UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of surface topography on gene expression of macrophages Wong, Angela Ting Ting 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2013_fall_wong_angela.pdf [ 4.14MB ]
Metadata
JSON: 24-1.0074276.json
JSON-LD: 24-1.0074276-ld.json
RDF/XML (Pretty): 24-1.0074276-rdf.xml
RDF/JSON: 24-1.0074276-rdf.json
Turtle: 24-1.0074276-turtle.txt
N-Triples: 24-1.0074276-rdf-ntriples.txt
Original Record: 24-1.0074276-source.json
Full Text
24-1.0074276-fulltext.txt
Citation
24-1.0074276.ris

Full Text

THE EFFECT OF SURFACE TOPOGRAPHY ON GENE EXPRESSION OF MACROPHAGES   by Angela Ting Ting Wong  D.M.D., The University of British Columbia, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Craniofacial Science)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) September 2013    ? Angela Ting Ting Wong 2013   ii Abstract Osseointegration, the direct attachment of living bone to an implant surface, is necessary for the success of dental implant therapy. The surface topography of an implant influences cellular behaviour and osseointegration. In-vivo studies have shown that rough surface topographies are associated with accumulation of macrophages and subsequent bone formation (Chehroudi et al., 2009). Macrophages can exhibit classic inflammatory (M1) or alternative (M2) phenotypes that secrete different patterns of cytokines, chemokines and growth factors that affect wound healing. These experiments aim to characterize the phenotypes of unstimulated RAW264.7 macrophages grown on polished and SLA (sandblasted and acid-etched) surfaces using gene expression microarray technology. Whole genome microarray analysis showed differential expression of 199 genes on day 1 and 4943 genes on day 5 on SLA compared to polished surfaces. These genes were assigned to network categories related to cellular movement, immune cell trafficking, inflammation, cell-to-cell signalling and tissue development. Gene profiles were also compared to M1 and M2 macrophages obtained by stimulation with LPS and IL-4 respectively. Stimulated macrophages exhibited gene expression profiles consistent with their predicted phenotypes. Confirmation of upregulated expression of CCL4 and CCL7, macrophage chemoattractants associated with early inflammatory responses, was performed with quantitative real time polymerase chain reaction (qPCR). CCL4 was significantly upregulated on SLA surfaces by a factor of 2.18 after 1 day, and CCL7 was significantly upregulated on SLA surfaces by a factor of 4.01 after 5 days. Both CCL4 and CCL7 were also significantly upregulated after 1 day by M1 and M2 stimulated macrophages on the microarray. These experiments show that unstimulated macrophages grown on SLA surfaces increase gene expression of macrophage chemoattractants, and exhibit an intermediate phenotype with a mixture of M1 and M2 characteristics.     iii Preface This dissertation is the original work of the author, A. Wong and several collaborators:  ? The identification and design of the project described in this thesis was conceived by D.M. Brunette, J.D. Waterfield and K. Barth ? Reproduction of surface topographies in epoxy, RAW264.7 macrophage cell culturing, RNA extraction and purification and RNA quality analysis, in Chapter 3-4 was performed by A. Wong (80%) and R. Kim (20%) ? The CLIP-CHIP? mouse microarray and analysis, in Chapter 3-4 was performed by R. Kappelhoff in C.M. Overall?s lab at the University of British Columbia ? The whole genome microarray, in Chapter 3-4 was performed by A. Haegert at the Laboratory for Advanced Genome Analysis (LAGA) using RNA supplied by A. Wong ? Analysis of the whole genome microarray using GeneSpring v12 (Agilent Technologies) and Ingenuity Pathway Analysis was performed by A. Haegert and A. Wong. The student?s t test for experimental samples was performed by A. Haegart using GeneSpring v12. The one sided t-test was performed by R. Bell at LAGA using statistics environment R v2.13.0 ?  Qualitative Polymerase Chain Reaction (qPCR) and analysis of qPCR in Microsoft Excel was performed by A. Wong (80%) and R. Kim (20%) ? The Valentine Model in Chapter 5 was illustrated by Ms. Quan Ho  Material in this thesis was presented at the 2013 International Association of Dental Research meeting in Seattle, Washington and the 2013 Canadian Society of Biomaterials meeting in Ottawa, Ontario.   iv Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ......................................................................................................................... iv List of Tables .............................................................................................................................. viii List of Figures ................................................................................................................................ x List of Abbreviations .................................................................................................................. xii Acknowledgements .................................................................................................................... xiv Dedication .................................................................................................................................... xv Chapter  1: Introduction .............................................................................................................. 1 1.1 Dental Implants .................................................................................................................. 3 1.2 Biocompatibility ................................................................................................................. 5 1.3 Titanium ............................................................................................................................. 6 1.4 Overview of Surface Topography ...................................................................................... 8 1.4.1 Surface Topography Fabrication ............................................................................... 11 1.4.1.1 Plasma Spraying .................................................................................................. 11 1.4.1.2 Anodization ......................................................................................................... 12 1.4.1.3 Blasting ................................................................................................................ 12 1.4.1.4 Etching ................................................................................................................. 12 1.4.1.5 SLA Surface Topography .................................................................................... 13 1.4.2 Nanotopography ........................................................................................................ 14   v 1.5 Dental Implant Wound Healing ....................................................................................... 14 1.5.1 Stages of Wound Healing .......................................................................................... 15 1.5.1.1 Coagulation .......................................................................................................... 15 1.5.1.2 Protein Adsorption ............................................................................................... 16 1.5.1.3 Cellular Adherence, Proliferation and Differentiation ........................................ 17 1.5.1.4 De Novo Bone Formation and Remodelling ....................................................... 18 1.6 Macrophages .................................................................................................................... 19 1.6.1 Macrophage Polarization ........................................................................................... 20 1.6.2 RAW264.7 Macrophage Cell Line ............................................................................ 22 1.6.3 Macrophages and Surface Topography ..................................................................... 23 1.7 Gene Expression ............................................................................................................... 25 1.7.1 Microarrays ............................................................................................................... 26 1.7.2 Polymerase Chain Reaction ...................................................................................... 27 Chapter  2: Statement of the Problem and Objectives ............................................................ 29 2.1 Problem ............................................................................................................................ 29 2.2 Hypotheses ....................................................................................................................... 30 2.3 Rationale ........................................................................................................................... 30 2.4 Specific Aims ................................................................................................................... 31 2.5 Significance ...................................................................................................................... 31 Chapter  3: Materials and Methods .......................................................................................... 32 3.1 Substrata ........................................................................................................................... 32 3.2 Cell Culture ...................................................................................................................... 33   vi 3.3 Proliferation Study ........................................................................................................... 34 3.4 Gene Expression Studies .................................................................................................. 36 3.4.1 RNA Extraction ......................................................................................................... 36 3.4.2 RNA Quality Assessment .......................................................................................... 37 3.4.3 cDNA Production ...................................................................................................... 38 3.4.4 Microarray ................................................................................................................. 38 3.4.4.1 Whole Genome Microarray ................................................................................. 38 3.4.4.2 CLIP-CHIP? Microarray ................................................................................... 40 3.4.5 Polymerase Chain Reaction ...................................................................................... 40 3.4.5.1 Primers ................................................................................................................. 41 3.4.5.2 qPCR Efficiency .................................................................................................. 43 3.5 Statistical Analysis ........................................................................................................... 43 Chapter  4: Results ...................................................................................................................... 45 4.1 Proliferation Study ........................................................................................................... 45 4.2 RNA Quality .................................................................................................................... 52 4.3 CLIP-CHIP? Microarray ................................................................................................ 54 4.4 Whole Genome Microarray .............................................................................................. 55 4.5 Quantitative Polymerase Chain Reaction ......................................................................... 66 4.6 Comparative Study on Titanium-coated P and SLA Surfaces ......................................... 69 Chapter  5: Discussion, Conclusions and Future Directions .................................................. 70 5.1 Discussion ........................................................................................................................ 70 5.1.1 Upregulation of Macrophage Chemoattractants CCL4 and CCL7 ........................... 70   vii 5.1.2 Gene Expression Profiles of Macrophages Stimulated with LPS and IL-4 .............. 73 5.1.3 Other Studies ............................................................................................................. 73 5.1.4 Microarray Data Analysis ......................................................................................... 75 5.1.4.1 Annotation Discrepancies .................................................................................... 78 5.1.5 False Positives ........................................................................................................... 79 5.1.6 Conclusions ............................................................................................................... 80 5.1.7 Future Directions ....................................................................................................... 80 Bibliography ................................................................................................................................ 82 Appendices ................................................................................................................................... 95 Appendix A Supplemental Data ............................................................................................... 95 A.1 qPCR Results for CysC, HSP, CTNND1, ADM and MMP12 .................................... 95 A.2 ANOVA Tables ........................................................................................................... 98 Appendix B Copyright Permission ......................................................................................... 100    viii List of Tables  Table 1.1: Mechanical Properties of Titanium Used in Implant Dentistry ..................................... 8	 ?Table 1.2: Roughness Parameters Used to Characterize Implant Surface Topographies ............. 10	 ?Table 1.3: Surface Topography of Implants from Three Major Companies ................................ 11	 ?Table 1.4: Characteristics of Murine M1 and M2 Phenotypes ..................................................... 22	 ?Table 3.1: Randomized Arrangement of Samples on Microarray Slides ..................................... 39	 ?Table 4.1: RAW264.7 Macrophages Grown on Polished Surface Topography for 1 Day .......... 47	 ?Table 4.2: RAW264.7 Macrophages Grown on SLA Surface Topography for 1 Day ................. 47	 ?Table 4.3: RAW264.7 Macrophages Grown on P Surface Topography for 5 Days .................... 48	 ?Table 4.4: RAW264.7 Macrophages Grown on SLA Surface Topography for 5 Days ............... 48	 ?Table 4.6: Top 10 Upregulated Genes on SLA vs P Surfaces, Day 1 .......................................... 56	 ?Table 4.7: Top 10 Downregulated Genes on SLA vs P Surfaces, Day 1 ..................................... 57	 ?Table 4.8: Top 10 Upregulated Genes on SLA vs P Surfaces, Day 5 .......................................... 57	 ?Table 4.9: Top 10 Downregulated Genes on SLA vs P Surfaces, Day 5 ..................................... 58	 ?Table 4.10: Top 10 Upregulated Genes over 5 Days of Growth on P Surfaces ........................... 58	 ?Table 4.11: Top 10 Downregulated Genes over 5 Days of Growth on P Surfaces ...................... 59	 ?Table 4.12: Top 10 Upregulated Genes over 5 Days of Growth on SLA Surfaces ...................... 59	 ?Table 4.13: Top 10 Downregulated Genes over 5 Days of Growth on SLA Surfaces ................. 60	 ?Table 4.14: Top 10 Upregulated Genes on LPS-stimulated Cells on P Surfaces after 1 Day ...... 60	 ?Table 4.15: Top 10 Upregulated genes on IL-4-Stimulated Cells on P Surfaces after 1 Day ...... 61	 ?Table 4.16: Top Molecular and Cellular Functions Modulated by SLA vs P .............................. 61	 ?Table 4.17: Top Canonical Pathways Modulated by SLA vs P, Day 1 ........................................ 62	 ?  ix Table 4.18: Top Canonical Pathways Modulated by SLA vs P, Day 5 ........................................ 62	 ?Table 4.19: Top Networks Significantly Modulated by SLA vs P ............................................... 63	 ?Table 5.1: IPA Gene Annotation Discrepancies ........................................................................... 79	 ?   x List of Figures Figure 1.1: Spectrum of Macrophage Phenotypes .......................................................................... 2	 ?Figure 1.2: SEM Image of SLA Surface Topography .................................................................. 13	 ?Figure 1.3: Contact and Distant Osteogenesis .............................................................................. 19	 ?Figure 3.1: Design for 1-Day Proliferation Study ........................................................................ 35	 ?Figure 3.2: Design for 5-Day Proliferation Study ........................................................................ 36	 ?Figure 4.1: Phase Contrast Micrograph of RAW264.7 Macrophages on a P Surface .................. 46	 ?Figure 4.2: Cell Counts for 1 Day Proliferation Study ................................................................. 46	 ?Figure 4.3: Cell Counts for 5 Day Proliferation Study ................................................................. 47	 ?Figure 4.4: DAPI-stained Cells on P Surfaces after 1 Day, Seeded at 5x105 cells/ml ................. 48	 ?Figure 4.5: DAPI-stained Cells on SLA Surfaces after 1 Day, Seeded at 5x105 cells/ml ............ 49	 ?Figure 4.6: DAPI-stained Cells on P and SLA Surfaces for 1 Day, Seeded at 2x105 cells/ml ..... 49	 ?Figure 4.7: DAPI-stained Cells on P and SLA Surfaces for 1 Day, Seeded at 1x104 cells/ml ..... 50	 ?Figure 4.8: DAPI-stained Cells on P and SLA Surfaces for 5 Days, Seeded at 2x104 cells/ml ... 50	 ?Figure 4.9: DAPI-stained Cells on P and SLA Surfaces for 5 Days, Seeded at 1x104 cells/ml ... 50	 ?Figure 4.10: DAPI-stained Cells on P and SLA Surfaces for 5 Days, Seeded at 5x103 cells/ml . 51	 ?Figure 4.11: Acceptable Absorbance Curves Generated by the Spectrophotometer .................... 52	 ?Figure 4.12: Unacceptable Absorbance Curves Generated by the Spectrophotometer ................ 53	 ?Figure 4.13: An Acceptable Electropherogram for a Polished Sample, RIN 9.0 ......................... 53	 ?Figure 4.14: CLIP-CHIP? Results Showing Expression of MMP1a, MMP7 and MMP12 ....... 54	 ?Figure 4.15 Canonical Pathway Analysis for SLA vs P, Day 1 ................................................... 64	 ?Figure 4.16 Canonical Pathway Analysis for SLA vs P, Day 5 ................................................... 65	 ?Figure 4.17: Gene Expression of CCL4 on P and SLA at Day 1 .................................................. 66	 ?  xi Figure 4.18: Gene Expression of CCL7 on P and SLA at Day 5 .................................................. 67	 ?Figure 4.19: Efficiency of qPCR Using CCL4, CCL7 and CTNND1 Primers ............................ 68	 ?Figure 4.20: Relative Expression of CCL4 on Ti-coated P and SLA Surfaces at Day 1 .............. 69	 ?Figure 5.1: The Valentine Model of Macrophage Accumulation on Rough Surfaces ................. 72	 ?   xii List of Abbreviations  Al2O3 Aluminium oxide APC Antigen presenting cell ATCC American Type Culture Collection ASTM American Society of Testing and Materials bFGF Basic fibroblast growth factor  BMP-2 Bone morphogenetic protein-2 CCL2 Cysteine:cysteine (C:C) ligand 2 chemokine Also known as MCP-1, monocyte chemotactic protein-1 CCL3 C:C ligand 3 chemokine Also known as MIP-1?, macrophage inflammatory protein alpha CCL4 C:C ligand 4 chemokine Also known as MIP-1?, macrophage inflammatory protein beta CCL5 C:C ligand 5 chemokine Also known as RANTES, regulated on activation, normal T cell expressed and secreted  CCL7 C:C ligand 7 chemokine Also known as MCP-3, monocyte chemotactic protein-3 CCL8 C:C ligand 8 chemokine Also known as MCP-2, monocyte chemotactic protein-2 CCL13 C:C ligand 13 chemokine Also known as MCP-4, monocyte chemotactic protein-4 COX-2 Cyclooxygenase 2 CTNND1 Catenin/cadherin-associated protein, delta 1 Cy3 Cyanine 3-CTP dye Cy5 Cyanine 5-CTP dye CysB Cystatin B CysC Cystatin C DAPI 4?,6-diamidino-2-phenylindol DNA Deoxyribonucleic acid  ECM Extracellular matrix HSP Heat shock protein  IGF-1 Insulin-like growth factor 1 IgG Immunoglobulin G IL-1 Interleukin-1 alpha IL-1? Interleukin-1 beta IL-4 Interleukin-4 IL-6 Interleukin-6 IL-10 Interleukin-10 IL-12 Interleukin-12 INF? Interferon gamma LPS Lipopolysaccharide   xiii M1 Classically activated macrophages M2 Alternatively activated macrophages MCP-1 Monocyte chemotactic protein-1, also known as CCL2 MCP-2 Monocyte chemotactic protein-2, also known as CCL8 MCP-3 Monocyte chemotactic protein-3, also known as CCL7 MCP-4 Monocyte chemotactic protein-3, also known as CCL13 MIP-1? Macrophage inflammatory protein-1 alpha, also known as CCL3 MIP-1? Macrophage inflammatory protein-1 beta, also known as CCL4 ul Micrometer ml Millilitre mm Millimetre NF-k? Nuclear factor-kappa beta nm Nanometer P Polished surface PBS Phosphate buffered saline PDEGF Platelet derived endothelial growth factor PDGF Platelet derived growth factor PMN  Polymononuclear Ra Average roughness over all points on a profile  Rq Root mean square deviation of the profile RzDIN Average of maximum peak to valley heights  Rt The maximum peak to valley height of a profile  RNA Ribonucleic acid  Sa Average roughness over all points on a surface Sq Root mean square deviation of the surface St The maximum peak to valley height of a surface Sdr Developed surface area  SD Standard deviation SEM Scanning electron microscope SLA Sandblasted and acid-etched TGF?  Transforming growth factor beta Th1 T helper cell 1 Th2 T helper cell 2 Ti Titanium Ti-6AL-4V Titanium alloy TiO2 Titanium oxide TNF? Tumour necrosis factor alpha     xiv Acknowledgements  I wish to thank the kind faculty and staff at the University of British Columbia for their continued patience, tolerance and support, particularly my supervisor, Dr. Don Brunette and committee members Dr. Doug Waterfield and Dr. Clive Roberts.   I appreciate the kindness of the Dr. C. Schuler lab for allowing me to use the nanodrop spectrophotomer, bioanalyzer and PCR machine, and the former laboratory of Mr. Andre Wong, manager of the Faculty of Dentistry?s confocal microscope for his help with fluorescence microscopy. I thank Dr. E. Putnins? lab for allowing shared use of the ultrasonic machine and the mastercycler.   I am very grateful to Dr. Katrin Barth for her advice, teaching, guidance, help, and for being a role model of laboratory organization. I give many thanks and appreciation towards Ms. Rosa Kim for her qPCR expertise and invaluable help with pipetting. I am thankful for the members of the Brunette lab, particularly Ms. Haisle Moon, Ms. Quan Ho and Ms. Kiana Kianoush for assisting with cell surface production, maintenance of cell cultures and eating my baked goods.     I would like to thank Dr. Reini Kappelhoff for all her patience and time spent on the CLIP-CHIP? experiment, as well as for making time for multiple meetings and helpful discussions. I would also like to thank Ms. Anne Haegart and Mr. Robert Bell from LAGA for spending the time to demonstrate and explain the analysis of the microarray data.     xv Dedication  I would like to dedicate this thesis to my parents. I am extremely grateful for their love and support of all my academic and personal endeavours.   1 Chapter  1: Introduction  Implants are medical devices that can replace, support or enhance a missing, damaged or deficient biological structure. Many tissues and organs lack the ability to regenerate following significant damage and require implants to help restore original form and function (Ratner, 2001a). Some examples of implants used today include orthopaedic hip and knee prostheses, dental implants, intraocular lenses, vascular grafts, cardiac pacemakers, heart valves, breast implants and implantable drug delivery devices (Elman, Ho Duc, & Cima, 2009; Ratner, 2001a).  Ideally, an implanted material should do no harm to adjacent tissue and promote a favourable healing response. Successful wound healing relies on the ability to promote tissue attachment to its surface and to reduce the occurrence of chronic inflammation and fibrosis (Stanford, 2010). Following implant placement, a layer of host plasma proteins assembles onto the implant surface and directs the interactions with surrounding cells (Anderson, Rodriguez, & Chang, 2008). Cells that adsorb to the protein layer release chemical mediators that result in the recruitment, migration and behavioural modulation of other cells (Anderson et al., 2008). Macrophages, key signaling cells that help organize the responses of other cells, are among the first cells to arrive at a wound site, and can perform a variety of functions, including elimination of debris and foreign pathogens, acting as antigen presenting cells to cytotoxic T cells, recruitment of other immune cells, tissue repair and wound healing (Lingen, 2001; J. W. Pollard, 2004). Macrophages exhibit plasticity, the ability to change their phenotype and function in response to the environment (Mantovani, Sica, & Locati, 2005). Two distinct phenotypes of macrophages have been described originally based on their functional similarities to Th1 and Th2 helper cells: M1,   2 associated with classical inflammatory responses and M2, an alternative phenotype associated with wound healing or immunoregulatory behaviour (Mantovani et al., 2005; Martinez, Helming, & Gordon, 2009; Mosser & Edwards, 2008). Many macrophages do not exhibit strictly M1 or M2 phenotypes but have characteristics of both states and lie on a spectrum between M1 and M2, illustrated in Figure 1.1 (Mosser & Edwards, 2008; Sica & Mantovani, 2012). It has been shown in-vitro and in-vivo that macrophages behave differently when cultured on smooth and rough surface topographies, exhibiting changes in cell morphology, gene and protein expression (Barth, Waterfield, & Brunette, 2013; Chehroudi et al., 2009; Refai, Textor, Brunette, & Waterfield, 2004; Rich & Harris, 1981; Takebe, Champagne, Offenbacher, Ishibashi, & Cooper, 2003; Tan, Qian, Rosado, Flood, & Cooper, 2006). Surface topography affects cell selection of pathways involved in producing pro-inflammatory cytokines (Ghrebi, Hamilton, Waterfield, & Brunette, 2013; Waterfield, Ali, Nahid, Kusano, & Brunette, 2010).  The aim of this thesis is to study the unstimulated macrophage response to a smooth and rough implant surface topography at the level of gene expression to further characterize the macrophage genotype and phenotype induced by changes in topography.     Figure 1.1: Spectrum of Macrophage Phenotypes    M1 M2 Wound Healing, immunoregulatory  Classically activated, inflammatory    3 1.1 Dental Implants  The introduction of dental implants in the mid-1960s revolutionized the dental specialty of prosthodontics, giving dentists the ability to replace teeth that they could not with conventional prostheses (Albrektsson & Wennerberg, 2005; Misch, 2007). Dental implants are also used to provide retention for prosthetic ears, eyes and noses in patients who may have lost these structures due to trauma, a congenital defect, or a result of surgical treatment for cancer (Albrektsson & Wennerberg, 2005; Beumer, Marunick, & Esposito, 2011). Replacing missing teeth with dental implants has become a well-accepted and popular treatment modality. In the United States of America alone, it is estimated that the number of implants placed per year by dentists is 5,505,720, and that in 2010, the sales of dental implants and dental implant prosthetics totalled 9.4 billion dollars (Slots, 2013).   The first evidence of the use of dental implants dates at least back to 200 AD in Europe, with the discovery of iron in the mandible of a well-preserved human body (Ratner, 2001a). Nacre, otherwise known as mother of pearl, was collected from seashells and used by the Mayans in 600 AD to replace missing teeth (Ratner, 2001b; Villar, Ba, & Mills, 2012). The first dental implants were poorly documented and generally unsuccessful (Albrektsson & Wennerberg, 2005). Implants were made in a variety of different designs and materials such as such as allografts or autografts of human teeth, sterilized bovine teeth, gold, platinum, porcelain, iridium, tantalum, stainless steel, carbon and sapphire (Ratner, 2001b; Villar et al., 2012).  Per-Ingvar Branemark is credited with treating the first edentulous patient with titanium dental implants in 1965 following his discovery of osseointegration in 1952 after implanting titanium   4 chambers in rabbit tibia and finding that he could not remove the titanium chambers at the end of his experiments (Albrektsson & Wennerberg, 2005; Br?nemark et al., 1977). Branemark defined osseointegration as the direct, structural and functional connection between organized vital bone and the surface of a load-bearing titanium implant (Br?nemark et al., 1977). Osseointegration essentially creates an ankylosis of the implant in bone, rendering it immobile but able to serve as a stable and reliable foundation for a prosthetic tooth. During the 1960s and 1970s, Branemark and his colleagues researched and developed a clinical placement protocol for dental implants (Albrektsson & Wennerberg, 2005). Clinical outcomes were favourable and implant survival rates were reported as 81% for the maxilla and 91% for the mandible after an observation period of 5 to 9 years (Adell, Lekholm, Rockler, & Br?nemark, 1981). Although Branemark?s studies initially received criticism for inconsistencies in reported data, his documented findings became accepted in the European scientific community and led to the introduction of osseointegration to North America at the Toronto conference in 1982 (Albrektsson & Wennerberg, 2005; James, Altman, Clem & Lozada, 1986). In 1990, Zarb and his colleagues published results of the Toronto study, a replica of the study performed by Branemark and his colleagues, and reported similar favourable surgical and prosthetic outcomes over an observation period of 4 to 9 years (Adell et al., 1981; Br?nemark et al., 1977; Zarb & Schmitt, 1990a; 1990b)  The Branemark implant used in these landmark studies was fabricated from commercially pure titanium, had a screw-shaped design and a machined surface containing microscopic grooves and pits (Misch, 2007). Although many different implant designs, materials, coatings and topographies have been developed to improve on the Branemark implant, a significant increase in bone to implant contact and implant survival is attributed to the use of rougher surface   5 topographies (Albrektsson & Wennerberg, 2005; Cochran, 1999). Recently published long-term clinical studies conducted by dental specialists report high implant survival rates with rough surface implants. Implants with SLA surfaces placed in 303 patients had a cumulative survival rate of 98.8% at 10 year follow-up (Buser et al., 2012). Long-term survival of titanium plasma sprayed implants is reported as 89.23% at 10 years and 82.94% at 16 years (Simonis, Dufour, & Tenenbaum, 2010). One should interpret these high survival rates with caution, as many published studies are conducted under optimal conditions in healthy patients by highly skilled dental practitioners. Patients with risk factors such as smoking, immunodeficiencies and a history of radiation treatment or chemotherapy, treated by general practitioners with more limited experience may have lower success rates (Goodacre, Bernal, Rungcharassaeng & Kan, 2003; Porter & Fraunhofer, 2005).  1.2 Biocompatibility  A biomaterial is a non-living substance or device designed to interact with and/or replace part of a living system (Ballanti et al., 2013; Lakes, 2007; Ratner, Hoffman, & Schoen, 2004; Tsirogianni, Moutsopoulos, & Moutsopoulos, 2006; D. F. Williams, 1987) Biocompatibility is often defined as the ability of a material to elicit an appropriate biological response when used for a specific application (Anusavice, 2003; Raghavendra, Wood, & Taylor, 2005; Ratner, 2001b; D. F. Williams, 2008). However, there is lack of consensus on the definition of biocompatibility. For example, Williams (2008) states that the sole requirement for biocompatibility is for the implanted device to ?do no harm? to the host tissues. Biocompatibility has also been defined by a material?s performance on International Organization for Standardization (ISO) tests, which include evaluations of genotoxicity, carcinogenicity,   6 cytotoxicity, systemic toxicity, irritation, sensitivity, interactions with blood and local effects after implantation (ISO, 2009; Kieswetter, Schwartz, Dean, & Boyan, 1996; Ratner, 2001b).  Titanium is considered a biocompatible material because it allows bone to heal right up to its surface in the absence of toxic, chronic inflammatory, allergic or mutagenic reactions (Anusavice, 2003; Harvey, Hill, & Bayat, 2013; Kieswetter et al., 1996; Raghavendra et al., 2005; Schenk & Buser, 1998). Numerous in-vivo animal and human clinical studies have demonstrated the biocompatibility of titanium over time (Carinci et al., 2003; Franz, Rammelt, Scharnweber, & Simon, 2011). According to Schenk and Buser (1998), titanium is bioinert, in that it does not release any toxic substances or cause adverse tissue reactions, but can be made bioactive, a material that can cause a favourable tissue reaction. A bioactive implant is designed to interact with its surrounding tissue to induce bone formation on its surface (Kieswetter et al., 1996; Kohavi, Klinger, Steinberg, & Sela, 1995; Sela, Badihi, & Rosen, 2007). Bioactivity of dental implants can be improved by altering chemical composition and surface topography (Davies, 2003; Harvey et al., 2013; Le Gu?hennec, Soueidan, Layrolle, & Amouriq, 2007; Villar et al., 2012).  1.3 Titanium Titanium, element number 22 on the periodic table of elements, is considered an extremely reactive transition metal, as it rapidly forms a tenacious oxide layer in the presence of air or water (Franz et al., 2011; Wang & Fenton, 1996). This approximately 3 to 5nm thick titanium dioxide (TiO2) layer renders titanium inert, electrochemically passive, resistant to corrosion and biocompatible (Franz et al., 2011; Ratner, 2001b; Villar et al., 2012; Wang & Fenton, 1996). Titanium is used in medicine and dentistry to replace hard tissue; for example, titanium is found   7 in artificial hip and knee joints, bone plates and fixation screws in bone fracture sites (Elias, Lima, Valiev, & Meyers, 2008; G. Mendon?a, Mendon?a, Arag?o, & Cooper, 2008; Stanford, 2010; D. F. Williams, 2008). The American Society of Testing and Materials (ASTM) classifies titanium materials into five grades with stronger mechanical properties increasing with each grade (Table 1.1). (Elias et al., 2008; G. Mendon?a et al., 2008; Niinomi, 1998; Park, Gemmell, & Davies, 2001). The elastic modulus of titanium (104 to 114 GPa) is closer to bone (17 to 25GPa) than other metals used to replace hard tissue such as stainless steel, cobalt chromium alloy which have moduli of 206GPa and 240GPa, respectively (Lingen, 2001; Niinomi, 1998). Large differences between elastic moduli of bone and implants may lead to micromotion at implant-bone interface and subsequent bone resorption and implant failure (Van Oosterwyck et al.; 1998).   There are four grades of commercially pure titanium (CpTi) that consist of 99% titanium with the remaining 1% composed of varying amounts of iron, oxygen, nitrogen, hydrogen and carbon that become incorporated into the material during the purification process (Villar et al., 2012). Grade five titanium, Ti-6Al-4V, is an alloy consisting of titanium, 4% vanadium and 6% aluminium (Villar et al., 2012; Wang & Fenton, 1996). Ti-6Al-4V has stronger mechanical properties than CpTi, but has the potential to elicit an undesired foreign body reaction if metal ions in the alloy are released into surrounding tissue via corrosion in bodily fluids (Villar et al., 2012). The most commonly used titanium materials in implant dentistry are Grade 4 CpTi and Ti-6Al-4V (Elias et al., 2008; Villar et al., 2012).      8 Titanium Type Tensile Strength (MPa) Yield Strength (MPa) Elastic Modulus (GPa) Grade 1 CpTi 240 175 103-107 Grade 2 CpTi 345 275 103-107 Grade 3 CpTi 450 380 103-107 Grade 4 CpTi 550 483 103-107 Grade 5 Ti-6Al-4V 860 795 101-114  Table 1.1: Mechanical Properties of Titanium Used in Implant Dentistry   1.4 Overview of Surface Topography Surface topography is defined by the degree of roughness of the surface and the orientation of its irregularities (Albrektsson & Wennerberg, 2004). Surface topography can influence osseointegration by affecting biological processes such as protein absorption, intracellular signalling, osteoblast attachment, cellular proliferation and differentiation, production of extracellular matrix, alkaline phosphatase activity, peri-implant bone formation and primary implant stability (Brunette, 1988; 2001; Harvey et al., 2013; Kulangara & Leong, 2009). Implant surface topography is one the six main factors affecting osseointegration, which also include biocompatibility of the implant material, the type of host bone, surgical technique, loading conditions and prosthetic design (Albrektsson, Zarb, & Worthington, 1986).   Surface roughness in the implant literature can also be classified by size into macro, micro and nano-sized topologies. Macro-retentive dental implant features such as screw threads engage bone and provide primary mechanical stability and compressive loading, and micro-retentive features are capable of influencing bone growth and remodelling (Stanford, 2010). Nano-sized features have been shown to alter cellular responses such as protein absorption, cell adhesion, proliferation and differentiation (G. Mendon?a et al., 2008). Macro-sized features are larger than 10um, micro-sized features are between 1 and 10um and nano-sized features have dimensions   9 between 1 to 100nm (Le Gu?hennec et al., 2007; Zaveri et al., 2010). Topographical features can be measured and characterized by a number of techniques including non-contact laser profilometry (LPM) for micron-sized features, and interference microscopy (IM) and stereo-scanning electron microscopy (stereo-SEM) for nano-sized features (Wieland, Textor, Spencer, & Brunette, 2001). Wavelength-dependent roughness evaluation using LPM, IM and stereo-SEM gives more accurate measurements and can differentiate features produced by different processes (Wieland et al., 2001). An excellent source of information detailing the chemical and topographic characterization of titanium surfaces can be found in the PhD thesis by M. Wieland (Wieland, 1999).   The height-descriptive parameters ?Sa? and ?Ra? are commonly used to depict the average roughness of a surface; the Sa parameter gives a measurement of the average roughness over a surface in 3 dimensions, whereas Ra parameter provides the measurement over a profile in 2 dimensions (Wennerberg & Albrektsson, 2010). The Sa value takes into account the height deviation when looking at the plane perpendicular to the surface, as well as deviations within the plane itself, or the texture of the surface (Hallgren, Reimers, Gold, & Wennerberg, 2001). Wennerberg (2004) believes that the Sa value provides a more accurate representation of the average surface roughness than the Ra value, and can differentiate between anisotropic or isotropic surfaces. An anisotropic surface is a surface with some orientation of the features, such as a machined implant surface, whereas an isotropic surface is one in which the features lack a clear orientation, seen with SLA surfaces (Wennerberg & Albrektsson, 2009). Two surfaces may have different morphologies but still share a common Ra value (Cooper, 2000). Nevertheless, Ra   10 and other roughness parameters such as Rq, Sq, RzDIN, Rt, St, and Sdr can be used describe surface topography (Table 1.2) (Wennerberg & Albrektsson, 2010).   Roughness parameter Description Ra Average roughness over all points on a profile Sa Average roughness over all points on a surface Rq Root mean square deviation of the profile  Sq Root mean square deviation of the surface  RzDIN Average of maximum peak to valley heights  Rt The maximum peak to valley height of a profile  St The maximum peak to valley height of a surface  Sdr Developed surface area, total surface area if peaks are flattened out   Table 1.2: Roughness Parameters Used to Characterize Implant Surface Topographies  Sdr is a measurement of the ?developed surface area?, which takes into account the number and height of the peaks within a surface and describes the surface enlargement after the surface is flattened out (Wennerberg & Albrektsson, 2010). Table 1.3 summarizes the Sa and Sdr values of implant surfaces from three major implant companies: Nobel Biocare?, Astra Tech? and Straumann? (Wennerberg & Albrektsson, 2010). Wennerberg and Albrektsson (2010) believe that a surface with a Sa of 1.5 micrometers and an Sdr of 50% promotes the strongest bone response, based on results from a series of animal studies conducted by their research group that evaluated bone response and removal torque of titanium implants with different topographies, but those values are best viewed as crude estimates of the optimal surfaces as so few topographies were examined.      11 Surface Sa (micrometers) Sdr (%) Machined Branemark (Nobel Biocare?) 0.9 34 TiUnite (Nobel Biocare?) 1.1 37 TiOblast (Astra Tech?) 1.1 31 OsseoSpeed (Astra Tech?) 1.4 37 SLA (Straumann?) 1.5 34 SLActive (Straumann?) 1.75 143  Table 1.3: Surface Topography of Implants from Three Major Companies    1.4.1 Surface Topography Fabrication Rough surface topographies on titanium dental implants can be produced by a myriad of techniques. One can create topography with additive procedures such as plasma spraying, producing convex or bumpy surfaces, or with subtractive procedures, such as blasting, etching and anodization, producing concave surfaces with pits or pores (Wennerberg & Albrektsson, 2009).   1.4.1.1 Plasma Spraying Implants may be surface coated with titanium or hydroxyapatite particles to increase the speed and amount of bone formation (Ong, Carnes, & Bessho, 2004). Fabrication involves injection of powders into a plasma torch at a high temperature and subsequently spraying of the particles onto implant surfaces, where they form a bumpy film of approximately 30um thickness with an average roughness of 7um (Le Gu?hennec et al., 2007; Wennerberg & Albrektsson, 2009).     12 1.4.1.2 Anodization  Anodization is carried out by immersing the implant surface in a hot solution of strong acids at a high current density or potential in a galvanic cell with the goal of thickening the oxide layer from 5nm to more than 1000nm (Le Gu?hennec et al., 2007). The thickness of the oxide layer is dependent on the amount of voltage applied (Classen, Lloberas, & Celeada, 2009; Mantovani, Biswas, Galdiero, Sica, & Locati, 2012; Puippe, 2003). According to Le Guehennec et al. (2007), the native oxide layer on the implant surface will be dissolved along current convection lines and thickened in other regions, creating macro or nano-pores along the surfaces.  1.4.1.3 Blasting  This process involves blasting alumina, titanium oxide or calcium phosphate particles at a high velocity with compressed air at the implant surface. The size of the roughness features depends on the size of particles used; for example, titanium oxide particles that are 25um in diameter produce surfaces with an average roughness between 1 to 2um and alumina oxide particles 250um in diameter produce 20 to 40um features (Le Gu?hennec et al., 2007; Schuler et al., 2009).  1.4.1.4 Etching  Strong acids, such as HCl, H2SO4, HNO3 and HF can be used to etch titanium to produce pits 0.5 to 2.0um in diameter (Le Gu?hennec et al., 2007). Acid etching is an isotropic subtractive process that removes particles to create surface features. Titanium is a corrosion-resistant metal and therefore requires strong acids to produce surface changes. Some surfaces are ?dual acid-  13 etched? and immersed in a mixture of two concentrated strong acids at a temperature of approximately 100?C (Le Gu?hennec et al., 2007).   1.4.1.5 SLA Surface Topography Etching can also be used in combination with blasting, to produce additional roughness within pits creating by the blasted particles. An example of a blasted and etched surface is the SLA (sandblasted, large-grit, acid-etched) surface by Straumann which uses Al2O3 beads 250um in size to blast the surface, which is then soaked in a hot solution of H2SO4 and HCl for several minutes (Taborelli et al., 1997). The fabrication processes produces two distinct topographies detectable with SEM: blasting with Al2O3 beads produces 20 to 40um pits and acid etching creates smaller 0.5 to 2um features within these pits (Wieland et al., 2001).   Figure 1.2: SEM Image of SLA Surface Topography  With kind permission from John Wiley and Sons. Barth, K.A., Waterfield, J.D. and D.M. Brunette. The effect of surface roughness on RAW264.7 macrophage phenotype. 2013. J. Biomed Mater Res Part A: 000-000    14 1.4.2 Nanotopography In the implant literature, nano-sized surface topographical features range from 1 to 100nm in size and are classified according to their form and structure as nanostructures, nanocrystals, nanocoatings, nanoparticles and nanofibres (G. Mendon?a et al., 2008). Specific and well-defined nanofeatures can be produced by a variety of methods including electron beam lithography, colloidal lithography, dip pen lithography and nanoimprinting (Kulangara & Leong, 2009).  Nanoscale roughness is an important determinant of protein interactions and can change the surface energy or wettability of the biomaterial. This is important in osseointegration because protein absorption occurs almost immediately after implantation and mediates cell attachment and proliferation (Harvey et al., 2013; G. Mendon?a et al., 2008). Most nanofeatures, are thought to produce randomly oriented features and occur within micro-sized surface topographies; therefore it is difficult to determine if effects are created by the nano-topography, the micro-topography, or both (Wennerberg & Albrektsson, 2010). Furthermore, some topographical modifications such as nanoparticle deposition, can alter the composition of the surface, and cause difficulties in determining whether effects are created by the topography or the surface chemistry (Lang et al., 2009).   1.5 Dental Implant Wound Healing  Implantation of medical devices involves a myriad of host reactions including local injury, coagulation, acute and chronic inflammation, granulation tissue formation, foreign body reaction and fibrous encapsulation of the device (Anderson et al., 2008). Wound healing around dental implants is unique in that it occurs in both hard and soft tissues and interacts with a variety of cells at different areas of the implant. Successful hard tissue healing results in osseointegration,   15 where bone formation occurs on the implant surface. Formation of a fibrous soft tissue capsule around the implant often leads to clinical failure (Villar et al., 2012). Soft tissue healing should only occur above bone where epithelial cells and connective tissue attach and form a tight marginal seal around the top of the implant, creating a physical barrier which functions to protect the implant-bone interface from bacteria in the oral environment and prevent downgrowth of epithelial cells which can lead to loosening, infection and loss of the implant (Kim, Murakami, Chehroudi, Textor, & Brunette, 2006; Rompen, Domken, Degidi, Farias Pontes, & Piattelli, 2006).   1.5.1 Stages of Wound Healing  The earliest stage of wound healing is coagulation, followed by a series of events that include protein adsorption, cellular adherence, proliferation and differentiation (Kieswetter et al., 1996). This stage is followed by de novo bone formation in which a calcifiable matrix is produced and followed by a prolonged stage of bone remodelling during which immature woven bone is gradually converted into mature cortical bone (Davies, 2003).  1.5.1.1 Coagulation Surgical placement of a dental implant creates a localized trauma that results in bleeding and activation of several biochemical pathways, including the coagulation cascade, kinin-kallikrein system and the complement system (Stanford, 2010). Activation of the extrinsic pathway of the coagulation cascade via tissue damage leads to the release of thrombin, which cleaves fibrinogen to form fibrin, producing a clot (Franz et al., 2011). The kinin-kallikrein system mediates the intrinsic pathway of the coagulation cascade, also leading to the formation of a fibrin clot and   16 produces bradykinin related peptides which promote vasodilation and increase vascular permeability (Kashuba, Bailey, Allsup, & Cawkwell, 2013). The complement system is part of the innate immune system and functions to protect the host by attracting neutrophils, stimulating other cells to perform phagocytosis and may also eliminate pathogens directly (Ballanti et al., 2013; Tsirogianni et al., 2006).  1.5.1.2 Protein Adsorption A newly placed dental implant is immediately exposed to serum proteins, mineral ions, sugars, lipids and cytokines produced by nearby immune cells (Raghavendra et al., 2005). Within minutes, the implant adsorbs proteins from these surrounding fluids and forms a thin proteinaceous coating on its surface (Kieswetter et al., 1996). Surface proteins allow cells to adsorb to the implant surface either by binding directly to the implant surface or to an arginine-glycine-aspartic acid (RGD) binding site on the surface proteins (Harvey et al., 2013; Kieswetter et al., 1996; Raghavendra et al., 2005). Immune cells can use adhesion receptors called integrins to attach to proteins such as fibrinogen, factor X, fibronectin and vitronectin (Franz et al., 2011). Fibronectin and vitronectin, proteins that contain RGD motifs, have been found to readily adsorb to implant surfaces in-vivo (Kieswetter et al., 1996; Kohavi et al., 1995; Sela et al., 2007). The composition, concentration and conformations of surface proteins can be altered by implant composition, chemistry and topography (Davies, 2003; Villar et al., 2012). Nano-sized features are capable of altering protein shape and can affect subsequent cell adherence, proliferation and differentiation (G. Mendon?a et al., 2008). Topographies in the micron-sized range have been shown to increase the recruitment and activation of platelets, which may result in the release of   17 granule contents and pro-coagulant microparticles and enhanced osteoconduction (Park et al., 2001).   1.5.1.3 Cellular Adherence, Proliferation and Differentiation As part of the innate immune response, neutrophils arrive at the implant site within the first 24 hours and phagocytose necrotic debris created by the surgical procedure (Lingen, 2001). Macrophages migrate within 24 to 48 hours and secrete a variety of cytokines and chemokines that attract other leukocytes or bone-forming cells, as well as growth factors that promote angiogenesis, the growth of new blood vessels (Lingen, 2001). Erythrocytes, neutrophils and macrophages reside in a fibrin coagulum in the space between the implant surface and pre-existing alveolar bone (Villar et al., 2012). According to Stanford (2010), macrophages play a complex role in wound healing, as they are capable of both promoting and dampening inflammation, due to their inherent plasticity and change in phenotype in response to their environment. Macrophages are involved in early wound healing responses and their accumulation on implant surfaces has been associated with larger amounts of bone formation in-vivo (Chehroudi et al., 2009). Angiogenesis precedes and promotes bone formation as it allows nutrients, inflammatory cells and oxygen access to the wound site (Villar et al., 2012). Growth factors, such as platelet derived growth factor (PDGF), transforming growth factor beta (TGF?), platelet-derived endothelial growth factor (PDEGF) and insulin-like growth factor 1 (IGF-1) released from activated platelets in the blood clot accelerate the wound healing process and attract fibroblast-like mesenchymal cells that adhere to the implant surface and differentiate into bone-forming cells called osteoblasts (Stanford, 2010; Villar et al., 2012).    18 1.5.1.4 De Novo Bone Formation and Remodelling Osteoinduction occurs when undifferentiated fibroblast-like mesenchymal cells differentiate into osteoblasts, and osteogenesis occurs when osteoblasts form new bone (Lingen, 2001). Smaller amounts of fully-differentiated osteoblasts may also be recruited to the implant site by cytokines and chemokines released by platelets and macrophages (Davies, 2003; Stanford, 2010). Immature woven bone is initially formed by osteoblasts as early as one week after placement, and is remodelled and replaced with mature trabecular bone over the next 1 to 2 years (Albrektsson, 2008; Berglundh, Abrahamsson, & Lang, 2003).   Osteogenesis may occur in two ways: contact osteogenesis, where de novo bone forms directly on the implant surface, and distant osteogenesis, where de novo bone forms on pre-existing alveolar bone (Davies, 2003; Misch, 2007). Implant surface topography has the potential to increase contact osteogenesis by affecting events in early implant wound healing (Davies, 2003) Increased contact osteogenesis can lead to accelerated healing and faster osseointegration, more predictable healing in challenging clinical situations such as placement of implants in poorer quality bone, and earlier or immediate loading of the implant (Davies, 2003).     19  Figure 1.3: Contact and Distant Osteogenesis  In distant osteogenesis, osteoblasts lie adjacent to pre-existing bone and form new bone in a direction towards the implant surface. In contact osteogenesis, osteoblasts produce bone directly on the implant surface. With kind permission from J.E. Davies. Davies, J.E. Understanding peri-implant endosseous healing. 2003. Journal of Dental Education. 67(8): 932-949. Available at http://www.ecf.utoronto.ca/~bonehead/  1.6 Macrophages  Macrophages are derived from blood monocytes, cells of the mononuclear phagocyte system that originate from the bone marrow and circulate in the blood for several days before entering tissues and differentiating into macrophages (Gordon & Taylor, 2005; Hume, 2006). Macrophages are capable of performing a wide variety of biological functions including antigen presentation, clearance of debris and pathogens, activation of the immune response, wound healing and fibrosis (Fairweather & Cihakova, 2009). There are many macrophage subtypes resident to different tissues in the body that perform specialized functions; for example macrophages may reside in bone as osteoclasts and exhibit bone resorbing activity, or as alveolar macrophages and function to identify and clear microorganisms, viruses and environmental particles in the lung (Gordon & Taylor, 2005; Mosser & Edwards, 2008).    20 1.6.1 Macrophage Polarization Macrophages can be activated by bacterial products, cytokines, chemokines and growth factors in the environment to adopt two polarized phenotypes, M1 or M2 (Sica & Mantovani, 2012; Varin & Gordon, 2009). These stimuli can arise from polymononuclear (PMN) cells, antigen-specific immune cells and from the macrophages themselves (Mosser & Edwards, 2008). LPS and INF? induce a classic inflammatory M1 phenotype, and IL-4 and IL-13 induce an alternative M2 phenotype associated with wound healing and immunoregulation (Classen et al., 2009; Mantovani et al., 2005; Mosser & Edwards, 2008). Increased production of proinflammatory cytokines, reactive nitrogen and oxygen intermediates, strong microbicidal and tumoricidal activity and promotion of Th1 responses are characteristic of the M1 phenotype (Sica & Mantovani, 2012). On the other hand, the M2 phenotype encompasses a wide range of characteristics and may be classified by function, into categories of host defence, wound healing and immune regulation (Mosser & Edwards, 2008). Th2 cytokines IL-4 and IL-13 can activate macrophages to adopt a general M2 phenotype characterized by increased expression of IL-10, IL-12, increased phagocytic activity and expression of scavenger, mannose and galactose receptors and cell surface specific markers Fizz1 and Ym1 (Biswas, Chittezhath, Shalova, & Lim, 2012). A summary of the most common characteristics of murine M1 and M2 polarized macrophages is given in Table 1.5.  Metabolism of glucose, iron and amino acids are distinctly different between M1 and M2 phenotypes. M1 cells often function in a hypoxic tissue environment, metabolizing glucose with the anaerobic glycolytic pathways, whereas M2 cells obtain their energy through oxidative glucose metabolism (Mantovani et al., 2012). Iron is essential for cell growth and is also   21 metabolized differently by M1 and M2 macrophage phenotypes. While M1 cells tend to sequester iron from the environment by expressing high levels of ferritin, a protein involved in iron storage, M2 cells release iron into the environment by upregulating their expression of ferroportin, a iron exporter protein (Biswas & Mantovani, 2012). An important distinguishing feature of murine M1 and M2 cells is the production of nitric oxide (NO) and polyamines. Classic inflammatory M1 cells have microbicidal activity and characteristically express NOS2 and produce NO to destroy ingested microbes (Mantovani et al., 2012). On the other hand, wound healing M2 cells express high levels of arginase I (Arg1), a protein that catalyses the production of polyamines needed for collagen synthesis, cell proliferation, tissue remodelling and fibrosis (Classen et al., 2009; Mantovani et al., 2012). Mantovani et al. (2012) believe that some macrophages adopt an ?M2-like? phenoype that exhibits some but not all the characteristics of M2 cells. Classification of phenotype by cytokine production remains controversial because the same cytokines, such as CCL2, also known as monocyte chemotactic protein-1 (MCP-1), may be produced by both phenotypes (Barth et al., 2013).                    22 Macrophage Phenotype Main inducers Marker Description M1 LPS, INF? IL-1? Interleukin 1 beta, pro-inflammatory cytokine IL-6, IL-23 Pro-inflammatory cytokines Il-12 Interleukin 12, pro-inflammatory cytokine, promotes Th1 differentiation CCL2 (MCP-1) Monocyte chemotaxis iNOS Produces NO for intracellular killing COX-2 Cyclooxygenase, produces prostaglandins TNF? Tumor necrosis factor alpha CXCL9, 10, 11 Th1 chemokines  M2    IL-4, IL-13  IL-10 Interleukin 10, promotes Th2 differentiation MRC1 Mannose receptor CCL2 (MCP-1) Monocyte chemotaxis CCL7 (MCP-3) Monocyte and macrophage chemotaxis Arg1 Arginase, upregulated following activation of STAT3 and STAT6 Ym1 Fizz1 Cell surface molecules upregulated following activation of STAT6  Table 1.4: Characteristics of Murine M1 and M2 Phenotypes      1.6.2 RAW264.7 Macrophage Cell Line The RAW264.7 cell line is derived from adult male BALB/c mice transformed by the Abelson murine leukemia virus (Hartley et al., 2008). It is the most commonly used mouse macrophage cell line in medical research and the primary experimental system for large-scale studies of signalling pathways (Hartley et al., 2008). RAW264.7 cells have properties of normal macrophages and are sensitive to macrophage activating agents such as lipopolysaccharide (LPS) However, RAW264.7 cells are murine macrophages that behave differently in some aspects than human macrophages and express unique markers of alternative activation (Murray & Wynn, 2011a). For example, the M1 marker NOS2, and M2 markers Arg1, Fizz1 and Ym2 markers are expressed by murine but not human macrophages (Murray & Wynn, 2011b). However,   23 conservation of function is apparent across these two species in that the different chemokines secreted by human and mouse macrophages tend to attract similar cell types (Martinez et al., 2009). Human macrophages can vary markedly due to variation among donor sources. The RAW264.7 macrophage cell line provides a stable, reproducible system to examine various phenomena such as macrophage proliferation, morphological changes, phenotype, cytokine production and signalling pathways and compare the effects of different surface topographies and treatments (Barth et al., 2013; Ghrebi et al., 2013; Refai et al., 2004; Waterfield et al., 2010).    1.6.3 Macrophages and Surface Topography Macrophages are among the first immune cells to migrate and enter the wound site, and are thought to be the key players in directing wound healing around dental implants (Stanford, 2010). Following implant placement, macrophages sterilize the wound site, remove necrotic debris created by the surgical procedure and mediate the early phase of inflammation, wound healing and bone formation by producing various growth factors, cytokines and chemokines (Alfarsi, Hamlet, & Ivanovski, 2013; Stanford, 2010). One may be able to control macrophage response by altering surface topography. Macrophages have been shown in-vitro and in-vivo to prefer rough surfaces, exhibiting a phenomenon termed ?rugophilia? (Rich & Harris, 1981; Salthouse, 1984). Chehroudi et al. (2009) investigated bone formation with machined, polished, finely blasted, coarsely blasted, acid etched, titanium plasma sprayed and SLA surfaces in-vivo and found that SLA surfaces were associated with the highest numbers of newly-recruited macrophages. The SLA surface also induced faster and greater amounts of bone formation (Chehroudi et al., 2009).    24 Several in-vitro studies have shown that surface topography affects macrophage gene and protein expression. Macrophages have been found to secrete pro-osteoinductive and pro-osteogenic growth factors as well as pro and anti-inflammatory cytokines when cultured on rougher titanium surfaces in-vitro (Barth et al., 2013; Paul et al., 2008; Takebe et al., 2003). Takebe et al. (2002) found higher levels of TGF? and BMP2 gene expression by murine J774A.1 macrophages cultured on a grit-blasted surface topography. TGF? is an anti-inflammatory cytokine and BMP2 is an osteogenic growth factor (Moura, Soares, Souza, & Zanetta-Barbosa, 2011; Takebe et al., 2003). Refai et al. (2004) studied responses of RAW264.7 macrophages at the protein level and observed downregulated production of pro-inflammatory cytokines IL-1? and IL-6, and chemokines MIP-1? and MCP-1 with unstimulated RAW264.7 macrophages on SLA surfaces. Paul et al. (2008) studied human macrophages on smooth, microstructured and nanotextured surfaces with microarrays and found that the microstructured surface, characterized by bumps with average heights of 1um that were 30um spaced apart, induced a specific gene expression pattern of both pro and anti-inflammatory molecules. Hamlet et al. (2011) studied gene expression of RAW264.7 macrophages with a DNA microarray and found downregulation of pro-inflammatory cytokines and chemokines on a nanoscale calcium phosphate modified surface. Also using RAW264.7 macrophages, Barth et al. (2013) did not find significant differences in gene expression of M1 and M2 markers NOS2 and Arg1, but performed further investigation at the protein level using ELISA and observed upregulation of chemokines MCP-1 (CCL2) and MIP-1? (CCL3) by both IL-4 stimulated macrophages and unstimulated macrophages on SLA surfaces. The above studies have shown that that unstimulated macrophages grown on rough topographies do not adopt the prototypical M1 or M2 phenotypes but exhibit mixtures of markers for each phenotype. Certain topographies such as the SLA are   25 associated with more macrophage recruitment and bone formation in-vivo, and in-vitro studies have provided some clues to explain these phenomena. The SLA surface may stimulate a unique macrophage phenotype that lies within a spectrum of extreme M1 or M2 phenotypes (see Figure 1.1), and is conducive to implant wound healing and osseointegration.   1.7 Gene Expression Every cell in a multicellular organism contains the same genomic sequence, but not every gene is expressed at any one time. Gene expression can be measured by examining the transcriptome, the complete set of mRNA produced at a given time, and evaluating the proteome, the complete set of proteins produced by a cell over a period of time (Adams, 2008).  Commonly used methods to measure gene expression include gel electrophoresis, DNA hybridization or southern blotting, RNA hybridization or northern blotting, microarray analysis and polymerase chain reaction (Watson, 2008). Gel electrophoresis involves separating DNA according to size in a gel matrix by using an electric field. Relative amounts of specific molecules, visualized through the use of fluorescent dyes, are detected as bands on the gel and represent accumulated DNA product (Watson, 2008). Hybridization takes advantage of the base pairing property of nucleic acids where complementary bases adenine and thymidine (uracil for RNA), and cytosine and guanine have a tendency to pair up with one another. Probes, complementary sequences to a gene of interest are used to detect the presence, for Northern and Southern blots, or quantify, for microarrays, target gene in a sample. Samples are labelled with a fluorescent dye and hybridized to the probe, and presence of the gene of interest results in detectable fluorescence on the blot or array (Watson, 2008).    26 1.7.1 Microarrays Microarrays can be thought of as large northern blots capable of analysing the expression of thousands of genes at the same time, allowing one to visualize a broad picture of the cellular response (D. J. Lockhart & Winzeler, 2000). A microarray consists of a glass slide onto which thousands of probes for specific genes are attached (Harrington, Rosenow, & Retief, 2000). There are many different types of arrays in use, but the most popular types are the spotted and high-density oligonucleotide arrays. Spotted, or printed oligonucleotide microarrays use longer pre-synthesized single or double-stranded DNA probes, called oligonucleotides, printed directly onto glass slides (Harrington et al., 2000; Sinicropi, Cronin, & Liu, 2006). Each oligonucleotide probe, approximately 30 to 80 oligonucleotides long, detects a specific gene of interest with high specificity due to its relatively long length and sequence that lacks homology with other gene transcripts (Sinicropi et al., 2006; Uzan, Villemin, & Garel, 2008). High-density, or in situ synthesized oligonucleotide arrays use a collection of different oligonucleotides with 11 perfectly matched (PM) probes and 11 mismatched (MM) sequences that are synthesized in situ on the microarray (Barth et al., 2013; Harrington et al., 2000; Sinicropi et al., 2006). These oligonucleotide probes have very short lengths (approximately 25 nucleotides long) and detect with perfect accuracy different sections of a specific gene. The MM probes contain a single base pair error to allow one to measure nonspecific hybridization (Harrington et al., 2000). Several commercial microarray platforms have been shown to give reliable and reproducible results. The microarray quality control (MAQC) project tested six commercially available microarray platforms and compared results with three additional gene expression measurement techniques and found that across all platforms, probes were accurately designed and correlation values between platforms ranged from 0.84 to 0.90 with qPCR validation (Shi et al., 2006)   27 1.7.2 Polymerase Chain Reaction Polymerase chain reaction (PCR) is a technique used to measure gene expression. Although there are numerous kinds of PCR techniques, real time quantitative polymerase chain reaction (qPCR) is considered the most sensitive, accurate, rapid and reproducible assay (Peinnequin et al., 2004). qPCR has two main advantages over other PCR techniques: 1) it does not require any post-PCR manipulation because all reactions are performed in closed tubes, minimizing the risk for potential errors due to contamination, and 2) results are normalized to a control gene that is constantly expressed in the cell, such as ?-actin, 18S or 28S ribosomal subunits or GAPDH (Heid, Stevens, Livak, & Williams, 1996). Normalizing to a control gene accounts for minor variations that may occur due to differences in the starting amount of RNA and efficiencies of the qPCR reaction (Giulietti et al., 2001). Minor differences in qPCR amplification efficiency can result in major errors as any discrepancy becomes exponentially larger with each cycle. If there are minor inefficiencies in the reaction, the amount of amplified cDNA end product may not correspond or relate to the amount of starting material and quantification of gene expression will be inaccurate (Giulietti et al., 2001; Pfaffl, 2001). The hallmark of qPCR is that is detects amplification product as the reaction progresses, combining the processes of amplification and detection into a single step (Wong & Medrano, 2005). Accurate quantification of amplification products is achieved by using a fluorescent dye that becomes incorporated in the amplicons and is detected over time. SYBR Green I is a commonly used dye that florescence when it is bound to double stranded DNA (Giulietti et al., 2001; Pfaffl, 2001). SYBR Green I dye offers a great degree of flexibility, as it can be used with any primer pair; however it may amplify nonspecific PCR products such as primer dimers, double stranded DNA molecules formed by primers that anneal to each other instead of the target   28 sequence (Giulietti et al., 2001). Examining the melting curves to ensure that a single product was the result of the qPCR reaction can rule out the possibility of a false positive result due to primer dimer formation (Giulietti et al., 2001).   The fluorescent signal emitted from the dye is proportional to the amount of DNA produced during each PCR cycle (Watson, 2008). The threshold cycle (Ct) is the cycle number at which the fluorescence of the sample exceeds a chosen threshold value based on the amount of background fluorescence (Bustin & Nolan, 2004). A lower Ct value is indicative of a greater amount of starting material, as fewer cycles are required to produce enough signal to exceed the threshold (Wong & Medrano, 2005). The comparative Ct method (2-??Ct) can be used to determine differences in gene expression. It presents differences in gene expression as fold changes and accounts for non-ideal qPCR amplification efficiencies but assumes similar efficiencies for the target and control gene (Wong & Medrano, 2005). The equation to calculate the relative expression or ratio of target to control gene is as follows:   Ratio = 2-[?Ctsample ? ?Ctcontrol] = 2-??Ct (Bustin et al., 2009)  qPCR is used as a validating tool for microarrays because it is thought to be more sensitive, precise and able to resolve smaller differences in gene expression (Sinicropi et al., 2006). In these experiments, qPCR is performed to check for potential false negative results due to its ability to detect very small amounts of cDNA, and to check for false positive results by using validated or specially designed primers specific to sequences of interest, and by repeating qPCR assays to demonstrate the reproducibility of the results.    29 Chapter  2: Statement of the Problem and Objectives 2.1 Problem There are currently several hundred implant systems available, of different material compositions, designs and surface topography (Albrektsson & Wennerberg, 2004; Brunette et al. 1988). The majority of dental implants used clinically today are made of grade IV cpTi or Ti-6Al-4V, have a screw design and a rough surface topography (Pye, Lockhart, Dawson, & Murray, 2009). Long-term survival rates of conventionally placed roughened dental implants are high, and reported in several studies as between 90-95% over 10 years (Buser et al., 2012; Simonis et al., 2010). Despite high success rates, implant failures still occur and are thought to be due to the inability to establish or maintain osseointegration (M. Esposito, Hirsch, Lekholm, & Thomsen, 1998). Many different factors, including surgical technique, implant design, prosthesis design, local host factors such as the quality and quantity of bone, and systemic host factors such as smoking, systemic illnesses and previous radiation therapy have been shown to be risk factors for implant loss (Goodacre et al., 2003; Pye et al., 2009). Factors affecting osseointegration that are under control of the dentist are the selection of the patients and the appropriate type of implant, as well as his or her surgical technique. Implant surface topography affects the degree of protein distribution, subsequent protein-cell interactions and implant wound healing (Harvey et al., 2013). Rough implant surface topographies are thought to perform better than turned or machined surfaces in-vivo due to their ability to enhance osteoconduction and induce more bone formation (Berglundh, Abrahamsson, Albouy, & Lindhe, 2007; Davies, 2003; Grassi et al., 2006). Macrophages grown on SLA surfaces have been shown to be associated with greater and faster amounts of bone formation in-vivo but the biological mechanisms leading towards osseointegration are poorly understood (Berglundh et al., 2007; Chehroudi et al., 2009). Use of   30 high-throughput genetic screening techniques can help characterize the macrophage response to different surface topographies. Microarray analyses will be used to explore which genes are being down or up-regulated and give clues to signalling pathways and cellular behaviour towards implant surfaces.    2.2 Hypotheses Surface topography influences changes in gene expression in RAW264.7 macrophages, in particular, between cells grown on polished and SLA surface topographies after 1 and 5 days.   2.3 Rationale Many implant surface designs were developed as a result of trial and error optimization, without knowledge of how cells would respond to the surfaces (Brunette, 2001). Current research has focused on what kinds of cells adhere and interact with surface features. Of particular interest are the macrophages, cells essential for the orchestration of wound healing via their production of growth factors, cytokines and chemokines (Lingen, 2001; Tan et al., 2006). The purpose of these studies is to determine which specific genes are differentially expressed on SLA compared to polished surfaces. Previous reports on gene and protein expression of murine macrophages grown in-vitro on smooth and rough surface topographies focused on pre-selected genes. My studies, however, probe the entire mouse genome and have the potential to identify patterns of gene responses and discover upregulated or downregulated genes which hither to have not been postulated as being influenced by topography.     31 2.4 Specific Aims 1. To quantify, using 4',6-diamidino-2-phenylindole (DAPI) fluorescent staining, the numbers of RAW264.7 macrophage cells present on P (polished) and SLA (sand-blasted and acid-etched) surface topographies after 1 and 5 days of growth  2. To determine which RAW264.7 macrophage genes are significantly up or downregulated on SLA compared to polished surfaced topographies after 1 and 5 days of growth  3. To determine whether RAW264.7 macrophages exhibit responses to molecules known to affect pathway selection in native macrophages 4. To document that the RAW264.7 cell line identified as macrophages on the basis of their morphology and phagocytic activity, do in fact exhibit gene expression patterns characteristic of macrophages   2.5 Significance Results from these exploratory studies will help us characterize the phenotype of macrophages grown in-vitro on the SLA surface topography.  Use of a whole genome microarray will provide a large list of genes that may play a role in macrophage response to rough topographies. These genes, which will be identified by their significant up or downregulation by the SLA surface topography will become targets of future investigations that could clarify the mechanisms whereby rough surfaces enhance osteogenesis.       32 Chapter  3: Materials and Methods  3.1 Substrata Grade 2 commercially pure titanium (cpTi) disks, 15mm in diameter and 1mm in width, of a smooth and rough surface topography were provided by Institute Straumann (Waldenburg, Switzerland). The smooth topography (polished, P) was produced by mechanical polishing, and the rough topography (sandblasted and acid-etched, SLA) was produced via sandblasting with Al2O3 beads approximately 250 micrometers in size followed by etching in a hot solution of H2SO4 and HCl (Wieland et al., 2001).   Impressions were made of the provided polished and SLA titanium disks with light-bodied consistency polyvinyl siloxane impression material (Provil Light, Dormagen, HK, Germany). These silicone impressions were used to cast the epoxy-resin (EPO-TEK 302-3; Epoxy Technology, Bellerica, MA) replicas of the titanium disks. After mixing, the epoxy resin was left to set for 24 hours in a fume hood before being placed in an oven with a temperature of 58?C for 7 days to ensure complete curing of the material and to minimize residual unreacted monomer content. Cured epoxy replicas were cleaned with a 50:50 mix of distilled water and 7X O-Matic Laboratory Detergent (MP Biomedicals, Irvine, CA) and placed in an ultrasonicator for 30 minutes. The epoxy replicas were then rinsed with distilled water ten times to dilute and remove remnants of the detergent, and left to dry in a sterile petri dish in a tissue-culture laminar-flow fume hood overnight.    Several polished and SLA epoxy surfaces were coated with titanium to use in an experiment comparing differences in gene expression between materials with the same topography (Section   33 4.6). Epoxy replicas were sputter-coated with 50nm of titanium using a Randex 3140 Sputtering System (Palo Alto, CA, USA).   Prior to cell culture, the epoxy replicas were treated for 4 minutes in an argon glow-discharge chamber. Glow-discharging the epoxy replicas with argon gas increases the surface energy of the material to facilitate cell attachment and proliferation (Aebi & Pollard, 1987).    3.2 Cell Culture The cells used in these studies were RAW264.7 macrophages, obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). When not in use, RAW264.7 cells were cryopreserved in liquid nitrogen vapor phase in complete growth medium supplemented with 5% (v/v) dimethyl sulfoxide (DMSO), a cryoprotectant. Cells were revived by first slowly bringing them to room temperature (20?C) and culturing in complete growth medium made of Dulbecco?s Modified Eagle?s Medium (Stemcell Technologies, Vancouver, Canada) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Cansera, Etobicoke, Canada) and an antibiotic mixture containing 66ug/ml penicillin G (Sigma-Aldrich), 33ug/ml gentamycin sulfate (Gibco, New York, USA) and 0.7ug/ml amphotericin B (Gibco, New York, USA) on tissue culture plastic 75cm2 flasks at 37?C in a humidified atmosphere of 90% air and 5% CO2  (Barth et al., 2013). Media was changed every 2 days and cells were subcultured at a ratio of 1:8 using a cell scraper or Detachin? (Genlantis, San Diego, CA, USA) a protease containing cell detachment solution before reaching confluence, approximately every 3 days. Cells used for experiments were passaged a minimum of 3 times to ensure that DMSO was no longer present in the cell   34 culture, and no more than 15 times to avoid detrimental effects of high passage numbers (ChangLiu & Woloschak, 1997).  3.3 Proliferation Study The numbers of macrophages on a surface were quantified with DAPI staining. RAW264.7 cells were grown on 36 polished and SLA epoxy surfaces at concentrations of 500,000 cells/ml, 200,000 cells/ml and 100,000 cells/ml for 1 day of growth, and at concentrations of 20,000 cells/ml, 10,000 cells/ml and 5,000 cells/ml for 5 days of growth (Figures 3.1 and 3.2).  These concentrations were selected based on previous literature (Ghrebi et al., 2013; Refai et al., 2004; Takebe et al., 2003; Tan et al., 2006). Initial cell concentrations to prepare dilutions were calculated using an electronic cell counter (Beckman Coulter, Pasadena, CA, USA). Cells were prepared for DAPI staining by first washing the surfaces with phosphate buffered saline (PBS) followed by fixation with cold 70% ethanol for 20 minutes. Following two washes with PBS, 0.5ng/ml of DAPI solution was added to stain nuclei for 30 minutes. Samples were washed five times with PBS and mounted onto labelled glass slides and covered with plastic coverslips. Glass slides containing DAPI stained surfaces were examined with a fluorescent microscope using the UV-2E-C filter at a 10x magnification. At least five images in different locations on the surface were taken with a digital camera connected to the microscope. ImageJ (Image Processing and Analysis in Java) software was used to count the number of cells present on the surfaces.   Counts were averaged and used in combination with visual analysis of the images for cell confluency to determine the ideal cell seeding concentration. The discs used had radii of 7.5mm and a total area of 176.63mm2, calculated using the formula ?r2. The area of the field of view   35 was 4.90mm2, calculated by multiplying the height (2.54mm) by the width (1.93mm). By dividing the total area of the disc by the area of the field of view, the average number of cells per disc could be calculated by multiplying the average number of cells in a field of view by a factor of 36.  (176.63 mm2/4.90 mm2 = 36). This study was repeated twice using two different passages of RAW264.7 macrophages.    Figure 3.1: Design for 1-Day Proliferation Study    36  Figure 3.2: Design for 5-Day Proliferation Study   3.4 Gene Expression Studies Effects of surface topography on gene expression of RAW264.7 macrophages were analysed using a protease microarray, the CLIP-CHIP?, and a whole genome microarray, the Agilent SurePrint G3 mouse gene expression microarray. Following analysis of the microarrays, quantitative real time polymerase chain reaction (qPCR) was employed to confirm that the same genes were differentially expressed on P and SLA surfaces.  3.4.1 RNA Extraction RNA was extracted from RAW264.7 macrophages using the RNeasy Mini Kit (Qiagen, Valencia, CA), following manufacturer?s instructions (Qiagen, 2010). Cells membranes were disrupted to release their RNA content with a lysis buffer containing guanidine thiocyanate. Cells were subsequently homogenized using QIAshredder spin columns, which shear high molecular   37 weight DNA and cellular components, creating solutions with reduced but uniform viscosities. 70% ethanol was added to the homogenized cell lysates to precipitate RNA. Each solution was mixed with a pipette tip and transferred to an RNAeasy spin column which was centrifuged and washed with a buffer containing a small amount of guanidine salt. The optional on-column DNase digestion step was carried out for 15 minutes to ensure removal of genomic DNA. The RNA was subsequently washed with buffer RW1, a wash buffer containing salts, and buffer RPE, a wash buffer containing ethanol, before being eluted from the column with water.   3.4.2 RNA Quality Assessment  Immediately following RNA extraction, quality and quantity was assessed with the Nanodrop ND-1000 Spectrophotometer (Thermo Scientific, Hudson, NH). This instrument measures the absorbance of all molecules in a sample. RNA and DNA absorb at a wavelength of 260nm, and carbohydrates and phenol absorb at a wavelength of 230nm. RNA with 260/280 ratios and 260/230 ratios greater than 1.8 were considered pure, without significant protein or phenol contamination (Nolan, Hands, & Bustin, 2006).   RNA used for microarray experiments was further assessed using a 2100 Bioanalyzer (Agilent Technologies). The bioanalyzer is a microfluidic capillary electrophoresis system that uses a software algorithm to evaluate RNA integrity (Bustin et al., 2009; Nolan et al., 2006; Schroeder, Mueller, & Stocker, 2006). RNA samples are first labelled with a dye that fluoresces when exposed to a laser. The bioanalyzer separates molecules of RNA according to molecule weight and subsequently detects each separated band of RNA with the laser. The amount of fluorescence correlates with the amount of RNA present and results are summarized in   38 electropherograms with RNA integrity numbers (RINs) based on its characteristics (Nolan et al., 2006; Schroeder et al., 2006)  3.4.3 cDNA Production cDNA was made from RNA using the iScript cDNA synthesis kit (Bio-Rad Laboratories, Hercules, CA) and a Mastercycler gradient thermal cycler (Eppendorf AG, Hamburg, Germany) according to manufacturer?s instructions. A total of 1ug of RNA was used for a 20ul reaction. 15ul of RNA and RNase-free water were added to 0.2ml tubes first, followed by 4ul of a mix of oligo deoxy-thymine (oligo-dT) primers and 1ul of reverse transcriptase enzyme. No reverse transcriptase and water controls were generated with each set of extracted RNA.   3.4.4 Microarray 3.4.4.1 Whole Genome Microarray Whole genome microarray experiments were performed with 16 microarrays from the Agilent mouse gene expression 4x44K v2 microarray kit, following manufacturer?s instructions (Agilent Technologies, 2008). Each microarray contains 39,340 pre-synthetized 60-mer oligonucleotide probes for the entire mouse genome (Agilent Technologies, 2008). There are four microarrays on each glass slide and RNA samples were randomized across the slides so that none of the biological replicates within one treatment group would be hybridized to the same slide to take into account variation between slides (see Table 3.1). Extracted RNA from RAW264.7 macrophages grown on polished and SLA surfaces for 1 and 5 days was used, as well as RNA from LPS and IL-4 stimulated RAW264.7 macrophages grown on polished surfaces. 3 biological   39 replicates were included for each polished and SLA condition whereas only 1 sample of each IL-4 and LPS stimulated condition was used.   Slide #1 Slide #2 Slide #3 Slide #4 Polished #3, Day 5  SLA #3, Day 1 Polished #3, Day 1 IL-4-stimulated, Day 5 SLA #3, Day 5 Polished #2, Day 1 LPS-stimulated, Day 5 Polished #2, Day 5 SLA #1, Day 1 Polished #1, Day 5 Polished #1, Day 1 LPS-stimulated, Day 1 IL-4-stimulated, Day 1 SLA #1, Day 5 SLA #2, Day 1 SLA #2, Day 5  Table 3.1: Randomized Arrangement of Samples on Microarray Slides  Samples were randomized across four glass slides containing four microarrays each to minimize effects of variations between slides    RNA samples were first prepared using the Low Input Quick Amp Labelling Kit (Agilent Technologies, Santa Clara, CA, USA) which uses an RNA polymerase to amplify and incorporate the dye, cyanine 3-CTP (Cy3) into RNA, generating labelled complementary RNA (cRNA) labelled with Cy3. The One-Color RNA Spike-in Kit (Agilent Technologies, Santa Clara, CA), a set of positive control transcripts designed to anneal to complementary probes on the microarray chip was used to generate quality control reports for each microarray. The RNeasy Mini Kit (Qiagen, Valencia, CA) as described in section 3.4.1, was used to purify amplified cRNA samples. A nanodrop ND-1000 Spectrophotometer (Thermo Scientific, Hudson, NH) was used to quantitate cRNA and ensure adequate quality (A260/280 > 1.8), yield (>1.65ug) and specific activity (>9.0pmol of Cy3 per ug of cRNA). cRNA samples were loaded onto microarray slides (Figure 3.3) and hybridized using SureHyb hybridization chambers (Agilent Technologies) and the Agilent hybridization oven. After washing the arrays to remove   40 non-specifically bound cRNAs, scanning, background correction, normalization and analysis was performed using Agilent DNA microarray scanner with Feature Extraction Software. Data analysis was performed using GeneSpring v12 (Agilent Technologies) and Ingenuity Pathway Analysis (Ingenuity) software. More detailed protocol descriptions can be found in Agilent user guides for the one-color microarray based gene expression analysis, hybridization oven and SureScan microarray scanner system (Agilent Technologies, 2008).   3.4.4.2 CLIP-CHIP? Microarray The CLIP-CHIP? microarray, developed in Dr. C. Overall?s laboratory, focuses on analysis of mRNA transcripts of proteases, non-proteolytic homologs and protease inhibitor genes (Kappelhoff, Wilson, & Overall, 2008). The mouse CLIP-CHIP? was used for these experiments. Extracted RNA from RAW264.7 macrophages grown on polished and SLA surfaces for 5 days was used, as this experiment had been performed previously with 1 day samples. RNA from samples was labelled with Cy3, amplified and converted to cRNA. A mix of RNA from all samples was labelled with Cy5. cRNA was hybridized to the microarrays which were scanned by a MWG 428 microarray scanner and data was analysed using ImaGene (BioDiscovery). A complete description of the CLIP-CHIP? protocol is given by Kappelhoff (2010). (Kappelhoff, auf dem Keller, & Overall, 2010)  3.4.5 Polymerase Chain Reaction qPCR was performed using a RG-3000 Rotor Gene PCR cycler (Corbett Research) and iQ SYBR Green Supermix Kit (Bio-Rad Laboratories) using Glyceraldehyde-3-Phosphate Dehydrogense (GAPDH), a common housekeeping gene to normalize cDNA samples. qPCR   41 was performed using 3 biological and 3 technical replicates, and 3 runs were performed for each gene to confirm reproducibility and reliability of the results. Biological replicates are derived from RAW264.7 macrophages grown on different surfaces of the same type under the same conditions, whereas technical replicates involve using cells from the same surface. Biological variability is attributed to inherent differences between individual cell culture samples, and technical variability can arise from using improperly calibrated instruments, human pipetting errors and during sample extraction and purification processes (S. Taylor, Wakem, Dijkman, Alsarraj, & Nguyen, 2010). A no reverse transcriptase (nRT) control was included to ensure the absence of contaminating genomic DNA, and negative water control were included in triplicate with each reaction. Machine settings are as follows: initial denaturation was performed at 95? for 5 minutes, followed by 40 cycles at 95? for 10 seconds and 60? for 60 seconds.   3.4.5.1 Primers 7 genes were investigated from the CLIP-CHIP? microarray results (CysC, CysB, ADAMTS3, HSP, MMP1a, MMP7, MMP12) and 6 genes were investigated from the whole gnome microarray results (CCL4, CCL7, CTNND1, C1Qa, ADM and CCL13). Primers were either found in the literature or designed using NCBI Primer-Blast, IDT DNA oligo analyser and mfold analyser from the RNA Institute of Albany to have a melting temperature between 55 and 60?C, an amplicon size less than 250 base pairs and absence of dimerization capability, hairpin formation and secondary priming sites (Tables 3.1-3.2).  The comparative CT (2-??Ct) method was used to analyze qPCT data (Schmittgen & Livak, 2008).     42 Primer Sequence Orientation Source Cys C 5?-CCA TGA CCA GCC CCA TCT GAT-3? Forward Designed 5?-CAC AAG TAA GGA ACA GTC TGC -3? Reverse Cys B 5?-GTC TTC AGC TTC TCC GTG CT-3? Forward Designed 5?-GTT GGT GCC AGC CAC TAT CT-3? Reverse ADAMTS3  5?-TGG TGT TCC TGT CAC TTT GG-3? Forward Designed 5?-TCA CCA GCT CAT ACT CTC TG-3? Reverse HSP 5?-CTC GGC TTT CCC GTC AAG AT-3? Forward Designed 5?-GTC CAG GGC ATC TGA AGC AT-3? Reverse MMP1a 5?-GTC TTT GAG GAG GAA GGC GAT ATT-3? Forward (Wells et al., 2003) 5?-AGT TAG GTC CAT CAA ATG GGT TGT T-3? Reverse MMP7 5?-TGGAGTGCCAGATGTTGCAG-3? Forward (Haro et al., 2000) 5?-TTTCCATATAACTTCTGAATGCCT-3? Reverse MMP12 5?- GCT AGA AGC AAC TGG GCA AC-3? Forward (Peinnequin et al., 2004) 5?- ACC GCT TCA TCC ATC TTG AC-3? Reverse GAPDH  5?-TGG CAA AGT GGA GAT TGT TGC C-3? Forward (Barth et al., 2013) 5?-AAG ATG GTG ATG GGC TTC CCG-3? Reverse  Table 3.1: Primers used for qPCR validation of CLIP CHIP microarray  Primer Sequence Orientation Source CCL4  5?-TTC TCT TAC ACC TCC CGG CAG-3? Forward (Carvalho-Gaspar, Billing, Spriewald, & Wood, 2005) 5?-GTA CTC AGT GAC CCA GGG CTC A-3? Reverse CCL7  5?-AGC TAC AGA AGG ATC ACC AG-3?  Forward (Michalec et al., 2002) 5?-CAC ATT CCT ACA GAC AGC TC-3?  Reverse CTNND1  5?- GGG TCT CAC CAC AAG ATG CC-3? Forward (Bartlett, Dobeck, Tye, & Perez-Moreno, 2010) 5? TCC TGG GGT CCG TTG AGT TT-3? Reverse C1Qa  5?-GCT GAC CAT GAC CCT AGT ATG GA-3? Forward (Lattin et al., 2009) 5?-GGC GGC CAG GAT TTC C-3? Reverse ADM  5?-CGA AAG AAG TGG AAT AAG TGG G-3? Forward (Uzan et al., 2008) 5?-GTT CAT GCT CTG GCG GTA GCG-3? Reverse CCL13  5?-GGC TGG AGC ATC CAG TTT GA-3? Forward Designed 5?-CTC CAT GGT GTC CCT GCC TC-3? Reverse GAPDH  5?-TGG CAA AGT GGA GAT TGT TGC C-3? Forward (Barth et al., 2013) 5?-AAG ATG GTG ATG GGC TTC CCG-3? Reverse  Table 3.2: Primers used for qPCR validation of whole genome microarray   43 3.4.5.2 qPCR Efficiency  The efficiency of qPCR reactions was calculated to ensure that all reaction conditions, buffers and primers were optimized and that these experiments would be reproducible (Taylor et al., 2010; Tichopad, Dzidic, & Pfaffl, 2002). Standard curves were generated using a series of dilutions of sample cDNA over six points, by plotting the cycle number of CT against the log of the concentration of cDNA and obtaining the best fit line (Pfaffl, 2001). The slope of the line was used to calculate the primer efficiency using the formula Efficiency = 10(-1/slope) -1 (Bustin et al., 2009) The ideal slope is -3.322, giving a primer efficiency of 1, indicative of a doubling of material for each cycle in the reaction (Bustin et al., 2009; Taylor et al. 2010) Optimization of the qPCR assay was determined by evaluating the equation of the linear regression line, and coefficient of determination (r2) (Taylor et al., 2010).   3.5 Statistical Analysis One-way ANOVA was performed using StatPlus for MacOS in Microsoft Excel 2011 to compare means of cell counts from the proliferation study. Following the general practice at LAGA, student?s t tests using GeneSpring version 12 (Agilent Technologies) were used to test for significant differences in gene expression between macrophages grown on polished and SLA surfaces and to compare gene expression profiles with LPS and IL-4 controls. All tests were performed on log transformed normalized data. One sample t-test Student?s t-test calculations for LPS and IL-4 controls were done using the statistics environment R version 2.13.0. qPCR data analysis was performed using StatPlus in Microsoft excel 2011. For all the tests, significance was set at a level of P ? 0.05, and a meaningful difference in gene expression was considered if there was a fold change greater or less than 2.    44  Two versions of Ingenuity Pathway Analysis (IPA) were used to perform network, functional and canonical pathway analysis for the whole genome microarray:  1) Build version 172788, content version 14197757, Release date: August 11th 2012 2) Build version 212183, content version 14855783, Release date: February 5th, 2013       45 Chapter  4: Results 4.1 Proliferation Study  Seeding concentrations of 1x105 cells/ml and 5x105 cells/ml have been used to culture murine macrophages on titanium and titanium coated epoxy discs for 24 hours (Refai et al., 2004; Tan et al., 2006). Observation of the effects of surface topography over a period of 5 days on RAW264.7 macrophage number was previously studied by Ghrebi et al. (2013) with an initial seeding concentration of 2x104 cells/ml. It was decided to create high, medium and low concentration categories based on the findings by Refai et al. (2004) and Ghrebi et al. (2013) to determine the optimal seeding concentration for the macrophages for 1 and 5 days of growth. RAW264.7 cells on polished surfaces were also examined under a light microscope to check for cell viability and confluence before DAPI staining and counting (Figure 4.1). The goal was to have the number of cells on the surface be below confluence in order to observe the effects of topography relatively independent of the influence of cell to cell contact   Figures 4.2 and 4.3, and tables 4.1 to 4.4 illustrate the average number of cells counted per field of view observed with the 10x fluorescent microscope as well as a calculation of the average number of cells present on the disc for 1 and 5 days of growth. 1-way ANOVA confirmed that the cell counts from the high, medium and low concentration were significantly different for each surface and time point with p<0.05.    46  Figure 4.1: Phase Contrast Micrograph of RAW264.7 Macrophages on a P Surface   Figure 4.2: Cell Counts for 1 Day Proliferation Study RAW264.7 macrophages were cultured on polished and SLA surfaces for 1 day at 3 different seeding concentrations. One-way ANOVA demonstrated that cell numbers differed significantly among seeding concentrations.   0	 ?20,000	 ?40,000	 ?60,000	 ?80,000	 ?100,000	 ?120,000	 ?140,000	 ?5x105	 ?cells/ml	 ? 2x105	 ?cells/ml	 ? 1x105	 ?cells/ml	 ?#	 ?of	 ?cells	 ?Seeding	 ?Density	 ?Polished	 ?1	 ?day	 ? SLA	 ?1	 ?Day	 ?  47  Figure 4.3: Cell Counts for 5 Day Proliferation Study  RAW264.7 macrophages were cultured on polished and SLA surfaces for 5 days at 3 different seeding concentrations. One-way ANOVA demonstrated that cell numbers differed significantly among seeding concentrations.  Seeding Concentration Average # cells per field of view (2.54 x 1.93mm) Standard deviation Average # cells on the epoxy disc 5x105 cells/ml 3606 377 129,816 2x105 cells/ml 1549 145 55,764 1x105 cells/ml 677 76 24,372   Table 4.1: RAW264.7 Macrophages Grown on Polished Surface Topography for 1 Day Averages were calculated from 3 replicates per seeding concentration   Seeding Concentration Average # cells per field of view (2.54 x 1.93mm) Standard deviation Average # cells on the epoxy disc 5x105 cells/ml 2651 674 95,436 2x105 cells/ml 1261 224 45,396 1x105 cells/ml 425 67 15,300  Table 4.2: RAW264.7 Macrophages Grown on SLA Surface Topography for 1 Day Averages were calculated from 3 replicates per seeding concentration   0	 ?10,000	 ?20,000	 ?30,000	 ?40,000	 ?50,000	 ?60,000	 ?70,000	 ?2x104	 ?cells/ml	 ? 1x104	 ?cells/ml	 ? 5x103	 ?cells/ml	 ?#	 ?of	 ?cells	 ?Seeding	 ?Density	 ?Polished	 ?5	 ?day	 ? SLA	 ?5	 ?day	 ?  48  Seeding Concentration Average # cells per field of view (2.54 x 1.93mm) Standard deviation Average # cells on the epoxy disc 2x104 cells/ml 1703 353 61,308 1x104 cells/ml 1273 658 45,828 5x103 cells/ml 214 91 7704  Table 4.3: RAW264.7 Macrophages Grown on P Surface Topography for 5 Days Averages were calculated from 3 replicates per seeding concentration   Seeding Concentration Average # cells per field of view (2.54 x 1.93mm) Standard deviation Average # cells on the epoxy disc 2x104 cells/ml 1579 1579 56,844 1x104 cells/ml 513 513 18,468 5x103 cells/ml 37 32 1332  Table 4.4: RAW264.7 Macrophages Grown on SLA Surface Topography for 5 Days Averages were calculated from 3 replicates per seeding concentration      Figure 4.4: DAPI-stained Cells on P Surfaces after 1 Day, Seeded at 5x105 cells/ml   49  Figure 4.5: DAPI-stained Cells on SLA Surfaces after 1 Day, Seeded at 5x105 cells/ml     Polished SLA  Figure 4.6: DAPI-stained Cells on P and SLA Surfaces for 1 Day, Seeded at 2x105 cells/ml    50   Polished SLA  Figure 4.7: DAPI-stained Cells on P and SLA Surfaces for 1 Day, Seeded at 1x104 cells/ml    Polished SLA   Figure 4.8: DAPI-stained Cells on P and SLA Surfaces for 5 Days, Seeded at 2x104 cells/ml    Polished SLA   Figure 4.9: DAPI-stained Cells on P and SLA Surfaces for 5 Days, Seeded at 1x104 cells/ml   51   Polished SLA   Figure 4.10: DAPI-stained Cells on P and SLA Surfaces for 5 Days, Seeded at 5x103 cells/ml  The best seeding concentration was determined to be 2x105 cells/ml for 1 day and 2.0x104 cells/ml for 5 days of growth. Approximately 50,000 RAW264.7 macrophages would be expected on both polished and SLA surfaces with these initial seeding concentrations. This concentration allows for sufficient cells to be available for RNA isolation and relatively few cells touching or overlapping one another so that measured changes in gene expression would be due to effects of surface topography rather than cell to cell contact. During the course of study, it was observed that a seeding concentration of and 2.0x104 cells/ml for 5 days resulted in excessive cellular proliferation, evidenced by the dense appearance of cells on the surface and the change in colour of the cell media to a yellow hue, indicating an acidic environment with a low pH (ATCC, 2012). It was decided to use a modified lower seeding concentration of 1.5x104 cells/ml was for 5 day experiments.     52 4.2 RNA Quality  RNA quality was assessed immediately after extraction using the NanoDrop ND-1000 spectrophotometer (Thermo Scientific). RNA used for all experiments had 260/280 and 260/230 ratios greater than 1.8, indicating a sufficiently pure sample free of genomic DNA and phenol contaminants (Figure 4.8). Integrity of the RNA for the microarray experiments was further assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA). Accepted RNA produced RNA integrity numbers (RINs) greater than 9.0 and electropherograms with distinct marker and 28S and 18S peaks (Figure 4.9).    Figure 4.11: Acceptable Absorbance Curves Generated by the Spectrophotometer 18 RNA samples, represented by different colours within the blue bracket demonstrate acceptable absorbance curves as well as 260/280 and 260/230 ratios greater than 1.8.  Acceptable RNA samples   53  Figure 4.12: Unacceptable Absorbance Curves Generated by the Spectrophotometer 8 of the 14 RNA samples lying within the red bracket exhibit unacceptable absorbance curves with 260/280 and 260/230 ratios below 1.8. 6 of the 14 RNA samples in the blue bracket have acceptable values.      Figure 4.13: An Acceptable Electropherogram for a Polished Sample, RIN 9.0 Peaks for the marker, 18S and 28S ribosomal units are distinct indicating acceptable RNA quality  18S 28S Marker Unacceptable RNA samples Acceptable RNA samples   54 4.3 CLIP-CHIP? Microarray The protease microarray did not find any differentially expressed genes on SLA compared to polished surfaces at 5 days. Potential candidate genes that were expressed on both polished and SLA surfaces were reported. These included CysB, CysC and HSP, ADAMTS3, MMP1a, MMP7 and MMP12. Differential expression of these genes on SLA and polished surfaces was analyzed with qPCR but no significant differences were found.     Figure 4.14: CLIP-CHIP? Results Showing Expression of MMP1a, MMP7 and MMP12  MMP1a (collagenase-like B), MMP7 (matrilysin) and MMP12 are expressed at a very low level on almost all polished and SLA samples and are not significantly upregulated or downregulated on either surface.      55 4.4 Whole Genome Microarray 199 and 4,943 genes were differentially expressed on SLA compared to polished surfaces after 1 and 5 days, respectively. After 1 day, 116 genes were upregulated and 83 genes were downregulated and after 5 days, 1849 genes were upregulated and 3094 were downregulated. The identities of the top ten differentially expressed genes are shown in Tables 4.6 to 4.9. Time comparisons were also made, and the top differentially expressed genes over 5 days on P and SLA surfaces are listed in Tables 4.10 to 4.13. Gene expression profiles of LPS and IL-4 stimulated cells were also performed via comparison with un-stimulated macrophages and summarized in Tables 4.14 and 4.15. Genespring v12 (Agilent Technologies) was used to extract, normalize and sort significant data from the microarray chips and Ingenuity Pathway Analysis (IPA, Ingenuity Systems, www.ingenuity.com) was used to identify the most significant biological functions and canonical signalling pathways to help understand the functional significance of differentially expressed genes. Top molecular and cellular functions of the macrophages grown on SLA surfaces were identified and are listed and ordered by p-value (P<0.05) in Table 4.16. Canonical signalling pathways are depicted in Tables 4.17 and 4.18, and Figures 4.15 and 4.16. Molecules may follow canonical pathways, which are well-described, classical biological pathways, or non-canonical pathways, which consist of all the other alternative biological pathways (Sahu, 2008). Canonical signalling pathways are ordered by p-value, which IPA calculates using the Fisher?s exact test to screen for non-random associations of genes assigned to the pathways (Savli, Szendr?i, Romics, & Nagy, 2008). Differentially expressed genes are also ordered into biological networks ranked by p-value (Table 4.19).     56 CCL4, CCL7 and CCL13 are chemokines involved in biological functions that include cellular movement, haematological system development and function, immune cell trafficking, and the inflammatory response. CCL4 and CCL13 were differentially upregulated on SLA vs polished surfaces after 1 day of growth, and CCL7 was differentially upregulated after 5 days of growth. CTNND1, a cadherin associated protein was differentially upregulated on SLA vs polished surfaces after 1 day of growth. According to pathway analysis, CTNND1 is involved in cell proliferation, cell movement, activation of cells and inflammation. C1qa, a complement associated protein was found to be downregulated on SLA vs polished surfaces after 1 day of growth. Its associated pathways include those involving cell death, necrosis, apoptosis and cell activation. Adrenomedullin, a vasodilatory peptide, was highly expressed over time on both P and SLA surfaces.   Gene Symbol Gene Description Fold Change DLG2 Discs, large homolog 2  9.376 CTNND1 Catenin (cadherin associated protein), delta 1 6.416 Prl2c2  Prolactin family 2, subfamily c, member 5  6.253  P4HA2  Proline 4-hydroxylase  4.855  SLAIN2* SLAIN Motif-Containing Protein 2 4.417  CCL4  Chemokine ligand 4  4.366  CCL13* Chemokine ligand 13 4.150  ZNF385B  Zinc finger protein 385B  4.051  CDA Cytidine deaminase  3.974  RBPMS2 RNA binding protein with multiple splicing 2  3.931  Table 4.6: Top 10 Upregulated Genes on SLA vs P Surfaces, Day 1 *IPA identified SLAIN2, which was unidentified by Genespring v12. IPA identified CCL2 (in Genespringv12) as CCL13.         57  Gene Symbol Gene Description Fold Change Cypt1 Cysteine-rich perinuclear theca 4 (Cypt4) 10.154 C1QA Complement component 1, q subcomponent, alpha 6.345 EPHA8 Eph receptor A8 4.993 C1QC Complement component 1, q subcomponent, C chain  4.433 Ms4a4b Membrane-spanning 4-domains, subfamily A, member 4B  3.971 TDRD9 Tudor domain containing 9  3.865 TGFB1 Transforming growth factor, beta induced 3.755 CELF2 CUGBP, Elav-like family member 2 3.433 SYT2 Synaptotagmin II  3.235 EIF2S3 B6-derived CD11 +ve dendritic cells cDNA 3.035  Table 4.7: Top 10 Downregulated Genes on SLA vs P Surfaces, Day 1   Gene Symbol Gene Description Fold Change ZNF618 Zinc fingerprotein 618 4.417 SYT6 Synaptotagmin VI  4.048 ESPN Espin 3.991 CCL7 Chemokine (C-C motif) ligand 7 3.868 CSF3 Colony stimulating factor 3  3.667 A3galt2 Alpha 1,3-galactosyltransferase 2 3.356 CYP2D6 Cytochrome P450, family 2, subfamily d, polypeptide 10 3.213 HOMER2 Homer homolog 2 3.030 IGHA1 Immunoglobulin heavy chain complex 2.961 SP2 Sp2 transcription factor  2.838  Table 4.8: Top 10 Upregulated Genes on SLA vs P Surfaces, Day 5         58 Gene Symbol Gene Description Fold Change DNM3 Dynamin 3  8.405 CLEC4C C-type lectin domain family 4, member b1 6.165 F10 Coagulation factor X (F10) 5.790 CXCL3*  Chemokine (C-X-C motif) ligand 3 Macrophage inflammatory protein 2-Beta 5.331 C16orf89* Chromosome 16 Open Reading Frame 89 4.577 RORA RAR-related orphan receptor alpha  4.567 LAMA2 Laminin, alpha 2 3.839 FILIP1L Filamin A interacting protein 1-like 3.805 C8orf48* Chromosome 8 Open Reading Frame 48 3.560 NME8* NME/NM23 family member 8 3.528  Table 4.9: Top 10 Downregulated Genes on SLA vs P Surfaces, Day 5 *Disagreements between IPA and Genespring v12. CXCL3 was identified as CXCL2, C160rf89 was identified as AU021092, C8orf48 was identified as AI429214 and NME8 was identified as Txndc3 in GeneSpring v12   Gene Symbol Gene Description Fold Change P4HA2 Proline 4-hydroxylase, alpha II polypeptide 106.027 ADM Adrenomedullin 65.519 MGARP Not identified by Genespring v12 (Agilent Technologies) 43.187 F3 Coagulation factor III 28.321 TNFRSF9 Tumor necrosis factor receptor superfamily, member 9 27.855 CA6 Carbonic anhydrase 6 2 26.513 EGLN3 EGL nine homolog 3 24.664 TRIOBP TRIO and F-actin binding protein 14.889 PYGL Liver glycogen phosphorylase 12.625 SELENBP1 Selenium binding protein 1 12.099  Table 4.10: Top 10 Upregulated Genes over 5 Days of Growth on P Surfaces *IPA identified MGARP       59 Gene Symbol Gene Description Fold Change PMEPA1 Prostate transmembrane protein, androgen induced 1 23.685 CD207 CD207 antigen 22.134 PLA2G7 Phospholipase A2, group VII 7.887 Trim12a Tripartite motif-containing 12A 7.734 ORL1 Oxidized low density lipoprotein (lectin-like) receptor 1 7.486 DUOXA1 Dual oxidase maturation factor 1 7.289 CLEC6A C-type lectin domain family 4, member n 6.852 FABP7 Fatty acid binding protein 7, brain 6.624 OFLML3 Olfactomedin-like 3 6.590 PLEKHS1 Not identified by Genespring v12 (Agilent Technologies) 6.432  Table 4.11: Top 10 Downregulated Genes over 5 Days of Growth on P Surfaces *IPA identified PLEKHS1   Gene Symbol Gene Description Fold Change ADM Adrenomedullin 12.478 EGLN3 Egl Nine Homolog 3 11.501 ABCA1 ATP-binding cassette, sub-family A (ABC1), member 1 8.062 RNASE6 ribonuclease, RNase A family, 6 7.966 SELENBP1 Selenium-binding protein 1 7.879 P4HA2 Prolyl 4-hydroxylase, alpha polypeptide II 7.449 MGARP Mitochondria-localized glutamic acid-rich protein 5.708 FAM101B RefilinB 5.080 CHRM1 Chromosome 1 cholinergic receptor, muscarinic 1 4.698 Cd24a CD24a antigen 4.620  Table 4.12: Top 10 Upregulated Genes over 5 Days of Growth on SLA Surfaces        60 Gene Symbol Gene Description Fold Change GSTA5 Glutathione S-transferases 14.829 DLG2 Discs, large homolog 2  13.788 CLEC6A Dectin-2 10.830 FBXL13 F-box and leucine-rich repeat protein 13 10.384 Serpinb1b Serine (or cysteine) peptidase inhibitor, clade B, member 1b 8.973 CTNND1 Catenin (cadherin associated protein), delta 1 8.388 MYC Myelocytomatosis oncogene 8.266 PMEPA1 Prostate transmembrane protein, androgen induced 1 7.972 CLEC4C Transgene insertion 956 7.419 CD207 CD207 antigen 6.955  Table 4.13: Top 10 Downregulated Genes over 5 Days of Growth on SLA Surfaces     Gene Symbol Gene Description Fold Change SERPINB2 Plasminogen Activator Inhibitor 2 1645.189 IL-1? Interleukin 1 beta  1303.721 IL-6 Interleukin 6 797.016 IL-1? Interleukin 1 alpha 510.226 IL-36? Interleukin 36 alpha 473.506 CXCL1 Chemokine ligand 1 453.508 COL5A3 Col5a3 collagen, type V, alpha 3 425.930 PDPN Podoplanin 395.178 CCL7 Chemokine ligand 7 338.446 1100001G20Rik RIKEN cDNA 1100001G20 gene 318.629 CCL4* Chemokine ligand 4 19.340  Table 4.14: Top 10 Upregulated Genes on LPS-stimulated Cells on P Surfaces after 1 Day *Not in the top 10, but significantly upregulated, p value <0.05              61 Gene Symbol Gene Description Fold Change CISH Cytokine-inducible SH2-containing protein 228.658 Vmn1r181 Vomeronasal 1 receptor 181 37.370 ARG1 Arginase 35.551 CCDC8 Coiled-coil domain containing 8 31.585 CRB1 Crumbs homolog 1 22.749 RNASE3 Ribonuclease, RNase family, 3 21.287 ABCA4 ATP-binding cassette transporter gene sub-family A, member 4 21.235 RASSF9 Ras association domain-containing protein 9 21.186 GLI3 Zinc finger protein 20.604 PHF3 PHD finger protein 3 20.006 CCL7* Chemokine ligand 7 12.720 CCL4* Chemokine ligand 4 3.370  Table 4.15: Top 10 Upregulated genes on IL-4-Stimulated Cells on P Surfaces after 1 Day *Not in the top 10, but significantly upregulated, p value <0.05   SLA vs P, Day 1 SLA vs P, Day 5 Molecular and Cellular Functions p-value # molecules Molecular and Cellular Functions p-value # molecules Cellular movement 1.26x10-6 ? 9.16x10-3 39 Cell-to-cell signalling and interaction 1.08x10-5 ? 3.58x10-3 29 Cell-to-cell signalling and interaction 3.86x10-6 ? 9.16x10-3 36 Cellular growth and proliferation 3.86x10-6 ? 9.16x10-3 9 Cellular function and maintenance 4.23x10-6 ? 9.16x10-3 35 Cellular movement 4.23x10-6 ? 9.16x10-3 13 Cell death and survival 6.65x10-6 ? 9.16x10-3 46 Gene expression 6.65x10-6 ? 9.16x10-3 10 Cell morphology 8.30x10-6 ? 9.16x10-3 14 Cell morphology 8.30x10-6 ? 9.16x10-3 22  Table 4.16: Top Molecular and Cellular Functions Modulated by SLA vs P       62  Canonical Pathway p-value Ratio Molecules Role of IL-17F in Allergic Inflammatory Diseases 5.38x10-3  3/48 (0.062) CCL4, CCL7, IL1B Aryl Hydrocarbon Receptor Signalling 6.59x10-3 5/161 (0.031) MYC,ALDH1L2,NQO1,IL1B,NFIB Differential Regulation of Cytokine Production in Macrophages and T-helper cells  8.07x10-3 2/18 (0.111) CCL4,IL1B LPS/IL-1 Mediated Inhibition of RXR Function 9.01x10-3 6/239 (0.025) ACSBG1,ALDH1L2,UST,FABP4,IL1B,SULT1A3/SULT1A4 Histamine Biosynthesis 9.16x10-3 1/3 (0.333) HDC   Table 4.17: Top Canonical Pathways Modulated by SLA vs P, Day 1 The ratio is of the number of genes from the data set assigned to the canonical pathway, divided by the total number of genes in the canonical pathway     Canonical Pathway p-value Ratio Molecules TREM1 Signalling 4.32x10-3 3/71 (0.042) CXCL3,CCL7,ITGAX Glycolysis I 8.75x10-3 2/45 (0.044) FBP1,GAPDHS Gluconeogenesis I 8.75x10-3 2/50 (0.04) FBP1,GAPDHS Glycerol-3-phosphate shuttle 1.29x10-2 1/9 (0.111) GPD1 Hematopoiesis from Pluripotent Stem Cells 1.40x10-2 2/63 (0.032)  IGHA1,CSF3  Table 4.18: Top Canonical Pathways Modulated by SLA vs P, Day 5 The ratio is of the number of genes from the data set assigned to the canonical pathway, divided by the total number of genes in the canonical pathway                     63 SLA vs P Day 1 SLA vs P Day 5 Network Score Network Score Gastrointestinal Disease, Inflammatory Disease, Cell Cycle 30 Cellular Movement, Haematological System Development and Function, Immune Cell Trafficking 33 Cellular Movement, Haematological System Development and Function, Immune Cell Trafficking 28 Cell Cycle, Lipid Metabolism, Small Molecule Biochemistry 30 Cell-To-Cell Signalling and Interaction, Cellular Assembly and Organization, Cellular Function and Maintenance 25 Cellular Development, Connective Tissue Development and Function, Cell Morphology 30 Organismal Development, Reproductive System Development and Function, Tissue Development 25 Cell Signalling, Nucleic Acid Metabolism, Small Molecule Biochemistry 24 Cancer, Reproductive System Disease, Cardiovascular System Development and Function 24 Cell Signalling, Molecular Transport, Nucleic Acid Metabolism 22  Table 4.19: Top Networks Significantly Modulated by SLA vs P  IPA calculates and assigns a network score from p-values; a higher network score indicates greater probability of findings not occurring due to chance alone    64   Figure 4.15 Canonical Pathway Analysis for SLA vs P, Day 1  IPA lists canonical pathways most significant to the data set (SLA Day 1 versus P Day 1). The numbers on the right hand side of each bar represent the total number of genes that map to the canonical pathway. The proportions of upregulated (red) and downregulated (green) genes are illustrated. For example, in histamine biosynthesis, there are three genes in the pathway and one of them (Histidine decarboxylase) is downregulated, so the percentage of downregulation is 33%. Each canonical pathway is ordered according to level of significance (p-value). The yellow line is a representation of the  ?log p value.    65  Figure 4.16 Canonical Pathway Analysis for SLA vs P, Day 5 IPA lists canonical pathways most significant to the data set (SLA Day 5 versus P Day 5). The numbers on the right hand side of each bar represent the total number of genes that map to the canonical pathway. The proportions of upregulated (red) and downregulated (green) genes are illustrated by colour. CXCL3, CCL7 and ITGAX are differentially expressed and members of the TREM1 signalling pathway, which consists of 71 known molecules. CXCL3 and ITGAX are downregulated and CCL7 is upregulated at day 5. Each canonical pathway is ordered according to level of significance (p-value). The yellow line is a representation of the  ?log p value.  66 4.5 Quantitative Polymerase Chain Reaction  To screen for false positive results from the microarrays, confirmation of expression of CysC, HSP, MMP12 from the CLIP-CHIP? microarray and CCL4, CCL7, CTNND1 and ADM from the whole genome microarray was performed with designed primers (see section 3.4.5). Primers for CysB, ADAMTS3, MMP1a and MMP7, CCL13 and C1qa did not detect enough cDNA product for qPCR reactions. No significant differences were found for CysC, HSP, MMP12, CTNND1 and ADM. Two results from the whole genome microarray were confirmed by qPCR: CCL4 was significantly upregulated on SLA surfaces by a factor of 2.18 at day 1, and CCL7 was significantly upregulated on SLA surfaces by a factor of 4.01 at day 5 (Figures 4.17 and 4.18). Primer efficiency curves for CCL4, CCL7 and CTNND1 are illustrated in Figure 4.11.     Figure 4.17: Gene Expression of CCL4 on P and SLA at Day 1 Results are averaged from 3 independent experiments and include SD error bars.  CCL4 was significantly upregulated on SLA surfaces by a factor of 2.18 after 1 day 0	 ?0.5	 ?1	 ?1.5	 ?2	 ?2.5	 ?3	 ?Polished	 ?  	 ?SLA	 ?2-????CT	 ?Surface	 ?Type	 ?  67   Figure 4.18: Gene Expression of CCL7 on P and SLA at Day 5 Results are averaged from 3 independent experiments and include SD error bars.  CCL7 was significantly upregulated on SLA surfaces by a factor of 4.01 after 5 days   0	 ?0.5	 ?1	 ?1.5	 ?2	 ?2.5	 ?3	 ?3.5	 ?4	 ?4.5	 ?5	 ?Polished	 ?  	 ?SLA	 ?2-????CT	 ?Surface	 ?Type	 ?  68  Figure 4.19: Efficiency of qPCR Using CCL4, CCL7 and CTNND1 Primers  Standard curves for CCL7, CTNND1 and CCL4 were generated by plotting the log of the cDNA concentration to the numbers of cycles required to obtain Ct. Slope values are related to qPCR efficiency. The equation E = 10(-1/slope) -1 was used to calculate efficiency. The qPCR efficiencies for CCL7, CTNND1 and CCL4 are 1.202, 1.382 and 1.263 respectively. Ideal efficiency approaches 1.0 (Bustin et al., 2009). The coefficient of determination, r2 may be used to evaluate variability within the assay, and a value >0.980 is desirable. Difficulty achieving these values usually requires deletion of endpoints of the standard curve (Taylor et al., 2010). The r2 values for CCL7, CTNND1 and CCL4 are 0.939, 0.971 and 0.990 respectively. No values have been deleted in this figure.          y = -2.9164x + 29.768 R? = 0.93905 y = -2.8191x + 22.854 R? = 0.97055 y = -2.6517x + 20.95 R? = 0.98985 0 5 10 15 20 25 30 35 40 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 CCL7 CTNND1 CCL4 Linear (CCL7) Linear (CTNND1) Linear (CCL4)   69 4.6 Comparative Study on Titanium-coated P and SLA Surfaces  Cells have been found to react to topographical features in a similar manner on many different materials (Curtis & Wilkinson, 1997). Surface topographies were reproduced in epoxy resin to create approximately two hundred samples required to perform the experiments in this thesis, and macrophage response to epoxy versus titanium substrata was called into question. To compare the response of RAW264.7 macrophages to epoxy and titanium-coated epoxy, qPCR was used to compare differential expression of CCL4 on Ti-coated P and SLA surfaces. CCL4 was shown to be upregulated by a factor of 3.78 on SLA surfaces after 1 day. These results are in rough agreement with the microarray (4.366 fold upregulation) and qPCR on epoxy (2.18 fold upregulation).     Figure 4.20: Relative Expression of CCL4 on Ti-coated P and SLA Surfaces at Day 1  0	 ?0.5	 ?1	 ?1.5	 ?2	 ?2.5	 ?3	 ?3.5	 ?4	 ?4.5	 ?5	 ?2-????CT	 ?	 ?Polished	 ? SLA	 ?  70 Chapter  5: Discussion, Conclusions and Future Directions 5.1 Discussion The analogy of a ?fishing expedition? has been used to describe microarray experiments (Lockhart & Winzeler, 2000). Thousands of fishing lines were drawn to ?catch? or identify specific genes of interest involved with implant wound healing responses to a smooth and rough surface topography. The protease and whole genome microarrray analysed expression of 891 and 39,340 genes, respectively. Microarrays were used as exploratory tools to help characterize RAW264.7 macrophage phenotype induced by implant surface topography. Due to the large number of gene comparisons to be made and the potential for false positives results, qPCR was used to verify findings from the microarray.  5.1.1 Upregulation of Macrophage Chemoattractants CCL4 and CCL7 Significant upregulation of CCL4, also known as macrophage inflammatory protein-1 beta (MIP-1?), at day 1 and CCL7, also known as monocyte chemotactic protein-3 (MCP-3), at day 5 on SLA versus polished surfaces was confirmed with qPCR. CCL4 and CCL7 are macrophage chemoattractants that send signals to recruit more macrophages.  A summary of the processes occurring on polished and SLA surface topographies is illustrated as the ?Valentine Model? in Figure 5.1.   The macrophage inflammatory protein-1 (MIP-1) family includes CCL3 (MIP1?), CCL4 (MIP-1?), CCL9/10 (MIP-1?) and CCL15 (MIP-1?) (Maurer & Stebut, 2004). MIP-1 proteins are chemokines that have chemoattractant and pro-inflammatory properties and act via G-protein coupled surface receptors present on monocytes, macrophages and lymphocytes (Maurer &   71 Stebut, 2004). CCL4 is highly related to CCL3, which is characterized as a monocyte chemoattractant produced by platelets and resident macrophages involved in inflammatory and wound healing responses (DiPietro, Burdick, Low, Kunkel, & Strieter, 1998; VanOtteren, Standiford, & Kunkel, 1994).   CCL7, monocyte chemotactic protein-3, was highly upregulated on SLA compared to polished surfaces at day 5. The monocyte chemoattractant proteins (MCPs) include CCL2 (MCP-1), CCL8 (MCP-2), CCL7 (MCP-3) and CCL13 (MCP-4) (McQuibban et al., 2002). Like CCL4, CCL7 can be induced by LPS, TNF?, IFN? and IL-1? (Dezerega et al., 2010). CCL7 is also a marker for an alternatively activated (M2) murine macrophage (Martinez et al., 2009). Chemokines CCL2, CCL7, CCL8 and CCL13 bind to the same chemokine receptor, CCR2 (Szymczak & Deepe, 2009). Chemokines binding the CCR2 receptor triggers signalling events that lead to the recruitment of macrophages (Szymczak & Deepe, 2009).     72  Figure 5.1: The Valentine Model of Macrophage Accumulation on Rough Surfaces RAW264.7 macrophages upregulate gene expression of chemokines CCL4 and CCL7 on SLA surfaces   A key characteristic of the SLA surface observed in-vivo is its ability to accumulate macrophages. These macrophages remain on the implant surface and are associated with faster and greater amounts of bone formation (Chehroudi et al., 2009). The increased gene expression of chemokines CCL4 and CCL7 observed in these experiments may explain macrophage accumulation on SLA surfaces in-vivo.      73 5.1.2 Gene Expression Profiles of Macrophages Stimulated with LPS and IL-4 LPS and IL-4 stimulated controls exhibited gene expression profiles consistent with M1 and M2 phenotypes (Table 4.14 and 4.15). After 1 day of growth, inflammatory cytokines IL-1?, IL6 and IL-1? had fold change upregulations of 1,303.72, 797.02 and 510.23 respectively. Arg1 was upregulated 35.55 times on IL-4 stimulated macrophages. These results indicate that cells of the RAW264.7 macrophage cell line respond in an appropriate manner and are capable of expressing markers of M1 and M2 phenotypes. Furthermore, these results are in agreement with Barth et al. (2013), who observed upregulation of another M1 marker, NOS2 and Arg1 on LPS and IL-4 stimulated RAW264.7 macrophages respectively. Use of M1 and M2 controls in the microarray allow for comparisons to be made with macrophages stimulated by surface topography. CCL4 and CCL7, upregulated on SLA surfaces, were also upregulated on LPS and IL-4 stimulated controls indicating that these chemokines may be involved in M1 and M2 responses. Overall, it appears that RAW264.7 macrophages are an excellent cell line to use for in-vitro experiments on surface topographies as they exhibit the same characteristics as murine macrophages, and act in predictable ways when stimulated by cytokines involved in macrophage phenotype regulation.   5.1.3 Other Studies Other in-vitro studies using different model systems have shown that surface topographies affect unstimulated macrophage gene expression. Tan et al. (2006) studied expression of pro-inflammatory cytokines by murine macrophages cultured on polished and grit-blasted titanium discs. Only a modest increase of pro-inflammatory cytokine expression was observed over both surfaces (Tan et al., 2006). Paul et al. (2008) used microarray technology to investigate the effects of different polyvinylidene fluoride (PVDF) surface topographies on gene expression of   74 human macrophages and found small upregulations of CCL4 and CCL7 on a microstructured surface containing 1um tall features spaced 30um apart compared to smooth surface control. However, Paul et al. (2008) concluded that surface topography had no influence on expression of CCL4 and CCL7, as fold changes were only 0.80 and 0.18 respectively. They concluded that the microstructured surface induced a gene expression profile that was predominantly pro-inflammatory, but also had properties of both M1 and M2 phenotypes (Paul et al., 2008). Hamlet et al. (2011) investigated the effects of different topographies on RAW264.7 macrophage cells at the gene and protein level and found that chemokines CCL2 and CCL3 were downregulated on a commercially pure titanium nanoscale calcium phosphate surface, when compared to a commercially pure microrough surface. No differences in CCL2, CCL3 or CCL4 secretion were found at the protein level among the nanoscale and microrough surfaces tested (Hamlet & Ivanovski, 2011). Barth et al. (2013), however, screened for cytokine and chemokine secretion using an antibody array and observed an accumulation of CCL2 and CCL3 on SLA surfaces after 5 days. Overall, these studies suggest that a microrough surface topography activates small changes in gene expression, and a change in macrophage phenotype that is neither M1 nor M2. Barth et al. (2013) performed qPCR to compare the expression of NOS2 and Arg1 on RAW264.7 macrophages grown on polished and SLA surface topographies and found no significant differences in gene expression of these M1 and M2 markers. In the absence of a stimulus such as LPS or IL-4, RAW264.7 macrophages exposed to surface topography alone will not activate cellular pathways to produce NOS2 or Arg1 (Barth et al., 2013).   There are no characteristic markers of macrophage phenotype induced by surface topography and identifying one would be difficult due to the variety of different cell lines and surfaces used   75 across experiments. Based on the available literature and results from the experiments in this thesis, we can conclude that microrough topographies induce a small upregulation of pro-inflammatory cytokines and chemokines and exhibit an intermediary phenotype between M1 and M2 (Barth et al., 2013; Paul et al., 2008).   Although epoxy replicas, which exhibit a different surface chemistry from titanium, were used as substrata for these experiments, they were observed to be capable of stimulating similar gene expression patterns as titanium-coated replicas. SLA topography on epoxy and Ti-coated discs both stimulated upregulation of CCL4 (Section 4.6).   5.1.4 Microarray Data Analysis    The Gene Ontology (GO) project is a manually curated electronic database that provides a representation of the biological knowledge of gene-product function through a controlled vocabulary of precisely defined and interrelated terms (Chan, Kishore, Sternberg, & Van Auken, 2012; Leonelli, Diehl, Christie, Harris, & Lomax, 2011; Rubin, Shah, & Noy, 2008). These terms are used to annotate biomedical databases and allow pathway analysis software to make biomedical insights from experimental results (Rubin et al., 2008; Werner, 2008). There are two components of GO: ontologies and annotations (Rubin et al., 2008). Ontologies are the defined terms and structured relationships between these terms, and annotations are the associations made between the terms and gene products (Rhee, Wood, Dolinski, & Draghici, 2008; Rubin et al., 2008). GO can be divided into three separate branches or hierarchies: biological processes, molecular function and cellular components (Sidhu, Dillon, & Chang, 2008; Smith, Williams, & Schulze-Kremer, 2003; Werner, 2008).    76  To help us characterize macrophage phenotype and evaluate the significance of thousands of genes that were differentially expressed on SLA surfaces, pathway analysis software, Ingenuity Pathway Analysis (IPA) was used. IPA uses information contained in the Ingenuity Knowledge Base (IKB), a large private database developed from millions of selected findings from public literature that is maintained and updated on a weekly basis by content scientists (Ficenec et al., 2003; Helleman, Smid, Jansen, van der Burg, & Berns, 2010; Valencia-Cruz et al., 2013). IPA can be used to annotate microarray data sets with GO terms, gene symbols or other identifiers, generate lists of top upregulated and downregulated genes, and perform gene ontology and functional analyses (Henderson-Maclennan, Papp, Talbot, McCabe, & Presson, 2010). IPA integrates findings across multiple dimensions from the gene and molecular level, to make connections with cellular and disease processes. It can also determine which canonical and non-canonical biological pathways are significantly associated with the genes of interest (Helleman et al., 2010).   GO analysis was applied to our experimental data to determine the functional significance of differentially expressed genes on SLA versus polished surface topographies. Some of the top molecular and cellular functions include cellular movement and cell-to-cell signalling (see Table 4.16), which is supportive of our findings that that SLA surface topography induces upregulations of pro-inflammatory cytokines and chemokines.   Differentially expressed genes were also sorted into canonical pathways, ranked by p-value. The top canonical pathway for SLA versus polished surfaces at day 1 (Table 4.17) was ?the role of   77 IL-17F in Allergic Inflammatory Diseases?, which included upregulations of CCL4 and CCL7. IL-17F is a cytokine produced by activated T cells that has been shown in the literature to induce other chemokines including CCL2 and CXCL2 (Iyoda et al., 2010). The top canonical pathway for SLA versus polished surfaces at day 5 (Table 4.18) was ?TREM1 signalling?, which included upregulation of CCL7. TREM-1 is a recently identified cell surface receptor expressed by neutrophils and monocytes. When stimulated, TREM-1 has been shown to trigger release of inflammatory chemokines and cytokines, including CCL2 and CCL7 (Bouchon, Dietrich, & Colonna, 2000; Dower, Ellis, Saraf, Jelinsky, & Lin, 2008). Glycolysis and gluconeogenesis, anaerobic and aerobic glucose metabolic pathways were top canonical pathways at day 5. Use of both M1 and M2 metabolic pathways by macrophages also supports findings that the SLA topography induces a mixed phenotype. However, IPA findings must be interpreted with caution. Upon closer examination (see Table 4.18), only two molecules, fructose-1,6-bisphosphonate (FBP1) and Glyceraldehyde-3-phosphate dehydrogenase, testis-specific (GAPDHS) were reported as differentially expressed within the glycolysis and gluconeogenesis canonical pathways.   Functional analysis with IPA determined that the top networks associated with SLA versus polished surface topographies included those related to inflammation, cellular movement, immune cell trafficking and cell-to-cell signalling (see Table 4.19). The top scoring network was ?Cellular Movement, Haematological System Development and Function, Immune Cell Trafficking?, which included the following molecules: ABCA1, ABCC2, Alp, BHLHE40, CCL7, CCL13, Cg, Collagen type IV, CSF3, CXCL3, DDR1, elastase, ERK1/2, F10, Fcer1, Fibrinogen, HPSE, HR, IgE, IL-1, LAMA2, LDL, MAP2K1/2, MMP, Nr1h, PAX7, PDGF, BB,   78 RAP1GAP, RORA, Rxr, SAA, SLPI, SNTG1, TGF? and trypsin. Overall, information generated by IPA analysis was used to choose candidates for qPCR validation of the microarray. CCL4, CCL7, CCL13, C1Qa, CTNND1 and ADM were selected based on fold change value, gene ontology and functional network analyses.   5.1.4.1 Annotation Discrepancies While comparing data generated by GeneSpringv12 to that generated by IPA, several gene annotation discrepancies were noted in the lists of top 10 differentially expressed genes. A brief list of detected discrepancies is summarized in Table 5.1. Annotation errors may occur due to errors in typography or misreading similar gene descriptions, and are often missed by program users due to the reliance on the accuracy of the software that they assume accurately transforms raw data into readable formats (Henderson-Maclennan et al., 2010). Microarray data is often exported as user-friendly formats such as a Microsoft Office Excel spreadsheet before being imported into IPA. This can cause problems because during this process, probename IDs containing numbers may be converted into numerical dates or formulas and become unrecognizable by IPA (Henderson-Maclennan et al., 2010). Misidentified genes can significantly alter downstream analysis of functional pathways and affect interpretation of data. Henderson et al. (2010) compared gene symbol annotations over four releases of IPA software and found that annotations of probeset IDs ranged from 86.7 to 97.3%. Since gene annotations and canonical pathways continue to change over time as scientific research progresses, one can avoid gene annotation problems by 1) verifying results with later versions of the software, 2) using one more than pathway analysis program, or 3) interpreting findings with caution   79 (Henderson-Maclennan et al., 2010). The errors in annotation discovered during the course of these experiments have been reported to IPA.   Agilent probeset ID GenBank Accession # Gene symbol (GeneSpring) Gene symbol (IPA) A_51_P286737 NM_011333 CCL2 CCL13 A_55_P2003321 BC019454 Not identified SLAIN2 A_51_P217463 NM_009140 CXCL2 CXCL3 A_52_P159429 NM_001033220 AU021092 C16orf89 A_55_P2069231 NM_001039220 AI429214 C8orf48 A_55_P2125801 NM_001167909 Txndc3 NME8  Table 5.1: IPA Gene Annotation Discrepancies    5.1.5 False Positives  Although CCL4 and CCL7 were confirmed as differentially expressed genes on the microarray by qPCR, a number of false positives were discovered. In particular, significant differences could not be found for CCL13, C1Qa, CTNND1 and ADM. These errors may have occurred during microarray data processing and analysis, or could be related to the particular window of time when the assays were performed. Furthermore, greater than 6x108 comparisons were made for each surface at each time point. Procedures for correction of type I error in multiple comparisons have been suggested such as Bonferroni, but can be overly conservative and fail to identify true positive differences. When the Benjamini-Hochberg false discovery correction was used, no significantly upregulated or downregulated genes were found, but in fact several differences were identified by qPCR (CCL4 was significantly upregulated on SLA surfaces by a factor of 2.18 at day 1, and CCL7 was significantly upregulated on SLA surfaces by a factor of 4.01 at   80 day 5). Overall, it appears that differences in gene expression between the polished and SLA surfaces were few in number but some are nevertheless real.   5.1.6 Conclusions The following conclusions can be made from the experiments outlined in this thesis:  1. The RAW264.7 macrophage cell line exhibits gene expression patterns characteristic of macrophages. Gene expression profiles of LPS and IL-4 stimulated macrophages were representative of M1 and M2 phenotypes.  2. SLA surface topography induces changes in gene expression and appears to up-regulate chemokines CCL4 and CCL7 that specifically attract macrophages  3. Differences between polished and SLA surfaces are small, and require qPCR validation to ensure a true positive result 4. Gene ontology and pathway analysis suggest that SLA surfaces induce a mixed M1 and M2 phenotype associated with increased levels of cellular movement, cell-to-cell signalling, aerobic and anaerobic metabolic pathways 5. Preliminary results indicate that the upregulation of CCL4 on SLA epoxy surfaces is also found on SLA titanium-coated surfaces, supporting the use of epoxy, a readily available and cost-effective material, for in-vitro studies on implant surface topography  5.1.7 Future Directions 1. Clarification of possible annotation errors with an updated version of IPA software or with another pathway analysis program is needed to confirm expression of CCL2,   81 CXCL2 and other potential genes that may have been absent on the version of the software used to analyse data in this thesis.  2. Validation of additional genes, such as CCL2, CCL3 and other members of the monocyte chemoattractant proteins and macrophage inflammatory protein families is needed to improve our understanding of macrophage phenotype induced by SLA surface topography, and to explain some results in conflict with previous literature 3. Validation of gene expression at the protein level is needed to determine if genes of interest are not only expressed, but translated into proteins and secreted into the environment 4. Microarrays may be used to develop improved implant surfaces by identifying desirable gene expression profiles.   82 Bibliography Adams, J. (2008). Transcriptome: Connecting the Genome to Gene Function. Nature Education. Retrieved July 13, 2013, from http://www.nature.com/scitable/topicpage/transcriptome-connecting-the-genome-to-gene-function-transcriptome-connecting-the-genome-to-gene-function-605  Adell, R., Lekholm, U., Rockler, B., & Br?nemark, P. I. (1981). A 15-year study of osseointegrated implants in the treatment of the edentulous jaw. International Journal of Oral Surgery, 10(6), 387?416. doi:10.1016/S0300-9785(81)80077-4  Aebi, U., & Pollard, T. D. (1987). A glow discharge unit to render electron microscope grids and other surfaces hydrophilic. Journal of Electron Microscopy Technique, 7(1), 29?33. doi:10.1002/jemt.1060070104  Agilent Technologies Inc. (2008, May 1). One- Color Microarray-Based Gene Expression Analysis (Quick Amp Labeling) with Tecan HS Pro Hybridization. Santa Clara, CA, USA.  Albrektsson, T. (2008). Hard tissue implant interface. Australian Dental Journal, 53, S34?S38. doi:10.1111/j.1834-7819.2008.00039.x  Albrektsson, T., & Wennerberg, A. (2004). Oral implant surfaces: Part 1--review focusing on topographic and chemical properties of different surfaces and in vivo responses to them. The International Journal of Prosthodontics, 17(5), 536?543.  Albrektsson, T., & Wennerberg, A. (2005). The impact of oral implants-past and future, 1966-2042. J Can Dent Assoc. 71(5), 327  Albrektsson, T., Zarb, G., & Worthington, P. (1986). The long-term efficacy of currently used dental implants: a review and proposed criteria of success. International Journal of Oral and Maxillofacial Implants, 1(1), 11?25.  Alfarsi, M. A., Hamlet, S. M., & Ivanovski, S. (2013). Titanium surface hydrophilicity modulates the human macrophage inflammatory cytokine response. Journal of Biomedical Materials Research Part A. doi:10.1002/jbm.a.34666  Anderson, J. M., Rodriguez, A., & Chang, D. T. (2008). Foreign body reaction to biomaterials. Seminars in Immunology, 20(2), 86?100. doi:10.1016/j.smim.2007.11.004  Anusavice, K. J. (2003). Phillips' Science of Dental Materials. 11th edition. Philadelphia, PA, USA: Saunders Co.   ATCC. (2012, July 2). ATCC Animal Cell Culture Guide. Manassas, VA. Retrieved July 2, 2013, from https://www.atcc.org/~/media/PDFs/Culture%20Guides/AnimCellCulture_Guide.pdf   83 Ballanti, E., Perricone, C., Greco, E., Ballanti, M., Di Muzio, G., Chimenti, M. S., & Perricone, R. (2013). Complement and autoimmunity. Immunologic Research, 56(2-3), 477?491. doi:10.1007/s12026-013-8422-y  Barth, K. A., Waterfield, J. D., & Brunette, D. M. (2013). The effect of surface roughness on RAW 264.7 macrophage phenotype. Journal of Biomedical Materials Research Part A, 101(9) 2679-88. doi:10.1002/jbm.a.34562  Bartlett, J. D., Dobeck, J. M., Tye, C. E., & Perez-Moreno, M. (2010). Targeted p120-Catenin Ablation Disrupts Dental Enamel Development. PLOS ONE, 5(9), 1?11.  Berglundh, T., Abrahamsson, I., & Lang, N. P. (2003). De novo alveolar bone formation adjacent to endosseous implants. Clinical Oral Implants Research, 14, 251?262.  Berglundh, T., Abrahamsson, I., Albouy, J. P., & Lindhe, J. (2007). Bone healing at implants with a fluoride-modified surface: an experimental study in dogs. Clinical Oral Implants Research, 18(2), 147?152. doi:10.1111/j.1600-0501.2006.01309.x  Beumer, J., Marunick, M. T., & Esposito, S. J. (2011). Maxillofacial Rehabilitation. Hanover Park, IL: Quintessence Publishing Company.  Biswas, S. K., & Mantovani, A. (2012). Orchestration of metabolism by macrophages. Cell Metabolism, 15(4), 432?437. doi:10.1016/j.cmet.2011.11.013  Biswas, S. K., Chittezhath, M., Shalova, I. N., & Lim, J.-Y. (2012). Macrophage polarization and plasticity in health and disease. Immunologic Research, 53(1-3), 11?24. doi:10.1007/s12026-012-8291-9  Bouchon, A., Dietrich, J., & Colonna, M. (2000). Cutting edge: inflammatory responses can be triggered by TREM-1, a novel receptor expressed on neutrophils and monocytes. Journal of immunology (Baltimore, Md. 1950), 164(10), 4991?4995.  Br?nemark, P. I., Hansson, B. O., Adell, R., Breine, U., Lindstr?m, J., Hall?n, O., & Ohman, A. (1977). Osseointegrated implants in the treatment of the edentulous jaw. Experience from a 10-year period. Scandinavian Journal of Plastic and Reconstructive Surgery. Supplementum, 16, 1?132.  Brunette, D. M. (1988). The effects of implant surface topography on the behavior of cells. International Journal of Oral Maxillofacial Implants. 3:231?46  Brunette, D. M. (2001). Principles of Cell Behavior on Titanium Surfaces and Their Application to Implanted Devices. In Titanium in Medicine (pp. 485?512). Berlin, Heidelberg: Springer Berlin Heidelberg. doi:10.1007/978-3-642-56486-4_15   84 Brunette, D. M., Hamilton, D. W., Chehroudi, B., & Waterfield, J. D. (2005). Update on improving the bio-implant interface by controlling cell behaviour using surface topography. International Congress Series, 1284, 229?238. doi:10.1016/j.ics.2005.06.093  Buser, D., Janner, S. F. M., Wittneben, J.-G., Br?gger, U., Ramseier, C. A., & Salvi, G. E. (2012). 10-Year Survival and Success Rates of 511 Titanium Implants with a Sandblasted and Acid-Etched Surface: A Retrospective Study in 303 Partially Edentulous Patients. Clinical Implant Dentistry and Related Research, 14(6), 839?851. doi:10.1111/j.1708-8208.2012.00456.x  Bustin, S. A., & Nolan, T. (2004). Pitfalls of Quantitative Real-Time Reverse-Transcription Polymerase Chain Reaction. Journal of Biomolecular Techniques : JBT, 15(3), 155.  Bustin, S. A., Benes, V., Garson, J. A., Hellemans, J., Huggett, J., Kubista, M., et al. (2009). The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Clinical Chemistry, 55(4), 611?622.  Carinci, F., Volinia, S., Pezzetti, F., Francioso, F., Tosi, L., & Piattelli, A. (2003). Titanium-cell interaction: analysis of gene expression profiling. Journal of biomedical materials research. Part B, Applied Biomaterials, 66(1), 341?346. doi:10.1002/jbm.b.10021  Carvalho-Gaspar, M., Billing, J. S., Spriewald, B. M., & Wood, K. J. (2005). Chemokine gene expression during allograft rejection: Comparison of two quantitative PCR techniques. Journal of Immunological Methods, 301(1-2), 41?52. doi:10.1016/j.jim.2005.03.003  Chan, J., Kishore, R., Sternberg, P., & Van Auken, K. (2012). The Gene Ontology: enhancements for 2011. Nucleic Acids Research, 40, 559?564. doi:10.1093/nar/gkr1028  ChangLiu, C. M., & Woloschak, G. E. (1997). Effect of passage number on cellular response to DNA-damaging agents: Cell survival and gene expression. Cancer Letters, 113(1-2), 77?86.  Chehroudi, B., Ghrebi, S., Murakami, H., Waterfield, J. D., Owen, G., & Brunette, D. M. (2009). Bone formation on rough, but not polished, subcutaneously implanted Ti surfaces is preceded by macrophage accumulation. Journal of Biomedical Materials Research Part A, 9999A, 724-737 doi:10.1002/jbm.a.32587  Classen, A., Lloberas, J., & Celeada, A. (2009). Macrophage Activation: Classical vs Alternative. Methods in Molecular Biology. 531, 29-43. doi: 10.1007/978-1-59745-396-7_3. Cochran, D. L. (1999). A Comparison of Endosseous Dental Implant Surfaces. Journal of Periodontology, 70(12), 1523?1539. doi:10.1902/jop.1999.70.12.1523  Cooper, L. F. (2000). A role for surface topography in creating and maintaining bone at titanium endosseous implants. Journal of Prosthetic Dentistry, 84, 522?534.     85 Curtis, A., & Wilkinson, C. (1997). Topographical control of cells. Biomaterials.  Davies, J. E. (2003). Understanding peri-implant endosseous healing. Journal of Dental Education, 67(8), 932?949.  Dezerega, A., Osorio, C., Mardones, J., Mundi, V., Dutzan, N., Franco, M., et al. (2010). Monocyte chemotactic protein-3: possible involvement in apical periodontitis chemotaxis. International Endodontic Journal, 43(10), 902?908. doi:10.1111/j.1365-2591.2010.01764.x  DiPietro, L. A., Burdick, M., Low, Q. E., Kunkel, S. L., & Strieter, R. M. (1998). MIP-1alpha as a critical macrophage chemoattractant in murine wound repair. Journal of Clinical Investigation, 101(8), 1693?1698. doi:10.1172/JCI1020  Dower, K., Ellis, D. K., Saraf, K., Jelinsky, S. A., & Lin, L.-L. (2008). Innate Immune Responses to TREM-1 Activation: Overlap, Divergence, and Positive and Negative Cross-Talk with Bacterial Lipopolysaccharide. Journal of Immunology, 180, 3520?3534.  Elias, C. N., Lima, J., Valiev, R., & Meyers, M. A. (2008). Biomedical applications of titanium and its alloys. Journal of the Minerals, Metals and Materials Society, 60, 46-49  Elman, N. M., Ho Duc, H. L., & Cima, M. J. (2009). An implantable MEMS drug delivery device for rapid delivery in ambulatory emergency care. Biomedical Microdevices, 11(3), 625?631. doi:10.1007/s10544-008-9272-6  Esposito, M., Hirsch, J. M., Lekholm, U., & Thomsen, P. (1998). Biological factors contributing to failures of osseointegrated oral implants. (II). Etiopathogenesis. European Journal of Oral Sciences, 106(3), 721?764.  Fairweather, D., & Cihakova, D. (2009). Alternatively activated macrophages in infection and autoimmunity. Journal of Autoimmunity, 33(3-4), 222?230. doi:10.1016/j.jaut.2009.09.012  Ficenec, D., Osborne, M., Pradines, J., Richards, D., Felciano, R., Cho, R. J., et al. (2003). Computational knowledge integration in biopharmaceutical research. Briefings in Bioinformatics, 4(3), 260?278.  Franz, S., Rammelt, S., Scharnweber, D., & Simon, J. C. (2011). Immune responses to implants - A review of the implications for the design of immunomodulatory biomaterials. Biomaterials, 32(28), 6692?6709. doi:10.1016/j.biomaterials.2011.05.078  Ghrebi, S., Hamilton, D. W., Waterfield, J.D., & Brunette, D. M. (2013). The effect of surface topography on cell shape and early ERK1/2 signaling in macrophages; linkage with FAK and Src. 101(7), 2118-28. Journal of Biomedical Materials Research Part A. doi:10.1002/jbm.a.34509    86 Ghrebi, S., Owen, G.R. & Brunette, D.M. (2007). Triton X-100 pretreatment of LR-white thin sections improves immunofluoresence specificity and intensity. Microsc Res Tech. 70(7), 55-562  Giulietti, A., Overbergh, L., Valckx, D., Decallonne, B., Bouillon, R., & Mathieu, C. (2001). An Overview of Real-Time Quantitative PCR: Applications to Quantify Cytokine Gene Expression. Methods, 25(4), 386?401. doi:10.1006/meth.2001.1261 Goodacre, C. J., Bernal, G.,  Rungcharassaeng, K. & Kan, J. Y. K. (2003). Clinical complications with implants and implant prostheses. Journal of Prosthetic Dentistry, 90, 121?132.  Gordon, S., & Taylor, P. R. (2005). Monocyte and macrophage heterogeneity. Nature Reviews Immunology, 5(12), 953?964. doi:10.1038/nri1733  Grassi, S., Piattelli, A., de Figueiredo, L. C., Feres, M., de Melo, L., Iezzi, G., et al. (2006). Histologic evaluation of early human bone response to different implant surfaces. Journal of Periodontology, 77(10), 1736?1743. doi:10.1902/jop.2006.050325  Hallgren, C., Reimers, H., Gold, J., & Wennerberg, A. (2001). The importance of surface texture for bone integration of screw shaped implants: an in vivo study of implants patterned by photolithography. Journal of Biomedical Materials Research Part A, 57(4), 485?496.  Hamlet, S., & Ivanovski, S. (2011). Inflammatory cytokine response to titanium chemical composition and nanoscale calcium phosphate surface modification. Acta Biomaterialia, 7(5), 2345?2353. doi:10.1016/j.actbio.2011.01.032  Haro, H., Crawford, H. C., Fingleton, B., Shinomiya, K., Spengler, D. M., & Matrisian, L. M. (2000). Matrix metalloproteinase-7-dependent release of tumor necrosis factor-alpha in a model of herniated disc resorption. Journal of Clinical Investigation, 105(2), 143?150. doi:10.1172/JCI7091  Harrington, C. A., Rosenow, C., & Retief, J. (2000). Monitoring gene expression using DNA microarrays. Current opinion in Microbiology, 3, 285?291.  Hartley, J. W., Evans, L. H., Green, K. Y., Naghashfar, Z., Macias, A. R., Zerfas, P. M., & Ward, J. M. (2008). Expression of infectious murine leukemia viruses by RAW264.7 cells, a potential complication for studies with a widely used mouse macrophage cell line. Retrovirology, 5(1), 1. doi:10.1186/1742-4690-5-1  Harvey, A. G., Hill, E. W., & Bayat, A. (2013). Designing implant surface topography for improved biocompatibility. Expert Review of Medical Devices, 10(2), 257?267. doi:10.1586/erd.12.82  Heid, C. A., Stevens, J., Livak, K. J., & Williams, P. M. (1996). Real time quantitative PCR. Genome Research, 6, 986?994.    87 Helleman, J., Smid, M., Jansen, M. P. H. M., van der Burg, M. E. L., & Berns, E. M. J. J. (2010). Pathway analysis of gene lists associated with platinum-based chemotherapy resistance in ovarian cancer: the big picture. Gynecologic oncology, 117(2), 170?176. doi:10.1016/j.ygyno.2010.01.010  Henderson-Maclennan, N. K., Papp, J. C., Talbot, C. C., McCabe, E. R. B., & Presson, A. P. (2010). Pathway analysis software: annotation errors and solutions. Molecular genetics and metabolism, 101(2-3), 134?140. doi:10.1016/j.ymgme.2010.06.005  Hume, D. A. (2006). The mononuclear phagocyte system. Current Opinion in Immunology.  ISO (2009). Biological evaluation of medical devices -- Part 1: Evaluation and testing within a risk management process. ISO 10993-1:2009.  Iyoda, M., Shibata, T., Kawaguchi, M., Hizawa, N., Yamaoka, T., Kokubu, F., & Akizawa, T. (2010). IL-17A and IL-17F stimulate chemokines via MAPK pathways (ERK1/2 and p38 but not JNK) in mouse cultured mesangial cells: synergy with TNF-? and IL-1?. American Journal of Renal Physiology, 298, 779?787.  Kappelhoff, R., auf dem Keller, U., & Overall, C. M. (2010). Matrix Metalloproteinase Protocols. (I. M. Clark, Ed.) (Vol. 622). Totowa, NJ: Humana Press. doi:10.1007/978-1-60327-299-5  Kappelhoff, R., Wilson, C. H., & Overall, C. M. (2008). The CLIP-CHIP?: A Focused Oligonucleotide Microarray Platform. In The Cancer Degradome. D. Edwards (Ed.), (pp. 1?19). Springer Science.  Kashuba, E., Bailey, J., Allsup, D., & Cawkwell, L. (2013). The kinin-kallikrein system: physiological roles, pathophysiology and its relationship to cancer biomarkers. Biomarkers: Biochemical Indicators of Exposure, Response, and Susceptibility to Chemicals, 18(4), 279?296. doi:10.3109/1354750X.2013.787544  Kieswetter, K., Schwartz, Z., Dean, D. D., & Boyan, B. D. (1996). The Role of Implant Surface Characteristics in the Healing of Bone. Critical Reviews in Oral Biology & Medicine, 7(4), 329?345. doi:10.1177/10454411960070040301  Kim, H., Murakami, H., Chehroudi, B., Textor, M., & Brunette, D. M. (2006). Effects of surface topography on the connective tissue attachment to subcutaneous implants. International Journal of Oral and Maxillofacial Implants, 21(3), 354?365.  Kohavi, D., Klinger, A., Steinberg, D., & Sela, M. N. (1995). Adsorption of salivary proteins onto prosthetic titanium components. The Journal of Prosthetic Dentistry, 74(5), 531?534.  Kulangara, K., & Leong, K. W. (2009). Substrate topography shapes cell function. Soft Matter, 5(21), 4072?4076. doi:10.1039/B910132M   88  Lakes, R. S. (2007). Biomaterials: An Introduction. (J. Park & R. S. Lakes, Eds.) (3rd ed.). New York: Springer Science.  Lang, N. P., Jepsen, S., Working Group 4. (2009). Implant surfaces and design (Working Group 4). Clinical Oral Implants Research, 20 Suppl 4, 228?231. doi:10.1111/j.1600-0501.2009.01771.x  Lattin, J. E., Greenwood, K. P., Daly, N. L., Kelly, G., Zidar, D. A., Clark, R. J., et al. (2009). Beta-arrestin 2 is required for complement C1q expression in macrophages and constrains factor-independent survival. Molecular Immunology, 47(2-3), 340?347. doi:10.1016/j.molimm.2009.09.012  Le Gu?hennec, L., Soueidan, A., Layrolle, P., & Amouriq, Y. (2007). Surface treatments of titanium dental implants for rapid osseointegration. Dental Materials, 23(7), 844?854. doi:10.1016/j.dental.2006.06.025  Leonelli, S., Diehl, A. D., Christie, K. R., Harris, M. A., & Lomax, J. (2011). How the gene ontology evolves. BMC Bioinformatics, 12(1), 325. doi:10.1186/gb-2005-6-5-r46  Lingen, M. W. (2001). Role of Leukocytes and Endothelial Cells in the Development of Angiogenesis in Inflammation and Wound Healing. Archives of Pathology & Laboratory Medicine. 125, 67-71  Lockhart, D. J., & Winzeler, E. A. (2000). Genomics, gene expression and DNA arrays. Nature, 405(6788), 827?836. doi:10.1038/35015701  Mantovani, A., Biswas, S. K., Galdiero, M. R., Sica, A., & Locati, M. (2012). Macrophage plasticity and polarization in tissue repair and remodelling. The Journal of Pathology, 229(2), 176?185. doi:10.1002/path.4133  Mantovani, A., Sica, A., & Locati, M. (2005). Macrophage Polarization Comes of Age. Immunity, 23, 344?346.  Martinez, F. O., Helming, L., & Gordon, S. (2009). Alternative activation of macrophages: an immunologic functional perspective. Annual Review of Immunology, 27, 451?483. doi:10.1146/annurev.immunol.021908.132532  Maurer, M., & Stebut, von, E. (2004). Macrophage inflammatory protein-1. The International Journal of Biochemistry & Cell Biology, 36(10), 1882?1886. doi:10.1016/j.biocel.2003.10.019  McQuibban, G. A., Gong, J.-H., Wong, J. P., Wallace, J. L., Clark-Lewis, I., & Overall, C. M. (2002). Matrix metalloproteinase processing of monocyte chemoattractant proteins generates CC chemokine receptor antagonists with anti-inflammatory properties in vivo. Blood, 100(4), 1160?1167.   89  Mendon?a, G., Mendon?a, D. B. S., Arag?o, F. J. L., & Cooper, L. F. (2008). Advancing dental implant surface technology ? From micron- to nanotopography. Biomaterials, 29(28), 3822?3835. doi:10.1016/j.biomaterials.2008.05.012  Michalec, L., Choudhury, B. K., Postlethwait, E., Wild, J. S., Alam, R., Lett-Brown, M., & Sur, S. (2002). CCL7 and CXCL10 Orchestrate Oxidative Stress-Induced Neutrophilic Lung Inflammation, 168, 846?852.  Misch, C. E. (2007). Contemporary Implant Dentistry. 3rd Edition. St Louis, MO, USA: Mosby.  Mosser, D. M., & Edwards, J. P. (2008). Exploring the full spectrum of macrophage activation. Nature Reviews Immunology, 8(12), 958?969. doi:10.1038/nri2448  Moura, C. C. G., Soares, P. B. F., Souza, M. A. de, & Zanetta-Barbosa, D. (2011). Effect of titanium surface on secretion of IL1? and TGF?1 by mononuclear cells. Brazilian Oral Research, 25(6), 500?505.  Murray, P. J., & Wynn, T. A. (2011a). Obstacles and opportunities for understanding macrophage polarization. Journal of Leukocyte Biology, 89(4), 557?563. doi:10.1189/jlb.0710409  Murray, P. J., & Wynn, T. A. (2011b). Protective and pathogenic functions of macrophage subsets. Nature Reviews Immunology, 11(11), 723?737. doi:doi:10.1038/nri3073  Niinomi, M. (1998). Mechanical properties of biomedical titanium alloys. Materials Science and Engineering: A, 231?236.  Nolan, T., Hands, R. E., & Bustin, S. A. (2006). Quantification of mRNA using real-time RT-PCR. Nature Protocols, 1(3), 1559?1582. doi:10.1038/nprot.2006.236  Ong, J. L., Carnes, D. L., & Bessho, K. (2004). Evaluation of titanium plasma-sprayed and plasma-sprayed hydroxyapatite implants in vivo. Biomaterials, 25, 4601?4606.  Park, J. Y., Gemmell, C. H., & Davies, J. E. (2001). Platelet interactions with titanium: modulation of platelet activity by surface topography. Biomaterials, 22, 2671?2682.  Paul, N. E., Skazik, C., Harwardt, M., Bartneck, M., Denecke, B., Klee, D., et al. (2008). Topographical control of human macrophages by a regularly microstructured polyvinylidene fluoride surface. Biomaterials, 29(30), 4056?4064. doi:10.1016/j.biomaterials.2008.07.010  Peinnequin, A., Mouret, C., Birot, O., Alonso, A., Mathieu, J., Claren?on, D., et al. (2004). Rat pro-inflammatory cytokine and cytokine related mRNA quantification by real-time polymerase chain reaction using SYBR green. BMC Immunology, 5(1), 3. doi:10.1186/1471-2172-5-3    90 Pfaffl, M. W. (2001). A new mathematical model for relative quantification in real-time RT?PCR. Nucleic Acids Research, 29(9), 2002?2007.  Pollard, J. W. (2004). Opinion: Tumour-educated macrophages promote tumour progression and metastasis. Nature Reviews Cancer, 4(1), 71?78. doi:10.1038/nrc1256  Porter, J. A., & Fraunhofer, von, J. A. (2005). Success or failure of dental implants? A literature review with treatment considerations. General dentistry, 53(6), 423?32; quiz 433, 446.  Puippe, J. C. (2003). Surface treatments of titanium implants. European Cells and Materials, 5(1), 32?33.  Pye, A. D., Lockhart, D., Dawson, M. P., & Murray, C. A. (2009). A review of dental implants and infection. Journal of Hospital Infection, 72, 104?110.  Qiagen. (2010). RNeasy Mini Handbook. Qiagen. Retrieved June 15, 2013, http://www.qiagen.com/Products/Catalog/Sample-Technologies/RNA-Sample-Technologies/Total-RNA/RNeasy-Mini-Kit#technicalspecification  Raghavendra, S., Wood, M. C., & Taylor, T. C. (2005). Early wound healing around endosseous implants: a review of the literature. International Journal of Oral and Maxillofacial Implants, 20(3), 425?431.  Ratner, B. D. (2001a). Replacing and renewing: synthetic materials, biomimetics, and tissue engineering in implant dentistry. Journal of Dental Education, 65(12), 1340?1347.  Ratner, B. D. (2001b). A Perspective on Titanium Biocompatibility. In Titanium in Medicine (pp. 1?12). Berlin, Heidelberg: Springer Berlin Heidelberg. doi:10.1007/978-3-642-56486-4_1  Ratner, B. D., Hoffman, A. S., & Schoen, F. J. (2004). Biomaterials science: a multidisciplinary endeavor in Biomaterials Science (2nd Edition) San Diego, CA, USA: Elsevier Academic Press  Refai, A. K., Textor, M., Brunette, D. M., & Waterfield, J. D. (2004). Effect of titanium surface topography on macrophage activation and secretion of proinflammatory cytokines and chemokines. Journal of Biomedical Materials Research Part A, 70A(2), 194?205. doi:10.1002/jbm.a.30075  Rhee, S. Y., Wood, V., Dolinski, K., & Draghici, S. (2008). Use and misuse of the gene ontology annotations. Nature Reviews Genetics, 9, 509?515.  Rich, A., & Harris, A. K. (1981). Anomalous preferences of cultured macrophages for hydrophobic and roughened substrata. Journal of Cell Science, 50, 1?7.  Rompen, E., Domken, O., Degidi, M., Farias Pontes, A. E., & Piattelli, A. (2006). The effect of material characteristics, of surface topography and of implant components and connections on   91 soft tissue integration: a literature review. Clinical Oral Implants Research, 17(S2), 55?67. doi:10.1111/j.1600-0501.2006.01367.x  Rubin, D. L., Shah, N. H., & Noy, N. F. (2008). Biomedical ontologies: a functional perspective. Briefings in Bioinformatics. 9(1), 75-90  Sahu, S. C. (2008). Toxicogenomics: A Powerful Tool for Toxicity Assessment (1st ed.). Chippenham, Wiltshire, UK: John Wiley & Sons, Ltd.  Salthouse, T. N. (1984). Some aspects of macrophage behavior at the implant interface. Journal of Biomedical Materials Research Part A, 18(4), 395?401. doi:10.1002/jbm.820180407  Savli, H., Szendr?i, A., Romics, I., & Nagy, B. (2008). Gene network and canonical pathway analysis in prostate cancer: a microarray study. Experimental & Molecular Medicine, 40(2), 176?185. doi:10.3858/emm.2008.40.2.176  Schenk, R. K., & Buser, D. (1998). Osseointegration: a reality. Periodontology 2000, 17(1), 22?35. doi:10.1111/j.1600-0757.1998.tb00120.x  Schmittgen, T. D., & Livak, K. J. (2008). Analyzing real-time PCR data by the comparative: C: T: method. Nature Protocols, 3(6), 1101?1108.  Schroeder, A., Mueller, O., & Stocker, S. (2006). The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC Molecular Biology 7(3)  Schuler, M., Kunzler, T. P., de Wild, M., Sprecher, C. M., Trentin, D., Brunette, D. M., et al. (2009). Fabrication of TiO 2-coated epoxy replicas with identical dual-type surface topographies used in cell culture assays. Journal of Biomedical Materials Research Part A, 88A(1), 12?22. doi:10.1002/jbm.a.31720  Sela, M. N., Badihi, L., & Rosen, G. (2007). Adsorption of human plasma proteins to modified titanium surfaces. Clin Oral Impl Res, 18, 630?638.  Shi, L., Reid, L. H., Jones, W. D., Shippy, R., Warrington, J. A., et al. (2006). The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nature Biotechnology, 24(9), 1151?1161. doi:10.1038/nbt1239  Sica, A., & Mantovani, A. (2012). Macrophage plasticity and polarization: in vivo veritas. Journal of Clinical Investigation, 122(3), 787?795. doi:10.1172/JCI59643DS1  Sidhu, A. S., Dillon, T. S., & Chang, E. (2008). Current Status of Biomedical Ontologies: Developments in 2007. Digital Ecosystems and Technologies, 538?541.    92 Simonis, P., Dufour, T., & Tenenbaum, H. (2010). Long-term implant survival and success: a 10-16-year follow-up of non-submerged dental implants. Clinical Oral Implants Research, 21(7), 772?777. doi:10.1111/j.1600-0501.2010.01912.x  Sinicropi, D., Cronin, M., & Liu, M.-L. (2006). Gene Expression Profiling Utilizing Microarray Technology and RT-PCR. In BioMEMS and Biomedical Nanotechnology (pp. 23?46). Boston, MA: Springer US. doi:10.1007/978-0-387-25843-0_2  Slots, J. (2013). Periodontology: past, present, perspectives. Periodontology 2000, 62(1), 7?19. doi:10.1111/prd.12011  Smith, B., Williams, J., & Schulze-Kremer, S. (2003). The Ontology of the Gene Ontology. AMIA Annual Symposium Proceedings, 2003, 609.  Stanford, C. M. (2010). Surface modification of biomedical and dental implants and the processes of inflammation, wound healing and bone formation. International Journal of Molecular Sciences, 11, 354?369.  Szymczak, W. A., & Deepe, G. S. (2009). The CCL7-CCL2-CCR2 axis regulates IL-4 production in lungs and fungal immunity. Journal of Immunology, 183(3), 1964?1974. doi:10.4049/jimmunol.0901316  Taborelli, M., Jobin, M., Fran?ois, P., Vaudaux, P., Tonetti, M., Szmukler-Moncler, S., et al. (1997). Influence of surface treatments developed for oral implants on the physical and biological properties of titanium. (I) Surface characterization. Clinical Oral Implants Research, 8(3), 208?216.  Takebe, J., Champagne, C. M., Offenbacher, S., Ishibashi, K., & Cooper, L. F. (2003). Titanium surface topography alters cell shape and modulates bone morphogenetic protein 2 expression in the J774A.1 macrophage cell line. Journal of Biomedical Materials Research Part A, 64A(2), 207?216. doi:10.1002/jbm.a.10275  Tan, K. S., Qian, L., Rosado, R., Flood, P. M., & Cooper, L. F. (2006). The role of titanium surface topography on J774A.1 macrophage inflammatory cytokines and nitric oxide production. Biomaterials, 27(30), 5170?5177. doi:10.1016/j.biomaterials.2006.05.002  Taylor, S., Wakem, M., Dijkman, G., Alsarraj, M., & Nguyen, M. (2010). A practical approach to RT-qPCR?Publishing data that conform to the MIQE guidelines. Methods, 50(4), S1?S5. doi:10.1016/j.ymeth.2010.01.005  Tichopad, A., Dzidic, A., & Pfaffl, M. W. (2002). Improving quantitative real-time RT-PCR reproducibility by boosting primer-linked amplification efficiency. Biotechnology Letters, 24(24), 2053?2056. doi:10.1023/A:1021319421153    93 Tsirogianni, A. K., Moutsopoulos, N. M., & Moutsopoulos, H. M. (2006). Wound healing: immunological aspects. Injury, 37 Suppl 1, S5?12. doi:10.1016/j.injury.2006.02.035  Uzan, B., Villemin, A., & Garel, J. M. (2008). Adrenomedullin is anti-apoptotic in osteoblasts  through CGRP1 receptors and MEK-ERK pathway. Journal of Cellular Physiology, 215, 122?128.  Valencia-Cruz, A. I., Uribe-Figueroa, L. I., Galindo-Murillo, R., Baca-L?pez, K., Guti?rrez, A. G., V?zquez-Aguirre, A., et al. (2013). Whole genome gene expression analysis reveals casiope?na-induced apoptosis pathways. PLOS ONE, 8(1), e54664. doi:10.1371/journal.pone.0054664  VanOtteren, G. M., Standiford, T. J., & Kunkel, S. L. (1994). Expression and regulation of macrophage inflammatory protein-1 alpha by murine alveolar and peritoneal macrophages. (thoracic). Am J Respir Cell Mol Biol, 10, 8?15.  Varin, A., & Gordon, S. (2009). Alternative activation of macrophages: Immune function and cellular biology. Immunobiology, 214, 630?641.  Villar, C. C., Ba, G. H., & Mills, M. P. (2012). Wound healing around dental implants. Endodontic Topics, 25, 44?62.  Wang, R. R., & Fenton, A. (1996). Titanium for prosthodontic applications: a review of the literature. Quintessence international (Berlin, Germany: 1985), 27(6), 401?408.  Waterfield, J. D., Ali, T. A., Nahid, F., Kusano, K., & Brunette, D. M. (2010). The effect of surface topography on early NF?B signaling in macrophages. Journal of Biomedical Materials Research Part A, 95A(3), 837?847. doi:10.1002/jbm.a.32857  Watson, J. D. (2008). Molecular Biology of the Gene (6 ed.). Menlo Park, CA: Benjamin-Cummings Publishing Company.  Wells, J. E. A., Rice, T. K., Nuttall, R. K., Edwards, D. R., Zekki, H., Rivest, S., & Yong, V. W. (2003). An adverse role for matrix metalloproteinase 12 after spinal cord injury in mice. Journal of Neuroscience: 23(31), 10107?10115.  Wennerberg, A., & Albrektsson, T. (2009). Effects of titanium surface topography on bone integration: a systematic review. Clinical Oral Implants Research, 20, 172?184. doi:10.1111/j.1600-0501.2009.01775.x  Wennerberg, A., & Albrektsson, T. (2010). On implant surfaces: a review of current knowledge and opinions. International Journal of Oral and Maxillofacial Implants 25(1), 63?74.  Werner, T. (2008). Bioinformatics applications for pathway analysis of microarray data. Current Opinion in Biotechnology, 19(1), 50?54. doi:10.1016/j.copbio.2007.11.005   94  Wieland, M. (1999, July 29). Experimental Determination and Quantitative Evaluation of the Surface Composition and Topography of Medical Implant Surfaces and Their Influence on Osteoblastic Cell-Surface Interactions. (N. D. Spencer, M. Wolfensberger, M. Textor, & D. M. Brunette). PhD Thesis, Swiss Federal Institute of Technology Zurich. http://e-collection.library.ethz.ch/view/eth:23240  Wieland, M., Textor, M., Spencer, N. D., & Brunette, D. M. (2001). Wavelength-dependent roughness: a quantitative approach to characterizing the topography of rough titanium surfaces. International Journal of Oral and Maxillofacial Implants, 16(2), 163?181.  Williams, D. F. (1987). Tissue-biomaterial interactions. Journal of Materials science, 22(10), 3421?3445. doi:10.1007/BF01161439  Williams, D. F. (2008). On the mechanisms of biocompatibility. Biomaterials, 29(20), 2941?2953. doi:10.1016/j.biomaterials.2008.04.023  Wong, M. L., & Medrano, J. F. (2005). Real-time PCR for mRNA quantitation. Biotechniques. 39(1), 1-11  Zarb, G. A., & Schmitt, A. (1990a). The longitudinal clinical effectiveness of osseointegrated dental implants: The Toronto study. Part II: The prosthetic results. The Journal of Prosthetic Dentistry, 64(1), 53?61.  Zarb, G. A., & Schmitt, A. (1990b). The longitudinal clinical effectiveness of osseointegrated dental implants: The Toronto study. Part III: Problems and complications encountered. The Journal of Prosthetic Dentistry, 64(1), 185?194.  Zaveri, T. D., Dolgova, N. V., Chu, B. H., Lee, J., Wong, J., Lele, T. P., et al. (2010). Contributions of surface topography and cytotoxicity to the macrophage response to zinc oxide nanorods. Biomaterials, 31(11), 2999?3007. doi:10.1016/j.biomaterials.2009.12.055     95 Appendices  Appendix A  Supplemental Data A.1 qPCR Results for CysC, HSP, CTNND1, ADM and MMP12  Relative Expression of CysC on P and SLA Surfaces at Day 5   Relative Expression of HSP on P and SLA Surfaces at Day 5 0	 ?0.2	 ?0.4	 ?0.6	 ?0.8	 ?1	 ?1.2	 ?1.4	 ?1.6	 ?1.8	 ?2-????CT	 ?Polished	 ? SLA	 ?0	 ?0.2	 ?0.4	 ?0.6	 ?0.8	 ?1	 ?1.2	 ?2-????CT	 ?Polished	 ? SLA	 ?  96   Relative Expression of CTNND1 on P and SLA Surfaces at Day 5     Relative Expression of ADM on P and SLA Surfaces at Day 5  0	 ?0.2	 ?0.4	 ?0.6	 ?0.8	 ?1	 ?1.2	 ?2-????CT	 ?	 ?Polished	 ? SLA	 ?0	 ?0.2	 ?0.4	 ?0.6	 ?0.8	 ?1	 ?1.2	 ?1.4	 ?2-????CT	 ?Polished	 ? SLA	 ?  97   Relative Expression of MMP12 on P and SLA Surfaces at Day 1 Note overlapping SD error bars 0	 ?0.5	 ?1	 ?1.5	 ?2	 ?2.5	 ?3	 ?3.5	 ?4	 ?4.5	 ?2-????CT	 ?Polished	 ? SLA	 ?  98 A.2 ANOVA Tables Groups Sample size Sum Mean Variance     5x105 3 10,816.6 3,605.53333 16,401.85333   2x105 3 5,006. 1,668.66667 28,341.17333   1x105 3 2,033.8 677.93333 3,300.05333          ANOVA       Source of Variation SS df MS F p-level F crit Between Groups 13,303,846.78222 2 6,651,923.39111 415.37242 0. 8.05209 Within Groups 96,086.16 6 16,014.36           Total 13,399,932.94222 8             One-Way ANOVA: Cell Seeding Concentration for 1-day Cell Proliferation Study, P Surfaces  Groups Sample size Sum Mean Variance     5x105 3 7,930.45 2,643.48333 53,655.55083   2x105 3 3,782. 1,260.66667 40,578.17333   1x105 3 1,275.2 425.06667 2,852.49333          ANOVA       Source of Variation SS df MS F p-level F crit Between Groups 7,531,781.80056 2 3,765,890.90028 116.36742 0.00002 8.05209 Within Groups 194,172.435 6 32,362.0725           Total 7,725,954.23556 8      One-Way ANOVA: Cell Seeding Concentration for 5-day Cell Proliferation Study, P Surfaces         99 Groups Sample size Sum Mean Variance     2x104 3 5,099.47 1,699.82333 17,687.95763   1x104 3 3,734.3 1,244.76667 124,193.50333   5x103 3 645.15 215.05 6,911.4325          ANOVA       Source of Variation SS df MS F p-level F crit Between Groups 3,471,944.83487 2 1,735,972.41743 35.00112 0.00049 8.05209 Within Groups 297,585.78693 6 49,597.63116           Total 3,769,530.6218 8      One-Way ANOVA: Cell Seeding Concentration for 1-Day Cell Proliferation Study, SLA Surfaces    Groups Sample size Sum Mean Variance     2x104 3 4,736.15 1,578.71667 36,724.16583   1x104 3 1,525.25 508.41667 28,752.27083   5x103 3 118.84 39.61333 1,209.10963          ANOVA       Source of Variation SS df MS F p-level F crit Between Groups 3,734,157.72602 2 1,867,078.86301 83.99476 0.00004 8.05209 Within Groups 133,371.0926 6 22,228.51543           Total 3,867,528.81862 8          One-Way ANOVA: Cell Seeding Concentration for 5-Day Cell Proliferation Study, SLA Surfaces  100 Appendix B  Copyright Permission Figure 1.2: With kind permission from John Wiley and Sons. Barth, K.A., Waterfield, J.D. and D.M. Brunette. The effect of surface roughness on RAW264.7 macrophage phenotype. 2013. J. Biomed Mater Res Part A: 000-000 Figure 1.3 With kind permission from J.E. Davies. Davies, J.E. Understanding peri-implant endosseous healing. 2003. Journal of Dental Education. 67(8): 932-949. Available at http://www.ecf.utoronto.ca/~bonehead/             

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0074276/manifest

Comment

Related Items