UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Gene expression profiles during scarless healing of human palatal mucosa Habijanac, Tara Marie 2005

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

Item Metadata


831-ubc_2005-0214.pdf [ 9.07MB ]
JSON: 831-1.0091955.json
JSON-LD: 831-1.0091955-ld.json
RDF/XML (Pretty): 831-1.0091955-rdf.xml
RDF/JSON: 831-1.0091955-rdf.json
Turtle: 831-1.0091955-turtle.txt
N-Triples: 831-1.0091955-rdf-ntriples.txt
Original Record: 831-1.0091955-source.json
Full Text

Full Text

GENE EXPRESSION PROFILES DURING SCARLESS HEALING OF HUMAN PALATAL MUCOSA By TARA MARIE HABIJANAC DDS, The University of Alberta, 2000 A THESIS IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF MASTER OF SCIENCE In THE FACULTY Of GRADUATE STUDIES (Dental Science)  THE UNIVERSITY OF BRITISH COLUMBIA April 2005  © Tara Marie Habijanac, 2005  11  ABSTRACT Scarless healing of adult wounds is not well understood. Unlike human skin, oral masticatory mucosa heals largely by regeneration rather than scar formation. For a better understanding of healing by regeneration, we compared gene expression profiles of human mucosal wounds 1-, 3-, and 7-day-postwounding to that of normal non-wounded mucosa in three healthy volunteers. Clinical and histological healing was followed in additional volunteers for up to 60 days. The majority of the mucosal wounds healed with regeneration as the wounds at 60 days or later were no longer clinically or histologically detectable. At day one, a large number of genes showed a change in expression (988 genes). Expression levels of extracellular matrix proteins were down-regulated while genes in several categories including cytokines, cell motility and immune response were up-regulated. Gene ontology analysis of 3-day-old wounds revealed up-regulation of many categories including proteinase inhibitors, calcium binding categories. In 7-day-old wounds, genes for the extracellular matrix proteins and cytoskeletal structural proteins were highly expressed. Interestingly, some genes with the largest increase in expression showed no major  decline from day one to seven. These included matrix  metalloproteinase-1, serine proteinase inhibitor B l and a novel gene in the context of wound healing, carcinoembryonic antigen-related cell adhesion molecule-6 (Ceacam-6). In early wounds, Ceacam-6 protein was expressed by basal and suprabasal keratinocytes at the wound margin.  After 7 days post-wounding, expression of Ceacam-6 in  keratinocytes declined but was induced in granulation tissuefibroblaststhat continued to express Ceacam-6 until day 28. The function of Ceacam-6 in oral cells during wounding remains unravelled but could involve regulation of self-defence against bacteria and  Ill  inhibition of anoikis. To conclude, the present study represents one of the first large-scale gene expression studies comparing different phases of regenerating human mucosal wounds. Future studies are needed to directly compare gene expression profiles of healing scar-forming skin wounds to regenerating oral wounds.  TABLE OF CONTENTS  ABSTRACT TABLE OF CONTENTS LIST OF FIGURES LIST OF TABLES ACKNOWLEDGEMENTS ABBREVIATIONS INTRODUCTION CHAPTER ONE  Review of the Literature  1.1. Normal composition and function of oral mucosa 1.2. Overview of wound healing 1.3. Fibrotic and anti-fibrotic mechanisms in wound healing 1.4. Microarray techniques and wound healing CHAPTER TWO CHAPTER THREE  Aim of the study Materials and Methods  Experimental wounding Transcriptional profiling Routine histology and immunohistochemistry CHAPTER FOUR  Results  Clinical healing Transcriptional profiling Histology and immunohistochemistry  CHAPTER FIVE  Discussion  CHAPTER SIX  Conclusion and Future Directions  BIBLIOGRAPHY  vi LIST OF FIGURES Figure 1.  Wound Preparation, Biopsy Technique and Armamentarium Used. Initial wound made on healthy palatal gingiva away from the gingival margins of teeth with a standardized double blade holder. Initial tissue sample then removed (A,B,C). After initial wound has healed for appropriate time, a 4 mm punch biopsy is placed over wound and the wound sample is removed carefully (D,E,F). Armamentarium (G) includes mirror, periodontal probe, tissue forceps, double blade holder with #15 blades, and 6900N mini-blade. Double blade Holder (H) and 6900N miniblade (I). Initial tissue sample with 10mm X 2mm X 2mm dimensions (J) and biopsy of wound sample with 4 mm diameter and 2 mm depth (K). 43  Figure 2.  Agglomerative Hierarchical Clustering of Gingival Wounds  Figure 3.  Initial palatal wound (A,C,E) and corresponding 1 -day-old wound (B), 3day-old wound (D), and 5-day-old wound (F). 48  Figure 4.  Initial palatal wound (A,C,E), and corresponding 7-day-old wound (B), 14-day-old wound (D), and 28-day-old wound (F). 49  Figure 5.  Baseline palatal gingiva prior to initial wound preparation (A,C,E), and corresponding area following more than 6 months healing after wound biopsy (B,D,F). Clinical differences are evident between A and B (area surrounded by arrows), but not evident between C and D, or E and F. 50  Figure 6.  Significant alterations in gene expression during wound healing  Figure 7.  Over-represented Ontologies (p <= 0.01). Day 1 Up-regulated genes  Figure 8.  Over-represented Ontologies (p <= 0.01). Dayl Down-regulated genes 55  Figure 9.  Over-represented Ontologies (p <= 0.01). Day 3 Up-regulated genes  Figure 10.  Over-represented Ontologies (p <= 0.01). Day 3 Down-regulated genes 57  Figure 11.  Over-represented Ontologies (p <= 0.01). Day 7 Up-regulated genes  Figure 12.  Over-represented Ontologies (p <= 0.01). Day 7 Down-regulated genes 59  Figure 13.  Hematoxylin & Eosin staining of healthy gingiva (A,B,C), 1-day-old wound (D,E,F), 3-day-old wound (G,H,I), and 5-day-old wound (J,K,L). Arrows in panel J mark epithelium that has started to proliferate prior to migration over the wound bed. EP, epithelium; CT, connective tissue; FC, fibrin clot; GT, granulation tissue. 68  45  53 54  56  58  Vll  Figure 14.  Hematoxylin and Eosin Staining of 7-day-old wound (A,B,C), 14-day-old wound (D,E,F), 28-day-old wound (G,H,I), and 60-day-old wound (J,K,L). Arrows in panel B show that the epithelium has completely covered the underlying granulation tissue at 7 days. The wound is no longer evident by 60 days. EP, epithelium; CT, connective tissue; GT, granulation tissue. 69  Figure 15.  Immunostaining of Ceacam-6 in oral gingival wounds. Healthy gingiva (A,B,C) does not express Ceacam-6 in either epithelium or connective tissue. In panel B and C, the white line divides the epithelium and connective tissue. The 1-day-old wound (D,E,F) shows positive staining for Ceacam-6 in the basal and spinous layers of the epithelium (denoted by arrows in E and F). This staining begins distant from the wound margin and continues to the epithelium adjacent to the wound margin where epithelial cells are beginning to proliferate. The 3-day -old wound (G,H,I) and the 5-day-old wound (J,K,L) reveal less staining in the basal layer of epithelium, but persistent staining in the suprabasal layers and the proliferating epithelium adjacent to the wound margin (denoted by arrows in G, H, J, and L). 70  Figure 16.  Immunostaining of Ceacam-6 in oral gingival wounds. The 7-day-old wound (A,B,C) shows some positive staining for Ceacam-6 in the new suprabasal layers of epithelium and an increase in staining of cells in the connective tissue (denoted by arrows in B and C). The 14-day-old wound continues to show expression of Ceacam-6 in cells of the connective tissue (denoted by arrows in F), but staining is no longer positive in the epithelium. The 28-day-old wound (G,H,I) and the 60-day-old wound (J,K,L) no longer exhibit positive staining for Ceacam-6 in either the epithelium or the connective tissue. The outline in panels H, K, and L divide the epithelium and connective tissue. 71  Vlll  LIST OF TABLES Table 1. Table 2. Table 3. Table 4. Table 5. Table 6.  Significantly Upregulated Genes at Day 1 Wound Compared to Baseline 60 Significantly Downregulated Genes at Day 1 Wound compared to Baseline 61 Significantly Upregulated Genes at Day 3 Wound Compared to Baseline 62 Significantly Downregulated Genes at Day 3 Wound Compared to Baseline 63 Significantly Upregulated Genes at Day 7 Wound Compared to Baseline 64 Significantly Downregulated Genes at Day 7 Wound Compared to Baseline 65  ix  ACKNOWLEDGEMENTS I would like to thank Dr. Hannu Larjava for all of his guidance, expertise, and patience during my studies. I would also like to thank Dr. Lari Hakkinen for all of his suggestions and help especially in the laboratory. I would like to thank Dr. Humphrey Gardner for all of his help regarding the Affyrnetrix microarrays and Ms. Suzanne Szak for her expertise in statistical analysis of microarray data. I am very thankful for Mr. Cristian Sperantia for the technical support throughout the past few years. I would like to also thank my many friends who were kind enough to be volunteers for this project. And lastly, to my family, and especially my husband... It may have taken many years to complete, but you were always there to encourage me!  ABBREVIATIONS Ceacam-6  carcinoembryonic antigen-related cell adhesion molecule-6  TGF-oc, -p  transforming growth factor -a, -(3  SB  stratum basale  SS  stratum spinosum  SG  stratum granulosum  SC  stratum corneum  PDGF  platelet derived growth factor  FMLP  N-formyl-methionyl-leucyl-phenylalanine  TNF-a  tumor necrosis factor-a  IL-1,-4  interleukin -1,-4  FGF  fibroblast growth factor  MMP  matrix metalloproteinase  TIMP  tissue inhibitor of matrix metalloproteinase  EGF  epidermal growth factor  KGF  keratinocyte growth factor  CTGF  connective tissue growth factor  ECM  extracellular matrix  VEGF  vascular endothelial growth factor  GAG  glycosaminoglycan  IGF  insulin growth factor  HA  hyaluronic acid  HGF  hepatocyte growth factor  1 INTRODUCTION Wound healing of damaged skin or mucosa is a complex multi-step process that is essential for the survival of all organisms. Human adult wound healing involves four major overlapping processes including inflammation, re-epithelialization, granulation tissue formation and angiogenesis, and tissue remodelling. The ultimate goal of wound healing is complete restoration of both the structure and function of the damaged tissues (regeneration). This type of healing can only be seen in certain adult vertebrates, such as newts, where complete regeneration of a missing appendage is indeed possible [1]. Interestingly, the early fetus of many mammals have the ability to heal skin wounds without scar formation [2]. In contrast, however, adult cutaneous wound healing results in scar-formation characterised by an abundance of collagen and disorganized fiber structure. Several cellular and molecular mechanisms have been proposed to explain why adult wounds form scars compared to scarless fetal wound healing [3]. These include increased levels of hyaluronan, less differentiated fibroblasts, lack of fibrin clot and a different inflammatory response between fetal and adult wounds. Fetal wounds demonstrate fewer inflammatory cells that are also cleared faster than in adult skin wounds [4]. As a consequence of this altered inflammatory response and the fetal tissue itself, fetal wounds contain less profibrotic cytokines such as TGFB-1 and -2 and more anti-fibrogenic cytokines such as TGFB-3, that could explain in part the scarless fetal wound healing [3]. Very little is known about cellular mechanisms and gene expression in adult tissues that do not scar. Gingival tissue and palatal masticatory mucosa have maintained their ability to regenerate and heal without forming a scar [5-7]. Some potential reasons for preferential oral wound healing may include a unique and moist  2 environment, the presence of oral bacteria, and fetal-like properties of oral cells [8, 9]. In the present study, we performed transcriptional profiling of human masticatory mucosal wounds at early time points (1-7 days) to screen thousands of genes without prior knowledge or bias for which genes may be involved in scarless healing. In addition, we followed healing for up to 60 days by histology, and clinically for up to 6 months. The results demonstrate that gingival wounds heal both histologically and clinically without scars. Hundreds of genes including well-known wound-associated genes and novel genes in this context showed significantly altered temporal expression during early healing.  3 CHAPTER I Review of the Literature 1.1. Normal composition and function of oral mucosa Oral mucosa can be divided into three basic categories: lining mucosa, which is found over mobile structures such as lips, cheeks, soft palate, alveolar mucosa, the vestibular fornix and floor of the mouth; masticatory mucosa, which covers the gingiva and the hard palate; and specialized mucosa, which is located on the dorsum of the tongue [10]. This review will focus specifically on masticatory mucosa. Oral masticatory mucosa is comprised of a stratified squamous epithelium and underlying connective tissue, including a lamina propria and submucosa. A basal lamina is interposed between the epithelium and the lamina propria. The mucosal epithelium forms rete ridges alternating with projecting connective tissue papillae. Much of the research on stratified squamous epithelium and basal laminae are based on the cutaneous model, which hold many similarities to masticatory mucosa. In the cutaneous model, the epithelium, or epidermis can be divided into four distinct layers. These layers are: the stratum basale (SB), stratum spinosum (SS), stratum granulosum (SG) and the stratum corneum (SC). The stratum basale is a single layer of columnar basal cells that attaches to a basement membrane by hemidesmosomes. These cells are primarily undifferentiated epidermal stem cells characterized by a high nuclear to cytoplasmic ratio, and cell organelles such as mitochondria and keratinfilaments(primarily cytokeratin 5 and 14) [11]. The basal cells synthesize and secrete type IV collagen, type VII collagen, laminin, perlecan, and cytokines [11,12].  4 The stratum spinosum has a spiny appearance due to an abundance of desmosomes[l 1]. In addition to the organelles found in the SB, the SS also contains lipidrich lamellar bodies known as membrane-coating granules. There is an increase in cellular keratin filaments with cytokeratins 1 and 10 being the biological markers for this layer [11, 12]. In the more superficial layers of the SS, the cells begin to flatten and elongate. The stratum granulosum is characterized by dark-staining keratohyalin granules. These granules consist of several proteins including filaggrin, loricrin, and involucrin (precursors of the cell envelope), while synthesis of cytokeratins 1 and 10 decline [12]. Keratohyalin granules become progressively larger in the upper granulocytes thus reflecting a quantitative increase in keratin synthesis[l 1]. The cells in this layer become more differentiated as indicated by an increase in protein synthesis and an increase in membrane-coating granules in the outer layers of the SG. These lamellar bodies play a role in the formation of the lipid permeability barrier of epidermis [13]. In vitro studies have shown that keratinocyte differentiation appears to be induced by an increase in cytosolic calcium [14]. It is thought that a rapid influx of calcium causes the cells of the stratum granulosum to transform into corneocytes (terminally differentiated keratinocytes), which are the cells that make up the last and most superficial layer of the epidermis, the stratum corneum [15]. Corneocytes are chemically and physically resistant cells that are flat, anucleated and packed with keratin filaments. They are surrounded by a cell envelope of cross-linked proteins (involucrin, loricrin, filaggrin, desmoplakin-like protein, and cysteine-rich protein) and covalently bound lipids [16]. The corneocytes are interconnected by desmomal attachments. The human  5  skin often has about 18-21 cell layers in the SC. Each cell may differ in their thickness, keratin packing, and number of desmosomes depending on the location in the body and the SC. The structure and differentiation of mucosal epithelium may hold many similarities to skin; however, differences do exist. Most obviously, mucosa is devoid of hair follicles and sweat glands. In addition, there is considerable site variability of oral mucosal differentiation ranging from a fully keratinized epithelium on the hard palate and gingiva (epidermal-like), to a non-keratinized epithelium covering the buccal mucosa [17]. Variations in epithelial thickness, cell proximity and arrangement, as well as the depth and number of rete ridges and connective tissue papillae also exist [18]. Cell turnover is much faster in the oral cavity than in the epidermis, showing over four to five times the rate of epithelial turnover than in the skin [12]. Interposed between the epithelial layer and connective tissue is the basement membrane or basal lamina. This region is a highly specialized interface functioning as a barrier to cells and some large molecules and as mechanical support for the adhesion of the epithelium to the underlying dermis or lamina propria [19]. The basal lamina also has important roles for the migration of keratinocytes during wound healing (refer to section 1.2). Under electron microscopy, this zone can be divided into four layers from superficial to deep: the basal cell plasma membrane with hemidesmosomes, an electron lucent region known as the lamina lucida, an electron dense region referred to as the lamina densa, and a reticular layer consisting of various fibrillar and nonfibrillar proteins and anchoring fibrils. Currently, the basal lamina has been well defined with over twenty macromolecules characterized at the biological and genomic level. Some key molecules  6 in each layer include: bullous pemphigoid antigen of 230 kDa and 180kDa and the a6f}4 integrin in the hemidesmosome complex, laminin 5 in the lamina lucida, type IV collagen in the lamina densa, and type VII collagen in the anchoring fibrils [20]. The connective tissue layer, or lamina propria, of masticatory mucosa is mainly fibrous in nature. The primary cell type is thefibroblast,which plays a major role in the development, maintenance and repair of gingival connective tissue [9]. The primary function of the fibroblast is to synthesize the extracellular matrix components of the connective tissue. In addition,fibroblastsplay a role in tissue homeostasis through collagen phagocytosis and production of collagenases, which allows tissue remodelling. Type I collagen is the primary collagen found in the connective tissue and is arranged in either large, dense bundles of thick fibers or loose, short, thinfibersarranged in a fine reticular network [21]. Type III collagen is also found in this fine reticular pattern and mostly adjacent to the basement membrane. Type V and VI collagen has also been localized in the gingival connective tissue [22]. Several proteoglycans have been identified in gingival connective tissue and include decorin, biglycan, and versican [2325]. These proteoglycans function as organizers of extracellular matrix, space fillers, and cell surface components [26]. Other key components of the gingival connective tissue include glycoproteins such asfibronectin,osteonectin, vitronectin, elastin, and tenascins [27-30]. When describing specifically masticatory or palatal mucosa, it has been shown that palatal mucosa has approximately two times the number of desmosomes than lining mucosa [31]. This is likely a response to a greater need for resistance against forces of mastication. The epithelial-connective tissue interface in the hard palate is also  7 characterized by more lamina propria papillae [12]. Although the hard palate epithelium more closely resembles epidermis than non-keratinized oral epithelium, quantitative comparison of epidermal and palatal stratum corneum have shown that palatal epithelium shows fewer lipid lamellae than epidermis, which can account for the relatively greater permeability of the palatal mucosa compared to skin [32]. The basal cells of the hard palate were also found to contain significantly moreribosomesand mitochondria than epidermal basal cells [31]. The ultrastructural appearance of epidermal cornified squames revealed keratin filaments appearing less dense than the surrounding matrix, whereas the keratin filaments of the hard palate could not be distinguished from the surrounding matrix [31]. In addition to the hard palate, the masticatory mucosa also includes oral gingival epithelium. Macroscopically, gingiva can be anatomically divided into three distinct regions: the free marginal gingiva, the interdental gingiva and the attached gingiva (which is continuous with the mucosa of the hard palate). Histologically, the gingiva consists of three different types of epithelium based on location and phenotypic differences. The junctional epithelium is located from the base of the gingival sulcus to a point approximately 2 mm coronal to the alveolar bone. The sulcular epithelium lines the gingival sulcus and the oral epithelium is located from the tip of the gingival crest to the mucogingival junction, comprising part of the free marginal and attached gingival (for a review on gingiva see Bartold et al., 2000) [9]. Similar to skin, the function of masticatory mucosa is largely protective in nature. In particular, the unique junctional epithelium plays a role in host defence against  8  bacterial infection as well as allowing the movement of substances between the connective tissue and the oral cavity [33].  1.2 Overview of Wound Healing Most tissues and organs of the body are capable of healing or repairing injuries. Healing can be divided into two different types: regeneration and repair. Regeneration, the most desirable healing outcome, is defined as the complete functional and structural restoration of the damaged tissue [34]. Repair is healing that occurs by replacement of the lost tissue by non-specialized connective tissue (granulation tissue) that can lead to scar formation [34]. Some animals, such as newts, are capable of epimorphic regeneration, or replacement of an amputated appendage [1]. Higher vertebrates, however, do not possess this ability for epimorphic regeneration and in most cases healing occurs through repair. Wound healing is a dynamic, interactive process that involves soluble mediators (growth factors and cytokines), blood elements, extracellular matrix molecules, and the participation of several cell types including inflammatory, epithelial, endothelial cells and fibroblasts [35]. Wound healing will occur as a response to tissue injury incurred by trauma, microbes or foreign materials, and can be described in four phases: hemostasis and inflammation, re-epithelialization, granulation tissue formation and angiogenesis, and tissue remodeling. These phases do not occur in sequential order, but instead show considerable overlap. In skin and mucosal wound healing, the goal is to restore the protective function of these tissues in an efficient manner in order to prevent microbial access into the tissues. 1.2.i Inflammation  9 The inflammatory phase can be divided into early and late inflammation. Early inflammation begins immediately upon tissue injury resulting in the damage of blood vessels and the extravasation of blood constituents. Blood clotting following extravasation will occur due to the surface activation of Hageman factor that can generate the fragments bradykinin and potent vasoactive agents as well as initiate the complement cascade [35]. These fragments will promote vasopermeability and vasodilation [36]. In addition, blood clotting will occur because of tissue pro-coagulant factors released from damaged cells, and surface membrane coagulation factors and phopholipids expressed on activated platelets and endothelial cells [37]. Following injury, the exposure of the subendothelial tissue and types IV and V collagen will promote the binding and aggregation of platelets [38]. Following activation, platelets will secrete adhesive proteins such as fibronectin, fibrinogen, thrombospondin and von Willebrand factor VIII [35]. The first three proteins act as ligands for platelet aggregation while von Willebrand factor functions as a mediator of adhesion between platelets and fibrillar collagen [39,40]. In addition, platelets release other soluble mediators such as serotonin, platelet-derived growth factor (PDGF), transforming growth factor- a (TGF-a), adenosine phosphate, platelet-activating factor, factor V, 1,2-hydroxy eicosatetranoic acid, and thromboxane A2, which promote vasocontriction, additional platelet aggregation and further growth factor release [35,41]. Platelets will undergo a series of functional and structural changes following activation, such as acquiring a spherical shape with spinous pseudopod projections that promotes increased aggregation with other platelets [41, 42]. The activated platelets undergo a rearrangement of their membrane phospholipids, which allow the interaction of clotting factors V and X [43]. This interaction will increase the  10  production of thrombin several fold, which catalyzes the formation of fibrin from fibrinogen and forms a meshwork between the aggregated platelets [43]. The aggregation of platelets and the formation of this fibrin-rich clot will fill in the discontinuity of the wounded tissue, preventing further bleeding and bacterial infection [35]. In addition, this clot will act as a provisional matrix for cell migration and a reservoir for growth factors, proteases, and protease inhibitors. Despite the important role that platelets play in the process of wound healing, a recent study comparing wound healing in thrombocytopenic to control mice revealed that despite a significant alteration in wound inflammation, the rate of the proliferative aspects of repair, including epithelialization, collagen synthesis and angiogenesis, were almost identical [44]. The level of key growth factors was also similar between the two groups of mice, which may suggest that other sources of cytokines and growth factors (such as the increased macrophage and T-cell population) compensated for the lack of platelets in this study. It is important that clot formation stay limited to the area of injury, and this occurs via intrinsic blood vessel activities. These include the production of prostacylin that inhibits platelet aggregation [45]; the binding of antithrombin III to thrombin, which stops its activity [46]; the production of protein C, which is an enzyme that degrades clotting factors V and VIII [47]; and the release of plasminogen activator, which initiates clot lysis through the conversion of plasminogen to plasmin [48]. Neutrophils are the key cell type involved in the early inflammatory phase and arrive at the wound site following platelets. Neutrophils cleanse the wounded area of foreign particles and bacteria by phagocytosis, enzymatic and oxygen radical mechanisms [49]. Neutrophils arrive from the surrounding broken blood vessels and are  11 recruited to the site of injury by various chemoattractants. Chemoattractants for neutrophils and other leukocytes include: fibrinopeptides cleaved from fibrinogen, fibrin degradation products, complement components C5a and C3a, bacterial products characterized by the synthetic peptide N-formyl-methionyl-leucyl-phenylalanine (FMLP), leukotriene B4, platelet activating factor, PDGF and platelet factor 4 [50-52]. Many of these chemoattractants also function in the adherence of neutrophils to endothelial cells by increasing the expression of CD11/CD18 receptors on the neutrophil's surface. In conjunction with sialyl lewis X, the CD11/CD18 complex mediates the adherence of the neutrophils to the blood vessel endothelium [53]. By way of a concentration gradient of chemoattractants, the adherent neutrophils begin the process of diapedesis or migration through the endothelial cells and into the site of inflammation. Neutrophil activation by chemoattractants also stimulates the release of elastase and collagenase, which facilitate cell penetration through the blood vessel basement membranes [35]. This action of diapedesis is further enhanced by the increased capillary permeability caused by various vasodilating agents such as serotonin, bradykinin, histamine and arachidonic acid metabolites [43]. The early inflammatory phase usually lasts for about three days, after which time the neutrophils become trapped within the wound clot and dessicated tissue. This eschar sloughs during tissue repair. Any neutrophils remaining within viable tissue will be phagocytosed by macrophages during the late inflammatory phase [54]. In cases of excessive wound contamination with foreign particles or bacteria, a persistent neutrophilinflammatory response may exist [35].  12 The late inflammatory phase overlaps with the early phase, beginning about one day after injury and persisting for about seven days. The primary cell type involved in this phase is the monocyte/macrophage. These cells may remain at the wound site for days to weeks [43]. Monocytes arrive after the neutrophils and are attracted to the site by chemoattractants. These chemoattractants include collagen fragments[55], elastin [56], fibronectin [57], enzymatically active thrombin [58] and TGF-0 [59]. The method of recruitment of monocytes into the site of injury is similar to the recruitment of neutrophils (adhesion to endothelium and migration through the blood vessels). The metamorphosis of monocyte into macrophage occurs by contact with the wound extracellular matrix [35]. The function of the wound macrophage is to debride the tissue through phagocytosis and digestion of pathogenic microbes, tissue debris, and remaining neutrophils [54]. Phagocytosis is stimulated by the interaction of the macrophages with the extracellular matrix proteins through integrin receptors. Campbell found that cultured macrophages and presumably wound macrophages also release proteolytic enzymes such as collagenase, which can facilitate wound debridement [60]. In addition, macrophages can induce the membrane-associated enzyme phospholipase which releases arachidonic acid from the cell membrane. Arachidonic acid metabolites such as prostaglandins and leukotrienes play a role in modulating the inflammatory response. Macrophages play an important role in growth factor and cytokine release. They are considered a major secretory cell with the ability to secrete more than 100 different molecules. Some of these include: colony stimulating factor-1 which is necessary for monocyte-macrophage survival, tumor necrosis factor-oc (TNF-a), a pro-inflammatory cytokine, PDGF, TGF-{3, TGF-a, interleukin-1 (IL-1), andfibroblastgrowth factor  13 (FGF)[61j. Macrophage-derived growth factors are necessary for the initiation of new tissue formation in wounds [35]. Through the release of soluble mediators, macrophages act on other inflammatory cells and on mesenchymal cells in order to regulate their phenotype and function [43]. It has been shown previously that macrophage-depleted animals have defective wound repair [43, 62]. In a more recent study, however, it was shown that the PU. 1 null mouse, which lacks functioning neutrophils and macrophages, was still capable of repairing skin wounds, with reduced scarring, yet in a similar time frame as the wild type mouse [63]. In this situation, other cells, such as fibroblasts, presumable took over some of the roles of the missing inflammatory cells, such as ' cytokine release and phagocytosis. Other inflammatory cell types involved in the late inflammatory phase include mast cells and lymphocytes. Both of these cells are involved in the release of various mediators. Lymphocytes arrive at the wound site at the same time as the monocyte. They are of more importance in the wound healing process if injury was caused by a foreign antigen or subject to secondary infection [64]. 1.2.U Re-epithelialization  Re-epithelialization of a wound will begin within hours after injury [65]. In order to cover the denuded wound, the epithelial cells will begin to migrate over the site. Migrating epithelial cells will arise from the wound periphery, and in cutaneous wounds, from the residual hair or sweat gland structures at the wound base [66]. During early epithelial cell migration, cell proliferation does not occur. In the later stages of reepithelialization (1-2 days after injury), proliferation of the epithelial cells will occur in order to provide an increased number of cells to cover the wound [66]. The basal cells of  14 the epithelium will respond to the wounding by dissolving their hemidesmosomal complexes and withdrawing their tonofilaments, which will release them from the basement membrane. The epithelial cells will also lose intercellular desmosomal attachments and form peripheral cytoplasmic actinfilamentswhich will allow for cell movement [67, 68]. The migrating keratinocytes also have a moreflattenedand elongated morphology [69]. The signals for these actions will arise from various soluble mediators released from the damaged cells, blood clot, and inflammatory cells. In order to migrate, keratinocytes will express matrixmetalloproteinases (MMP's), which are capable of degrading components of the extracellular matrix in order to facilitate migration below or through the clot. Epidermal cells, and other cell types such asfibroblastsand macrophages, can produce MMP's which function in the degradation of collagen, fibrin andfibronectin[70]. MMP activity is of great importance in reepithelialization and studies in cell culture have shown that blocking MMP activity prevents keratinocyte migration [70, 71]. In addition, the epithelial cells produce plasmin, which is activated by plasminogen activator, and functions in the degradation of the fibrin-fibronectin clot [72]. The presence of the plasminogen gene is critical for wound healing, as animals that lack this gene do not re-epithelialize [73]. Plasminogen activator further activates collagenase and therefore further enhances the degradation of extracellular matrix proteins [74]. The migration of epithelial cells along a provisional wound matrix composed mainly of fibrin,fibronectin,and vitronectin is highly regulated by integrin cell-surface receptors [66, 75]. Integrins are cell-surface transmembrane ap heterodimers that are responsible for most cell-extracellular matrix interactions in keratinocytes [76]. Their  15 functions include regulation of cellular function during growth, development, differentiation, and the immune response [77, 78]. During wound healing, there is a change in the expression of existing integrins of epithelial cells and increased expression of other integrins that are able to adhere to and migrate along a provisional wound matrix [78]. The common components of the basement membrane, such as collagen type IV and VII, laminin-1 and heparan sulphate proteoglycan are missing from beneath the migrating keratinocytes [78]. In their place, the migrating keratinocytes appear to be depositing fibronectin, tenascin-C and laminin-5 into the provisional wound matrix [79-81]. Fibronectin is a key component in the formation of the initial clot and granulation tissue during wound healing. In wounds, fibronectin can exist in three alternatively spliced variants that have different roles. Fibronectin containing the EDA or EDB regions (which are omitted from plasma fibronectin), are found to be upregulated in embryonic development and wounding, with EDA fibronectin being expressed by migrating keratinocytes [76]. Laminin-5 is thought to function in keratinocyte migration [82]. Tenascin-C is found in the connective tissue underneath the migrating epithelial front which suggests a role in re-epithelialization during wound healing [79]. Vitronectin is also present in the provisional matrix and has roles in cell adhesion, cell migration and invasion [83]. Migrating keratinocytes have been found to express 3 new fibronectin receptors, namely ct5pl, avP6, and avpi [78]. a5pi and avp6 are absent in normal healthy adult epithelium. The increased expression of fibronectin receptors is likely induced by cytokines such as TGF-pi [84]. Keratinocytes can also utilize other extracellular matrix  16 molecules for migration including tenascin-C and laminin-5 [85]. The keratinocytes recognize these molecules by other specific integrin receptors. There are two proposed models for epithelial cell migration. In the sliding model, the epithelial cells move as a sheet. In this sheet, the cells at the margin appear to be actively motile while the cells behind these appear to be dragged along [86]. If the attachment of the marginal cells to the wound matrix substrate is disturbed, the migrating sheet will withdraw. This model of epithelial cell migration has been demonstrated in tissue culture, for embryonic epithelial movement, amphibian wound closure and corneal wound closure [66]. It is currently accepted that most simple epithelium moves by the sliding model. For multilayered epithelium such as the skin and mucosa, a different model, known as the leapfrog model, has been proposed [87]. This model suggests that cells at the migrating front adhere to the substrate and the cells above and behind this front crawl over these newly adherent basal cells. The data for this model is mostly indirect, however there are many studies supporting this model [88, 89]. The specific stimulus for migration and proliferation during epithelialization has not yet been determined; however, it is known that migration is not dependent on cell proliferation [90]. It is thought that the absence of marginal neighbor cells or the "free edge" effect can signal both migration and proliferation of epithelial cells [65]. The local release of growth factors and cytokines by various cell types including epithelial cells, and the increased number of growth factor receptors may also stimulate these processes. Some of the key growth factors implicated in migration and proliferation of epithelium are: epidermal growth factor (EGF), TGF-ct, keratinocyte growth factor (KGF), and TGF-P,  which stimulates cell motility but inhibits epithelial cell proliferation [35]. TGF-  17 P plays a very important role in re-epithelialization as well as other important phases of wound healing. Three mammalian isotypes have been identified (TGF-pi, 2, 3) and all have been found in healing wounds of animals [91]. TGF-pl is the predominant isoform in wound keratinocytes while both TGF-pl and 2 have been found in wound fibroblasts and macrophages adjacent to the wound [92]. Once the migrating epithelial cells meet each other over the middle of the wound, the migration will stop according to the concept of "contact inhibition" [66]. This will occur around five to seven days following injury. Basement membrane proteins (type IV and VII collagen, laminin-1 and laminin-5 and heparan sulfate proteoglycans) will then begin to reappear in an ordered sequence from the margin of the wound inward in a "zipperlike" fashion in epidermal wounds [93]. In contrast, small gingival wounds show nucleation of the basement membrane in multiple sites at the same time [85]. Epithelial cells will revert to their normal stationary phenotype and firmly attach to the reestablished basement membrane through hemidesmosomes and type VII collagen fibrils [94]. The epithelial cells will differentiate and then reform their stratified structure. The basement membrane is usually completely reorganized by four weeks following injury [76]. 1.2. iii Granulation Tissue Formation and Angiogenesis  New stroma, which is also called granulation tissue, will begin to form beneath the epithelium approximately four days following injury [35]. Clinically this tissue has a somewhat granular appearance that bleeds easily [95]. This initial restructuring of the damaged tissue functions as a temporary barrier against the external environment. Granulation tissue is primarily composed of inflammatory cells (mainly macrophages),  18 fibroblasts, newly formed and forming blood vessels, and loose connective tissue. The provisional extracellular matrix promotes granulation tissue formation by providing a scaffold for contact guidance (fibronectin and collagen), low impedance for cell mobility (hyaluronic acid), a reservoir for cytokines, and direct signals to the cells through integrin receptors [96, 97]. Macrophages, blood vessels, andfibroblastswill occupy the wound space at the same time [98]. Macrophages function in providing a continuous source of cytokines, which are necessary to stimulatefibroplasiaand angiogenesis [65]. Fibroblasts are the major cellular component that supply the new extracellular matrix necessary to support cellular ingrowth, while blood vessels will carry oxygen and nutrients to the cells to sustain their metabolism [35]. Fibroplasia is defined as the components of the granulation tissue that are derived fromfibroblasts[95]. Fibroplasia is therefore a mixture of fibroblasts and extracellular matrix (ECM). The stimulus forfibroplasiaarises from the chemotactic, mitogenic, and modulatory activities of cytokines; many of which are released from platelets and macrophages. Fibroblasts themselves are also capable of producing cytokines and these cells can respond in an autocrine manner [99]. Asfibroblastsrespond to cytokines, many biological responses occur, including the induction of additional cytokines [100], and the modulation of cytokine receptor number or affinity [101]. Some of the key cytokines involved infibroplasiainclude: PDGF, basicfibroblastgrowth factor (b-FGF), TGF-a, TGF-p, and connective tissue growth factor (CTGF), which is induced by TGF-P and modulates fibroblast growth and extracellular matrix secretion [65,102]. Fibroblasts synthesize, deposit, and remodel the ECM, while the E C M regulates this function of the fibroblasts. Fibroblasts can usefibronectinin the provisional matrix for movement  19 through the wound. In addition,fibroblastsalso use fibrin and vitronectin as a substratum for movement [103]. The ability offibroblaststo bind tofibronectin,fibrin and vitronectin are mediated by integrin receptors. In studies onfibroblast-likecells lines, it was found that the ct4pi integrin receptor facilitates fibroblast movement while the fibronectin receptor a5pl retards movement, especially when overexpressed [104, 105]. It was also found that cc5pi was maximally expressed in wound fibroblasts after they ceased migration and established links to the extracellular matrix [106]. Both PDGF and TGF-P can stimulatefibroblaststo migrate [107] and upregulate integrin receptors [108]. These growth factors are likely responsible for promoting a migratoryfibroblastphenotype. In vitrofibroblastmigration has also been observed in response to other chemoattractants including: fragments of the 5 component of th  complement, types I, II, and III collagen-derived peptides, fibronectin fragments, elastinderived peptides and interleukin 4 (IL-4) [55, 109-111]. Fibroblasts will move towards a chemotactic gradient by extending lamellipodia in the direction of the stimulus while their opposite pole will then be released. This process of movement is very similar to haptotaxis [112,113]. In haptotaxis, thefibroblastwill extend lamellipodia randomly and one lamellipodium will become dominant so that the cell will spread in its direction. The cell will ooze forward and break old adhesions and the excess membrane at the trailing edge will then become available to resume random protruding lamellipodia. Extracellular matrix fibrils strongly influence the direction offibroblastmigration as the fibroblasts tend to align and migrate along and not across the substrata fibrils [114]. This process is called contact guidance [115]. Fibroblast migration can therefore be influenced by chemotactic, haptotactic, and contact guidance signals.  20 The movement of fibroblasts into the provisional matrix also requires proteases to create a pathway for migration. These enzymes are derived from fibroblasts and serum and include plasminogen activator and MMP's including collagenases, gelatinase A, and stromelysin [74]. Once the fibroblasts have migrated into the wound, their major function changes from migration to protein synthesis [106]. The provisional matrix is gradually replaced with a collagenous matrix, which is likely the result of TGF-pi [116]. The profibrotic phenotype of the fibroblast is characterized by an abundant rough endoplasmic reticulum and golgi apparatus [35]. Production of collagen appears maximal at seven days following injury [95]. The formation of new blood vessels, a process called angiogenesis, is necessary to sustain the formation of granulation tissue. This process relies upon the migration and mitogenesis of endothelial cells as well as the appropriate extracellular matrix in the wound bed. Initially, the acidic and basic forms of fibroblast growth factor were thought to initiate the induction of angiogenesis [65]. Many other molecules have also been found to have angiogenic properties including: vascular endothelial growth factor (VEGF), TGF-pi, angiogenin, angiotropin, angiopoietin 1, thrombospondin, interleukin-8, and lactic acid [35, 117, 118]. In addition, low oxygen tension may also stimulate angiogenesis [119]. Most of these molecules induce angiogenesis by stimulating the production of b-FGF or VEGF by macrophages or endothelial cells. The action of bFGF is critical during the first three days of wound repair whereas VEGF plays a more critical role in angiogenesis during the formation of granulation tissue [120]. bFGF is released by macrophages and injured tissue cells and stimulates endothelial cells to release  21 plasminogen activator and procollagenase [35]. VEGF can stimulate the proliferation of endothelial cells and cause increased vasopermeability [121]. In addition to the factors which induce angiogenesis, appropriate extracellular matrix and endothelial receptors for the provisional matrix are necessary. Proliferating endothelial cells adjacent to and within wounds deposit fibronectin within the vessel wall [122]. Angiogenesis also requires the expression of fibronectin receptors by endothelial cells [123] and therefore, a perivascular matrix abundant in fibronectin may act as a conduit for endothelial cell movement into the wound. The expression and activity of proteases are also necessary for angiogenesis [124]. A general hypothesis regarding the series of events leading to angiogenesis is as follows [65]: Injury causes tissue destruction and hypoxia and angiogenic factors, such as bFGF, are immediately released by macrophages (attracted to the wound site by fibronectin fragments and other degradation products). The production of VEGF from epidermal cells is stimulated by hypoxia. The production of plasminogen activator from endothelial cells will convert plasminogen to plasmin. Plasmin and activated collagenase will then digest the basement membrane. This fragmentation will allow endothelial cells to migrate into the wound site and form new blood vessels. Once the wound site is filled with new granulation tissue, angiogenesis will cease and many blood vessels will disintegrate by apoptosis. 1.2. iv Tissue  Remodelling  The process of extracellular matrix remodeling, cell maturation, and cell apoptosis create the phase of wound healing known as tissue remodelling. This phase of wound repair overlaps granulation tissue formation and can be occurring at the margins of a  22 wound while granulation tissue is still invading the middle of the wound space [35]. By day seven following injury, actin fibrils begin to appear within the cytoplasm of the fibroblasts [65]. In addition, intercellular and cell-matrix connections begin to appear [106]. There is a change in integrin expression on the fibroblast surfacefromctV containing receptors to oc5pi fibronectin receptors [95]. After day nine, the fibroblasts begin to arrange themselves in a radial array across the wound and lose their collagensynthesis phenotype to become the myofibroblast phenotype [95].The myofibroblast phenotype is characterized by abundant cytoplasmic microfilament bundles and asmooth muscle actin [125]. The differentiation of myofibroblasts occurs between day six and fifteen after wounding and is induced by TGF-P [125, 126]. After day fifteen, approximately 70% of fibroblasts in mature granulation tissue express a-smooth muscle actin [125]. Myofibroblasts are often present in tissues under mechanical stress, but are only occasionally found in normal tissue [127]. The attachment of myofibroblasts to fibronectin and collagen by integrins, the intercellular adherens junctions between myofibroblasts, and the cross-links between collagen bundles allow contraction by the myofibroblasts to be transmitted across the wound [128]. Myofibroblasts are aligned along the lines of contraction in the wound. Wound contraction by myofibroblasts parallels myofibroblast differentiation as it occurs between days seven and fourteen following injury and is stimulated by TGF-p and PDGF [129,130]. When wound contraction ceases, myofibroblasts will disappear by apoptosis and the scar tissue will become less cellular while newfibroblasts,typical of normal connective tissue, will emerge [125, 131]. It is still not entirely known whether these new fibroblasts  23 arise from connective tissue adjacent to the wound or from a primitivefibroblastpopulation that originally populated the wound. Apoptosis of myofibroblasts will begin at day twelve, peak at day twenty and resolve by day sixty following wounding [131]. Since myofibroblasts can persist in certain pathologic conditions (fibrotic lesions), it is important that they undergo apoptosis when wound contraction is completed. Collagen remodelling occurs during the transition from granulation tissue to scar formation and is dependant on both a low rate of synthesis and degradation of collagen [65]. The deposition of new extracellular matrix proteins during granulation tissue formation is stimulated by TGF-P [76]. Another potent modulator of protein synthesis is mechanical tension. Signals initiated by tension in a three dimensional collagen matrix require integrin-collagen interactions. Tension mediated from the matrix seems to balance production versus degradation. For example, collagen gels under tension exhibit a relatively high protein synthesis rate [132] while stress relaxation down-regulates type I collagen expression and up-regulates MMP-1 and MMP-13 expression [133]. A healing wound will generally gain about 20% of its strength by the end of the third week of healing, at which time fibrillar collagen has accumulated rapidly [65]. After this time, wounds will accumulate collagen and increased tensile strength at a much slower rate. Further increase in tensile strength also has more to do with collagen remodelling and formation of larger collagen bundles and alteration of intermolecular cross-links rather than increased collagen deposition [134]. In skin, wounded tissue will fail to gain the same strength as uninjured tissue, with a scar being about 70% as strong as intact skin [135].  24 1.3 Fibrotic and Anti-fibrotic Mechanisms in Wound Healing The healing of both cutaneous and oral mucosal wounds will proceed through the same basic phases of wound healing, including inflammation, epithelialization, granulation tissue formation and angiogenesis, and tissue remodelling. Adult cutaneous wound healing is considered reparative in nature as dermal integrity is re-established at the expense of scar formation. Scar formation can be defined as the macroscopic disturbance of normal skin structure and function as a consequence of wound repair, with such changes arising from alterations in epidermal, dermal, and subcutaneous tissues [136]. Cutaneous scarring is characterized by inflammation, a disorganized dermal collagen pattern, and failure of hair follicles and other dermal appendages to regenerate [3, 35]. In contrast to normal skin , where collagen bundles show a basket-weave like pattern with a random structure, scar tissue shows smaller bundles aligned in a parallel fashion to the epidermis [3]. Scar tissue following trauma and surgical procedures can be considered a significant clinical problem as it can impair normal growth and function and cause disfigurement. In contrast, fetal cutaneous wound healing is fundamentally different from the adult because early fetal wounds heal without scarring [137]. Oral mucosal wounds also differ from adult cutaneous wounds as healing is often more rapid and show reduced scar formation [138].  1.3.i Fibrogenic Mechanisms  Currently, there is no method to measure the exact amount of scarring either clinically or histologically. The difference between "normal" scarring and excessive scarring (such as keloids or hypertrophic scars) may only be the degree or extent of the  25 same basic process [139]. Scar formation can be essentially described as the permanent deposition of connective tissue made up of extracellular matrix components, with collagen being most abundant, and a lack of large blood vessels and elastin [140]. Initially the dermal wound is composed mainly of fibrin, fibronectin, and hyaluronan [141]. Following this, sulphated glycosaminoglycans (GAGs) and collagens are synthesized [35, 142]. The nature of the collagen in the later stages of the wound is predominantly type I [142].The final amounts of collagen and GAGs seem to contribute to the resilience and tensile strength of scar tissue [141]. Variousfibroticlesions and hypertrophic scars may accumulate an excess of extracellular matrix several years beyond initial injury [143]. Myofibroblasts, which arise fromfibroblastsand are characterized by a-smooth muscle actinfibers,are believed to play a role in wound contraction and scarring [144]. The transformation offibroblastto myofibroblast is influenced by the presence of mechanical stress, TGF-P 1, and cellularfibronectin(ED-A splice variant) [145]. Myofibroblasts are abundant in the wound between six and fifteen days [125] and function in extracellular matrix production and wound contraction [145]. Following wound contraction, the granulation tissue myofibroblasts, along with vascular cells, undergo apoptosis following the completion of epithelialization [131]. During pathologic fibrotic phenomena, such as keloids and hypertrophic scars, apoptosis of myofibroblasts may be diminished or aberrant in comparison with normal healing wounds [145]. This aberrant response may be due to altered expression of p53 and Bel family proteins [146, 147]. Keloids are described as thick scar tissues which have escaped the boundaries of the original wound [148]. They rarely exhibit erythema and are often asymptomatic.  26 Keloids can occur in all races, but a predilection for African Americans has been established [149]. Hypertrophic scars are different from keloids as they are often erythematous, nodular, and itchy. They also remain within the confines of the original wound [150]. Both of these conditions result in a prolonged inflammatory state, which therefore prolong the myofibroblast population. Eosinophils, in particular, will produce abundant irrflammatory cytokines (most notably TGF-P), which will promote the differentiation of myofibroblasts and synthesis of extracellular matrix [151]. Different cytoskeletal protein expression has been found in myofibroblasts during wound healing and fibrocontractive disorders. For example, vimentin is present in early granulation tissue myofibroblasts whereas both vimentin and cc-SM actin -type cells are found primarily during the retraction phase [125]. Granulation tissue myofibroblasts are devoid of desmin and SM-myosin heavy chains, whereas hypertrophic scar myofibroblasts express these proteins [125,152]. In addition, keloids do not have a-SM actin -expressing myofibroblasts when compared to hypertrophic scars [153], which may mean that a-SM actin is a key element in the pathogenesis of contraction in hypertrophic scars. It has also been shown that a-SM actin-expressing myofibroblasts are the key cells involved in collagen synthesis during pulmonary fibrosis [154]. The turnover of extracellular matrix is regulated by matrix metalloproteinases, which are secreted by keratinocytes, macrophages, and other cells [155]. MMP's are secreted as inactive proenzymes and are zinc dependant proteinases capable of degrading a wide variety of extracellular matrix components [156]. The activity of MMP's, is counteracted by tissue inhibitors of MMP's (TIMP's). The expression of MMP's and TIMP's must be carefully regulated during normal wound healing. In the case of keloids  and hypertrophic scars, excess collagen production has been attributed to a decrease in MMP-9 and no change in the TIMP expression [157]. An increase in TIMP expression and a decrease in MMP expression has also been implicated in excessive fibrosis such as in scleroderma [158]. In a recent study by Dang, it was shown that scars formed in fetal rats after nineteen days gestation whereas scarless wound healing was found at sixteen days gestation [155]. In this study, there was an increase in TIMP-1 and TIMP-3 in the scar-forming wounds where as the scarless wounds exhibited no change in the expression of these molecules. It is well known that TGF-P plays a very important role in scarring. Reduced or more transient expression of TGF-Ps and their receptors have been found in non-scarring fetal wounds as compared with adult wounds [159,160]. A strong and persistent expression of TGF-Ps and their receptors have been found in fibroblasts of postburn hypertrophic scars [161]. Both keloid tissues and keloid-derived fibroblasts have been associated with an overexpression of TGF-P 1 and -p2 [162]. When fetal wounds were treated with TGF-P 1, a large increase in scarring occurred in these cutaneous wounds [163]. Conversely, when incisional rat wounds were treated with neutralizing antibodies to TGF-pl and TGF-P2, a marked reduction in extracellular matrix deposition and scarring was evident [138]. In this study, the neutralizing antibodies reduced the magnitude of monocyte and macrophage infiltration, but in doing so, healing progressed more slowly. The application of antibodies was most effective in the early stages of wound healing, which indicates that the fibroproliferative phenotype of the fibroblast may be established in early healing. Transgenic mice lacking intracellular Smad3, which is part of the signalling cascade after TGF-p binds to its receptor, have been found to  28 have reduced monocyte recruitment and a shorter inflammatory phase [164]. Mice lacking Smad3 have shown significantly decreased fibrosis in both cutaneous and pulmonary fibrosis [165,166]. A study by Sumiyoshi et al. found that over-expression of Smad3 caused increased contraction of collagen gels by dermal fibroblasts, possibly by regulating TGB-p signalling [167]. There are currently five known isoforms of TGF-P, but only three of these, TGFp i , TGF-P2, and TGF-P3, are expressed in mammalian tissues [168]. These three isoforms behave similarly in most assays; however, their expression during embryogenesis, fibrosis, and wound healing are different [138]. All three isoforms are expressed within wounds, with TGFpi making up 85% of this expression [151]. While  TGF-P 1 and TGF-P2 are most often associated with scarring, TGF-P3 is associated with fetal wound healing and antagonizing the scarring effects of the other TGF-P isoforms [138, 169]. There are other cytokines withfibrogenicproperties that have been implicated in pathologic scarring. These included insulin-like growth factors (IGF), which act in synergy with TGF-P to increasefibroblastproliferation and matrix synthesis [170]. IGF has also been found in large quantities infibroticdermis [171]. Other cytokines that increase proliferation and differentiation offibroblastsand synthesis of matrix include: fibroblast growth factor, epidermal growth factor, IL-1, IL-4 and CTGF [151]. CTGF is expressed in many different tissues and organs and can stimulate the proliferation and chemotaxis offibroblastsdirectly [172]. CTGF is considered a downstream mediator of TGF-P 1, and has been implicated in the pathogenesis of fibrotic  29 processes [173]. The degree of C T G F expression has been correlated with the degree of fibrosis [174]. Activins, which are members of the  TGF-P superfamily, are involved in various  aspects of cell growth and differentiation and have also been implicated in enhanced wound repair and scarring. Over-expression of activins in transgenic mice revealed epidermal thickening and dermal fibrosis, possibly by inducing expression of growth factors in dermal fibroblasts with stimulation of keratinocyte proliferation [175].  1.3.U Antijibrotic Mechanisms: Fetal Wound Healing Extensive research has shown that the embryo and early fetus respond to wounding in a fundamentally different mechanism than the adult. In general, the fetal wound will show rapid closure, lack of inflammation, a reduction in cytokine expression, and a change in the ratio of different cytokines [176]. The fetal wound environment is much different than in an adult, as the wound is bathed in fetal serum and amniotic fluid rich in growth factors, hyaluronic acid, and fibronectin. Hyaluronic acid (HA) has been found in high concentrations in regions of tissue repair, proliferation, and regeneration [176]. Although the factor that causes persistent elevation of H A in the fetus is unknown, the presence of H A cause fibroblasts to migrate more rapidly, thereby accelerating fetal wound healing [177]. Fetal wounds also heal without the formation of a fibrin clot, which is thought to be the result of healing in a fluid environment [178]. It is not entirely known whether the presence of an eschar in the adult plays a role in scarring, although it most likely serves as a reservoir for growth factors. Amniotic fluid has been found to regulate matrix degrading enzymes. It has been shown to enhance collagenase activity, and inhibit  30  hyaluronidase, elastase, and cathepsin-B, suggesting a role i n scar-free healing [179]. M a n y other studies have shown, however, that scarless fetal healing can proceed i n the absence o f amniotic fluid, such as when human fetal skin was transplanted to a subcutaneous location on an adult athymic mouse [180]. In addition, when adult sheep skin was wounded i n the intrauterine environment, the skin continued to heal with a scar [181]. M a n y studies now suggest that scarless repair is intrinsic to fetal tissue rather than being dependant on the environment. Scarless healing i n the fetus is dependant on both the gestational age and the size o f the wound. In humans, an early fetal incisional wound (less than 24 weeks) w i l l heal without a scar whereas a late fetal wound w i l l heal with fibrosis [182]. A t each gestational age, it has been found that increasing the wound size also increases the frequency o f scarring [183]. Scarless healing has also been found to be dependant on the tissue type. Skin and bone heal without a scar during early gestational age; however, other regions such as nerve tissue and intestinal wounds do not [184,185]. The fetal wound is exposed to a reduced early inflammatory signal when compared to adult wounds. It has been demonstrated that fetal platelets aggregrate poorly in response to collagen and contain less  TGF-P and P D G F than adult platelets [186, 187].  Fetal wounds also lack the large neutrophil infiltrate that characterizes the early stage o f adult wounds [188]. A s healing progresses, fetal macrophages do enter the wound, but they are present i n lower numbers [189]. The fetal fibroblast plays a crucial role i n scarless healing and it exhibits many differences from its adult counterpart. Fetal fibroblasts synthesize more total collagen, exhibit greater prolyl hydroxylase activity (a rate limiting enzyme i n collagen synthesis),  31  synthesize greater proportions of types III and V collagen, and migrate faster than adult fibroblasts [2,190,191]. The high amount of H A and the increased amount of H A receptors on fetal fibroblasts may be related to the increased migration of these fibroblasts [192]. The myofibroblast has been implicated in wound contraction and scarring. In fetal lambs, the myofibroblast was found to be absent in the 75-day fetus, and present after 100 days gestation, when scar formation begins [193]. This suggests that the absence of these cells may have a role in scarless fetal healing. The fetal extracellular matrix also contributes to scarless wound healing. Fetal wounds heal with collagen in a fine reticular pattern which is indistinguishable from the surrounding uninjured dermis [194]. In addition, fetal wounds have a higher proportion of collagen type III in the early gestational age, which is smaller and finer than type I collagen. Cell adhesion molecules such as fibronectin and tenascin, and cell surface receptors (integrins) seem to appear earlier in fetal wounds than in adult wounds [195]. This may allow for a more rapid cellular infiltration and a faster rate of fetal fibroblast migration thus accelerating fetal wound healing. Higher proportions of MMP's to TIMP's have been found in early gestational wounds when compared to later gestational wounds [155]. This ratio of MMP's and their inhibitors in early fetal wounds favors extracellular matrix turnover and fetal cell migration, thus promoting scarless repair. Growth factors have also been studied in the fetal environment.  TGF-P, including  all three isoforms, has been found in unwounded fetal skin. At early gestational ages, however, lower levels of  TGF-P 1 and high levels of TGF-p3 are expressed [169]. Fetal  wound experiments have shown that a high TGF-P3 to associated with scarless healing [196].  TGF-P 1 expression ratio is  32 The unique properties of fetal tissue have also been studied at the genetic level. Some recent studies have demonstrated the activity of specific genes in human fetal skin known as homeobox genes. These genes are transcription factors that regulate pattern formation, such as the formation of limbs. After wounding, two of these homeobox genes, HOXB13 and PRX-2, show changes in expression that correlate with healing [197,198]. In addition, these two genes are not found in adult skin. This may suggest a role for such genes in fetal skin regeneration. 1.3. iii Comparison of Oral and Cutaneous Wound Healing  Many observations have been made to suggest that oral wounds heal more rapidly and with less scarring than cutaneous wounds. The characteristics which lead to this type of healing in the oral cavity may include a unique environment and intrinsic differences in the cells involved. Oral mucosal wounds are continuously bathed in saliva, which may be favorable for more rapid tissue repair. Studies have shown that desalivation of mice results in slowed healing of cutaneous wounds suggesting a favourable role of saliva in wound healing [199]. In addition, individuals with xerostomia have shown delayed oral wound healing [200]. Features of saliva such as proper pH, ionic strength, and the presence of ions seem to favour oral mucosal healing [201]. Saliva provides a moist environment which prevents tissue dessication, allows accelerated angiogenesis, and increased breakdown of fibrin and tissue debris [202]. In addition to a moist ionic environment, the majority of the healing potential of saliva is due to the presence of growth factors and bacteria [203]. Epidermal growth factor and vascular endothelial growth factor have both been found in saliva and play important roles in the repair of oral mucosa and  33 angiogenesis respectively [204, 205]. The presence of bacteria in the oral cavity may also increase the rate of wound healing by attracting macrophages into the wound site followed by the release of cytokines. The formation of granulation tissue and blood supply will therefore be accentuated in the wound, and a greater tensile strength of the contaminated wound will result due to increased connective tissue synthesis [206]. Studies have indicated that gingival fibroblasts, unlike adult skin fibroblasts, exhibit similar characteristics to fetal fibroblasts in three dimensional extracellular matrices [207,208].Oral fibroblasts were able to contract a three dimensional collagen matrix faster than dermal fibroblasts [209]. In addition, gingival fibroblasts secrete different proteolytic enzymes in comparison to dermal fibroblasts, such as MMP -13 and increased levels of tissue plasminogen activator [210, 211]. Oral mucosal fibroblasts were also found to exhibit increased levels of active MMP-2 and lattice contraction when compared to skin fibroblasts [6]. These differences in the proteolytic enzyme profile of oral and skin fibroblasts likely contribute to the preferential remodelling of oral wounds. Oral fibroblasts have also been found to produce more hepatocyte growth factor and keratinocyte growth factor than skin fibroblasts [212]. The heterogeneity of HGF and KGF expression in oral and skin fibroblasts may play a role in the different rates of repair in the two tissues. In a recent study by Szpaderska et al, it was found that oral wounds contained significantly less interleukin-6 and interleukin-8 (pro-inflammatory cytokines), and TGF-pl in mucosa versus skin [7]. In addition the inflammatory cell infiltrate consisting of neutrophils, macrophages, and T-cell was reduced in oral versus dermal wounds. This  34 study suggests that a reduced inflammatory response leads to an improved healing response in oral versus dermal wounds. 1.4 Microarray Techniques and Wound Healing Over the past decade, the use of large scale gene expression studies utilizing DNA microarrays has grown in popularity. These measurement techniques include the two most commonly used DNA microarrays: complementary DNA microarrays [213] and oligonucleotide microarrays [214]. A l l microarrays consist of distinct DNA sequences known as "probes" which are fixed to an immobile surface such as glass or nylon. This probe is then exposed to a labelled "target" of complementary RNA (cRNA) derived from mRNA of an experimental sample, which will then hybridize to the immobilized DNA. The amount of label associated with each DNA probe location will allow the inference of the abundance of each mRNA species from the experimental sample; therefore measuring gene expression. Microarray technology is useful because the expression of thousands of genes can be measured simultaneously in a single assay. This technique is in contrast to more traditional gene expression analyses, such as Northern blot, which can analyze up to 20 mRNAs at one time [215]. Microarray studies are essentially considered as "discovery research" and therefore are not considered as "hypothesis driven" research [216]. The information derived from microarrays, however, can help to narrow the search for genes that can later be used in hypothesis driven studies. To help better understand the function of genes, the analysis of the expression of mRNA under a certain set of conditions, known as functional genomics, has become important [215]. Using new large scale gene expression technologies, functional genomics will allow the comparison of different sets of genes used in different  35 conditions, such as in the comparison of health and disease, or healthy and wounded tissues. In addition to the use of microarrays in functional genomics, this technology has also proven useful for: biomarker determination, finding genes that correlate with disease progression; pharmacological research, by exposing tissues to varying doses or types of compounds and determining differences in gene expression; toxicogenomics, by finding gene expression patterns in a model organism exposed to a compound and their use as a predictor for adverse effects; target selectivity, to define a compound by the gene expression profile that it evokes in a target tissue and then compare it to other compound gene expression profiles: prognostic tests, to determine sets of genes that can distinguish one disease from another: and disease subclass determination, to find multiple subcategories of tumours in a single clinical diagnosis [217]. The use of microarray techniques in the area of wound healing is growing, although it is still currently limited in the area of cutaneous and especially oral wound healing. 1.4.i Microarray Studies in the Area of Cutaneous Wound Healing In 2000, Tsou et al compared gene expression of hypertrophic scar tissue to normal scar tissue and uninjured skin in humans [218]. It was found that 44 genes were over-expressed and 124 genes were under-expressed in hypertrophic scars compared with normal scars. Differential expression was also found between normal scars and uninjured skin and hypertrophic scars and uninjured skin. A more recent study compared cultured fibroblasts from hypertrophic scars and normal skin, and the response of these cells to interleukin-6 stimulation [219]. Comparison of the gene expression profiles, using  36 Affymetrix microarrays, revealed an increase in expression of 12 genes and a decrease in expression of 14 genes in the hypertrophic scar fibroblasts versus normal skin fibroblasts. It was also found that there was an absence of upregulation of matrix-metallo proteinases in hypertrophic scarfibroblastsin response to IL-6, which may play a role in excess collagen formation in these scars. Thisfindingwas in contrast to the effect of IL-6 on normalfibroblasts.Other recent studies have also found different gene expression patterns in hypertrophic scars and in keloids when compared to normal skin, including the upregulation of TGF-P 1 in keloids [220, 221]. Studies in 2001 utilizing cDNA microarrays, were conducted to determine the baseline gene expression in uninjured skin, and to determine the gene expression at thirty minutes and one hour following initial injury [222, 223]. It was found that when comparing uninjured normal human skin in nine different individuals, seventy one genes (1.7%) were substantially variable in their expression [223]. Included in this list of genes was expression variability for many regulatory proteins such as DNA binding proteins. Mitochondrial genes were among the seventy-one variable genes, which may suggest differences in energy and oxidative metabolism. Although gene expression across the nine individuals was 98% similar, this study indicates that differential gene expression in normal skin exists between individuals, and such variation therefore must be considered when comparing normal skin to injured skin. The follow-up study comparing uninjured skin to early injury (at thirty minutes and 1 hour), showed a consistent increase in gene expression in 124 of 4000 genes (3%) at thirty minutes [222]. No significant down regulation was noted for any gene at thirty minutes. Most of the up-regulated genes were involved in transcription and signalling. By one hour, only 46 of 4000 genes (1.15%)  37 were significantly increased; however, 264 of 4000 (6.6%) of genes were significantly down-regulated, which included many structural genes and many other genes with unknown functions. Affymetrix oligonucleotide microarrays have been used to analyze burn wounds in rats and mice. Spies utilized this microarray technique to determine gene expression at various time points following a burn wound (2, 6, 24, and 240 hours) as compared to uninjured skin in rats [224], In this study, differential expression of genes were associated with events such as cell survival and death, cell growth regulation, cell metabolism, inflammation and immune response. The time of appearance of various genes matched the known time sequence of a healing burn wound. A similar study was conducted on mice with scald wounds, and revealed significant alterations in the expression of genes, with the greatest changes between 3 and 14 days after wounding [225]. Upregulation of gene expression in this study was both transient and time-dependant. Gene ontologies showed an integrated upregulation of inflammatory and protease genes at early time points in the scald wound. Nowinski et al. utilized the affymetrix microarray technique to determine how keratinocytes may affect the activity of fibroblasts during the process of reepithelialization [226]. Cocultures of these two cell types, physically separated by a membrane, were carried out for 48 hours after which fibroblasts were analyzed for gene expression. 243 genes were up-regulated two -fold or more, while 100 genes were downregulated two-fold or more. This in-vitro model demonstrated that keratinocytes regulated the expression of a total of 343 fibroblasts genes, with the majority of these genes being stimulated.  38 Another recent study investigated the differences in gene expression between human neutrophils in peripheral blood, and those that have migrated to the site of a skin wound [227]. This study demonstrated that the migration of neutrophils into a skin lesion is associated with an extensive change in gene expression, including up-regulation of anti-apoptotic genes, down-regulation of pro-apoptotic genes, and up-regulated genes for cytokines and chemokines critical for chemotaxis. The cellular fate and function of neutrophils in skin wounds is therefore partially regulated at the transcriptional level. Many of the cutaneous wound healing studies incorporating microarray techniques are still preliminary studies only. These studies have dealt with differences between baseline and injured tissue, injuries of various forms, and cell types involved in wound healing. Many of the studies utilize animal models, in vitro models, and have small sample sizes. In addition, not all of the studies use the same microarray technique. There are critical differences between oligonucleotide and cDNA microarrays. cDNA microarrays provide the amount of each transcript relative to another sample, while oligonucleotide arrays provide an absolute amount for each transcript [228]. This basic difference creates challenges in the ability to compare observations between the two techniques [229]. In the past, there have been few attempts to replicate and/or compare expression data between different microarray techniques, and there is a general lack of standardized controls for such experiments. Some studies that do compare the two techniques [230, 231] have found large discrepancies in the quantification of gene expression. Caution should therefore be exercised when attempting to compare results between different microarray platforms. Currently, recommendations do exist to increase the reliability of microarray data [216]. These include: adequate sample size, pooling  39  samples before hybridization, and using other statistical methods besides arbitrary fold increase to analyze the data. Confirmational studies are recommended (such as Northern blot, quantitative RT-PCR, in situ hybridization or immunohistochemistry) to verify expression data. It would not be feasible to verify all genes from a microarray data sheet; however, key genes of interest should be verified in some way. While more microarray studies are slowly emerging in the area of wound healing, very limited data has been gathered specifically on oral wound healing. In addition, comparisons between oral and cutaneous tissues and wounds using microarray techniques have not been conducted.  40 CHAPTER II Aim of the study The aim of this study was to characterize the gene expression profiles of early human masticatory mucosal wounds after one, three, and seven days of healing. In addition, histologic and clinical analysis was carried out for these wounds for sixty days and six months, respectively. Based on the transcriptional profiling of the mucosal wounds, a novel gene product in the context of oral wound healing was further assessed by localization of the protein within the wounds.  41 CHAPTER III Materials and methods Experimental wounding  The experimental protocol for this study was approved by the Clinical Research Ethics Board of the University of British Columbia. The study involved twenty-three volunteer subjects. All subjects signed a written consent form to participate in the study. The subjects ranged between 24 and 66 years of age (mean 40.8 years) and were all healthy, non-smoking individuals. Full thickness wounds were made in the palatal masticatory mucosa at least 3 mm away from the gingival margins of the teeth and distal to the canine. The wounds were 10 mm long with the standard width of 2 mm, made with parallel blades. The tissue was dissected off at the depth of 2 mm using mini-blades (6900N Hartzell) (Fig. 1). The tissue harvested from the initial wound (control sample) was rinsed in physiologic saline and placed in either a cryotube (Nalge Nunc Int, Rochester, NY) or embedded in Tissue-Tek O.C.T. (Sakura Finetek Inc, Torrance, CA) and immediately frozen in liquid nitrogen (0 day or control sample). All control tissue specimens were stored at -86°C until further use. The wounds were then left uncovered to heal for a period of 1, 3, 7, 14, 28, or 60 days. Two punch biopsies (4 mm diameter) were taken from each wound and rinsed with physiologic saline. One biopsy was placed into a cryotube and immediately frozen in liquid nitrogen. The second biopsy was embedded in Tissue-Tek O.C.T. and immediatelyfrozenin liquid nitrogen. All samples were then stored at -86°C until further use. Tissue samples (control, 1, 3, and 7 day wounds) that were stored in cryotubes were used for transcriptional profiling (see below).  42  Tissue samples (control and all wound time points) embedded in Tissue-Tek O.C.T. were prepared for routine histology and immunohistochemistry (see below). For evaluation of clinical scarring, a survey was conducted at the end of the experiment. Eight randomly organized palatal clinical photos (four pre-wounding samples and four wound sites after 6 months of healing) were shown to three dentists unaware of the purpose of this study. Each dentist was asked to describe whether palatal masticatory mucosa in the photographs was normal or abnormal in clinical appearance.  Initial wound made on healthy palatal gingiva away from the gingival margins of teeth with a standardized double blade holder. Initial tissue sample then removed (A,B,C).  After initial wound has healed for appropriate time, a 4 mm punch biopsy is placed over wound and the wound sample is removed carefully (D,E,F).  Armamentarium (G) includes mirror, periodontal probe, tissue forceps, double blade holder with # 15 blades, and 6900N mini-blade. Double blade Holder (H) and 6900N mini-blade (I). Initial tissue sample with 10mm X 2mm X 2mm dimensions (J) and biopsy of wound sample with 4 mm diameter and 2 mm depth (K).  Figure 1. Wound Preparation, Biopsy Technique and Armamentarium Used  44  Transcriptional  profiling  Preparation of labelled cRNA was performed as recommended by the manufacturer of the microarrays (Affymetrix Inc). Total RNA was extracted from each gingival sample using TRIzol reagent. Poly(A) mRNA was isolated by reverse transcription with a T7Oligo (dT) promoter primer to generate double stranded cDNA. The cDNA was then used as a template for in-vitro transcription (IVT) to yield biotin-labelled cRNA's. The biotinylyated cRNA's were then fragmented for hybridization to a gene chip HGU95Av2 array for 16 hours at 45°C. After hybridization, the probe arrays were then washed in non-stringent buffer and stained with streptavidin phycoerythrin (SAPE) followed by an antibody amplification step. Arrays were then scanned for fluorescence intensities using Affymetrix protocols. Scanned output files were analyzed using Microarray Analysis Suite version 5.0V Agglomerative hierarchical clustering was performed to show correlation of gene expression profiles between individuals (Fig. 2). Results are presented as the fold changes between non-wounded control samples and test samples (for each 1-, 3-, and 7- day palatal wound). Analysis of variance was then conducted to determine the significantly up-regulated and down-regulated genes for all patients at each wounding time point (p-value<0.01). Genes whose expression did not differ from the background level were omitted. In addition, functional clustering of genes at each wound time-point was determined using the Gene Ontology (GO) Consortium. The "observed" gene ontology classes at each time point were compared to the "expected gene ontology classes that would occur by chance alone using bootstrapping statistical methods (p-value<0.01).  Figure 2 . Agglomerative Hierarchical Clustering of Gingival Wounds o 10  o  o  -ift-  CTi  1-day wound i. ~l-day wound ii "1-day wound iii. "3-day wound i. -3-day wound ii. -7-day wound i. -3-day wound iii. _7-day wound ii. _7-day wound iii. -Control i. -Control ii. -Control iii. -Control iv. -Control v. -Control vi. -Control vii.  46  Routine histology and immunohistochemistry  The control and wound specimens were sectioned (8 um) in a cryostat and stored at 86°C until used. After fixing with -20°C acetone for 5 minutes, a set of the sections were stained by hematoxylin and eosin for morphological analysis of the wounds. These sections were then mounted with Krazy glue (Elmer's Products Can Inc, Brampton, ON). For immunolocalization studies, sections were fixed with -20°C acetone for five minutes, after which sections were rinsed in phosphate buffered saline (PBS) containing 1 mg/ml bovine serum albumin (BSA) for five minutes. Sections were then incubated in normal blocking serum (Vectastain;Vector Laboratories Inc; Burlingame, CA) for 30 minutes at room temperature and then incubated with primary monoclonal antibody against carcinoembryonic antigen-related cell adhesion molecule 6 (Ceacam-6) (Genovac; Freiberg, Germany). After rinsing the sections with PBS/BSA (lmg/ml), sections were then incubated with biotinylated anti-mouse secondary antibody for 1 hour at room temperature and then reacted with ABC avidin/peroxidase reagent (Vectastain Elite kit; Vector Laboratories Inc; Burlingame, CA). Sections were rinsed again and reacted with Vector VIP substrate. The reaction was stopped when the desired colour was detected in the cells under a light microscope by immersing the sections in distilled water for five minutes. A l l sections were then dried and mounted with Vectamount (Vector Laboratories, Burlingame, CA). Control staining was performed without the primary antibodies and no specific reaction was obtained (data not shown).  47 CHAPTER IV Results Clinical healing  All palatal mucosal wounds healed uneventfully and with no complications. Clinical observations of 1-day to 28-day wounds show a continuous progression of healing (changes in the color, outline and size of the wound) (Fig. 3 and 4). The outline of the wound margin could still be delineated for certain individuals after 28 days of healing (Fig. 4). Clinical comparison of same-day wounds between different individuals revealed x  some variation in healing speed (data not shown). Observations of the palatal mucosa were made following at least 6 months after the biopsy procedure (second wounding) and compared to the baseline palatal mucosa. The majority of the wounds (about 75%, see below) healed by regeneration and wound sites were indistinguishable from the surrounding tissue (Fig. 5). In some cases, however, the wound site demonstrated an outline consistent with a biopsy wound. The clinical survey comparing pre- and postwounded palates revealed that clinical changes associated with the experimental wounds were not obvious to dental-trained individuals. The only observation regarding four prewound and post-wound palates was a possible fibrosis or scar tissue noted in one palate (Fig. 5) by two out of the three examiners.  Figure 3. Initial palatal wound (A,C,E) and corresponding 1-day-old wound (B), 3-day-old wound (D), and 5-day-old wound (F).  49  Figure 4. Initial palatal wound (A,C,E), and corresponding 7-day-old wound (B), 14-day-old wound (D), and 28-day-old wound (F).  50  A.  •  —. •' ri  :  v  B.  9.  __E____L_ ____p^ _,  ,  al  p  C. m  E.  F.  m  J_l " ' i f  Figure 5 . Baseline palatal gingiva prior to initial wound preparation (A,C,E), and corresponding area following more than 6 months healing after wound biopsy (B,D,F). Clinical differences are evident between A and B (area surrounded by arrows), but not evident between C and D, or E and F.  •  51  Transcriptional  profiling  Seven control tissue samples (0 day) and three samples for each wound time point (1-, 3-, and 7-day) were analyzed by Affymetrix protocols as described above. A total of 988 genes showed a significant fold change in expression in the one-day wound samples compared to baseline non-wounded gingiva. Of this total, 471 (47.7%) were up-regulated and 517 (52.3%) were down-regulated (Fig. 6). At day three, 421 genes showed a change in expression with 270 of these (64.1%) showing up-regulation and 151 (35.9%) showing down-regulation (Fig. 6). By day seven, 543 genes showed a change in expression with 298 (54.9%) being up-regulated and 245 (45.1%) being down-regulated (Fig. 6). No wound time points beyond seven days were analyzed. Table 1 and 2 list the twenty most up-regulated and down-regulated genes at the 1-day wound. The largest increase in expression was seen for serine proteinase inhibitor B l , matrix metalloproteinase-1 (MMP-1) and carcinoembryonic antigen-related cell adhesion molecule-6 (Ceacam-6) at the 1-day wound. In addition, an increase in expression of certain growth factors and cytokines was seen (IL-6, IL-8, endothelial cell growth factor1). In contrast, a down-regulation in the expression of collagen type I, III, VI and cytokeratin 2 and 15 were seen at the 1-day wound. Trends in ontology analysis revealed an up-regulation of genes in 24 categories in the 1-day wound including cytokines, negative control of cell proliferation, cell motility, inflammatory and immune response categories, as well as many others (Fig. 7). A general downward trend in 13 categories was seen in the 1-day wound including such categories as extracellular matrix, epidermal differentiation, and collagen categories (Fig. 8).  52 The 3-day wounds showed some similar changes in gene expression to the 1-day wound in that MMP-1, Ceacam-6, and serine proteinase inhibitor B l showed the greatest increase in expression (Table 3). Gene ontology analysis showed an upward trend of 10 categories in the 3-day wound including proteinase inhibitors, calcium binding and cytoplasm categories (Fig. 9). Table 4 reveals the most significantly down-regulated genes at the 3-day wound. Ontology analysis revealed a downward trend in six groups in the 3-day-old wounds including epidermal differentiation (as in the day-1 wound), and embryogenesis and morphogenesis categories (Fig. 10). Tables 5 and 6 reveal the most significantly up-regulated and down-regulated genes expressed following 1 week of healing. Some notable changes in the gene ontology trends were seen at the 7-day wound including both a downward and upward trend in epidermal differentiation, an upward trend in extracellular matrix genes, collagen and collagen binding genes and cytoskeletal structural proteins (Fig. 11 and 12). An upward trend for proteinase inhibitors (with serine proteinase inhibitor B l still being highly expressed) persisted in the 7-day wound. In total, ontology analysis revealed an upward trend in gene expression for 23 categories and a downward trend in 5 categories.  Figure 6. Significant Alterations in Gene Expression During Wound Healing  1200 2  1000  O  800  c  down-regulated up-regulated  600 400 =  200 0 DAY 1  DAY3  DAY7  Time Figure 6. The total number of genes showing a significant change in expression is denoted by the entire bar for each wound time point. Each bar is then divided into up-regulated or down regulated genes.  Number of Genes O W O U l O U i O U l O U l cytoplasm inflammatory response cytoskeletal structural  A orq  protein  *-r)  3 s §  chemotaxis  C- jj- •  chemokine  °* « ffl  o « g o •a cr  cytokine microtubule  O  HJ  interleukin-8 receptor  O  n a m S5 w t= O  ft »(2 SL E.  O  (D O  3  <-  n  S.  8 ft  ft OQ 3 O  ligand cell cycle control  o  ft s re o  O ft  ft ta  O  n o ft X  <-jg  signal cytoskeleton  3  O QTQ  3  cell surface receptor linked  QTQ O  cell motility protein glycosylation response to pest/pathogen/parasite soluble fraction  5'  GTPase  s cr ft 2 e ^  actin modulating  ft)  & ft 3  o  o  extracellular space mitochondrion protein biosynthesis  O A  ft  B5 (JQ Pt H fB &  00  P  ^ cx  negative control of cell proliferation methenyltetrahydrofolate cyclohydrolase methy ten etetrahy dro fo late dehydrogenase immune response  Figure 8. Over-represented Ontologies (p<0.01) Day 1 - Down-regulated genes  Observed Expected  __ "3  ed O  =:"C x u  o  ed  _  ed  ed  E  u  3J  '5. cu  —  C3  =  V-  —  -a  ©  D. p u SO  c .2 a £ O 2 .2-  11 c 00  3 C B Jd 00 cd u  s  1  ed S  I i  E o-  C u Sp __ 09  o  Q  cu E  2  | &*  c c  CU *3 sC C CA  £ c cu CJ  cd  cu CL  a cu  c  5 cu o cu — CL u en  00  "3 CJ  cd o  Gene Ontology Category  Figure 8. Each gene ontology category consists of a certain number of down-regulated genes. The expected results are the number of genes expressed in each GO category that would occur by chance alone.  ed  00  c d cd  B  Figure 9. Over-represented Ontologies (p<0.01) Day 3 - Up-regulated genes l Observed m Expected £  30  SO  _c CL  o  43  u  le  E = |_ r  ccs CJ  CD 95  CD  a u  C3  c  CD  SO  o  cd *— TJ  x: CD  TJ  C  2 2 CD  °  ^  M  IE  O  J=  9§  >> J3 U D. CCj aj  TJ JS  o  c CJ  cd ,s-  _5J  cd  c  o  3 _=  CD  c c  E  CD  a  CD  CL  o  o  ft  CN  CJ  Gene Ontology Category Figure 9. Each gene ontology category consists of a certain number of up-regulated genes. The expected results are the number of genes expressed in each GO category that would occur by chance alone.  Figure 10. Over-represented Ontologies (p<0.01) Day 3 - Down-regulated genes • Observed i Expected  transcription regulation  creatine kinase  heterochromatin  embryogenesis and morphogenesis  mitochondrial matrix  epidermal differentiation  Gene Ontology Category  Figure 10. Each gene ontology category consists of a certain number of down-regulated genes. The expected results are the number of genes expressed in each GO category that would occur by chance alone.  OC in  Figure 11. Over-represented Ontologies (p<0.01) Day 7 - Up-regulated genes I Observed Expected  ti  g 20 5 10 E  a u  g_ a.  E  -o  s  Gene Ontology Category Figure 11. Each gene ontology category consists of a certain number of up-regulated genes. The expected results are the number of genes expressed in each GO category that would occur by chance alone.  '5 jo O JS  a.  .8  Figure 12. Over-represented Ontologies (p<0.01) Day 7 - Down-regulated genes • Observed  m Expected 10 ,  epidermal differentiation  organic anion transporter  multidrug transporter  cell shape and cell size control  intermediate filament  Gene Ontology Category Figure 12. Each gene ontology category consists of a certain number of down-regulated genes. The expected results are the number of genes expressed in each GO category that would occur by chance alone.  Table 1. The Twenty Most Upregulated Genes at Day 1 Wound Compared to Baseline Gene Fold Change Gene Description SERPINB1 100 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 1 MMP1 40.11 matrix metalloproteinase 1 (interstitial collagenase) NCA 33.67 carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross reacting antigen) IL6 21.66 interleukin 6 (interferon, beta 2) DEFB2 18.43 EMBL: Homo sapiens beta defensin 2 (HBD2) gene, complete eds. MMP9 16.71 matrix metalloproteinase 9 (gelatinase B, 92kDa gelatinase, 92kDa type IV collagenase) GROl 15.4 EMBL: Human gene for melanoma growth stimulatory activity (MGSA) GPR37 13.1 Human putative endothelin receptor type B-like protein mRNA, complete eds G0S2 12.98 EMBL: Human G0S2 protein gene, complete eds. UP 12.74 uridine phosphorylase IL8 12.34 interleukin 8 GR02 11.18 chemokine (C-X-C motif) ligand 2 ANXA3 9.61 annexinA3 SERPINB4 9.39 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 4 ECGF1 9.24 endothelial cell growth factor 1 (platelet-derived) SLC16A1 8.31 solute carrier family 16 (monocarboxylic acid transporters), member 1 T1A-2 7.93 lung type-I cell membrane-associated glycoprotein LTBR 7.78 lymphotoxin beta receptor (TNFR superfamily, member 3). NNMT 7.35 nicotinamide N-methyltransferase ACTN1 7.32 actinin, alpha 1  Table 2. The Twenty Most Downregulated Genes at Day 1 Wound compared to Baseline Gene Fold Change Gene Description COL1A2 -21.13 collagen, type I, alpha 2 | HSA8SEQ -13.88 EMBL: Human 3' nucleotide sequence mRNA. AZGP1 -12.47 alpha-2-glycoprotein 1, zinc MT4 -9.83 EMBL: Human metallothionein IV (MTIV) gene, complete eds. FHL1 -9.26 four and a half LIM domains 1 RNASE4 -6.82 ribonuclease, RNase A family, 4 HPGD -6.81 hydroxyprostaglandin dehydrogenase 15-(NAD) MYH11 -6.72 myosin, heavy polypeptide 11, smooth muscle AQP1 -6.61 aquaporin 1 (channel-forming integral protein, 28kDa) DSIPI ' -6.43 delta sleep inducing peptide, immunoreactor COL3A1 -6.01 collagen, type III, alpha 1 (Ehlers-Danlos syndrome type IV, autosomal dominant) HUMCYT2 -5.81 cytokeratin 2 IGHG3 -5.59 immunoglobulin heavy constant gamma 3 (G3m marker) MN1 -5.51 meningioma (disrupted in balanced translocation) 1 FBLN2 -5.5 fibulin 2 AADAC -5.37 arylacetamide deacetylase (esterase) KRT15 -5.06 keratin 15 THBS2 -5.05 thrombospondin 2 PPAP2B -4.94 phosphatidic acid phosphatase type 2B COL6A1 -4.84 collagen, type VI, alpha 1  Table 3.The Twenty Most Upregulated Genes at Day 3 Wound Compared to Baseline Gene Fold Change Gene Description SERPINB1 55.89 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 1 NCA 48.12 carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross reacting antigen) MMP1 21.72 matrix metalloproteinase 1 (interstitial collagenase) UP 18.21 uridine phosphorylase KLK6 17.2 kallikrein 6 (neurosin, zyme) SERPINB4 16.17 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 4 S100A7 15.14 S100 calcium binding protein A7 (psoriasin 1) SLC16A1 11.56 solute carrier family 16 (monocarboxylic acid transporters), member 1 HBB 9.32 hemoglobin, beta PI3 7.75 EMBL: Homo sapiens elafin precursor, gene, complete eds. ODC1 7.62 EMBL: Human gene for ornithine decarboxylase ODC (EC DEFB4 7.13 EMBL: Homo sapiens beta defensin 2 (HBD2) gene, complete eds. TIMP1 6.76 EMBL: Homo sapiens TIMP gene for tissue inhibitor of metalloproteinases, partial eds. PRSS3 6.55 protease, serine, 3 (mesotrypsin) PDXK 6.28 pyridoxal (pyridoxine, vitamin B6) kinase F2RL1 5.6 coagulation factor II (thrombin) receptor-like 1 S100A13 5.52 S100 calcium binding protein A13 HBAP1 5.03 EMBL: Human alpha globin psi-alpha-1, alpha-2 and alpha-1 genes, complete eds. MYH11 4.98 myosin, heavy polypeptide 11, smooth muscle TCRBC1 4.9 human germline T-cell receptor beta chain  Table 4. The Twenty Most Downregulated Genes at Day 3 Wound Compared to Baseline Gene Fold Change Gene Description APOD -8.4 apolipoprotein D MT4 ' -6.15 E M B L : Human metallothionein IV (MTIV) gene, complete eds. DSIPI -3.72 delta sleep inducing peptide, immunoreactor NR1D2 -3.52 nuclear receptor subfamily 1, group D, member 2 CYP3A5 -3.49 cytochrome P450, subfamily IIIA (niphedipine oxidase), polypeptide 5 MAPT -3.47 microtubule-associated protein tau CLU -3.24 clusterin MFAP4 -3.22 microfibrillar-associated protein 4 KIAA0802 -3.2 KIAA0802 protein WEE 1 -3.19 W E E 1 + homolog (S. pombe) DBP -3.11 E M B L : Human D-site binding protein gene, exon 4 and complete eds. ID2 -2.91 inhibitor of D N A binding 2, dominant negative helix-loop-helix protein RPL23AP1 -2.89 E M B L : Human D N A sequence from clone RP3-377H14 on chromosome 6p21.32-22.1. PTN -2.65 pleiotrophin (heparin binding growth factor 8, neurite growth-promoting factor 1) FAT2 -2.59 F A T tumor suppressor homolog 2 (Drosophila) KIAA0318 -2.54 KIAA0318 protein D4S234E -2.49 E M B L : Homo sapiens 21 kDa protein gene, complete eds, clone D4S234. CKB -2.43 E M B L : Human gene for creatine kinase B (EC ABCC5 -2.39 ATP-binding cassette, sub-family C (CFTR/MRP), member 5 IGFBP5 -2.38 insulin-like growth factor binding protein 5  Table 5. The Twenty Most Upregulated Genes at Day 7 Wound Compared to Baseline Gene Fold Change Gene Description SERPINB1 22.54 serine (or cysteine) proteinase inhibitor, clade B (ovalbumin), member 1 MMP1 17.07 matrix metalloproteinase 1 (interstitial collagenase) KLK6 17.03 kallikrein 6 (neurosin, zyme) NNMT 15.62 nicotinamide N-methyltransferase SPP1 14.97 secreted phosphoprotein 1 (osteopontin, bone sialoprotein I, early T-lymphocyte activation 1) MMP9 14.48 matrix metalloproteinase 9 (gelatinase B, 92kDa gelatinase, 92kDa type IV collagenase) DKFZP5861 13.77 DKFZP586L151 protein FN1 12.67 fibronectin 1  NCA  12.46 carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross reacting antigen)  PTHLH T1A-2 FAP SLC16A1 UP FCGR3B IL6 SLC16A3 COL4A1 LILRB1 NT5  12.17 11.71 11.57 10.9 8.87 8.7 8.65 7.65 7.33 7.12 6.29  parathyroid hormone-like hormone lung type-I cell membrane-associated glycoprotein fibroblast activation protein, alpha solute carrier family 16 (monocarboxylic acid transporters), member 1 uridine phosphorylase Fc fragment of IgG, low affinity Ilia, receptor for (CD16) interleukin 6 (interferon, beta 2) solute carrier family 16 (monocarboxylic acid transporters), member 3 EMBL: Human alpha-1 collagen type IV gene, exon 52. Homo sapiens monocyte/macrophage Ig-related receptor MIR-7 (MIR cl-7) mRNA, complete eds 5'-nucleotidase, ecto (CD73)  Table 6. The Twenty Most Downregulated Genes at Day 7 Wound Compared to Baseline Gene Fold Change Gene Description MT4 -22.99 EMBL: Human metallothionein IV (MTIV) gene, complete eds. HSA8SEQ -11.01 EMBL: Human 3'nucleotide sequence mRNA. ARG1 -10.58 arginase, liver DSC1 -8.36 desmocollin 1 KRT2A -7.98 EMBL: Homo sapiens keratin 2e (KRT2E) gene, complete eds. CYP3A5 -5.66 cytochrome P450, subfamily IIIA (niphedipine oxidase), polypeptide 5 HTR3A -5.27 5-hydroxytryptamine (serotonin) receptor 3A COCH -5.1 coagulation factor C homolog, cochlin (Limulus polyphemus) FLG -4.84 filaggrin SLAC2-B -3.91 SLAC2-B ABCC5 -3.8 ATP-binding cassette, sub-family C (CFTR/MRP), member 5 EPHB6 -3.73 EphB6 EPHA4 -3.69 EphA4 PNUTL1 -3.68 peanut-like 1 (Drosophila) COMP -3.54 cartilage oligomeric matrix protein (pseudoachondroplasia, epiphyseal dysplasia 1, multiple) BMP7 -3.17 bone morphogenetic protein 7 (osteogenic protein 1) LY6G6C -3.01 EMBL: Homo sapiens genes encoding RNCC protein, DDAH protein, Ly6-C protein and immunoglobulin receptor DSIPI -2.97 delta sleep inducing peptide, immunoreactor RNASE4 -2.89 ribonuclease, RNase A family, 4 KRT10 -2.83 EMBL: Human gene for acidic (type I) cytokeratin 10  66 Histology and immunohistochemistry All wounds were processed for histological analysis with hematoxylin and eosin. The 1day wound margins were clearly defined with a fibrin clot and a multitude of j inflammatory cells established within the wound (Fig. 13). The 3-day-old wound still showed evidence of a fibrin clot with some apparent migration of epithelium along the wound margins (Fig. 13). In some of the 5-day wounds, fibrin clot was still apparent at the top of the wound. Migration of epithelium was clearly evident at the wound edge and epithelium showed increased thickness at the wound margin consistent with strong proliferation. In some, but not in all 5-day-old wounds epithelial sheets were fused in the middle of the wound. Granulation tissue was present in the centre of the wound below the fibrin clot and migrating epithelium (Fig. 13). By day 7, the epithelium had covered all the wounds. Granulation tissue filled the wound below the newly formed epithelium. The wound outline could still be demarcated (Fig. 14). The 14-day-old wounds showed still an increase in the layers of epithelium and rete ridges were being restored. The granulation tissue is still apparent beneath the new epithelium, although the wound margins are not well defined (Fig. 14). By 28 days, the outline of the wound was not readily apparent and the cellularity of the new connective tissue was reduced (Fig. 14). By day 60, the connective tissue is totally regenerated and the wound site cannot be distinguished from the surrounding tissue (Fig. 14). Epithelial rete ridges have been reestablished in these wounds (Fig. 14).  To verify the  transcriptional  profiling at the  protein  level, we  conducted  immunolocalization of Ceacam-6 (one of the mostly up-regulated novel genes in all time  67 points) in healthy gingiva and all gingival wounds. No Ceacam-6 was present in either epithelium or connective tissue in the non-wounded healthy gingiva (Fig. 15). Ceacam-6 was localized in keratinocyte cell membranes of the basal and suprabasal layers of epithelium especially at the wound margin but also at a distance from the margin in the 1day wound (Fig. 15). Ceacam-6 expression was reduced in the basal layer of epithelium in the 3-day and 5-day wounds with more notable localization of this protein in the suprabasal epithelial layers and in the proliferating and migrating epithelium (Fig. 15). In addition, the staining is seen in closer proximity to the wound margin as compared to the 1-day wound (Fig. 15). No Ceacam-6 is evident in the connective tissue of any of the 1, 3, and 5- day wounds (Fig. 15). By day 7, Ceacam-6 is still noted in the suprabasal layers of the newly formed epithelium (Fig. 16). Positive staining for Ceacam-6 is now evident in the fibroblasts of the newly forming connective tissue in the 7-day wound (Fig. 16). Ceacam-6 staining expression persists in the fibroblasts in the 14-day wound but it is no longer expressed in the keratinocytes (Fig. 16). By day 28, Ceacam-6 expression was no longer seen in either the epithelium or connective tissue (Fig. 16).  68  Figure 13. Hematoxylin & Eosin staining of healthy gingiva (A,B,C), 1-day-old wound (D,E,F), 3-day-old wound (G,H,I), and 5-day-old wound (J,K,L). Arrows in panel J mark epithelium that has started to proliferate prior to migration over the wound bed. EP, epithelium; CT, connective tissue; FC, fibrin clot; GT, granulation tissue. Bar 260 urn: A , D, G, J; Bar 100 jj.m; B , E, H , K.; Bar 50 u.m:C, F, I, L.  69  Figure 14. Hematoxylin and Eosin Staining of 7-day-old wound (A,B,C), 14-day-old wound (D,E,F), 28-day-old wound (G,H,1), and 60-day-old wound (J,K,L). Arrows in panel B show that the epithelium has completely covered the underlying granulation tissue at 7 days. The wound is no longer evident by 60 days. EP, epithelium; CT, connective tissue; GT, granulation tissue. Bar 260 mm: A, D, G, J; Bar 100 mm: B, E, H, K; Bar 50 mm: C, F, I, L.  70  Figure 15. Immunostaining of Ceacam-6 in oral gingival wounds. Healthy gingiva (A,B,C) does not express Ceacam-6 in either epithelium or connective tissue. In panel B and C, the white line indicates the separation of epithelium and connective tissue. The 1-day-old wound (D,E,F) shows positive staining for Ceacam-6 in the basal and spinous layers of the epithelium (denoted by arrows in E and F). This staining begins distant from the wound margin and continues to the epithelium adjacent to the wound margin where epithelial cells are beginning to proliferate. The 3-day -old wound (G,H,I) and the 5-dayold wound (J,K,L) reveal less staining in the basal layer of epithelium, but persistent staining in the suprabasal layers and the proliferating epithelium adjacent to the wound margin (denoted by arrows in G , H , J, and L). Bar 260 p.m: A , D, G , J; bar 100 u,m: B, E, H, K ; bar 50 urn: C, F, I, L.  71  '  Figure 16.  Immunostaining  I""  of Ceacam-6 in oral gingival  ^11 III  wounds. The 7-day-old wound  (A,B,C) shows some positive staining for Ceacam-6 in the new suprabasal layers of epithelium and an increase in staining of cells in the connective tissue (denoted by arrows in B and C). The 14-day-old wound continues to show expression of Ceacam-6 in cells of the connective tissue (denoted by arrows in F), but staining is no longer positive in the epithelium. The 28-day-old wound (G,H,I) and the 60-day-old wound (J,K,L) no longer exhibit positive staining for Ceacam-6 in either the epithelium or the connective tissue. The outline in panels H, K, and L show the separation between epithelium and connective tissue. Bar 260 pm: A , D, G , J; bar 100 pm: B, E, H, K; bar 50 pm: C , F, I, L.  72 CHAPTER V Discussion The mechanism of adult scarless wound healing is not well understood. As demonstrated in this study, healing of palatal masticatory mucosa in the early stages (1-7 days) involved up-regulation and down-regulation of several genes in a highly coordinated fashion. This study revealed that the gene expression profiles between individuals showed good correlation even though some expected clinical variation in the healing speed was noted between individuals. Wound healing rate and histological healing of this study was comparable to those we have documented previously using the same type of experimental setting [79, 85]. Following more than six months of healing, most wounds healed by regeneration i.e. did not show clinical evidence of scarring. This agrees with the clinicians' observations and some data that even relatively large wound sites in masticatory mucosa heal rapidly and with limited or lack of scar formation [7, 76].  Gene-ontology analysis in this study is consistent with general knowledge of the phases of wound healing. Expected early upregulation of genes involved in processes such as inflammation, cell motility and chemotaxis, and a downregulation in extracellular matrix and collagen formation and epidermal differentiation was observed. Genes involved in cell proliferation, extracellular matrix formation, and epidermal differentiation were upregulated at a later stage which is consistent with a lower abundance of epithelial cells and fibroblasts at early stages of healing, an increase in layers of epithelium seen after one week and the early stages of fibroplasia. Some of these changes in gene expression found during the first week of wound healing have been documented previously in gene  73 and protein expression studies over several decades. Therefore, it was not surprising to observe the up-regulation of many extracellular matrix genes, such as collagen type I, III, VI and several proteoglycans that associate with granulation tissue formation in 7-dayold wounds. Equally expected was the relative down-regulation of these genes in earlier wounds that can be explained by the low abundance of mRNA derived from fewer fibroblasts present in the wound biopsy relative to that from inflammatory cells. Indeed, inflammatory cells are likely to contribute to up-regulation of various cytokines and growth factors during early wound healing. For example, the present study revealed IL-8 co-expression with GRO-a/CXC ligand 1 one day after injury in the superficial wound bed, which has been previously determined in cutaneous wound healing [232]. GROl and IL-8 are both potent regulators/activators of neutrophil chemotaxis and therefore play a critical role in the early inflammation phase of wound healing [233, 234]. Expression of matrix metalloproteinases (MMP-1, MMP-3, MMP-9, and MMP-10) was highly up-regulated in 1-day-old wounds suggesting the need for active matrix degradation to initiate cell migration, clot degradation and clearance of wound debris. This data agrees with previous reports demonstrating MMP expression profiles in both connective tissue and epithelium during 1, 3, and 7 days post- wounding [235] . Interestingly, the expression of MMP-1 was one of the highest of all genes in all time points studied. The simplest explanation is that the levels in non-wounded tissue are negligible but this should be true for many genes in non-wounded tissue. One would have also expected the expression of MMP-1 to show decline in the 7-day-old wounds in which high synthesis of extracellular matrix is a characteristic feature. Several previous reports have indicated that MMP-1 expression is strongly induced by migrating  74 keratinocytes in human skin wounds and gradually declines when re-epithelialization is complete [236, 237]. Many different cell types are likely to contribute to the total MMP1 mRNA up-regulation and further in situ hybridization studies are necessary to pinpoint the cellular origin of MMP-1 in regenerating mucosal wounds. For normal wound healing, it is important that MMP activity stays well controlled as chronic non-healing wounds are characterized by excessive MMP activity [238]. In palatal wounds, an upregulation of TIMP-1 was seen only in 3-day-old wounds. MMP-1 activity is regulated at various levels and, therefore, no conclusions can be drawn from the relative expression imbalance of MMP-1 and TIMP-1. General proteolytic control in these wounds was sustained as evidenced by a large increase in serine protease inhibitors and proteinase inhibitors in all wounds. High expression of TGF63 has been linked to scarless wound healing, and TGFB1 for scar formation, respectively [3]. TGFB1 and TGF63 were both up-regulated at the same time in 7-day-old wounds (4.7 and 2.6 fold increase respectively). Interestingly, CTGF was up-regulated at all time points (2.7, 4.2, 4.7 fold increase for 1, 3, 7 day wounds, respectively). CTGF has been considered as a downstream mediator of TGFB1 [173] but certainly their gene expression profiles are not linked in oral wounds. More studies are needed to compare gene expression levels in non-wounded and wounded tissue that heal with scars or with regeneration, to unravel whether these processes are inbuilt to the tissue itself. The gene expression findings of the present study are largely confirmatory which is of great importance in order to lend validity to the use of microarray technologies. The major goal of this gene profiling study was to uncover novel genes that may play key  75 roles in scarless oral wound healing. To explore whether the alterations in mRNA levels seen in the array would reflect an induction of respective protein levels and to find out which cells are involved, we investigated the tissue distribution of Ceacam-6 protein that is a novel gene in the wound healing context. This gene was chosen for further study based on such a significant increase in expression at the early time points in the wound. Ceacam-6 is a member of the carcinoembryonic antigen family together with Ceacam-1 and Ceacam-7. Ceacam-6 is expressed in normal tissues including the epithelium lining the gastrointestinal tract, eccrine sweat glands, lung epithelia, testes, granulocytes and monocytes but its expression in wounds has not been investigated [239]. Ceacam-6 is upregulated, however, in colorectal tumors [240]. Biological functions of Ceacam-6 include homotypic or heterotypic cell adhesion [241], inhibition of anoikis [242] and binding to a variety of bacteria [243-245]. Ceacam-6 may therefore play a role in the innate immune response by binding and trapping microorganisms and preventing bacterial invasion [246]. It is possible that induction of Ceacam-6 expression in epithelial cells at the wound margin could play a role in self-defence against bacterial challenge of the oral cavity. Interestingly, Ceacam-6 was also detected in connective tissue cells (fibroblasts) in the later phases of wounding (1-2 weeks post-injury), where it may also play a role in inhibiting anoikis when the cells need to detach from the matrix for migration. Further studies of Ceacam-6 in oral wound healing will be necessary to clarify the function of this protein. Previous gene profiling studies in human wounds have been limited. The baseline gene expression profiles in uninjured skin, and the gene expression changes at thirty minutes and one hour following initial injury have been determined [222, 223]. It was  76 found that gene expression between different individuals was largely identical except for variable expression of 71 genes (1.7%), including many regulatory proteins such as DNA binding proteins and mitochondrial genes[223]. The follow-up study comparing uninjured skin to early injury (at thirty minutes and 1 hour), showed a consistent increase in gene expression in 124 of 4000 genes (3%) at thirty minutes [222]. No significant down regulation was noted for any gene at thirty minutes. Most of the up-regulated genes were involved in transcription and signalling. By one hour, only 46 of 4000 genes (1.15%) were significantly increased; however, 264 of 4000 (6.6%) of genes were significantly down-regulated, which included many structural genes and many other genes with unknown functions. Tsou et al. compared gene expression of hypertrophic scar tissue to normal scar tissue and uninjured skin and found that 44 genes were overexpressed and 124 genes were under-expressed in hypertrophic scars compared with normal scars [218]. Other recent studies have also found different gene expression patterns in hypertrophic scars and in keloids when compared to normal skin, including upregulation of TGF-P 1 in keloids [220, 221]. In addition to humans, Affymetrix oligonucleotide microarrays have been used to analyze burn wounds in rats [224, 225]. Differential expression of genes was associated with events such as cell survival and death, cell growth regulation, cell metabolism, inflammation and immune response. The time of appearance of various genes matched the known time sequence of a healing burn wound. A study conducted on mice with scald wounds, revealed significant alterations in the expression of genes, with the greatest changes between 3 and 14 days after wounding [225]. It was also found that gene expression was a transient and time-dependant upregulation. Gene ontologies in this study showed an integrated upregulation of  77 inflammatory and protease genes at early time points in the scald wound. The present study is currently the only human large-scale gene expression study comparing the different phases of oral wound healing for greater than 48 hours. A recent study by Warburton et al. utilized cDNA microarrays to characterize early human mucosal wound healing at 2-4 hours and 48 hours post-wounding [247]. Constitutive gene expression was also determined for the normal, pre-wounded mucosa. After 2 to 4 hours of healing, an upregulation of genes involved in transcription, such as c-fos and jun-D, was noted. After 48 hours, it was found that genes involved in tissue remodelling, such as MMP's, were significantly upregulated, while genes involved in cell adhesion were downregulated. This latter finding is consistent with the one and three day wounds of the present study. Some of the limitations of the present study include the small sample size analyzed for each gingival wound and the use of microarrays for the early phases of wound healing only. Further studies are needed to corroborate the findings of this study and should incorporate gene expression profiles of later wounds. In addition, it is imperative to compare gene expression profiles of regenerating mucosal wounds to skin wounds that heal by scar formation in an animal model that closely reflects human skin and oral mucosa.  78  CHAPTER VI \  Conclusions and Future Directions 6.1 Conclusions 1. Clinical healing occurred with regeneration or lack of scarring in the masticatory mucosa of the majority of the individuals in the study. 2. Morphologic analysis of all wounds revealed consistent progression in healing from 1 to 60 days, with the final wound (60 days of healing) showing complete regeneration (restored epithelium, rete ridges, and connective tissue). 3. Microarray data revealed large changes in gene expression during the early phases of wound healing, many of which were confirmatory of previous wound healing studies, and others which were novel in the context of wound healing. 4. The expression of one such novel gene, Ceacam-6, was confirmed by the localization of the protein in 1-day to 14-day wounds and may play important roles in the oral wound healing process.  6.2 Future Directions Future studies will be necessary to corroborate the findings of the present study and should include microarray analysis of later wound time points. It will be important to compare gene expression profiles in regenerating oral mucosal wounds to scar forming skin wounds in both an animal model and in humans. Further studies clarifying the role of Ceacam-6 in both oral and cutaneous wounds should also be conducted. Finally, it may be of interest to compare wound healing of regenerating masticatory mucosa to other oral tissues that have been shown to scar (such as alveolar mucosa).  79 BIBLIOGRAPHY 1. 2. 3. 4. 5. 6.  Sicard, R., Leukocyte and immunological influence on regeneration of amphibian forelimbs. Regulation of vertebrate limb regeneration, ed. R. Sicard. 1985, Cambridge: Oxford University Press. 128-145. Bullard, K.M.j M.T. Longaker, and H.P. Lorenz, Fetal wound healing: current biology. World J Surg, 2003. 27(1): p. 54-61. Ferguson, M.W. and S. O'Kane, Scar-free healing:fromembryonic mechanisms to adult therapeutic intervention. Philos Trans R Soc Lond B Biol Sci, 2004. 359(1445): p. 839-50. Cowin, A.J., et al., Endogenous inflammatory response to dermal wound healing in the fetal and adult mouse. Dev Dyn, 1998. 212(3): p. 385-93. Sciubba, J.J., J.P. Waterhouse, and J. Meyer, A fine structural comparison of the healing of incisional wounds of mucosa and skin. J Oral Pathol, 1978. 7(4): p. 214-27. Stephens, P., et al., Skin and oralfibroblastsexhibit phenotypic differences in extracellular matrix reorganization and matrix metalloproteinase activity. Br J  7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.  Dermatol, 2001.144(2): p. 229-37. Szpaderska, A.M., J.D. Zuckerman, and L.A. DiPietro, Differential injury responses in oral mucosal and cutaneous wounds. J Dent Res, 2003. 82(8): p. 621-6. Irwin, C.R., et al., Regulation offibroblast-inducedcollagen gel contraction by interleukin-1 beta. J Oral Pathol Med, 1998. 27(6): p. 255-9. Bartold, P.M., L.J. Walsh, and A.S. Narayanan, Molecular and cell biology of the gingiva. Periodontol 2000, 2000. 24: p. 28-55. Gartner, L.P., Oral anatomy and tissue types. Semin Dermatol, 1994.13(2): p. 6873. Menon, G.K., New insights into skin structure: scratching the surface. Adv Drug Deliv Rev, 2002. 54 Suppl 1: p. S3-17. Garant, P., Oral Cells and Tissues. 2003, Chicago: Quintessence Publishing Co. Madison, K.C., Barrier function of the skin: "la raison d'etre" of the epidermis. J Invest Dermatol, 2003.121(2): p. 231-41. Hennings, H., et al., Calcium regulation ofgrowth and differentiation of mouse epidermal cells in culture. Cell, 1980.19(1): p. 245-54. Menon, G.K., S. Grayson, and P.M. Elias, Ionic calcium reservoirs in mammalian epidermis: ultrastructural localization by ion-capture cytochemistry. J Invest Dermatol, 1985. 84(6): p. 508-12. Jarnik, M . , M.N. Simon, and A.C. Steven, Cornified cell envelope assembly: a model based on electron microscopic determinations of thickness and projected density. J Cell Sci, 1998. I l l ( Pt 8): p. 1051-60. Schroeder, H.E. and M . Amstad-Jossi, Type and variability of the stratum corneum in normal and diseased human oral stratified epithelia. J Biol Buccale,  1984.12(2): p. 101-16. Philipsen, H.P., P. Cleaton-Jones, and A.V. Fisker, Correlative light microscopic and scanning electron microscopic study of completely and incompletely orthokeratinized rat oral epithelium. Scand J Dent Res, 1982. 90(4):'p. 255-62.  80 19.  22.  Kanitakis, J., Anatomy, histology and immunohistochemistry of normal human Eur J Dermatol, 2002.12(4): p. 390-9; quiz 400-1. Ghohestani, R.F., et al., Molecular organization of the cutaneous basement membrane zone. Clin Dermatol, 2001.19(5): p. 551-62. Chavrier, C , et al., Connective tissue organization of healthy human gingiva. Ultrastructural localization of collagen types I-III-IV. J Periodontal Res, 1984. 19(3): p. 221-9. Romanos, G., et al., Immunohistochemical distribution of the collagen types IV, V,  23.  Viand glycoprotein laminin in the healthy rat, marmoset (Callithrix jacchus) and human gingivae. Matrix, 1991. 11(2): p. 125-32. Bratt, P., et al., Isolation and characterization of bovine gingival proteoglycans  skin.  20. 21.  versican and decorin. Int J Biochem, 1992. 24(10): p. 1573-83.  24.  Kahari, V.M., H. Larjava, and J. Uitto, Differential regulation of extracellular matrix proteoglycan (PG) gene expression. Transforming growth factor-beta 1 up-regulates biglycan (PGI), and versican (large fibroblast PG) but downregulates decorin (PGII) mRNA levels in human fibroblasts in culture. J Biol  25. 26.  27. 28. 29.  Chem, 1991.266(16): p. 10608-15. Larjava, H., L. Hakkinen, and F. Rahemtulla, A biochemical analysis of human periodontal tissue proteoglycans. Biochem J, 1992. 284 ( Pt 1): p. 261-1 A. Gallagher, J.T., The extendedfamily ofproteoglycans: social residents of the pericellular zone. CurrOpinCell Biol, 1989.1(6): p. 1201-18. Pitaru, S., et al., Immunoelectron microscopic studies on the distributions of fibronectin and actin in a cellular dense connective tissue: the periodontal ligament of the rat. J Periodontal Res, 1987. 22(1): p. 64-74. Bartold, P.M., Connective tissues of the periodontium. Research and clinical implications. Aust Dent J, 1991. 36(4): p. 255-68. Salonen, J., et al., Immunohistochemical localization of SPARC (osteonectin) and denatured collagen and their relationship to remodelling in rat dental tissues.  Arch Oral Biol, 1990.35(5): p. 337-46. 30.  Becker, J., D. Schuppan, and S. Muller, Immunohistochemical distribution of  31.  collagens type I, III, IV and VI, of undulin and of tenascin in oral fibrous hyperplasia. J Oral Pathol Med, 1993. 22(10): p. 463-7. Schroeder, H., Differentiation of Human Oral Stratified Epithelia. 1981, Basel:  32.  Karger. Swartzendruber, D.C., et al., Organization of the intercellular spaces ofporcine  33. 34. 35.  epidermal and palatal stratum corneum: a quantitative study employing ruthenium tetroxide. Cell Tissue Res, 1995. 279(2): p. 271-6. Page, R.C., et al., Advances in the pathogenesis of periodontitis: summary of developments, clinical implications and future directions. Periodontol 2000,1997.  14: p. 216-48. Takata, T., Oral wound healing concepts in periodontology. Curr Opin Periodontol, 1994: p. 119-27. Clark, R., Wound Repair. Overview and General Considerations, in The Molecular and Cellular Biology of Wound Repair, R. Clark, Editor. 1996, Plenum Press: New York. p. 3-50.  81 36. 37.  Yamamoto, T. and C.G. Cochrane, Guinea pig Hageman factor as a vascular permeability enhancement factor. Am J Pathol, 1981.105(2): p. 164-75. Furie, B. and B.C. Furie, The molecular basis of blood coagulation. Cell, 1988.  53(4): p. 505-18. 38.  39. 40.  41. 42.  43. 44.  Legrand, Y.J., et al., Activation of platelets by microfibrils and collagen. A comparative study. Lab Invest, 1986. 54(5): p. 566-73. Ginsberg, M.H., J.C. Loftus, and E.F. Plow, Cytoadhesins, integrins, and platelets. Thromb Haemost, 1988. 59(1): p. 1-6. Ruggeri, Z.M., von Willebrandfactor andfibrinogen. Curr Opin Cell Biol, 1993.  5(5): p. 898-906. Weksler, B., Platelets. Inflammation: Basic Principle and Clinical Correlates, ed. J. Gallin, Goldstein IM, Snyderman R. 1992, New York: Raven Press. Weksler, B.B., Roles for human platelets in inflammation. Prog Clin Biol Res, 1988. 283: p. 611-38. Wahl, L., Inflammation. Wound Healing. Biochemical and Clinical Aspects, ed. K. Cohen, Diegelmann RF, Lindblad WJ. 1992, Philadelphia: WB Saunders. Szpaderska, A . M . , et al., The effect of thrombocytopenia on dermal wound  healing. J Invest Dermatol, 2003.120(6): p. 1130-7. 45.  Moncada, S., et al., An enzyme isolatedfromarteries transforms prostaglandin endoperoxides to an unstable substance that inhibits platelet aggregation. Nature,  46.  Stern, D., et al., Interaction of antithrombin III with bovine aortic segments. Role of heparin in binding and enhanced anticoagulant activity. J Clin Invest, 1985.  1976. 263(5579): p. 663-5.  47.  75(1): p. 272-9. Loedam, J., Meijers, JCM, Sixma, JJ,and Bouma, BN, Inactivation of human factor VIII by activated protein CCofactor activity ofprotein S and protective effect of von Willebrandfactor. Journal of Clinical Investigation, 1988. 82: p.  1236-1243. 48. 49.  Loskutoff, D.J. and T.E. Edgington, Synthesis of a fibrinolytic activator and inhibitor by endothelial cells. Proc Natl Acad Sci U S A , 1977. 74(9): p. 3903-7. Klebanoff, S.J., Oxygen metabolitesfromphagocytes. Inflammation: Basic  50.  Principles and Clinical Correlates, ed. J. Gallin, Goldstein IM, Snyderman R. 1992, New York: Raven Press. 541-601. Smith, R.P., J.M. Lackie, and P.C. Wilkinson, The effects of chemotactic factors on the adhesiveness of rabbit neutrophil granulocytes. Exp Cell Res, 1979.  51.  52.  53.  122(1): p. 169-77. Gimbrone, M.A., Jr., A.F. Brock, and A.I. Schafer, Leukotriene B4 stimulates polymorphonuclear leukocyte adhesion to cultured vascular endothelial cells. J Clin Invest, 1984. 74(4): p. 1552-5. Tonnesen, M.G., L.A. Smedly, and P.M. Henson, Neutrophil-endothelial cell interactions. Modulation of neutrophil adhesiveness induced by complement fragments C5a and C5a des arg andformyl-methionyl-leucyl-phenylalanine in vitro. J Clin Invest, 1984. 74(5): p. 1581 -92. Albelda, S.M. and C A . Buck, Integrins and other cell adhesion molecules. Faseb J, 1990. 4(11): p. 2868-80.  82 54.  55. 56. 57.  58. 59.  60.  61. 62.  63. 64. 65. 66.  67.  68.  69. 70.  71.  Newman, S.L., J.E. Henson, and P.M. Henson, Phagocytosis of senescentneutrophils by human monocyte-derived macrophages and rabbit inflammatory macrophages. J Exp Med, 1982.156(2): p. 430-42. Postlethwaite, A.E. and A.H. Kang, Collagen-and collagenpeptide-induced chemotaxis of human blood monocytes. J Exp Med, 1976.143(6): p. 1299-307. Senior, R.M., G.L. Griffin, and R.P. Mecham, Chemotactic activity of elastinderivedpeptides. J Clin Invest, 1980. 66(4): p. 859-62. Clark, R.A., et al., Cryptic chemotactic activity offibronectin for human monocytes resides in the 120-kDa fibroblastic cell-binding fragment. J Biol Chem, 1988. 263(24): p. 12115-23. Bar-Shavit, R., et al., Chemotactic response of monocytes to thrombin. J Cell Biol, 1983. 96(1): p. 282-5. Wahl, S.M., et al., Transforming growth factor type beta induces monocyte chemotaxis and growth factor production. Proc Natl Acad Sci U S A , 1987. 84(16): p. 5788-92. Campbell, E.J., et al., Monocyte procollagenase and tissue inhibitor of metalloproteinases. Identification, characterization, and regulation of secretion. J Biol Chem, 1987. 262(33): p. 15862-8. Nathan, C.F., Secretory products of macrophages. J Clin Invest, 1987. 79(2): p. 319-26. Leibovich, S.J. and R. Ross, The role of the macrophage in wound repair. A study with hydrocortisone and antimacrophage serum. Am J Pathol, 1975. 78(1): p. 71100. Martin, P., et al., Wound healing in the PU.l null mouse—tissue repair is not dependent on inflammatory cells. Curr Biol, 2003.13(13): p. 1122-8. Gillitzer, R. and M . Goebeler, Chemokines in cutaneous wound healing. J Leukoc Biol, 2001. 69(4): p. 513-21. Singer, A.J. and R.A. Clark, Cutaneous wound healing. N Engl J Med, 1999. 341(10): p. 738-46. Stenn, K.S., Epithelialization, in Wound Healing. Biochemical and Clinical Aspects, K. Cohen, Diegelmann R.F., Lindblad W.J., Editor. 1992, WB Saunders: Philadelphia, p. 115-127. Goliger, J.A. and D.L. Paul, Wounding alters epidermal connexin expression and gap junction-mediated intercellular communication. Mol Biol Cell, 1995. 6(11): p. 1491-501. Gabbiani, G., C. Chaponnier, and I. Huttner, Cytoplasmic filaments and gap junctions in epithelial cells and myofibroblasts during wound healing. J Cell Biol, 1978. 76(3): p. 561-8. Larjava, H., et al., Keratinocyte integrins in wound healing and chronic inflammation of the human periodontium. Oral Dis, 1996. 2(1): p. 77-86. Steffensen, B., L. Hakkinen, and H. Larjava, Proteolytic events of wound-healing-coordinated interactions among matrix metalloproteinases (MMPs), integrins, and extracellular matrix molecules. Crit Rev Oral Biol Med, 2001. 12(5): p. 37398. Makela, M . , et al., Matrix metalloproteinase 2 (gelatinase A) is related to migration of keratinocytes. Exp Cell Res, 1999. 251(1): p. 67-78.  83 72. 73. 74. 75.  76. 77. 78. 79.  80.  81. 82.  83. 84.  85. 86. 87. 88.  89. 90.  Bugge, T.H., et al., Loss offibrinogen rescues mice from the pleiotropic effects of plasminogen deficiency. Cell, 1996. 87(4): p. 709-19. Romer, J., et al., Impaired wound healing in mice with a disrupted plasminogen gene. Nat Med, 1996. 2(3): p. 287-92. Mignatti, P. and D.B. Rifkin, Plasminogen activators and matrix metalloproteinases in angiogenesis. Enzyme Protein, 1996. 49(1-3): p. 117-37. Zhang, K., etal., Restricted expression andfunction oflaminin 1-binding integrins in normal and malignant oral mucosal keratinocytes. Cell Adhes Commun, 1996. 4(3): p. 159-74. Hakkinen, L., V.J. Uitto, and H. Larjava, Cell biology of gingival wound healing. Periodontal 2000, 2000. 24: p. 127-52. Giancotti, F.G. and E. Ruoslahti, Integrin signaling. Science, 1999. 285(5430): p. 1028-32. Larjava, H., Keratinocyte Interactions with Fibronectin during Wound Healing, in Cell Invasion, J. Heino, Kahari V-L, Editor. 2001. p. 42-64. Hakkinen, L., et al., Immunolocalization of tenascin-C, alpha9 integrin subunit, and alphavbeta6 integrin during wound healing in human oral mucosa. J Histochem Cytochem, 2000. 48(7): p. 985-98. Kainulainen, T., et al., Laminin-5 expression is independent of the injury and the microenvironment during reepithelialization of wounds. J Histochem Cytochem, 1998. 46(3): p. 353-60. Larjava, H., et al., Anti-integrin antibodies induce type IVcollagenase expression in keratinocytes. J Cell Physiol, 1993. 157(1): p. 190-200. Goldfinger, L.E., et al., The alpha3 laminin subunit, alpha6beta4 and alpha3betal integrin coordinately regulate wound healing in cultured epithelial cells and in the skin. J Cell Sci, 1999.112 ( Pt 16): p. 2615-29. Preissner, K.T., Structure and biological role of vitronectin. Annu Rev Cell Biol, 1991. 7: p. 275-310. Koivisto, L., et al., Different integrins mediate cell spreading, haptotaxis and lateral migration of HaCaT keratinocytes on fibronectin. Cell Adhes Commun, 1999. 7(3): p. 245-57. Larjava, H., et al., Expression of integrins and basement membrane components by wound keratinocytes. J Clin Invest, 1993. 92(3): p. 1425-35. Vaughan, R.B., Trinkaus J.P., Movements of epithelial cell sheets in vitro. Journal of Cell Science, 1966.1: p. 407-413. Winter, G.D., Movement of epidermal cells over the wound surface. Developmental Biology of Skin, 1964. 5: p. 113-127. Ortonne, J.P., et al., Immunomorphological and ultrastructural aspects of keratinocyte migration in epidermal wound healing. Virchows Arch A Pathol Anat Histol, 1981. 392(2): p. 217-30. Sciubba, J.J., Regeneration of the basal lamina complex during epithelial wound healing. J Periodontal Res, 1977.12(3): p. 204-17. Winter, G.D., Epidermal regeneration studied in the domestic pig. Epidermal Wound Healing, ed. H.I. Maibach, Rovee, D.T. 1972, Chicago: Yearbook Medical Publishing.  84 91.  92. 93.  94.  95. 96. 97.  98. 99. 100.  101. 102.  103.  104.  105.  106.  107.  Levine, J.H., et al., Spatial and temporal patterns of immunoreactive transforming growth factor beta 1, beta 2, and beta 3 during excisional wound repair. Am J Pathol, 1993.143(2): p. 368-80. Van Waes, C , Cell adhesion and regulatory molecules involved in tumor formation, hemostasis, and wound healing. Head and Neck, 1995.17: p. 140-147. Clark, R.A., et al., Fibronectin andfibrin provide a provisional matrix for epidermal cell migration during wound reepithelialization. J Invest Dermatol, 1982.79(5): p. 264-9. Gipson, I.K., S.J. Spurr-Michaud, and A.S. Tisdale, Hemidesmosomes and anchoringfibrilcollagen appear synchronously during development and wound healing. Dev Biol, 1988. 126(2): p. 253-62. Clark, R.A., Regulation offibroplasiain cutaneous wound repair. Am J Med Sci, 1993. 306(1): p. 42-8. Nathan, C. and M . Sporn, Cytokines in context. J Cell Biol, 1991.113(5): p. 9816. Damsky, C H . and Z. Werb, Signal transduction by integrin receptors for extracellular matrix: cooperative processing of extracellular information. Curr Opin Cell Biol, 1992. 4(5): p. 772-81. Hunt, T.K., Wound Healing and Wound Infection: Theory and Surgical Practice. 1980, New York: Appleton-Century-Crofts. Sporn, M.B. and A.B. Roberts, Peptide growth factors and inflammation, tissue repair, and cancer. J Clin Invest, 1986. 78(2): p. 329-32. Loef, E.B., Proper, J.A., Goustin, A.S., Shipley, G.D., DiCorleto, P.E., Moses, H.L., Induction of c-sis RNA and activity similar to platelet-derived growth factor by transforming growth factor-B: A proposed model for indirect mitogenesis involving autocrine activity. Proc. Natl. Acad. Sci. USA, 1986. 83: p. 2453-2457. Assoian, R.K., et al., Transforming growth factor-beta controls receptor levels for epidermal growth factor in NRKfibroblasts. Cell, 1984. 36(1): p. 35-41. Igarashi, A., et al., Regulation of connective tissue growth factor gene expression in human skinfibroblastsand during wound repair. Mol Biol Cell, 1993. 4(6): p. 637-45. Singer, II, et al., Cell surface distribution offibronectinand vitronectin receptors depends on substrate composition and extracellular matrix accumulation. J Cell Biol, 1988.106(6): p. 2171-82. Giancotti, F.G. and E. Ruoslahti, Elevated levels of the alpha 5 beta I fibronectin receptor suppress the transformed phenotype of Chinese hamster ovary cells. Cell, 1990. 60(5): p. 849-59. Chan, B.M., et al., Adhesion to vascular cell adhesion molecule 1 andfibronectin. Comparison of alpha 4 beta 1 (VLA-4) and alpha 4 beta 7 on the human B cell HneJY. J Biol Chem, 1992. 267(12): p. 8366-70. Welch, M.P., G.F. Odland, and R.A. Clark, Temporal relationships of F-actin bundle formation, collagen andfibronectinmatrix assembly, and fibronectin receptor expression to wound contraction. J Cell Biol, 1990.110(1): p. 133-45. Seppa, H., et al., Platelet-derived growth factor in chemotactic forfibroblasts.J Cell Biol, 1982. 92(2): p. 584-8.  85 108.  Gailit, J., Bueller, H., Clark, R., Platelet-derived growth factor and inflammatory cytokines have differential effects on the expression of integrins alfil and a5fil by human dermal fibroblasts. J. Invest Dermatol., 1995. 109. Postlethwaite, A.E., R. Snyderman, and A.H. Kang, Generation of a fibroblast chemotactic factor in serum by activation of complement. J Clin Invest, 1979. 64(5): p. 1379-85. 110. Postlethwaite, A.E., et al., Induction offibroblast chemotaxis by fibronectin. Localization of the chemotactic region to a 140,000-molecular weight nongelatin-binding fragment. J Exp Med, 1981. 153(2): p. 494-9. 111. Postlethwaite, A.E. and J.M. Seyer, Fibroblast chemotaxis induction by human recombinant interleukin-4. Identification by synthetic peptide analysis of two chemotactic domains residing in amino acid sequences 70-88 and 89-122. J Clin Invest, 1991. 87(6): p. 2147-52. 112. Harris, A.K., P. Wild, and D. Stopak, Silicone rubber substrata: a new wrinkle in the study of cell locomotion. Science, 1980. 208(4440): p. 177-9. 113. Carter, S.B., Cell movement and cell spreading: a passive or an active process? Nature, 1970. 225(5235): p. 858-9. 114. Hsieh, P. and L.B. Chen, Behavior of cells seeded in isolatedfibronectin matrices. J Cell Biol, 1983. 96(5): p. 1208-17. 115. Trinkaus, J.P., Cells into Organs. The forces That Shape the Embryo. 1984, Englewood Cliffs, NJ: Prentice-Hall. 116. Clark, R.A., et al., Collagen matrices attenuate the collagen-synthetic response of culturedfibroblasts to TGF-beta. J Cell Sci, 1995.108 ( Pt 3): p. 1251-61. 117. Folkman, J. and P. A. D'Amore, Blood vessel formation: what is its molecular basis? Cell, 1996. 87(7): p. 1153-5. 118. Risau, W., Mechanisms of angiogenesis. Nature, 1997. 386(6626): p. 671-4. 119. Detmar, M . , et al., Hypoxia regulates the expression of vascular permeability factor/vascular endothelial growth factor (VPF/VEGF) and its receptors in human skin. J Invest Dermatol, 1997.108(3): p. 263-8. 120. Nissen, N.N., et al., Vascular endothelial growth factor mediates angiogenic activity during the proliferative phase of wound healing. Am J Pathol, 1998. 152(6): p. 1445-52. 121. Keck, P.J., et al., Vascular permeability factor, an endothelial cell mitogen related to PDGF. Science, 1989.246(4935): p. 1309-12. 122. Clark, R.A., et al., Blood vessel fibronectin increases in conjunction with endothelial cell proliferation and capillary ingrowth during wound healing. J Invest Dermatol, 1982. 79(5): p. 269-76. 123. Brooks, P.C., R.A. Clark, and D.A. Cheresh, Requirement of vascular integrin alpha v beta 3 for angiogenesis. Science, 1994. 264(5158): p. 569-71. 124. Pintucci, G., et al., Angiogenesis and the fibrinolytic system. Semin Thromb Hemost, 1996. 22(6): p. 517-24. 125. Darby, I., O. Skalli, and G. Gabbiani, Alpha-smooth muscle actin is transiently expressed by myofibroblasts during experimental wound healing. Lab Invest, 1990. 63(1): p. 21-9.  86 126.  127. 128.  129. 130.  131.  132.  133.  134. 135. 136. 137.  138.  139. 140.  141. 142. 143.  Desmouliere, A., et al., Transforming growth factor-beta 1 induces alpha-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing culturedfibroblasts. J Cell Biol, 1993.122(1): p. 103-11. Desmouliere, A., et al., Apoptosis during wound healing, fibrocontractive diseases and vascular wall injury. Int J Biochem Cell Biol, 1997. 29(1): p. 19-30. Singer, II, et al., In vivo co-distribution offibronectin and actin fibers in granulation tissue: immunofluorescence and electron microscope studies of the fibronexus at the myofibroblast surface. J Cell Biol, 1984. 98(6): p. 2091-106. Clark, R.A., et al., Platelet isoforms ofplatelet-derived growth factor stimulate fibroblasts to contract collagen matrices. J Clin Invest, 1989. 84(3): p. 1036-40. Montesano, R. and L. Orci, Transforming growth factor beta stimulates collagenmatrix contraction byfibroblasts:implications for wound healing. Proc Natl Acad Sci U S A , 1988. 85(13): p. 4894-7. Desmouliere, A., et al., Apoptosis mediates the decrease in cellularity during the transition between granulation tissue and scar. Am J Pathol, 1995.146(1): p. 5666. Mochitate, K., P. Pawelek, and F. Grinnell, Stress relaxation of contracted collagen gels: disruption of actinfilamentbundles, release of cell surface fibronectin, and down-regulation of DNA and protein synthesis. Exp Cell Res, 1991.193(1): p. 198-207. Riikonen, T., et al., Integrin alpha 2 beta 1 is a positive regulator of collagenase (MMP-1) and collagen alpha 1(1) gene expression. J Biol Chem, 1995. 270(22): p. 13548-52. Bailey, A.J., et al., Collagen polymorphism in experimental granulation tissue. Biochem Biophys Res Commun, 1975. 66(4): p. 1160-5. Levenson, S.M., et al., The Healing of Rat Skin Wounds. Ann Surg, 1965. 161: p. 293-308. Ferguson, M.W., et al., Scar formation: the spectral nature offetal and adult wound repair. Plast Reconstr Surg, 1996. 97(4): p. 854-60. Longaker, M.T., et al., Studies in fetal wound healing, VI. Second and early third trimester fetal wounds demonstrate rapid collagen deposition without scar formation. J Pediatr Surg, 1990. 25(1): p. 63-8; discussion 68-9. Shah, M . , D.M. Foreman, and M.W. Ferguson, Neutralisation of TGF-beta 1 and TGF-beta 2 or exogenous addition of TGF-beta 3 to cutaneous rat wounds reduces scarring. J Cell Sci, 1995.108 (Pt 3): p. 985-1002. Robson, M . C , Proliferative scarring. Surg Clin North Am, 2003. 83(3): p. 55769. Wijdeveld, M.G., et al., A histological study of tissue response to simulated cleft palate surgery at different ages in beagle dogs. Arch Oral Biol, 1991. 36(11): p. 837-43. Cornelissen, A . M . , et al., Palatal mucoperiosteal wound healing in the rat. Eur J Oral Sci, 1999.107(5): p. 344-51. Witte, M.B. and A. Barbul, General principles of wound healing. Surg Clin North Am, 1997. 77(3): p. 509-28. Serini, G. and G. Gabbiani, Mechanisms of myofibroblast activity and phenotypic modulation. Exp Cell Res, 1999. 250(2): p. 273-83.  87 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158.  Desmouliere, A., Gabbiani, G., The Role of the Myofibroblasts in Wound Healing and Fibrocontractive Diseases, in The Molecular and Cellular Biology of Wound Repair, R. Clark, Editor. 1996: New York. p. 391-423. Gabbiani, G., The myofibroblast in wound healing andfibrocontractive diseases. J Pathol, 2003. 200(4): p. 500-3. Teofoli, P., et al., Expression of Bcl-2, p53, c-jun and c-fos protooncogenes in keloids and hypertrophic scars. J Dermatol Sci, 1999.22(1): p. 31-7. Wassermann, R.J., et al., Differential production of apoptosis-modulatingproteins in patients with hypertrophic burn scar. J Surg Res, 1998. 75(1): p. 74-80. Datubo-Brown, D.D., Keloids: a review of the literature. Br J Plast Surg, 1990. 43(1): p. 70-7. Marneros, A.G., et al., Clinical genetics of familial keloids. Arch Dermatol, 2001. 137(11): p. 1429-34. Rudolph, R., Wide spread scars, hypertrophic scars, and keloids. Clin Plast Surg, 1987.14(2): p. 253-60. O'Leary, R., E.J. Wood, and P.J. Guillou, Pathological scarring: strategic interventions. Eur J Surg, 2002.168(10): p. 523-34. Baur, P.S., D.L. Larson, and T.R. Stacey, The observation of myofibroblasts in hypertrophic scars. Surg Gynecol Obstet, 1975.141(1): p. 22-6. Ehrlich, H.P., et al., Morphological and immunochemical differences between keloid and hypertrophic scar. Am J Pathol, 1994.145(1): p. 105-13. Zhang, K., et al., Myofibroblasts and their role in lung collagen gene expression during pulmonary fibrosis. A combined immunohistochemical and in situ hybridization study. Am J Pathol, 1994.145(1): p. 114-25. Dang, C M . , et al., Scarless fetal wounds are associated with an increased matrix metalloproteinase-to-tissue-derived inhibitor of metalloproteinase ratio. Plast  Reconstr Surg, 2003.111(7): p. 2273-85. Pilcher, B.K., et al., Role of matrix metdlloproteinases and their inhibition in cutaneous wound healing and allergic contact hypersensitivity. Ann N Y Acad Sci, 1999. 878: p. 12-24. Neely, A.N., et al., Gelatinase activity in keloids and hypertrophic scars. Wound Repair Regen, 1999. 7(3): p. 166-71. Kuroda, K. and H. Shinkai, Gene expression of types land III collagen, decorin, matrix metalloproteinases and tissue inhibitors of metalloproteinases in skin fibroblasts from patients with systemic sclerosis. Arch Dermatol Res, 1997.  159. 160. 161.  162.  289(10): p. 567-72. Cowin, A.J., et al., Expression ofTGF-beta and its receptors in murine fetal and adult dermal wounds. Eur J Dermatol, 2001.11(5): p. 424-31. Nath, R.K., et al., The expression of transforming growth factor type beta in fetal and adult rabbit skin wounds. J Pediatr Surg, 1994. 29(3): p. 416-21. Ghahary, A., et al., Enhanced expression of mRNA for transforming growth factor-beta, type land type III procollagen in human post-burn hypertrophic tissues. J Lab Clin Med, 1993.122(4): p. 465-73. Lee, T.Y., et al., Expression of transforming growth factor beta 1, 2, and 3 proteins in keloids. Ann Plast Surg, 1999. 43(2): p. 179-84.  scar  I  88 163.  164. 165. 166. 167. 168. 169.  170.  171.  172.  173.  174.  175.  176.  177.  178.  Sullivan, K.M., et al., A model of scarless human fetal wound repair is deficient in transforming growth factor beta. J Pediatr Surg, 1995. 30(2): p. 198-202; discussion 202-3. Ashcroft, G.S., et al., Mice lacking Smad3 show accelerated wound healing and an impaired local inflammatory response. Nat Cell Biol, 1999.1(5): p. 260-6. Flanders, K.C., et al., Mice lacking Smad3 are protected against cutaneous injury induced by ionizing radiation. Am J Pathol, 2002.160(3): p. 1057-68. Zhao, J., et al., Smad3 deficiency attenuates bleomycin-inducedpulmonary fibrosis in mice. Am J Physiol Lung Cell Mol Physiol, 2002. 282(3): p. L585-93. Sumiyoshi, K., et al., Smads regulate collagen gel contraction by human dermal fibroblasts. Br J Dermatol, 2003.149(3): p. 464-70. Govinden, R. and K.D. Bhoola, Genealogy, expression, and cellular function of transforming growth factor-beta. Pharmacol Ther, 2003. 98(2): p. 257-65. Hsu, M . , et al., Ontogeny of expression of transforming growth factor-beta 1 (TGF-beta 1), TGF-beta 3, and TGF-beta receptors I and II in fetal rat fibroblasts and skin. Plast Reconstr Surg, 2001.107(7): p. 1787-94; discussion 1795-6. Ghahary, A., et al., Mannose-6-phosphate/IGF-II receptors mediate the effects of IGF-1-induced latent transforming growth factor beta 1 on expression of type I collagen and collagenase in dermal fibroblasts. Growth Factors, 2000.17(3): p. 167-76. Telasky, C , et al., IFN-alpha2b suppresses the fibrogenic effects of insulin-like growth factor-1 in dermal fibroblasts. J Interferon Cytokine Res, 1998.18(8): p. 571-7. Bradham, D.M., et al., Connective tissue growth factor: a cysteine-rich mitogen secreted by human vascular endothelial cells is related to the SRC-induced immediate early gene product CEF-10. J Cell Biol, 1991.114(6): p. 1285-94. Brigstock, D.R., The connective tissue growth factor/cysteine-rich 61/nephroblastoma overexpressed (CCN) family. Endocr Rev, 1999. 20(2): p. 189-206. Igarashi, A., et al., Connective tissue growth factor gene expression in tissue sectionsfromlocalized scleroderma, keloid, and other fibrotic skin disorders. J Invest Dermatol, 1996.106(4): p. 729-33. Munz, B., et al., Overexpression of activin A in the skin of transgenic mice reveals new activities of activin in epidermal morphogenesis, dermal fibrosis and wound repair. Embo J, 1999.18(19): p. 5205-15. McCallion, R.L., Ferguson, M.W.J., Fetal Wound Healing and the Development ofAntiscarring Therapies for Adult Wound Healing. Second ed. The Molecular and Cellular Biology of Wound Repair, ed. R.A.F. Clark. 1996, New York: Plenum Press. Longaker, M.T., et al., Studies in fetal wound healing: II A fetal environment accelerates fibroblast migration in vitro. J Pediatr Surg, 1989. 24(8): p. 793-7; discussion 798. Somasundaram, K. and K. Prathap, The effect of exclusion of amniotic fluid on intra-uterine healing of skin wounds in rabbit foetuses. J Pathol, 1972.107(2): p. 127-30.  89 179.  180. 181. 182. 183. 184.  185. 186.  187.  188. 189.  190.  191. 192.  193. 194. 195.  196.  Gao, X., L.D. Devoe, and K.S. Given, Effects of amniotic fluid on proteases: a possible role of amniotic fluid in fetal wound healing. Ann Plast Surg, 1994. 33(2): p. 128-34; discussion 134-5. Lorenz, H.P., et al., Scarless wound repair: a human fetal skin model. Development, 1992.114(1): p. 253-9. Longaker, M.T., et al., Adult skin wounds in the fetal environment heal with scar formation. Ann Surg, 1994. 219(1): p. 65-72. Adzick, N.S. and M.T. Longaker, Scarless fetal healing. Therapeutic implications. Ann Surg, 1992. 215(1): p. 3-7. Cass, D.L., et al., Wound size and gestational age modulate scar formation in fetal wound repair. J Pediatr Surg, 1997. 32(3): p. 411-5. Mast, B.A., C T . Albanese, and S. Kapadia, Tissue repair in the fetal intestinal tract occurs with adhesions, fibrosis, and neovascularization. Ann Plast Surg, 1998. 41(2): p. 140-4; discussion 144-7. Lin, K.Y., et al., Fetal nerve healing: an experimental study. Plast Reconstr Surg, 1994. 93(7): p. 1323-33. Olutoye, O.O., et al., Aggregatory characteristics and expression of the collagen adhesion receptor in fetal porcine platelets. J Pediatr Surg, 1995. 30(12): p. 164953. Olutoye, O.O., et al., Lower cytokine release by fetal porcine platelets: a possible explanation for reduced inflammation after fetal wounding. J Pediatr Surg, 1996. 31(1): p. 91-5. Mast, B.A., et al., Scarless wound healing in the mammalian fetus. Surg Gynecol Obstet, 1992.174(5): p. 441-51. Adzick, N.S., et al., Comparison offetal, newborn, and adult wound healing by histologic, enzyme-histochemical, and hydroxyproline determinations. J Pediatr Surg, 1985.20(4): p. 315-9. Duncan, B.W., Qian J., Liu X., et al., Regulation ofprolyl hydroxylase activity in fetal and adult fibroblasts. Fetal Wound Healing, ed. N.S. Adzick, Longaker, M.T. 1992, New York: Elsevier. Thomas, B.L., Krurnmel, T.M., Melang, M . , Collagen Synthesis and type expression by fetal fibroblasts in vitro. Surg. Forum, 1988.39: p. 642-644. Chen, W.Y., et al., Differences between adult andfoetal fibroblasts in the regulation ofhyaluronate synthesis: correlation with migratory activity. J Cell Sci, 1989. 94(Pt3): p. 577-84. Estes, J.M., et al., Phenotypic andfunctional features of myofibroblasts in sheep fetal wounds. Differentiation, 1994. 56(3): p. 173-81. Bullard, K.M., Longaker, M.T., Lorenz, H.P., Fetal Wound Healing: Current Biology. World J Surg, 2003. 27: p. 54-61. Cass, D.L., Bullard, K.M., Sylvester, K.G., Epithelial integrin expression is rapidly upregulated in human fetal wound repair. Surg. Forum, 1995. 47: p. 765767. Dang, C , Beanes, S., Soo, C , Hedrick, M . , Lorenz, H.P., A high ratio ofTGF/33 to TGF/31 expression in wounds is associated with scarless repair. Wound Repair Regen,2001.9: p. 153.  90 197. 198. 199. 200. 201. 202. 203. 204. 205.  206.  207.  208.  209.  210.  211.  212.  213. 214.  Mack, J.A., et al., Hoxbl3 knockout adult skin exhibits high levels ofhyaluronan and enhanced wound healing. Faseb J, 2003.17(10): p. 1352-4. White, P., et al., Deletion of the homeobox gene PRX-2 affects fetal but not adult fibroblast wound healing responses. J Invest Dermatol, 2003.120(1): p. 135-44. Bodner, L., et al., The effect of selective desalivation on wound healing in mice. Exp Gerontol, 1991. 26(4): p. 357-63. Epstein, J.B. and C. Scully, The role of saliva in oral health and the causes and effects of xerostomia. J Can Dent Assoc, 1992. 58(3): p. 217-21. Edgar, W.M., Saliva: its secretion, composition and functions. Br Dent J, 1992. 172(8): p. 305-12. Field, F.K. and M.D. Kerstein, Overview of wound healing in a moist environment. Am J Surg, 1994.167(1 A): p. 2S-6S. Zelles, T., et al., Saliva and growth factors: the fountain of youth resides in us all. J Dent Res, 1995. 74(12): p. 1826-32. Noguchi, S., Y. Ohba, and T. Oka, Effect of salivary epidermal growth factor on wound healing of tongue in mice. Am J Physiol, 1991. 260(4 Pt 1): p. E620-5. Dvorak, H.F., et al., Vascular permeability factor/vascular endothelial growth factor and the significance of microvascular hyperpermeability in angiogenesis. Curr Top Microbiol Immunol, 1999. 237: p. 97-132. Kilcullen, J.K., et al., Nonviable Staphylococcus aureus and its peptidoglycan stimulate macrophage recruitment, angiogenesis,fibroplasia,and collagen accumulation in wounded rats. Wound Repair Regen, 1998. 6(2): p. 149-56. Sloan, P., Current concepts of the role offibroblasts and extracellular matrix in wound healing and their relevance to oral implantology. J Dent, 1991.19(2): p. 107-9. Irwin, C.R., et al., Inter- and intra-site heterogeneity in the expression offetal-like phenotypic characteristics by gingivalfibroblasts:potential significance for wound healing. J Cell Sci, 1994.107 ( Pt 5): p. 1333-46. Stephens, P., et al., A comparison of the ability of intra-oral and extra-oral fibroblasts to stimulate extracellular matrix reorganization in a model of wound contraction. J Dent Res, 1996. 75(6): p. 1358-64. Ravanti, L., et al., Transforming growth factor-beta induces collagenase-3 expression by human gingivalfibroblastsvia p38 mitogen-activated protein kinase. J Biol Chem, 1999. 274(52): p. 37292-300. Lorimier, S., et al., Tissue origin and extracellular matrix control neutral proteinase activity in humanfibroblastthree-dimensional cultures. J Cell Physiol, 1996.168(1): p. 188-98. Gron, B., et al., Oral fibroblasts produce more HGF and KGF than skin fibroblasts in response to co-culture with keratinocytes. Apmis, 2002.110(12): p. 892-8. Schena, M . , et al., Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science, 1995. 270(5235): p. 467-70. Lockhart, D.J., et al., Expression monitoring by hybridization to high-density oligonucleotide arrays. Nat Biotechnol, 1996.14(13): p. 1675-80.  91 215.  216. 217. 218. 219. 220. 221.  222. 223. 224. 225. 226.  227.  228.  229. 230. 231.  232.  233.  Joos, L., E. Eryuksel, and M.H. Brutsche, Functional genomics and gene microarrays—the use in research and clinical medicine. Swiss Med Wkly, 2003. 133(3-4): p. 31-8. Murphy, D., Gene expression studies using microarrays: principles, problems, and prospects. Adv Physiol Educ, 2002. 26(1-4): p. 256-70. Butte, A., The use and analysis of microarray data. Nat Rev Drug Discov, 2002. 1(12): p. 951-60. Tsou, R., et al., Analysis of hypertrophic and normal scar gene expression with cDNA microarrays. J Burn Care Rehabil, 2000. 21(6): p. 541-50. Dasu, M.R., et al., Gene expression profilesfromhypertrophic scar fibroblasts before and after IL-6 stimulation. J Pathol, 2004. 202(4): p. 476-85. Paddock, H.N., et al., Analysis ofgene expression patterns in human postburn hypertrophic scars. J Burn Care Rehabil, 2003. 24(6): p. 371-7. Chen, W., et al., Development ofgene microarray in screening differently expressed genes in keloid and normal-control skin. Chin Med J (Engl), 2004. 117(6): p. 877-81. Cole, J., et al., Early gene expression profile of human skin to injury using highdensity cDNA microarrays. Wound Repair Regen, 2001. 9(5): p. 360-70. Cole, J., et al., Comparison of normal human skin gene expression using cDNA microarrays. Wound Repair Regen, 2001. 9(2): p. 77-85. Spies, M . , et al., Gene expression analysis in burn wounds of rats. Am J Physiol Regul Integr Comp Physiol, 2002. 283(4): p. R918-30. Feezor, R.J., et al., Temporal patterns ofgene expression in murine cutaneous burn wound healing. Physiol Genomics, 2004.16(3): p. 341-8. Nowinski, D., et al., Analysis ofgene expression in fibroblasts in response to keratinocyte-derivedfactors in vitro: potential implications for the wound healing process. J Invest Dermatol, 2004.122(1): p. 216-21. Theilgaard-Monch, K., et al., The transcriptional activation program of human neutrophils in skin lesions supports their important role in wound healing. J Immunol, 2004.172(12): p. 7684-93. Kuo, W.P., et al., Deciphering gene expression profiles generatedfrom DNA microarrays and their applications in oral medicine. Oral Surg Oral Med Oral Pathol Oral Radiol Endod, 2004. 97(5): p. 584-91. Kuo, W.P., et al., Analysis of matched mRNA measurementsfromtwo different microarray technologies. Bioinformatics, 2002.18(3): p. 405-12. Kothapalli, R., et al., Microarray results: how accurate are they? BMC Bioinformatics, 2002. 3(1): p. 22. Li, J., M . Pankratz, and J.A. Johnson, Differential gene expression patterns revealed by oligonucleotide versus long cDNA arrays. Toxicol Sci, 2002. 69(2): p. 383-90. Engelhardt, E., et al., Chemokines IL-8, GROalpha, MCP-1, IP-10, and Mig are sequentially and differentially expressed during phase-specific infiltration of leukocyte subsets in human wound healing. Am J Pathol, 1998.153(6): p. 184960. Christopherson, K., 2nd and R. Hromas, Chemokine regulation of normal and pathologic immune responses. Stem Cells, 2001.19(5): p. 388-96.  92 234. 235. 236.  237.  238.  239. 240.  241.  242. 243.  244. 245.  246.  247.  Rennekarnpff, H.O., et al., Bioactive interleukin-8 is expressed in wounds and enhances wound healing. J Surg Res, 2000. 93(1): p. 41-54. Salo, T., et al., Expression of matrix metalloproteinase-2 and -9 during early human wound healing. Lab Invest, 1994. 70(2): p. 176-82. Saarialho-Kere, U.K., et al., Cell-matrix interactions modulate interstitial collagenase expression by human keratinocytes actively involved in wound healing. J Clin Invest, 1993. 92(6): p. 2858-66. Inoue, M . , et al., Collagenase expression is rapidly induced in wound-edge keratinocytes after acute injury in human skin, persists during healing, and stops at re-epithelialization. J Invest Dermatol, 1995.104(4): p. 479-83. Trengove, N.J., et al., Analysis of the acute and chronic wound environments: the role of proteases and their inhibitors. Wound Repair Regen, 1999. 7(6): p. 44252. Thompson, J.A., Molecular cloning and expression of carcinoembryonic antigen gene family members. Tumour Biol, 1995.16(1): p. 10-6. Neumaier, M . , et al., Biliary glycoprotein, a potential human cell adhesion molecule, is down-regulated in colorectal carcinomas. Proc Natl Acad Sci U S A , 1993. 90(22): p. 10744-8. Kuroki, M . , et al., Identification and comparison of residues critical for celladhesion activities of two neutrophil CD66 antigens, CEACAM6 and CEACAM8. J Leukoc Biol, 2001. 70(4): p. 543-50. Ordonez, C , et al., Human carcinoembryonic antigen functions as a general inhibitor of anoikis. Cancer Res, 2000. 60(13): p. 3419-24. Leusch, H.G., et al., The putative role of members of the CEA-gene family (CEA, NCA an BGP) as ligands for the bacterial colonization of different human epithelial tissues. Zentralbl Bakteriol, 1991. 275(1): p. 118-22. Chen, T., et al., Several carcinoembryonic antigens (CD66) serve as receptors for gonococcal opacity proteins. J Exp Med, 1997.185(9): p. 1557-64. Virji, M . , et al., Carcinoembryonic antigens (CD66) on epithelial cells and neutrophils are receptors for Opa proteins ofpathogenic neisseriae. Mol Microbiol, 1996. 22(5): p. 941-50. Hammarstrom, S., The carcinoembryonic antigen (CEA) family: structures, suggestedfunctions and expression in normal and malignant tissues. Semin Cancer Biol, 1999. 9(2): p. 67-81. Warburton, G., et al., Transcriptional events in a clinical model of oral mucosal tissue injury and repair. Wound Repair Regen, 2005.13(1): p. 19-26.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items