potential lymphokine mediator of previously observed enhancement of adherent

monocyte/macrophage/FBGC behavior.

Introduction

The use of synthetic biomaterials as components in implanted prostheses, medical devices, and tissue-engineered constructs requires that we gain a full understanding of the complex cellular responses and interactions that occur at implant sites. Monocytes and

macrophages have received much of the attention due to their direct interactions with biomaterials in terms of their degradative capabilities and their involvement in the foreign body reaction. The foreign body reaction describes the process of monocyte adhesion, differentiation into macrophages, macrophage activation, and macrophage fusion to form foreign body giant cells (FBGC). Lymphocytes appear at the implant site for a limited period of time during the chronic inflammatory phase,1,2 but our understanding of the

lymphocyte response to biomaterials is limited. There is evidence of lymphocyte

activation from exposure to implanted prostheses such as silicone gel breast implants and

left ventricular assist devices which can have implications on immune function.1,3

Patients with left ventricular assist devices have shown an increased risk of infection as well as the presence of auto-reactive antibodies.4 We previously demonstrated that

lymphocytes were capable of enhancing monocyte adhesion, macrophage activation, and

macrophage fusion in vitro through indirect mechanisms of interaction.5,6 However, the specific lymphocyte-derived soluble factors involved have not been determined. The elucidation of lymphocyte behavior and interactions are vital as we continue to develop

- 148 - Chapter V

novel biomaterials, devices, and other therapeutic technologies utilizing synthetic

materials.

The lymphocyte population performs a variety of functions and plays an integral

role in the immune response. The major lymphocyte subtypes include CD4+ and CD8+

T lymphocytes (T cells), B lymphocytes (B cells), and natural killer (NK) cells.

Activated CD4+ T cells when secreting IFN-γ and TNF-β are type 1 T helper (Th1) cells

while CD4+ type 2 T helper (Th2) cells secrete cytokines such as IL-4, IL-5, and IL-13.7

CD8+ T cells are referred to as T suppressor or T cytotoxic cells. CD8+ T cells and NK cells are both capable of killing cells through apoptosis as well as secreting cytokines such as those from CD4+ T cells.7,8 Additionally, T cells also produce IL-2 upon

activation to induce proliferation. B cells produce specific antibodies targeting foreign

antigens.

IFN-γ, IL-4, and IL-13 are lymphokines with the ability to modulate macrophage

responses. IFN-γ is known to activate macrophages and polarize them towards a pro-

inflammatory state which is reflected by their upregulation in capability to present

antigen, phagocytose, and produce pro-inflammatory cytokines and effector molecules.9

IL-4 and IL-13, on the other hand, induces macrophage activation to promote a downregulatory response which is shown by enhanced production of anti-inflammatory

IL-10 and expression of non-opsonic receptors such as the mannose receptor.9

Using cytokine protein arrays, we previously found that lymphokines such as IL-2 and IL-5 were not produced in direct lymphocyte co-cultures with macrophages and

FBGCs but we did detect interferon-inducible protein (IP)-10.10 To further investigate

biomaterial-mediated lymphocyte behavior in vitro, we investigated the quantitative

- 149 - Chapter V

production of additional lymphokines IFN-γ, IL-4, and IL-13 from lymphocytes cultured alone and in co-culture with biomaterial-adherent monocytes, macrophages, and FBGCs.

Our hypothesis is that lymphocytes are activated by interactions with adherent macrophages and FBGCs to produce lymphokines that in turn, guide adherent macrophage and FBGC behavior. The aim of this study was to evaluate biomaterial- induced lymphocyte activation and identify potential lymphokines involved in influencing adherent macrophage and FBGC behavior.

Materials and Methods

Preparation of Biomaterial Surfaces

Model biomaterial surfaces were PET-based surfaces identical to those utilized in

Chapter II – IV and described in Chapter I.11,12 Briefly, Mylar® PET, hydrophobic surfaces, were initially coated with poly(benzyl N,N-diethyldithiocarbamate-co-styrene)

(BDEDTC) to provide an additional hydrophobic surface. Subsequent UV polymerization to form polyacrylamide (PAAm), sodium salt of polyacrylic acid

(PAANa), and methiodide of poly(dimethylaminopropylacrylamide) (DMAPAAmMeI) provided material substrates with distinct hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic surface chemistries, respectively. These surfaces were cut into 1.5 cm diameter disks and dipped into 100% ethanol for sterilization prior to placement into

24-well TCPS plates. Silicone rings were cut from silicone tubing of O.D. 5/8 and I.D.

3/8 (Cole Parmer, Vernon Hills, IL), sonicated for 5 minutes in 100% ethanol, autoclave sterilized, and inserted into each well to secure the material surfaces at the bottom of the

- 150 - Chapter V

well. Phosphate-buffered saline containing magnesium chloride and calcium chloride

(Invitrogen) (PBS++) was used to remove residual ethanol from the material surfaces.

Monocyte and Lymphocyte Cell Culture

Monocyte and lymphocyte cell populations were isolated as previously

described.13 Briefly, monocytes and lymphocytes were isolated from blood donated by

healthy non-medicated human donors via a non-adherent centrifugation procedure. Each

population was suspended separately in serum free medium (SFM) supplemented with L-

glutamine, antibiotics, antimycotics, and 20% autologous serum (AS). The isolated

monocyte populations contained, on average, 60.5% ± 9.1% monocytes and 39.5 ± 9.1%

contaminating lymphocytes while the isolated lymphocyte population consisted of, on

average, 51.3% ± 4.1% CD4+ T lymphocytes, 30.8% ± 5.2% CD8+ T lymphocytes,

10.7% ± 1.1% NK cells, 5.6% ± 0.5% B lymphocytes, and 1.6% ± 0.1% contaminating

monocytes as previously determined by flow cytometry (See Chapter II).10 5 x 105

monocytes in 0.25 mL SFM with 20% AS were plated onto PET, BDEDTC, PAAm,

PAANa, and DMAPAAmMeI surfaces and allowed to incubate for 2 hours at 37˚C and

++ 5% CO2. After the 2 hour incubation, non-adherent cells were removed by a PBS

wash leaving adherent monocytes which were cultured with and without 1.5 x 106 lymphocytes. The 1.5 x 106 lymphocytes in 0.5 mL SFM with 20% AS were plated alone

and then co-cultured directly with adherent monocytes on the 5 biomaterial surfaces, and

co-cultured indirectly with adherent monocytes via 0.02 μm membrane pore transwell

inserts (Nunc, Naperville, IL). Supernatants were collected after 3, 7, and 10 days of

culture, centrifuged at 3000 rpm to remove cells/debris, and cell-free supernatant stored

at -20˚C for cytokine analysis.

- 151 - Chapter V

Quantification of Lymphokine Production

Production of IFN-γ, IL-4, and IL-13 were quantified by ELISA (R&D systems,

Minneapolis, MN), performed according to manufacturer’s instructions, and measured by

an EL808 ultra microplate reader with KC Junior software (Bio-Tek Instruments, Inc.,

Winooski, Vermont).

Statistical Analysis

All results were presented as an average ± the standard error of the mean (SEM)

with multiple human donors (n = 3). Statistical analysis was performed utilizing Minitab

statistical software (Minitab Inc., State College, PA) with statistically significant

differences determined by ANOVA and the Tukey post hoc test.

Results

Lymphokine Production

Lymphokines IFN-γ, IL-4, and IL-13 were assayed in order to evaluate

lymphocyte activation in the response to biomaterials. IFN-γ production over 3, 7, and

10 days from lymphocytes and monocytes in individual as well as co-cultures on PET- based photograft copolymerized surfaces was quantified and shown in Figure 5.1 (Table

5.1). IL-4 and IL-13 were not detected in any cultures by ELISA with detection sensitivity limits of less than 10 and 32 pg/mL, respectively.

Adherent monocytes after 3 days of cultures, develop and differentiate into adherent macrophages, and have the potential to fuse into foreign body giant cells.

Lymphocytes in co-culture have the opportunity to interact with adherent monocytes, macrophages, and FBGCs but after 3 days, the population is predominantly macrophages

- 152 - Chapter V

and FBGCs. Over 10 days of culture, virtually no IFN-γ was produced in lymphocyte

and monocyte-only cultures while the co-cultures elicited IFN-γ. Indirect lymphocyte

and adherent monocyte co-cultures evoked a relatively steady production of IFN-γ near or below 100 pg/mL over time and showed no material dependence or trends.

Lymphocytes over 3, 7, and 10 days in direct co-culture with adherent macrophages and FBGCs produced approximately 2-2.5 fold (PET), 1.5-2.5 fold

(BDEDTC), 0.8-2.5 fold (PAAm), 3.5-5 fold (PAANa), and 1.3-1.5 fold

(DMAPAAmMeI) greater IFN-γ than lymphocytes in indirect co-cultures. Thus, direct co-cultures elicited IFN-γ production which varied with material surface chemistry.

Differences in hydrophilicity and hydrophobicity as well as charge affected the production of IFN-γ. The hydrophilic/anionic surface (PAANa) demonstrated the highest level of production showing a greater than 5 fold increase compared to the hydrophilic/neutral surface (PAAm) which evoked the least IFN-γ and greater than 2.8 fold increase relative to the level on the hydrophilic/cationic surface (DMAPAAmMeI) over all time points. PET and BDEDTC were both hydrophobic surfaces and showed similar levels of IFN-γ production (<1.4 fold differences). The levels on these hydrophobic surfaces were lower (1.4 – 2.3 fold) than on the hydrophilic/anionic surface but greater than the amount of IFN-γ elicited (1.4 – 2 fold) on the two other hydrophilic surfaces, PAAm (neutral) and DMAPAAmMeI (cationic). Except for the hydrophilic/anionic surface (PAANa), the hydrophobic surfaces showed slightly higher

IFN-γ levels than the hydrophilic surfaces. The levels of IFN-γ increased slightly after

3 days.

- 153 - Chapter V

PET BDEDTC 600 A Day 3 PAAm 500 PAANa DMAPAAmMeI 400

300

200 Concentration (pg/mL) Concentration γ 100 IFN- 0 Lymphocyte Monocyte Direct Indirect

600 B Day 7

500

400

300

200 Concentration (pg/mL) Concentration γ 100 IFN- 0 Lymphocyte Monocyte Direct Indirect

600 C Day 10 * 500

400

300

200 Concentration (pg/mL) Concentration

γ 100

IFN- 0 Lymphocyte Monocyte Direct Indirect Culture

Figure 5.1: Quantification of IFN-γ production from lymphocyte and monocyte single and co-cultures over A) 3, B) 7, and C) 10 days. Data represents average ± the standard error of the mean (n=3). *p < 0.01

- 154 - Chapter V

Table 5.1: IFN-γ production (pg/mL) from Lymphocyte and Monocyte Cultures Exposed to Biomaterial Surfaces over 3, 7, 10 days

Culture Biomaterial Day 3 Day 7 Day 10 Type Lymphocyte 0 ± 0 0 ± 0 0 ± 0 Monocyte 0 ± 0 0 ± 0 0 ± 0 PET Direct 180 ± 124 196 ± 150 187 ± 75 Indirect 50 ± 37 95 ± 71 79 ± 77 Lymphocyte 0 ± 0 0 ± 0 0 ± 0 Monocyte 0 ± 0 0 ± 0 0 ± 0 BDEDTC Direct 185 ± 142 238 ± 152 258 ± 115 Indirect 76 ± 53 103 ± 84 172 ± 157 Lymphocyte 0 ± 0 4 ± 4 3 ± 3 Monocyte 0 ± 0 0 ± 0 0 ± 0 PAAm Direct 46 ± 36 60 ± 44 78 ± 69 Indirect 58 ± 48 49 ± 36 32 ± 20 Lymphocyte 0 ± 0 8 ± 8 2 ± 1 Monocyte 0 ± 0 2 ± 2 2 ± 2 PAANa Direct 315 ± 125 339 ± 231 423 ± 88 Indirect 61 ± 43 93 ± 48 105 ± 51 Lymphocyte 0 ± 0 0 ± 0 0 ± 0 Monocyte 0 ± 0 2 ± 2 3 ± 1 DMAPAAmMeI Direct 105 ± 54 122 ± 30 134 ± 36 Indirect 73 ± 43 95 ± 52 90 ± 50 Mean ± SEM, n=3

Discussion

We previously investigated production of cytokine/chemokines, including some

lymphokines, utilizing proteins arrays and ELISAs (Chapter II).10 Lymphokines IL-4 and

IL-5 were undetected (< 1 pg/mL) in direct lymphocyte co-culture with monocytes,

macrophages, and FBGCs using the protein arrays. Other lymphokines were below the

minimum detection limits and undetected by the arrays: IL-2 (25 pg/mL), IL-3 (100 pg/mL), IL-7 (100 pg/mL), IL-13 (100 pg/mL), and IFN-γ (100 pg/mL). IL-2 was further evaluated by ELISA with a minimum detection limit of less than 7 pg/mL and was again

- 155 - Chapter V

undetected. The cytokine array screening showed production of interferon inducible

protein (IP)-10, which as its name implies can be induced by IFN-γ, in lymphocyte and

monocyte co-cultures. The finding prompted further investigation of IFN-γ production despite the lack of detection by the cytokine arrays. Therefore, in this study IFN-γ, IL-4, and IL-13 production were quantified by ELISA in order to further evaluate lymphocyte behavior and interactions at material surfaces. Lymphocytes and monocytes were cultured alone and in co-cultures on different biomaterial surfaces in order to investigate lymphokine production as a function of material surface chemistries.

The importance of IFN-γ, IL-4, and IL-13 is that first, they are secreted primarily by lymphocytes (i.e. lymphokines) and are indicative of lymphocyte activation. These cytokines are secreted by subpopulations of CD4+ T cells, CD8+ T cells, and NK cells.

IFN-γ is secreted by CD4+ Th1 cells, CD8+ Tc1 cells, and NK1 cells while IL-4 and IL-

13 are generally secreted by CD4+ Th2 cells, CD8+ Tc2 cells and NK2 cells.7 Secondly,

these lymphokines play roles in activating and polarizing macrophages. IFN-γ is pro- inflammatory and polarizes macrophages towards a classical (type I) phenotype, while

IL-4 and IL-13 are anti-inflammatory cytokines and induce an alternative (type II) macrophage activation.9

We showed previously that IL-2, IL-4, and IL-5 were not produced at a level

above 1-10 pg/mL nor IFN-γ at a minimum 100 pg/mL level (Chapter II) and suggested

that this was indicative of a lack of lymphocyte activation.10 In this study, we confirmed

the lack of IL-4 detection by protein array with ELISA and found that neither IL-4 nor

IL-13 was produced in any lymphocyte and monocyte cultures whether alone or in co-

- 156 - Chapter V

culture. The lack of IL-2, IL-4, IL-5, and IL-13 production points towards a lack of

classic T lymphocyte activation in these cultures.

In this study, we did, however, demonstrate the production of IFN-γ over 3, 7, and

10 days in lymphocyte and monocyte co-cultures suggesting that there is indeed induction of activation in at least a subset of the lymphocyte population. The lack of production in lymphocyte and monocyte-only cultures indicated lymphocyte interactions with biomaterial-adherent macrophage and FBGCs were required. However, because

IFN-γ was induced in indirect co-cultures, direct lymphocyte and macrophage cell-cell interactions were not necessary. The results suggest that the lymphocyte population is activated via non-contact mechanisms. This is consistent with previous findings where induction of lymphocyte proliferation occurred through macrophage-derived soluble factors (i.e. indirect paracrine interactions).5,14 The specific proliferating lymphocyte

population in these in vitro cultures, however, has not been determined. Rodriguez et al.

showed that T lymphocytes in a population of mononuclear cells cultured on biomaterial

surfaces, did not proliferate nor express cell surface markers (CD69 and CD25)15 The authors found donor variability, but minimal production of IL-2 and IFN-γ in the cultures of mononuclear cells; only 1 donor showed production of IFN-γ. These findings are somewhat inconsistent with the finding in this study. The isolation procedure for mononuclear cell populations is different and thus results in cell ratios and co-culture conditions that are different than the ones in this study. The finding by Brodbeck et al. that there was optimum co-culture ratios for examining lymphocyte/macrophage interactions perhaps explains this inconsistency.5

- 157 - Chapter V

Although not found in these co-cultures, IL-4 and IL-13 have been shown to

induce foreign body giant cell formation.16-18 Kao et al. showed that the blocking of IL-4 led to a reduction in FBGC formation while the addition of recombinant IL-4 increased

FBGC formation in vivo.17 Obviously, the in vitro cultures do not fully represent the in

vivo conditions that showed the presence of IL-4 and its role in inducing macrophage fusion to form FBGCs. It is likely that either the cellular constituents in these co-culture

populations are missing key players or the interactions occurring in these co-cultures are insufficient to simulate the in vivo situation.

As mentioned previously, IFN-γ is strongly produced only by certain subpopulations of activated T cells and NK cells and plays an integral role in modulating lymphocyte differentiation into Th1 cells, effector T cell responses, and upregulating antigen presentation through MHC molecules for enhanced T cell activity.19 In the

immune response, T cells and NK cells are generally activated through contact with

antigen; however, lymphocytes can also be activated via cytokine stimulation (i.e.

antigen-independent mechanisms). A summary and description of different lymphocyte

activators is provided in Table 5.2. Cytokines capable of activating T and NK cells to

secrete IFN-γ independent of T cell receptor (TCR) ligation include a combination of IL-

2, TNF-α, and IL-6 along with IL-12, IL-15, IL-18, and IL-21.20,21 IL-2 and IL-21 are

products of activated lymphocytes while IL-12, IL-15, IL-18 can be derived from

activated macrophages. The lymphocyte population in this study consisted of primarily T

lymphocytes in addition to NK cells. Since they were both capable of interacting with

adherent macrophages and FBGCs, these cells could all contribute to IFN-γ production.

- 158 -

Table 5.2: Modulators of Lymphocyte Activation and Proliferation

Inductors Functions Mechanisms of Upregulated Downregulated Unchanged action Molecules Molecules Molecules Combination of IL-7, - Proliferation (selective T p38 – proliferation - CD40L – (fraction - CCR7 IL-15, TNF-α, IL-6, lymphocyte subsets) of cells) - CD25 IL-10 cytokines22 - Differentiation (Activation?) - IL-2/IL-15Rβ - IL-2/IL-15Rγc - CD45RA - IFN-γ and IL-4 (fraction of cells) - CCR5 IL-15, IL-6, TNF-α23 - Activation - CD45RO - CD45RA - CD40L - CD25 - HLA-DR

- 159 - CD69

Adhesion molecules - CD11a (LFA-1) - CD11b (MAC-1) - ICAM-1 - CD44

TCR stimulation22 - Proliferation - Calcineurin - CD40L - CD45RA (naïve T - differentiation - PKC - CD25 cells) - ERK MAP kinase - CCR7 - Proliferation - DR, CD4, CD3 - CD45RO - CD45RA - CD45RA+ T TCR stimulation24 - differentiation cells do not produce IFN-γ or help B cells TNF-α25 - Induction of hyporesponsive - Impaired TCR - CD69 - TCR/CD3 cell (chronic T cells signal transduction surface inflammation) - TCR transduction molecules - Calcium flux - IL-2 production

Inductors Functions Mechanisms of Upregulated Downregulated Unchanged action Molecules Molecules Molecules IL-15 - Activation (T cells except - IL-2Rβ chain - CD69 - CD69 on CD4+ (similar to IL-2) – naive CD4+ T cells) CD45RO- (naive share IL-2Rβ,γ - Proliferation (T cells except CD4+ cells) chains26 CD4+ naive T cells)

Cocktail of IL-2, TNF- - Activated purified naïve - Lymphokine - CD45RA+ T α, IL-624 (CD45RA+), memory synthesis (mRNA cells do not (CD45RO+), and resting IFN-γ and IL-4) produce IFN-γ or CD4+ T cells to express - CD69 help B cells activation markers and - Incorporate proliferate, display effector thymidine - 160 function (lymphokine (proliferation) synthesis and help for Ig - CD25 (IL-2R p55 production by B cells) chain) on fraction of T cells) IL-21 + IL-1527 - Enhances IFN-γ from human - Stat3 (IL-21) - NF-kB binding Or NK and T cells - Stat5 (IL-15) - IFN-γ production IL-21 + IL1827 - IL-18 and IL15 (not IL-21) induces NF-kB binding CD16 (FcγRIII) - Fc receptor activating - Signaling through - Antibody dependent receptor21 FcεRIγ cellular cytotoxicity - IFN-γ and GM-CSF production - NK cell degranulation

Chapter V

In general macrophage-derived inducers of IFN-γ production induce both T and

NK cell activation. If indeed T lymphocytes are not activated in response to biomaterial surfaces, either the signals are specific for NK cells or there may also be other unknown complicating factors involved that may be specifically inhibitory to T lymphocytes.

Because of evidence suggesting that T lymphocytes may not be activated on biomaterial surfaces, we hypothesize that NK cells may be the primary activated lymphocyte in the population. Therefore, future studies will need to focus on NK cell behavior in response to biomaterial surfaces as well as the specific cytokine inductors of IFN-γ production.

These in vitro studies are biologically relevant since both T and NK cells have been shown to be present at biomaterial implant sites in vivo.1 The possibility that NK cells

are playing a role in responding to biomaterial surfaces is interesting in that in addition to

cytokine production, they have the capability to perform cytotoxic functions at

biomaterial surfaces. NK cells are involved in controlling infections and tumors as well

as regulating the magnitude of the immune or inflammatory responses by inducing

apoptosis of target cells.21 Activated NK cells are capable of killing antigen presenting

cells (e.g. macrophages and dendritic cells), fibroblasts and bone marrow stromal

cells.28,29 The killing mechanism is induced by target cell ligation of NK natural

cytotoxicity receptors or other activating receptors which results in lytic granule

polarization and degranulation of effector molecules such as perforins and granzymes.30

The results also showed that lymphocyte cell contact with biomaterial-adherent

macrophages and FBGCs can enhance IFN-γ production. We previously showed that T cells as well as NK cells physically interact with adherent macrophages and FBGCs on biomaterial surfaces (See Chapter IV). Although the mechanisms of interaction are

- 161 - Chapter V

currently unknown in the context of biomaterial surfaces, inflammatory states and

immune responses provide some insight into potential interactions. These potential

interactions between lymphocytes and monocytes, macrophage, and FBGC are described

in Table 4.7 of Chapter IV. The next step will be to determine the pertinent T cell and

NK cell interactions with adherent macrophages and FBGCs that elicit or enhance IFN-γ

production in response to biomaterial surfaces.

Synthetic materials utilized in clinical applications have shown evidence of

lymphocyte activation. These clinical examples along with in vitro studies on

lymphocyte responses to biomaterials are summarized in Table 5.3. Katzin et al. found

that lymphocytes in the fluid and tissue surrounding silicone gel breast implants were

predominantly T cells, and relative to peripheral blood, a greater percentage of the T cells

were HLA-DR+ and CD29+ indicating a state of immune activation.1 Additionally, left ventricular assist device (LVAD) patients have shown the development of B cell hyperreactivity and immune dysfunction leading to heightened risk of infection and potential autoimmune disorders due to biomaterial-activated T cells.1,3,4 Schuster et al.

demonstrated the elevated presence of anti-HLA antibodies and soluble CD40L in LVAD patients indicating B cell activation.3 The results here are consistent in showing that

lymphocytes can be activated by exposure to biomaterials through interactions with

macrophages. However, the scenario involving the LVAD and silicone implants is

different in that this current study examines the response to an intact biomaterial surface

that is not phagocytosable by macrophages. In the LVAD and silicone implant scenario,

the cellular infiltrates such as lymphocytes and macrophages have the potential to also

respond to small phagocytosable biomaterial particles. In vitro studies concerning the

- 162 - Chapter V

LVAD response were performed with cells exposed to biomaterial particles that

permitted macrophage phagocytosis.3 In fact, studies have shown that smaller

phagocytosable particles elicit stronger stimulated responses than particles of larger

sizes.31 These differences may contribute to some of the inconsistencies regarding

lymphocyte activation illustrated in Table 5.3.

Table 5.3 also contains information on lymphocyte responses to metal prosthetics

for reference. Lymphocytes are shown to be activated in recipients of joint replacements.

Metallic debris/ions are capable of complexing with serum proteins creating haptens

recognizable by lymphocytes resulting in a cell mediated type IV immune reaction.32

The specific mechanisms for lymphocyte activation in regard to implants composed of synthetic polymer are as yet unclear, but direct lymphocyte activation by biomaterials does not appear likely. We hypothesize that lymphocytes may be activated by soluble factors released as a result of biomaterial-induced macrophage activation.

- 163 -

Table 5.3: Findings for Biomaterial Lymphocyte Activation Studies

Study Biomaterial Lymphocyte Activation? Mechanism Activation References Studied Markers Evaluated

LVAD Polyurethane T cells Yes - Calcium influx 3,4 - CD95 (Fas) Upregulation - CD40L upregulation (CD25-dependent)

B cells Yes CD40/CD40L - CD86 Upregulation 3,4 (T cell - Anti-phospholipid interactions) anti-HLA antibodies

- 164 Hemodialysis Cellulose triacetate No T cell No - CD25 (IL-2R) 33 Cuprammonium activation - HLA-DR Polysulphone

Silicone Breast Implants Silicone T cells Yes - HLA-DR expression 1 - CD29 expression Lymphocyte/Macrophage PET-based surfaces T lymphocytes Yes Macrophage- - Proliferation 5 Co-cultures (in vitro) B lymphocytes derived Cytokines NK cells Yes Macrophage- - IFN-γ production Chapter V derived Cytokines No - IL-2, IL-4, IL-5, IL-13 10

Production Chapter II, V

Silicone Rubber (SR) T cell No - Proliferation by CSFE - PBMC culture 15 Polyetherurethane - Activation markers (PEU) (CD69 and CD25 PET expression) TCPS - IL-2 and IFN-γ production

Joint Replacement Prosthetics

- Total Joint Replacement Chromium and Cobalt T cells Yes - CD69 expression 34 extract

Chromium, cobalt, T cells - Metal-specific - Proliferation 35 nickel, titanium adaptive immune - IL-2 and IFN-γ response production

Cobalt-chrome alloy T cells Yes - Potential cell - IL-2 receptor - Revision Hip Replacement 32 Titanium alloy mediated Type IV expression immune reaction - HLA-DR expression (metal-protein hapten formation) - 165

Cement, polyethylene T cells Yes - CD28 (T cells) - Peri-implant and Revision 36 Macrophage - B7-1 and B7-2 (APC) Surgery (APC)

FeCrNi and TiAlV T cells Yes - CD28 (T cells) 37 Macrophage - CD86 (B7-2) on and macrophages and Multinucleated multinucleated giant giant cells cells

Table 5.4: Modulators of Monocyte and Macrophage Activation

Inductor Upregulated Downregulated Mechanism of Additional Effects Molecules Molecules Action Interactions

1. IL-2, IL-6, TNF- α - TNF-α23,38 - Cytokine activated - IFN-γ or GM- - Enhanced TNF-α23 stimulated T - not IL-1023 Lymphocyte CSF23 lymphocytes23,38 - IL-1β39 - PI3 kinase and NFkB 2. IL-15 alone23 - IL-1α39 kinase α–dependent - Endogenous IL- - Inhibit TNF-α23 3. IL-15, IL-6 and TNF-α induction of TNF- α 38 1023 stimulated T lymphocytes23

TCR-activated - TNF-α20,40 - T cell contact40 20,40

- 166 lymphocytes - IL-10 - IL-10 production-partly 1. Anti-CD3 - IL-1β39 dependent on 2. Anti-CD3 and anti- - MMP-139 endogenous TNF-α and CD28 or specific antigen - Low IL-1Ra39 IL-140 - Lymphocyte membrane TNF-α enhance monocyte-derived TNF- α40

Cytokine production Soluble CD2339 - IL-1β - TNF-α GM-CSF, IL-3, or IFN-γ41 - CD40

LPS - Stress-inducible - NK cell co-culture - High LPS resulted in ligands for NKG2D: direct NK-mediated cell ULBP1,2,328 lysis28

- Same as INF-γ - Same as INF-γ - Classical Macrophage activation9 activation9 Activation9

Inductor Upregulated Downregulated Mechanism of Additional Effects Molecules Molecules Action Interactions - Antigen presentation - IL-10 Classical Macrophage INF-γ9 (MHC) - IL-1RII Activation - IL-12, IL-23, TNF-α, - CCR2 IL-1, IL-6, Type I IFN - CCR5 - Toxic intermediates - CCR1 (NO, ROI) - TLR2,4 - IL-1RI - IL-1R accessory protein - CCL2,3,4,5 CXCL8,16 CXCL9,10,11 - 167 - CCR7 - non-opsonic receptors - TLR4 - Alternative activation IL-4 and IL-139 (e.g. mannose receptor) - IL-12 - IL-10 - IL-1 - IL-1RII - TNF - IL-1Ra - IL-6 - CCL16, 17, 18, 22, 24 - NF-kB/STAT1 - CXCR1, 2 - IL-10 - Deactivated activation IL-109 - Non-opsonic receptors (e.g. mannose receptor) - CCL16,18 - CXCL13 - CCR2, 5

Activin A from Anti-CD3 - Arginase-1 - Inducible NO cMaf - Alternative activation (M2 stimulated Th2 cells synthase NF-AT phenotype)

Chapter V

In addition to providing an indication of lymphocyte activation, IFN-γ is capable

of modulating cellular inflammatory responses. We have previously shown that

lymphocyte interactions, through soluble factor production, could enhance monocyte

adhesion, macrophage fusion, and macrophage/FBGC activation.5,6 This lymphokine may play a role in these observations as IFN-γ has been shown to activate macrophages to produce higher levels of inflammatory cytokines such as IL-1 and TNF-α, enhance antigen presentation and secrete reactive molecules9 as well as promote and enhance

fusion to form multinucleated giant cells.16,42 A description of macrophage activation by

both lymphocyte contact and noncontact mediated mechanisms, including IFN-γ are

described in Table 5.4.

The level of IFN-γ production was found to vary on the different biomaterial

surface chemistries suggesting that lymphocytes were differentially activated. This

dependency on biomaterial surfaces is in agreement with many studies evaluating cellular

responses to biomaterials. The hydrophilic/anionic (PAANa) surfaces have been shown to promote macrophage activation (Chapter III),6 fusion to form giant cells (Chapter

II),10,12 and lymphocyte interactions with macrophages and FBGCs (Chapter IV). In this study, we found that this surface also evoked a higher level of IFN-γ relative to the other surfaces. This result makes sense when we consider that the production of IFN-γ is induced by mediators produced by activated cells such as lymphocytes and macrophages, and PAANa was found to be an activating surface for macrophages. Furthermore, the functional capabilities of this lymphokine as a fusion inductor or promoter and cellular activator provide a potential soluble factor mediating the observed enhancement of macrophage behavior on this particular surface. Additionally, IFN-γ was found to be

- 168 - Chapter V

significantly enhanced when lymphocytes were permitted to engage in direct cell-cell

contact with adherent macrophages and giant cells on the PAANa surface more so than

on all other surfaces investigated. This implies that the phenotypes of the lymphocytes,

macrophages and FBGCs on PAANa in regards to surface molecules may be distinct

relative to the phenotypes on the other surfaces. For instance, lymphocyte surface

molecules such as integrins can change in conformation as well as localization based on

direct cell-cell interaction with macrophages and indirect interactions via cytokines and chemokines.43,44 Additional studies are required to investigate lymphocyte cell surface

changes that occur upon interactions at biomaterial surfaces.

One of the primary differences in cellular behavior observed on PAANa

compared to the other surfaces is its induction of macrophage fusion. As previously

demonstrated in Chapter IV, lymphocytes were observed to adhere to FBGCs in addition

to macrophages primarily on this fusion promoting hydrophilic/anionic surface. We currently do not have a full understanding of the functional capabilities of foreign body

giant cells. In addition to promoting degradation of biomaterial surfaces,45 FBGCs have

been shown to produce cytokines,46 and display a wide range of cell surface antigens

similar to macrophages.47 FBGCs display many cell surface molecules such as major

histocompatibility complex molecules (e.g. HLA-DR) capable of providing potential

molecular interactions with lymphocytes.47 It is possible that the FBGCs could possess

enhanced lymphocyte activating capabilities. Therefore, future studies must continue to

investigate FBGC functions, particularly in relation to lymphocyte interactions at implant

sites.

- 169 - Chapter V

The biological response to biomaterials is a complex process involving

interactions between multiple cell types. We have found that the production of IFN-γ which suggest the activation of lymphocytes upon indirect paracrine interactions with

adherent macrophages and foreign body giant cells. The direct engagement of lymphocytes with biomaterial-adherent macrophages and FBGCs results in enhancement

of IFN-γ production. These findings prompt future investigation into the specific molecular mechanisms involved in these interactions. Biomaterial-induced and dependent processes must be fully understood to ensure optimal functionality in the design of novel biomaterials or implantable devices in addition to preventing the adverse consequences associated with many of the current clinical implants.

References

1. Katzin WE, Feng LJ, Abbuhl M, Klein MA. Phenotype of lymphocytes associated with the inflammatory reaction to silicone gel breast implants. Clin Diagn Lab Immunol 1996;3(2):156-61.

2. Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81-110.

3. Schuster M, Kocher A, John R, Hoffman M, Ankersmit J, Lietz K, Edwards N, Oz M, Itescu S. B-cell activation and allosensitization after left ventricular assist device implantation is due to T-cell activation and CD40 ligand expression. Hum Immunol 2002;63(3):211-20.

4. Itescu S, John R. Interactions between the recipient immune system and the left ventricular assist device surface: immunological and clinical implications. Ann Thorac Surg 2003;75(6 Suppl):S58-65.

5. Brodbeck WG, Macewan M, Colton E, Meyerson H, Anderson JM. Lymphocytes and the foreign body response: lymphocyte enhancement of macrophage adhesion and fusion. J Biomed Mater Res A 2005;74(2):222-9.

6. Chang DT, Colton E, Anderson JM. Paracrine and juxtacrine lymphocyte enhancement of adherent macrophage and foreign body giant cell activation. J Biomed Mater Res A 2008.

- 170 - Chapter V

7. Santana MA, Rosenstein Y. What it takes to become an effector T cell: the process, the cells involved, and the mechanisms. J Cell Physiol 2003;195(3):392- 401.

8. Mosmann TR, Li L, Sad S. Functions of CD8 T-cell subsets secreting different cytokine patterns. Semin Immunol 1997;9(2):87-92.

9. Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 2004;25(12):677-86.

10. Chang DT, Jones JA, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Lymphocyte/macrophage interactions: Biomaterial surface-dependent cytokine, chemokine, and matrix protein production. J Biomed Mater Res A 2008.

11. Nakayama Y, Anderson JM, Matsuda T. Laboratory-scale mass production of a multi-micropatterned grafted surface with different polymer regions. J Biomed Mater Res 2000;53(5):584-91.

12. Jones JA, Chang DT, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Proteomic analysis and quantification of cytokines and chemokines from biomaterial surface-adherent macrophages and foreign body giant cells. J Biomed Mater Res A 2007;83(3):585-96.

13. McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 1994;91(21):10119-23.

14. MacEwan MR, Brodbeck WG, Matsuda T, Anderson JM. Student Research Award in the Undergraduate Degree Candidate category, 30th Annual Meeting of the Society for Biomaterials, Memphis, Tennessee, April 27-30, 2005. Monocyte/lymphocyte interactions and the foreign body response: in vitro effects of biomaterial surface chemistry. J Biomed Mater Res A 2005;74(3):285-93.

15. Rodriguez A. T Cell Interactions in the Foreign Body Response to Biomaterials. Dissertation. Case Western Reserve University, 2007.

16. McNally AK. Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 1995;147(5):1487-99.

17. Kao WJ, McNally AK, Hiltner A, Anderson JM. Role for interleukin-4 in foreign- body giant cell formation on a poly(etherurethane urea) in vivo. J Biomed Mater Res 1995;29(10):1267-75.

18. DeFife KM, Jenney CR, McNally AK, Colton E, Anderson JM. Interleukin-13 induces human monocyte/macrophage fusion and macrophage mannose receptor expression. J Immunol 1997;158(7):3385-90.

- 171 - Chapter V

19. Maher SG, Romero-Weaver AL, Scarzello AJ, Gamero AM. Interferon: cellular executioner or white knight? Curr Med Chem 2007;14(12):1279-89.

20. Monaco C, Andreakos E, Kiriakidis S, Feldmann M, Paleolog E. T-cell-mediated signalling in immune, inflammatory and angiogenic processes: the cascade of events leading to inflammatory diseases. Curr Drug Targets Inflamm Allergy 2004;3(1):35-42.

21. Hallett WH, Murphy WJ. Positive and negative regulation of Natural Killer cells: therapeutic implications. Semin Cancer Biol 2006;16(5):367-82.

22. Geginat J, Sallusto F, Lanzavecchia A. Cytokine-driven proliferation and differentiation of human naive, central memory, and effector memory CD4(+) T cells. J Exp Med 2001;194(12):1711-9.

23. Sebbag M, Parry SL, Brennan FM, Feldmann M. Cytokine stimulation of T lymphocytes regulates their capacity to induce monocyte production of tumor necrosis factor-alpha, but not interleukin-10: possible relevance to pathophysiology of rheumatoid arthritis. Eur J Immunol 1997;27(3):624-32.

24. Unutmaz D, Pileri P, Abrignani S. Antigen-independent activation of naive and memory resting T cells by a cytokine combination. J Exp Med 1994;180(3):1159- 64.

25. Isomaki P, Clark JM, Panesar M, Cope AP. Pathways of T cell activation and terminal differentiation in chronic inflammation. Curr Drug Targets Inflamm Allergy 2005;4(3):287-93.

26. Kanegane H, Tosato G. Activation of naive and memory T cells by interleukin- 15. Blood 1996;88(1):230-5.

27. Strengell M, Matikainen S, Siren J, Lehtonen A, Foster D, Julkunen I, Sareneva T. IL-21 in synergy with IL-15 or IL-18 enhances IFN-gamma production in human NK and T cells. J Immunol 2003;170(11):5464-9.

28. Nedvetzki S, Sowinski S, Eagle RA, Harris J, Vely F, Pende D, Trowsdale J, Vivier E, Gordon S, Davis DM. Reciprocal regulation of human natural killer cells and macrophages associated with distinct immune synapses. Blood 2007;109(9):3776-85.

29. Poggi A, Prevosto C, Zancolli M, Canevali P, Musso A, Zocchi MR. NKG2D and natural cytotoxicity receptors are involved in natural killer cell interaction with self-antigen presenting cells and stromal cells. Ann N Y Acad Sci 2007;1109:47- 57.

30. Moretta A, Marcenaro E, Parolini S, Ferlazzo G, Moretta L. NK cells at the interface between innate and adaptive immunity. Cell Death Differ 2008;15(2):226-33.

- 172 - Chapter V

31. Jacobs JJ, Hallab NJ, Urban RM, Wimmer MA. Wear particles. J Bone Joint Surg Am 2006;88 Suppl 2:99-102.

32. Goodman SB. Wear particles, periprosthetic osteolysis and the immune system. Biomaterials 2007;28(34):5044-8.

33. Grooteman MP, Nube MJ, van Limbeek J, Schoorl M, van Houte AJ. Lymphocyte subsets in dialyser eluates: a new parameter of bioincompatibility? Nephrol Dial Transplant 1996;11(6):1073-8.

34. Granchi D, Ciapetti G, Savarino L, Stea S, Filippini F, Sudanese A, Rotini R, Giunti A. Expression of the CD69 activation antigen on lymphocytes of patients with hip prosthesis. Biomaterials 2000;21(20):2059-65.

35. Hallab NJ, Anderson S, Stafford T, Glant T, Jacobs JJ. Lymphocyte responses in patients with total hip arthroplasty. J Orthop Res 2005;23(2):384-91.

36. Farber A, Chin R, Song Y, Huie P, Goodman S. Chronic antigen-specific immune-system activation may potentially be involved in the loosening of cemented acetabular components. J Biomed Mater Res 2001;55(3):433-41.

37. Bainbridge JA, Revell PA, Al-Saffar N. Costimulatory molecule expression following exposure to orthopaedic implants wear debris. J Biomed Mater Res 2001;54(3):328-34.

38. Brennan FM, Hayes AL, Ciesielski CJ, Green P, Foxwell BM, Feldmann M. Evidence that rheumatoid arthritis synovial T cells are similar to cytokine- activated T cells: involvement of phosphatidylinositol 3-kinase and nuclear factor kappaB pathways in tumor necrosis factor alpha production in rheumatoid arthritis. Arthritis Rheum 2002;46(1):31-41.

39. Burger D. Cell contact-mediated signaling of monocytes by stimulated T cells: a major pathway for cytokine induction. Eur Cytokine Netw 2000;11(3):346-53.

40. Parry SL, Sebbag M, Feldmann M, Brennan FM. Contact with T cells modulates monocyte IL-10 production: role of T cell membrane TNF-alpha. J Immunol 1997;158(8):3673-81.

41. Alderson MR, Armitage RJ, Tough TW, Strockbine L, Fanslow WC, Spriggs MK. CD40 expression by human monocytes: regulation by cytokines and activation of monocytes by the ligand for CD40. J Exp Med 1993;178(2):669-74.

42. Enelow RI, Sullivan GW, Carper HT, Mandell GL. Induction of multinucleated giant cell formation from in vitro culture of human monocytes with interleukin-3 and interferon-gamma: comparison with other stimulating factors. Am J Respir Cell Mol Biol 1992;6(1):57-62.

- 173 - Chapter V

43. Miyamoto YJ, Andruss BF, Mitchell JS, Billard MJ, McIntyre BW. Diverse roles of integrins in human T lymphocyte biology. Immunol Res 2003;27(1):71-84.

44. Hogg N, Laschinger M, Giles K, McDowall A. T-cell integrins: more than just sticking points. J Cell Sci 2003;116(Pt 23):4695-705.

45. Zhao Q, Topham N, Anderson JM, Hiltner A, Lodoen G, Payet CR. Foreign-body giant cells and polyurethane biostability: in vivo correlation of cell adhesion and surface cracking. J Biomed Mater Res 1991;25(2):177-83.

46. Hernandez-Pando R, Bornstein QL, Aguilar Leon D, Orozco EH, Madrigal VK, Martinez Cordero E. Inflammatory cytokine production by immunological and foreign body multinucleated giant cells. Immunology 2000;100(3):352-8.

47. Athanasou NA, Quinn J. Immunophenotypic differences between osteoclasts and macrophage polykaryons: immunohistological distinction and implications for osteoclast ontogeny and function. J Clin Pathol 1990;43(12):997-1003.

- 174 - Chapter VI

Dynamic Systems Model for Lymphocyte Interactions with Macrophages at Biomaterial Surfaces

Introduction

Biomedical materials are being increasingly called upon to perform many

different functions in many applications such as prostheses, medical devices, tissue

engineering, and regenerative medicine. These applications call for the implantable

materials to provide many functions such as structural strength, biodegradability, selective cellular adhesiveness, encapsulation, enhancement of tissue integration, and/or minimization of the host response. Introduction of any foreign material will result in a biological response involving inflammation and wound healing. If the implanted biomaterial provides a surface with which the body can interact, a host response called the foreign body reaction (FBR) occurs at the biomaterial surface which could end with fibrous encapsulation. The foreign body reaction involves a series of cellular events after biomaterial implantation including monocyte adhesion, monocyte differentiation to macrophages, and ultimately macrophage fusion to form FBGCs. These multinucleated cells have been observed to exist at the tissue/material interface for the duration of the implant and to mediate degradation of the biomaterial.1 Depending on the application,

prolonged degradation of the material may then result in malfunction and ultimate failure

of the implant. From implantation until removal of the foreign substance, the biological

response involves many complex processes and interactions. The mechanisms of which

are not fully understood.

One of the ways to analyze a biological system such as the processes and

interactions occurring at a biomaterial surface and gain a better mechanistic

- 175 - Chapter VI

understanding of the processes that occur is to utilize a mechanistic model. This tool requires an understanding of the system and the essential processes that are involved.

The model utilizes mathematical expressions to describe the system and abides by the

fundamental laws of conservation. Parameters of the model represent essential

mechanisms/processes of the system. The model allows investigation of the relative

sensitivities of the system processes and their importance in determining the outputs.

Additionally, the model can be utilized to test hypotheses as well as generate hypotheses

for experimental study.

The surface chemistry and structure influence the biological response to the

implanted biomaterials, devices, or tissue-engineered constructs. Depending on the

application or purpose, particular biomaterial surface chemistries may be desirable. For instance, a long term drug delivery devices would function optimally with material surface chemistries or modified material surfaces that suppressed the foreign body reaction and fibrous encapsulation. On the other hand, biodegradable systems could

utilize hydrophobic biomaterials that favored the foreign body reaction and the

macrophage- and FBGC-derived molecules (e.g. enzymes) or hydrophilic biomaterials

that favor water absorption for hydrolytic degradative mechanisms. Also, tissue-

engineered constructs containing specific autologous cells would require a specific

material allowing adhesion or retention of those cells. Therefore, the goal is to ultimately

tailor design materials or select appropriate materials for the specific application. A

model can contribute to this aim by providing a tool for predicting how altering certain

parameters by biomaterial properties can direct the biological response.

- 176 - Chapter VI

There are many models that describe cellular interactions as well as cellular

adhesion but fewer that describe the behaviors on or in response to biomaterials. Zhu et al. described ligand-receptor binding of cellular adhesion while N’Dri et al modeled leukocyte rolling and migration in response to stimulus.2,3 Bigerelle et al. developed a

kinetic model that correlated osteoblast adhesion on metallic surfaces to material

composition, surface chemistry, and topography.4 However, the model describes

osteoblast behavior which includes proliferation and thus is not relevant to macrophage

responses to biomaterial which include adhesion and fusion. Other models do describe

lymphocyte behavior and interactions with other cells types in terms of proliferation and

cell turnover of subpopulations, autocrine and paracrine cell-cell interactions, cell-cell

interactions through adhesive molecules, and cytotoxic lymphocyte response through

dendritic cell and cytotoxic T lymphocyte interactions.5-8 These models, although

describing cellular processes of interest, are not relevant to the foreign body reaction or lymphocyte interactions with monocytes and macrophages in response to biomaterial surfaces.

There are only a few models pertaining to the foreign body reaction. Kao et al. and Zhao et al. modeled the fusion process based on probability and fusion kinetics to describe the FBGC density and size distribution on biomaterial surfaces in vivo over time.9-11 These models address adhesion and fusion parameters but do not integrate other

essential cellular mechanisms that occur at the material surface aside from fusion such as

apoptosis and differentiation. Recently, our laboratory has developed a model of our in vitro monocyte culture system describing the process of monocyte adhesion, macrophage differentiation, and macrophage fusion to form foreign body giant cells (i.e. foreign body

- 177 - Chapter VI

reaction) and incorporating additional monocyte/macrophage behavior such as apoptosis

and detachment.12

Lymphocytes are transiently present at the implant site and have been shown in

vitro to interact with monocytes, macrophages, and FBGCs resulting in mutual effects on

each other in terms of activation and enhanced inflammatory responses (See Chapters III

and V).13-15 Therefore, the aim is to develop a model that extends the monocyte in vitro

culture model to include direct lymphocyte interactions and provide quantitative

predictability of the in vitro co-culture system. To accomplish this, differential equations

were developed to describe the time varying non-adherent and adherent population of

cells at the biomaterial surface. This dynamic systems model of multiple cellular

interactions will be applied to experimental data from 5 different biomaterial surfaces

displaying distinct surface chemistries in order to analyze how the system parameters

varied with hydrophobicity/hydrophilicity and charge. The model fit to experimental

data will be performed by a least squares optimization procedure. This dynamic systems

model provides a tool for analyzing the complex events occurring at the biomaterial

surface and provides some predictive capabilities.

Experimental Methods

In vitro Co-cultures

Lymphocyte and monocyte populations were isolated from peripheral blood of healthy human donors by a non-adherent centrifugation method as previously

described.16 As described in Chapter II, 1.0x106 and 1.5x106 cells in 1 mL SFM and 20%

autologous serum of the monocyte and lymphocyte populations, respectively, were

- 178 - Chapter VI

simultaneously plated into each well on material surfaces. PET-based photograft co-

polymerized surfaces were stabilized to the bottom of 24-well TCPS plates by silicone

rings. The material surfaces were described in Chapter I and provided distinct

chemistries featuring hydrophobic (PET and BDEDTC), hydrophilic/neutral (PAAm), hydrophilic/anionic (PAANa), and hydrophilic/cationic (DMAPAAmMeI) properties.

The lymphocytes and monocytes were directly co-cultured for periods of 3, 7, and 10 days. Subsequently, the supernatant and non-adherent cells were removed and adherent cells fixed by methanol for 5 minutes at room temperature.

Visualization and Quantification of Adherent Populations

After culture periods of 3, 7, and 10 days, adherent cells were fixed by methanol and stained May-Grünwald and Giemsa. The adherent density of monocyte/macrophage,

FBGC, and lymphocyte were quantified by counting nuclei in five representative 20X

objective fields. Percent fusion was determined by dividing the number of nuclei in

FBGCs by the total number of adherent nuclei in each field. Acquired experimental data

consisted of total adherent nuclei and cells, total FBGC nuclei and cells, percent fusion,

and total adherent lymphocyte density at 3, 7, and 10 days of culture for each of the

biomaterial surfaces.

Model Development

When exposed to biomaterial surfaces in vitro or in vivo, monocytes adhere onto the surface, differentiate into macrophages, and ultimately fuse into foreign body giant cells (FBGCs). Concurrently, lymphocytes adhere onto the surface as well as interact with the monocytes, macrophages, and FBGCs. The adhered monocytes and

- 179 - Chapter VI

macrophages are capable of undergoing apoptosis or detaching from the surface while

adhered lymphocytes can proliferate, undergo apoptosis, or also detach from the surface.

A simplified schematic of the in vitro co-culture is provided by Figure 6.1.

Blood- Adhesion Adherent Differentiation Adherent Fusion Adherent Foreign Isolated Monocyte Macrophage Body Giant Cell Monocyte (MO) (Mφ) (FBGC) (MO) Lymphocyte Lymphocyte Lymphocyte Lymphocyte (L) (L) (L) (L) Non-adherent Non-adherent Non-adherent Monocytes Monocytes and Monocytes and Macrophages Macrophages

Non-adherent Non-adherent Non-adherent Lymphocytes Lymphocytes Lymphocytes

Figure 6.1: Schematic of the direct lymphocyte/monocyte co-culture system.

The model was developed by determining the system variables and constructing a

system diagram of the variables and their interactions. Figure 6.2 shows the overall

system diagram which includes the progression of the foreign body reaction to form

foreign body giant cells on the biomaterial surface along with adherent and non-adherent

lymphocyte interactions and behavior. System assumptions for the model included cell behavior and interactions which were most likely (hypothesized) in addition to

simplifications/approximations. The assumptions and the system diagrams were used to

develop the parameters, system equations, inputs, and outputs for implementation in

MATLAB 7.2.0.232 (R2006a) (Mathworks, Natick, MA). The program codes are

included in Appendix I – III. The model parameters (i.e. processes or rate constants) that

are involved in the system diagram are listed in Table 6.1.

- 180 - Chapter VI

MO1 (0): Monocyte AaL: Apoptosis non-adherent (1 nucleus)

αL η M1-2 La: π Lna Adherent κ1 MO : Monocyte LMO : Lymphocyte- 1 lymphocyte 1 adherent monocyte complex κ−1 (1 nucleus) Lna α0 δ0

ηL δL γL γ AMO: DMO: Apoptosis Detachment Lna κ Lna: LMφ : Lymphocyte- 1 Μφ : Macrophage π 1 1 N Nonadherent macrophage (1 nuclei) α δ1 lymphocyte complex κ−1 1 Lna β AMφ: DMφ: βL Apoptosis Detachment κ Lna LMφ : Lymphocyte- 1 2 Μφ : Macrophages macrophage 2 (2 nuclei) complex κ−1 Lna

βL β

Lna F L F : κ LF : Lna 3-N 2 3 1 3 κ Lymphocyte (2) Lymphocyte 1 -FBGC -FBGC F3: FBGC κ−1 Legend complex Lna complex κ−1 (3 nuclei) Lna β βL β State Variables L Lna Lna κ L F : κ 1 Observed variables 2 N 1 LF : Lymphocyte (2) N FN: FBGC Lymphocyte (N nuclei) -FBGC κ−1 κ−1 -FBGC complex Lna Lna complex

Figure 6.2: Overall system diagram.

- 181 - Chapter VI

Table 6.1: Dynamic Systems Model Parameters

Symbols Rate Constant

η Monocyte Adhesion

ηL Lymphocyte Adhesion γ Monocyte Differentiation

γL Differentiation due to Lymphocyte Interaction

αo Monocyte Apoptosis

α1 Macrophage Apoptosis

αL Adherent Lymphocyte Apoptosis

δ0 Monocyte Detachment

δ Macrophage Detachment

δL Lymphocyte Detachment β Fusion

βL Fusion due to Lymphocyte Interaction π Adherent Lymphocyte Proliferation

πN Nonadherent Lymphocyte Proliferation

κ1 Lymphocyte dissociation

κ-1 Lymphocyte Binding

Cellular behavior

Monocytes, macrophages, and FBGCs interact with lymphocytes to form 1 lymphocyte complex while FBGCs can also form 2 lymphocyte complexes. For simplification, we assume a maximum lymphocyte interaction of 2 and a single FBGC can have a maximum of 30 nuclei (N). Complexes cannot fuse with each other. The processes that are important here are the same as for noncomplexed cells except apoptosis and detachment are not considered. This is because we assume that only individual adherent monocytes, macrophages, and lymphocytes undergo apoptosis and detachment. Complexes with 2 lymphocyte interactions involve the same processes as

- 182 - Chapter VI

the complexes with 1 lymphocyte interaction. However, we assume that only FBGCs can have more than 1 lymphocyte interaction. Therefore, monocyte and macrophage

complexes with 2 lymphocyte interactions do not exist. The nomenclature of the model

is provided in Table 6.2. The following are representations of the mechanisms that are

occurring in this biological system.

Table 6.2: Model Nomenclature

Adherent Cells

Mo: monocyte M k: macrophage (k=1,2 nuclei) M k: FBGC (k ≥ 3 nuclei) L: lymphocyte L 1Mo: lymphocyte-monocyte complex L 1Mk: lymphocyte-macrophage (k=1,2 nuclei) complex L jMk: lymphocyte (j =1,2)-FBGC (k ≥ 3 nuclei) complex

Non-adherent Cells

DMo: detached monocyte DM1: detached macrophage AMo: apoptotic monocyte AM1: apoptotic macrophage AL: apoptotic lymphocyte NL: non-adherent lymphocyte

Loss of adherent monocyte by apoptosis, detachment, differentiation:

α0 ,δ0 ,γ M o ⎯⎯→⎯ AM o + DM o + M 1

Complexing of lymphocyte and monocyte:

κ1 ,κ −1 M o + NL ←⎯→⎯ L1M o

Loss of adherent macrophage by apoptosis, detachment:

- 183 - Chapter VI

α1 ,δ1 M 1 ⎯⎯→⎯ AM 1 + DM 1

Lymphocyte apoptosis, proliferation, reversible surface adhesion:

⎧⎯⎯→α L AL ⎫ ⎪ π ⎪ L ⎨⎯⎯→ 2L ⎬

⎪ δ L ,η L π N ⎪ ⎩←⎯→⎯ NL ⎯⎯→ 2NL⎭

Fusion of macrophages & FBGC:

β M i + M j ⎯⎯→ M i+ j (i, j ≥ 1)

Reversible fusion of non-adherent lymphocytes with macrophage, FBGC & lymphocyte

complexes:

κ1 ,κ −1 ⎧M j ←⎯→⎯ L1M j ⎫ ⎪ ⎪ NL + ⎨ ⎬ ( j ≥ 1)

⎪ κ1 ,κ −1 ⎪ ⎩L1M j ←⎯→⎯ L2M j ⎭

Irreversible fusion of macrophage/FBGC with lymphocyte complexes:

β L ⎧L1M j ⎯⎯→ L1M i+ j (i, j ≥ 1) ⎪ M i + ⎨

⎪ β L ⎩L2M j ⎯⎯→ L2M i+ j (i ≥ 1; j ≥ 3)

Model Equations

A dynamic systems model is developed to analyze cellular interactions as well as

cell interactions with biomaterial surfaces in an in vitro cell culture. The number density

of adherent monocytes varies with time as:

dM 0 = −[]α + γ + δ M − κ NL * M + κ L M (1) dt 0 0 0 −1 0 1 1 0

- 184 - Chapter VI

where the rate coefficients α0, γ, δ0, κ-1, κ1 relate to cell apoptosis, differentiation, detachment, and reversible complexing, respectively. Initially, plated monocytes adhere onto the surface with an initial number density [Mo(0)]:

M 0 ()0 = η * Mplated where Mplated is the number of monocytes plated. The number density of adherent macrophages with one nucleus varies according to: dM 1 = γM − []α + δ M − κ NL * M + κ L M dt 0 1 1 1 −1 1 1 1 1 (2) N −1 ⎡N −1 N −1 ⎤ − βM1 * ∑ M v − βLM1 ⎢∑ L1M v − ∑ L2M v ⎥ v=1 ⎣ v=1 v=3 ⎦ where the rate coefficients characterize monocyte differentiation (γ), apoptosis (α1), detachment (δ1), binding (κ-1), and unbinding (κ1) to lymphocytes, fusion with each other

(β) , and fusion with lymphocyte complexes. The fusion processes are bounded by

FBGCs with maximum N nuclei. In this case, macrophages with 1 nuclei can fuse with all other cells up to N-1 nuclei. The number density of apoptotic cells on the surface increases as:

dAM j = α M ( j = 0,1) (3) dt j j

The concentration of detached cells in solution increases as:

dDM j = δ M (j = 0,1) δ = δ (4) dt j j j where we assume that the detachment rates for monocytes and macrophages are lumped into one parameter (δ).

The lymphocyte populations consist of those that are adherent and those that are non-adherent. The number density of adherent lymphocytes can be described by:

- 185 - Chapter VI

dL = ()π −α − δ L +η NL (5) dt L L L where apoptosis (αL) and detachment (δL) contribute to losses while proliferation (π) and attachment (ηL) contribute to gains. The number density of the apoptotic lymphocytes increases as:

dAL = α L (6) dt L

The number density of non-adherent lymphocytes changes according to: dNL ⎡ N N ⎤ = ()π N −η L NL + δ L L − κ −1 NL * ⎢∑ M v + ∑ L1M v ⎥ dt ⎣ v=0 v=3 ⎦ (7) ⎡ N N ⎤ + κ1 ⎢∑ L1M v + ∑ L2 M v ⎥ ⎣ v=0 v=3 ⎦

Gains occur from proliferation (π N ) , detachment (δ L ) , unbinding of complexes (κ1) and losses from attachment (ηL ) and complexing (κ-1). We assume only non-adherent lymphocytes interact with the other cell types (e.g. monocytes, macrophages, and

FBGCs) and only adherent lymphocytes undergo apoptosis. Lymphocyte can interact with macrophages and FBGCs with up to N nuclei.

Macrophages with 2 nuclei vary with time as:

N −2 dM 2 ⎡ ⎤ = −κ −1 NL * M 2 + κ1L1M 2 + β ⎢M 1 * M 1 − M 2 * ∑ M v ⎥ dt ⎣ v=1 ⎦ (8) ⎡N −2 N −2 ⎤ − β L M 2 ⎢∑ L1M v + ∑ L2 M v ⎥ ⎣ v=1 v=3 ⎦

The processes include binding (κ-1) and unbinding (κ1) with lymphocytes, fusion with macrophages and FBGC (β) , and fusion with complexes (βL ) . Similarly, the number density of FBGC ( ≥ 3 nuclei) is determined by:

- 186 - Chapter VI

p N − j dM j ⎡ ⎤ = −κ −1 NL * M j + κ1L1M j + β ⎢∑ M j−w * M w − M j * ∑ M v ⎥ dt ⎣ w=1 v=1 ⎦ (9) ⎡N − j N − j ⎤ − β L M j * ⎢∑ L1M v − M j ∑ L2 M v ⎥ ⎣ v=1 v=3 ⎦ where

⎧ j ⎫ for j even ⎪ 2 ⎪ p = ⎨ ⎬ to include all fusion combinations that yield j −1 ⎪ for j odd⎪ ⎩⎪ 2 ⎭⎪ the specific number of nuclei. The fusion processes are bounded by FBGCs with maximum N nuclei. Thus, macrophages with j nuclei can fuse with all other cells up to

N-j nuclei. Only fused cells with greater than 3 nuclei are considered foreign body giant cells because we have previously experimentally determined that we can be more certain of determining a cell with 3 nuclei to be fused rather than unfused. The accuracy of differentiating between a fused cell with 2 nuclei and 2 unfused cells is lower. Cells with

2 nuclei have a greater probability of merely being associated rather than fused.

Therefore, fused cells with 2 nuclei are still considered macrophages (1-2 nuclei).

Complexes of a monocyte with 1 lymphocyte change according to:

dL M 1 0 = −()γ + κ L M + κ NL * M (10) dt L 1 1 0 −1 0 where rate constants γL, κ-1, κ1 relate to differentiation due to lymphocyte interaction, binding, and unbinding, respectively. The number density of macrophages with 1 nuclei and 1 lymphocyte interaction expressed as:

N −1 dL1M 1 = γ L L1M 0 + κ −1 NL * M 1 − κ1L1M 1−β L L1M 1 * ∑ M v (11) dt v=1

- 187 - Chapter VI

increases through monocyte differentiation (γL) and lymphocyte binding (κ-1), and decreases through unbinding (κ1) and fusion with individual monocyte, macrophages, and

FBGCs (βL). For a macrophage with 2 nuclei and 1 lymphocyte interactions, changes occur through fusion:

N −2 dL1M 2 ⎡ ⎤ = κ −1 NL * M 2 − κ1L1M 2 + β L ⎢L1M 1 * M 1 − L1M 2 * ∑ M v ⎥ (12) dt ⎣ v=1 ⎦

Complexes with FBGC that occur with 1 lymphocyte vary according to:

dL M 1 j = κ NL * M − κ NL * L M + κ L M − κ L M dt −1 j −1 1 j 1 2 j 1 1 j (13) ⎡ j−1 N − j ⎤ + β L ⎢∑ L1M j−w * M w − L1M j * ∑ M v ⎥ j ≥ 3 ⎣ w=1 v=1 ⎦

Complexes with 2 lymphocyte interactions and FBGC change according to:

N −3 dL2 M 3 ⎡ ⎤ = κ −1 NL * L1M 3 − κ1L2 M 3 − β L ⎢L2 M 3 * ∑ M v ⎥ (14) dt ⎣ v=1 ⎦ which includes complexing and uncomplexing with lymphocytes and fusion with individual cells of any number of nuclei. The number density of complexes with more nuclei can change through fusion with macrophages and FBGCs and binding or unbinding with lymphocytes:

dL M 2 j = κ NL * L M − κ L M dt −1 1 j 1 2 j (15) ⎡ j −3 N − j ⎤ + βL ⎢∑ L2M j −w *M w − L2M j * ∑ M v ⎥ j ≥ 3 ⎣w=1 v=1 ⎦

We assumed that only FBGCs (≥ 3 nuclei) can have more than one lymphocyte interaction with max of 2 lymphocytes, complexes do not fuse with other complexed cells, and fused or complexed cells do not detach. Like uncomplexed macrophages and

FBGCs, the fusion processes for complexed macrophages and FBGCs are bounded by

- 188 - Chapter VI

FBGCs with maximum N nuclei. Thus, macrophages with j nuclei can fuse with all other cells up to N-j nuclei.

These differential equations were solved utilizing the ODE15s function in

MATLAB.

Model Variables Related to Experimental Measurements

Model outputs are computed for comparison with observed variables obtained from in vitro experiments. Adherent monocytes and macrophages with fewer than 2 nuclei are grouped together and named FBGC precursors (P):

2 P = ∑()M q + L1M q + L2 M q (16) q=0

The number density of FBGCs (F) is

N F = ∑(M q + L1M q + L2 M q ) (17) q=3 which includes individual as well as complexed cells. The total number of surface- adherent cells is determined by:

AT = P + F (18)

The number density of nuclei in the FBGC precursor population is

1 N M = ∑()M q + L1M q +2()L1M 2 (19) q=0 while the number density of nuclei in the FBGCs is

N N F = ∑ q[]M q + L1M q + L2 M q (20) q=3

The number density of nuclei of all adherent cells is then

NT = N M + N F (21)

- 189 - Chapter VI

The percent fusion on the surface can be determined by:

N F X F = (22) NT

There are two types of adhered lymphocytes: surface-adherent and cell-associated. The adherent lymphocyte population can be determined and accounted for with:

LS = L (23)

While the cell-associated lymphocytes can have single interactions or double interactions in complexes expressed by:

N N LC = ∑ L1M q + 2∑ L2 M q (24) q=0 q=3

Finally, the total number density of adherent lymphocytes can be determined by:

LT = LS + LC (25)

Therefore, the simulated model output provided total adherent nuclei and cells, total

FBGC nuclei and cells, percent fusion, and total adherent lymphocyte density over 10 days of culture for comparison with experimental data.

Sensitivity Analysis and Parameter Estimation

After implementation of the model equations by the MATLAB software, a sensitivity analysis of model parameters was performed. Model parameters were varied by factors of ±2, ±5, and ±10 to examine how the model output changed. With this analysis we determined parameters that greatly influenced model output and parameters with less or little effect on the model output. The objective is to ensure that the model emphasizes the parameters (i.e. mechanisms) that play the most important roles in the biological system. Additionally, previous research or literature provided a basis for making assumptions (e.g. hypotheses) on certain biological processes (i.e. parameters).

- 190 - Chapter VI

These parameter assumptions along with sensitivity analysis allowed development of a model with some parameters that could be set to a constant. By reducing the number of estimated parameters, the optimization procedure could be more precise. Based on sensitivity analysis, the literature, and assumptions/hypotheses, the rate coefficients (i.e. parameters) were either set to a constant (Table 6.3) or estimated (Table 6.4).

- 191 -

Table 6.3: Model Parameter Names and Values used in Simulations

Symbols Rate Constant Value (hr-1) Comments Sources Direct biomaterial adhesion is limited and η Lymphocyte Adhesion 0.00035 Chapter V L independent of biomaterial surface Based on morphological visualization, the majority of adherent monocytes have differentiated to γ Monocyte Differentiation 0.75 macrophages. Hypothesized to be independent of biomaterial surface chemistries Lymphocyte interaction has not been shown to Differentiation due to Lymphocyte influence monocyte differentiation on biomaterials. γ 0.75 L Interaction Hypothesized to be the same as without direct lymphocyte interaction - 192 Output not very sensitive to parameter variation α Adherent Lymphocyte Apoptosis 0.00008 L and no experimental data available at this time

Lymphocytes predominantly detach from the δ Lymphocyte Detachment 0.925 Chapter V L surface since direct biomaterial adhesion is limited

Lymphocytes proliferation occurs in in vitro co- cultures. Only lymphocyte population affected by π Adherent Lymphocyte Proliferation 0.0000005 parameter changes. Since other outputs are 13 minimally affected and limited lymphocyte experimental data, parameter was set to a constant.

Lymphocytes proliferation occurs in in vitro co- cultures. Affects lymphocyte population but the Nonadherent Lymphocyte majority of model outputs are not greatly affected. π 0.0000005 13 N Proliferation Sensitivity analysis along with limited lymphocyte experimental data provided justification to set to a constant value

Table 6.4: Model Parameters for Estimation

Symbols Rate Constant Value (hr-1) Comments Sources η Monocyte Adhesion Estimated

α0 Monocyte Apoptosis Estimated

α1 Macrophage Apoptosis Estimated Monocyte and macrophage detachments were δ Monocyte Detachment Estimated 0 assumed to be equal Monocyte and macrophage detachments were δ Macrophage Detachment Estimated 1 assumed to be equal - 193 β Fusion Estimated Equal to fusion rate without direct lymphocyte β Fusion due to Lymphocyte Interaction Estimated 13 L interaction

κ1 Lymphocyte dissociation Estimated

κ-1 Lymphocyte Binding Estimated

Chapter VI

The least-squares fitting of model output to experimental data (i.e. minimization of the objective function) was performed by the MATLAB lsqcurvefit optimization function. All data points for each donor were utilized for the optimization of fit between experimental data and model output. Because experimental data of the different variables were of vastly different magnitudes, particular variables with smaller values might not play a significant role in the optimization procedure. Therefore, experimental data for each condition was scaled to relatively similar magnitudes prior to lsqcurvefit optimization in order for all values to make similar contributions to the final fit. The scaling and fit was an iterative procedure involving scaling, parameter estimation, and analysis of simulated model fit via objective function and residuals. The objective function for model fitting to experimental data is

n n n * 2 * 2 * 2 Φ = w1 ∑()ATi − ATi + w2 ∑ ()Fi − Fi + w3 ∑(NTi − NTi ) i=1 i=1 i=1 (26) n n n * 2 * 2 * 2 w4 ∑()()()N Fi − N Fi + w5 ∑ X Fi − X Fi + w6 ∑ LTi − LTi i=1 i=1 i=1 where

w = Scaling factor for each variable and is dependent on the biomaterial surface

n = Number of data points for each variable [i.e. timepoints (3) and donors (3-4)]

A, F, N, X, L = Model output

A*, F*, N*, X*, L* = Corresponding experimental data

Table 6.5 shows the scaling factors for each biomaterial.

- 194 - Chapter VI

Table 6.5: Objective Function Scaling Factors for each Biomaterial

Biomaterial w1 w2 w3 w4 w5 w6 PET 1.0E-06 1.0E-04 1.0E-06 1.0E-05 1 1.0E-05 BDEDTC 1.0E-06 1.0E-04 1.0E-06 1.0E-05 1 1.0E-05 PAAm 4.0E-06 1.0E-04 4.0E-06 1.3E-05 1 2.5E-05 PAANa 4.0E-06 1.0E-04 3.3E-06 1.0E-05 1 3.3E-05 DMAPAAmMeI 9.1E-07 1.0E-04 9.1E-07 1.0E-05 1 5.7E-06

This procedure allowed determination of parameter values that minimized the error between the simulated model output and experimental data and provided the best model fit for all the experimentally-observed variables. These procedures were all performed to obtain a consistent basis for interpretation of results. Initial parameters were varied around values determined from a manual fit of the data in order to increase the likelihood that estimated parameter values yielded a global minimum in the objective function rather than a local minimum. The final parameter choices were those that most closely fit the data (i.e. lowest objective function).

Results

The model simulates lymphocyte, monocyte, macrophage, and FBGC numbers at the biomaterial surface over the course of direct in vitro co-culture taking into account the various processes that are known to occur. The contributions of the processes are represented by the parameters listed in Table 6.1. The model was capable of simulating the time varying adherent densities we observe in our in vitro lymphocyte/monocyte co- culture system. Figure 6.3 represents some of quantitative information provided by model simulations. We can see that after the initial monocyte adhesion, the monocyte population diminishes giving rise to macrophages at the biomaterial surface. The

- 195 - Chapter VI macrophage population also diminishes over time as they apoptose or fuse. FBGCs are the terminal cells of the foreign body reaction and gradually increase over time. The fraction of fused cells increases as the foreign body reaction increases. The model provides quantitative predictions on the cell nuclei distribution at the end of the co- culture. In addition, the model can provide the density of lymphocytes at the biomaterial surface whether biomaterial-adherent or cell-adherent. Finally, the model can also provide quantitative information on the apoptotic lymphocyte, monocyte, and macrophage populations as well as the non-adherent lymphocyte, monocyte, and macrophage populations.

Parameters were estimated by fitting the model outputs to the experimental data for each of the 5 biomaterial surfaces. Since nonlinear systems have local minima, the estimated parameter values may be misleading. Varying initial parameter values changes the convergence path. This iterative process of varying initial parameters and examining residuals and objective function was done to ensure convergence to the neighborhood of the global minimum. The initial parameters were varied sometimes by an order of magnitude to decrease the likelihood of missing the global minimum. The variation of initial parameter values for the optimization procedures for each of the surfaces along with the variation in the estimated parameter values are shown in Table 6.6. The resulting estimated parameters demonstrated variation but generally showed consistency.

- 196 - Chapter VI

Figure 6.3: Representative quantitative simulations provided by the developed model.

- 197 -

Table 6.6: Initial Parameter Values and Final Parameter Estimates for all 5 Material Surfaces

Initial Values Parameter PET BDEDTC PAAm PAANa DMAPAAmMeI η 0.69 ± 0.14 0.62 ± 0.17 0.54 ± 0.18 0.75 ± 0.16 0.68 ± 0.16

α0 0.26 ± 0.22 0.14 ± 0.10 0.28 ± 0.21 0.13 ± 0.09 0.32 ± 0.29

α1 0.25 ± 0.22 0.14 ± 0.10 0.28 ± 0.21 0.16 ± 0.13 0.12 ± 0.07 δ 0.15 ± 0.14 0.09 ± 0.05 0.16 ± 0.15 0.21 ± 0.22 0.17 ± 0.14 β 4.9E-07 ± 2.9E-07 3.5E-07 ± 3.1E-07 7.8E-07 ± 4.9E-07 1.2E-06 ± 9.3E-07 3.5E-07 ± 2.9E-07

κ1 0.08 ± 0.06 0.06 ± 0.04 0.03 ± 0.03 0.07 ± 0.04 0.11 ± 0.09 κ-1 1.1E-08 ± 7.4E-09 8.1E-09 ± 4.1E-09 9.1E-09 ± 5.4E-09 1.0E-08 ± 6.0E-09 3.9E-08 ± 3.5E-08

- 198 Final Estimated Values Parameter PET BDEDTC PAAm PAANa DMAPAAmMeI η 0.61 ± 0.10 0.57 ± 0.13 0.44 ± 0.16 0.69 ± 0.14 0.47 ± 0.10

α0 0.34 ± 0.20 0.09 ± 0.08 0.50 ± 0.18 0.15 ± 0.12 0.61 ± 0.24

α1 0.12 ± 0.08 0.13 ± 0.06 0.29 ± 0.17 0.28 ± 0.19 0.03 ± 0.04 δ 0.11 ± 0.08 0.07 ± 0.05 0.25 ± 0.11 0.24 ± 0.12 0.09 ± 0.05 β 2.7E-07 ± 7.8E-08 1.8E-07 ± 7.3E-08 1.4E-06 ± 3.9E-07 1.0E-06 ± 4.6E-07 3.2E-07 ± 1.6E-07

κ1 0.12 ± 0.03 0.10 ± 0.08 0.04 ± 0.04 0.11 ± 0.04 0.26 ± 0.10 κ-1 1.9E-08 ± 2.2E-09 1.5E-08 ± 6.8E-09 1.2E-08 ± 5.9E-09 1.3E-08 ± 4.3E-09 8.6E-08 ± 3.4E-08

Chapter VI

The final parameters for each of the biomaterial surfaces were determined to be the ones showing the minimum objective function of all the trials. These final parameters are shown with the material’s corresponding contact angles in Table 6.7. The experimental data comparison with final model output is shown in Figure 6.4. In terms of monocyte adhesion (η), the hydrophilic/anionic and hydrophilic/cationic surfaces show a greater than 2 fold difference. The apoptosis rates (α0 and α1) showed similar values on the PET, BDEDTC, PAAm, and DMAPAAmMeI surfaces. The rates of apoptosis on these surfaces are highest on PAAm which are over 4 times greater than PET and

BDEDTC, and lowest on DMAPAAmMeI. PAANa showed a large discrepancy between monocyte and macrophage apoptosis rates (> 1000 fold). Rate of monocyte apoptosis

(α0) was very low while the rate of macrophage apoptosis (α1) was higher than on PAAm.

Detachment rates (δ) were the highest on the hydrophilic/neutral PAAm surface with greater than 2 fold increase over the other surfaces. The lowest rate of detachment was on the hydrophilic/anionic (PAANa) surface which showed an almost 10 fold difference compared to the hydrophilic/neutral (PAAm) surface. The fusion coefficient (β) showed the highest levels on the hydrophilic surface regardless of charge but hydrophilic/neutral and hydrophilic/anionic surfaces were at the highest levels. In terms of rate of lymphocyte binding (κ-1) and dissociation (κ1), the hydrophilic/neutral (PAAm) and hydrophilic/anionic (PAANa) surfaces showed the lowest levels in both while hydrophilic/cationic surface showed the highest level by greater than 2 fold. All the parameter values were similar for the two hydrophobic surfaces, PET and BDEDTC.

- 199 - Chapter VI

Table 6.7: Summary of Estimated Parameter Values (hr-1)

PET BDEDTC PAAm PAANa DMAPAAmMeI η 0.49 0.65 0.61 0.80 0.29

α0 0.15 0.11 0.76 1.2E-04 0.07 α1 0.10 0.10 0.50 0.98 0.03 δ 0.13 0.12 0.24 0.03 0.09 β 3.7E-07 2.1E-07 1.3E-06 8.2E-07 5.1E-07

κ1 0.13 0.07 0.04 0.07 0.26 κ-1 2.0E-08 1.2E-08 1.1E-08 7.0E-09 9.0E-08 Surface Properties: a Contact angle (˚) 72 ± 1 77 ± 3 37 ± 2 21 ± 4 16 ± 2 Wettability Hydrophobic Hydrophobic Hydrophilic Hydrophilic Hydrophilic Charge Neutral Neutral Neutral Anionic Cationic aValues obtained from Chapter I

- 200 - Chapter VI

- 201 - Chapter VI

- 202 - Chapter VI

Figure 6.4: Modeling results from parameter estimation for (A) PET, ( B) BDEDTC, (C) PAAm, (D) PAANa, (E) DMAPAAmMeI. Individual data points represent experimental data and is shown as a mean ± the standard error of the mean (n = 3-4). Lines represent model outputs.

As Figure 6.4D shows, one area where the model did not consistently fit experimental data was in fusion (i.e. fused nuclei and percent fusion). On surfaces with low levels of fusion, the model simulated the experimental results well. However, on

PAANa, specifically, where the levels of fusion were significantly higher, the model failed to account for the fast rise in fusion observed experimentally by 3 days. The model simulations regardless of initial parameters values, consistently underestimated the experimental data.

Discussion

The purpose of this work was to develop a model that described the events occurring at the biomaterial surface of an in vitro lymphocyte and monocyte co-culture

- 203 - Chapter VI system. The developed model incorporates all of these essential processes (e.g. adhesion, differentiation, apoptosis, detachment, proliferation, fusion, binding, and dissociation) as model parameters to provide a quantitative simulation of the cellular populations at the biomaterial surface. This work builds on previous models of the FBR by incorporating lymphocyte interactions and behaviors. The model captured the important time varying cell populations on biomaterial surfaces and provided a way to analyze the effects of different biomaterial surface chemistries and properties on the model parameters.

Adhesion Parameter Estimates

Optimization of the fit between experimental data and model outputs allowed determination of the parameter values specific to each biomaterial surface. The monocyte adhesion rate was lowest on the hydrophilic/cationic surfaces compared to the other surfaces which were relatively similar. This can possibly be explained by the mechanisms of cellular adhesion. In the presence of serum proteins, a layer of adsorbed proteins initially contact the surface. It is to this protein layer that cells generally adhere.

Different biomaterial surfaces have been shown to dictate the type, conformation, and distribution of adsorbed proteins.17,18 These protein layer variations can explain the differential biological observations such as monocyte adhesion. Monocyte adhesion occurs within hours of contact with biomaterial surfaces and previous studies have shown that regardless of surface properties (hydrophobic or charged) of plasma-modified polystyrene surfaces, monocytes adhered at similarly high levels initially.16 Jenney et al. demonstrated that even hydrophilic PEO-immobilized surfaces, which are known to inhibit adhesion, showed initially comparable monocyte adhesion levels and found that inhibition of adhesion occurred in the long term (i.e. past 3 days).19 Other than the

- 204 - Chapter VI estimated monocyte adhesion rate on hydrophilic/cationic (DMAPAAmMeI) all of the other parameter estimates were comparable high. Thus, in general the model accurately predicted the similarly high monocyte adhesion rates on all surfaces despite lack of experimental data at early time points (i.e. < 3 days). The underestimated monocyte adhesion rate on hydrophilic/cationic surfaces may be due to limited experimental data particularly during the early culture time points.

Apoptosis Parameter Estimates

Analysis of the monocyte and macrophage apoptosis rate constants showed that the model predicted the highest rates of apoptosis on hydrophilic/neutral (PAAm) and hydrophilic/anionic (PAANa) surfaces compared to the other hydrophobic (PET and

BDEDTC) and hydrophilic/cationic (DMAPAAmMeI) surfaces. This is consistent with previous studies on biomaterial-induced apoptosis where these hydrophilic/neutral and hydrophilic/anionic surfaces were shown to increase adherent macrophage apoptosis compared to hydrophobic and hydrophilic/cationic surfaces. Apoptosis has been shown to be mediated by TNF-α as well as by disruption of adhesion mechanisms.20,21 We have found that on these hydrophilic/neutral and hydrophilic/anionic surfaces, macrophages are activated to produce higher levels of TNF-α compared to the other surfaces (See

Chapters II and III).14,15 This provides some potential mechanisms for why certain surfaces may induce higher levels of apoptosis. Additionally, we have potential surface properties or molecules that we may be able to exploit in biomaterial design in order to dictate the desired cellular adhesion response. The experimental data utilized for parameter estimation in model fitting did not include apoptosis data. Despite this, the model was capable of making accurate predictions of how certain biomaterials would

- 205 - Chapter VI influence apoptosis. This provides validation for the use of this model in describing the lymphocyte/monocyte in vitro co-culture system, interpreting the results, and making predictions.

Detachment Parameter Estimates

The hydrophilic/anionic surface showed the lowest detachment rate coefficient.

This suggests that the adhesion mechanisms on this surface may be more specific or stronger compared to the others. As mentioned, adherent cells make interactions with the adsorbed protein layer. It is possible that the type and conformation of the adsorbed proteins are favorable for monocyte and macrophage integrins. Another possibility is that the adherent monocytes and macrophages are activated and accelerated along the path to fusion and formation of foreign body giant cells thus increasing the cellular surface area for biomaterial contact and adhesion. These are hypotheses that could be addressed in future studies.

Fusion Parameter Estimates

The fusion rate constants were all higher on hydrophilic surfaces than the hydrophobic counterparts regardless of surface charge. However, hydrophilic/cationic surfaces showed a lower fusion rate relative to the hydrophilic/neutral and hydrophilic/cationic surfaces. These two surfaces were shown previously in Chapters II and III to induce higher levels of activation and macrophage-derived inflammatory mediators.14,15 Although fusion mediators and mechanisms have not been fully elucidated, soluble factors such as IL-4, IL-13, and MCP-1 have been shown to induce fusion.22-25 Therefore, we can speculate that the elevated levels of activation and cytokine production would include fusion inducers. Additionally, McNally et al.

- 206 - Chapter VI demonstrated vitronectin adsorption on biomaterial surfaces, rather than other proteins

(i.e. complement C3bi, collagen types I or IV, fibrinogen, plasma fibronectin, fibroblast fibronectin, laminin, thrombospondin, or von Willebrand factor), to be critical in supporting prolonged adhesion and subsequent development of IL-4-induced foreign body giant cells in vitro.26 Keselowsky et al. found that plasma fibronectin may play a role in the fusion process in vivo.27 However, the specific mechanism of involvement

(e.g. adsorption) was not determined. Although vitronectin is generally thought to be universally adsorbed,17,28 it is possible that the hydrophilic/neutral and hydrophilic/anionic surfaces may enhance adsorption of vitronectin or induce distinct conformations. Bale et al. determined that vitronectin adsorbed onto carboxylic acid- containing copolymers showed an enhanced reactivity to conformationally sensitive vitronectin antibodies compared to vitronectin adsorbed onto polystyrene.28 The chemical structure of the hydrophilic/anionic surface (PAANa) is sodium salt of polyacrylic acid which has carboxylic acid as one of its functional groups. Additionally, fibronectin has been demonstrated to retain functionality on hydrophilic surfaces and shown to adsorb more readily on negatively charged surface compared to hydrophilic

29 neutral (OH) but not on hydrophobic nonpolar (CH3) surfaces. Thus, additional studies into protein adsorption, particularly vitronectin and fibronectin, and protein conformational states on varying biomaterial surface chemistries is warranted.

Lymphocyte Interactions Parameter Estimates

Lymphocyte interactions with monocytes, macrophages, and foreign body giant cells were also analyzed through the binding and dissociation rate constants. The hydrophilic/neutral and hydrophilic/anionic surfaces once again showed differences

- 207 - Chapter VI compared to the other surfaces. Here, the parameter estimates for lymphocyte binding and dissociation were lower than on other surfaces. This suggests that the membrane intercellular adhesion or stimulatory molecules may have lower affinities on these two surfaces; however, when they do interact, the adhesion between the two cells may be stronger and more stable. Many surface molecules that mediate cellular interactions (e.g.

MHC molecules and integrins) on both macrophages and lymphocytes are upregulated upon cellular activation or chemokine ligation.30-34 These two surfaces have previously been shown to induce increased macrophage activation as well as cytokine and chemokine levels (refer to Chapters II and III).14,15,35 Additionally, these surfaces were shown in Chapter IV to promote macrophage- and FBGC-adherent lymphocytes. The developed model provides some hypotheses into biomaterial-dependent interactions that prompt validation by further experimental studies. At present the relevant molecular mechanisms of interaction are unknown in the context of biomaterial surfaces. Future studies will need to focus on identifying these molecules as well as investigating the effect of varying biomaterial surface properties on these molecular interaction times, affinities, and strengths.

Model Limitations

Assumptions and simplifications were made in the development of this model.

The detachment rates for monocytes and macrophages were assumed to be constant and equal. This distinction should be considered in future modeling studies. The monocyte differentiation rate was also assumed to be constant but some studies have shown differentiation of monocytes to be influenced by biomaterials.36,37 Therefore, this parameter should be considered to be biomaterial-dependent for future work.

- 208 - Chapter VI

Lymphocyte apoptosis and effects on monocyte differentiation in response to biomaterials are currently unknown. Due to limited experimental data for model development in these areas in addition to sensitivity analysis, lymphocyte apoptosis was set to a constant while monocyte differentiation rates were the same regardless of direct lymphocyte interaction. These processes require further experimental investigation and future modeling studies should consider estimation of these parameters. In addition, studies have shown induction of lymphocyte proliferation on these biomaterial surfaces with differential effects by biomaterial surface chemistries.13,38 Again, due to limited experimental data for model fitting and the fact that analysis of the parameter sensitivity revealed that only the lymphocyte population of the model output appeared to be affected, proliferation was held to a constant value.

In analyzing the optimized model outputs and the experimental data, one area of inconsistency between the model output and the experimental data was in simulating responses to biomaterials with high fusion states. A limitation in representing this behavior was the relatively high variability in the levels of fusion for each donor. The model was not capable of accounting for relatively high early levels of fusion at day 3 and resulted in underestimation. Experimentally, fusion occurred by day 3 with subsequent small increases over time. This was true for both the hydrophilic/neutral and hydrophilic/cationic surfaces. This suggests that the early behaviors during the progression to fusion including the fusion rate could potentially be time dependent. The rate of differentiation or fusion for instance could be greater in the initial stages (i.e. < 3 days) than the later stages (i.e. > 3 days). This is possible as some inflammatory mediators such as IL-1β, TNF-α, MIP-1β have been shown to be produced at a high level

- 209 - Chapter VI early in time (i.e. by 3 days) with subsequent decrease over time (Chapters II and

III).14,15,35 Therefore, through analysis of inconsistencies or deficiencies in the model, we are still provided with some information on the fusion process/mechanism that we may not have discerned simply by looking at the experimental data.

Effects of Biomaterial Surface Properties

Overall, despite some differences between hydrophilic/neutral and hydrophilic/anionic surfaces, these two surfaces appear to set themselves apart from the hydrophobic and hydrophilic/cationic surfaces. The hydrophobic surfaces, PET and

BDEDTC, showed similar parameter estimates which was expected since other than their surface chemistry, biomaterial surface properties were similar. From analysis of the parameters compared to the corresponding biomaterial properties, we find that the relationship between a biomaterial surface property (e.g. contact angle) and a biological process is generally not an exact correlation. These results show that the processes in the in vitro co-cultures are influenced by multiple surface properties, namely ionic character and degree of surface wettability. Hydrophilicity/hydrophobicity surface classification may correlate to a particular response such as fusion in this case but is generally not sufficient to determine the complete biological response. Surface charge is also important. Perhaps if the sole desire for a biomaterial application is modulation of macrophage fusion, we could focus on hydrophobicity and hydrophilicity. Therefore, future modeling studies for quantitative predictability of the biological response to biomaterials will require acquisition of experimental data from biomaterials of differing surface wettability and charge (i.e. ionic potentials). Although this study focused on these two surface properties, other material properties such as surface chemistry and

- 210 - Chapter VI

39 other physical properties (e.g. Tg) may play a role. Abramson et al. predicted the growth rate of fibroblasts on biomaterial surfaces according to various biomaterial

39 properties: contact angle, Tg, and chemical composition. If we can correlate parameters to biomaterial properties, then we can potentially predict the biological response to novel biomaterial surfaces or contribute to the design of specific biomaterial surfaces to achieve the desired responses. The ultimate goal would be to be able to input a set of biomaterial parameters into a model which would then output the predicted biological responses.

Conclusions

The dynamic systems model that we have developed aimed to describe an in vitro lymphocyte/monocyte co-culture and provide outputs simulating the cell populations at the biomaterial surface over time. However, the model is limited by sparse experimental data, experimental variability, and assumptions/simplifications. Despite these limitations the model provides a tool for analyzing the cellular response to biomaterials in an in vitro co-culture of lymphocytes and monocytes. The model was capable of making predictions that were validated by experimental results in the literature (e.g. biomaterial-dependent macrophage apoptosis) as well as providing predictions/hypotheses for future experimental studies. Future modeling studies will also require acquisition of more experimental biological response data over the time course. In addition, more materials

(i.e. biomaterials with a wide range of physical and chemical characteristics) along with the corresponding material properties data are also warranted to provide better predictive capabilities. To do so, there is a need for developing easier and faster methods of acquiring quantitative biological data. Addressing these various concerns will help move us closer to the goal of rational biomaterial design.

- 211 - Chapter VI

References

1. Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81-110.

2. Zhu C, Williams TE. Modeling concurrent binding of multiple molecular species in cell adhesion. Biophys J 2000;79(4):1850-7.

3. N'Dri NA, Shyy W, Tran-Son-Tay R. Computational modeling of cell adhesion and movement using a continuum-kinetics approach. Biophys J 2003;85(4):2273- 86.

4. Bigerelle M, Anselme K. A kinetic approach to osteoblast adhesion on biomaterial surface. J Biomed Mater Res A 2005;75(3):530-40.

5. Bajzer Z, Vuk-Pavlovic S. Modeling positive regulatory feedbacks in cell-cell interactions. Biosystems 2005;80(1):1-10.

6. Macallan DC, Asquith B, Irvine AJ, Wallace DL, Worth A, Ghattas H, Zhang Y, Griffin GE, Tough DF, Beverley PC. Measurement and modeling of human T cell kinetics. Eur J Immunol 2003;33(8):2316-26.

7. Simon SI, Chambers JD, Sklar LA. Flow cytometric analysis and modeling of cell-cell adhesive interactions: the neutrophil as a model. J Cell Biol 1990;111(6 Pt 1):2747-56.

8. Ludewig B, Krebs P, Junt T, Metters H, Ford NJ, Anderson RM, Bocharov G. Determining control parameters for dendritic cell-cytotoxic T lymphocyte interaction. Eur J Immunol 2004;34(9):2407-18.

9. Zhao OH, Anderson JM, Hiltner A, Lodoen GA, Payet CR. Theoretical analysis on cell size distribution and kinetics of foreign-body giant cell formation in vivo on polyurethane elastomers. J Biomed Mater Res 1992;26(8):1019-38.

10. Kao WJ, Hiltner A, Anderson JM, Lodoen GA. Theoretical analysis of in vivo macrophage adhesion and foreign body giant cell formation on strained poly(etherurethane urea) elastomers. J Biomed Mater Res 1994;28(7):819-29.

11. Kao WJ, Zhao QH, Hiltner A, Anderson JM. Theoretical analysis of in vivo macrophage adhesion and foreign body giant cell formation on polydimethylsiloxane, low density polyethylene, and polyetherurethanes. J Biomed Mater Res 1994;28(1):73-9.

12. Jones JA. Biomaterials and the Foreign Body Reaction: Surface Chemistry Dependent Macrophage Adhesion, Fusion, Apoptosis, and Cytokine Production. Dissertation. Case Western Reserve University, 2007.

- 212 - Chapter VI

13. Brodbeck WG, Macewan M, Colton E, Meyerson H, Anderson JM. Lymphocytes and the foreign body response: lymphocyte enhancement of macrophage adhesion and fusion. J Biomed Mater Res A 2005;74(2):222-9.

15. Chang DT, Colton E, Anderson JM. Paracrine and juxtacrine lymphocyte enhancement of adherent macrophage and foreign body giant cell activation. J Biomed Mater Res A 2008.

16. McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 1994;91(21):10119-23.

17. Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of biomaterial-cell interactions by adsorbed proteins: a review. Tissue Eng 2005;11(1-2):1-18.

18. Xu LC, Siedlecki CA. Effects of surface wettability and contact time on protein adhesion to biomaterial surfaces. Biomaterials 2007;28(22):3273-83.

19. Jenney CR, Anderson JM. Effects of surface-coupled polyethylene oxide on human macrophage adhesion and foreign body giant cell formation in vitro. J Biomed Mater Res 1999;44(2):206-16.

20. Brodbeck WG, Shive MS, Colton E, Ziats NP, Anderson JM. Interleukin-4 inhibits tumor necrosis factor-alpha-induced and spontaneous apoptosis of biomaterial-adherent macrophages. J Lab Clin Med 2002;139(2):90-100.

21. Frisch SM, Screaton RA. Anoikis mechanisms. Curr Opin Cell Biol 2001;13(5):555-62.

22. DeFife KM, Jenney CR, McNally AK, Colton E, Anderson JM. Interleukin-13 induces human monocyte/macrophage fusion and macrophage mannose receptor expression. J Immunol 1997;158(7):3385-90.

23. McNally AK. Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 1995;147(5):1487-99.

24. Kao WJ, McNally AK, Hiltner A, Anderson JM. Role for interleukin-4 in foreign- body giant cell formation on a poly(etherurethane urea) in vivo. J Biomed Mater Res 1995;29(10):1267-75.

25. Kyriakides TR, Foster MJ, Keeney GE, Tsai A, Giachelli CM, Clark-Lewis I, Rollins BJ, Bornstein P. The CC chemokine ligand, CCL2/MCP1, participates in

- 213 - Chapter VI

macrophage fusion and foreign body giant cell formation. Am J Pathol 2004;165(6):2157-66.

26. McNally AK, Jones JA, Macewan SR, Colton E, Anderson JM. Vitronectin is a critical protein adhesion substrate for IL-4-induced foreign body giant cell formation. J Biomed Mater Res A 2007.

27. Keselowsky BG, Bridges AW, Burns KL, Tate CC, Babensee JE, LaPlaca MC, Garcia AJ. Role of plasma fibronectin in the foreign body response to biomaterials. Biomaterials 2007;28(25):3626-31.

28. Bale MD, Wohlfahrt LA, Mosher DF, Tomasini B, Sutton RC. Identification of vitronectin as a major plasma protein adsorbed on polymer surfaces of different copolymer composition. Blood 1989;74(8):2698-706.

29. Scotchford CA, Gilmore CP, Cooper E, Leggett GJ, Downes S. Protein adsorption and human osteoblast-like cell attachment and growth on alkylthiol on gold self- assembled monolayers. J Biomed Mater Res 2002;59(1):84-99.

30. Miyamoto YJ, Andruss BF, Mitchell JS, Billard MJ, McIntyre BW. Diverse roles of integrins in human T lymphocyte biology. Immunol Res 2003;27(1):71-84.

31. Hogg N, Laschinger M, Giles K, McDowall A. T-cell integrins: more than just sticking points. J Cell Sci 2003;116(Pt 23):4695-705.

32. Maher SG, Romero-Weaver AL, Scarzello AJ, Gamero AM. Interferon: cellular executioner or white knight? Curr Med Chem 2007;14(12):1279-89.

33. Ashany D, Song X, Lacy E, Nikolic-Zugic J, Friedman SM, Elkon KB. Th1 CD4+ lymphocytes delete activated macrophages through the Fas/APO-1 antigen pathway. Proc Natl Acad Sci U S A 1995;92(24):11225-9.

34. Santana MA, Rosenstein Y. What it takes to become an effector T cell: the process, the cells involved, and the mechanisms. J Cell Physiol 2003;195(3):392- 401.

35. Jones JA, Chang DT, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Proteomic analysis and quantification of cytokines and chemokines from biomaterial surface-adherent macrophages and foreign body giant cells. J Biomed Mater Res A 2007;83(3):585-96.

36. Collier TO, Anderson JM, Brodbeck WG, Barber T, Healy KE. Inhibition of macrophage development and foreign body giant cell formation by hydrophilic interpenetrating polymer network. J Biomed Mater Res A 2004;69(4):644-50.

37. Dinnes DL, Santerre JP, Labow RS. Influence of biodegradable and non- biodegradable material surfaces on the differentiation of human monocyte-derived macrophages. Differentiation 2008;76(3):232-44.

- 214 - Chapter VI

38. MacEwan MR, Brodbeck WG, Matsuda T, Anderson JM. Student Research Award in the Undergraduate Degree Candidate category, 30th Annual Meeting of the Society for Biomaterials, Memphis, Tennessee, April 27-30, 2005. Monocyte/lymphocyte interactions and the foreign body response: in vitro effects of biomaterial surface chemistry. J Biomed Mater Res A 2005;74(3):285-93.

39. Abramson SD, Alexe G, Hammer PL, Kohn J. A computational approach to predicting cell growth on polymeric biomaterials. J Biomed Mater Res A 2005;73(1):116-24.

- 215 - Chapter VII

Conclusions and Future Directions

Summary and Conclusions

The goal of this work was two-fold: (1) to gain an understanding of the

interactions between lymphocytes and monocytes/macrophages in the response to

biomaterial surfaces and (2) to identify the relationship between material surface

properties and corresponding biological responses for optimizing the utilization of

biomaterials as an enabling technology in implant applications. The conducted studies

utilized in vitro lymphocyte and monocyte co-cultures to investigate the activities between these two important inflammatory leukocytes when confronted with a foreign body such as an implanted biomaterial. Influences of biomaterial surface properties on cellular responses such as adhesion, cytokine production, and activation were examined with model PET-based molecularly-engineered surfaces displaying distinct hydrophobic/neutral, hydrophilic/neutral, hydrophilic/anionic, and hydrophilic/cationic

surfaces. Surface wettability and ionic character are two potential material surface

parameters that can be modulated to control cellular behavior and achieve desired

biological responses for the biomaterial application. With the current understanding of

the foreign body reaction and lymphocyte interactions at biomaterial surfaces, a

mechanistic model of our in vitro co-culture was developed for quantitative predictability

and analysis of the important cellular processes that exist in lymphocyte and macrophage

responses to biomaterial surfaces.

- 216 - Chapter VII

Inflammatory Mediator Response to Biomaterials

In Chapter II, we examined the protein products secreted by lymphocytes and macrophages in co-culture to gain insight on the cellular activities and interactions that occur at biomaterial surfaces. Utilizing protein cytokine arrays and ELISA, we found that essentially all the inflammatory mediators were macrophage-derived and capable of controlling cellular responses such at recruitment, migration, and activation at the biomaterial surface. The protein arrays allowed detection of a multitude of inflammatory mediators, but the limitation is that some proteins may be produced below the minimum detection levels. This tool was useful for assessing the role of soluble mediators in the response to biomaterials as we were able to determine potential factors that may be important in the fusion process including GRO-α, RANTES, MIP-1β, TNF-α, MIP-3α, and IP-10. These represent potential targets for inhibiting the foreign body reaction.

ELISA quantification of select inflammatory mediators suggested that the hydrophilic/neutral and hydrophilic/anionic surfaces promote pro-inflammatory responses and reduced ECM degradation relative to other surface properties while hydrophilic/cationic surfaces favored enhanced potential for ECM breakdown. The lack of lymphokine production as measured by protein array suggested that the lymphocyte population was not activated in these co-cultures when exposed to biomaterial surfaces.

Lymphocyte Effects on Macrophages

Due to previous work in our laboratory demonstrating lymphocyte enhancement of monocyte adhesion and macrophage fusion (i.e. the FBR),1 we wanted to determine whether lymphocytes played a role in the general inflammatory response through effects on macrophage-derived products. To this aim, we examined how lymphocytes affected

- 217 - Chapter VII macrophage activation as determined by the production of inflammatory mediators

(Chapter III). The results showed that lymphocytes enhanced pro-inflammatory cytokine production but did not significantly affect anti-inflammatory or ECM-modifying proteins.

The mechanisms of this enhancement occurred initially through paracrine interactions

(i.e. indirect lymphokine factors) and subsequently through juxtacrine interactions (i.e. direct cell-cell contact). In agreement with Chapter II results, hydrophilic/neutral and hydrophilic/anionic surfaces evoked the highest levels of activation but showed differences in the profile of highly produced cytokines and chemokines. The finding that the hydrophilic surface, which minimizes monocyte and macrophage adhesion, is highly activating challenges the idea that using or designing a hydrophilic surface to minimize adhesion of inflammatory cells will result in a more biocompatible surface.

Lymphocyte Adhesion and Interactions

On the flipside, in Chapter IV, we addressed the adhesion behavior of specific lymphocyte subpopulations when exposed to foreign synthetic materials and how adherent monocytes during the progression of the foreign body reaction and different biomaterial surface properties influenced this lymphocyte response. We found that lymphocytes showed limited direct biomaterial adhesion which was consistent across all surfaces utilized (i.e. hydrophobic, hydrophilic/neutral, and hydrophilic/anionic) and were predominantly adherent to monocytes, macrophages, and foreign body giant cells.

In other words, cell to cell interactions were the primary mode of lymphocyte adhesion at the biomaterial surface. CD4+ and CD8+ T cells were the major lymphocytes engaged in physical interaction while the very small number of NK cells found at the biomaterial surface were all found to be adherent to macrophages. In addition, biomaterial surface

- 218 - Chapter VII property influenced not only the density of adherent-lymphocyte but also selectivity of adherent lymphocytes. The hydrophilic/neutral surface inhibited both monocyte/macrophage adhesion but promoted lymphocyte interactions, particularly

CD4+ T cells, with the sparsely adherent cells. The hydrophilic/anionic surface promoted monocyte and macrophage adhesion and fusion to form FBGCs as well as lymphocyte adhesion, with no selectivity. The hydrophobic surface showed low levels of adherent lymphocytes, selective for CD8+ T cells, despite significant monocyte and macrophage adhesion. The findings suggest that biomaterial-induced phenotypic differences in the cell populations are playing a role and not simply a matter of adherent monocyte, macrophage, and FBGC density differences. The differential capability of lymphocytes to engage in stable and strong interactions with monocytes, macrophages, and FBGCs on the biomaterial surfaces further emphasizes the potential for biomaterial surface properties to dictate certain desired cellular processes. The fact that FBGCs are interacting with lymphocytes is a significant finding as we have not found evidence in the literature demonstrating this observation. There have only been limited studies of FBGC phenotypic characteristics demonstrating potential lymphocyte interactions through expression of HLA-DR and co-stimulatory B7 molecules.2,3

Macrophages Effects on Lymphocytes

The findings from Chapter III showed that macrophage-derived inflammatory mediators were enhanced by lymphocyte-derived soluble factors. Although we did not initially detect IFN-γ by protein array, we did not rule this activator out due to array sensitivity limitations. The protein array did, however, indicate the production of IP-10.

In addition to being a chemokine for activated lymphocytes, this factor is also induced by

- 219 - Chapter VII interferon-γ. In Chapter V we identified and quantified IFN-γ production from both direct and indirect lymphocyte/monocyte co-cultures but not in individual cultures. The significance of this is two-fold. First, it suggests lymphocyte activation at biomaterial surfaces as a result of macrophage-derived factors (i.e. non-specific activation).

Secondly, it represents a potential modulator of monocyte, macrophage, and FBGC behavior including adhesion, activation, and fusion. The finding that indirect lymphocyte interactions with monocyte, macrophages, and FBGCs showed biomaterial independent

IFN-γ production while direct lymphocyte interactions with adherent monocytes, macrophages, and FBGCs enhanced IFN-γ secretion and evoked a differential IFN-γ response was also significant. Relative to all other surfaces, the increase due to direct cell-cell interactions on the hydrophilic/anionic surface was much larger. This is the same surface that evoked the highest level of FBGC formation and monocyte-, macrophage, and FBGC-adherent lymphocytes (Chapter IV). Clearly, the profile of cell types and phenotypes are significantly biomaterial dependent. The correlation between adherent FBGC density, FBGC-adherent lymphocyte density, and lymphocyte activation has implications for potential FBGC involvement.

Dynamic Systems Model

The mechanistic model describing lymphocyte and monocyte responses and interactions at biomaterial surfaces presented in Chapter VI provided a tool for analysis of the in vitro co-culture system. This tool allowed insight into the relative importance of particular processes to the model output. Additionally, the model was capable of making predictions that were validated by experimental results in the literature (e.g. biomaterial- dependent macrophage apoptosis) as well as providing predictions/hypotheses for future

- 220 - Chapter VII

experimental studies. Thus, the model can be useful for gaining a better understanding of the overall system.

The studies described in Chapters II – V have shown that lymphocytes and monocytes/macrophages can mutually affect each other in responding to biomaterial

surfaces. Rather than simply being innocent bystanders at biomaterial implant sites, lymphocytes play a role in the biological response. The lymphocytes do not have the capability or the predilection to directly engage foreign bodies as phagocytes such as neutrophils or macrophages would. The work indicates that lymphocytes do not prefer to

directly adhere to biomaterial surfaces, but act through interactions with biomaterial-

adherent monocytes, macrophages, and FBGCs. These interactions include both direct

juxtacrine as well as indirect paracrine interactions and include lymphocyte effects on

monocyte, macrophage, and FBGC activation as well as monocyte, macrophage, and

FBGC effects on lymphocyte interactions and activation.

Biological Implications

The finding that lymphocytes are capable of becoming activated by noncontact

mechanisms has immunological and clinical implications. The specific activated

lymphocyte populations in these co-cultures have not been determined, but different

responses would be expect according to the particular activated lymphocyte

subpopulations. There is evidence in the literature for clinical consequences of

biomaterial implant induced lymphocyte activation. Specifically, patients with left

ventricular assist devices (LVAD) have developed decreased numbers of CD4+ Th1 cell

subpopulations due to T lymphocyte activation induced cell death resulting in increased

susceptibility to infection.4 Moreover, these patients show hyperreactive B cells with

- 221 - Chapter VII increased production of anti-HLA antibodies.5 Although Itescu et al. have demonstrated that T lymphocytes are indeed activated by LVAD biomaterials as measured by CD40L expression and CD25 and determined that B cells are subsequently stimulated by activated T cells,4 they have not focused on the mechanisms occurring at the biomaterial surface. The findings shown in Chapters III – V describe the effects of interactions between lymphocytes and cells of the monocyte lineage and contribute to potential mechanisms for the biomaterial-induced lymphocyte activation states observed clinically.

Significance to Biomaterials

Throughout the previous chapters, we illustrated the significant effects biomaterial surface properties can have on a variety of cellular responses, namely adhesion, fusion, activation, interactions, and cytokine production. The model biomaterial surfaces utilized in these studies provide homogeneous modified surfaces for examination of hydrophilicity/hydrophobicity and charge. As demonstrated, both surface properties have influenced lymphocyte and monocyte activities and interactions.

Therefore, these surface properties represent parameters that can be modulated in biomaterial design to achieve the desired set of responses.

At this stage, the biological response to biomaterials is not fully characterized nor is the precise relationship between material surface properties and cellular responses fully understood. Therefore, the particular biomaterial selected and utilized in a specific application is not an exact science and is more often than not, suboptimal. Failure to consider the biological response to specific surface properties can lead to devastating consequences. For instance, a drug delivery device that fails to account for the foreign body reaction and associated degradative responses could end up releasing drug too

- 222 - Chapter VII

quickly. The findings presented in Chapters II and III show that despite inhibition of

monocyte and macrophage adhesion on the model hydrophilic surface (PAAm), the

minimally adherent cells were still highly activated. Unless the surface were ideal and

completely abolished protein adsorption and cellular adhesion, an inflammatory response

could still be elicited leading to FBGC formation, subsequent degradation, as well as the

wound healing response. Researchers intending to design a biocompatible biomaterial by minimizing cellular adhesion must therefore, take this into account.

The biomaterial surface chemistry dependent findings presented in the previous chapters can be exploited in future biomaterial applications. The hydrophilic/anionic

PET-modified PAANa chemistry can potentially be used for biodegradable scaffolds and

taking advantage of the FBR and the degradative response associated with it. Whether all

hydrophilic/anionic surface chemistries elicit this same response remains to be seen.

Most likely, the degree of hydrophilicity and charge will play a role. Additionally, the

differential lymphocyte activation and interactions evoked on the varying biomaterial

surfaces present another opportunity for material surface property control of desired

responses. Biomaterials are being investigated as adjuvants due to their capability to

induce dendritic cell maturation and enhance the immune response to antigen delivered in

a combination product.6-9 An effective adjuvant is one where antigen is efficiently

delivered to a large population of antigen presenting cells such as dendritic cells and enhances the antigen-specific activation of immune cells. Macrophages are antigen presenting cells as well and the introduction of biomaterials into the body would results in abundant macrophage interactions at the site of the biomaterial. The hydrophilic/anionic surface promotes monocyte and macrophage adhesion as well as

- 223 - Chapter VII

lymphocyte interactions with biomaterial-adherent monocytes, macrophages, and FBGCs

leading to enhanced macrophage and lymphocyte activation (Chapter III – V). The

charged nature of the surface chemistry is also advantageous as adsorbed antigen to

charged microparticles has been shown to enhance the immune responses.10 Therefore,

this biomaterial surface chemistry has potential for application as an adjuvant.

Conclusions

Overall, this work provides mechanistic insight into lymphocyte interactions with

the monocyte lineage of cells involved in the foreign body reaction and how these cells

influence one another. In addition, biomaterial surface chemistries were shown to induce

significant differences in cellular responses. These findings contribute to an

understanding of how biomaterial-induced responses correlate with material surface

properties which is pivotal for novel biomaterial design and application.

Future Directions

The work presented in the preceding chapters contribute towards elucidating the complexities associated with the biological response to biomaterial surfaces as well as

builds on the database of knowledge on the relationship between biomaterial surface

properties and associated cellular responses. This work extends the current state of

knowledge, which has focused primarily on macrophage responses (e.g. adhesion,

activation, apoptosis, and fusion) to biomaterial surfaces, to include lymphocyte

interactions with adherent monocytes, macrophages, and FBGCs and their mutual effects.

However, we are still in the process of gaining a complete mechanistic understanding of

- 224 - Chapter VII the foreign body reaction to biomaterials as well as detailed mechanisms of lymphocyte interactions with monocytes, macrophages, and FBGCs.

Fusion and Foreign Body Reaction

In terms of the foreign body reaction, we are still in search of specific molecular mediators of fusion. Several potential targets were presented in Chapter II. Inhibition studies could help to determine whether they play a role. In terms of lymphocyte responses to biomaterial surfaces, there are still many unknowns. Lymphocyte activation-induced cell death has been described by Itescu et al.4 However, lymphocyte effects on biomaterial-adherent monocyte, macrophage, and FBGC apoptosis have not been investigated. Activated NK cells, for instance, have the capability to induce macrophage apoptosis.11,12 The direct and indirect interaction mechanisms involved in lymphocyte enhancement of macrophage behavior such as adhesion, fusion, and activation along with reciprocal activation of lymphocytes remain unknown at this time.

Potential molecular interactions between lymphocytes and macrophages described in

Table 4.7 in Chapter IV should be considered. Targeting potential interactions with inhibitions studies along with an immunocytochemical staining approach and subsequent visualization of molecular target co-localization would allow identification of the specific molecular mediators and the roles they play.

Cellular Sources of Cytokines and Role of IFN-γ

IFN-γ is a potential target as a soluble mediator in these responses. The source of this cytokine must also be identified. This could be accomplished via flow cytometric staining of intracellular cytokines and cell specific markers. Previous research identified lymphokine IL-4 and IL-13 as fusion inducers.13-15 These cytokines were not found in in

- 225 - Chapter VII vitro co-cultures indicating that the in vitro setup does not correlate to the in vivo conditions. T lymphocytes, hypothesized producers, were not required for FBGC formation in T cell deficient mice indicating the presence of alternative sources of IL-4 and IL-13.16 Prospective sources include NKT cells, NK cells, mast cells, basophils, and eosinophils. Normal and deficient animal models could be exploited for these studies.

Foreign Body Giant Cell and Lymphocyte Interactions

Foreign body giant cells were implicated in Chapters IV and V as interacting with lymphocytes and potentially playing a role in enhancing lymphocyte activation.

Lymphocytes were shown to adhere to FBGCs on biomaterial surfaces indicating specific molecular mediators that could have functional significance. Previous findings have shown potential interactions to include MHC and B7 involvement.2,3 The exact functions of FBGCs, as it relates to interactions with lymphocytes, are still unknown and thus present another avenue for future research.

Dynamic Systems Model

The developed model described in Chapter VI was extremely beneficial for analyzing the lymphocyte and monocyte co-culture and the important cellular processes involved. The discussion in Chapter VI presents numerous hypotheses and unknowns in these interactions that require study and validation. For example, analysis of model parameters and simulations suggest that the fusion rate and/or monocyte differentiation rate could possibly vary with time which could be examined in future studies.

Biomaterials

Additionally, a wide array of biomaterial surfaces will be necessary to allow generalization of biomaterial surface property relationships with the cellular responses

- 226 - Chapter VII

presented in the preceding chapters. We are currently limited by experimental data for

responses to varying degrees of biomaterial surface properties. Future studies should

explore how hydrophobicity/hydrophilicity ranging from extremely hydrophobic to

extremely hydrophilic as well as charge potential ranging from highly positive to highly

negative correlate to biological responses. In addition to surface physical properties,

chemical composition will also need to be examined. Undoubtedly, all of these material characteristics will play a role in determining the biological outcome. The deficiency in

experimental data over time and material surface properties are also limitations in the mechanistic model as it currently stands.

Mechanisms Mediating Biomaterial Surface Effects

Cells such as monocytes, macrophages, and lymphocytes interact not with the synthetic biomaterials but with proteins that adsorb onto the surface. The properties of the biomaterial surface determine the type and conformation of the adsorbed protein layer which ultimately determine the cellular responses (i.e. biological response). Therefore,

the mechanism by which biomaterial surfaces dictate the biological response is through

the adsorbed protein layer. Surface chemistries with similar properties can induce

adsorption of similar proteins which permits the potential correlation of the presented

findings to a variety of biomaterial surfaces and their corresponding responses in vitro

and in vivo. For instance, hydrophilic surfaces have shown the capability to

preferentially adsorb particular proteins (e.g. fibronectin) and retain functionality

compared to hydrophobic surfaces.17 In order to truly achieve the capability of designing

biomaterials to control biological responses, future investigations must examine the

mechanisms mediating biomaterial surface effects, namely how biomaterial surfaces can

- 227 - Chapter VII

influence the specific adsorbed protein type, conformation, and orientation in addition to

how the specific characteristics of the adsorbed protein affects cellular behavior.

It is with understanding of the biological response to biomaterial surfaces as well

as the relationship between these responses and material characteristics that will allow us

control over these responses. The developed model provides a step towards predictability

of biological response based on particular material surface parameters. Ultimately, the

goal is to be able to design a material to meet the needs of the specific application.

References

1. Brodbeck WG, Macewan M, Colton E, Meyerson H, Anderson JM. Lymphocytes and the foreign body response: lymphocyte enhancement of macrophage adhesion and fusion. J Biomed Mater Res A 2005;74(2):222-9.

2. Athanasou NA, Quinn J. Immunophenotypic differences between osteoclasts and macrophage polykaryons: immunohistological distinction and implications for osteoclast ontogeny and function. J Clin Pathol 1990;43(12):997-1003.

3. Bainbridge JA, Revell PA, Al-Saffar N. Costimulatory molecule expression following exposure to orthopaedic implants wear debris. J Biomed Mater Res 2001;54(3):328-34.

4. Itescu S, John R. Interactions between the recipient immune system and the left ventricular assist device surface: immunological and clinical implications. Ann Thorac Surg 2003;75(6 Suppl):S58-65.

5. Schuster M, Kocher A, John R, Hoffman M, Ankersmit J, Lietz K, Edwards N, Oz M, Itescu S. B-cell activation and allosensitization after left ventricular assist device implantation is due to T-cell activation and CD40 ligand expression. Hum Immunol 2002;63(3):211-20.

6. Yoshida M, Mata J, Babensee JE. Effect of poly(lactic-co-glycolic acid) contact on maturation of murine bone marrow-derived dendritic cells. J Biomed Mater Res A 2007;80(1):7-12.

7. Yoshida M, Babensee JE. Differential effects of agarose and poly(lactic-co- glycolic acid) on dendritic cell maturation. J Biomed Mater Res A 2006;79(2):393-408.

- 228 - Chapter VII

8. Babensee JE, Paranjpe A. Differential levels of dendritic cell maturation on different biomaterials used in combination products. J Biomed Mater Res A 2005;74(4):503-10.

9. Yoshida M, Babensee JE. Poly(lactic-co-glycolic acid) enhances maturation of human monocyte-derived dendritic cells. J Biomed Mater Res A 2004;71(1):45- 54.

10. Singh M, Kazzaz J, Ugozzoli M, Chesko J, O'Hagan DT. Charged polylactide co- glycolide microparticles as antigen delivery systems. Expert Opin Biol Ther 2004;4(4):483-91.

11. Nedvetzki S, Sowinski S, Eagle RA, Harris J, Vely F, Pende D, Trowsdale J, Vivier E, Gordon S, Davis DM. Reciprocal regulation of human natural killer cells and macrophages associated with distinct immune synapses. Blood 2007;109(9):3776-85.

12. Poggi A, Prevosto C, Zancolli M, Canevali P, Musso A, Zocchi MR. NKG2D and natural cytotoxicity receptors are involved in natural killer cell interaction with self-antigen presenting cells and stromal cells. Ann N Y Acad Sci 2007;1109:47- 57.

13. McNally AK. Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 1995;147(5):1487-99.

14. DeFife KM, Jenney CR, McNally AK, Colton E, Anderson JM. Interleukin-13 induces human monocyte/macrophage fusion and macrophage mannose receptor expression. J Immunol 1997;158(7):3385-90.

15. Kao WJ, McNally AK, Hiltner A, Anderson JM. Role for interleukin-4 in foreign- body giant cell formation on a poly(etherurethane urea) in vivo. J Biomed Mater Res 1995;29(10):1267-75.

16. Rodriguez A. T Cell Interactions in the Foreign Body Response to Biomaterials. Dissertation. Case Western Reserve University, 2007.

17. Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of biomaterial-cell interactions by adsorbed proteins: a review. Tissue Eng 2005;11(1-2):1-18.

- 229 - Appendix I

% Main program code %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % This program uses the cell mode. % Calls the model output function to solve model differential equations % Calls lsqcurvefit function for model fitting to experimental data % Set up for estimating 7 parameters % David T. Chang % March 2008 %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ clear; close; clear all; clc; clf; close all;

%% Global items %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Global Parameters and Variables % This cell defines the global constants %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ global gamma_diff; global nu_L_adhesion; global gamma_L_diff; global alpha_aL_apop; global delta_L_detach; global beta_L_fusion; global pi_aL_prolif; global pi_naL_prolif; global pi_prolif; global donors; global scale; global data_sets; global adherent_cells_scale; global adherent_FBGCs_scale; global adherent_nuclei_scale; global adherent_FBGC_nuclei_scale; global adherent_lymphs_scale; global tspan width y0 length_y MaxN; global Mplated; global end_N end_Complex1 end_Complex2 end_N_na end_L_a end_L_na ... end_Apoptosed end_Apoptosed_aL end_Proliferated_aL ... end_Proliferated_naL; global end_L end_Proliferated; global t; global Totalcells_Distribution; global Nuclei_Array; global timepoints;

- 230 - Appendix I global Nuclei_Matrix; global Totalnuclei_Distribution; global Mono Mono_Nuc; global Macro1 Macro2 Macro2_Nuc Macros Macros_Nuc; global Mono_Macros Mono_Macros_Nuc; global FBGCs_Distribution FBGCs_Distribution_nuclei; global FBGC FBGC_Nuc; global Surface_Adherent_Lymphs Cell_Associated_Lymphs; global Total_Adherent_Lymphs Lymphocytes; global Total_Cells Total_Nuc; global Percent_Fusion; global Final_Cell_Distribution; global Apoptosed; global ModelOutput_Timeslots; global ModelOutput_Analysis;

%% Constants and inputs %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Constants and inputs for ODEs % This cell specificies the inputs to the code %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ donors=4; % Specifies how many donors in ydata data_sets=6; % Determines how many data sets (i.e. variables) for lsqcurvefit to compare % 4 -- total cell, total fbgc, percent fusion, adherent lymphs % 6 -- 4 + total nuclei and fbgc nuclei biomaterial=5; % Determines which biomaterial data to model fit % 1=PET % 2=BDEDTC % 3=PAAm % 4=PAANa % 5=DMAPAAmMeI plot_model_points=0; % 1 - to plot model points for comparison check % 0 - to not plot model points

% Inputs and set-ups for ode15s or ode45s: t0=0; % Initial timepoint tf=10; % Final timepoint tspan=[t0 tf]; % Defines timespan for using ode15s or ode45s MaxN=30; % Specifies maximum # of nuclei in fused FBGC width=MaxN+1; % Defines width of distribution of cells Mplated=500000; % Number of monocytes plated at day 0 Lplated=1500000; % Number of lymphocytes plated at day 0

%% Experimental Data %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Experimental data % This cell defines the experimental data for model fitting %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

- 231 - Appendix I

tdata=[3 7 10]; % Specifies data time points if biomaterial==1 % PET ydata_adherent_nuclei(1,:)=[524 1648 1169]*71.26; % # adherent nuclei at 3,7,10 days ydata_adherent_nuclei(2,:)=[1021 810 1724]*71.26; ydata_adherent_nuclei(3,:)=[3030 1138 766]*71.26;

ydata_adherent_FBGC_nuclei(1,:)=[1 20 30]*71.26; % # adherent FBGC nuclei at 3,7,10 days ydata_adherent_FBGC_nuclei(2,:)=[7 22 214]*71.26; ydata_adherent_FBGC_nuclei(3,:)=[46 328 145]*71.26;

ydata_adherent_cells(1,:)=[523 1635 1147]*71.26; % # adherent cells at 3,7,10 days ydata_adherent_cells(2,:)=[1016 794 1541]*71.26; ydata_adherent_cells(3,:)=[2995 883 654]*71.26;

ydata_adherent_FBGCs(1,:)=[0 6 8]*71.26; % # adherent FBGCs at 3, 7,10 days ydata_adherent_FBGCs(2,:)=[2 7 31]*71.26; ydata_adherent_FBGCs(3,:)=[10 73 34]*71.26;

ydata_adherent_lymphs(1,:)=[64 227 159]*71.26; % # adherent lymphocytes at 3,7,10 days ydata_adherent_lymphs(2,:)=[55 96 305]*71.26; ydata_adherent_lymphs(3,:)=[473 193 85]*71.26; if donors==4 ydata_adherent_nuclei(4,:)=[2042 1324 1271]*71.26; ydata_adherent_FBGC_nuclei(4,:)=[49 76 261]*71.26; ydata_adherent_cells(4,:)=[2005 1264 1041]*71.26; ydata_adherent_FBGCs(4,:)=[11 16 30]*71.26; end % Scaling of PET data adherent_cells_scale=10^6; adherent_FBGCs_scale=10^4; adherent_nuclei_scale=10^6; adherent_FBGC_nuclei_scale=10^5; adherent_lymphs_scale=10^5; elseif biomaterial==2 % BDEDTC ydata_adherent_nuclei(1,:)=[2073 1865 1458]*71.26; % # adherent nuclei at 3,7,10 days ydata_adherent_nuclei(2,:)=[1493 1587 673]*71.26; ydata_adherent_nuclei(3,:)=[3246 1401 1271]*71.26;

ydata_adherent_FBGC_nuclei(1,:)=[0 9 41]*71.26; % # adherent FBGC nuclei at 3,7,10 days ydata_adherent_FBGC_nuclei(2,:)=[0 4 6]*71.26; ydata_adherent_FBGC_nuclei(3,:)=[15 413 259]*71.26;

ydata_adherent_cells(1,:)=[2073 1859 1425]*71.26; % # adherent cells at 3,7,10 days ydata_adherent_cells(2,:)=[1493 1585 670]*71.26; ydata_adherent_cells(3,:)=[3235 1080 1072]*71.26;

ydata_adherent_FBGCs(1,:)=[0 3 8]*71.26; % # adherent FBGCs at 3, 7,10 days ydata_adherent_FBGCs(2,:)=[0 1 2]*71.26; ydata_adherent_FBGCs(3,:)=[4 93 60]*71.26;

ydata_adherent_lymphs(1,:)=[115 274 190]*71.26; % # adherent lymphocytes at 3,7,10 days ydata_adherent_lymphs(2,:)=[111 154 85]*71.26; ydata_adherent_lymphs(3,:)=[136 353 227]*71.26; if donors==4 ydata_adherent_nuclei(4,:)=[2415 1905 699]*71.26;

- 232 - Appendix I

ydata_adherent_FBGC_nuclei(4,:)=[37 130 97]*71.26; ydata_adherent_cells(4,:)=[2387 1806 623]*71.26; ydata_adherent_FBGCs(4,:)=[9 31 21]*71.26; end % Scaling of BDEDTC data adherent_cells_scale=10^6; adherent_FBGCs_scale=10^4; adherent_nuclei_scale=10^6; adherent_FBGC_nuclei_scale=10^5; adherent_lymphs_scale=10^5; elseif biomaterial==3 % PAAm ydata_adherent_nuclei(1,:)=[495 561 434]*71.26; % # adherent nuclei at 3,7,10 days ydata_adherent_nuclei(2,:)=[316 282 132]*71.26; ydata_adherent_nuclei(3,:)=[938 387 751]*71.26;

ydata_adherent_FBGC_nuclei(1,:)=[116 91 135]*71.26; % # adherent FBGC nuclei at 3,7,10 days ydata_adherent_FBGC_nuclei(2,:)=[5 55 0]*71.26; ydata_adherent_FBGC_nuclei(3,:)=[186 94 287]*71.26;

ydata_adherent_cells(1,:)=[403 492 325]*71.26; % # adherent cells at 3,7,10 days ydata_adherent_cells(2,:)=[313 241 132]*71.26; ydata_adherent_cells(3,:)=[792 308 516]*71.26;

ydata_adherent_FBGCs(1,:)=[24 21 25]*71.26; % # adherent FBGCs at 3, 7,10 days ydata_adherent_FBGCs(2,:)=[1 15 0]*71.26; ydata_adherent_FBGCs(3,:)=[40 15 52]*71.26;

ydata_adherent_lymphs(1,:)=[32 89 75]*71.26; % # adherent lymphocytes at 3,7,10 days ydata_adherent_lymphs(2,:)=[41 32 44]*71.26; ydata_adherent_lymphs(3,:)=[122 52 76]*71.26; if donors==4 ydata_adherent_nuclei(4,:)=[714 464 32]*71.26; ydata_adherent_FBGC_nuclei(4,:)=[97 15 10]*71.26; ydata_adherent_cells(4,:)=[637 453 24]*71.26; ydata_adherent_FBGCs(4,:)=[19 4 2]*71.26; end % Scaling of PAAm data adherent_cells_scale=0.25*10^6; adherent_FBGCs_scale=10^4; adherent_nuclei_scale=0.25*10^6; adherent_FBGC_nuclei_scale=0.75*10^5; adherent_lymphs_scale=0.4*10^5; elseif biomaterial==4 % PAANa ydata_adherent_nuclei(1,:)=[2205 1990 1598]*71.26; % # adherent nuclei at 3,7,10 days ydata_adherent_nuclei(2,:)=[2538 1627 495]*71.26; ydata_adherent_nuclei(3,:)=[2835 1667 926]*71.26;

ydata_adherent_FBGC_nuclei(1,:)=[1140 1215 1082]*71.26; % # adherent FBGC nuclei at 3,7,10 days ydata_adherent_FBGC_nuclei(2,:)=[21 68 37]*71.26; ydata_adherent_FBGC_nuclei(3,:)=[367 869 460]*71.26;

ydata_adherent_cells(1,:)=[1171 903 616]*71.26; % # adherent cells at 3,7,10 days ydata_adherent_cells(2,:)=[2522 1579 469]*71.26; ydata_adherent_cells(3,:)=[2507 867 507]*71.26;

- 233 - Appendix I

ydata_adherent_FBGCs(1,:)=[106 128 99]*71.26; % # adherent FBGCs at 3, 7,10 days ydata_adherent_FBGCs(2,:)=[6 20 11]*71.26; ydata_adherent_FBGCs(3,:)=[39 69 41]*71.26;

ydata_adherent_lymphs(1,:)=[79 143 123]*71.26; % # adherent lymphocytes at 3,7,10 days ydata_adherent_lymphs(2,:)=[96 138 63]*71.26; ydata_adherent_lymphs(3,:)=[167 149 85]*71.26; if donors==4 ydata_adherent_nuclei(4,:)=[2550 2393 1606]*71.26; ydata_adherent_FBGC_nuclei(4,:)=[1915 1994 1305]*71.26; ydata_adherent_cells(4,:)=[783 487 367]*71.26; ydata_adherent_FBGCs(4,:)=[148 89 66]*71.26; end % Scaling of PAANa data adherent_cells_scale=0.25*10^6; adherent_FBGCs_scale=10^4; adherent_nuclei_scale=0.3*10^6; adherent_FBGC_nuclei_scale=10^5; adherent_lymphs_scale=0.3*10^5; elseif biomaterial==5 % DMAPAAmMeI ydata_adherent_nuclei(1,:)=[1007 1528 844]*71.26; % # adherent nuclei at 3,7,10 days ydata_adherent_nuclei(2,:)=[673 1215 658]*71.26; ydata_adherent_nuclei(3,:)=[1576 2440 1401]*71.26;

ydata_adherent_FBGC_nuclei(1,:)=[10 42 41]*71.26; % # adherent FBGC nuclei at 3,7,10 days ydata_adherent_FBGC_nuclei(2,:)=[2 98 6]*71.26; ydata_adherent_FBGC_nuclei(3,:)=[10 101 299]*71.26;

ydata_adherent_cells(1,:)=[1000 1493 810]*71.26; % # adherent cells at 3,7,10 days ydata_adherent_cells(2,:)=[671 1145 654]*71.26; ydata_adherent_cells(3,:)=[1569 2356 1152]*71.26;

ydata_adherent_FBGCs(1,:)=[3 8 8]*71.26; % # adherent FBGCs at 3, 7,10 days ydata_adherent_FBGCs(2,:)=[0 28 2]*71.26; ydata_adherent_FBGCs(3,:)=[3 17 49]*71.26;

ydata_adherent_lymphs(1,:)=[363 314 217]*71.26; % # adherent lymphocytes at 3,7,10 days ydata_adherent_lymphs(2,:)=[109 164 111]*71.26; ydata_adherent_lymphs(3,:)=[660 935 473]*71.26; if donors==4 ydata_adherent_nuclei(4,:)=[1919 1639 1084]*71.26; ydata_adherent_FBGC_nuclei(4,:)=[3 85 77]*71.26; ydata_adherent_cells(4,:)=[1917 1574 1023]*71.26; ydata_adherent_FBGCs(4,:)=[1 21 17]*71.26; end % Scaling of DMAPAAmMeI data adherent_cells_scale=1.1*10^6; adherent_FBGCs_scale=10^4; adherent_nuclei_scale=1.1*10^6; adherent_FBGC_nuclei_scale=10^5; adherent_lymphs_scale=1.75*10^5; end ydata_percent_fusion=ydata_adherent_FBGC_nuclei./ydata_adherent_nuclei;

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Scale data to similar magnitudes

- 234 - Appendix I

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ydata_adherent_cells_S=ydata_adherent_cells./adherent_cells_scale; ydata_adherent_FBGCs_S=ydata_adherent_FBGCs./adherent_FBGCs_scale; ydata_adherent_nuclei_S=ydata_adherent_nuclei./adherent_nuclei_scale; ydata_adherent_FBGC_nuclei_S=ydata_adherent_FBGC_nuclei./adherent_FBGC_nuclei_scale; ydata_adherent_lymphs_S=ydata_adherent_lymphs./adherent_lymphs_scale; ydata_percent_fusion_S=ydata_percent_fusion; if data_sets==3 ydata=[ydata_adherent_cells; ydata_adherent_FBGCs; ydata_adherent_lymphs]; % Scaled ydata for model fit ydata_S=[ydata_adherent_cells_S; ydata_adherent_FBGCs_S; ydata_adherent_lymphs_S]; elseif data_sets==4 ydata=[ydata_adherent_cells; ydata_adherent_FBGCs; ydata_percent_fusion; ydata_adherent_lymphs]; % Scaled ydata for model fit ydata_S=[ydata_adherent_cells_S; ydata_adherent_FBGCs_S; ydata_percent_fusion_S; ydata_adherent_lymphs_S]; elseif data_sets==6 ydata=[ydata_adherent_cells; ydata_adherent_FBGCs; ydata_adherent_nuclei; ydata_adherent_FBGC_nuclei; ydata_percent_fusion; ydata_adherent_lymphs]; % Scaled ydata for model fit ydata_S=[ydata_adherent_cells_S; ydata_adherent_FBGCs_S; ydata_adherent_nuclei_S;ydata_adherent_FBGC_nuclei_S; ydata_percent_fusion_S; ydata_adherent_lymphs_S]; end

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Calculate data averages %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Mean_nuclei=mean(ydata_adherent_nuclei); Mean_FBGC_nuclei=mean(ydata_adherent_FBGC_nuclei); Mean_cells=mean(ydata_adherent_cells); Mean_FBGCs=mean(ydata_adherent_FBGCs); Mean_lymphs=mean(ydata_adherent_lymphs); Mean_fusion=mean(ydata_percent_fusion);

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Calculate standard deviation of data %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ STD_nuclei=std(ydata_adherent_nuclei); STD_FBGC_nuclei=std(ydata_adherent_FBGC_nuclei); STD_cells=std(ydata_adherent_cells); STD_FBGCs=std(ydata_adherent_FBGCs); STD_lymphs=std(ydata_adherent_lymphs); STD_fusion=std(ydata_percent_fusion);

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Calculate standard error of the mean %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [m,n]=size(ydata_adherent_nuclei); SEM_nuclei=STD_nuclei/sqrt(m);

[m,n]=size(ydata_adherent_FBGC_nuclei); SEM_FBGC_nuclei=STD_FBGC_nuclei/sqrt(m);

- 235 - Appendix I

[m,n]=size(ydata_adherent_cells); SEM_cells=STD_cells/sqrt(m);

[m,n]=size(ydata_adherent_FBGCs); SEM_FBGCs=STD_FBGCs/sqrt(m);

[m,n]=size(ydata_adherent_lymphs); SEM_lymphs=STD_lymphs/sqrt(m);

[m,n]=size(ydata_percent_fusion); SEM_fusion=STD_fusion/sqrt(m);

%% Guesstimates and Evaluation of parameters %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % This cell allows manual manipulation of parameters to determine model % output and plots model output with all experimental data %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Rate constants involving only monocytes in culture %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ nu_adhesion=0.75; % Monocyte adhesion rate constant gamma_diff=0.75; % Mono differentiation into macs rate constant alpha_mono_apop=0.1; % Monocyte apoptosis rate constant alpha_macro_apop=0.1; % Macrophage apoptosis rate constant delta_detach=0.05; % Cell detachment rate constant beta_fusion=0.85*10^-6; % Fusion rate constant

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Rate constants involving Lymphocyte and monocyte interactions %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

nu_L_adhesion=0.000035; % Lymphocyte adhesion rate constant gamma_L_diff=0.75; % mono diff into macs due to Lymph rate constant alpha_aL_apop=0.00008; % Adherent lymphocyte apoptosis rate constant delta_L_detach=0.925; % Lymphocyte detachment rate constant beta_L_fusion=beta_fusion; % Fusion due to lymphocyte rate constant pi_aL_prolif=0.0000005; % Adherent lymph proliferation rate constant pi_naL_prolif=0.0000005; % Nonadherent lymph proliferation rate constant kappa1=0.5*10^-1; % unbinding of complexes rate constant kappa_1=0.8*10^-8; % binding to form complexes rate constant

% Combine parameters into 1 array for input into model output function P0(1,1)=nu_adhesion; P0(1,2)=alpha_mono_apop; P0(1,3)=alpha_macro_apop; P0(1,4)=delta_detach; P0(1,5)=beta_fusion; P0(1,6)=kappa1; P0(1,7)=kappa_1; % Scale 7 parameters scale=[1 1 1 1 1000000 10 100000000]; P0_S=P0.*scale; % Scaled P0

- 236 - Appendix I

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Set-up to solve differential equations %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

% Initializing all the variables N0=zeros(1,width); % Initial values for each cell size N0(1,1)=nu_adhesion*Mplated; % Defines initial # of adherent cells with 1 nuclei % Array of length width

Complex1_0=zeros(1,width); % Initial # of 1 lymphocyte to monos, macs, FBGCs complexes Complex2_0=zeros(1,width); % Initial # of 2 lymphocyte to FBGC complexes % 2 arrays of length width % First 3 spots of complex2 will never change because only % FBGCs can have more than 1 lymphocyte interaction

N_na0=zeros(1,2); % Initial values for nonadherent monocytes and macrophages N_na0(1,1)=Mplated-N0(1,1); % Initial # of nonadherent monocytes % Array of length 2

L_a0=nu_L_adhesion*Lplated; % Initial # of adherent lymphocytes L_na0=Lplated-L_a0; % Initial # of nonadherent lymphocytes % 2 arrays of length 2 total

Apoptosed0=zeros(1,2); % Initial # of apoptosed adherent 1 nuclei monocytes and macrophages Apoptosed_aL_0=0; % Initial # of apoptosed adherent lymphocytes % 2 arrays of length 3 total

Proliferated_aL_0=0; % Initial # of proliferated adherent lymphocytes Proliferated_naL_0=0; % Initial # of proliferated nonadherent lymphocytes % 2 arrays of length 2 total

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Combine all initial values into one array %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

y0=[N0 Complex1_0 Complex2_0 N_na0 L_a0 L_na0 Apoptosed0 Apoptosed_aL_0 Proliferated_aL_0 Proliferated_naL_0];

% End of each piece in complete array end_N=length(N0); end_Complex1=end_N+length(Complex1_0); end_Complex2=end_Complex1+length(Complex2_0); end_N_na=end_Complex2+length(N_na0); end_L_a=end_N_na+length(L_a0); end_L_na=end_L_a+length(L_na0); end_Apoptosed=end_L_na+length(Apoptosed0); end_Apoptosed_aL=end_Apoptosed+length(Apoptosed_aL_0); end_Proliferated_aL=end_Apoptosed_aL+length(Proliferated_aL_0); end_Proliferated_naL=end_Proliferated_aL+length(Proliferated_naL_0);

length_y=length(y0); % Determines length of complete initial array

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Calls modeloutput function %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

- 237 - Appendix I

ModelOutput=difeqsolver_modeloutput_7P2(P0_S,tdata);

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Plots Cell types over time, Nuclei counts over time, Percent Fusion over time, and Cell Nuclei % Distribution at Final Time % Plots model output with all experimental data for comparison %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ for i=2:donors tdata(i,:)=[tdata(1,:)]; end figure subplot(2,1,1); plot(t,Total_Cells(:,1), 'k-',t,FBGC(:,1), 'g--',t, Mono(:,1), 'b--', t, Macro1(:,1), 'b-.', t, Macro2, 'b:', t, Mono_Macros(:,1), 'r--'); hold; plot(tdata,ydata_adherent_cells,'ks',tdata,ydata_adherent_FBGCs,'g^'); legend('Total Cells','FBGCs','Monocytes','Macrophages (1 Nucleus)','Macrophages (2 Nuclei)','Cells (<3 Nuclei)'); xlabel('Time(Days)'); ylabel('# of Adherent Cells'); title('Adherent Cell Population'); subplot(2,1,2); plot(t,Total_Nuc(:,1),'k-',t,FBGC_Nuc(:,1),'g--',t,Mono(:,1),'b--',t,Macro1(:,1),'b- .',t,Macro2_Nuc(:,1),'b:',t,Mono_Macros_Nuc(:,1),'r--'); hold; plot(tdata,ydata_adherent_nuclei,'ks',tdata,ydata_adherent_FBGC_nuclei,'g^'); legend('Total Nuclei','Nuclei within FBGCs','Nuclei in Monocytes','Nuclei in Macrophages w/ 1 Nucleus','Nuclei in Macrophages w/ 2 Nuclei','Nuclei in Monocytes & Macrophages'); xlabel('Time(Days)'); ylabel('# of Adherent Nuclei'); title('Adherent Nuclei over Time');

figure; subplot(2,1,1); plot(t,Percent_Fusion); hold; plot(tdata,ydata_percent_fusion,'ks'); xlabel('Time(Days)'); ylabel('Percent fusion'); title('Percent Fusion');

subplot(2,1,2); sizes=1:MaxN; plot(sizes,Final_Cell_Distribution(:),'x'); xlabel('Number of Nuclei per Cell'); ylabel('# of Cells'); title('Cell Size Distribution at Final Time');

figure; subplot(2,2,1); plot(t,Surface_Adherent_Lymphs); xlabel('Time(Days)'); ylabel('Surface Adherent Lymphocytes');

- 238 - Appendix I title('Lymphocyte Surface Adhesion over Time'); subplot(2,2,2); plot(t,Cell_Associated_Lymphs); xlabel('Time(Days)'); ylabel('Cell-associated Lymphocytes'); title('Cell-associated Lymphocytes over Time'); subplot(2,2,3); plot(t,Total_Adherent_Lymphs); hold; plot(tdata(1:3,:),ydata_adherent_lymphs(1:3,:),'ks'); xlabel('Time(Days)'); ylabel('Total Adherent Lymphocytes'); title('Total Lymphocyte on Surface over Time');

%% Plotting Model Simulations %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Plotting figures of representative model outputs (no experimental data) % Utilized in Dissertation Chapter %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ figure % Plot adherent cell populations subplot(2,1,1); plot(t,Total_Cells(:,1),'k-',t,FBGC(:,1),'k--',t,Mono(:,1),'k-.',t,Macros(:,1),'k:'); hold; legend('Total Cells','FBGCs','Monocytes','Macrophages'); xlabel('Time (Days)'); ylabel('Biomaterial-Adherent Cells');

% Plot adherent nuclei subplot(2,1,2); plot(t,Total_Nuc(:,1),'k-',t,FBGC_Nuc(:,1),'k--',t,Mono(:,1),'k-.',t,Macros_Nuc(:,1),'k:'); hold; legend('Total Nuclei','FBGC Nuclei','Monocyte Nuclei','Macrophage Nuclei'); xlabel('Time (Days)'); ylabel('Biomaterial-Adherent Nuclei'); figure; % Plot percentage fusion subplot(2,3,1); plot(t,Percent_Fusion,'k-'); hold; xlabel('Time (Days)'); ylabel('Percent fusion');

% Plot apoptosed monocytes and macrophages subplot(2,3,2); plot(t,Apoptosed(:,1),'k-',t,Apoptosed(:,2),'k:'); hold; legend('Monocytes','Macrophages'); xlabel('Time (Days)'); ylabel('Apoptosed Cells');

- 239 - Appendix I

% Plots cell nuclei distribution subplot(2,3,3); sizes=1:MaxN; bar(sizes,Final_Cell_Distribution(:),'k'); hold; xlabel('Number of Nuclei per Cell'); ylabel('Number of Cells');

% Plots surface-adherent lymphocytes subplot(2,3,4); plot(t,Surface_Adherent_Lymphs,'k-'); xlabel('Time (Days)'); ylabel('Biomaterial-Adherent Lymphocytes'); subplot(2,3,5); % Plots cell-adherent lymphocytes plot(t,Cell_Associated_Lymphs,'k-'); xlabel('Time (Days)'); ylabel('Cell-Adherent Lymphocytes'); subplot(2,3,6); % Plots total adherent lymphocytes plot(t,Total_Adherent_Lymphs,'k-'); hold; xlabel('Time (Days)'); ylabel('Total Lymphocytes');

%% Parameter estimation with lsqcurvefit %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Run cells: Global Parameters, Constants and inputs, Experimental data % and this parameter estimation with lsqcurvefit % % This cell will estimate parameters by comparing model output with % experimental data %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Establish parameters and scale them for lsqcurvefit %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

% Input unscaled P0 nu_adhesion=0.35; % Monocyte adhesion rate constant alpha_mono_apop=0.05; % Monocyte apoptosis rate constant alpha_macro_apop=0.05; % Macrophage apoptosis rate constant delta_detach=0.1; % Cell detachment rate constant beta_fusion=0.1*10^-6; % Fusion rate constant kappa1=0.5*10^-1; % Unbinding of complexes rate constant kappa_1=0.5*10^-8; % Binding to form complexes rate constant

% Combine all parameters into one array for lsqcurvefit % 7 parameters P0(1,1)=nu_adhesion; P0(1,2)=alpha_mono_apop; P0(1,3)=alpha_macro_apop; P0(1,4)=delta_detach;

- 240 - Appendix I

P0(1,5)=beta_fusion; P0(1,6)=kappa1; P0(1,7)=kappa_1; % Scale 7 parameters scale=[1 1 1 1 1000000 10 100000000]; P0_S=P0.*scale; % Scaled P0

% Fixed Rate constants involving only monocyte culture gamma_diff=0.75; % Mono differentiation into macs rate constant

% Fixed Rate constants involving Lymphocyte and monocyte interactions nu_L_adhesion=0.000035; % Lymphocyte adhesion rate constant gamma_L_diff=0.75; % Mono diff into macs due to Lymph rate constant alpha_aL_apop=0.00008; % Adherent lymphocyte apoptosis rate constant delta_L_detach=.925; % Lymphocyte detachment rate constant beta_L_fusion=beta_fusion; % Fusion due to lymphocyte rate constant pi_aL_prolif=0.0000005; % Adherent lymph proliferation rate constant pi_naL_prolif=0.0000005; % Nonadherent lymph proliferation rate constant

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Upper and lower bounds for lsqcurvefit %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ lb=0.*scale; ub=1.*scale;

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Set-up to solve differential equations %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

% Initializing all the variables N0=zeros(1,width); % Initial values for each cell size N0(1,1)=nu_adhesion*Mplated; % Defines initial # of adherent cells with 1 nuclei % Array of length width

Complex1_0=zeros(1,width); % Initial # of 1 lymphocyte to monos,macs, FBGCs complexes Complex2_0=zeros(1,width); % Initial # of 2 lymphocyte to FBGC complexes % 2 arrays of length width % First 3 spots of complex2 will never change because only % FBGCs can have more than 1 lymphocyte interaction

N_na0=zeros(1,2); % Initial values for nonadherent monocytes and macrophages N_na0(1,1)=Mplated-N0(1,1); % Initial # of nonadherent monocytes % Array of length 2

L_a0=nu_L_adhesion*Lplated; % Initial # of adherent lymphocytes L_na0=Lplated-L_a0; % Initial # of nonadherent lymphocytes % 2 arrays of length 2 total

Apoptosed0=zeros(1,2); % Initial # of apoptosed adherent 1 nuclei monocytes and macrophages Apoptosed_aL_0=0; % Initial # of apoptosed adherent lymphocytes % 2 arrays of length 3 total

Proliferated_aL_0=0; % Initial # of proliferated adherent lymphocytes Proliferated_naL_0=0; % Initial # of proliferated nonadherent lymphocytes % 2 arrays of length 2 total

- 241 - Appendix I

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Combine all initial values into one array %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

y0=[N0 Complex1_0 Complex2_0 N_na0 L_a0 L_na0 Apoptosed0 Apoptosed_aL_0 Proliferated_aL_0 Proliferated_naL_0];

length_y=length(y0); % Determines length of complete initial array

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Run lsqcurvefit %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ options=optimset('Display','iter','Tolfun',1e-8,'MaxFunEvals',3000,'MaxIter',2000); [Pfinal_S,resnorm,residual,exitflag,output,lambda,jacobian]=lsqcurvefit(@difeqsolver_modeloutput_7P2, P0_S,tdata,ydata_S,lb,ub,options);

% Outputs after optimization procedure for analysis Pfinal_S P0_S scale Pfinal=Pfinal_S./scale;

% Plot adherent cells and nuclei subplot(2,1,1); % Model output plot(t,Total_Cells(:,1), 'k-', t, FBGC(:,1), 'g--', t, Mono(:,1), 'b--', t, Macro1(:,1), 'b-.', t, Macro2, 'b:', t, Mono_Macros(:,1), 'r--'); hold on; % Experimental data plot(tdata,ydata_adherent_cells(:,:),'ks',tdata,ydata_adherent_FBGCs(:,:),'g^'); if plot_model_points==1 plot(ModelOutput_Timeslots(1,:), ModelOutput_Analysis(1,:),'ko',ModelOutput_Timeslots(1,:), ModelOutput_Analysis(4,:),'rs');

- 242 - Appendix I end legend('Total Cells','FBGCs','Monocytes','Macrophages (1 Nucleus)','Macrophages (2 Nuclei)','Cells (<3 Nuclei)'); xlabel('Time(Days)'); ylabel('# of Adherent Cells'); title('Adherent Cell Population'); hold off; subplot(2,1,2); % Model output plot(t, Total_Nuc(:,1), 'k-', t, FBGC_Nuc(:,1), 'g--', t, Mono(:,1), 'b--', t, Macro1(:,1), 'b-.', t, Macro2_Nuc(:,1), 'b:', t, Mono_Macros_Nuc(:,1), 'r--'); hold on; % Experimental data plot(tdata,ydata_adherent_nuclei(:,:),'ks',tdata,ydata_adherent_FBGC_nuclei(:,:),'g^'); if plot_model_points==1 && data_sets==6 plot(ModelOutput_Timeslots(1,:), ModelOutput_Analysis(7,:),'ko',ModelOutput_Timeslots(1,:), ModelOutput_Analysis(10,:),'rs'); end legend('Total Nuclei','Nuclei within FBGCs','Nuclei in Monocytes','Nuclei in Macrophages w/ 1 Nucleus','Nuclei in Macrophages w/ 2 Nuclei','Nuclei in Monocytes & Macrophages'); xlabel('Time(Days)'); ylabel('# of Adherent Nuclei'); title('Adherent Nuclei over Time'); hold off;

figure; % Plot percent fusion and nuclei distribution subplot(2,1,1); % Model output plot(t,Percent_Fusion); hold on; % Experimental data plot(tdata,ydata_percent_fusion(:,:),'ks'); if plot_model_points==1 if data_sets==4 plot(ModelOutput_Timeslots(1,:), ModelOutput_Analysis(7,:),'ko'); elseif data_sets==6 plot(ModelOutput_Timeslots(1,:), ModelOutput_Analysis(13,:),'ko'); end end xlabel('Time(Days)'); ylabel('Percent fusion'); title('Percent Fusion'); hold off;

subplot(2,1,2); sizes=1:MaxN; % Model output plot(sizes,Final_Cell_Distribution(:),'x'); xlabel('Number of Nuclei per Cell'); ylabel('# of Cells'); title('Cell Size Distribution at Final Time');

figure; % Plot surface-adherent lymphocytes (model output)

- 243 - Appendix I subplot(2,2,1); plot(t,Surface_Adherent_Lymphs); xlabel('Time(Days)'); ylabel('Surface Adherent Lymphocytes'); title('Lymphocyte Surface Adhesion over Time'); subplot(2,2,2); % Plot cell-adherent lymphocytes (model output) plot(t,Cell_Associated_Lymphs); xlabel('Time(Days)'); ylabel('Cell-associated Lymphocytes'); title('Cell-associated Lymphocytes over Time'); subplot(2,2,3); % Plot total adherent lymphocytes % Model output plot(t,Total_Adherent_Lymphs); hold on; % Experimental data plot(tdata(1:3,:),ydata_adherent_lymphs(:,:),'ks'); if plot_model_points==1 if data_sets==4 plot(ModelOutput_Timeslots(1,:), ModelOutput_Analysis(10,:),'ko'); elseif data_sets==6 plot(ModelOutput_Timeslots(1,:), ModelOutput_Analysis(16,:),'ko'); end end xlabel('Time(Days)'); ylabel('Total Adherent Lymphocytes'); title('Total Lymphocyte on Surface over Time'); hold off;

% Plot residuals figure; hold on for i=1:donors:size(ydata) plot(tdata(1,:),residual(i,:),'k--',tdata(2,:),residual(i+1,:),'g-',tdata(3,:),residual(i+2,:),'r:'); end xlabel('Time (days)'); ylabel('Residual'); title('Residuals'); hold off

%% Plot only pertinent model and data comparisons %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % These figures were used in the dissertation chapter %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ tdata=ModelOutput_Timeslots; figure

% Plot total cells and FBGCs subplot(2,2,1); plot(t,Total_Cells(:,1),'k-',t,FBGC(:,1),'k:');

- 244 - Appendix I hold on; errorbar(tdata(1,:),Mean_cells,SEM_cells,'ko','MarkerSize',3,'MarkerFaceColor','k'); errorbar(tdata(1,:),Mean_FBGCs,SEM_FBGCs,'kd','MarkerSize',3,'MarkerFaceColor','k'); legend('Model - Total Cells','Model - FBGCs','Experimental - Total Cells','Experimental - FBGCs'); xlabel('Time (Days)'); ylabel('Biomaterial-Adherent Cells'); hold off;

% Plot total nuclei and FBGC nuclei subplot(2,2,2); plot(t,Total_Nuc(:,1),'k-',t,FBGC_Nuc(:,1),'k:'); hold on; errorbar(tdata(1,:),Mean_nuclei,SEM_nuclei,'ko','MarkerSize',3,'MarkerFaceColor','k'); errorbar(tdata(1,:),Mean_FBGC_nuclei,SEM_FBGC_nuclei,'kd','MarkerSize',3,'MarkerFaceColor','k'); legend('Model - Total Nuclei','Model - FBGC Nuclei','Experimental - Total Nuclei','Experimental - FBGC Nuclei'); xlabel('Time (Days)'); ylabel('Biomaterial-Adherent Nuclei'); hold off;

% Plot percent fusion over time subplot(2,2,3); plot(t,Percent_Fusion,'k-'); hold on; errorbar(tdata(1,:),Mean_fusion,SEM_fusion,'ks','MarkerSize',4,'MarkerFaceColor','k'); xlabel('Time (Days)'); ylabel('Percent fusion'); hold off;

% Plot total (surface and cell) adherent lymphocytes subplot(2,2,4); plot(t,Total_Adherent_Lymphs,'k-'); hold on; errorbar(tdata(1,:),Mean_lymphs, SEM_lymphs,'ks','MarkerSize',4,'MarkerFaceColor','k'); xlabel('Time (Days)'); ylabel('Total Adherent Lymphocytes'); hold off;

- 245 - Appendix II function ModelOutput=difeqsolver_modeloutput_7P2(P,tdata)

% Model output function called by main code % Determines output utilizing ode15s and differential equations function % Solves model differential equations with ODE15s % David T. Chang % March 2008

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Global Variables %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ global gamma_diff; global nu_L_adhesion; global gamma_L_diff; global alpha_aL_apop; global delta_aL_detach; global beta_L_fusion; global pi_aL_prolif; global pi_naL_prolif; global pi_prolif; global donors; global scale; global data_sets; global adherent_cells_scale; global adherent_FBGCs_scale; global adherent_nuclei_scale; global adherent_FBGC_nuclei_scale; global adherent_lymphs_scale; global tspan width y0 length_y MaxN; global Mplated; global end_N end_Complex1 end_Complex2 end_N_na end_L_a end_L_na ... end_Apoptosed end_Apoptosed_aL end_Proliferated_aL ... end_Proliferated_naL; global end_L end_Proliferated; global t; global Totalcells_Distribution; global Nuclei_Array; global timepoints; global Nuclei_Matrix; global Totalnuclei_Distribution; global Mono Mono_Nuc; global Macro1 Macro2 Macro2_Nuc Macros Macros_Nuc; global Mono_Macros Mono_Macros_Nuc; global FBGCs_Distribution FBGCs_Distribution_nuclei; global FBGC FBGC_Nuc; global Surface_Adherent_Lymphs Cell_Associated_Lymphs; global Total_Adherent_Lymphs Lymphocytes; global Total_Cells Total_Nuc; global Percent_Fusion; global Final_Cell_Distribution;

- 246 - Appendix II global Apoptosed; global ModelOutput_Timeslots; global ModelOutput_Analysis;

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Scale back parameters %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

P=P./scale;

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Calls ODE15s to solve model differential equations %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ nu_adhesion=P(1,1); y0(1,1)=nu_adhesion*Mplated; [t,y]=ode15s(@difeq_fxn_direct_7P2,tspan,y0,[],P);

%Splitting up the combined array N=y(:,1:end_N); Complex1=y(:,end_N+1:end_Complex1); Complex2=y(:,end_Complex1+1:end_Complex2); N_na=y(:,end_Complex2+1:end_N_na); L_a=y(:,end_N_na+1:end_L_a); L_na=y(:,end_L_a+1:end_L_na); Apoptosed=y(:,end_L_na+1:end_Apoptosed); Apoptosed_aL=y(:,end_Apoptosed+1:end_Apoptosed_aL); Proliferated_aL=y(:,end_Apoptosed_aL+1:end_Proliferated_aL); Proliferated_naL=y(:,end_Proliferated_aL+1:end_Proliferated_naL);

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Defines cell types from returned matrix and percent fusion over time %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Totalcells_Distribution=N+Complex1+Complex2; % Determination of distribution of total adherent cell nuclei of each type (mono, macs, FBGCs)

Nuclei_Array=[1 1 2:MaxN]; % Array of how many nuclei in each cell type timepoints=length(t); Nuclei_Matrix=repmat(Nuclei_Array,timepoints,1);

Totalnuclei_Distribution=Totalcells_Distribution.*Nuclei_Matrix;

Mono=Totalcells_Distribution(:,1); % Determines monocyte population Mono_Nuc=Totalnuclei_Distribution(:,1); % Determines # of monocyte nuclei

Macro1=Totalcells_Distribution(:,2); % Determines # of macrophages with 1 nuclei Macro2=Totalcells_Distribution(:,3); % Determines # of macrophages with 2 nuclei Macro2_Nuc=Totalnuclei_Distribution(:,3); % Determines # of nuclei in 2 nuclei macrophages

Macros=Macro1+Macro2; % Determines total # of macrophages Macros_Nuc=Macro1+Macro2_Nuc; % Determines total nuclei # in macrophages

- 247 - Appendix II

Mono_Macros=Mono+Macros; % Determines total # of cells with <3 nuclei Mono_Macros_Nuc=Mono+Macros_Nuc; % Determines total # of nuclei in cells with <3 nuclei

FBGCs_Distribution=Totalcells_Distribution(:,4:MaxN+1); % Determines # of cells with >=3 nuclei (FBGCs) FBGCs_Distribution_nuclei=Totalnuclei_Distribution(:,4:MaxN+1); % Determines # of nuclei in all FBGCs

FBGC=sum(FBGCs_Distribution')'; % Determines the total # of FBGCs FBGC_Nuc=sum(FBGCs_Distribution_nuclei')'; % Determines the total # of FBGC nuclei

Surface_Adherent_Lymphs=L_a; % Determines surface adherent lymphocytes Cell_Associated_Lymphs=sum(Complex1')'+sum(Complex2')'; % Determines cell-associated lymphocytes Total_Adherent_Lymphs=Surface_Adherent_Lymphs+Cell_Associated_Lymphs; % Determines total adherent lymphocytes

Total_Cells=Mono_Macros+FBGC; % Defines total number of adherent cells Total_Nuc=Mono_Macros_Nuc+FBGC_Nuc; % Defines total number of adherent nuclei

Percent_Fusion=FBGC_Nuc./Total_Nuc; % Defines percent fusion as a function of time

Final_Cell_Distribution=[Mono(end)+Macro1(end) Totalcells_Distribution(end,3:end)]; % Defines the final time distribution combining all 1 nuclei cells

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Defines Model Output %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

% Determine timeslots for day 3 and day 7 count_d3=0; count_d7=0; temp_D3=0; temp_D7=0; for tt=1:length(t) if t(tt)>=3 && t(tt)<=3.5 count_d3=count_d3+1; temp_D3(count_d3)=tt; end if t(tt)>=7 && t(tt)<=7.5 count_d7=count_d7+1; temp_D7(count_d7)=tt; end end timeslot_D3=temp_D3(1); timeslot_D7=temp_D7(1); timeslots=length(t);

for i=1:donors Total_Cells_model(i,:)=[Total_Cells(timeslot_D3,1) Total_Cells(timeslot_D7,1) Total_Cells(timeslots,1)]; FBGC_model(i,:)=[FBGC(timeslot_D3,1) FBGC(timeslot_D7,1) FBGC(timeslots,1)]; Total_Nuc_model(i,:)=[Total_Nuc(timeslot_D3,1) Total_Nuc(timeslot_D7,1) Total_Nuc(timeslots,1)];

- 248 - Appendix II

FBGC_Nuc_model(i,:)=[FBGC_Nuc(timeslot_D3,1) FBGC_Nuc(timeslot_D7,1) FBGC_Nuc(timeslots,1)]; Percent_Fusion_model(i,:)=[Percent_Fusion(timeslot_D3,1) Percent_Fusion(timeslot_D7,1) Percent_Fusion(timeslots,1)]; Total_Adherent_Lymphs_model(i,:)=[Total_Adherent_Lymphs(timeslot_D3,1) Total_Adherent_Lymphs(timeslot_D7,1) Total_Adherent_Lymphs(timeslots,1)]; ModelOutput_Timeslots(i,:)=[t(timeslot_D3) t(timeslot_D7) t(timeslots)]; end

% Scaling of model outputs for comparison with scaled experimental data Total_Cells_model_S=Total_Cells_model./adherent_cells_scale; FBGC_model_S=FBGC_model./adherent_FBGCs_scale; Total_Nuc_model_S=Total_Nuc_model./adherent_nuclei_scale; FBGC_Nuc_model_S=FBGC_Nuc_model./adherent_FBGC_nuclei_scale; Total_Adherent_Lymphs_model_S=Total_Adherent_Lymphs_model./adherent_lymphs_scale; Percent_Fusion_model_S=Percent_Fusion_model;

% Model output for comparison of 3 variables if data_sets==3 ModelOutput=[Total_Cells_model; FBGC_model; Total_Adherent_Lymphs_model(1:3,:)]; ModelOutput_Analysis=ModelOutput; % Scaled modeloutputs for comparison with scaled ydata ModelOutput=[Total_Cells_model_S; FBGC_model_S; Total_Adherent_Lymphs_model_S(1:3,:)];

% Model output for comparison of 4 variables elseif data_sets==4 ModelOutput=[Total_Cells_model; FBGC_model; Percent_Fusion_model; Total_Adherent_Lymphs_model(1:3,:)]; ModelOutput_Analysis=ModelOutput; % Scaled modeloutputs for comparison with scaled ydata ModelOutput=[Total_Cells_model_S; FBGC_model_S; Percent_Fusion_model_S; Total_Adherent_Lymphs_model_S(1:3,:)];

% Model output for comparison of 6 variables elseif data_sets==6 ModelOutput=[Total_Cells_model; FBGC_model; Total_Nuc_model; FBGC_Nuc_model; Percent_Fusion_model; Total_Adherent_Lymphs_model(1:3,:)]; ModelOutput_Analysis=ModelOutput; % Scaled modeloutputs for comparison with scaled ydata ModelOutput=[Total_Cells_model_S; FBGC_model_S; Total_Nuc_model_S; FBGC_Nuc_model_S; Percent_Fusion_model_S; Total_Adherent_Lymphs_model_S(1:3,:)]; end

- 249 - Appendix III function dy=difeq_fxn_direct_7P2(t,y,P)

% This function defines the model differential equations that will be solved by ODE15s. % David T. Chang % March 2008

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Global Constants %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ global tspan width length_y MaxN; global end_N end_Complex1 end_Complex2 end_N_na end_L_a end_L_na ... end_Apoptosed end_Apoptosed_aL end_Proliferated_aL ... end_Proliferated_naL; global end_L end_Proliferated; global gamma_diff; global nu_L_adhesion; global gamma_L_diff; global alpha_aL_apop; global delta_L_detach; global beta_L_fusion; global pi_aL_prolif; global pi_naL_prolif; global pi_prolif;

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Separate Parameter Array %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ nu_adhesion=P(1,1); alpha_mono_apop=P(1,2); alpha_macro_apop=P(1,3); delta_detach=P(1,4); beta_fusion=P(1,5); kappa1=P(1,6); kappa_1=P(1,7);

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy=zeros(length_y,1); % Defines a matrix, dy, filled with zeros

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations for single monocyte to FBGC progression (1 to MaxN+1) %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy(1)=-alpha_mono_apop*y(1)-gamma_diff*y(1)-delta_detach*y(1) -kappa_1*y(end_L_a+1)*y(1)+kappa1*y(end_N+1); % Rate of change of monocytes dy(2)=gamma_diff*y(1)-alpha_macro_apop*y(2)-delta_detach*y(2) -kappa_1*y(end_L_a+1)*y(2)+kappa1*y(end_N+2); % Rate of change of 1 nuclei macrophages for v=2:3 dy(2)=dy(2)-beta_fusion*y(2)*y(v)-beta_L_fusion*y(2)*y(end_N+v);

- 250 - Appendix III end for v=4:MaxN dy(2)=dy(2)-beta_fusion*y(2)*y(v)-beta_L_fusion*y(2)*y(end_N+v) -beta_L_fusion*y(2)*y(end_Complex1+v); end for k=3:end_N if rem(k,2)==0 % For even differential equations p=(k/2)-1; elseif rem(k,2)==1 % For odd differential equations p=(k-1)/2; end dy(k)=-kappa_1*y(end_L_a+1)*y(k)+kappa1*y(end_N+k); for w=1:p dy(k)=dy(k)+beta_fusion*y(k-w)*y(w+1); end for v=2:3 dy(k)=dy(k)-beta_fusion*y(k)*y(v)-beta_L_fusion*y(k)*y(end_N+v); end for v=4:MaxN-k+2 dy(k)=dy(k)-beta_fusion*y(k)*y(v)-beta_L_fusion*y(k)*y(end_N+v) -beta_L_fusion*y(k)*y(end_Complex1+v); end end

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations for one lymphocyte interaction with single monocyte to FBGC progression (1 to MaxN+1) %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy(end_N+1)=-gamma_L_diff*y(end_N+1)+kappa_1*y(end_L_a+1)*y(1)-kappa1*y(end_N+1); dy(end_N+2)=gamma_L_diff*y(end_N+1)+kappa_1*y(end_L_a+1)*y(2)-kappa1*y(end_N+2); dy(end_N+3)=kappa_1*y(end_L_a+1)*y(3)-kappa1*y(end_N+3)+beta_L_fusion*y(end_N+2)*y(2); for v=2:MaxN dy(end_N+2)=dy(end_N+2)-beta_L_fusion*y(end_N+2)*y(v); dy(end_N+3)=dy(end_N+3)-beta_L_fusion*y(end_N+3)*y(v); end for k=4:end_N dy(end_N+k)=kappa_1*y(end_L_a+1)*y(k)-kappa1*y(end_N+k)+kappa1*y(end_Complex1+k)- kappa_1*y(end_L_a+1)*y(end_N+k); for w=1:k-2 dy(end_N+k)=dy(end_N+k)+beta_L_fusion*y(end_N+k-w)*y(w+1); end for v=2:MaxN-k+2 dy(end_N+k)=dy(end_N+k)-beta_L_fusion*y(end_N+k)*y(v); end end

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations for 2 lymphocyte interactions with single monocyte to FBGC progression % (1 to MaxN+1) %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy(end_Complex1+1)=0;

- 251 - Appendix III dy(end_Complex1+2)=0; dy(end_Complex1+3)=0; for k=4:end_N dy(end_Complex1+k)=-kappa1*y(end_Complex1+k)+kappa_1*y(end_L_a+1)*y(end_N+k); for w=1:k-4 dy(end_Complex1+k)=dy(end_Complex1+k)+beta_L_fusion*y(end_Complex1+k-w)*y(w+1); end for v=2:MaxN-k+2 dy(end_Complex1+k)=dy(end_Complex1+k)-beta_L_fusion*y(end_Complex1+k)*y(v); end end

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations for nonadherent monocyte and macrophage %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy(end_Complex2+1)=delta_detach*y(1); % Rate of change of nonadherent monocytes dy(end_Complex2+2)=delta_detach*y(2); % Rate of change of nonadherent macrophages

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations for adherent and nonadherent lymphocytes %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy(end_N_na+1)=-alpha_aL_apop*y(end_N_na+1) + pi_aL_prolif*y(end_N_na+1) + nu_L_adhesion*y(end_L_a+1) - delta_L_detach*y(end_N_na+1); % Rate of change of adherent lymphocytes dy(end_L_a+1)=pi_naL_prolif*y(end_L_a+1)-nu_L_adhesion*y(end_L_a+1) +delta_L_detach*y(end_N_na+1); % Rate of change of nonadherent lymphocytes for v=1:3 dy(end_L_a+1)=dy(end_L_a+1)-kappa_1*y(end_L_a+1)*y(v)+kappa1*y(end_N+v) -kappa_1*y(end_L_a+1)*y(end_N+v)+kappa1*y(end_Complex1+v); end for v=4:MaxN+1 dy(end_L_a+1)=dy(end_L_a+1)-kappa_1*y(end_L_a+1)*y(end_N+v)+kappa1*y(end_Complex1+v); end

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations for Apoptosed cells %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy(end_L_na+1)=alpha_mono_apop*y(1); % Rate of change of apoptosed monocytes dy(end_L_na+2)=alpha_macro_apop*y(2); % Rate of change of apoptosed macrophages dy(end_Apoptosed+1)=alpha_aL_apop*y(end_N_na+1); % Rate of change of apoptosed adherent lymphocytes

%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ % Differential equations for Proliferated cells %~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dy(end_Apoptosed_aL+1)=pi_aL_prolif*y(end_N_na+1); % Proliferated adherent lymphocytes dy(end_Proliferated_aL+1)=pi_naL_prolif*y(end_L_a+1); % Proliferated nonadherent lymphocytes

- 252 - Works Cited

Abbas AK. The control of T cell activation vs. tolerance. Autoimmun Rev 2003;2(3):115-8.

Abbondanzo SL, Young VL, Wei MQ, Miller FW. Silicone gel-filled breast and testicular implant capsules: a histologic and immunophenotypic study. Mod Pathol 1999;12(7):706-13.

Abramson S, Alexander H, Best S, Bokros J, Brunski J, Colas A, Cooper S, Curtis J, Haubold A, LL H. Classes of materials used in medicine. In: Ratner BD HA, Schoen FJ, Lemons JE, editor. Biomaterial Science: An Introduction to Materials in Medicine. San Diego: Elsevier Academic Press; 2004. p 67-225.

Abramson SD, Alexe G, Hammer PL, Kohn J. A computational approach to predicting cell growth on polymeric biomaterials. J Biomed Mater Res A 2005;73(1):116-24.

Alderson MR, Armitage RJ, Tough TW, Strockbine L, Fanslow WC, Spriggs MK. CD40 expression by human monocytes: regulation by cytokines and activation of monocytes by the ligand for CD40. J Exp Med 1993;178(2):669-74.

Allen LT, Tosetto M, Miller IS, O'Connor DP, Penney SC, Lynch I, Keenan AK, Pennington SR, Dawson KA, Gallagher WM. Surface-induced changes in protein adsorption and implications for cellular phenotypic responses to surface interaction. Biomaterials 2006;27(16):3096-108.

Al-Saffar N, Revell PA, Kobayashi A. Modulation of the phenotypic and functional properties of phagocytic macrophages by wear particles from orthopaedic implants. J Mater Sci Mater Med 1997;8(11):641-8.

Anderson JM, Rodriguez A, Chang DT. Foreign body reaction to biomaterials. Semin Immunol 2007;20(2):86-100.

Anderson JM. Biological responses to materials. Annu Rev Mater Res 2001;31:81-110.

Anderson JM. Inflammation and the foreign body response. Problems in General Surgery 1994;11(2):147-60.

Anderson JM. Inflammatory reaction: the nemesis of implants. In: Zilla P, Greisler H, editors. Tissue Engineering of Vascular Prosthetic Grafts. Austin: R.G. Landes Company; 1999. p 197-206.

Anderson JM. Multinucleated giant cells. Curr Opin Hematol 2000;7(1):40-7.

Andersson J, Ekdahl KN, Lambris JD, Nilsson B. Binding of C3 fragments on top of

- 253 - adsorbed plasma proteins during complement activation on a model biomaterial surface. Biomaterials 2005;26(13):1477-85.

Ankersmit HJ, Edwards NM, Schuster M, John R, Kocher A, Rose EA, Oz M, Itescu S. Quantitative changes in T-cell populations after left ventricular assist device implantation: relationship to T-cell apoptosis and soluble CD95. Circulation 1999;100(19 Suppl):II211-5.

Ashany D, Song X, Lacy E, Nikolic-Zugic J, Friedman SM, Elkon KB. Th1 CD4+ lymphocytes delete activated macrophages through the Fas/APO-1 antigen pathway. Proc Natl Acad Sci U S A 1995;92(24):11225-9.

Athanasou NA, Quinn J. Immunophenotypic differences between osteoclasts and macrophage polykaryons: immunohistological distinction and implications for osteoclast ontogeny and function. J Clin Pathol 1990;43(12):997-1003.

Babensee JE, Paranjpe A. Differential levels of dendritic cell maturation on different biomaterials used in combination products. J Biomed Mater Res A 2005;74(4):503- 10.

Bacon KB, Premack BA, Gardner P, Schall TJ. Activation of dual T cell signaling pathways by the chemokine RANTES. Science 1995;269(5231):1727-30.

Bainbridge JA, Revell PA, Al-Saffar N. Costimulatory molecule expression following exposure to orthopaedic implants wear debris. J Biomed Mater Res 2001;54(3):328- 34.

Bajzer Z, Vuk-Pavlovic S. Modeling positive regulatory feedbacks in cell-cell interactions. Biosystems 2005;80(1):1-10.

Bale MD, Wohlfahrt LA, Mosher DF, Tomasini B, Sutton RC. Identification of vitronectin as a major plasma protein adsorbed on polymer surfaces of different copolymer composition. Blood 1989;74(8):2698-706.

Bender BS, Curtis JL, Nagel JE, Chrest FJ, Kraus ES, Briefel GR, Adler WH. Analysis of immune status of hemodialyzed adults: association with prior transfusions. Kidney Int 1984;26(4):436-43.

Bergman AJ, Zygourakis K. Migration of lymphocytes on fibronectin-coated surfaces: temporal evolution of migratory parameters. Biomaterials 1999;20(23-24):2235-44.

Berton G, Lowell CA. Integrin signalling in neutrophils and macrophages. Cell Signal 1999;11(9):621-35.

Bigerelle M, Anselme K. A kinetic approach to osteoblast adhesion on biomaterial surface. J Biomed Mater Res A 2005;75(3):530-40.

- 254 - Biltresse S, Attolini M, Marchand-Brynaert J. Cell adhesive PET membranes by surface grafting of RGD peptidomimetics. Biomaterials 2005;26(2):4576-87.

Brennan FM, Foey AD, Feldmann M. The importance of T cell interactions with macrophages in rheumatoid cytokine production. Curr Top Microbiol Immunol 2006;305:177-94.

Brennan FM, Hayes AL, Ciesielski CJ, Green P, Foxwell BM, Feldmann M. Evidence that rheumatoid arthritis synovial T cells are similar to cytokine-activated T cells: involvement of phosphatidylinositol 3-kinase and nuclear factor kappaB pathways in tumor necrosis factor alpha production in rheumatoid arthritis. Arthritis Rheum 2002;46(1):31-41.

Brevig T, Holst B, Ademovic Z, Rozlosnik N, Rohrmann JH, Larsen NB, Hansen OC, Kingshott P. The recognition of adsorbed and denatured proteins of different topographies by beta2 integrins and effects on leukocyte adhesion and activation. Biomaterials 2005;26(16):3039-53.

Brodbeck WG, Colton E, Anderson JM. Effects of adsorbed heat labile serum proteins and fibrinogen on adhesion and apoptosis of monocytes/macrophages on biomaterials. J Mater Sci Mater Med 2003;14(8):671-5.

Brodbeck WG, Macewan M, Colton E, Meyerson H, Anderson JM. Lymphocytes and the foreign body response: lymphocyte enhancement of macrophage adhesion and fusion. J Biomed Mater Res A 2005;74(2):222-9.

Brodbeck WG, Nakayama Y, Matsuda T, Colton E, Ziats NP, Anderson JM. Biomaterial surface chemistry dictates adherent monocyte/macophage cytokine expression in vitro. Cytokine 2002;18(6):311-19.

Brodbeck WG, Patel J, Voskerician G, Christenson E, Shive MS, Nakayama Y, Matsuda T, Ziats NP, Anderson JM. Biomaterial adherent macrophage apoptosis is increased by hydrophilic and anionic substrates in vivo. Proc Natl Acad Sci U S A 2002;99(16):10287-92.

Brodbeck WG, Shive MS, Colton E, Nakayama Y, Matsuda T, Anderson JM. Influence of biomaterial surface chemistry on the apoptosis of adherent cells. J Biomed Mater Res 2001;55(4):661-8.

Brodbeck WG, Shive MS, Colton E, Ziats NP, Anderson JM. Interleukin-4 inhibits tumor necrosis factor-alpha-induced and spontaneous apoptosis of biomaterial- adherent macrophages. J Lab Clin Med 2002;139(2):90-100.

Brodbeck WG, Voskerician G, Ziats NP, Nakayama Y, Matsuda T, Anderson JM. In vivo leukocyte cytokine mRNA responses to biomaterials are dependent on surface chemistry. J Biomed Mater Res A 2003;64(2):320-9.

- 255 - Broughton G, 2nd, Janis JE, Attinger CE. The basic science of wound healing. Plast Reconstr Surg 2006;117(7 Suppl):12S-34S.

Burger D, Dayer JM. Cytokines, acute-phase proteins, and hormones: IL-1 and TNF- alpha production in contact-mediated activation of monocytes by T lymphocytes. Ann N Y Acad Sci 2002;966:464-73.

Burger D, Dayer JM. The role of human T-lymphocyte-monocyte contact in inflammation and tissue destruction. Arthritis Res 2002;4 Suppl 3:S169-76.

Burger D. Cell contact-mediated signaling of monocytes by stimulated T cells: a major pathway for cytokine induction. Eur Cytokine Netw 2000;11(3):346-53.

Campbell DJ, Kim CH, Butcher EC. Chemokines in the systemic organization of immunity. Immunol Rev 2003;195:58-71.

Canonico P, Cassinelli C, Morra M. Surface-modified polyester fabrics. Med Device Technol 1997;8(3):16-21.

Castiglione A, Pagliaro P, Romagnoni M, Beccari M, Cavaliere G, Veneroni G, Sorgato G. Flow cytometric analysis of leukocytes eluted from haemodialysers. Nephrol Dial Transplant 1991;6 Suppl 2:31-5.

Chang DT, Colton E, Anderson JM. Paracrine and juxtacrine lymphocyte enhancement of adherent macrophage and foreign body giant cell activation. J Biomed Mater Res A 2008.

Chang DT, Jones JA, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Lymphocyte/macrophage interactions: biomaterial surface-dependent cytokine, chemokine, and matrix protein production. J Biomed Mater Res A 2008.

Charo IF, Ransohoff RM. The many roles of chemokines and chemokine receptors in inflammation. N Engl J Med 2006;354(6):610-21.

Chellat F, Grandjean-Laquerriere A, Le Naour R, Fernandes J, Yahia L, Guenounou M, Laurent-Maquin D. Metalloproteinase and cytokine production by THP-1 macrophages following exposure to chitosan-DNA nanoparticles. Biomaterials 2005;26(9):961-70.

Chinn J, Sauter J, Phillips R, Jr, Kao W, Anderson J, Hanson S, Ashton T. Blood and tissue compatibility of modified polyester: thrombosis, inflammation, and healing. J Biomed Mater Res 1998;39(1):130-40.

Chow CC, Clermont G, Kumar R, Lagoa C, Tawadrous Z, Gallo D, Betten B, Bartels J, Constantine G, Fink MP and others. The acute inflammatory response in diverse shock states. Shock 2005;24(1):74-84.

- 256 - Collier TO, Anderson JM, Brodbeck WG, Barber T, Healy KE. Inhibition of macrophage development and foreign body giant cell formation by hydrophilic interpenetrating polymer network. J Biomed Mater Res A 2004;69(4):644-50.

Collier TO, Anderson JM. Protein and surface effects on monocyte and macrophage adhesion, maturation, and survival. J Biomed Mater Res 2002;60(3):487-96.

Damsky CH, Ilic D. Integrin signaling: it's where the action is. Curr Opin Cell Biol 2002;14(5):594-602.

Daniels AU, Barnes FH, Charlebois SJ, Smith RA. Macrophage cytokine response to particles and lipopolysaccharide in vitro. J Biomed Mater Res 2000;49(4):469-78. de Waal Malefyt R, Abrams J, Bennett B, Figdor CG, de Vries JE. Interleukin 10(IL-10) inhibits cytokine synthesis by human monocytes: an autoregulatory role of IL-10 produced by monocytes. J Exp Med 1991;174(5):1209-20.

DeFife KM, Jenney CR, McNally AK, Colton E, Anderson JM. Interleukin-13 induces human monocyte/macrophage fusion and macrophage mannose receptor expression. J Immunol 1997;158(7):3385-90.

DeFife KM, Yun JK, Azeez A, Stack S, Ishihara K, Nakabayashi N, Colton E, Anderson JM. Adhesion and cytokine production by monocytes on poly(2- methacryloyloxyethyl phosphorylcholine-co-alkyl methacrylate)-coated polymers. J Biomed Mater Res 1995;29(4):431-9.

Delon I, Brown NH. Integrins and the actin cytoskeleton. Curr Opin Cell Biol 2007;19(1):43-50.

Ding L, Linsley PS, Huang LY, Germain RN, Shevach EM. IL-10 inhibits macrophage costimulatory activity by selectively inhibiting the up-regulation of B7 expression. J Immunol 1993;151(3):1224-34.

Dinnes DL, Santerre JP, Labow RS. Influence of biodegradable and non-biodegradable material surfaces on the differentiation of human monocyte-derived macrophages. Differentiation 2008;76(3):232-44.

Doussis IA, Puddle B, Athanasou NA. Immunophenotype of multinucleated and mononuclear cells in giant cell lesions of bone and soft tissue. J Clin Pathol 1992;45(5):398-404.

Enelow RI, Sullivan GW, Carper HT, Mandell GL. Induction of multinucleated giant cell formation from in vitro culture of human monocytes with interleukin-3 and interferon-gamma: comparison with other stimulating factors. Am J Respir Cell Mol Biol 1992;6(1):57-62.

- 257 - Epstein NJ, Bragg WE, Ma T, Spanogle J, Smith RL, Goodman SB. UHMWPE wear debris upregulates mononuclear cell proinflammatory gene expression in a novel murine model of intramedullary particle disease. Acta Orthop 2005;76(3):412-20.

Esche C, Stellato C, Beck LA. Chemokines: key players in innate and adaptive immunity. J Invest Dermatol 2005;125(4):615-28.

Eskin S, Horbett T, McIntire L, Mitchell R, Ratner B, Schoen F, Yee A. Some Background Concepts. In: Ratner BD, Hoffman AS, Schoen FJ, Lemons JE, editors. Biomaterial Sciences: An Introduction to Materials in Medicine. San Diego: Elsevier Academic Press; 2004. p 237-291.

Farber A, Chin R, Song Y, Huie P, Goodman S. Chronic antigen-specific immune- system activation may potentially be involved in the loosening of cemented acetabular components. J Biomed Mater Res 2001;55(3):433-41.

Friedl P, den Boer AT, Gunzer M. Tuning immune responses: diversity and adaptation of the immunological synapse. Nat Rev Immunol 2005;5(7):532-45.

Frisch SM, Screaton RA. Anoikis mechanisms. Curr Opin Cell Biol 2001;13(5):555-62.

Garcia AJ. Get a grip: integrins in cell-biomaterial interactions. Biomaterials 2005;26(36):7525-9.

Geginat J, Sallusto F, Lanzavecchia A. Cytokine-driven proliferation and differentiation of human naive, central memory, and effector memory CD4(+) T cells. J Exp Med 2001;194(12):1711-9.

Gerard C, Rollins BJ. Chemokines and disease. Nat Immunol 2001;2(2):108-15.

Giancotti FG, Ruoslahti E. Integrin signaling. Science 1999;285(5430):1028-32.

Goodman SB. Wear particles, periprosthetic osteolysis and the immune system. Biomaterials 2007;28(34):5044-8.

Gorbet MB, Sefton MV. Biomaterial-associated thrombosis: roles of coagulation factors, complement, platelets and leukocytes. Biomaterials 2004;25(26):5681-703.

Granchi D, Ciapetti G, Filippini F, Stea S, Cenni E, Pizzoferrato A, Toni A. In vitro cytokine production by mononuclear cells exposed to bone cement extracts. Biomaterials 2000;21(17):1789-95.

Granchi D, Ciapetti G, Savarino L, Stea S, Filippini F, Sudanese A, Rotini R, Giunti A. Expression of the CD69 activation antigen on lymphocytes of patients with hip prosthesis. Biomaterials 2000;21(20):2059-65.

- 258 - Grooteman MP, Nube MJ, van Limbeek J, Schoorl M, van Houte AJ. Lymphocyte subsets in dialyser eluates: a new parameter of bioincompatibility? Nephrol Dial Transplant 1996;11(6):1073-8.

Groth T, Altankov G, Klosz K. Adhesion of human peripheral blood lymphocytes is dependent on surface wettability and protein preadsorption. Biomaterials 1994;15(6):423-8.

Groth T, Klosz K, Campbell EJ, New RR, Hall B, Goering H. Protein adsorption, lymphocyte adhesion and platelet adhesion/activation on polyurethane ureas is related to hard segment content and composition. J Biomater Sci Polym Ed 1994;6(6):497-510.

Groth T, Zlatanov I, Altankov G. Adhesion of human peripheral lymphocytes on biomaterials preadsorbed with fibronectin and vitronectin. J Biomater Sci Polym Ed 1994;6(8):729-39.

Groux H, Bigler M, de Vries JE, Roncarolo MG. Interleukin-10 induces a long-term antigen-specific anergic state in human CD4+ T cells. J Exp Med 1996;184(1):19- 29.

Groux H, Cottrez F. The complex role of interleukin-10 in autoimmunity. J Autoimmun 2003;20(4):281-5.

Gupta B, Plummer C, Bisson I, Frey P, Hilborn J. Plasma-induced graft polymerization of acrylic acid onto poly(ethylene terephthalate) films: characterization and human smooth muscle cell growth on grafted films. Biomaterials 2002;23(3):863-71.

Hallab NJ, Anderson S, Stafford T, Glant T, Jacobs JJ. Lymphocyte responses in patients with total hip arthroplasty. J Orthop Res 2005;23(2):384-91.

Hallett WH, Murphy WJ. Positive and negative regulation of Natural Killer cells: therapeutic implications. Semin Cancer Biol 2006;16(5):367-82.

Henson PM. The immunologic release of constituents from neutrophil leukocytes. I. The role of antibody and complement on nonphagocytosable surfaces or phagocytosable particles. J Immunol 1971;107(6):1535-46.

Henson PM. The immunologic release of constituents from neutrophil leukocytes. II. Mechanisms of release during phagocytosis, and adherence to nonphagocytosable surfaces. J Immunol 1971;107(6):1547-57.

Hernandez-Pando R, Bornstein QL, Aguilar Leon D, Orozco EH, Madrigal VK, Martinez Cordero E. Inflammatory cytokine production by immunological and foreign body multinucleated giant cells. Immunology 2000;100(3):352-8.

- 259 - Hogg N, Laschinger M, Giles K, McDowall A. T-cell integrins: more than just sticking points. J Cell Sci 2003;116(Pt 23):4695-705.

Horbett TA. Principles underlying the role of adsorbed plasma proteins in blood interactions with foreign materials. Cardiovascular Pathology 1999;2(3 (Suppl.)):137S-148S.

Houssiau FA, Coulie PG, Van Snick J. Distinct roles of IL-1 and IL-6 in human T cell activation. J Immunol 1989;143(8):2520-4.

Hoves S, Krause SW, Schutz C, Halbritter D, Scholmerich J, Herfarth H, Fleck M. Monocyte-derived human macrophages mediate anergy in allogeneic T cells and induce regulatory T cells. J Immunol 2006;177(4):2691-8.

Hu WJ, Eaton JW, Ugarova TP, Tang L. Molecular basis of biomaterial-mediated foreign body reactions. Blood 2001;98(4):1231-8.

Hwang JJ, Jelacic S, Samuel NT, Maier RV, Campbell CT, Castner DG, Hoffman AS, Stayton PS. Monocyte activation on polyelectrolyte multilayers. J Biomater Sci Polym Ed 2005;16(2):237-51.

Isler P, Vey E, Zhang JH, Dayer JM. Cell surface glycoproteins expressed on activated human T cells induce production of interleukin-1 beta by monocytic cells: a possible role of CD69. Eur Cytokine Netw 1993;4(1):15-23.

Isomaki P, Clark JM, Panesar M, Cope AP. Pathways of T cell activation and terminal differentiation in chronic inflammation. Curr Drug Targets Inflamm Allergy 2005;4(3):287-93.

Itescu S, John R. Interactions between the recipient immune system and the left ventricular assist device surface: immunological and clinical implications. Ann Thorac Surg 2003;75(6 Suppl):S58-65.

Ito T, Iwasaki Y, Narita T, Akiyoshi K, Ishihara K. Controlled adhesion of human lymphocytes on electrically charged polymer surface having phosphorylcholine moiety. Science and Technology of Advanced Materials 2003;4(2):99-104.

Jacobs JJ, Hallab NJ, Urban RM, Wimmer MA. Wear particles. J Bone Joint Surg Am 2006;88 Suppl 2:99-102.

Janeway CA, Travers P, Walport M, Shlomchik M. Immunobiology: The Immune System in Health and Disease. New York: Garland Publishing; 2001. 1-708 p.

Jenney CR, Anderson JM. Adsorbed serum proteins responsible for surface dependent human macrophage behavior. J Biomed Mater Res 2000;49(4):435-47.

- 260 - Jenney CR, Anderson JM. Effects of surface-coupled polyethylene oxide on human macrophage adhesion and foreign body giant cell formation in vitro. J Biomed Mater Res 1999;44(2):206-16.

Jones JA, Chang DT, Meyerson H, Colton E, Kwon IK, Matsuda T, Anderson JM. Proteomic analysis and quantification of cytokines and chemokines from biomaterial surface-adherent macrophages and foreign body giant cells. J Biomed Mater Res A 2007;83(3):585-96.

Jones JA, Dadsetan M, Collier TO, Ebert M, Stokes KS, Ward RS, Hiltner PA, Anderson JM. Macrophage behavior on surface-modified polyurethanes. J Biomater Sci Polym Ed 2004;15(5):567-84.

Jones JA, McNally AK, Chang DT, Qin LA, Meyerson H, Colton E, Kwon IL, Matsuda T, Anderson JM. Matrix metalloproteinases and their inhibitors in the foreign body reaction on biomaterials. J Biomed Mater Res A 2008;84(1):158-66.

Jones JA. Biomaterials and the Foreign Body Reaction: Surface Chemistry Dependent Macrophage Adhesion, Fusion, Apoptosis, and Cytokine Production. Dissertation. Case Western Reserve University, 2007.

Jurkowski MK, Bobek-Billewicz B, Cwiklinska-Jurkowska M, Jurkowski P. Effect on peripheral blood natural killer cytotoxic cell activity in rats after intraperitoneal implantation of double veloured polyester (Dacron) prosthesis. Arch Immunol Ther Exp (Warsz) 2004;52(2):129-36.

Kadoya Y, al-Saffar N, Kobayashi A, Revell PA. The expression of osteoclast markers on foreign body giant cells. Bone Miner 1994;27(2):85-96.

Kanegane H, Tosato G. Activation of naive and memory T cells by interleukin-15. Blood 1996;88(1):230-5.

Kao WJ, Hiltner A, Anderson JM, Lodoen GA. Theoretical analysis of in vivo macrophage adhesion and foreign body giant cell formation on strained poly(etherurethane urea) elastomers. J Biomed Mater Res 1994;28(7):819-29.

Kao WJ, Lee D, Schense JC, Hubbell JA. Fibronectin modulates macrophage adhesion and FBGC formation: the role of RGD, PHSRN, and PRRARV domains. J Biomed Mater Res 2001;55(1):79-88.

Kao WJ, McNally AK, Hiltner A, Anderson JM. Role for interleukin-4 in foreign-body giant cell formation on a poly(etherurethane urea) in vivo. J Biomed Mater Res 1995;29(10):1267-75.

Kao WJ, Zhao QH, Hiltner A, Anderson JM. Theoretical analysis of in vivo macrophage adhesion and foreign body giant cell formation on polydimethylsiloxane, low density

- 261 - polyethylene, and polyetherurethanes. J Biomed Mater Res 1994;28(1):73-9.

Katzin WE, Feng LJ, Abbuhl M, Klein MA. Phenotype of lymphocytes associated with the inflammatory reaction to silicone gel breast implants. Clin Diagn Lab Immunol 1996;3(2):156-61.

Keegan AD. IL-4. In: Oppenheim JJ, Feldman M, editors. Cytokine Reference. San Diego, CA: Academic Press; 2001.

Keselowsky BG, Bridges AW, Burns KL, Tate CC, Babensee JE, LaPlaca MC, Garcia AJ. Role of plasma fibronectin in the foreign body response to biomaterials. Biomaterials 2007;28(25):3626-31.

Keselowsky BG, Collard DM, Garcia AJ. Integrin binding specificity regulates biomaterial surface chemistry effects on cell differentiation. Proc Natl Acad Sci U S A 2005;102(17):5953-7.

Keselowsky BG, Collard DM, Garcia AJ. Surface chemistry modulates focal adhesion composition and signaling through changes in integrin binding. Biomaterials 2004;25(28):5947-54.

Kiener PA, Moran-Davis P, Rankin BM, Wahl AF, Aruffo A, Hollenbaugh D. Stimulation of CD40 with purified soluble gp39 induces proinflammatory responses in human monocytes. J Immunol 1995;155(10):4917-25.

Kim JK, Scott EA, Elbert DL. Proteomic analysis of protein adsorption: serum amyloid P adsorbs to materials and promotes leukocyte adhesion. J Biomed Mater Res A 2005;75(1):199-209.

Kim Y, Kang I, Huh M, Yoon S. Surface characterization and in vitro blood compatibility of poly(ethylene terephthalate) immobilized with insulin and/or heparin using plasma glow discharge. Biomaterials 2000;21(2):121-30.

Kumar R, Clermont G, Vodovotz Y, Chow CC. The dynamics of acute inflammation. J Theor Biol 2004;230(2):145-55.

Kyriakides TR, Foster MJ, Keeney GE, Tsai A, Giachelli CM, Clark-Lewis I, Rollins BJ, Bornstein P. The CC chemokine ligand, CCL2/MCP1, participates in macrophage fusion and foreign body giant cell formation. Am J Pathol 2004;165(6):2157-66.

Le JM, Fredrickson G, Reis LF, Diamantstein T, Hirano T, Kishimoto T, Vilcek J. Interleukin 2-dependent and interleukin 2-independent pathways of regulation of thymocyte function by interleukin 6. Proc Natl Acad Sci U S A 1988;85(22):8643-7.

Lee MH, Ducheyne P, Lynch L, Boettiger D, Composto RJ. Effect of biomaterial

- 262 - surface properties on fibronectin-alpha5beta1 integrin interaction and cellular attachment. Biomaterials 2006;27(9):1907-16.

Ley K, Laudanna C, Cybulsky MI, Nourshargh S. Getting to the site of inflammation: the leukocyte adhesion cascade updated. Nat Rev Immunol 2007;7(9):678-89.

Lodoen MB, Lanier LL. Natural killer cells as an initial defense against pathogens. Curr Opin Immunol 2006;18(4):391-8.

Ludewig B, Krebs P, Junt T, Metters H, Ford NJ, Anderson RM, Bocharov G. Determining control parameters for dendritic cell-cytotoxic T lymphocyte interaction. Eur J Immunol 2004;34(9):2407-18.

Luttikhuizen DT, Harmsen MC, Van Luyn MJ. Cellular and molecular dynamics in the foreign body reaction. Tissue Eng 2006;12(7):1955-70.

Ma N, Petit A, Yahia L, Huk OL, Tabrizian M. Cytotoxic reaction and TNF-alpha response of macrophages to polyurethane particles. J Biomater Sci Polym Ed 2002;13(3):257-72.

Macallan DC, Asquith B, Irvine AJ, Wallace DL, Worth A, Ghattas H, Zhang Y, Griffin GE, Tough DF, Beverley PC. Measurement and modeling of human T cell kinetics. Eur J Immunol 2003;33(8):2316-26.

MacEwan MR, Brodbeck WG, Matsuda T, Anderson JM. Student Research Award in the Undergraduate Degree Candidate category, 30th Annual Meeting of the Society for Biomaterials, Memphis, Tennessee, April 27-30, 2005. Monocyte/lymphocyte interactions and the foreign body response: in vitro effects of biomaterial surface chemistry. J Biomed Mater Res A 2005;74(3):285-93.

Mach F, Schonbeck U, Bonnefoy JY, Pober JS, Libby P. Activation of monocyte/macrophage functions related to acute atheroma complication by ligation of CD40: induction of collagenase, stromelysin, and tissue factor. Circulation 1997;96(2):396-9.

Maeda M, Kimura M, Inoue S, Kataoka K, Okano T, Sakurai Y. Adhesion behavior of rat lymphocyte subpopulations (B cell and T cell) on the surface of polystyrene/polypeptide graft copolymer. J Biomed Mater Res 1986;20(1):25-35.

Maghazachi AA, Al-Aoukaty A, Schall TJ. CC chemokines induce the generation of killer cells from CD56+ cells. Eur J Immunol 1996;26(2):315-9.

Maher SG, Romero-Weaver AL, Scarzello AJ, Gamero AM. Interferon: cellular executioner or white knight? Curr Med Chem 2007;14(12):1279-89.

Manie S, Kubar J, Limouse M, Ferrua B, Ticchioni M, Breittmayer JP, Peyron JF,

- 263 - Schaffar L, Rossi B. CD3-stimulated Jurkat T cells mediate IL-1 beta production in monocytic THP-1 cells. Role of LFA-1 molecule and participation of CD69 T cell antigen. Eur Cytokine Netw 1993;4(1):7-13.

Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol 2004;25(12):677-86.

Marchant R, Wang I. Physical and chemical aspects of biomaterials used in humans. In: RS G, editor. Implantation Biology. Boca Raton: CRC Press; 1994. p 13-39.

Marques AP, Reis RL, Hunt JA. Cytokine secretion from mononuclear cells cultured in vitro with starch-based polymers and poly-L-lactide. J Biomed Mater Res A 2004;71(3):419-29.

McKenzie ANJ, Matthews DJ. IL-13. In: Oppenheim JJ, Feldman M, editors. Cytokine Reference. San Diego, CA: Academic Press; 2001.

McNally AK, Anderson JM. Beta1 and beta2 integrins mediate adhesion during macrophage fusion and multinucleated foreign body giant cell formation. Am J Pathol 2002;160(2):621-30.

McNally AK, Anderson JM. Complement C3 participation in monocyte adhesion to different surfaces. Proc Natl Acad Sci U S A 1994;91(21):10119-23.

McNally AK, Anderson JM. Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 1995;147(5):1487-99.

McNally AK, DeFife KM, Anderson JM. Interleukin-4-induced macrophage fusion is prevented by inhibitors of mannose receptor activity. Am J Pathol 1996;149(3):975- 85.

McNally AK, Jones JA, Macewan SR, Colton E, Anderson JM. Vitronectin is a critical protein adhesion substrate for IL-4-induced foreign body giant cell formation. J Biomed Mater Res A 2007.

McNally AK, Macewan SR, Anderson JM. alpha subunit partners to beta1 and beta2 integrins during IL-4-induced foreign body giant cell formation. J Biomed Mater Res A 2007;82(3):568-74.

McNally AK. Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am J Pathol 1995;147(5):1487-99.

- 264 - Miller KM, Anderson JM. Human monocyte/macrophage activation and interleukin 1 generation by biomedical polymers. J Biomed Mater Res 1988;22(8):713-31.

Miyamoto YJ, Andruss BF, Mitchell JS, Billard MJ, McIntyre BW. Diverse roles of integrins in human T lymphocyte biology. Immunol Res 2003;27(1):71-84.

Monaco C, Andreakos E, Kiriakidis S, Feldmann M, Paleolog E. T-cell-mediated signalling in immune, inflammatory and angiogenic processes: the cascade of events leading to inflammatory diseases. Curr Drug Targets Inflamm Allergy 2004;3(1):35- 42.

Moretta A, Marcenaro E, Parolini S, Ferlazzo G, Moretta L. NK cells at the interface between innate and adaptive immunity. Cell Death Differ 2008;15(2):226-33.

Mosmann TR, Li L, Sad S. Functions of CD8 T-cell subsets secreting different cytokine patterns. Semin Immunol 1997;9(2):87-92.

Mott JD, Thomas CL, Rosenbach MT, Takahara K, Greenspan DS, Banda MJ. Post- translational proteolytic processing of procollagen C-terminal proteinase enhancer releases a metalloproteinase inhibitor. J Biol Chem 2000;275(2):1384-90.

Murch AR, Grounds MD, Marshall CA, Papadimitriou JM. Direct evidence that inflammatory multinucleate giant cells form by fusion. J Pathol 1982;137(3):177-80.

Nagase H, Visse R, Murphy G. Structure and function of matrix metalloproteinases and TIMPs. Cardiovasc Res 2006;69(3):562-73.

Nakayama Y, Anderson JM, Matsuda T. Laboratory-scale mass production of a multi- micropatterned grafted surface with different polymer regions. J Biomed Mater Res 2000;53(5):584-91.

Nakayama Y, Matsuda T, Irie M. A novel surface photo-graft polymerization method for fabricated devices. Asaio J 1993;39(3):M542-4.

Nakayama Y, Matsuda T. Photo induced surface heparin immobilization. Asaio J 1993;39(3):M754-7.

Nakayama Y, Matsuda T. Surface Macromolecular Architectural Designs Using Photo- Graft Copolymerization Based on Photochemistry of Benzyl N,N- Diethyldithiocarbamate. Macromolecules 1996;29:8622-8630.

Nakayama Y, Miyamura M, Hirano Y, Goto K, Matsuda T. Preparation of poly(ethylene glycol)-polystyrene block copolymers using photochemistry of dithiocarbamate as a reduced cell-adhesive coating material. Biomaterials 1999;20:963-970.

N'Dri NA, Shyy W, Tran-Son-Tay R. Computational modeling of cell adhesion and

- 265 - movement using a continuum-kinetics approach. Biophys J 2003;85(4):2273-86.

Neale SD, Athanasou NA. Cytokine receptor profile of arthroplasty macrophages, foreign body giant cells and mature osteoclasts. Acta Orthop Scand 1999;70(5):452- 8.

Nedvetzki S, Sowinski S, Eagle RA, Harris J, Vely F, Pende D, Trowsdale J, Vivier E, Gordon S, Davis DM. Reciprocal regulation of human natural killer cells and macrophages associated with distinct immune synapses. Blood 2007;109(9):3776- 85.

Nilsson B, Ekdahl KN, Mollnes TE, Lambris JD. The role of complement in biomaterial-induced inflammation. Mol Immunol 2007;44(1-3):82-94.

O'Connor GM, Hart OM, Gardiner CM. Putting the natural killer cell in its place. Immunology 2006;117(1):1-10.

Olson MW, Gervasi DC, Mobashery S, Fridman R. Kinetic analysis of the binding of human matrix metalloproteinase-2 and -9 to tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2. J Biol Chem 1997;272(47):29975-83.

Oppenheim J, Feldman M, Durum S, Hirano T, Vilcek J, Nicola N, editors. Cytokine reference: a compendium of cytokines and other mediators of host defense. Volume 1. San Diego: Academic Press; 2001. 2000 p.

Osthues A, Knauper V, Oberhoff R, Reinke H, Tschesche H. Isolation and characterization of tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2) from human rheumatoid synovial fluid. FEBS Lett 1992;296(1):16-20.

Otsu T, Yamashita K, K T. Synthesis, reactivity, and role of 4-vinylbenzyl N,N- diethyldithiocarbamate as a monomer-iniferter in radical polymerization. Macromolecules 1986;19:287-290.

Otsu T. Iniferter Concept and Living Radical Polymerization. Journal of Polymer Science: Part A: Polymer Chemistry 2000;38:2121-2136.

Park S, Bearinger J, Lautenschlager E, Castner D, Healy K. Surface modification of poly(ehtylene tererphthalate) angioplasty balloons with a hydrophilic poly(acrylamide-co-ethylene glycol) interpenetrating polymer network coating. J Biomed Mater Res 2000;53(5):568-76.

Parry SL, Sebbag M, Feldmann M, Brennan FM. Contact with T cells modulates monocyte IL-10 production: role of T cell membrane TNF-alpha. J Immunol 1997;158(8):3673-81.

Piglowski J, Gancarz I, Staniszewska-Kus J, Paluch D, Szymonowicz M, Konieczny K.

- 266 - Influence of plasma modification on biological properties of poly(ethylene terephthalate). Biomaterials 1994;15(11):909-16.

Poggi A, Prevosto C, Zancolli M, Canevali P, Musso A, Zocchi MR. NKG2D and natural cytotoxicity receptors are involved in natural killer cell interaction with self- antigen presenting cells and stromal cells. Ann N Y Acad Sci 2007;1109:47-57.

Pueyo ME, Darquy S, Capron F, Reach G. In vitro activation of human macrophages by alginate-polylysine microcapsules. J Biomater Sci Polym Ed 1993;5(3):197-203.

Quinn JM, Athanasou NA, McGee JO. Extracellular matrix receptor and platelet antigens on osteoclasts and foreign body giant cells. Histochemistry 1991;96(2):169- 76.

Raij L, Kay NE. Effect of hemodialysis membranes on human lymphocyte natural killer function. Contrib Nephrol 1987;59:17-25.

Ratner BD HA, Schoen FJ, Lemons JE, editor. Biomaterial Science: An Introduction to Materials in Medicine. 2nd ed. San Diego: Elsevier Academic Press; 2004.

Reddig PJ, Juliano RL. Clinging to life: cell to matrix adhesion and cell survival. Cancer Metastasis Rev 2005;24(3):425-39.

Refai AK, Textor M, Brunette DM, Waterfield JD. Effect of titanium surface topography on macrophage activation and secretion of proinflammatory cytokines and chemokines. J Biomed Mater Res A 2004;70(2):194-205.

Rhodes NP, Hunt JA, Williams DF. Macrophage subpopulation differentiation by stimulation with biomaterials. J Biomed Mater Res 1997;37(4):481-8.

Rodriguez A. T Cell Interactions in the Foreign Body Response to Biomaterials. Dissertation. Case Western Reserve University, 2007.

Romagnani S. Th1/Th2 cells. Inflamm Bowel Dis 1999;5(4):285-94.

Roncarolo MG, Bacchetta R, Bordignon C, Narula S, Levings MK. Type 1 T regulatory cells. Immunol Rev 2001;182:68-79.

Santana MA, Rosenstein Y. What it takes to become an effector T cell: the process, the cells involved, and the mechanisms. J Cell Physiol 2003;195(3):392-401.

Schall TJ, editor. The Cytokine Handbook. New York: Academic Press; 1994. 419-460 p.

Schuster M, Kocher A, John R, Hoffman M, Ankersmit J, Lietz K, Edwards N, Oz M, Itescu S. B-cell activation and allosensitization after left ventricular assist device

- 267 - implantation is due to T-cell activation and CD40 ligand expression. Hum Immunol 2002;63(3):211-20.

Schwartz RH. Models of T cell anergy: is there a common molecular mechanism? J Exp Med 1996;184(1):1-8.

Scotchford CA, Gilmore CP, Cooper E, Leggett GJ, Downes S. Protein adsorption and human osteoblast-like cell attachment and growth on alkylthiol on gold self- assembled monolayers. J Biomed Mater Res 2002;59(1):84-99.

Scott MJ, Hoth JJ, Stagner MK, Gardner SA, Peyton JC, Cheadle WG. CD40-CD154 interactions between macrophages and natural killer cells during sepsis are critical for macrophage activation and are not interferon gamma dependent. Clin Exp Immunol 2004;137(3):469-77.

Sebbag M, Parry SL, Brennan FM, Feldmann M. Cytokine stimulation of T lymphocytes regulates their capacity to induce monocyte production of tumor necrosis factor- alpha, but not interleukin-10: possible relevance to pathophysiology of rheumatoid arthritis. Eur J Immunol 1997;27(3):624-32.

Sethi RK, Neavyn MJ, Rubash HE, Shanbhag AS. Macrophage response to cross-linked and conventional UHMWPE. Biomaterials 2003;24(15):2561-73.

Shen M, Garcia I, Maier RV, Horbett TA. Effects of adsorbed proteins and surface chemistry on foreign body giant cell formation, tumor necrosis factor alpha release and procoagulant activity of monocytes. J Biomed Mater Res A 2004;70(4):533-41.

Shenton M, Lovell-Hoare M, Stevens G. Adhesion enhancement of polymer surfaces by atmospheric plasma treatment. J Phys D: Appl Phys 2001;34:2754-2760.

Shive MS, Brodbeck WG, Anderson JM. Activation of caspase 3 during shear stress- induced neutrophil apoptosis on biomaterials. J Biomed Mater Res 2002;62(2):163- 8.

Shu U, Kiniwa M, Wu CY, Maliszewski C, Vezzio N, Hakimi J, Gately M, Delespesse G. Activated T cells induce interleukin-12 production by monocytes via CD40- CD40 ligand interaction. Eur J Immunol 1995;25(4):1125-8.

Simon SI, Chambers JD, Sklar LA. Flow cytometric analysis and modeling of cell-cell adhesive interactions: the neutrophil as a model. J Cell Biol 1990;111(6 Pt 1):2747- 56.

Singh M, Kazzaz J, Ugozzoli M, Chesko J, O'Hagan DT. Charged polylactide co- glycolide microparticles as antigen delivery systems. Expert Opin Biol Ther 2004;4(4):483-91.

- 268 - Smetana K, Jr. Multinucleate foreign-body giant cell formation. Exp Mol Pathol 1987;46(3):258-65.

Stetler-Stevenson WG, Liotta LA, Kleiner DE, Jr. Extracellular matrix 6: role of matrix metalloproteinases in tumor invasion and metastasis. Faseb J 1993;7(15):1434-41.

Strengell M, Matikainen S, Siren J, Lehtonen A, Foster D, Julkunen I, Sareneva T. IL-21 in synergy with IL-15 or IL-18 enhances IFN-gamma production in human NK and T cells. J Immunol 2003;170(11):5464-9.

Tan W, Teo G, Liao K, Leong K, Mao H, Chan V. Adhesion contact dynamics of primary hepatocytes on poly(ethylene terephthalate) surface. Biomaterials 2005;26(8):891-8.

Tang L, Jennings TA, Eaton JW. Mast cells mediate acute inflammatory responses to implanted biomaterials. Proc Natl Acad Sci U S A 1998;95(15):8841-6.

Trindade MC, Lind M, Sun D, Schurman DJ, Goodman SB, Smith RL. In vitro reaction to orthopaedic biomaterials by macrophages and lymphocytes isolated from patients undergoing revision surgery. Biomaterials 2001;22(3):253-9.

Tsai AT, Rice J, Scatena M, Liaw L, Ratner BD, Giachelli CM. The role of osteopontin in foreign body giant cell formation. Biomaterials 2005;26(29):5835-43.

Unutmaz D, Pileri P, Abrignani S. Antigen-independent activation of naive and memory resting T cells by a cytokine combination. J Exp Med 1994;180(3):1159-64.

Visse R, Nagase H. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ Res 2003;92(8):827- 39.

Vodovotz Y, Clermont G, Chow C, An G. Mathematical models of the acute inflammatory response. Curr Opin Crit Care 2004;10(5):383-90.

Vodovotz Y. Deciphering the complexity of acute inflammation using mathematical models. Immunol Res 2006;36(1-3):237-45. von Andrian UH, Mackay CR. T-cell function and migration. Two sides of the same coin. N Engl J Med 2000;343(14):1020-34.

Vroman L. The life of an artificial device in contact with blood: initial events and their effect on its final state. Bull N Y Acad Med 1988;64(4):352-7.

Wagner C, Hansch GM, Stegmaier S, Denefleh B, Hug F, Schoels M. The complement receptor 3, CR3 (CD11b/CD18), on T lymphocytes: activation-dependent upregulation and regulatory function. Eur J Immunol 2001;31(4):1173-80.

- 269 - Wagner C, Hansch GM. Receptors for complement C3 on T-lymphocytes: relics of evolution or functional molecules? Mol Immunol 2006;43(1-2):22-30.

Wagner DH, Jr., Stout RD, Suttles J. Role of the CD40-CD40 ligand interaction in CD4+ T cell contact-dependent activation of monocyte interleukin-1 synthesis. Eur J Immunol 1994;24(12):3148-54.

Wagner VE, Bryers JD. Monocyte/macrophage interactions with base and linear- and star-like PEG-modified PEG-poly(acrylic acid) co-polymers. J Biomed Mater Res A 2003;66(1):62-78.

Walzak M, Flynn S, Foerch R, Hill J, Karbashewski E, Lin A, Strobel M. UV and ozone treatment of polypropylene and poly(ethylene terephthalate). In: Mittal K, editor. Polymer Surface Modification: Relevance to Adhesion. Utrecht: VSP; 1995. p 253- 72.

Ward SG, Westwick J. Chemokines: understanding their role in T-lymphocyte biology. Biochem J 1998;333 ( Pt 3):457-70.

Wilson CJ, Clegg RE, Leavesley DI, Pearcy MJ. Mediation of biomaterial-cell interactions by adsorbed proteins: a review. Tissue Eng 2005;11(1-2):1-18.

Xu LC, Siedlecki CA. Effects of surface wettability and contact time on protein adhesion to biomaterial surfaces. Biomaterials 2007;28(22):3273-83.

Xue M, Thompson PJ, Clifton-Bligh R, Fulcher G, Gallery ED, Jackson C. Leukocyte matrix metalloproteinase-9 is elevated and contributes to lymphocyte activation in type I diabetes. Int J Biochem Cell Biol 2005;37(11):2406-16.

Yagi M, Miyamoto T, Sawatani Y, Iwamoto K, Hosogane N, Fujita N, Morita K, Ninomiya K, Suzuki T, Miyamoto K and others. DC-STAMP is essential for cell- cell fusion in osteoclasts and foreign body giant cells. J Exp Med 2005;202(3):345- 51.

Ying L, Yin C, Zhuo R, Leong K, Mao H, Kang E, Neoh K. Immobilization of galactose ligands on acrylic acid graft-copolymerized poly(ehtylene terephthalate) film and its application to hepatocyte culture. Biomacromolecules 2003;4(1):157-65.

Yokoyama M, Nakahashi T, Nishimura T, Maeda M, Inoue S, Kataoka K, Sakurai Y. Adhesion behavior of rat lymphocytes to poly(ether)-poly(amino acid) block and graft copolymers. J Biomed Mater Res 1986;20(7):867-78.

Yoshida M, Babensee JE. Differential effects of agarose and poly(lactic-co-glycolic acid) on dendritic cell maturation. J Biomed Mater Res A 2006;79(2):393-408.

Yoshida M, Babensee JE. Poly(lactic-co-glycolic acid) enhances maturation of human

- 270 - monocyte-derived dendritic cells. J Biomed Mater Res A 2004;71(1):45-54.

Yoshida M, Mata J, Babensee JE. Effect of poly(lactic-co-glycolic acid) contact on maturation of murine bone marrow-derived dendritic cells. J Biomed Mater Res A 2007;80(1):7-12.

Zdolsek J, Eaton JW, Tang L. Histamine release and fibrinogen adsorption mediate acute inflammatory responses to biomaterial implants in humans. J Transl Med 2007;5:31.

Zhao OH, Anderson JM, Hiltner A, Lodoen GA, Payet CR. Theoretical analysis on cell size distribution and kinetics of foreign-body giant cell formation in vivo on polyurethane elastomers. J Biomed Mater Res 1992;26(8):1019-38.

Zhao Q, Topham N, Anderson JM, Hiltner A, Lodoen G, Payet CR. Foreign-body giant cells and polyurethane biostability: in vivo correlation of cell adhesion and surface cracking. J Biomed Mater Res 1991;25(2):177-83.

Zhao Q, Topham N, Anderson JM, Lodoen G, Payet CR. Foreign-body giant cells and polyurethane biostability: in vivo correlation of cell adhesion and surface cracking. J Biomed Mater Res 1991;25(2):177-83.

Zhao QH, McNally AK, Rubin KR, Renier M, Wu Y, Rose-Caprara V, Anderson JM, Hiltner A, Urbanski P, Stokes K. Human plasma alpha 2-macroglobulin promotes in vitro oxidative stress cracking of Pellethane 2363-80A: in vivo and in vitro correlations. J Biomed Mater Res 1993;27(3):379-88.

Zhu C, Williams TE. Modeling concurrent binding of multiple molecular species in cell adhesion. Biophys J 2000;79(4):1850-7.

Ziats NP, Pankowsky DA, Tierney BP, Ratnoff OD, Anderson JM. Adsorption of Hageman factor (factor XII) and other human plasma proteins to biomedical polymers. J Lab Clin Med 1990;116(5):687-96.

- 271 -